Effects of buprenorphine and an alternative nondrug reinforcer, alone and in combination on smoked cocaine self-administration in monkeys

ELSEVIER

Drug and Alcohol Dependence 45 (1997) 21~29

Effects of buprenorphine and an alternative nondrug reinforcer, alone and in combination on smoked cocaine self-administration in monkeys Joshua S. Rodefer, Adande J. Mattox, Sherry S. Thompson, Marilyn E. Carroll * Department of Psychiatry,

University of Minnesota Medical School, Box 392 UMHC,

Minneapolis,

MN 55455, USA

Received 15 August 1996; accepted 14 January 1997

Abstract

The abuse of smoked cocaine base, also known as ‘crack’, continues to be a major public health problem and to date the success of pharmacological or behavioral interventions has been limited. The purpose of this study was to evaluate the efficacy of a behavioral (alternative reinforcer-saccharin) and pharmacological (0.01 mg/kg buprenorphine) treatment alone and in combination. Five adult male rhesus monkeys self-administered cocaine base (1.0 mg/kg/delivery) via the smoking/inhalation route. Each day ten smoke deliveries were available contingent upon completion of a chained FR (lever press), FR (inhalation response) response schedule during 4 hr sessions.The data were analyzed using a behavioral economic framework in which the lever press response requirements were varied from 64 to 1024 to generate a demand function (consumption x FR) for cocaine under the following conditions: (1) buprenorphine pretreatment alone (0.01 mg/kg, i.m., 30 min presession); (2) concurrent access to saccharin alone (0.03% wt/vol); and (3) buprenorphine pretreatment in the presence of concurrent accessto saccharin. Under all conditions, increases in the lever FR resulted in significant decreasesin smoked cocaine base deliveries. Neither buprenorphine pretreatment alone nor concurrent saccharin alone produced significant decreasesin smoked cocaine deliveries; however, the combination of buprenorphine pretreatment and concurrent saccharin significantly decreasedthe mean number of smoked cocaine deliveries from the no treatment baseline and from the buprenorphine alone condition. These data suggest that the combination of pharmacotherapy and alternative reinforcers may be an effective treatment strategy to alter smoked cocaine self-administration. 0 1997 Elsevier Science Ireland Ltd. Keywords: Alternative reinforcer; Behavioral economics; Buprenorphine; Cocaine; Inhalation; Rhesus monkeys: Saccharin; Self-administration; Smoking

1. Introduction Cocaine base (crack) smoking continues to be a major public health problem. According to the most recent information available from the National Institute on Drug Abuse, 4 million individuals tried crack cocaine in 1994, an increase of 5% from 1993 (National Institute on Drug Abuse, 1995). Further, increases in the number of people who tried crack cocaine once in the previous month (520000) and in the previous year (1.3 million) were also noted. The percentage of cocaine * Corresponding author. Tel.: + 1 612 6266289; fax: + 1 612 6248935; e-mail: [email protected]

users smoking cocaine base ranged from 70 to 85%; whereas, the number of individuals injecting cocaine ranged from 11 to 30% (Office of National Drug Control Policy, 1995). That the use of smoked cocaine poses significant social problems is highlighted by the mention of cocaine in 46% of drug-associated deaths in 1993 (USDHHS, 1995). Despite much research, few effective medications exist for treating crack cocaine abuse. Recently, a number of studies have evaluated buprenorphine, a partial agonist at the mu opioid receptor, as a potential pharmacological treatment for cocaine abuse. Buprenorphine decreased the self-administration of many drugs, including smoked cocaine base (Carroll et al., 1992;

0376-8716/97/$17.008 1997 Elsevier Science Ireland Ltd. All rights reserved PIISO376-8716(97)01341-O

Comer et al., 1994) i.v. cocaine and opiates (Mello et al., 1983, 1989; Mello and Negus, 1996; Winger et al.. 1992), and oral phencyclidine (PCP) and ethanol (Carroll et al.. 1992: Rawleigh et al., 1996) in rhesus monOf drug keys. Buprenorphine’s reduction self-administration occurred across a range of doses (0.0050.5 mglkg) and was non-selective in that buprenorphine also decreased intake of saccharin (Carroll et al., 1992) a highly preferred nondrug reinforcer, across a range of doses (0.003-0.8 mg/kg) in monkeys. Buprenorphine produced marked reductions in drug intake in the animal laboratory when drug was available through either the i.v. or oral route of administration (Carroll et al., 1992; Mello et al., 1989). Reductions in smoked cocaine-reinforced behavior were relatively modest and inconsistent at low buprenorphine doses (0.003 -0.01 mg/kg) (Carroll et al., 1992; Comer et al.. 1994). but higher doses (O.lLO.8 mg,ikg) appeared to yield greater reductions smoked cocaine intake (Carroll et al., 1992; Comer et al., 1994). This finding closely models buprenorphine’s limited ability to reduce cocaine intake in humans (Compton et al., 1995; Strain et al., 1994). Schottenfeld and colleages (1993) suggest that while buprenorphine has only a modest effect on cocaine use, higher doses and/or more chronic treatment may prove more successful. Yet, these clinical studies have primarily examined cocaine use in opioid-dependent individuals (Compton et al., 1995; Kosten et al., 1992; Schottenfeld et al., 1993; Strain et al., 1994) while most of the animal studies have investigated buprenorphine’s effect on purely cocaine-using subjects. The ability of non-drug substances such as increased food (Carroll and Carmona, 1991; Nader and Woolverton, 1991). saccharin (Carroll, 1985) or sucrose (Kanarek and Marks-Kaufman, 1988) have been demonstrated to alter drug-reinforced behavior. This use of alternative reinforcers has also been investigated in human research (Higgins et al., 1993, 1994; Silverman et al.. 1996). In these studies, recreational cocaine users received vouchers that could be exchanged for prosocial alternative reinforcers (e.g. movie tickets) contingent upon drug-free urine samples. The data from these preclinical and clinical studies suggest that nondrug reinforcers effectively decrease the intake of drug and reduce drug-reinforced behaviors. Furthermore, clinical studies on the effect of alternative reinforcers on drug-taking behavior demonstrate that this behavioral intervention is as successful as pharmacological approaches to treating drug abuse. A number of previous studies evaluated the efficacy of combining buprenorphine treatment and access to an alternative nondrug reinforcer (e.g. saccharin). Comer et al. (1996) studied the effects of combining buprenorphine pretreatment and access to a glucose and saccharin solution on i.v. cocaine self-administration in rats

across a range of response requirements. Both buprenorphine alone and access to the glucose and saccharin mixture significantly decreased cocaine intake. However, when the pharmacological and behavioral manipulations were combined, there was an interaction that resulted in significantly less cocaine intake than either treatment condition alone. Moreover, this effect grew in magnitude as response requirement or fixed-ratio (FR) increased (Comer et al., 1996). These findings suggest that a combination of increased response requirements with both pharmacological and behavioral interventions may produce the greatest reduction in drug self-administration. In a similar fashion, Rawleigh et al. (1996) investigated buprenorphine pretreatment and access to saccharin alone and in combination in rhesus monkeys that were self-administering orally-delivered PCP. The authors reported that both buprenorphine pretreatment and saccharin access decreased PCP self-administration. In addition the combination of pharmacological and behavioral treatments had a nearly additive effect. Comer et al. (1994) examined a different route of administraton when they compared the effects of buprenorphine and the availability of saccharin separately on smoked cocaine self-administration in rhesus monkeys. Pretreatment with buprenorphine (0.01 and 0.1 mg/kg) resulted in decreased cocaine intake in some monkeys, but only the higher dose (0.1 mg/kg) of huprenorphine significantly decreased deliveries of smoked cocaine over a range of operant requirements. Moreover, concurrent saccharin, while serving as an economic substitute for cocaine, did not alter smoked cocaine intake. Thus, the use of buprenorphine and an alternative reinforcers to reduce drug intake has produced mixed results. It appears that cocaine-base smoking is particularly resistant to either treatment. The goal of the present study was to extend the findings of Comer and colleagues (1994) and examine a combination of two treatments, 0.01 mg/kg buprenorphine pretreatment and concurrent access to saccharin, that were previously demonstrated only to be minimally effective alone in reducing cocaine-base smoking. Demand for cocaine base was assessed by examining a wide range of FR schedules for smoke deliveries to compare the reinforcing efficacy of cocaine under each of the individual and combined treatment conditions. A within-subject design was used in which the behavioral (saccharin) and pharmacological (buprenorphine) interventions and their combination were given in a nonsystematic order across monkeys. Since buprenorphine doses (Carroll et al., 1992) and saccharin concentrations (Carroll, 1985) had been examined previously, a single parameter of each treatment was chosen for its near threshold effect.

2. Methods 2.1. Subjects Five adult male rhesus monkeys (Mucaca mulutta) served as subjects (M-LR, M-S, M-N, M-R, and MM4). The monkeys were maintained at 85% of their free feeding body weights during the duration of the study. All subjects had previous experience with self-administering smoked cocaine base and one subject (M-M4) also had experience with oral drug self-administration (ethanol and phencyclidine). Three subjects (M-LR, M-S and M-M4) had previous exposure to drinking saccharin and to buprenorphine pretreatment. All subjects were housed individually in custom-made stainless-steel chambers (Lab Products, Maywood, NJ). Each chamber was positioned such that the subjects could maintain auditiory, olfactory and visual contact with the other subjects. The colony room was temperature-controlled and maintained at between 22-23°C. Lighting in the colony room was set on a 12 h light/ dark cycle with the lights on at 07:OO h. Use of the animals in this experimental protocol was approved by the University of Minnesota Animal Use and Care Committee under protocol number 9511016, and the University of Minnesota is an AAALAC accredited institution. 2.2. Apparatus The individual cages had a wire grid floor and solid side and back walls with a vertical barred door. An operant panel was attached to the exterior of one side chamber wall. The panel contained one smoking spout and one drinking spout each mounted on clear Plexiglas units that covered holes punched in the chamber walls to accomodate the drinking and smoking devices. Mounted behind the clear Plexiglas plates were 2 small 1.1 W cue lights that, when illuminated, signaled to the subject that an appropriate response had been made. The smoking and drinking spouts each protruded 2.5 cm into the chamber. The spouts were 15 cm left and right of the panel midline and 48 cm above the chamber floor. A standard primate lever was located equidistant between the spouts and 37.5 cm above the chamber floor. The operant panel also contained three stimulus lights located in a row above the level of the two spouts. Illumination of the left green light, directly above the drinking spout, signaled liquid availability (fixed-ratio 1 schedule; FRl); the red light above the lever (center) signaled the lever phase (FR 64-1024) of the experiment and the right green light, located directly above the smoking spout, flashed (10 Hz) and signaled the puff phase (FR 5) of the experiment. Each monkey’s operant panel was controlled by a microcomputer (MicroInterfaces, Minneapolis, MN) that was located in an adjacent room.

Liquids were dispensed from 2000 ml liquid reservoirs that were attached to and suspended above the operant panel. Each lip contact (FR 1) resulted in the closing of an electronic drinking circuit and activated a solenoid valve that dispensed about 0.60 ml of liquid from the gravity fed drinking spout. Drugs were placed onto nichrome wire coils and inserted in the back of the smoking spout. Subjects could self-administer cocaine base following completion of a chained schedule that consisted of lever pressing followed by inhalation responses. Puffs on the smoking spout were detected by a vacuum sensor (Coventry Specialty Corp., Westfield, MA model 505-3) and were indicated to the subject by brief illumination of the cue lights surrounding the smoking spout. The first four puffs required an inhalation of approximately 1 s each in order to be detected by the vacuum sensor. The fifth puff required an inhalation duration of 0.1 s and served to trigger the heating of the drug coil. Using this method, the drug was volatilized at the beginning of the fifth puff and was signaled to the subject by a noticeable ‘click’ (the opening of the solenoid valve) and the illumination of the cue lights surrounding the smoking spout for 15 s. Both stimulus and cue lights associated with the smoking spout were extinguished following the end of this 15 s drug signal period. The actual physical duration of the monkey’s fifth puff typically lasted between 5- 10 s depending upon the individual smoking topography of the monkey. Physiological data and cocaine blood levels resulting from this experimental procedure have been published previously (Carroll et al., 1990) and are comparable to those reported in humans (Fischman et al., 1983; Hatsukami et al., 1990). Detailed descriptions of test chambers and drinking spouts (Henningfield and Meisch, 1976; Meisch and Henningfield, 1977) microcomputers (Carroll et al., 1981) and smoking devices (Carroll et al., 1990; Comer et al., 1994; Mattox and Carroll, 1996) have been published. 2.3. Drugs Cocaine base was dissolved in 95% ethanol ( 100 mgjml) on a weekly basis and stored in an airtight flask at room temperature. Drug coils were prepared by dripping a precise amount of cocaine solution onto the nichrome coil wires which were then permitted to dry at room temperature for at least 24 h to allow adequate ethanol evaporation. The coils were weighed before and after drug application to ensure correct dosing. Saccharin was dissolved (0.03% wt/vol) in tap water and stored in 2000 ml bottles. New solutions were made every other day. Cocaine HCl was obtained from the National Institute on Drug Abuse (Research Triangle Institute, Research Triangle Park. NC). Buprenorphine HCl (0.3 mg:‘ml), as Buprenex”, (Reckitt and Colman

Pharmaceuticals, Richmond, VA) was obtained from the University of Minnesota Hospital Pharmacy. Saccharin was purchased from Sigma (St. Louis, MO) and 95% ethanol was purchased from the University of Minnesota Chemical Storehouse. 2.4. Procedure

Monkeys were initially trained to self-administer cocaine base (1 .O mg/kg per coil) by requiring an inhalation response of 1 (FRl). This unit-dose of cocaine was selected as it falls near the peak of the dose-response curve (Carroll et al., 1990) and would permit comparison to previous work (Comer et al., 1994). When the self-administration behavior was reliably established, the inhalation requirement was gradually increased and then the lever press response was introduced and increased in a similar fashion. Further details of the training procedure have previously been reported (Carroll et al., 1990). Under training conditions the lever press requirement was increased by doubling the FR value (e.g. 1. 2, 4, 8, 16, 32, 64) until behavior was reliable and stable at FR 64. Experimental data were collected 7 days per week. Each day began with a 1 h time-out at 07:30 h when all chamber lights were extinguished and all responding had no programmed consequences. During this timeout overnight water consumption was measured and monkey chambers were prepared for the upcoming session by filling reservoirs with water and loading drug coils into the smoking spouts. A 4 h session (08:3012:30 h) immediately followed the time-out. Each daily session was composed of a maximum of ten trials. To gain access to a smoke delivery, each subject had to complete a chained operant schedule that contained both a lever and inhalation response requirement. Initiation of the lever pressing phase was indicated to the subjects by illumination of the red light over the primate lever. When the lever requirement was completed, the red light was extinguished and the flashing green stimulus light over the smoking spout was illuminated. Each trial ended with either a successful drug delivery or with a trial termination (failure to complete schedule requirements in the 30 min period). Each successfully completed trial was followed by a 15 min intertrial interval (ITI) to permit direct-effects of the drug to dissipate, during which time responding on the lever or smoking spout had no programmed consequences. However, subjects were able to respond on the drinking spout to obtain liquid under an FR 1 schedule throughout the ITI. Spent drug coils were replaced during the ITI by the experimenter who entered the colony room during breaks in responding. In contrast, terminated trials were not followed by the 15 min ITI, but immediately progressed into the next trial. Two consecutive trial terminations ended the daily session. If the subject

did not terminate out of session, the session ended after ten smoke deliveries or at 12:30 h. At this time the session water consumption was measured, smoking spouts were cleaned and monkeys were fed. Subjects had accessto water under an FR 1 schedule during the intersession period that began at 12:30 h and lasted until 19:30 h the next day. The 2 x 2 design of the experiment was such that each FR was tested in each treatment condition before progressing to the next condition. Order of initial liquid condition, either water (wat) or saccharin (sacc), was counterbalanced across subjects, and subjects completed the the two pharmacological conditions (no bup vs bup) within the initial liquid component before changing liquids. However, subjects always experienced the no bup condition first. The lever requirements (FR 64, 128, 256, 512 and 1024) were varied across sessions and were presented in a nonsystematic order with retests at selected FRs to ensure levels of responding had not changed. Behavioral stability was defined as a 3-day period with no increasing or decreasing trend in smoke deliveries; furthermore, stability did not vary as a function of treatment condition. 2.5. Buprenorphine pretreatment

Buprenorphine (0.01 mg/kg, i.m.) was injected 30 min before the daily experimental session. Buprenorphine injections were typically administered for a 3-5 day period, of which the final 3 days of stable behavior represent data used in calculating individual and group means. Following completion of a buprenorphine injection series, cocaine-reinforced behavior was allowed to return to pre-injection levels. While some subjects immediately returned to their pre-injection level, subjects were typically permitted at least 7 days before any subsequent buprenorphine injections were administered. 2.6. Concurrent saccharin access

For these tests, saccharin (0.03% wt/vol) was placed in the liquid reservoirs instead of water during the daily 4 h experimental sessions and was available under an FR 1 schedule of reinforcement across the range of cocaine FR levels. The entire range of cocaine FRs were assessedwith concurrent saccharin before progressing to the bup + sacc condition. 2.7. Buprenorphine pretreatment saccharin access

with concurrent

The combination of buprenorphine pretreatment (0.01 mg/kg, i.m. 30 min presession) and the concurrent availability of saccharin (0.03% wt/vol) instead of water was tested across the range of cocaine FR levels.

J.S. Rodcfer et al. /Drug

15

and Alcohol Dependence 4.5 (1997) 21-29

Buprenorphine was typically administered for a 3-5 day period, of which the final 3 days of stable behavior represent data used in calculating individual and group means. 2.8. Data analysis

The mean number of smoke deliveries and liquid (water or saccharin) consumption was based on data from 3 consecutive days from each monkey. The group means and standard errors of the mean are calculated from the individual data from each of the five monkeys. Repeated measures ANOVAs were performed using SuperANOVA (Abacus Concepts, Berkeley, CA). Significance levels for statistical results were set a priori at P < 0.05.

Smoking data were presented in the form of demand functions whereby consumption (drug deliveries) was plotted as a function of drug price (lever FR requirement). Previous studies have demonstrated that demand is a positively decelerating function in which consumption decreasesas price increases (Comer et al., 1994; Hursh and Bauman, 1987; Mattox and Carroll, 1996; Rawleigh et al., 1996).

phine condition there was a significant main effect of cocaine FR (FC4.,6j= 16.419, P <0.05), but there was not a significant main effect of buprenorphine pretreatment alone (F,,,d, = 2.923, P > 0.05) on cocaine-reinforced behavior. When the combination of the two treatments was compared to the no treatment baseline there was a significant main effect of cocaine FR on cocaine smoke deliveries (FC4.,hJ= 21.723, P < 0.05). Examination of the bup + sacc treatment condition revealed a significant decreasein cocaine smoke deliveries (F,,.,, = 9.345. P < 0.05) compared to the no treatment condition that was evident across all but the largest cocaine FR. Furthermore, there was a significant interaction (F,,,,,, = 3.140, P < 0.05) between the main effects of cocaine FR and treatment. Each of the treatments, buprenorphine or saccharin, was then compared to the combined bup + sacc treatment in order to assess relative treatment’ efficacy. When the bup treatment (with water) was compared to u -A-

loo0

Water Water + 0.01 BIJP

= 1

3. Results 3.1. Liquid intake 1

Liquid intake is presented in Fig. 1 as a function of the treatment conditions. An ANOVA comparing saccharin and water intake (without buprenorphine) indicated that signficantly more saccharin was consumed V’w,, = 8.18; P < 0.05) suggesting saccharin functioned as a reinforcer in this protocol. Buprenorphine treatment did not significantly reduce either water (F,,,a, = 7.209, P > 0.05) or saccharin intake from the no treatment condition (FC1,4j= 6.988, P > 0.05). In addition, there was no significant effect of the cocaine base FR value on liquid intake under any treatment condition (P > 0.05).

‘j

a

,,,p 10

.,__ ~ 100

.-5 t2 c% ui

1llBB

-o-A-

loo0

1WBB

sacchuin Sacchain+O.OlBup

1

3.2. Smoked cocaine self-administration

The mean number of smoked cocaine deliveries in each condition significantly decreased in a negatively accelerating manner as a function of cocaine FR (Fig. 2). There was a significant main effect of FR such that increases in cocaine FR resulted in decreased smoked cocaine deliveries across both liquid conditions (F,,,,, = 21.108, P < 0.05). However, the addition of saccharin did not significantly change the number of smoke deliveries compared to the no treatment condition (FC,,4j= 0.002, P > 0.05). Similarly, when the no treatment condition was compared to the buprenor-

Cocaine

FR

Fig. 1. Mean ( i S.E.) liquid intake is plotted as a function of the FR requirement or price of cocaine base smoke deliveries and by treatment condition. The ‘top frame’ displays water intake with and without buprenorphine pretreatment. The ‘bottom frame’ displays saccharin intake with and without buprenorphine pretreatment. The baseline or no treatment condition is indicated by ‘open circles’. The effect of buprenorphine (0.01 mg/kg) on water intake is indicated by the ‘open triangles’. Saccharin intake prior to buprenorphine treatment is shown in the ‘filled circles’ and saccharin intake during buprenorphine treatment is represented by the ‘tilled triangles’. Each point represents a mean of five monkeys. Each monkey’s data point was a mean of the last 3 days of stable behavior.

Cocaine

FR

Fig. 2. Mean ( + S.E.) of cocaine-base smoke deliveries are plotted as a function of the FR lever press schedule requirement or price. The left panel compares the no treatment condition (open circles and dashed line) with the condition in which saccharin was concurrently available as a nondrug alternative reinforcer (filled circles and solid line). The center panel compares the no treatment baseline condition (open circles and dashed line) with the condition in which buprenorphine injections (0.01 mg/kg) were given prior to the smoking sessions(filled squares and solid line). The right panel compares the no treatment baseline condition (open circles and dashed line) with the combined saccharin and buprenorphine treatment condition (filled triangles and solid line). Each point is a mean five monkeys and the data point for each monkey consists of the mean of the last 3 days of stable behavior at each condition

the combined treatment of bup + sacc, there was again a significant main effect of cocaine FR (F,,,,,, = 10.347, P < 0.05) however no significant main effect of saccharin (F,,.,, = 2.499, P > 0.05) was found. In the alternative comparison, sacc versus bup + sacc, a significant main effect of cocaine FR was also found (F,,,,, = 18.64, P < 0.05). Moreover, there was a significant main effect of buprenorphine treatment (F,,.,, = 8.792, P < 0.05) when saccharin was concurrently available. The effect of the bup, sacc, and bup + sacc treatments at each cocaine FR as a function of treatment day are plotted in Fig. 3. The functions are based on group mean data (error bars removed for clarity) and show that saccharin alone had almost no effect on cocaine smoke deliveries, except at FR 256. The 21% reduction in smoking deliveries observed at FR 256 was largely a function of two subjects (M-M4 and M-R) that demonstrated a large decrease in cocaine smoke deliveries from FR 128 to 256. In contrast, the effect of bup alone was small but consistent across the range of cocaine FRs. The relative lack of any decrease in cocaine smoke deliveries at FR 1024 is due to the fact that almost no deliveries were taken at this FR, a floor effect. The reductions in smoking produced by the combined bup + sacc treatment were consistently greater than with bup alone, and this effect was more pronounced at FR 256 than at either lower or higher FRs.

4. Discussion Increasing

the FR requirements

for cocaine base

resulted in positively decelerating decrease in cocaine smoke deliveries and the rate of responding increased with FR (not shown for present experiment; see Mattox and Carroll, 1996). These data are compatible with previously published behavioral economic studies (Comer et al., 1994; Mattox and Carroll, 1996) and

FR64

FR128

0

12

FR255

3

FR512

FRl024 10 8 6 4 2 0L 0

12

3

Treatment Day Fig. 3. Mean cocaine-base smoke deliveries at FR 64-1024 are plotted as a function of the 3 days of data for each of the treatment conditions. The water baseline, or no treatment condition is indicated by ‘open circles’. Data from when saccharin was concurrently available are indicated by ‘closed circles’. Pretreatment with 0.01 mg/kg buprenorphine is indicated by ‘open triangles’ and the combined treatment of concurrent saccharin and buprenorphine pretreatment is indicated by ‘closed triangles’. Standard error bars are omitted for reasons of clarity.

J.S. Rod&r

ct ul. II Drug und Alcol~ol Deptwitwe

support the notion that this smoking procedure in rhesus monkeys is a viable model of drug-taking behavior. One limitation of this protocol is that the behavior required to obtain reinforcement increases more than loo-fold over the duration of the experiment. Thus, a subject that responded at a constant rate of 0.56 responses/s might obtain all possible deliveries between FR 64-512 but none at the FR 1024 level. However, the observation that few deliveries of cocaine smoke were obtained at the highest FR is not a result of the subjects not being able to complete the response requirement. In previous work, subjects demonstrated the ability to respond at a rate of 2.28 responses/set and achieve multiple deliveries at FR 4096 during a daily session (Carroll et al., 1990). Thus, the observed decreases in deliveries at higher FRs in this and previous studies are more likely associated with increased response latencies or pauses than with inadequate response rates. Substituting saccharin for water as the concurrently available liquid resulted in greater intake in milliliters (than water) increased responding for (not shown) and delivery of liquid (Fig. 1) indicating that saccharin was functioning as a reinforcer in this protocol. However, the presence of concurrent saccharin had a negligible effect on cocaine intake across the range of cocaine FRs examined. Comer and colleagues (1994) documented this same effect using a similar procedure. However, previous studies have reported marked decreases in drug consumption in rhesus monkeys that self-administered oral PCP (Carroll and Rodefer, 1993) and ethanol when saccharin replaced concurrently available water (Carroll et al., 1995). Similar decreases in drug-reinforced responding as a result of access to sweetened liquids have been reported in rats that selfadministered i.v. cocaine (Carroll and Lac, 1993; Carroll et al., 1989; Comer et al., 1996). This evidence suggests that the efficacy of certain alternative reinforcers in drug treatment may depend upon the form of the drug that is functioning as a reinforcer and/or the route of administration. One way that alternative reinforcers such as saccharin can contribute to decreased drug-reinforced behavior is by functioning as an economic substitute; consumption of the alternative reinforcer increases as the consumption of drug decreases. Some reports have suggested that concurrently available saccharin can serve as an economic substitute for cocaine. In a similar protocol, Comer and coworkers (1994) demonstrated decreased smoked cocaine consumption as a function of cocaine FR. Moreover, saccharin intake (available concurrently under FR 1 schedule of reinforcement) increased with corresponding decreases in cocaine-reinforced behavior. However, only a small but nonsignificant increase in saccharin intake was observed in the present study. This difference may in part be due to the intrasubject variability in responding for saccharin.

4.5 (19Y7) 21 24,

11

Under buprenorphine pretreatment, levels of both saccharin and water intake decreased across most of the cocaine FR range, but these results did not reach statistical significance. Likewise, buprenorphine (0.01 mg/kg) pretreatment did not significantly decrease intake of cocaine smoke across the range of cocaine FRs (Figs. 2 and 3). This result is consistent with the finding of the Comer et al. (1994) report of no significant decrease in cocaine intake following buprenorphine (0.01 mg/kg) pretreatment when group data were analyzed. However, Comer and coworkers did note a significant decrease in cocaine intake at the 0.01 mg/kg buprenorphine dose in some of the individual data. The 0.01 mg/kg dose of buprenorphine may represent a threshold therapeutic dose, or its effects may not emerge under certain methodological conditions (i.e. cocaine smoking). Rawleigh et al. (1996) demonstrated a significant decrease in oral intake of both PCP and saccharin when monkeys had been pretreated with 0.005 mgikg buprenorphine. In addition, Carroll and colleagues (1992) observed decreases in cocaine smokereinforced behavior at even lower doses (0.003 mg/kg) of buprenorphine, under a progressive-ratio schedule of reinforcement. These varied effects of buprenorphine on drug-reinforced behavior have been observed previously (Carroll et al., 1992; Mello et al., 1990) and appear to be a function of schedule requirements and route of drug administration. While neither the behavioral or pharmacological treatments individually produced significant decreases in cocaine-reinforced behavior, the bup + sacc treatment produced a significant decrease in deliveries of cocaine smoke. This effect was observed over a wide range of response requirements but the decrease due to the bup + sacc was much less than has been previously reported when drugs were either self-administered through the oral (Carroll and Rodefer, 1993) or iv. routes of administration (Carroll et al., 1989). Furthermore, an asymmetrical treatment effect was observed when the combined treatment was compared to the two individual treatments. The lack of significant difference between bup and bup + sacc conditions suggests that the addition of saccharin did not decrease cocaine-reinforced behavior any more than was already resulting from buprenorphine pretreatment. However, the significant finding between the sacc and bup + sacc conditions suggests that the addition of buprenorphine served to decrease (or further decrease) cocaine-reinforced behavior that was slightly suppressed by concurrent saccharin. Thus, a pharmacological reinforcer may have greater impact following the addition of alternative reinforcers into a treatment program rather than before they are introduced. However, a limitation of the current study was the use of only one parameter for each of the two treatments examined. As such. it it not possible to adequately determine whether pretreatment

2x

J.S. Rodejer rt ul. IIDrug und Alcohol Dependence 45 (1997) 21-29

with buprenorphine potentiated or antagonized any effects of cocaine with concurrent saccharin. The lack of decrease in either saccharin or water intake due to bup treatment was not unexpected, as the dose of buprenorphine was one that has previously been demonstrated to have little effect on cocaine smoke- or saccharin-reinforced behavior (Comer et al., 1994). However this result is promising as it demonstrates that buprenorphine did not suppress behavior in a non-selective fashion, a goal for any pharmacological treatment. This finding might resemble results from a report on buprenorphine’s effect on food- and cocainemaintained responding (Lukas et al., 1995). Lukas and colleagues reported that buprenorphine pretreatment reduced i.v. cocaine intake at doses that did not alter food-maintained responding. In summary, increasing FR values decreased cocaine smoke deliveries, and the addition of either saccharin or pretreatment with buprenorphine did not modify the ability of increasing FR to decrease cocaine smoke deliveries. However, the combination of bup + sacc treatment significantly decreasedcocaine smoke deliveries. While previous reports have found buprenorphine to have a mixed ability to decrease drug-reinforced behavior, buprenorphine appears to be a more successful agent when the abused drug efficacy is low (i.e. either with low doses of the abused drug or routes of administration that have a less immediate onset of action) rather than with a highly efficacious procedure (e.g. higher doses or smoke self-administration). When combined with the present results, this suggests that individuals that abuse drugs through the i.v. or smoked routes of administration may be better served with a combined behavioral and pharmacological treatment paradigm.

Acknowledgements The authors gratefully acknowledge the technical assistance of Vincent Hunt. This research was supported by NIDA Grants ROl DA07716, PO1 DA08131 and T 32 DA07097. Some of these results were presented at the annual meeting of the College on Problems of Drug Dependence, Scottsdale, AZ, 1995.

References Carroll, M.E. (1985) Concurrent phencyclidine and saccharin access: presentation of an alternative reinforcer reduces drug intake. J. Exp. Anal. Behav. 43, 131-44. Carroll, M.E. and Carmona, G. (1991) Effects of food FR and food deprivation on disruptions in food-maintained performance of monkeys during phencyclidine withdrawal. Psychopharmacology 104, 143-9.

Carroll, M.E., Carmona, G.N., May, S.A., Buzalsky, S. and Larson, C. (1992) Buprenorphine’s effects on self-administration of smoked cocaine base and orally delivered phencyclidine, ethanol and saccharin in rhesus monkeys. J. Pharmacol. Exp. Ther. 261. 2631. Carroll, M.E., Krattiger, K.L., Gieske, D. and Sadoff, D.A. ( 1990) Cocaine-base smoking in rhesus monkeys: reinforcing and physiological effects. Psychopharmacology 102, 443 -50. Carroll, M.E. and Lac, ST. (1992) Effects of buprenorphine on self-administration of cocaine and a nondrug reinforcer in rats. Psychopharmacology 106, 439- 46. Carroll, M.E. and Lac, S.T. (1993) Autoshaping i.v. cocaine self-administration in rats: effects of nondrug alternative reinforcers on acquisition, Psychopharmacology 110, 5-~-12. Carroll, M.E., Lac, ST. and Nygaard, S.L. (1989) A concurrently available nondrug reinforcer prevents the acquisition or decreases the maintenance of cocaine-reinforced behavior. Psychopharmacology 97, 23-9. Carroll, M.E. and Rodefer, J.S. (1993) Income alters choice between drug and an alternative nondrug reinforcer in monkeys. Exp. Clin. Psychopharmacol. 1, 110 120. Carroll, M.E., Rodefer, J.S. and Rawleigh, J.M. (1995) Concurrent self-administration of ethanol and an alternative nondrug reinforcer in monkeys: effects of income (session length) on demand for drug. Psychopharmacology 120, l-9. Carroll, M.E., Santi, P.A. and Rudell, R.L. (1981) A microcomputer system for the control of behavioral experiments. Pharmacol. Biochem. Behav. 14, 4155417. Comer, S.D., Hunt, V.R. and Carroll, M.E. (1994) Effects of concurrent saccharin availability and buprenorphine pretreatment on demand for smoked cocaine base in rhesus monkeys. Psychopharmacology 115, 15-23. Comer, S.D., Lac, S.T., Wyvell, C.L. and Carroll, M.E. (1996) Combined effects of buprenorphine and a nondrug reinforcer on iv. cocaine self-administration in rats maintained under FR schedules. Psychopharmacology 125, 355-360. Compton, P.A., Ling, W., Charuvastra, V.C. and Wesson, D.R. (1995) Buprenorphine as a pharmacotherapy for cocaine abuse: a review of the evidence. J. Addict. Dis. 14, 97 114. Fischman, M.W., Schuster, C.R. and Hatano, Y. (1983) A comparison of the subjective and cardiovascular effects of cocaine and lidocaine in humans. Pharmacol. Biochem. Behav. 18, 123-7. Hatsukami, D.. Keenan, R., Carroll, M., Colon, E., Geiske, D., Wilson, B. and Huber, M. (1990) A method for delivery of precise doses of smoked cocaine-base to humans. Pharmacol. Biochem. Behav. 36, l-7. Henningfield, J.E. and Meisch, R.A. (1976) Drinking device for rhesus monkeys. Pharmacol. Biochem. Behav. 4, 609-10. Higgins, ST., Budney, A.J., Bickel, W.K., Foerg, F.E., Donham, R., and Badger, G.J. (1994) Incentives improve outcome in outpatient behavioral treatment of cocaine dependence. Arch. Gen. Psychiatry 51. 568-76. Higgins, S.T., Budney, A.J., Bickel, W.K., Hughes, J.R., Foerg, F. and Badger, G. (1993) Achieving cocaine abstinence with a behavioral approach. Am. J. Psychiatry 150, 763-9. Hursh, S.R. and Bauman, R.A. (1987) The behavioral analysis of demand. In: Advances in Behavioral Economics (Green. L. and Kagel, J.H., eds.). Ablex, Norwood, NJ. Kanarek, R.B. and Marks-Kaufman, R. (1988) Dietary modulation of oral amphetamine intake in rats. Physiol. Behav. 44, 501-5. Kosten, T.R., Rosen, M.I., Schottenfeld, R. and Ziedonis, D. (1992) Buprenorphine for cocaine and opiate dependence. Psychopharmacol. Bull. 28, 15-9. Lukas, S.E., Mello, N.K., Drieze, J.M. and Mendelson, J.H. (1995) Buprenorphine-induced alterations of cocaine’s reinforcing effects in rhesus monkey: a dose-response analysis. Drug Alcohol Depend. 40. 87- 98.

J.S. Rodefer

et al. /Drug

and Alcohol

Mattox, A.J. and Carroll, M.E. (1996) Smoked heroin self-administration in rhesus monkeys. Psychopharmacology 125, 195-201. Meisch, R.A. and Henningfield, J.E. (1977) Drinking of ethanol by rhesus monkeys: experimental strategies for establishing ethanol as a reinforcer. Adv. Exp. Med. Biol. 85B, 443-63. Mello, N.K., Bree, M.P. and Mendelson, J.H. (1983) Comparison of buprenorphine and methadone effects on opiate self-administration in primates. J. Pharmacol. Exp. Ther. 254, 926-939. Mello. N.K., Mendelson, J.H., Bree, M.P. and Lukas, SE. (1989) Buprenorphine suppresses cocaine self-administration by rhesus monkeys. Science 245, 859962. Mello, N.K., Meldelson, J.H., Bree, M.P. and Lucas, SE. (1990) Buprenorphine and naltrexone effects on cocaine self-administration by rhesus monkeys. J. Pharmacol. Exp. Ther. 254, 9266939. Mello, N.K. and Negus, S.S. (1996) Preclinical evaluation of pharmacotherapies for treatment of cocaine and opioid abuse using drug self-administration procedures [Review]. Neuropsychopharmacology 14, 375-424. Nader, M.A. and Woolverton, W.L. (1991) Effects of increasing the magnitude of an alternative reinforcer on drug choice in a discrete-trials choice procedure. Psychopharmacology 105, 16974. National Institute on Drug Abuse (1995) Drugs and Drug Abuse Education Newsletter. XXVI, Office of National Drug

Dependence

45 (1997)

21-29

29

Control Policy (1995) National trends in drug abuse. Pulse Check. 28. Rawleigh, J.M., Rodefer, J.S., Hansen, J.J. and Carroll, M.E. (1996) Combined effects of buprenorphine and an alternative nondrug reinforcer on phencyclidine self-administration in rhesus monkeys. Exp. Chn. Psychopharmacol. 4, 68-76. Schottenfeld, R.S., Pakes, J., Ziedonis, D. and Kosten, T.R. (1993) Buprenorphine: Dose-related effects on cocaine and opioid use in cocaine-abusing opioid-dependent humans. Biol. Psychiatry 34, 66-74. Silverman, K., Higgins, S. T., Brooner, R.K., Montoya, I.D., Cone, E.J., Schuster, C.R. and Preston, K.L. (1996) Sustained cocaine abstinence in methadone maintenance patients through voucherbased reinforcement therapy. Arch. Gen. Psychiatry 53, 409-15. Strain, E.C., Stitzer, M.L., Liebson, LA. and Bigelow, G.E. (1994) Buprenorphine versus methadone in the treatment of opioid-dependent cocaine users. Psychopharmacology 116, 401-6. USDHHS (1995) Annual Medical Examiner Data, 1993b. Drug Abuse Warning Network (DAWN). Series 1, USDHHS Publication No. (SMA) 95-3019. Winger, G., Skjoldager, P. and Woods, J.H. (1992) Effects of buprenorphine and other opioid agonists and antagonists on alfentanil- and cocaine-reinforced responding in rhesus monkeys. J. Pharmacol. Exp. Ther. 261, 31 l-7.