Secondary Logo

Journal Logo

Intravenous self-administration of heroin/cocaine combinations (speedball) using nose-poke or lever-press operant responding in mice

David, V.; Polis, I.; McDonald, J.; Gold, L.H.

Research Papers

Acquisition and dose-related self-administration of heroin (H)/cocaine (C) combinations in C57BL/6 × SJL mice were studied in nose-poke or lever-press operant responding procedures. C57BL/6 × SJL mice readily acquired self-administration in both operant procedures with a combination of doses known to be ineffective when each drug was used alone (H:15 μg/kg and C:150 μg/kg per injection). Similar numbers of infusions were obtained under conditions of fixed-ratio (FR) 3 versus 1 for the nose-poke and lever-press responses, respectively. Dose–effect curves for heroin:cocaine combinations revealed a pattern corresponding to a leftward shift of the dose-response for intravenous cocaine self-administration. Curves were similar whether generated with 1 or 3 days of availability per dose, or including subjects that did not respond preferentially (>70% responses) to the hole or lever associated with drug delivery, along with those that did. Motor activity induced by a combination of low doses for each drug was examined (H: 0.375 mg/kg and C: 3.75 mg/kg, i.p.). Under these conditions, the combination of both drugs induced an initial cocaine-like stimulation of horizontal activity, in contrast to the tendency of heroin to decrease activity. It is concluded that heroin:cocaine combinations used in the present study had reinforcing effects in C57BL/6 × SJL mice, and produced a cocaine-like effect in the early part of drug-induced activity sessions, followed by a locomotor profile corresponding to the average of both drugs.

Department of Neuropharmacology CVN-7, The Scripps Research Institute, La Jolla, USA

Correspondence to L. Gold, CNS/CV Biological Research, Schering-Plough Research Institute, K-15-2 MS2600, 2015 Galloping Hill Rd, Kenilworth, NJ 07033, USA. E-mail:

Received 4 February 2000

Revised 23 October 2000 accepted 23 October 2000

Back to Top | Article Outline


In many cases, drug abuse would be better described as polydrug abuse. Cocaine use by opiate abusers and methadone-maintained patients has increased since the 1980s, and concurrent cocaine/heroin use, also known as ‘speedballs’, is now a predominant form of polydrug abuse (Kosten et al., 1986; Greberman and Wada, 1994; National Institute on Drug Abuse, 1998). Based on clinical observations, several non-exclusive hypotheses have been proposed to explain abuse of cocaine/opiate mixtures. It has been postulated that compared to either drug alone, the drug combination may produce enhanced or prolonged euphoric effects, decreased negative side-effects or an entirely unique effect (Ellinwood et al., 1976; Kosten et al., 1986; Foltin and Fischman, 1992; Tutton and Crayton, 1993). Despite the increased popularity of speedball abuse, the neuropharmacological effects of the simultaneous injection of cocaine and heroin in animal models have received less attention.

In preclinical studies, self-administration of cocaine/heroin combinations has been reported in rhesus monkeys (Mello et al., 1995; Rowlett and Woolverton, 1997), rats and mice using fixed- (FR) or progressive-ratio (PR) schedules of reinforcement (Hemby et al., 1996; Roberts et al., 1997; Duvauchelle et al., 1998; Ranaldi and Munn, 1998). Mello et al. (1995) have shown that combination of high heroin doses with non-reinforcing doses of cocaine increased self-administration of cocaine in rhesus monkeys, as measured by rate of intake. In contrast, using the PR paradigm in the same species, Rowlett and Woolverton (1997) did not observe increased breaking points for cocaine self-administration when heroin was added. However, they found that non-reinforcing doses of cocaine maintained self-administration after the addition of low doses of heroin. When non-reinforcing doses of cocaine were combined with heroin, the heroin dose–response function was shifted to the left, and naltrexone dose-dependently antagonized self-administration of heroin and speedball, but not cocaine (Rowlett et al., 1998). More recently, the dose–effect curve for speedball self-administration under a second-order schedule of reinforcement was shifted to the right following treatment with a mixture of dopamine–opioid antagonists (flupenthixol–quadazocine:Mello and Negus, 1999).

In rats, a heroin/cocaine mixture produced a leftward shift of the cocaine dose–response curve in a self-administration study, using an FR schedule, and this effect was shown to be sensitive to systemic injections of naltrexone (Hemby et al., 1996). Studies using PR schedules have indicated increased breaking points for cocaine/heroin combinations and, consistent with studies of FR responding, a leftward shift of the cocaine dose–response curve was produced by the addition of heroin (Duvauchelle et al., 1998; Ranaldi and Munn, 1998). Recently, Hemby et al. (1999) have reported that self-administration of a cocaine/heroin combination potentiated the extracellular dopamine release in the nucleus accumbens of rats induced by intravenous cocaine, but not heroin, self-administration. However, this synergistic effect was not associated with any change in the pattern of responding.

Compared with these other laboratory species, self-administration of drug combinations in mice has been less well studied. Roberts et al. (1997) showed that the addition of heroin to cocaine led to self-administration in BALB/cByJ mice that did not exhibit this behavior under conditions of cocaine availability alone. However, self-administration of the combination was obtained with a dose of heroin maintaining self-administration when used alone in the BALB/cByJ strain, and drug intake remained low. Based on the high emotional reactivity of BALB/cByJ mice, these results suggest that a negative reinforcement process (removal or reduction of cocaine aversive effects) may underlie intravenous self-administration of some heroin:cocaine combinations.

Because the mouse self-administration paradigm has great potential for studying genetic influences on behavior, further characterization of self-administration of heroin/cocaine mixtures in mice is warranted. Extending the prior work from BALB/c mice to a common background strain used to create transgenic mice, such as the hybrid C57BL/6 × SJL, will help to provide the necessary background information for the future study of the neuropharmacological effects of speedball in genetically modified animals. Specifically, the current studies were designed to assess further the reinforcing effects of heroin/cocaine combinations by studying the acquisition and dose-related responding for intravenous self-administration in C57BL/6 × SJL mice. To test whether a particular operant response would influence acquisition of speedball self-administration in mice, independent groups of animals were compared, using nose-poke or lever-press manipulanda. Additionally, in the same strain of mice, the unconditioned motor response to low doses of each drug alone, or in combination, was tested.

Back to Top | Article Outline


2.1. Subjects

Adult male F 1 hybrid mice from a C57BL/6 × SJL cross were obtained from The Scripps Research Institute breeding colony, and housed 2–4 mice/cage in Plexiglas cages (28 × 17 × 11.5 cm). Animals were then housed singly in these cages 1 week prior to beginning the intravenous self-administration experiments and throughout the experiments, in order to protect catheterized subjects. Colony temperature was maintained at 24°C. Subjects were provided with free access to food and maintained under conditions of a reverse light/dark cycle (lights on at 22:00 hours, lights off at 10:00 hours). The animals were aged 3–6 months (25–30 g) at the beginning of the experiments.

Back to Top | Article Outline

2.2. Apparatus

2.2.1. Catheter construction

The chronic intravenous catheter for mice was similar to that described previously (Deroche et al., 1997). Briefly, one end of a 7-cm length of soft Silastic tubing (i.d. 0.30 mm, o.d. 0.64 mm) was immersed in Hemo-De solvent (Fisher Scientific, Pittsburgh, PA, USA), to expand the tip for fitting to a 22-gauge steel cannula with a plastic screw-threaded pedestal (collar), and bent at a right angle. The cannula was then encased in dental cement with a piece of durable mesh secured to the bottom. The catheters were immersed in 70% ethanol for 20 min and then washed in 0.9% bacteriostatic saline, prior to the surgical implantation. After implantation, the exposed tip of the external cannula was secured with a stopper consisting of a small piece of removable Tygon tubing (0.5 mm i.d./1.5 mm o.d.), closed at one end with monofilament and encased in slightly larger copper tubing.

Back to Top | Article Outline

2.2.2. Intravenous self-administration unit

Twelve Plexiglas chambers (14.9 × 15.2 × 18.3 cm each, constructed in-house) were located within larger boxes (Coleman coolers) equipped with exhaust fans that also functioned to mask background noise. Each chamber contained one wall with either two small holes (0.9 cm diameter, 4.2 cm apart, 1.5 cm from the floor) or two levers (4.2 cm apart, 1.5 cm from the floor) equipped with photocell beams to detect responses. One hole or lever was defined as active, in a counterbalanced right and left order across subjects. When the photocell in the active hole was triggered by a nose-poke, or when the active lever was depressed (requiring a force of 5 ± 1 g), the infusion pump (Razel syringe pump with a 5 r.p.m. motor) delivered an infusion of 25 μl/5 s from a 10 ml syringe. Responses to the hole or lever defined as inactive, were without scheduled consequence at any time. The syringe was connected via Tygon tubing to a swivel (Instech, Plymouth Meeting, PA, USA). The infusion system was mounted outside the exterior box (cooler).

Back to Top | Article Outline

2.2.3. Motor activity

Locomotor activity was measured in large Plexiglas cages placed into frames (29.2 × 50.5 cm) mounted with photocell beams (San Diego Instruments, San Diego, CA, USA). The horizontal locomotion frames consisted of a 4 × 8 array of beams. Zone entries, defined as movement into one of eight equal-sized squares, rather than the repeated breaking of a single beam, reflected horizontal locomotion. A second-tier frame (7 cm in height), consisting of eight beams equally spaced along the long axis, was used to record vertical activity (rearing).

Back to Top | Article Outline

2.3. Catheter implantation surgery

The mouse was anesthetized with a halothane/oxygen vapor mixture and maintained at approximately 1.5% halothane, delivered via a miniature nose-cone with tubes for vapor entry and exhaust to a ventilation system. Throughout the procedure, the animal was supported by a heating pad to prevent hypothermia. After shaving and application of 70% alcohol and iodine preparatory solution, incisions were made in the midscapular region, as well as anteromedial to the forearm above the external right jugular vein. A catheter was passed subcutaneously from the dorsal incision to the ventral incision. Following isolation of the jugular vein, a 23-gauge needle was inserted into the vein 0.5 cm above the pectoral muscle, to facilitate insertion of a 22-gauge modified needle that had been filed to be used as a guide for the Silastic tubing. Once the tubing was inserted approximately 0.6 cm into the vein, the modified needle was removed and the tubing was further inserted to the level of a small mark (1.2 cm from the open tip of the Silastic). The catheter was then tied gently to the vein with two sutures (surgical silk 4-0; Ethicon). Approximately 0.01 ml of physiological saline was flushed through the catheter to displace the blood and to sustain catheter patency by avoiding clotting. The catheter was then capped with a Tygon stopper. Animals were allowed a minimum of 5 days’ recovery before the beginning of self-administration testing.

Back to Top | Article Outline

2.3.1. Catheter evaluation and maintenance

Methohexital sodium is an ultra-short-acting barbiturate that, when flushed through a patent catheter, produces overt signs of sedation within seconds. Aqueous 1% methohexital sodium was dissolved in 0.9% bacteriostatic saline to obtain a concentration of 5 mg/ml. A methohexital test was performed once prior to the first self-administration session, and once after every five self-administration sessions thereafter. At least 4 h after a cocaine self-administration session, or 12 h prior to the next session, approximately 0.02 ml of the solution was flushed through the catheter. Animals that showed no immediate signs of sedation were removed from the experiment.

Back to Top | Article Outline

2.4. Intravenous self-administration procedure

2.4.1. Acquisition study

Self-administration sessions (one/day; 2 h each, 5 days/week) were conducted between 12:00 and 18:00 hours. During the first 2 h session, the animals were not connected to the injection system (chamber habituation). Starting with the second session, drug solutions were available under a fixed-ratio 1 (FR 1) schedule, with a 20 s time-out during which responses, nose-pokes or lever-presses, were recorded but not reinforced. Mice were given access to a unit dose of heroin (15 μg/ kg per injection) and cocaine (150 μg/kg per injection) (training dose). The choice of a 1:10 ratio was based on clinical observations and preclinical studies showing that heroin was approximately ten times more potent than cocaine in inducing rate-decreasing effects (Mello et al., 1995). The choice of unit dose of heroin (15 μg/kg/injection) and cocaine (150 μg/kg/injection) was based on previous observations that cocaine supported acquisition of self-administration at a dose of 1 but not 0.25 mg/kg/injection in C57BL/6 × SJL mice (Deroche et al., 1997). In pilot experiments, heroin alone at the dose of 15 μg/kg/injection did not maintain self-administration in the C57BL/6 × SJL strain.

To compare operant responding for drug versus vehicle, two additional groups of animals (nose-poke, n  = 9; and lever-press, n  = 10) were given the opportunity to self-administer saline during 10 daily sessions.

The FR was raised from 1 to 3 on the second day of self-administration training in the nose-poke group responding for drug. A FR1 schedule was maintained throughout the study for all other groups. The acquisition phase of self-administration was studied for 10 consecutive sessions. Data were evaluated for two performance variables: (1) stability of intake, measured as three consecutive self-administration sessions in which the total number of injections (total cocaine intake) remained constant within 20% deviation of the mean of these sessions (stable responding or baseline); and (2) percent responses on the active manipulandum, defined as nose-pokes in the active holes or presses on the active lever/total responses, calculated over 3 consecutive days.

Back to Top | Article Outline

2.4.2. Dose–effect study

Following the acquisition phase, doses of speedball were varied such that the heroin/cocaine ratio (1:10) remained constant (7.5/75; 15/150; 30/300; 60/600 μg/kg/injection). The order of dose-substitutions, including training dose, was randomized among animals. Each dose was tested for three consecutive sessions. Saline was substituted at the end of the dose–effect study to minimize the potential influence of extinction (Deroche et al., 1997; Rocha et al., 1997; Rocha, 1999). Subjects were put into the dose–response study if they exhibited stable responding as defined above. Subjects demonstrating greater than 70% percent responses on the active manipulandum were compared with those that did not.

Back to Top | Article Outline

2.5. Locomotor activity

Locomotor activity testing was also conducted during the dark portion of the light/dark cycle. Mice were brought into the motor activity room 1 h prior to the start of testing. After a 2-h habituation period to the motor activity cages, the mice were injected with either saline, heroin (0.375 mg/kg), cocaine (3.75 mg/kg), or a combination of both (speedball), in a between-subjects design, n  = 8 mice/group. These doses of cocaine and heroin were expected to stimulate activity minimally (Deroche et al., 1997). Drug-induced locomotor activity was recorded for an additional 2 h. All injections were administered i.p. in a volume of 1 ml/100 g body weight.

Back to Top | Article Outline

2.6. Drugs

Cocaine HCl, 3,6-diacetylmorphine HCl and combinations of cocaine HCl and 3,6-diacetylmorphine HCl (NIDA, Rockville, MD, USA) were dissolved in sterile 0.9% saline. Doses were calculated as μg/kg/injection, based on a body weight of 25 g. Methohexital sodium (Eli Lilly, Indianapolis, IN, USA) was dissolved in sterile 0.9% saline to obtain a concentration of 5 mg/ml.

Back to Top | Article Outline

2.7. Data analysis

Two-way ANOVAs with one repeated factor (session or day for self-administration experiments and minutes for the motor activity study) were used to analyze between-group (operant or drug) differences as well as group × session interactions (Winer, 1971). Significant differences were followed by appropriate post hoc analyses detailed in the results section and figure legends. One-way analyses of variance (ANOVAs) were used for the within-subjects repeated factor day (= session), to analyze the time-course of the number of self-injections for each experiment in each group. Unpaired Student's t-test was used to compare the percent responses on the active manipulandum of a particular session with chance level (50%). A significance level of P  < 0.05 was used for all statistical analyses.

Back to Top | Article Outline


3.1. Intravenous self-administration

3.1.1. Saline-injected animals (

Figure 1A)

Fig. 1

Fig. 1

Subjects trained under conditions of saline availability earned 25 to 50 injections/2-hour session, regardless of the operant response (nose-poke, n  = 10; lever, n  = 9). No subjects achieved stability in the number of injections over the 10 sessions under a FR-1 schedule, in either nose-poking or lever-pressing groups [group effect: F(1,8) = 1.4, NS; session effect: F(9,18) = 1.3, NS; group × session interaction: F(9,162) = 0.8, NS]. A high between-subject variability, as expressed by standard error mean (SEM), was observed in both groups. Mice also did not show evidence of a preference for the active manipulandum. Both nose-poking and lever-pressing animals exhibited around 50% responding on the active hole or lever (chance level) for the 10 sessions of the acquisition phase.

Back to Top | Article Outline

3.1.2. Acquisition of speedball intravenous self-administration (

Figure 1B)

Out of 36 catheterized mice, a total of 33 animals (nose-poke group, n  = 15; lever-press group, n  = 18) entered the study after the first methohexital test. During the first session, in which responding had no scheduled consequences (chamber habituation), the number of responses in the nose-poke group was significantly greater than that of the lever-press group [unpaired Student t-test on the mean total nose-pokes versus lever-presses:t (31) = 3.61, P  < 0.001, data not shown]. Of the 15 subjects in the nose-poke group and 18 in the lever-press group, 10 and 11, respectively, completed the acquisition study. Mice in nose-poke (n  = 10) as well as lever-press groups (n  = 11) exhibited an increase in responding over days and met the criterion for stability by day 10 [one-way ANOVA on the number of injections over the nine training sessions, nose-poke: F(8,72) = 2.3, P  < 0.02; lever-press: F(8,80) = 6.8, P  < 0.001]. The number of infusions earned was similar in both groups under schedules of FR-3 for nose-poking and FR-1 for lever-pressing (no main effect of operant), although lever-pressing animals earned fewer injections in the first 3–4 sessions despite the lower FR requirement [operant × session interaction: F(8,152) = 2.05, P  < 0.05].

Since there was no overall main effect between animals nose-poking or lever-pressing for speedball injections, these groups were pooled and compared with mice responding under conditions of saline availability. Significant group differences were found for the number of injections obtained by saline versus speedball animals over the course of training sessions [main effect of group: F(1,39) = 7, P  < 0.01; session: F(8,39) = 3.1, P  < 0.01; and drug × session interaction: F(8,312) = 2.8, P  < 0.01].

Whereas every subject in both groups exhibited stable responding after 10 training sessions, only 7 and 6 animals, in the nose-poke and lever-press groups respectively, displayed greater than 70% responses to the active manipulandum for 3 days. However, comparison of all animals in the nose-poke or lever-press groups with stable responding (pooled) for speedball versus saline, revealed that drug availability significantly increased the preference for the active manipulandum [main effect of drug on percent responses to the active hole or lever: F(1,39) = 6.8, P <  0.01; no main effect of session or drug × session interaction] (Fig. 1). Moreover, during the first complete session of drug availability, some animals in both the lever-press and nose-poke groups exhibited responding to the active manipulandum at a level significantly higher than chance. Interestingly, during the chamber habituation session, the proportion of responding to the manipulandum subsequently designated as active was not different from chance level in nose-poking [t (10) = 0.8, NS] or lever-pressing subjects [t (10) = 0.4, NS].

Back to Top | Article Outline

3.1.3. Dose–effect study

Dose–effect curves were generated by varying simultaneously the concentrations of heroin and cocaine and keeping the ratio constant (1:10). Under these conditions, substitution during 3 consecutive days of each of the different combinations (7.5/75; 15/150; 30/300; 60/600 μg/kg/injection) resulted in dose-dependent changes in the number of injections, similar in subjects of the nose-poke and lever-press groups. A two-way ANOVA comparing the number of self-injections averaged over each 3-day substitution revealed a main effect of dose [F(3,21) = 38.28, P <  0.001], and confirmed the absence of any effect of the different operant responses. In both groups, the dose-regulation of intravenous self-administration behavior was reflected in a linear monotonic descending curve (Fig. 2). In contrast, the percent responses to the active manipulandum was not sensitive to dose substitution, except following saline substitution. Although 4 out of 10 animals of the lever group did not exhibit greater than 70% responses to the active lever at the end of the acquisition phase, they were included in the dose–effect study on the basis of their response stability. These subjects exhibited a dose-related self-administration profile identical to the subjects responding at greater than 70% to the active lever (data not shown). Therefore, these data were pooled for statistical analyses and graphical presentation of the number of injections (Figure 2A, right). For either manipulandum type, the number of injections earned during the first substitution day versus the mean of the 3 substitution days was identical (Fig. 2). It is of interest to compare the speedball dose–effect curve obtained in the present study with the dose–effect curve for cocaine alone in the same strain (C57BL/6 × SJL) previously reported by Deroche et al. (1997). As can be observed in Fig. 2, addition of heroin to cocaine resulted in a leftward shift of the dose–effect curve.

Fig. 2

Fig. 2

Back to Top | Article Outline

3.2. Locomotor activity

All four groups exhibited a similar habituation within the 2-h session before injections were given, as shown by a significant decrease over time in horizontal activity defined as zone entries [main effect of group: F(3,28) = 0.1, NS; main effect of time: F(11,33) = 119.26, P  < 0.001; no significant interaction] and rearing [main effect of group: F(3,28) = 0.3, NS; main effect of time: F(11,33) = 25.24, P <  0.001; no significant interaction] (Fig. 3

Fig. 3

Fig. 3


Cocaine administration increased locomotion, while heroin decreased locomotion, and the speedball profile corresponded to an average of the cocaine and heroin responses during most of the session (Fig. 3) [main effect of drug on zone entries: F(3,28) = 3.2, P  < 0.05); time: F(11,308) = 2.1, P  < 0.02; and drug × time interaction: F(33,308) = 1.6, P  < 0.01] (Fig. 3). However, during the first 10 min of the session, speedball-injected animals exhibited a stimulated activity similar to that of cocaine-injected mice. While cocaine and speedball induced a fast stimulant effect after only 10 min, heroin actually tended to decrease horizontal (number of zone entries) as well as vertical (number of rears) activity. Time-course analysis of the rearing behavior also revealed a difference across drug groups [main effect of drug: F(3,28) = 3.1, P <  0.05; no significant effect of time, or of drug × time interaction]. Both horizontal activity and rearing were significantly different in cocaine- and heroin-injected animals at the beginning and the end of the session (Fig. 3). Thus, speedball produced a cocaine-like stimulant effect during the early part of the session, followed by a profile that reflected the average of the effect of each drug alone.

Back to Top | Article Outline


The present study is one of several establishing the feasibility of intravenous drug self-administration behavior in mice (Carney et al., 1991; Grahame et al., 1995; Deroche et al., 1997; Roberts et al., 1997; Rocha et al., 1997; Stolerman et al., 1999). Our results demonstrate acquisition and dose-related responding for intravenous self-administration of heroin/cocaine combinations in the C57BL/6 × SJL strain, indicating that the heroin/cocaine combinations used in the present study functioned as a reinforcer in these mice, as has been previously reported for rats and monkeys (Mello et al., 1995; Hemby et al., 1996; Rowlett and Woolverton, 1997; Ranaldi and Munn, 1998).

Acquisition of self-administration was characterized by an initial increase in the number of injections, followed by stable responding over 7–10 days, and was associated with responding to the active manipulandum at a level significantly greater than chance in most subjects. Mice given the opportunity to self-administer speedball exhibited a significant preference for the drug-associated manipulanda as a group (both nose-poke and lever-press), although not all animals displayed this preference. It should be noted that only naive mice were used for the present experiments, whereas previous intravenous drug self-administration studies have often used mice trained first for operant discrimination with a food reinforcer (Rocha et al., 1997; Donny et al., 1998; Rocha, 1999; Stolerman et al., 1999). Naive subjects were used in order to interpret directly operant response rate as being supported by available drugs (Bolles, 1972). In the absence of prior discrimination training, an association between the operant response and the speedball-induced interoceptive cues may be difficult in some subjects. Moreover, in our studies no conditional stimuli were associated with drug availability, and nothing prevented the animals from responding to the inactive manipulandum during the time-out period. Therefore, complex responses involving one or the other manipulandum may have been established. In any case, most of the subjects exhibited increased responding to the active manipulandum when drug was available, and this preference was disrupted by saline substitution. Finally, while 3.75 mg/kg cocaine produced greater locomotor stimulation than the speedball combination (H 0.375:C 3.75 mg/kg), the dose–response curve for speedball self-administration was shifted to the left of that for cocaine. These results suggest that speedball-induced locomotor activity alone does not account for the pattern of self-administration observed.

When doses of speedball were varied so that the heroin/cocaine ratio remained constant (1:10), mice exhibited a clear dose-related behavior, similar in lever-press and nose-poke groups. Other than the fact that the number of responses required to maintain the same number of injections was 3 times greater with nose-poking (FR3) than lever-pressing (FR1), the type of operant response did not have a significant effect on the acquisition of intravenous speedball self-administration or dose-related responding. This observation is consistent with a previous report of nose-poke and lever-press operant responding for cocaine (Caine et al., 1999). Interestingly, speedball dose–effect curves were remarkably similar when constructed of the average of 3 substitution days, or only the first substitution day. Stabilization of drug-maintained responding and an inverse relationship between rate and dose are typically observed in self-administration paradigms under limited access conditions, and have been considered to reflect titration of drug exposure (Yokel, 1987).

It is of interest to note that acquisition of speedball self-administration occurred with a dose of cocaine that did not maintain self-administration when used alone. In fact, acquisition of cocaine self-administration in the C57BL/6 × SJL strain of mice was previously observed only with a dose of 1 mg/kg/injection, although acquisition was also tested with a dose of 0.25 mg/kg/injection (Deroche et al., 1997). Moreover, mice exhibited a clear dose-related behavior qualitatively similar to the cocaine dose–effect curve previously obtained in the same strain, but quantitatively corresponding to a shift to the left of the cocaine dose–effect curve (Fig. 2 and Deroche et al., 1997). Although there were some differences in self-administration procedure between the present study and the one of Deroche et al. (1997), i.e. the volume of injections (50 μl instead of 25 μl in the present study) or the number of training sessions (7 instead of 10), it is unlikely that these minor methodological differences account for a leftward shift in the cocaine dose–response curve. Therefore, acquisition and dose–effect data are convergent in suggesting that adding a low dose of heroin to a non-reinforcing dose of cocaine resembled using a higher or more reinforcing dose of cocaine. The present findings are consistent with previous self-administration studies in rats and monkeys, reporting a leftward shift of the cocaine dose–effect curve using FR or PR schedules. Combination of cocaine and heroin doses that alone elicited saline-level responding were found to produce robust increases in the number of bar-presses and increased breakpoints (Duvauchelle et al., 1998; Ranaldi and Munn, 1998). Similarly, doses of cocaine that did not maintain self-administration in rhesus monkeys did so after the addition of low doses of heroin (Rowlett and Woolverton, 1997).

Interestingly, in a previous study BALB/c mice did not exhibit self-administration of low doses of cocaine, but showed drug-seeking behavior when heroin was combined with cocaine (Roberts et al., 1997). Thus, these mice have been proposed to model a subset of speedball users who do not like cocaine when used alone (Roberts et al., 1997). In humans, Foltin and Fishman (1992) and Walsh et al. (1996) reported that cocaine/mu agonist combinations produced greater scores on a measure of subjective ‘high’ than either drug used alone. A cocaine/heroin synergism is further supported by recent electrophysiological and neurochemical evidence, showing that the same mesolimbic subpopulation of neurons was activated by both drugs (Chang et al., 1998) and that mesolimbic dopamine release was potentiated during speedball self-administration (Hemby et al., 1999). Both opiates and stimulants share the ability to increase extracellular levels of mesolimbic dopamine, a neurochemical effect thought to underlie their abuse potential (Di Chiara and Imperato, 1988; Spanagel et al., 1990). However, other studies have found that discriminative stimulus or reinforcing effects of heroin and cocaine co-administration were not additive (Mello et al., 1995; Lamas et al., 1998). It is possible that the observation of an additive effect may depend on the dose ratios used in the different studies.

Finally, the present study compared locomotor stimulation induced by low doses of heroin and cocaine alone or in combination. Whereas cocaine increased horizontal and vertical activity, heroin tended to decrease those behaviors, resulting in a significantly different response to the two drugs. Mice injected with speedball exhibited a cocaine-like stimulation during the initial minutes of the session, followed by a level of activity corresponding to the average effect of both drugs, comparable to that of saline-injected animals. Therefore, locomotor activation produced by the heroin:cocaine combination is unlikely to explain the leftward shift of the dose–effect curve for cocaine observed in the self-administration experiments (present study and Deroche et al., 1997). Furthermore, in paradigms using FR schedules of reinforcement, a leftward shift of the dose–effect curve is typically related to an increase in the reinforcing properties of the drug which is self-administered (Koob et al., 1987; Yokel, 1987).

In conclusion, the present study provides evidence for reinforcing properties of concurrent heroin and cocaine administration in C57BL/6 × SJL mice. Our findings further suggest that a cocaineheroin synergism may underlie the reinforcing properties of speedball, as well as the fast and short-acting stimulating effect observed during the locomotor activity test. Because this hybrid genetic background is used to generate transgenic animals, the present findings are important for the future analysis of interactions between speedball use and genetic factors with clinical relevance.

Back to Top | Article Outline


We would like to thank Drs A. Roberts, R. Picetti and M. Taffe for their helpful comments on an earlier version of the manuscript, M. Arends for skillful editorial assistance and B. Lintz for excellent technical assistance. This investigation was supported by Grant DA 10191 (LHG) from the National Institute on Drug Abuse. This is publication number 13027-NP from The Scripps Research Institute.

Back to Top | Article Outline


1. Bolles RC (( 1972 )). Reinforcement, expectancy, and learning . Psychol Rev 79 : 394 – 409 .
2. Caine SB , Negus SS , Mello NK (( 1999 )). Method for training operant responding and evaluating cocaine self-administration behavior in mutant mice . Psychopharmacology 147 : 22 – 24 .
3. Carney JM , Landrum RW , Cheng MS , Seale TW (( 1991 )). Establishment of chronic intravenous drug self-administration in the C57BL/6J mouse . Neuroreport 2 : 477 – 480 .
4. Chang JY , Janak PH , Woodward DJ (( 1998 )). Comparison of mesocorticolimbic neuronal responses during cocaine and heroin self-administration in freely moving rats . Neuroscience 18 : 3098 – 3115 .
5. Deroche V , Caine SB , Heyser CJ , Polis I , Koob GF , Gold LH (( 1997 )). Differences in the liability to self-administer intravenous cocaine between C57BL/6 × SJL and BALB/cByJ mice . Pharmacol Biochem Behav 57 : 429 – 440 .
6. Di Chiara G , Imperato A (( 1988 )). Opposite effects of mu and kappa opiate agonists on dopamine release in the nucleus accumbens and in the dorsal caudate of freely moving rats . J Pharmacol Exp Ther 244 : 1067 – 1080 .
7. Donny EC , Caggiula AR , Mielke MM , Jacobs KS , Rose C , Sved AF (( 1998 )). Acquisition of nicotine self-administration in rats: the effects of dose, feeding schedule, and drug contingency . Psychopharmacology 136 : 83 – 90 .
8. Duvauchelle CL , Sapoznik T , Kornetsky C (( 1998 )). The synergistic effects of combining cocaine and heroin (‘speedball’) using a progressive-ratio schedule of drug reinforcement . Pharmacol Biochem Behav 61 : 297 – 302 .
9. Ellinwood Jr EH , Eibergen RD , Kilbey MM (( 1976 )). Stimulants: interaction with clinically relevant drugs . Ann NY Acad Sci 281 : 393 – 408 .
10. Foltin RW , Fischman MW (( 1992 )). The cardiovascular and subjective effects of intravenous cocaine and morphine combinations in humans . J Pharmacol Exp Ther 261 : 623 – 632 .
11. Grahame NJ , Phillips TJ , Burkhart-Kasch S , Cunningham CL (( 1995 )). Intravenous cocaine self-administration in the C57BL/6J mouse . Pharmacol Biochem Behav 51 : 827 – 834 .
12. Greberman SB , Wada K (( 1994 )). Social and legal factors related to drug abuse in the United States and Japan . Public Health Rep 109 : 731 – 737 .
13. Hemby SE , Smith JE , Dworkin SI (( 1996 )). The effects of eticlopride and naltrexone on responding maintained by food, cocaine, heroin and cocaine/heroin combinations in rats . J Pharmacol Exp Ther 277 : 1247 – 1258 .
14. Hemby SE , Co C , Dworkin SI , Smith JE (( 1999 )). Synergistic elevations in nucleus accumbens extracellular dopamine concentrations during self-administration of cocaine/heroin combinations (speedball) in rats . J Pharmacol Exp Ther 288 : 274 – 280 .
15. Koob GF , Vaccarino F , Amalric M , Bloom FE (( 1987 )). Positive reinforcement properties of drugs: search for neural substrates . In: Brain Reward Systems and Abuse Engel J , Oreland L (editors): New York : Raven Press , pp. 35 – 50 .
16. Kosten TR , Gawin FH , Rounsaville BJ , Kleber HD (( 1986 )). Cocaine abuse among opioid addicts: demographic and diagnostic factors in treatment . Am J Drug Alcohol Abuse 12 : 1 – 16 .
17. Lamas X , Negus SS , Gatch MB , Mello NK (( 1998 )). Effects of heroin/cocaine combinations in rats trained to discriminate heroin or cocaine from saline . Pharmacol Biochem Behav 60 : 257 – 264 .
18. Mello NK , Negus SS (( 1999 )). Effects of flupenthixol and quadazocine on self-administration of speedball combinations of cocaine and heroin by rhesus monkeys . Neuropsychopharmacology 21 : 575 – 588 .
19. Mello NK , Negus SS , Lukas SE , Mendelson JH , Sholar JW , Drieze J (( 1995 )). A primate model of polydrug abuse: cocaine and heroin combinations . J Pharmacol Exp Ther 274 : 1325 – 1337 .
20. National Institute on Drug Abuse (NIDA) (1998). Epidemiologic Trends in Drug Abuse. Washington DC: NIH Publ. No. 98-4300, p. 79.
21. Ranaldi R , Munn E (( 1998 )). Polydrug self-administration in rats: cocaineheroin is more rewarding than cocaine-alone . Neuroreport 9 : 2463 – 2466 .
22. Roberts AJ , Polis IY , Gold LH (( 1997 )). Intravenous self-administration of heroin, cocaine, and the combination in Balb/c mice . Eur J Pharmacol 326 : 119 – 125 .
23. Rocha BA (( 1999 )). Methodology for analyzing the parallel between cocaine psychomotor stimulant and reinforcing effects in mice . Psychopharmacology 147 : 27 – 29 .
24. Rocha BA , Ator R , Emmett-Oglesby MW , Hen R (( 1997 )). Intravenous cocaine self-administration in mice lacking 5-HT1B receptors . Pharmacol Biochem Behav 57 : 407 – 412 .
25. Rowlett JK , Woolverton WL (( 1997 )). Self-administration of cocaine and heroin combinations by rhesus monkeys responding under a progressive-ratio schedule . Psychopharmacology 133 : 363 – 371 .
26. Rowlett JK , Wilcox KM , Woolverton WL (( 1998 )). Self-administration of cocaineheroin combinations by rhesus monkeys: antagonism by naltrexone . J Pharmacol Exp Ther 286 : 61 – 69 .
27. Spanagel R , Herz A , Shippenberg TS (( 1990 )). The effects of opioid peptides on dopamine release in the nucleus accumbens: an in vivo microdialysis study . J Neurochem 55 : 1734 – 1740 .
28. Stolerman IP , Naylor C , Elmer GI , Goldberg SR (( 1999 )). Discrimination and self-administration of nicotine by inbred strains of mice . Psychopharmacology 141 : 297 – 306 .
29. Tutton CS , Crayton JW (( 1993 )). Current pharmacotherapies for cocaine abuse: a review . Addict Dis 12 : 109 – 127 .
30. Walsh SL , Sullivan JT , Preston KL , Garner JE , Bigelow GE (( 1996 )). Effects of naltrexone on response to intravenous cocaine, hydromorphone and their combination in humans . J Pharmacol Exp Ther 279 : 524 – 538 .
31. Winer BJ (( 1971 )). Statistical Principles in Experimental Design . 2nd ed : New York : McGraw-Hill .
32. Yokel RA (( 1987 )). Intravenous self-administration: response rate, the effects of pharmacological challenges, and drug preference . In: Methods of Assessing the Reinforcing Properties of Abused Drugs Bozarth MA (editor): New York : Springer , pp. 117 – 141 .

speedball; cocaine; heroin; intravenous self-administration; reward; locomotor activity; mice

© 2001 Lippincott Williams & Wilkins, Inc.