Psychopharmacology (2003) 167:203–210 DOI 10.1007/s00213-002-1367-9

ORIGINAL INVESTIGATION

David A. White · Stephen G. Holtzman Discriminative stimulus effects of acute morphine followed by naltrexone in the squirrel monkey

Received: 8 September 2002 / Accepted: 27 November 2002 / Published online: 18 March 2003 Springer-Verlag 2003

Abstract Rationale The discriminative stimulus effects Keywords Acute dependence · Conditioned behavior · of a combination of acute morphine followed by naltrex- Dextrorphan · Drug discrimination · Levorphanol · one have been described in rats. Objective: The purpose Naloxone · Non-human primate · Opioid · U69,593 of this study was to extend observations to a non-human primate. Methods: Eight squirrel monkeys were trained in a discrete-trial avoidance/escape procedure to discrimi- Introduction nate morphine (1.7 mg/kg, IM, 4 h) followed by naltrexone (0.1 mg/kg, IM, 0.25 h) (MOR!NTX) versus In physically dependent subjects, abstinence or opioid- saline (1.0 ml/kg, IM, 4 h) followed by naltrexone antagonist administration following chronic treatment (0.1 mg/kg, IM, 0.25 h) (SAL!NTX). Results: Seven with morphine-like drugs results in a number of promi- subjects acquired the discrimination in an average of nent physiological and behavioral changes in addition to 108€14 sessions. MOR!NTX-appropriate responding subjective symptoms (e.g. negative mood states) associ- increased as an orderly function of increasing dose of ated with the syndrome of withdrawal from drug treat- morphine (0.56–1.7 mg/kg) and of naltrexone (0.01– ment (Martin 1983). Single doses of a morphine-like 10 mg/kg). The discrimination was also dependent upon agonist followed by the administration of an opioid interval between morphine and naltrexone administration. antagonist result in many of the same changes comprising The MOR!NTX cue was fully generalized to the the withdrawal syndrome, suggesting a common mecha- combination of levorphanol (0.3 mg/kg) followed by nism. For example, in humans a single dose of morphine naltrexone, but not to the non-opioid stereoisomer of followed by the antagonist naloxone elicits physical signs levorphanol, dextrorphan (0.3 and 3.0 mg/kg) or the that are qualitatively similar to those experienced after kappa-opioid-receptor-selective agonist U69,593 (0.3 mg/ withdrawal from chronic morphine treatment (Bickel et kg) followed by naltrexone. Naltrexone administered al. 1988; Heishman et al. 1989; June et al. 1995). 15 min before morphine dose-dependently blocked Similarities in antagonist-precipitated changes after MOR!NTX-appropriate responding. Conclusions: This chronic and acute administration of morphine also have is the first non-rodent study of the discriminative effects been observed for a variety of dependent measures in of MOR!NTX. MOR!NTX produces a unique intero- several animal models (Kosersky et al. 1974; Eisenberg ceptive stimulus that is pharmacologically selective, 1982; Krystal and Redmond Jr 1983; Ramabadran 1983; requires occupation of opioid receptors, presumably mu, Schnur 1991; Easterling and Holtzman 1997). Addition- for some minimum period of time, and is reversible. This ally, naloxone after a single dose of morphine induces discrimination procedure might provide new insights into subjective symptoms that are identified as withdrawal by the early drug-receptor interactions that underlie the subjects with a history of opiate use (Bickel et al. 1988; development of physical dependence upon morphine-like Heishman et al. 1989; June et al. 1995). These studies drugs. indicate that even a single dose of a morphine-like agonist can induce a state of physical dependence, which is revealed following challenge with an opioid antagonist. D. A. White ()) · S. G. Holtzman Drug discrimination affords a useful animal model for Department of Pharmacology, studying subjective drug effects, including those associ- Emory University School of Medicine, Rollins Research Center, ated with morphine withdrawal (Holtzman 1990). Rats 1510 Clifton Road NE, Suite 5074, Atlanta, GA 30322, USA treated chronically with morphine can be trained to e-mail: [email protected] discriminate saline from 0.1 mg/kg of the opioid antag- Tel.: +1-404-7270356 onist naltrexone (Gellert and Holtzman 1979; Holtzman Fax: +1-404-7270365 204 1985), which is significantly less than the discriminable Materials and methods dose in non-morphine dependent animals (Valentino et al. 1983). This interoceptive stimulus is generalized to other Subjects opioid antagonists but not to opioid agonists or to non- Eight male squirrel monkeys (Saimiri sciureus), weighing between opioid drugs. Additionally, rats abruptly withdrawn from 550 and 700 g at the beginning of experimentation, were pair- chronic morphine treatment respond as though they have housed with unlimited access to food and water. Monkeys were received naltrexone, suggesting similarities in the inter- provided with fresh fruit, peanuts, or a vitamin supplement mixture oceptive states elicited by naltrexone-precipitated and each day in the home cage. All of the monkeys were experimentally naive at the beginning of this study. Animals were maintained spontaneous withdrawal. according to the “Guide for the Care and Use of Laboratory The discriminative effects of antagonist-precipitated Animals” (National Academy of Sciences, 1996), and all proce- withdrawal from acute opioid dependence have also been dures were approved by the Institutional Animal Care and Use characterized in rats (Easterling and Holtzman 1999). Committee of Emory University. Subjects were trained to discriminate a combination of 10 mg/kg morphine and 0.3 mg/kg naltrexone adminis- Apparatus tered (SC) 4 and 0.25 h before a session, respectively (MOR!NTX), from a combination of saline followed by During experimental sessions, monkeys were seated in small 0.3 mg/kg naltrexone (SAL!NTX) at the same pretreat- primate chairs housed in ventilated and sound-attenuated chambers (BRS/LVE Inc., Laurel, Md., USA). The chairs were equipped with ment times. MOR!NTX-appropriate responding was an a small stock and two brass electrodes through which electric orderly function of the doses of naltrexone and morphine, current was delivered to a shaved portion of the monkey’s tail. Two with increased responding generally occurring with response levers were mounted 9.5 cm apart on the front panel and increased dose of drug. Additionally, MOR!NTX- 3 cm from the sidewalls. A Plexiglas partition extended from the ceiling to the waist-plate of the chair, creating a wall 6 cm out from appropriate responding was a function of the interval of the front panel. Two slots approximately 10 cm apart, measuring time between the administration of morphine and nal- 45 cm each, were cut out of this partition just in front of each lever trexone, with peak responding occurring at an interval of with approximately 10 cm between the slots to prevent the monkey 3.75 h. Rats trained with MOR!NTX also generalized to from reaching and pressing both levers simultaneously. A red morphine followed by naloxone. Finally, when training stimulus light was mounted at eye level and centered between the two response levers on the front panel. A white house light was was suspended and morphine was administered to the positioned above the red stimulus light or on the rear wall of the subjects continuously via osmotic pumps (20 or 40 mg/kg chamber, depending on the chamber used. The chambers were per day), subsequent injections of naltrexone occasioned equipped with white noise to mask extraneous sounds. dose-dependent increases in MOR!NTX-appropriate responding and complete generalization with the Drug discrimination procedure MOR!NTX training state. This further suggests that the interoceptive stimuli produced by an opioid antagonist Experimental sessions were conducted daily Monday through following either acute or prolonged pretreatment with an Friday and consisted of 25 trials. Monkeys were trained to press the response levers under a fixed-ratio 1 (FR1) schedule of stimulus opioid agonist are qualitatively similar. termination/avoidance. At the beginning of each trial, the house The purpose of this study was to extend observations on light was illuminated and the monkey had 5 s to press the lever interoceptive stimuli associated with antagonist-precipitat- appropriate for the injection combination received before the ed withdrawal from acute morphine dependence to a non- session in order to avoid a 2–4 mA electrical stimulus to the tail. If human primate: the squirrel monkey. Compared to rats, the monkey failed to press the correct lever within 5 s, the electrical stimulus was delivered in 1-s pulses every 2 s until the monkey monkeys show differences in the relative ratios and responded on the correct lever or until 15 stimuli were delivered. At distribution of opioid receptor subtypes (Mansour et al. the end of the trial, the house light was turned off and the red 1988). Species-related behavioral differences exist as well. stimulus light was turned on for a 30-s time-out period. Each For example, pretreatment with a single dose of morphine response on the incorrect lever resulted in an electrical stimulus and a 3-s changeover delay, during which responses on the correct lever markedly potentiates the response-rate-decreasing effect of did not end the trial. Each response during the 30-s time-out period opioid antagonists in rats (Young 1986; Adams and also resulted in the delivery of an electrical stimulus to the tail in Holtzman 1990; Oliveto et al. 1991; Easterling and order to discourage responding between trials. Holtzman 1997), an effect not observed in monkeys Monkeys received an IM injection of either saline or 1.7 mg/kg morphine 4 h before each daily session, followed by 0.1 mg/kg (France and Morse 1989). Therefore, we first determined if naltrexone 3.75 h later (SAL!NTX and MOR!NTX, respective- stimulus control of behavior could be established in the ly). These training doses were selected on the basis of the results of absence of those marked effects on operant behavior. pilot experiments that had been conducted in our laboratory. For When we found that monkeys could be trained to half of the monkeys, the left lever was correct when morphine discriminate morphine followed by naltrexone preceded naltrexone and the right lever was correct when saline preceded naltrexone; lever assignments were reversed for the other (MOR!NTX) from a combination of saline followed by half of the animals. Drug discrimination training continued until naltrexone (SAL!NTX), our second objective was to each monkey achieved a criterion of emitting the first response on begin characterizing the discrimination pharmacologically. the injection-appropriate lever in 88% (i.e. 22) of the trials in a session for four consecutive daily sessions, two sessions with each of the two training-drug combinations. Monkeys then underwent two test sessions, one for each condition. Test sessions were similar to training sessions, except there was no “correct” lever, so that a trial was terminated by a response on either lever. If monkeys 205 responded on the injection-appropriate lever in 88% of the trials or saline preceded an injection of saline, mean response latencies on both test days, they were considered to have met criteria for were compared with either a one-factor ANOVA followed by acquisition of the discrimination. Newman-Keuls test or by Student’s t-test, as appropriate. P-values Following acquisition of the discrimination, novel doses and/or 0.05 were accepted as statistically significant. drugs were tested once or twice per week, with at least 3 days between successive tests. Between test sessions, monkeys were required to perform at criterion (88% injection-appropriate responding) in at least two consecutive training sessions, one Results preceded by morphine and naltrexone injections and one preceded by saline and naltrexone injections. Discrimination acquisition Initially, dose-response curves for naltrexone following saline or 1.7 mg/kg morphine pretreatment were obtained for each monkey (n=7). On the basis of the variance of the results of that Seven monkeys (of eight) reached the criterion for experiment, subsequent experiments were performed on three to acquisition of the discrimination in an average (€SEM) five subjects. The next two series of experiments involved varying of 108€14 training sessions. One monkey failed to reach the dose (0.56 and 1.0 mg/kg) and pretreatment interval (0.5–16 h) criterion after more than 300 training sessions and was of morphine to determine whether the interoceptive effects of withdrawal from acute dependence are a function of the dose and removed from the study. All of the monkeys that were duration of morphine exposure. Novel drugs were then tested in trained successfully responded within 5 s of the start of a place of either morphine or naltrexone. The effects of drug doses trial, thereby avoiding the electrical stimulus to the tail. and drug combinations were tested in a systematic order, with the This performance was not altered within the dose ranges starting dose of naltrexone varying for each subject. All agonists tested; therefore avoidance data are not presented. were administered 4 h before a session (except for time course experiment) and antagonists were administered 0.25 h before. Drugs were tested up to doses that either substituted completely for MOR!NTX or produced observable side effects (e.g. muscle Stimulus-generalization curves following saline tremors, sedation, and/or respiratory depression) that suggested that or morphine pretreatment higher doses could be a threat to the well-being of the monkeys. Following pretreatment with 1.7 mg/kg morphine, in- Drugs creasing doses of naltrexone produced dose-dependent increases in MOR!NTX-appropriate responding (Fig. 1), Morphine sulfate (Penick, Newark, N.J., USA), naltrexone hydro- chloride, naloxone hydrochloride, dextrorphan tartrate (Sigma and full generalization occurred at 0.1 and 1.0 mg/kg. Chemical Co., St Louis, Mo., USA), and levorphanol tartrate Pretreatment with 1.0 or 0.56 mg/kg morphine resulted in (Roche Laboratories, Nutley, N.J., USA) were dissolved in 0.9% rightward shifts of the naltrexone stimulus-generalization saline. {[5a,7a,8b)-(+)-N-{7-(1-pyrrolidinyl)-1-oxaspiro(4,5)dec-8- curve relative to the curve after pretreatment with 1.7 mg/ yl)]benzene-acetamide} (U69,593; Research Biochemicals, Inc., kg morphine (Fig. 1), and increased the ED of Natick, Mass., USA) was dissolved in 3 parts 8.5% lactic acid and 2 50 parts 1.0 N sodium hydroxide. All drugs were injected IM in a volume of 0.25 ml/kg body weight to minimize irritation. Drug doses are expressed as the free base.

Data analysis

Stimulus-generalization data are expressed as the mean (n=4–7) number of trials completed on the response lever appropriate for the MOR!NTX condition; the remaining trials of the session were completed on the lever appropriate for SAL!NTX. Complete substitution (generalization) for MOR!NTX was defined as completion of 88% of the trials in the test session on the MOR!NTX-appropriate lever. Antagonist ED50s (i.e. the dose at which 50% of the trials are completed on the MOR!NTX- appropriate lever) following agonist pretreatment and antagonist AD50 (i.e. the dose that reduces MOR!NTX-appropriate respond- ing by 50% when administered before morphine) values were estimated for each monkey by log-linear interpolation of ascending or descending portions of the dose-response curve; group means and SEM were calculated from those data. ED50 data were analyzed with either a one-factor ANOVA and Newman-Keuls test post hoc Fig. 1 MOR!NTX discrimination is a function of MOR and NTX or with Student’s t-test, as appropriate. The time until responding dose. The discriminative stimulus effects of SAL or MOR (0.56– on the MOR!NTX-appropriate lever reached and returned to one- 1.7 mg/kg, SC; 4 h pretreatment) followed by SAL or NTX (0.01– 10 mg/kg, SC; 0.25 h pretreatment) were determined in monkeys half its maximal value (t1/2) was calculated by linear regression of the ascending or descending part of the time-course curve. These trained to discriminate 1.7 mg/kg MOR!0.1 mg/kg NTX from served as estimates of the onset and offset rates of acute morphine SAL!NTX. Each point represents a mean of one observation in dependence. each of four to seven monkeys. The dashed lines at the top and The latency to emit the first response was recorded and bottom of the figure represent criteria for full MOR!NTX and averaged over the 25-trial session. Mean group (saline or drug SAL!NTX appropriate responding, respectively. Possible sensiti- pretreatment) response latencies were calculated and analyzed zation to NTX was determined by re-testing the stimulus effects of SAL!1.0 mg/kg NTX following the completion of the MOR using a one-factor repeated measures ANOVA and Dunnett’s test ' post hoc. For control sessions in which an injection of either a drug stimulus generalization curves ( ). There was no increase in sensitivity to NTX 206

Table 1 The effects of opioid agonist pretreatment 4 h prior to testing on the ED50 of naltrexone or naloxone administered 15 min prior to testing

Agonist dose (mg/kg) Antagonist dose (mg/kg) Antagonist ED50 Potency relative n and 95% CI (mg/kg) to training condition Saline Naltrexone 0.1–10 – – 7 Naloxone 1.0, 10 – – 4 Morphine 0.56 Naltrexone 0.1–10 1.53 (0.21, 11.04) 0.016* 5 1.0 Naltrexone 0.03–1.0 0.094 (0.012, 0.72) 0.26* 4 1.7 Naltrexone 0.01–1.0 0.024 (0.012, 0.045) 1.0 7 1.7 Naloxone 0.01–1.0 0.073 (0.051, 0.11) 0.33* 4 Levorphanol 0.3 Naltrexone 0.01–1.0 0.031 (0.014, 0.07) 0.77 4 Dextrorphan 0.3 Naltrexone 0.01–1.0 – – 4 3.0 Naltrexone 0.01–1.0 – – 4 U69,593 0.3 Naltrexone 0.01–1.0 – – 4

* Significantly different from naltrexone ED50 after morphine training dose, P<0.05

Table 2 Naltrexone and naloxone dose-dependently decrease the main effect of antagonist dose for both naltrexone (NTX) and average latency to respond. The data represent the average latency naloxone (NX)[F(2,46)=15.29 and F(2,6)=32.1, respectively] (s) to respond per trial. Values are the mean€SEM. There was a 4 h Ptx 0.25 h Ptx n Antagonist dose (mg/kg) 0 (SAL) 0.01 0.03 0.1 0.3 1.0 10 Saline Naltrexone 7 1.82€0.31 1.47€0.28 1.46€0.29* 1.23€0.24* Naloxone 4 2.36€0.27 1.49€0.38* 1.05€0.15* Morphine 0.56 Naltrexone 5 1.69€0.31 0.97€0.14* 0.90€0.11* 0.78€0.10* 1.0 Naltrexone 4 2.04€0.43 1.58€0.26 1.53€0.23 1.35€0.23* 1.33€0.23* – 1.7 Naltrexone 4 1.62€0.36 1.57€0.33 1.15€0.23* 1.44€0.27 – 1.03€0.19* – 1.7 Naloxone 4 1.98€0.39 1.60€0.37 1.74€0.39 1.30€0.27* – 0.94€0.15* – * Significantly different from within-group saline control, P<0.05 naltrexone by 4- and 42-fold, respectively [Table 1; kg naltrexone decreased the latency to respond by 31€8% F(2,5)=44.87; P<0.05]. after saline and 32€4% after 1.7 mg/kg morphine. After 4-h pretreatment with saline, doses of naltrexone Following completion of the experiments described as high as 10 mg/kg resulted in fewer than six trials, on above, we re-tested saline (4 h pretreatment) followed by average, being completed on the MOR!NTX-appropri- 1.0 mg/kg naltrexone (15 min pretreatment) to determine ate lever. This dose of naltrexone was 100 times higher if the monkeys had become sensitized to naltrexone as the than the training dose (i.e. 0.1 mg/kg) and more than 400 result of repeated exposure to morphine. No apparent times higher than the ED50 of naltrexone after the training sensitization occurred as the subjects responded exclu- dose of morphine (Table 1). Therefore, an ED50 could not sively on the SAL!NTX lever (Fig. 1). be calculated for naltrexone after pretreatment with saline. Administration of saline 3.75 h after pretreatment with saline or any of the three morphine doses tested Temporal dependency of MOR!NTX discrimination failed to produce MOR!NTX-appropriate responding (Fig. 1). To assess the dependency of the discrimination on the There were no effects of 4 h saline or morphine interval between injection of the two training drugs, pretreatment on the average latency to respond in sessions pretreatment with morphine (1.7 mg/kg) was systemati- that began 15 min after an injection of saline (Table 2). cally varied from 0.2516 h while the naltrexone dose However, naltrexone dose-dependently decreased re- (0.1 mg/kg) and pretreatment time (0.25 h) remained sponse latencies, regardless of whether saline or morphine constant (Fig. 2). Responding on the MOR!NTX- (0.56, 1.0 or 1.7 mg/kg) had been given 3.45 h earlier appropriate lever reached 50% of peak value at 2 h after [Table 1; F(3,18)=6.53; F(3,18)=7.19; F(4,24)=6.36; and the morphine injection and peaked at 4 h. At 16 h after F(4,24)=5.63, respectively; P<0.05]. For example, 1.0 mg/ morphine pretreatment, responding occurred exclusively 207

Fig. 2 MOR!NTX discrimination is a function of MOR pretreat- Fig. 3 NTX pretreatment dose-dependently blocks MOR!NTX ment time. The discriminative stimulus effects of NTX (0.1 mg/kg, discrimination. NTX (0.01–1 mg/kg) was given 0.25 h before the SC; 0.25 h pretreatment) following MOR (1.7 mg/kg, SC) given normal training regiment of 1.7 mg/kg MOR (4 h pretreat- 0.5–16 h prior to testing were assessed. Each point represents a ment)!0.1 mg/kg NTX (0.25 h pretreatment). Each point repre- mean of four monkeys. Full MOR!NTX responding occurred with sents a mean of five monkeys. Other details are as in Fig. 1 4 h of MOR administration but was gone by 16 h following MOR. Other details are as in Fig. 1 on the SAL!NTX lever. The onset and offset of MOR!NTX-appropriate responding occurred with half- lives of 2.50€0.31 and 2.17€0.22 h, respectively.

Antagonism of stimulus effects

To determine if the interoceptive effects of MOR!NTX are mediated by opioid receptors, saline or naltrexone (0.01–1.0 mg/kg) was administered 0.25 h prior to the normal MOR!NTX training regimen (i.e. 4.25 h before the test session). Saline administered prior to the training drug combination did not alter the amount of MOR!NTX responding (Fig. 3). However, pretreatment Fig. 4 NX substitutes for NTX following 4 h pretreatment with with increasing doses of naltrexone dose-dependently MOR. MOR!NTX and SAL!NTX curves are reproduced from attenuated MOR!NTX-appropriate responding, with a Fig. 1. Each point of the NX curves represents a mean of four dose of 1.0 mg/kg reducing the stimulus effects of monkeys. Other details are as in Fig. 1 MOR!NTX by more than 75%. The AD50 of naltrexone for blocking the discriminative effects of MOR!NTX was 0.084 (CL 0.0094, 0.76) mg/kg. Pharmacological selectivity of stimulus control of behavior

Generalization with morphine and naloxone Pretreatment with levorphanol (0.3 mg/kg) in place of morphine followed by various doses of naltrexone Similar to the results obtained with SAL!NTX, the produced dose-dependent MOR!NTX-appropriate re- combination of 4 h pretreatment with saline and 15 min sponding, with full substitution occurring after the pretreatment with 1.0 or 10 mg/kg naloxone occasioned administration of 0.1 mg/kg naltrexone (Fig. 5). However, little or no MOR!NTX-appropriate responding (Fig. 4). the curve was biphasic: 1.0 mg/kg naltrexone resulted in In contrast, when 1.7 mg/kg morphine was given 4 h >40% less MOR!NTX-appropriate responding than did before testing, the combination of morphine followed 0.1 mg/kg naltrexone. The ED50 of naltrexone for the naloxone fully and dose-dependently generalized with the ascending part of the curve was equivalent to that MOR!NTX cue. However, naloxone was significantly observed after the training dose morphine (i.e. 1.7 mg/ less potent than naltrexone was by a factor of three kg; Table 1). There was no effect of levorphanol followed [t(1,6)=4.40; P<0.05; Table 2]. Naloxone, like naltrex- by saline on the average response latency. one, dose-dependently decreased the average latency to Pretreatment with either of two doses of dextrorphan respond [Table 1; F(2,12)=32.1; P<0.05]. (0.3 and 3.0 mg/kg) followed by 0.01–1.0 mg/kg 208 were approximately one-sixth and one-third, respectively, the doses required for rats. In contrast, monkeys and rats are approximately equisensitive to the discriminative effects of morphine alone (Shannon and Holtzman 1976; Schaefer and Holtzman 1977). Unlike rats, squirrel monkeys do not become sensitized to the response-rate-decreasing effects of an opioid antagonist after pretreatment with single or even repeated doses of an opioid agonist (France and Morse 1989). Despite the absence of grossly observable behavioral effects (i.e. suppressed responding), squirrel monkeys were brought under stimulus control by the combination of morphine and naltrexone, demonstrating the manifes- tation of a unique discriminable interoceptive state. This is consistent with the results of a study in which the Fig. 5 MOR!NTX discrimination is stereoselective and opioid “behavioral/motivational” signs of withdrawal (e.g. con- receptor subtype dependent. Generalization curves for SAL and ditioned place aversion) occurred in the absence of overt NTX (0.01–1.0 mg/kg; 0.25 h) were constructed following 4 h somatic signs (Schulteis et al. 1994). These data suggest pretreatment (SC) with levorphanol (0.3 mg/kg) dextrorphan (0.3 and 3.0 mg/kg), and U69,593 (U69; 0.3 mg/kg). Each point that the discriminative stimuli associated with antagonist- represents a mean of four monkeys. Other details are as in Fig. 1 precipitated withdrawal from acute opioid dependence occur independently of the observable somatic and behavioral signs of withdrawal. naltrexone failed to produce discriminative effects com- A requirement for the stimulus control of behavior by parable to those produced by MOR!NTX. The maxi- MOR!NTX was that opioid receptors had to be occupied mum number of trials completed on the MOR!NTX by morphine for some minimum period of time before lever averaged only 6.3, even after a dose of dextrorphan naltrexone was administered. Little MOR!NTX-appro- 10 times higher than the effective dose of levorphanol and priate responding occurred when morphine was adminis- a dose of naltrexone more than 30 times higher than the tered within 1 h of the session (15–45 min before naltrexone ED50 after levorphanol pretreatment (Fig. 5). naltrexone). Doses of naltrexone higher than the training Pretreatment with U69,593 (0.3 mg/kg) followed by dose were not tested at those early time points. Rats tested naltrexone (0.1–1.0 mg/kg) also did not induce with doses of naltrexone as much as two orders of MOR!NTX-like discriminative effects; responding oc- magnitude greater than the training dose following 15– curred only on the lever appropriate for SAL!NTX at all 45 min pretreatment with the training dose morphine doses of naltrexone (Fig. 5). responded almost exclusively on the choice lever appro- There was no main effect of dextrorphan or U69,593 priate for SAL!NTX (Easterling and Holtzman 1999). pretreatment on the average latency to respond in sessions Furthermore, MOR!NTX-appropriate responding was that followed an injection of saline 15 min earlier. reduced dose-dependently when naltrexone was adminis- tered 15 min before the training dose of morphine, decreasing the access of the agonist to opioid receptors. Discussion Thus, the discriminative effects of MOR!NTX required a finite period of agonist-receptor interaction and were The results of this study show that squirrel monkeys can not a consequence of non-receptor-mediated effects of be trained to discriminate the combination of 1.7 mg/kg either morphine or its metabolites. However, stimulus morphine (4 h pretreatment) followed by 0.1 mg/kg control of behavior was not based on the mere presence naltrexone (15 min pretreatment) from a combination of (or absence) of morphine at the time of testing, as might saline and naltrexone. Stimulus control of behavior was occur in animals discriminating between saline and an orderly function of the dose of morphine, the dose of naltrexone and pretreated with morphine before each naltrexone, and the interval between the two drugs. In (France and Woods 1985). When our monkeys were addition, the discriminative stimulus appears to be pretreated with morphine followed by saline as well as pharmacologically selective and mediated by opioid when they were pretreated with a combination of saline receptors. These data are consistent with previous find- followed by saline or by naltrexone training dose they ings with rats trained to discriminate 10 mg/kg morphine responded exclusively on the lever appropriate for the (4 h pretreatment) followed by 0.3 mg/kg naltrexone SAL!NTX state. These results support the view that the (15 min pretreatment) versus saline followed by naltrex- discriminative effects of MOR!NTX reflect interocep- one (Easterling and Holtzman 1999). However, the tive stimuli associated with antagonist-precipitated with- squirrel monkeys were far more sensitive to the stimulus drawal from acute physical dependence upon morphine, effects of the training drugs than the rats were: stimulus consistent with findings of studies on rats (Easterling and control of behavior in the monkeys was induced and Holtzman 1999) and humans (Bickel et al. 1988; Heish- maintained with doses of morphine and naltrexone that man et al. 1989; June et al. 1995). 209

An alternative explanation for the results of this study (t1/2=2–3 h) (Berkowitz 1976; Bigler et al. 1990; Got et al. is that morphine pretreatment simply enhanced a pre- 1994). In clinical studies, naloxone challenge after a existing effect of naltrexone. This seems unlikely. First, in single high dose of morphine elevated subjective with- water-deprived rats, potentiation of naltrexone-induced drawal scores even after morphine plasma levels had drinking suppression was an inverse function of time declined to nearly undetectable levels (June et al. 1995). following morphine pretreatment (White and Holtzman These results suggest that agonist occupation of mu- 2001); maximal effects occurred 2 h after morphine. opioid receptors results in changes to the receptor that Potentiation of the effects of naltrexone was evident 4 and persist long after the peak of agonist tissue levels. One 6 h after morphine but was significantly less than it was 2 h hypothesis is that receptor occupation by agonists results after. Second, our study showed that the combination of in phosphorylation of the receptors, which then become morphine followed by naloxone fully generalized to the constitutively active. Naloxone and naltrexone function as MOR!NTX cue, albeit at a 3-fold lower potency. This inverse agonists at the constitutively activated receptors, demonstrates that the antagonist action of a drug is critical precipitating withdrawal effects (Wang et al. 1994; Wang for the discrimination, consistent with the notion of et al. 2001). Another possibility is that endogenous antagonist-precipitated withdrawal. Similar substitution opioids released by morphine continue to occupy the mu- of naloxone for naltrexone occurred in rats treated acutely opioid receptor while tissue concentrations of morphine or chronically with morphine (Gellert and Holtzman 1979; are declining (Schlen and Bentley 1980; Olive et al. Easterling and Holtzman 1999). Third, saline followed by 1995), and it is the removal of them by antagonists that naltrexone did not occasion substantial responding on the results in interoceptive withdrawal stimuli. MOR!NTX-appropriate lever, even at a dose of naltrex- Whatever the cellular bases of the discrimination, the one (10 mg/kg) more than 400 times higher than the ED50 changes induced by the agonist-receptor interaction were of naltrexone after pretreatment with the training dose of completely reversible, paralleling acute dependence stud- morphine. Virtually complete dissociation between the ies on humans (Stitzer et al. 1991; June et al. 1995) and two training conditions suggests that MOR!NTX engen- rats (Easterling and Holtzman 1999). This was evidenced ders discriminative stimuli that are truly unique, rather in several ways, including 1) the time course of than simply enhancing effects of naltrexone that might be MOR!NTX-appropriate responding, 2) the stabile be- present after pretreatment with saline. havior on training days when the combination of saline Results of the initial pharmacological characterization and naltrexone were administered, and 3) the absence of of the MOR!NTX discrimination suggest that it is MOR!NTX-appropriate responding when the monkeys stereoselective and is at least partially mediated by the were tested at the end of the study with saline and 1.0 mg/ mu-opioid receptor. First, the combination of the morphine kg naltrexone. Furthermore, there was no indication of the congener levorphanol (0.3 mg/kg) and naltrexone were sensitization to behavioral effects of antagonists that can fully generalized with MOR!NTX. However, pretreat- occur from repeated administration of opioid agonists or ment with dextrorphan, the non-opioid stereoisomer of opioid antagonists (Goldberg et al. 1981; Dykstra 1983; levorphanol, followed by naltrexone occasioned little Young 1986; Adams and Holtzman 1990). MOR!NTX-appropriate responding, even at a dose of Our experiments revealed a shortening of response dextrorphan 10-fold higher than that of levorphanol. latency by naltrexone and naloxone. To the best of our Second, pretreatment with the combination of the kappa- knowledge, this is a novel finding. This finding is at odds selective agonist U69,593 and naltrexone failed to elicit a with that of another study where neither naloxone (0.1– MOR!NTX-like stimulus. The dose of U69,593 tested, 30 mg/kg) nor naltrexone (0.1–10 mg/kg) altered the 0.3 mg/kg, is 2–4 times the ED50 for suppressing in squirrel average cumulative response latency during generaliza- monkeys responding maintained by aversive stimuli (Pitts tion tests in squirrel monkeys trained to discriminate and Dykstra 1994; Powell and Holtzman 1999), and 0.1 mg/kg diprenorphine from vehicle (DeRossett and produced signs of respiratory depression in our monkeys. Holtzman 1986). We have no explanation for the different The possible mediation of this effect by delta-opioid findings of the two studies. receptors or even certain classes of non-opioid receptors This is the first study to assess the interoceptive effects cannot be excluded at this time. However, experimental of withdrawal from acute opioid dependence in a non- evidence suggests that the withdrawal-like effects associ- human primate. Our data add further support to the ated with acute opioid dependence are not a common conclusion that acute pretreatment with mu-opioid-receptor characteristic of all receptors systems. For example, in rats, agonists does not simply enhance the existing stimulus flumazenil-induced suppression of intracranial self-stimu- effects of antagonists, but rather induces a state of acute lation was not potentiated by acute pretreatment with the physical dependence that is revealed by antagonist admin- positive modulators of the GABAA receptor chlordiaz- istration (Easterling and Holtzman 1999). This discrimi- epoxide or diazepam (Easterling et al. 2000). nation procedure could afford a valuable animal model for Stimulus control of behavior was maximum at a time studying mechanisms that underlie the initial development when tissue levels of morphine are reported to be of physical dependence upon opioids and of the subjective declining in a number of species, including rats (t1/2= symptoms of withdrawal from morphine-like drugs. 0.9€ 0.2 h) (Barjavel et al. 1995), rhesus monkeys (t1/2 approximately 2 h) (Hartvig et al. 1984), and humans 210 Acknowledgements We thank Dr. Kelly R. Powell for conducting Holtzman SG (1985) Discriminative stimulus effects of morphine the preliminary experiments that led the way to this study. This withdrawal in the dependent rat: suppression by opiate and research was supported by NIH grants DA00541 and K05 nonopiate drugs. J Pharmacol Exp Ther 233:80–86 DA00008. Holtzman SG (1990) Discriminative stimulus effects of drugs: relationship to potential for abuse. In: Adler MW, Cowan A (eds) Modern methods in pharmacology, testing and evaluation of drugs of abuse. Wiley-Liss, New York, pp 193–210 References June HL, Stitzer ML, Cone E (1995) Acute physical dependence: time course and relation to human plasma morphine concen- Adams JU, Holtzman SG (1990) Pharmacologic characterization of trations. Clin Pharmacol Ther 57:270–280 the sensitization to the rate-decreasing effects of naltrexone Kosersky DS, Harris RA, Harris LS (1974) Naloxone-precipitated induced by acute opioid pretreatment in rats. J Pharmacol Exp jumping activity in mice following the acute administration of Ther 253:483–489 morphine. Eur J Pharmacol 26:122–124 Barjavel MJ, Scherrmann JM, Bhargava HN (1995) Relationship Krystal JH, Redmond DE Jr (1983) A preliminary description of between morphine analgesia and cortical extracellular fluid acute physical dependence on morphine in the vervet monkey. levels of morphine and its metabolites in the rat: a microdialysis Pharmacol Biochem Behav 18:289–291 study. Br J Pharmacol 116:3205–3210 Mansour A, Khachaturian H, Lewis ME, Akil H, Watson SJ (1988) Berkowitz BA (1976) The relationship of pharmacokinetics to Anatomy of CNS opioid receptors. Trends Neurosci 11:308– pharmacological activity: morphine, methadone and naloxone. 314 Clin Pharmacokinet 1:219–230 Martin WR (1983) Pharmacology of opioids. Pharmacol Rev Bickel WK, Stitzer ML, Liebson IA, Bigelow GE (1988) Acute 35:283–323 physical dependence in man: effects of naloxone after brief Olive MF, Bertolucci M, Evans CJ, Maidment NT (1995) morphine exposure. J Pharmacol Exp Ther 244:126–132 Microdialysis reveals a morphine-induced increase in pallidal Bigler D, Christensen CB, Eriksen J, Jensen NH (1990) Morphine, opioid peptide release. Neuroreport 6:1093–1096 morphine-6-glucuronide and morphine-3-glucuronide concen- Oliveto AH, Picker MJ, Dykstra LA (1991) Acute and chronic trations in plasma and cerebrospinal fluid during long-term morphine administration: effects of mixed-action opioids in rats high- dose intrathecal morphine administration. Pain 41:15–18 and squirrel monkeys responding under a schedule of food DeRossett SE, Holtzman SG (1986) Discriminative stimulus effects presentation. J Pharmacol Exp Ther 257:8–18 of the opioid antagonist diprenorphine in the squirrel monkey. J Pitts RC, Dykstra LA (1994) Antinociceptive and response rate- Pharmacol Exp Ther 237:437–444 altering effects of kappa opioid agonists, spiradoline, enadoline Dykstra LA (1983) Development of enhanced sensitivity to and U69,593, alone and in combination with opioid antagonists naloxone. Life Sci 33:2079–2089 in squirrel monkeys. J Pharmacol Exp Ther 271:1501–1508 Easterling KW, Holtzman SG (1997) Intracranial self-stimulation Powell KR, Holtzman SG (1999) Differential antagonism of the in rats: sensitization to an opioid antagonist following acute or rate-decreasing effects of kappa-opioid receptor agonists by chronic treatment with mu opioid agonists. J Pharmacol Exp naltrexone and norbinaltorphimine. Eur J Pharmacol 377:21–28 Ther 281:188–199 Ramabadran K (1983) Naloxone-precipitated abstinence in mice, Easterling KW, Holtzman SG (1999) Discriminative stimulus rats and gerbils acutely dependent on morphine. Life Sci effects of naltrexone after a single dose of morphine in the rat. J 33:385–388 Pharmacol Exp Ther 288:1269–1277 Schaefer GJ, Holtzman SG (1977) Discriminative effects of Easterling KW, Plovnick RM, Holtzman SG (2000) Acute opioid morphine in the squirrel monkey. J Pharmacol Exp Ther but not benzodiazepine dependence in rats responding for 201:67–75 intracranial self-stimulation. Psychopharmacology 148:263– Schlen H Bentley GA (1980) The possibility that a component of 271 morphine-induced analgesia is contributed indirectly via the Eisenberg RM (1982) Short-term tolerance to morphine: effects of release of endogenous opioids. Pain 9:73–84 diazepam, phenobarbital, and amphetamine. Life Sci 30:1615– Schnur P (1991) Acute morphine dependence in the hamster. 1623 Pharmacol Biochem Behav 38:711–713 France CP, Morse WH (1989) Pharmacological characterization of Schulteis G, Markou A, Gold LH, Stinus L, Koob GF (1994) supersensitivity to naltrexone in squirrel monkeys. J Pharmacol Relative sensitivity to naloxone of multiple indices of opiate Exp Ther 250:928–936 withdrawal: a quantitative dose-response analysis. J Pharmacol France CP Woods JH (1985) Opiate agonist-antagonist interactions: Exp Ther 271:1391–1398 application of a three-key drug discrimination procedure. J Shannon HE, Holtzman SG (1976) Evaluation of the discriminative Pharmacol Exp Ther 234:81–89 effects of morphine in the rat. J Pharmacol Exp Ther 198:54–65 Gellert VF, Holtzman SG (1979) Discriminative stimulus effects of Stitzer ML, Wright C, Bigelow GE, June HL, Felch LJ (1991) Time naltrexone in the morphine-dependent rat. J Pharmacol Exp course of naloxone-precipitated withdrawal after acute meth- Ther 211:596–605 adone exposure in humans. Drug Alcohol Depend 29:39–46 Goldberg SR, Morse WH, Goldberg DM (1981) Acute and chronic Valentino RJ, Herling S, Woods JH (1983) Discriminative stimulus effects of naltrexone and naloxone on schedule-controlled effects of naltrexone in narcotic-naive and morphine-treated behavior of squirrel monkeys and pigeons. J Pharmacol Exp pigeons. J Pharmacol Exp Ther 224:307–313 Ther 216:500–509 Wang D, Raehal KM, Bilsky EJ, Sadee W (2001) Inverse agonists Got P, Baud FJ, Sandouk P, Diamant-Berger O, Scherrmann JM and neutral antagonists at mu opioid receptor (MOR): possible (1994) Morphine disposition in opiate-intoxicated patients: role of basal receptor signaling in narcotic dependence. J relevance of nonspecific opiate immunoassays. J Anal Toxicol Neurochem 77:1590–1600 18:189–194 Wang Z, Bilsky EJ, Porreca F, Sadee W (1994) Constitutive mu Hartvig P, Bergstrom K, Lindberg B, Lundberg PO, Lundqvist H, opioid receptor activation as a regulatory mechanism underly- Langstrom B, Svard H, Rane A (1984) Kinetics of 11C-labeled ing narcotic tolerance and dependence. Life Sci 54:L339–L350 opiates in the brain of rhesus monkeys. J Pharmacol Exp Ther White DA, Holtzman SG (2001) Acute opioid pretreatment 230:250–255 potentiates naltrexone-induced drinking suppression in water- Heishman SJ, Stitzer ML, Bigelow GE, Liebson IA (1989) Acute deprived rats. J Pharmacol Exp Ther 298:156–164 opioid physical dependence in postaddict humans: naloxone Young AM (1986) Effects of acute morphine pretreatment on the dose effects after brief morphine exposure. J Pharmacol Exp rate-decreasing and antagonist activity of naloxone. Psycho- Ther 248:127–134 pharmacology 88:201-208