Intravenous self-administration of heroin, cocaine, and the combination in Balb/c mice

Abstract

Polydrug abuse, including the abuse of cocaine+heroin combinations (or `speedballs') is an increasingly significant problem. The use of genetically defined populations of mice has the potential to add considerably to the study of polydrug abuse. Balb/cByJ (Balb/c) mice have been shown to self-administer opiates, but not cocaine, therefore these mice were chosen for the initial characterization of intravenous self-administration of cocaine+heroin combinations. Mice were implanted with chronic indwelling jugular catheters and given the opportunity to self-administer heroin, cocaine or heroin+cocaine combinations. Heroin was self-administered, while, under the same conditions, none of the mice tested acquired cocaine self-administration. However, heroin+cocaine combinations were self-administered in naive mice as well as in mice that had failed to self-administer cocaine alone. The heroin+cocaine combination dose–effect curve resembled the heroin dose–effect curve. It is hypothesized that heroin may interact with effects of cocaine that function to limit self-administration in Balb/c mice, facilitating the acquisition and maintenance of self-administration of cocaine+heroin combinations.

Introduction

The use of genetically defined populations of mice such as inbred, recombinant inbred, transgenic and knockout strains along with behavioral genetic analytical techniques has added considerably to the study of drug abuse (Crabbe et al., 1994). The Balb/cByJ (Balb/c) inbred mouse strain has been highly utilized in the study of drugs of abuse such as cocaine, heroin and ethanol. These mice are unique in that they do not appear to acquire intravenous cocaine self-administration (Deroche et al., 1997) at doses equal to and lower than those that support acquisition in other mouse strains (Carney et al., 1991; Grahame et al., 1995; Grahame and Cunningham, 1995). In addition, a blunted locomotor response to cocaine has been observed in Balb/c mice as compared to other inbred strains (Shuster et al., 1977; Ruth et al., 1988; Weiner and Reith, 1990; Reith and Selmeci, 1991; Seale and Carney, 1991; Elmer et al., 1996; Deroche et al., 1997). It has been suggested that this may be due to lower brain concentrations of cocaine in Balb/c mice relative to other strains (Weiner and Reith, 1990), but this has not been reliably observed (Ruth et al., 1988). In fact, it appears that cocaine is quite potent in these mice as low doses are associated with significant increases in locomotor activity. However, its efficacy is low, as activity is only increased 100–150% over baseline (Elmer et al., 1996; Deroche et al., 1997). It does not appear that Balb/c mice are hyporesponsive to all behavioral effects of cocaine and other dopaminergic manipulations as they show pronounced cocaine-induced conditioned place preference (Seale and Carney, 1991) and haloperidol-induced catalepsy (Fink et al., 1982; Kanes et al., 1993). Thus, the phenomenon of low efficacy despite moderate potency of cocaine's effects in Balb/c mice may be related to an increased influence of cocaine-induced competing behaviors which disrupt measures of locomotion and self-administration.

In contrast to the effects of cocaine, opioids are self-administered and produce profound behavioral effects in Balb/c mice. For example, these mice have been shown to self-administer the opioid, etonitazene, orally in an operant paradigm (Elmer et al., 1995). They also self-administer morphine directly into several discrete brain regions in a site- and dose-specific manner (David and Cazala, 1994a, David and Cazala, 1994b, David and Cazala, 1996; Cazala and David, 1995). One report noted a failure of Balb/c mice to intravenously self-administer morphine in a single 30 min trial; however, conditioned place preference to this drug was exhibited (Semenova et al., 1995). Conditioned place preference has also been observed to the opioid etonitazene (Seale and Carney, 1991). Balb/c mice appear to be quite sensitive to several other behavioral effects of opiates including analgesia, respiratory depression, and enhanced locomotion (Castellano and Oliverio, 1975; Pick et al., 1991; Semenova et al., 1995; Elmer et al., 1995).

A difference with regard to behavioral sensitivity to cocaine versus opioids is evident from previous studies with Balb/c mice. Therefore this strain offers a unique model for examining the self-administration of combinations of cocaine and the opioid, heroin. Co-injection of cocaine and heroin, referred to as a `speedball', is increasing in popularity (Pollack et al., 1989; Schütz et al., 1994). The precise nature of the simultaneous use of two pharmacologically distinct substances remains unclear. However, discussions with speedball users suggest that speedballs enhance the positive effects of opiates and/or decrease the aversive properties of cocaine (Rosen and Kosten, 1991). These observations are supported by empirical evidence from a human behavioral pharmacology study indicating that speedballs have subjective properties which are unique to either cocaine or heroin alone (Foltin and Fischman, 1992).

The goal of the present experiments was to characterize intravenous heroin self-administration in Balb/c mice and to examine self-administration of combinations of heroin+cocaine. Based on the literature, it was predicted that a model of intravenous heroin self-administration could be developed in this mouse strain. Furthermore it was assumed that these mice would not acquire cocaine self-administration using criteria similar to those used previously in our laboratory (Deroche et al., 1997). Thus the Balb/c mouse strain offers a unique model for investigation of the nature of intravenous self-administration of combinations of heroin and cocaine. Furthermore, this study forms the basis for further development of mouse models of polydrug self-administration, allowing for a detailed investigation of the contribution of genetic and neuropharmacological influences to this type of drug abuse.