Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Butorphanol and nalbuphine are used clinically as analgesics with presumed decreased abuse liability due to their low efficacy at µ-opioid receptor sites. Interestingly, these compounds have substantial affinity for, and effects through, -opioid receptor sites, results that are reasonably well documented in the rodent literature (Leander, 1983; Pick et al., 1992; Jaw et al., 1993) but not unequivocally documented in the primate and human literature. In monkeys, butorphanol and nalbuphine demonstrated respiratory-depressant, antinociceptive, discriminative stimulus and reinforcing properties consistent with their identification as agonists with intermediate efficacy at µ-opioid receptors (Young et al., 1984; Gerak et al., 1994; Butelman et al., 1995; Zernig et al., 1997). Alternatively, butorphanol did not produce diuresis and generalized to nalbuphine in assays designed to more adequately reveal -opioid activity (Butelman et al., 1995; Gerak and France, 1996). Furthermore, the µ-opioid receptor agonists fentanyl, morphine, and methadone, but not the -opioid receptor agonists ethylketocyclazocine methanesulfonate (EKC), U50488 [(trans)-3,4-dichloro-N-methyl-N-[2-(1-pyrrolindinyl)-cyclohexyl]benzeneacetamide], enadoline, and spiradoline, generalized to the nalbuphine discriminative stimulus (Gerak and France, 1996).

In humans, butorphanol and nalbuphine produce complex effects. In methadone-maintained subjects, discriminating among the µ-opioid receptor agonist hydromorphone, the opioid receptor antagonist naltrexone and saline, butorphanol, and nalbuphine engendered naltrexone responses, consistent with a µ antagonist or low-efficacy µ agonist activity (Preston et al., 1990). In postaddicted subjects, discriminating among hydromorphone, pentazocine and saline, butorphanol engendered pentazocine responses, perhaps consistent with a  mechanism of action. However, subjects also identified these latter drugs as belonging to the barbiturate/benzodiazepine class (Preston et al., 1989), and the barbiturate secobarbital and the benzodiazepine receptor agonist lorazepam engendered partial pentazocine-appropriate responding in a subsequent study (Bickel et al., 1989). Although butorphanol, nalbuphine, and pentazocine share similar discriminative stimulus properties, the pharmacological specificity of these discriminative stimulus effects is not clear.

Behavioral results involving rodents have demonstrated predominantly µ, but also , activity for butorphanol and nalbuphine. Antinociceptive effects of butorphanol and nalbuphine were reversed with naltrexone; subsequently, apparent pA₂ values indicative of an interaction at the µ receptor were determined (Walker et al., 1994; Garner et al., 1997). Additional evidence for µ activity included the conditioning of a place preference by, and the self-administration of, butorphanol (Steinfels et al., 1982; Mamoon et al., 1995); finally, nalbuphine engendered morphine lever selection in rats trained to discriminate morphine from saline (Walker et al., 1997). Support for -opioid activity included the demonstration of the diuretic effects of acute butorphanol (Leander 1983) and a withdrawal syndrome precipitated by the -opioid receptor antagonist norbinaltorphimine after chronic butorphanol (Feng et al., 1997), and antinociceptive effects of acute nalbuphine were reversed with norbinaltorphimine (Pick et al., 1992).

One explanation for the lack of reliable in vivo -like activity after butorphanol and nalbuphine administration is their relative affinities for, and efficacies at, µ and -opioid receptors. Both compounds demonstrate substantial affinity for, and low efficacy at, µ and -opioid receptors (Emmerson et al., 1996; Zhu et al., 1997), yet their relative affinities and efficacies marginally favor binding and activity at µ over  receptors, resulting in in vivo observations consistent with a µ mechanism of action. In the present investigation, µ-opioid receptor-mediated effects were blocked through the use of the insurmountable µ-opioid receptor antagonist clocinnamox (Butelman et al., 1996; Zernig et al., 1996) in an attempt to demonstrate the -like activity of opioids with affinity for multiple receptors. The objective of present investigation was to characterize the effects of butorphanol and nalbuphine in the presence of clocinnamox in rhesus monkeys using two assays sensitive to -opioid receptor agonist activity: EKC discrimination and diuresis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects

Eight male and four female adult rhesus monkeys (Macaca mulatta) with complex experimental histories (including exposure to opioids and other behaviorally active drugs), weighing 6 to 11 kg, were individually housed in a vivarium maintained at 21 ± 1°C with 40% to 60% humidity and with a 12-h light/dark cycle. Monkeys in discrimination experiments were maintained at 90% free-feeding weight. All monkeys were fed 10 to 25 biscuits (Purina Monkey Chow) and fresh fruit two or three times weekly. Water was available ad libitum, except during diuresis test sessions, which took place between 10:00 AM and 1:00 PM. [Animals used in these studies were maintained in accordance with the University of Michigan Committee on Animal Care and "Guidelines of the Committee on the Care and Use of Laboratory Animal Resources" (National Health Council, Department of Health, Education and Welfare, ISBN 0-309-05377-3, revised 1996).]

Apparatus and Procedure

EKC Discrimination. All experiments used similar operant panels consisting of two primate response levers (model PRL-001; BRS-LVE, Laurel, MD), a panel of 7.5-W stimulus lights located above, and a food receptacle located between the levers, mounted on one wall of a testing chamber. The delivery of 300-mg banana-flavored food pellets (formula G/T: Noyes, Lancaster, NH) was controlled by an externally mounted food dispenser (model G5210; Gerbrands, Arlington, MA). A PC-compatible computer connected to an interface controlled the scheduling of events and recorded data. A lever press defined a response (n = 4; minimum 3/drug).

Diuresis. Accumulated urine (measured in milliliters) was collected from pans in the monkey's home cage 3 h after drug administration. Test sessions with single doses of agonists were conducted two times per week (typically Tuesday and Friday), and saline sessions were conducted two times per week (typically Monday and Thursday). In antagonist experiments, quadazocine was administered 30 min before agonist administration. In clocinnamox experiments, agonists were tested at 1, 24, and 72 h and 1 and 2 weeks after single doses of clocinnamox were administered (n = 8; minimum 4/drug).

Drugs

Butorphanol tartate (0.001-0.32 mg/kg; Bristol-Myers Squibb, Wallingford, CT), EKC (0.0001-0.32 mg/kg; Sanofi-Winthrop, Malvern, PA), fentanyl HCl (0.0001-0.01 mg/kg; National Institute on Drug Abuse, Bethesda, MD), and quadazocine methanesulfonate (0.1, 0.32, and 1 mg/kg; Sanofi-Winthrop, Malvern, PA) were dissolved in sterile water. Clocinnamox mesylate (0.1 mg/kg; J. W. Lewis, Bristol University, Bristol, UK), nalbuphine HCl (0.01-32 mg/kg; Dupont Merck, Wilmington, DE), and U69593 (+)-(5,7,8)-N-methyl-N-[7-(1-pyrrolidinyl)-1-oxaspiro[4,5]dec-8-yl]-benzeneacetamide, 0.0001-0.1 mg/kg; Pharmacia-Upjohn, Kalamazoo, MI) were dissolved in sterile water with a few drops of lactic acid. All drugs were administered i.m. in a volume of 1 ml/10 kg b.wt.

Data Analysis

The percentage of EKC lever selection [(EKC lever presses/total lever presses)*100] was determined for each subject; mean EKC lever selection below 20% and above 80% was designated as full saline and EKC generalization, respectively. EKC lever selection between 20% and 80% was designated as partial EKC generalization. Response rate and urine volume were converted to percentage of control [(drug response rate or urine volume/baseline response rate or urine volume)*100] for each subject; mean agonist data deviating from baseline control by more than 2 S.D. were accepted as significantly different. In addition, ED₅₀ values (i.e., the dose at which a 50% effect was observed) and 95% confidence intervals for EKC lever selection and response rate were calculated from first order regression equations for each subject, and nonoverlapping confidence intervals were accepted as significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Figure 1 depicts the effects of various µ- and -opioid receptor agonists in EKC discrimination (A) and diuresis (B) in rhesus monkeys. EKC training drug stimuli engendered drug-appropriate lever selection (saline, 1.6 ± 1% EKC lever selection; EKC, 97.2 ± 1.5% EKC lever selection; mean ± S.E.M.) with a response rate of 1.9 ± 0.1 responses/s, and baseline urine output was 52.1 ± 6.0 ml. The  receptor agonists EKC and U69593 produced and the µ and nonselective µ/-opioid receptor agonists fentanyl and butorphanol were without EKC-like discriminative stimulus and diuretic effects. The nonselective µ/-opioid agonist nalbuphine produced variable between- and within-subject results.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   EKC discriminative stimulus (A) and diuretic (B) effects of µ- and -opioid receptor agonists in rhesus monkeys. Each value represents the mean, and error bars represent 1 S.E.M. For discrimination, full EKC generalization is demonstrated above 80% EKC lever selection. For diuresis, horizontal dashed bars indicate 2 S.D. from control (saline administration) diuresis.

EKC. Table 1 depicts estimations of the ED₅₀ value for the discriminative stimulus and rate-suppressive effects of EKC, as well as the other tested compounds, in rhesus monkeys. EKC generalized to the EKC discriminative stimulus, and full generalization was observed at the 0.0032 mg/kg training dose (Fig. 2A and Table 1). Quadazocine (0.32 and 1 mg/kg i.m.) dose-dependently decreased the discriminative stimulus effects of EKC; at the highest quadazocine dose tested, full generalization was observed at 0.1 mg/kg. Pretreatment with the insurmountable µ antagonists clocinnamox (0.1 mg/kg i.m.) did not alter the discriminative stimulus effects of EKC at any of the time points evaluated.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   EKC discriminative stimulus (A) and diuretic (B) effects of EKC. Each value represents the mean, and error bars represent 1 S.E.M. For discrimination, full EKC generalization is demonstrated above 80% EKC lever selection. For diuresis, horizontal dashed bars indicate 2 S.D. from control (saline administration) diuresis.

U69593. U69593 generalized to the EKC discriminative stimulus; full generalization was observed at 0.01 mg/kg (Fig. 1A and Table 1). Quadazocine (0.32 mg/kg) modestly decreased (N.S.) the discriminative stimulus effects of U69593; a greater antagonism was demonstrated with the highest dose of quadazocine (1 mg/kg) but was characterized in only one subject as this higher dose became rate suppressive as the experiment progressed. Pretreatment with clocinnamox did not alter the discriminative stimulus effects of U69593 at any of the time points evaluated.

Butorphanol. Butorphanol failed to generalize to the EKC discriminative stimulus when administered alone and in the presence of quadazocine (0.1 and 0.32 mg/kg); a maximal 30% EKC lever selection was observed in one monkey at 0.032 mg/kg (Fig. 3A and Table 1). At 24 h after the administration of clocinnamox, butorphanol (0.1 mg/kg) fully generalized to the EKC discriminative stimulus in all monkeys, and butorphanol continued to engender EKC lever selection at 72 h in two of three monkeys with a reduced potency. Preclocinnamox responding returned at 1 week (i.e., determination of a butorphanol dose-effect curve revealed saline responding through doses at which behavior was suppressed).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   EKC discriminative stimulus (A) and diuretic (B) effects of butorphanol. Each value represents the mean, and error bars represent 1 S.E.M. For discrimination, full generalization is demonstrated above 80% EKC lever selection. For diuresis, horizontal dashed bars indicate 2 S.D. from control (saline administration) diuresis.

Nalbuphine. Nalbuphine produced variable results in each of the dependent measures. Nalbuphine fully generalized to the EKC discriminative stimulus in two of three monkeys, and EKC lever selection was abolished in the presence of quadazocine (0.1 and 0.32 mg/kg; Fig. 4A and Table 1). Nalbuphine produced different effects in each of the monkeys after clocinnamox administration. At 24 h, 1) nalbuphine generalized to the EKC discriminative stimulus in one monkey that had previously demonstrated a nalbuphine-EKC substitution, 2) nalbuphine produced 50% EKC lever selection in a second monkey that had previously demonstrated a nalbuphine-EKC substitution, and 3) nalbuphine engendered saline responding in a third monkey that had not previously demonstrated a nalbuphine-EKC substitution. Nalbuphine engendered predominantly saline responding at 72 h and 1 week after clocinnamox administration in all monkeys.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   EKC discriminative stimulus (A) and diuretic (B) effects of nalbuphine. Each value represents the mean, and error bars represent 1 S.E.M. For discrimination, full generalization is demonstrated above 80% EKC lever selection. For diuresis, horizontal dashed bars indicate 2 S.D. from control (saline administration) diuresis.

Fentanyl. Fentanyl failed to produce EKC lever selection (Fig. 1A) and dose-dependently decreased response rate (data not shown) and urine output (Fig. 1B).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The primary objective of the present study was to characterize the -like opioid receptor effects of the nonselective µ/-opioid agonists butorphanol and nalbuphine. When administered alone or in the presence of the competitive opioid antagonist quadazocine, butorphanol and nalbuphine did not produce (or did not produce consistent) EKC discriminative stimulus or diuretic effects. Conversely, butorphanol generalized to the EKC discriminative stimulus and produced diuresis after the administration of the insurmountable µ-opioid receptor antagonist clocinnamox, effects that are consistent with -opioid receptor activity. Nalbuphine continued to produce inconsistent results in the presence of clocinnamox. Taken together, these results demonstrate substantial -like opioid activity of butorphanol and demonstrate the use of insurmountable antagonists such as clocinnamox in the in vivo investigation of nonselective receptor agonists.

In nonhuman primates, butorphanol has been characterized as a low-efficacy µ-opioid receptor agonist (Butelman et al., 1995; Liguori et al., 1996), consistent with in vitro evidence of its affinity for, and low efficacy at, µ-opioid receptors (K_iµ = 0.5 nM; 12% maximal effect compared with fentanyl; Emmerson et al., 1996; Butelman et al., 1998). These behavioral effects are demonstrated despite the fact that butorphanol has equal affinity for, and low efficacy at, -opioid receptors (K_i = 0.68 nM; 27% maximal effect compared with the full  agonist U50488; Butelman et al., 1998; Remmers et al., 1999). When administered alone or in the presence of the competitive opioid antagonist quadazocine, butorphanol produced few discriminative stimulus or diuretic effects that were consistent with -opioid receptor activity. Nevertheless, after blocking the potential µ-opioid receptor-mediated effects with clocinnamox, -like opioid effects of butorphanol were demonstrated. These -opioid effects were dose and time related, with peak discriminative stimulus and diuretic effects observed at 24 and 72 h after clocinnamox administration, respectively. Additionally, the diuretic effects of butorphanol in the presence of clocinnamox were less than those observed with the traditional and full -opioid receptor agonist U69593; these former effects are consistent with the identification of butorphanol as a partial -opioid receptor agonist (Leander, 1983; Remmers et al., 1999).

Given the -like opioid receptor effects of butorphanol reported above and the similar effects of nalbuphine and butorphanol in antinociception, discrimination, and respiration assays, the discriminative stimulus and diuresis results obtained for nalbuphine in the presence of clocinnamox might be considered discouraging. Nalbuphine produced EKC discriminative stimulus effects in two of three monkeys, but the generalization was not improved after the administration of clocinnamox, and it is likely that EKC lever selection was due to other common stimulus properties of nalbuphine and EKC (i.e., sedation, dysphoria). In humans, complex and equivocal behavioral effects of nalbuphine have been reported, including its characterization as a low-efficacy µ-opioid receptor agonist (or µ antagonist) and a -opioid receptor agonist, as well as the identification of nalbuphine as a barbiturate/benzodiazepine-like agent (Preston et al., 1990).

Furthermore, nalbuphine alone and in the presence of clocinnamox did not produce reliable diuretic effects in the present experiment: neither dose- or time-related effects were demonstrated. Interestingly, although the affinities for nalbuphine are similar for µ- and -opioid receptors (K_iµ = 1.4 nM; K_i = 8.5 nM; Butelman et al., 1998), the reported efficacies are at variance and dependent on the cell line on which the receptor is expressed. In C₆ cells, the relative efficacy of nalbuphine is 22% (EC₅₀ = 5-8 nM) of that seen with U50488 (Remmers et al., 1999). In contrast, nalbuphine shows almost no efficacy in Chinese hamster ovary cells expressing the  receptor. Indeed, although a 38% maximal effect compared with U50488 was reported (i.e., the former being designated as a low-efficacy  agonist; Zhu et al., 1997), this estimate was obtained at very high nalbuphine concentrations (30 µM) and is likely to be a nonspecific effect. Taken together, the failure to demonstrate unequivocal EKC lever selection and diuretic effects agrees with the finding at  cloned opioid receptors expressed in Chinese hamster ovary cells and thus are likely due to the minimal efficacy of nalbuphine at -opioid receptor sites.

In vitro, EKC has substantial affinity for µ- and -opioid receptors (K_iµ = 1.7 nM; K_i = 1.2 nM; Butelman et al., 1998) and high-efficacy at -opioid receptors (100% maximal stimulation compared with U50488; Remmers et al., 1999). These in vitro findings parallel in vivo demonstrations of respiratory depression through µ-opioid receptor sites (Butelman et al., 1993) and discriminative and diuretic effects through -opioid receptor sites (Dykstra et al., 1987). Although the -opioid receptor activity in EKC discrimination and diuresis is well documented, it remained possible that these EKC effects were compromised due to its activity at the µ-opioid receptor: decreased potency or efficacy might be revealed through a leftward shift in the dose-effect function (discrimination and diuresis) or a larger maximal effect (diuresis) in the presence of clocinnamox. Because clocinnamox pretreatment failed to alter the discriminative stimulus and diuretic effects of EKC, there was no evidence for µ-opioid involvement in these effects.

Although -like opioid discriminative stimulus and diuretic effects were demonstrated in the presence of the insurmountable antagonist clocinnamox, it is puzzling why the use of the competitive antagonist quadazocine, at doses that targeted µ (0.1 mg/kg) and some of the  (0.32 and 1 mg/kg) population of opioid receptors (Negus et al., 1993), did not also eliminate the µ receptor population through which butorphanol exerted its action. This failure to demonstrate -like opioid receptor activity was not due to an insufficient quadazocine dose as the potency of butorphanol to suppress responding was decreased. Alternatively, it is likely that quadazocine was attenuating (i.e., shifting to the right) the µ- and -opioid effects of butorphanol, whereas clocinnamox was selectively abolishing the µ-opioid effects of butorphanol.

In conclusion, previous primate research characterized butorphanol and nalbuphine as low-efficacy µ-opioid receptor agonists, and the present investigation demonstrated an in vivo -like opioid activity of butorphanol in two assays sensitive to -opioid receptor activity: EKC discrimination and diuresis. Furthermore, these results support the use of insurmountable antagonists such as clocinnamox as tools to investigate the pharmacology of nonselective opioid agonists.