Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Insurmountable antagonism and repeated agonist treatment are two in vivo pharmacological methods to quantify the ability of an agonist to stimulate its biological receptors, i.e., agonist efficacy. Insurmountable antagonists directly remove a portion of the receptor population from interaction with agonists, either permanently through alkylation or temporarily through long-term or pseudoirreversible receptor binding (Tallarida and Jacob, 1979; Kenakin, 1997). The mechanisms underlying the consequences of repeated agonist treatment are less clear but may involve such processes as receptor down-regulation, desensitization, internalization, and/or phosphorylation (Law et al., 2000; Taylor and Fleming, 2001). The consequences of increasing doses of insurmountable antagonists or repeated agonists are similar, a progressive loss of agonist potency and an eventual reduction in maximal agonist effect.

Many investigators have shown that higher efficacy agonists seem less sensitive to receptor removal or inactivation by an insurmountable antagonist (Zimmerman et al., 1987; Adams et al., 1990; Pitts et al., 1996; Walker et al., 1998) or repeated agonist treatment (Paronis and Holtzman, 1992; Duttaroy and Yoburn, 1995; Walker and Young, 2001) than do lower efficacy agonists. Theoretically, this relation may arise because higher efficacy agonists have a larger receptor reserve or are more efficient at stimulating receptors than are lower efficacy agonists (Stephenson, 1956; Kenakin, 1997; Selley et al., 1997). For example, insurmountable antagonists such as -funaltrexamine (-FNA), clocinnamox, or buprenorphine produce larger alterations in dose-response curves for morphine than for the higher efficacy agonists fentanyl, methadone, alfentanil, or etonitazene (Adams et al., 1990; Comer et al., 1992; Zernig et al., 1994; Walker et al., 1995). Similarly, repeated etonitazene, morphine, or buprenorphine treatments produce larger alterations in buprenorphine or GPA 1657 dose-response curves than for those of the higher efficacy agonists etonitazene, fentanyl, or etorphine (Young et al., 1991; Paronis and Holtzman, 1992; Walker and Young, 2001). Therefore, agonists demonstrated to act through similar receptor populations could be identified as high-, intermediate-, or low-efficacy agonists based on their sensitivity to insurmountable blockade or repeated agonist treatment.

Drug discrimination assays are unusual in that quite high doses of insurmountable antagonists or frequent administration of high doses of agonists are required to alter the potency for opioid generalization (Holtzman, 1997; Morgan and Picker, 1998; Young et al., 1992; but see France and Woods, 1987). In addition, reductions in maximal discriminable effects are difficult to obtain for all but the lowest efficacy agonists in drug discrimination assays. For example, doses of -FNA or clocinnamox that markedly decrease the maximal effects of morphine in antinociceptive assays in rats fail to decrease its discriminative stimulus effects (Adams et al., 1990; Walker et al., 1996; Holtzman, 1997; Walker and Young, 2001). These observations indicate that the drug discrimination assay may have low-efficacy requirements; i.e., may require elimination or disabling of a majority of receptors to decrease maximal agonist effects (Holtzman, 1997).

A strategy to remove or disable a large enough proportion of receptors to alter dose-response curves for both high- and low-efficacy agonists in the drug discrimination assay may be to combine insurmountable antagonism and repeated agonist treatment in the same subjects. This strategy has been applied in vitro using the alkylating agent -chlornaltrexamine in normal and morphine-tolerant guinea pig ileum myenteric plexus to examine the role of receptor reserve in the magnitude of tolerance. In morphine-tolerant ilea, the dose of chlornaltrexamine needed to reduce maximum effects of normorphine is 10-fold lower than that required in nontolerant ilea (Chavkin and Goldstein, 1982, 1984). Similarly, clocinnamox and fentanyl infusions produce additive reductions in fentanyl potency in vivo in mice (Chan et al., 1997).

In the present in vivo study, the effects of: 1) the insurmountable antagonist clocinnamox alone, 2) repeated morphine treatment alone, and 3) the combination of clocinnamox and repeated morphine treatment on the discriminative stimulus effects of opioids were examined in rats trained to discriminate 3.2 mg/kg morphine from saline. Although the efficacy requirements of an opioid discrimination assay are controlled by the training agonist and dose, it is well established that discriminations based on 3.2 mg/kg morphine reflect µ-agonist-mediated processes (Holtzman, 1982; Young et al., 1992; Walker et al., 1996). Furthermore, whereas doses of insurmountable antagonists or repeated morphine regimens that reduce maximal effects of µ-agonists in antinociception assays in rats (Paronis and Woods, 1997; Walker et al., 1998; Walker and Young, 2001) reduce agonist potency in this discrimination assay, they produce few changes in maximal discriminative effects (Young et al., 1991; Walker et al., 1996; Holtzman, 1997; Walker et al., 1997; Morgan and Picker, 1998). Potency estimates for antagonists of morphine's discriminative stimulus effects in rats treated with clocinnamox or repeated morphine are similar to estimates obtained in control rats, suggesting that such treatments do not alter the receptor class through which morphine exerts discriminative stimulus effects (Walker et al., 1996). Therefore, clocinnamox and repeated morphine treatment were combined in the same rats to both remove and disable a large enough proportion of µ-opioid receptors to alter the potency and maximal effect of fentanyl, morphine, buprenorphine, and nalbuphine in a drug discrimination assay.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Forty, male Sprague-Dawley rats were housed individually in a colony room maintained under a 12-h-light/dark cycle. Water was freely available in the home cage. Rats received 14 to 25 g of Purina rat chow daily to maintain body weights of approximately 320 to 370 g.

Apparatus. Experiments were performed in chambers housed in ventilated, sound-attenuating cubicles. Two stimulus lights were mounted on one wall of each chamber above two response levers positioned 7 to 8 cm above the floor. A recessed food receptacle was located between the response levers. A minimal downward weight of 28 to 35 g was recorded as a response, and food pellets (45 mg; PJ Noyes, Lancaster, NH) were delivered by a pellet dispenser mounted outside the chamber. White noise was present in the experimental room. Experimental contingencies were arranged, and data were recorded by AIM microprocessors.

Procedure. Saline and 3.2 mg/kg morphine were established as discriminative stimuli for food-reinforced responses using a multiple-trial training procedure (Young et al., 1992). Each training session was divided into three discrete trials, each consisting of a 15-min pretreatment period followed by a 5-min ratio component. Before each trial, an injection of saline or morphine (s.c.) was administered, and the rats were then placed in darkened experimental chambers. After 15 min, stimulus lights were illuminated, and food pellets were delivered under a fixed-ratio schedule of reinforcement. Left lever responses were reinforced after an injection of morphine, and right lever responses were reinforced after an injection of saline. Incorrect lever responses reset the ratio counter to 0. Response requirements on both levers were increased steadily to an fixed-ratio 15 over a number of daily training sessions. The ratio component terminated after 5 min or 50 reinforcers, whichever occurred first. After the 5-min ratio component, rats were removed from the chamber, injected with morphine or saline, and returned to the chamber as the next pretreatment period began. Saline trials in the first component were followed by two additional saline trials, one saline trial and one morphine trial, or two morphine trials (on the third trial saline was injected). Morphine trials in the first component were followed by additional morphine training trials, each of which was preceded by a saline injection. Note that throughout multiple-trial training, responding on the morphine lever was reinforced on all trials after a morphine injection.

Pharmacological Procedures. During multiple-trial, cumulative-dosing tests, saline or a dose of agonist was administered before the pretreatment period, and 15 consecutive responses on either lever produced food pellets during a 5-min ratio component. After each ratio component, rats were removed, injected with the next cumulative dose of agonist (0.25 or 0.5 log₁₀-unit increments), and placed back into the chamber for the next pretreatment period. Testing continued for four to eight successive trials or until a rat exhibited markedly suppressed response rates.

Data Analysis. Discriminative performance is presented as the percentage of responses emitted on the morphine-appropriate lever. These data were analyzed only if 15 or more responses were emitted during the ratio component. Rates of responding for the session are expressed as a percentage of the average rate of responding from the saline training session immediately before the test or the tolerance regimen. All dose-response curves for individual rats were fitted using the following semilogarithmic form of the logistic dose-response equation:

(1)

where E is either percentage of morphine responses or percentage of control rate of responding, and E_min and E_max are the minimum and maximum of the sigmoid dose-response curve. E_min and E_max were generally set at 0 and 100 for discriminative effects, respectively, unless a compound failed to produce 100% morphine responding. In these cases, E_max was not constrained to 100%. X is the dose of agonist (in milligrams per kilogram) at a particular effect level. ED₅₀ is the agonist dose producing 50% morphine responses or 50% control rate of responding, and h is a slope factor. For each test, individual log ED₅₀ values were averaged for the group and a 95% CL calculated with the t statistic. To compare potency changes within groups, individual log ED₅₀ values and slopes were analyzed by separate one-way repeated measures analyses of variance. To compare potency changes across groups, a potency ratio was calculated for each rat by dividing the ED₅₀ after treatment by the ED₅₀ determined before treatment. Log potency ratios were analyzed by one- or two-way analysis of variance, as appropriate. Tukey's multiple comparison test was used for post-hoc analyses. Significance was set at P < 0.05.

Drugs. The following compounds were used: morphine sulfate, buprenorphine hydrochloride, fentanyl hydrochloride (National Institute on Drug Abuse, Rockville, MD), nalbuphine hydrochloride (Research Biochemicals, Inc., Natick, MA), and clocinnamox (gift from John W. Lewis, University of Bristol, Bristol, UK). Morphine, fentanyl, and nalbuphine were dissolved in physiological saline. Buprenorphine and clocinnamox were dissolved in sterile water. Solutions were prepared to administer each injection in a volume of 0.1 to 3.0 ml/100 g body weight. Doses are expressed as the forms listed above. Saline was injected in a volume of 1 ml/kg body weight.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Clocinnamox Antagonism of Morphine. In nontreated rats, morphine produced dose-dependent increases in percentage of morphine responding (Fig. 1, upper panels), with control ED₅₀ values ranging from 0.56 to 0.99 mg/kg (Table 1). In addition, morphine produced dose-dependent decreases in rate of responding (Fig. 1, bottom panels, with control ED₅₀ values ranging from 2.3 to 3.3 mg/kg (Table 1). Twenty-four-hour pretreatment with 10 mg/kg clocinnamox increased the ED₅₀ value for morphine to produce stimulus or rate-decreasing effects by 6.5-fold (Fig. 1, left panels).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Clocinnamox antagonism of the stimulus and rate-decreasing effects of morphine in normal (left panels) and morphine-tolerant rats (center and right panels) trained to discriminate 3.2 mg/kg morphine from saline. Ordinate, upper panels, percentage of total responses on the morphine-appropriate lever. Ordinate, lower panels, response rate expressed as a percentage of response rates from the saline training day before testing or repeated morphine treatment. Saline control values ranged from 0.34 to 1.4 responses/s. Data from rats making fewer than 15 total responses were included in the response-rate panels but not the discrimination panels. Abscissae, cumulative doses of morphine in milligrams per kilogram. Points above S represent the effects of a saline injection administered before the determination of the morphine dose-response curves. Left panels, effects of 24-h pretreatment of 10 mg/kg clocinnamox in normal rats (n = 5). Center panels, effects of 20 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 3.2 mg/kg clocinnamox (n = 7). Right panels, effects of 20 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 10 mg/kg clocinnamox (n = 6). In all panels, open circles represent the effects of morphine alone in two tests conducted before clocinnamox or repeated morphine treatment. All dose-response curves after repeated morphine treatment were determined 36 h after the last morphine injection on day 6. Vertical lines represent S.E.M.

Clocinnamox Antagonism of Fentanyl. In nontreated rats, fentanyl produced dose-dependent increases in percentage of morphine responding (Fig. 2, upper panels) with control ED₅₀ values of from 0.0054 or 0.0063 mg/kg (Table 2). In addition, fentanyl produced dose-dependent decreases in rate of responding (Fig. 2, bottom panels, with control ED₅₀ values of 0.019 or 0.020 mg/kg (Table 2). Twenty-four-hour pretreatment with 10 mg/kg clocinnamox increased the ED₅₀ value for fentanyl to produce stimulus or rate-decreasing effects by 2.7- or 9.7-fold, respectively (Fig. 2, left panels). Three rats suffered seizures and were euthanized within 2 days after clocinnamox treatment; the remaining five rats recovered initial sensitivity to fentanyl's stimulus but not rate-decreasing effects within 2 weeks.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Clocinnamox antagonism of the stimulus and rate-decreasing effects of fentanyl in normal (left panels) and morphine-tolerant rats (right panels) trained to discriminate 3.2 mg/kg morphine from saline. Ordinate, lower panels, saline control values ranged from 0.37 to 1.7 responses/s. Abscissae, cumulative doses of fentanyl in milligrams per kilogram. Points above S represent the effects of a saline injection administered before the determination of the fentanyl dose-response curves. Left panels, effects of 24-h pretreatment of 10 mg/kg clocinnamox in normal rats (n = 8). Right panels, effects of 20 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 10 mg/kg clocinnamox (n = 9). In all panels, open circles represent the effects of fentanyl alone. Other details as in Fig. 1.

Clocinnamox Antagonism of Buprenorphine. In nontreated rats, buprenorphine produced dose-dependent increases in percentage of morphine responding (Fig. 3, upper panels) with control ED₅₀ values of 0.0091 or 0.015 mg/kg (Table 3). In addition, buprenorphine produced dose-dependent decreases in rate of responding (Fig. 3, lower panels) with control ED₅₀ values of 0.048 mg/kg. Twenty-four-hour pretreatment with 10 mg/kg clocinnamox increased the ED₅₀ value for buprenorphine to produce stimulus or rate-decreasing effects by 6.1- or 41-fold, respectively.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Clocinnamox antagonism of the stimulus and rate-decreasing effects of buprenorphine in normal (left panels) and morphine-tolerant rats (right panels) trained to discriminate 3.2 mg/kg morphine from saline. Ordinate, lower panels, saline control values ranged from 0.54 to 1.8 responses/s. Abscissae, cumulative doses of buprenorphine in milligrams per kilogram. Points above S represent the effects of a saline injection administered before the determination of the buprenorphine dose-response curves. Left panels, effects of 24-h pretreatment of 10 mg/kg clocinnamox in normal rats (n = 6). Right panels, effects of 20 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 10 mg/kg clocinnamox (n = 4). In all panels, open circles represent the effects of buprenorphine alone. Other details as in Fig. 1.

Clocinnamox Antagonism of Nalbuphine. In nontreated rats, nalbuphine produced dose-dependent increases in percentage of morphine responding (Fig. 4, upper panels) with control ED₅₀ values ranging from 0.38 to 1.2 mg/kg (Table 4). In addition, nalbuphine produced dose-dependent decreases in rate of responding (Fig. 4, bottom panels) with control ED₅₀ values ranging from 17 to 19 mg/kg. Twenty-four-hour pretreatment with 10 mg/kg clocinnamox increased the ED₅₀ value for nalbuphine to produce stimulus effects by 8.1-fold. The ED₅₀ value for nalbuphine to produce rate-decreasing effects did not seem to be altered after 10 mg/kg clocinnamox pretreatment (Fig. 4, left panels).

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Clocinnamox antagonism of the stimulus and rate-decreasing effects of nalbuphine in normal (left panels) and morphine-tolerant rats (center and right panels) trained to discriminate 3.2 mg/kg morphine from saline. Ordinate, lower panels, saline control values ranged from 0.40 to 1.8 responses/s. Abscissae, cumulative doses of nalbuphine in milligrams per kilogram. Points above S represent the effects of a saline injection administered before the determination of the nalbuphine dose-response curves. Left panels, effects of 24-h pretreatment of 10 mg/kg clocinnamox in normal rats (n = 6). Center panels, effects of 10 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 3.2 mg/kg clocinnamox (n = 9). Right panels, effects of 10 mg/kg per day morphine treatment for 6 days alone or 12 days with a 24-h pretreatment of 10 mg/kg clocinnamox (n = 6). In all panels, open circles represent the effects of nalbuphine alone. Other details as in Fig. 1.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The rank order of relative efficacy for four µ-agonists in this drug discrimination assay in rats as determined by sensitivity to a 24-h pretreatment with the insurmountable antagonist clocinnamox was fentanyl > morphine  buprenorphine > nalbuphine. This rank order is consistent with previous data obtained with the irreversible antagonist -FNA in the drug discrimination assay (Holtzman, 1997; Picker, 1997). For example, in pigeons trained to discriminate morphine from saline, -FNA produced 0.6-, 4.5-, 5.0-, or 17-fold changes in the ED₅₀ values for fentanyl, morphine, buprenorphine, or nalbuphine to produce morphine-like stimulus effects, respectively (Morgan and Picker, 1998). In addition, treatment with the insurmountable antagonists clocinnamox or -FNA yields the same rank order of relative efficacy for this series of µ-agonists in antinociception assays (Zimmerman et al., 1987; Adams et al., 1990; Comer et al., 1992; Pitts et al., 1996; Tiano et al., 1998).

In contrast to results in antinociception assays, however, differences in discriminative potency after clocinnamox were not accompanied by differences in maximal discriminative stimulus effects. As has been reported for other insurmountable antagonists in drug discrimination assays (Holtzman, 1997; Picker, 1997; Morgan and Picker, 1998; but see France and Woods, 1987) clocinnamox produced parallel shifts in the dose-response curves for all µ-agonists, including the low-efficacy agonist nalbuphine, without altering slopes or maximal discriminative stimulus effects. This observation probably reflects the low-efficacy requirement of the discrimination assay relative to other test systems such as antinociception (Walker et al., 1996; Holtzman, 1997). For example, 10 mg/kg clocinnamox reduces the maximal effects of morphine below 50% in antinociception assays in rats (Paronis and Woods, 1997; Walker et al., 1998) yet failed to alter the maximum effects for percentage of morphine responding in the present drug discrimination assay (see also Walker et al., 1996).

In vitro and ex vivo radioligand binding experiments indicate that 24-h pretreatment with 10 mg/kg clocinnamox reduces the B_max values for µ-opioid receptors by 70 to 95% of control values in rats (Paronis and Woods, 1997) and mice (Burke et al., 1994; Zernig et al., 1995, 1996; Chan et al., 1995). A dose of 10 mg/kg clocinnamox yields in vivo estimates of q, the fraction of receptors remaining after inactivation by an insurmountable antagonist (Furchgott, 1966), of approximately 0.10 in the rat warm-water tail-withdrawal assay (Walker et al., 1998). This in vivo estimate indicates that approximately 90% of the µ-opioid receptors are removed by 24-h pretreatment with 10 mg/kg clocinnamox in rats. The evidence that approximately 70 to 95% of µ-opioid receptors can be eliminated from interaction with agonists by an insurmountable antagonist, yet an agonist can still produce full discriminative stimulus effects, indicates that drug discrimination can be a quite sensitive assay in which only limited agonist efficacy is needed for full agonist effect.

One method to increase the efficacy requirement of a test system, so that only high-efficacy agonists can produce maximal effects, is to treat repeatedly with an agonist (Kenakin, 1997). Repeated treatment with an agonist could decrease the stimulus-response capacity of the test system or reduce receptor density by mechanisms as diverse as receptor down-regulation, desensitization, internalization, and/or phosphorylation (Kenakin, 1997; Law et al., 2000; Taylor and Fleming, 2001). In vivo, repeated high-dose agonist treatment can suppress the maximal antinociceptive effects produced by both morphine (Fernandes et al., 1977; Blasig et al., 1979) and lower efficacy agonists such as dezocine and buprenorphine (Tiano et al., 1998; Walker and Young, 2001). In the present study, repeated treatment with 20 mg/kg per day morphine produced parallel shifts in the fentanyl, morphine, and buprenorphine dose-response curves without any apparent decrease in maximal discriminative effects, again indicating that the drug discrimination assay is a test system with low-efficacy requirements. In a previous study, however, repeated treatment with 20 mg/kg per day for 7 days decreased the maximal percentage of morphine responding produced by low-efficacy agonist nalbuphine but not that produced by buprenorphine or morphine in rats (Young et al., 1991). Thus, treatment with 20 mg/kg per day morphine seems to alter stimulus-response capacity sufficiently to decrease maximal effects for the lowest efficacy agonist, nalbuphine, but not for the higher efficacy agonists fentanyl, morphine, or buprenorphine.

The combination of clocinnamox and repeated morphine treatment produced a further antagonism of the percentage of morphine responding produced by all four µ-agonists, coupled with a markedly diminished maximal effect for buprenorphine and nalbuphine. The effects of clocinnamox in morphine-treated rats were disproportionately larger for the lower efficacy agonists buprenorphine and nalbuphine than for the higher efficacy agonists fentanyl and morphine. For example, 10 mg/kg clocinnamox alone produced a parallel, 6-fold change in the ED₅₀ value for buprenorphine in normal rats. However, in morphine-treated rats, 10 mg/kg clocinnamox decreased to less than 30% the maximal percentage of morphine responding achieved by buprenorphine, so that even doses 3500 times greater than the ED₅₀ value for buprenorphine alone could not produce significant discriminative stimulus effects.

In all cases, the effects of clocinnamox in morphine-treated rats were greater than those in normal rats and preserved the apparent differences in efficacy among the agonists. In morphine-treated rats, a dose of 10 mg/kg clocinnamox produced a parallel, rightward shift in the fentanyl dose-response curve, a large, nonparallel shift in the morphine-dose-response curve, and a large shift in the dose-response curve and reduction in the maximal effect for buprenorphine and nalbuphine. The rank order of relative efficacy for fentanyl, morphine, buprenorphine, and nalbuphine was preserved despite the large reduction in stimulus-response capacity of the drug discrimination assay produced by combined insurmountable antagonism and repeated agonist treatment. These observations are consistent with the following propositions: 1) the combined treatments had produced a maximum rightward shift; and 2) further decreases in receptor number or stimulus-response efficiency (i.e., higher doses of clocinnamox or higher, more frequent doses of morphine treatment) would produce only decreases in maximal effect without further reductions in potency.

The consequences of clocinnamox alone, repeated morphine treatment alone, and the combination of these two treatments on the rate-decreasing effects of fentanyl, morphine, and buprenorphine were generally similar to the effects of these treatments on discriminative stimulus effects. For example, larger alterations in potency to produce rate-decreasing effects were observed for buprenorphine than for fentanyl when morphine-treated rats were administered clocinnamox. Interestingly, however, the rate-decreasing effects of nalbuphine were not obviously altered by any treatment in normal or morphine-treated rats. This outcome is consistent with considerable evidence that the rate-decreasing effects of high doses of nalbuphine (Walker and Young, 1993; Walker et al., 1997), dezocine (Picker, 1997), meperidine (Leander and McMillan, 1977; Izenwasser et al., 1996), or butorphanol (Picker et al., 1996; Pitts et al., 1996) are not mediated simply through µ-opioid receptors.

The magnitudes of tolerance and cross-tolerance to the morphine-like discriminative stimulus effects of fentanyl, morphine, and buprenorphine were similar to those reported previously under slightly different procedures (Young et al., 1990, 1991; Walker et al., 1997). For nalbuphine, however, the minimal impact of repeated treatment with the lower dose of 10 mg/kg per day morphine, in tests 36 h after the last morphine injection, differs from the small degree of tolerance to nalbuphine reported in tests 12 h after the last morphine injection (Young et al., 1991). Although the nalbuphine dose-response curves in the present study failed to reveal any overt changes after repeated treatment with 10 mg/kg per day morphine, this treatment regimen clearly altered the sensitivity of the µ-opioid receptor population because clocinnamox produced greater changes in nalbuphine sensitivity in morphine-treated than in nontreated rats. Therefore, in a dynamic µ-opioid receptor system recovering from repeated low-dose morphine treatment, alterations in agonist potency may be revealed only if the system is pharmacologically challenged, as occurred with the insurmountable antagonist clocinnamox.

It is unlikely that the greater effects of clocinnamox in morphine-treated rats were the result of opioid abstinence. Mild weight loss but no other signs of withdrawal were observed in the morphine-treated rats receiving either saline or clocinnamox. This lack of abstinence in the present study probably reflects the moderate morphine treatment regimen rather than an inability of insurmountable antagonists to provoke abstinence. For example, in highly dependent rhesus monkeys, -FNA precipitates a severe abstinence syndrome that lasts 72 h and is not suppressed by injections of morphine (Gmerek and Woods, 1985).

The systematic, predictable changes observed in the fentanyl, morphine, buprenorphine, and nalbuphine dose-response curves suggest that insurmountable antagonism and repeated agonist treatment do not alter the receptor subtypes through which agonists act, but instead reduce receptor number and/or stimulus-response capacity of the receptor complex. The rank order of agonist potency changes after clocinnamox alone and of agonist potency changes and reductions in maximum effects after clocinnamox in morphine-treated rats was directly and consistently related to the efficacy of the opioid agonist being tested. These data suggest that, with heroic efforts, the µ-opioid receptor population can be challenged sufficiently to alter the slope and potency of agonist dose-response curves even in an assay as sensitive as drug discrimination.