Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Although research has characterized the role of the NMDA receptor in the development of morphine tolerance (Trujillo and Akil, 1991; Tiseo and Inturrisi, 1993; Kolesnikov et al., 1994), few studies have examined the role of opioid pharmacology in this effect. Data demonstrate that NMDA receptor antagonists attenuate the development of tolerance to the µ-opioid morphine, however, studies do not always support a role for the NMDA receptor in the development of tolerance to the antinociceptive effects of - and -opioids (Kolesnikov et al., 1993, 1994; Elliott et al., 1994; Bilsky et al., 1996a). More surprisingly, the competitive NMDA receptor antagonist LY235959 attenuates tolerance to the µ-opioid morphine, but does not attenuate tolerance to the µ-opioids fentanyl and DAMGO (Bilsky et al., 1996a).

Research also shows that factors such as the duration of chronic treatment (Adams and Holtzman, 1990; Allen and Dykstra, 1999), the frequency of drug administration, and the maintenance dose of an opioid (Mucha et al., 1971; Fernandes et al., 1982; Schuh et al., 1996) can influence the development of opioid tolerance. Recently, we reported that the maintenance dose of morphine is related to the magnitude of morphine tolerance and the attenuation of that tolerance with an NMDA receptor antagonist (Allen and Dykstra, 2000).

In addition to the parameters of a dosing regimen, tolerance magnitude is also influenced by the particular opioid agonist that is administered chronically. It is generally accepted that chronic treatment with equi-effective doses of µ-opioids that differ in intrinsic efficacy produce varying degrees of tolerance. Specifically, studies have shown an inverse relationship between the magnitude of tolerance produced by an opioid and that opioid's intrinsic efficacy (Stevens and Yaksh, 1989; Sosnowski and Yaksh, 1990; Duttaroy and Yoburn, 1995). Thus, differences in the efficacy of an opioid agonist represent a parameter that may determine the effectiveness of an NMDA receptor antagonist to attenuate tolerance.

Research also suggests a relationship between the development of cross-tolerance and the efficacy of a test agonist. For example, different degrees of cross-tolerance develop to morphine, buprenorphine, and dezocine, even when the chronic treatment condition is the same (Craft and Dykstra, 1990; Paronis and Holtzman, 1992; Tiano et al., 1998). Again, there appears to be an inverse relationship between opioid efficacy and cross-tolerance magnitude. These differences may represent another variable that determines the extent to which an NMDA receptor antagonist modulates opioid tolerance.

The purpose of this study was to determine the role of the competitive NMDA receptor antagonist LY235959 in the attenuation of tolerance and cross-tolerance to µ-opioid agonists with different efficacies. These experiments tested the hypothesis that the attenuation of opioid tolerance by LY235959 is related to the magnitude of tolerance that develops to an opioid agonist. Research suggests that the least tolerance should develop to high-efficacy agonists and the most tolerance should develop to low-efficacy agonists. Therefore, the effectiveness of LY235959 at attenuating tolerance should be directly related to the efficacy of the treatment agonist. In addition, these experiments tested the hypothesis that attenuation of cross-tolerance with LY235959 is not related to cross-tolerance magnitude, but to the effectiveness of LY235959 to attenuate tolerance to the chronically administered opioid. If NMDA receptor activation represents the mechanism by which opioid tolerance develops, then attenuating the development of tolerance to an opioid with an NMDA receptor antagonist should also attenuate the cross-tolerance conferred to other opioid agonists.

Etorphine and dezocine were selected as opioids with high- and low-efficacy relative to morphine. Biochemical data show that etorphine is maximally effective at inhibiting prostaglandin E₁- and forskolin-stimulated cyclic AMP production in SH-SY5Y cells (Yu and Sadée, 1988). Data showing that the insurmountable µ-opioid antagonist clocinnamox (C-CAM) (Comer et al., 1992) antagonizes the antinociceptive effects of etorphine to a lesser degree than the antinociceptive effects of morphine (Pitts et al., 1998; Walker et al., 1998) also suggest that etorphine is a high-efficacy agonist relative to morphine. On the other hand, behavioral studies suggest that µ-specific irreversible and insurmountable antagonists, such as -funaltrexamine and C-CAM, are more effective antagonists of the behavioral effects of dezocine than of morphine (Picker, 1997; Tiano et al., 1998). Therefore, dezocine was selected as an agonist with low-efficacy relative to morphine.

The antinociceptive effects of etorphine, morphine, and dezocine were measured using the rat warm-water tail-withdrawal procedure, a procedure sensitive to the effects of opioids with different efficacies (Morgan and Picker, 1998; Morgan et al., 1999b; Smith et al., 1999).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. A total of 159 naive male Sprague-Dawley rats (Charles River Laboratories, Raleigh, NC) were used in these experiments. Rats were housed individually in Plexiglas cages and kept in a colony room on a 12-h light/dark cycle. Rats weighed approximately 250 g upon arrival and were allowed free access to Purina Rat Chow for approximately 2 weeks. During the 2 weeks of each experiment, body weight was restricted to 300 to 360 g by controlling daily food intake. Water was available ad libitum throughout the course of the experiment.

Antinociceptive Procedure. A warm-water tail-withdrawal procedure was used to measure the antinociceptive effects of morphine and LY235959. Rats were placed in Plexiglas restraining tubes that allowed their tails to hang freely. The distal 8 cm of each rat's tail was immersed into a thermos containing either 40°C (non-noxious) or 55°C (noxious) water. Tail-withdrawal latency was measured using a hand-held stopwatch. A 15-s cut-off time was imposed to prevent tissue damage.

Experimental Design. A total of 18 groups of six to eight rats were used in the tolerance and cross-tolerance experiments (Table 1). Three agonists were selected for chronic administration, the high-efficacy agonist etorphine, the intermediate-efficacy agonist morphine, and the low-efficacy agonist dezocine. Rats received chronic injections of a maximally effective dose of each drug alone or in combination with the competitive NMDA receptor antagonist LY235959. A dose of 3.0 mg/kg LY235959 was chosen based on the effectiveness of this dose of LY235959 to completely prevent the tolerance produced by a maximally effective dose of morphine under similar experimental conditions (Allen and Dykstra, 2000). Thus, there were six possible chronic treatment conditions: chronic 0.003 mg/kg etorphine alone; chronic 0.003 mg/kg etorphine + 3.0 mg/kg LY235959; chronic 10 mg/kg morphine alone; chronic 10 mg/kg morphine + 3.0 mg/kg LY235959; chronic 3.0 mg/kg dezocine alone; and chronic 3.0 mg/kg dezocine + 3.0 mg/kg LY235959. For each chronic treatment, tolerance and/or cross-tolerance to the antinociceptive effects of each opioid was assessed.

Induction of Tolerance. Etorphine, morphine, and dezocine dose-effect curves were first determined in the laboratory on the morning of day 1 (prechronic). Tolerance was then induced by administering maximally effective doses of etorphine (0.003 mg/kg), morphine (10 mg/kg), or dezocine (3.0 mg/kg) either alone or in combination with LY235959 (3.0 mg/kg) twice daily for 7 days (each injection or pair of injections in the home cage separated by 12 h). An etorphine, morphine, or dezocine dose-effect curve was then redetermined in the laboratory on the morning of day 8 (chronic). Thus, all chronic drug injections were administered in the home cage and testing of morphine's antinociceptive effects under prechronic and chronic conditions occurred in the laboratory. Testing on day 8 began 1 h before rats would normally receive their morning injection. Thus, the antinociceptive effects of the opioids were measured close to the time of the regularly scheduled morning injection. No signs of opiate withdrawal were observed during testing or at any time during the course of the tolerance experiment.

Pharmacokinetic and Pharmacodynamic Variables. A group of 24 rats was used in control experiments designed to rank order etorphine, morphine, and dezocine according to pharmacokinetic and pharmacodynamic variables. After the standard habituation procedures described above, rats were tested over 3 weeks in the following way. First, time-effect curves for 0.003 mg/kg etorphine, 3.0 mg/kg dezocine, and 10 mg/kg morphine were determined, each in a distinct group of rats (n = 6-9). For this time course assessment, our standard tail-withdrawal procedure was used and modified in the following way: After baseline tail-withdrawal latencies were recorded, rats were injected only once with either of 0.003 mg/kg etorphine, 3.0 mg/kg dezocine, or 10 mg/kg morphine. Rats were exposed to both the 40°C and 55°C water at 15, 30, 60, 120, 180, and 240 min. Rats were removed from the restraining tubes after 60 min and returned to the tubes 5 min before testing at the 120, 180, and 240 min time points. Individual time-effect curves were analyzed for area under the curve (AUC; Tallarida and Murray, 1987) and the individual AUC values for rats tested with etorphine, morphine, and dezocine were compared by ANOVA. Drug availability (a pharmacokinetic variable) was inferred from these duration of action curves based on the assumption that a relationship exists between the pharmacological response to a drug over time and the accessible concentration of that drug.

Data Analysis. Baseline data were analyzed using The SAS System for Windows, version 6.12. Individual ED₅₀ values for rats were derived from regression lines fit to the ascending limbs of the opioid dose-response curves. Potency ratios were determined using the method of Tallarida and Murray (1987). Before determining potency ratios, group prechronic and chronic dose-effect curves were first tested for parallelism (Tallarida and Murray, 1987). After confirmation of parallelism, potency ratios and 95% confidence intervals were calculated. If the 95% confidence interval for a potency ratio included 1.0 (a 1-fold shift), then it was concluded that no tolerance developed in that condition. Differences in tolerance magnitude were assessed by comparing the confidence intervals for different dosing conditions. Individual AUC values were determined using the method of Tallarida and Murray (1987) and compared with an ANOVA performed in The SAS System for Windows, version 6.12.

Drugs. Morphine sulfate and etorphine hydrochloride were generously supplied by the National Institute on Drug Abuse. Dezocine hydrochloride was generously supplied by Astra Pharmaceuticals (Westborough, MA) and LY235959 [()-6-phosphonomethyl-decahydroisoquinoline-3-carboxylic acid] by The Lilly Research Laboratories (Indianapolis, IN). All drugs were dissolved in physiological saline and injected subcutaneously (volume = 1 ml/kg) in the dorsal flank. Dezocine hydrochloride required the use of lactic acid to bring the drug into solution.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Etorphine Tolerance. Figure 1 shows that etorphine increased tail-withdrawal latencies from the 55°C water in a dose-dependent manner. Maximal or near-maximal antinociceptive effects in nontolerant rats occurred after administration of 0.003 mg/kg etorphine (Fig. 1, closed symbols). Thus, 0.003 mg/kg etorphine was selected as the dose for chronic treatment.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Dose-effect curves for rats treated twice daily for 7 days with 0.003 mg/kg etorphine (top) or 0.003 mg/kg etorphine + 3.0 mg/kg LY235959 (bottom). Ordinates, %MPE. Abscissae, dose in milligrams per kilogram, log scale.

Morphine Tolerance. Figure 2 shows that morphine also increased tail-withdrawal latencies from the 55°C water in a dose-dependent manner. A dose of 10 mg/kg morphine produced maximal or near-maximal antinociceptive effects in nontolerant rats (Fig. 2, closed symbols). Thus, 10 mg/kg morphine was chosen as the maintenance dose in this experiment.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Dose-effect curves for rats treated twice daily for 7 days with 10 mg/kg morphine (top) or 10 mg/kg morphine + 3.0 mg/kg LY235959 (bottom). Ordinates, %MPE. Abscissae, dose in milligrams per kilogram, log scale.

Dezocine Tolerance. Figure 3 shows that dezocine also increased tail-withdrawal latencies from the 55°C water in a dose-dependent manner. A dose of 3.0 mg/kg dezocine produced maximal or near-maximal effects in nontolerant rats (Fig. 3, closed symbols), and was therefore chosen as the maintenance dose.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Dose-effect curves for rats treated twice daily for 7 days with 3.0 mg/kg dezocine (top) or 3.0 mg/kg dezocine + 3.0 mg/kg LY235959 (bottom). Ordinates, %MPE. Abscissae, dose in milligrams per kilogram, log scale.

Cross-Tolerance. Figure 4 shows the effects of chronic administration of 0.003 mg/kg etorphine alone or in combination with 3.0 mg/kg LY235959 on the antinociceptive effects of morphine and dezocine. As noted previously, tolerance did not develop to etorphine under these dosing conditions and administration of LY235959 with etorphine did not alter the antinociceptive potency of etorphine (Fig. 1). However, cross-tolerance to morphine and to dezocine was conferred after seven twice-daily injections of 0.003 mg/kg etorphine (Fig. 4). Cross-tolerance was not conferred to morphine or to dezocine when 3.0 mg/kg LY235959 was coadministered with etorphine during the chronic drug regimen. For example, the potency ratios for morphine were 2.2 (1.5-3.1) after the chronic administration of etorphine and 1.2 (0.9-1.6) after the chronic administration of etorphine + LY235959. The potency ratios for dezocine were 3.4 (1.9-6.2) when animals received twice-daily injections of 0.003 mg/kg etorphine and 1.2 (0.5-2.9) when 3.0 mg/kg LY235959 was combined with 0.003 mg/kg etorphine.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effects of morphine and dezocine before and after repeated administration with 0.003 mg/kg etorphine alone or in combination with 3.0 mg/kg LY235959. Ordinates, %MPE. Abscissae, dose of etorphine in milligrams per kilogram, log scale. Labels above each panel indicate the chronic treatment condition.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Effects of etorphine and dezocine before and after repeated administration with 10 mg/kg morphine alone or in combination with 3.0 mg/kg LY235959. Ordinates, %MPE. Abscissae, dose of morphine in milligrams per kilogram, log scale. Labels above each panel indicate the chronic treatment condition.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Effects of etorphine and morphine before and after repeated administration with 3.0 mg/kg dezocine alone or in combination with 3.0 mg/kg LY235959. Ordinates, %MPE. Abscissae, dose of dezocine in milligrams per kilogram, log scale. Labels above each panel indicate the chronic treatment condition.

Baseline Data. Baseline tail-withdrawal latencies are presented in Table 3. Mean baseline latencies for the 18 groups of rats in the tolerance and cross-tolerance experiments (six treatment conditions × three test conditions) ranged from 3.5 (±0.5) to 6.0 (±0.2) s before chronic drug administration and from 3.3 (±0.3) to 4.8 (±0.4) s after chronic drug administration. Mean difference scores (day 8 baseline  day 1 baseline) are also presented in Table 3. A repeated-measures three-way ANOVA was performed using baseline latencies as the within-subjects variable and chronic opioid treatment (etorphine, morphine, dezocine), chronic LY235959 treatment (no LY235959, 3.0 mg/kg LY235959), and test agonist (etorphine, morphine, dezocine) as the between-subjects variables. The results of this analysis are presented in Table 4. The analysis revealed a main effect of time (F_1,111 = 9.66, P = .0024). The mean baseline latency for day 1 was 4.6 (±0.1) s and the mean baseline latency for day 8 was 4.2 (±0.1) s. Thus, there was a small but significant decrease in baseline latencies between days 1 and 8 of the study. However, there were no interaction effects between baseline latency and any of the between-subjects variables. In other words, the decrease in baseline latency from day 1 to day 8 did not differ between rats across any treatment and/or testing conditions or combination of treatments and testing conditions. Significant differences among the between subjects variables were also small and were not related to the experimental results in the tolerance and cross-tolerance experiments in any systematic way.

View this table: [in this window] [in a new window]   TABLE 3 Mean baseline latencies in seconds for the 18 groups of rats in the tolerance and cross-tolerance experiments on day 1 (Prechronic) and day 8 (Chronic) The values in parentheses represent the standard error of the mean (S.E.M.). The difference in baseline (day 8  day 1) is also presented ( BL). Rats were treated chronically with etorphine (ET), morphine (MS), or dezocine (DZ) either alone or in combination with LY.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Scatterplot of the relationship between an individual rat's change in baseline latency and that rat's tolerance ratio (day 8 ED50/day 1 ED50) for rats that received repeated injection with etorphine, morphine, and or dezocine either alone (A) or in combination with 3.0 mg/kg LY235959 (B). Presented of each scatterplot are the best-fit line (least-squares method) and the coefficient of determination (r2). Ordinates, tolerance ratio expressed as log unit shift. Abscissae, change in baseline latency in seconds.

Pharmacokinetic and Pharmacodynamic Variables. Although it was not the purpose of this study to thoroughly characterize etorphine, morphine, or dezocine according to various pharmacokinetic and pharmacodynamic variables, two experiments were performed to provide comparative data that would inform interpretation of the data related to magnitude of tolerance. Figure 8 shows the effects of 0.003 mg/kg etorphine, 3.0 mg/kg dezocine, and 10 mg/kg morphine on tail-withdrawal latencies from 55°C at 15, 30, 60, 120, 180, and 240 min. All drugs were maximally effective at 30 min. Morphine had a longer duration of action than both etorphine and dezocine. An AUC analysis was performed on the time-effect curves for individual rats. The results of this analysis are presented in Table 5. The ANOVA of the individual AUC values revealed a significant main effect of drug (F_2,21 = 8.21, P = .0023). Student-Newman-Keuls post hoc comparisons showed that the AUC for morphine was significantly greater than the AUC for etorphine or dezocine.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Time-effect curves for 0.003 mg/kg etorphine, 3.0 mg/kg dezocine, and 10 mg/kg morphine using the tail-withdrawal procedure. Ordinate, %MPE. Abscissa, time in minutes.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Dose-effect curves for etorphine (top), morphine (middle), and dezocine (bottom) from rats before () and 24 h after () a subcutaneous injection with 1.0 mg/kg C-CAM. Ordinates, %MPE. Abscissae, dose in milligrams per kilogram, log scale.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates that the competitive NMDA receptor antagonist LY235959 completely or partially attenuates the development of tolerance and the cross-tolerance conferred to opioids that produce their antinociceptive effects at the µ-opioid receptor. Prevention of tolerance was only limited by the magnitude of tolerance and cross-tolerance that developed after a chronic opioid treatment. These results support the hypothesis that the development of tolerance to µ-opioids in general is prevented by NMDA receptor antagonists, and that the prevention of morphine tolerance by NMDA receptor antagonists does not represent an idiosyncratic interaction between morphine and the glutamate system.

In this study, tolerance developed to the effects of morphine and dezocine, but not to the effects of the high-efficacy opioid etorphine. A comparison of potency ratios reveals that there were significant differences in the magnitude of tolerance that developed to etophine, morphine, and dezocine. The least tolerance (i.e., no tolerance) developed to etorphine, followed in increasing magnitude by morphine (2.9-fold shift) and dezocine (9.6-fold shift). Thus, there was an inverse relationship between the relative efficacy of the opioid and the magnitude of tolerance that developed to that opioid.

Coadministration of LY235959 with morphine and dezocine completely prevented the 2.9-fold rightward shift in the morphine dose-response curve but only partially prevented the 9.6-fold rightward shift in the dezocine dose-response curve. The etorphine dose-effect curve did not shift after chronic administration of etorphine, and no alterations in the etorphine dose-effect curve occurred when rats received 3.0 mg/kg LY235959 along with chronic etorphine injections. These data indicate that the magnitude of tolerance that develops to an opioid can limit the effectiveness of a dose of LY235959 to prevent the development of tolerance. Similar results were found with rats maintained on three different doses of morphine under similar experimental conditions (Allen and Dykstra, 2000). Potency ratios for rats maintained on 10, 20 and 40 mg/kg morphine were 2.9 (2.0-4.2), 5.9 (4.4-7.7), and 12.4 (8.9-17.7), respectively. A dose of 3.0 mg/kg LY235959 coadministered with each maintenance dose of morphine completely prevented the 2.9-fold shift produced by chronic treatment with 10 mg/kg morphine, but only partially prevented the development of tolerance induced with the higher morphine maintenance doses (Allen and Dykstra, 2000).

It is important to note that chronic treatment with LY235959 and etorphine did not result in changes in etorphine's antinociceptive potency. If administration of LY235959 prevented tolerance by increasing a rat's sensitivity to opioids (an effect that would oppose the rightward shift in an opioid dose-response curve), then chronic administration of LY235959 with etorphine should result in a leftward shift in the etorphine dose-effect curve. This would appear as a potency ratio with an upper confidence limit lower than 1.0 (e.g., 5-, 10-, and 100-fold shifts to the left would produce potency ratios of 0.2, 0.1, and 0.01, respectively). That the potency ratio for etorphine was unchanged after chronic treatment with etorphine and LY235959 further suggests that prevention of opioid tolerance by NMDA receptor antagonists is not the result of increases in opioid sensitivity after chronic opioid/NMDA antagonist treatment.

This study shows that chronic treatment with equi-effective doses of µ-opioids can produce varying degrees of tolerance, and the magnitude of tolerance is inversely related to the efficacy of the opioid. This relationship has been demonstrated with various behavioral assays (Emmett-Oglesby et al., 1988; Stevens and Yaksh, 1989; Paronis and Holtzman, 1992; Paronis and Holtzman, 1994). It suggests that high-efficacy agonists are the least likely to produce tolerance when treatments are matched for effective dose. Therefore, when low-efficacy agonists are used as toleragens, NMDA receptor antagonists may not completely prevent tolerance; however, if lower doses are administered, less tolerance may develop to low-efficacy opioids and NMDA receptor antagonists may prevent tolerance completely under these conditions.

In opioid cross-tolerance studies, the chronic opioid treatment is the same but the change in the potency of test agonists differs. High-efficacy agonists show less cross-tolerance than low-efficacy agonists given a common treatment (Craft and Dykstra, 1990; Sosnowski and Yaksh, 1990; Paronis and Holtzman, 1992; Tiano et al., 1998). In this study, cross-tolerance was measured under three different opioid treatment conditions (chronic etorphine, morphine, or dezocine, either with or without LY235959). Based on the efficacy of these agonists, we would predict less cross-tolerance to etorphine than morphine tolerance or dezocine tolerance given chronic treatment with morphine or dezocine, respectively. This is in fact what the data reveal. Conversely, the data suggest that although chronic etorphine does not produce etorphine tolerance, the opioid system is still compromised as a result of this chronic treatment because rats are cross-tolerant to morphine and dezocine.

Chronic treatment with µ-opioid agonists has been shown to produce such changes as alterations in receptor number (Yoburn et al., 1993; Díaz et al., 1995), alterations in the functional state of the receptor (Sadée and Wang, 1995; Bilsky et al., 1996b), and changes in the ratio of opioid receptor binding to inhibitory and stimulatory G-protiens (Wu et al., 1998). It is possible that all of these phenomena play some role in the reduced potency of opioid agonists to produce behavioral effects. Given that high-efficacy opioids are more resistant than low-efficacy opioids to decreases in the pool of functional receptors, any of these changes could result in the differential degree of cross-tolerance observed in this study.

The results from the cross-tolerance experiment are particularly intriguing. If NMDA receptor antagonists prevent tolerance by completely preventing the receptor-level changes that result from chronic treatment, then NMDA receptor antagonists should prevent cross-tolerance to other µ-opioid agonists, regardless of magnitude. It is possible that LY235959 partially prevents receptor-level changes, because a partial prevention of these changes could explain the differential prevention of tolerance and cross-tolerance after chronic morphine and LY235959 administration. It is interesting to note that even when tolerance to a µ-opioid agonist (such as morphine) is completely prevented, changes may still be present in the opioid system. This may prove important in the clinical application of opioid/NMDA receptor antagonist drug combinations.

Although receptor theory is an attractive basis for understanding the development of opioid tolerance, other theories have attempted to describe the loss of analgesic effects that follow chronic opioid administration. Another hypothesis that increasingly gains recognition is that tolerance to opioid analgesic effects is related to the development of hyperalgesia (Mao et al., 1994, 1995; Larcher et al., 1998; Laulin et al., 1999). Several studies show that repeated opioid treatment results in a progressive increase in sensitivity to a noxious stimulus (Larcher et al., 1998; Laulin et al., 1999). This hyperalgesic response, expressed as a decreased baseline, is prevented when animals are administered NMDA receptor antagonists along with repeated opioid administration (Mao et al., 1994).

The results from this study might be predicted based on a hyperalgesia model. In essence, hyperalgesia, or an increased sensitivity to a noxious stimulus, is functionally equivalent to an increase in stimulus intensity. Several authors have described relationships between opioid efficacy and stimulus intensity (Morgan et al., 1999a,b; Smith et al., 1999). In general, there are greater changes in the potency and effectiveness of low-efficacy opioids when stimulus intensities increase compared with high-efficacy opioids. High-efficacy opioids are less responsive to changes in stimulus intensity. Thus, if chronic treatment with an opioid produced a hyperalgesic state, it would be expected that shifts in the dose-effect curves for low-efficacy opioids would be larger when compared with high-efficacy opioids. This would appear as a differential development of cross-tolerance.

In the present study, however, changes in baseline latency after chronic drug administration were small (average difference = 1.26 s). More importantly, according to the hyperalgesia model, a decrease in latency should be associated with an increase in tolerance magnitude. Thus, there should be an inverse relationship between change in baseline and tolerance magnitude. There was no relationship between the direction of change in baseline latency and the magnitude of tolerance that developed in individual animals, further suggesting that the development of hyperalgesia is not related to the development of tolerance in these rats.

In summary, chronic treatment with maximally effective doses of three µ-opioid agonists produced various degrees of tolerance and cross-tolerance and the magnitude of tolerance and cross-tolerance produced by an opioid treatment varied inversely with the relative efficacy of both the toleragen and the test agonist. The coadministration of LY235959 with an opioid during chronic treatment completely prevented or reduced the magnitude of tolerance or cross-tolerance that developed. The effectiveness of LY235959 depended on both the magnitude of tolerance and cross-tolerance that developed after a chronic opioid treatment. These results suggest that NMDA receptor antagonists prevent the changes that occur after chronic administration of µ-opioid agonists in general and that the prevention of tolerance is a function of the magnitude of change in the opioid system.