Association for Behavior Analysis International

Over the past decade, pre-clinical research has made great strides in understanding the pharmacological underpinnings of benzodiazepine-maintained behavior. A major determinant of benzodiazepine self-administration by human subjects is antecedent conditions, e.g., drug history, anxiety disorders. Studies with non-human subjects offer unique opportunities for understanding the intersection of antecedent conditions and pharmacological mechanisms underlying benzodiazepine self-administration. Benzodiazepines are challenging to establish as reinforcers, most likely due to modest reinforcing effectiveness compared with other drugs of abuse (e.g., stimulants). Unequivocal reinforcing effects of benzodiazepines have been demonstrated with non-human primates, and we recently have shown that drug history is a key determinant of the pharmacological mechanisms underlying reinforcing effects of benzodiazepines. Rodent models have proven more difficult to establish. In one example, C57Bl/6J mice were given unlimited access to midazolam plus sucrose in one bottle and sucrose alone in a second bottle, and approximately 50% chose the midazolam bottle significantly more than the sucrose-alone bottle. Because C57Bl/6J mice are genetically identical, these findings illustrate how environmental factors can robustly determine the extent to which a benzodiazepine is consumed. Overall, these observations suggest that the reinforcing effects of benzodiazepines may be uniquely reliant on exogenous factors, such as drug history and behavioral phenotype.

Animal models of self-administration have been useful in understanding factors that contribute to the abuse of benzodiazepines, but there are few demonstrations of benzodiazepine self-administration in rodents, the most widely-used preclinical model. Rats with chronic intravenous (i.v.) catheters were trained to self-administer the short-acting benzodiazepine midazolam on a fixed-ratio schedule of reinforcement. The response-dependent i.v. delivery of midazolam maintained steady, but relatively low rates of responding. Following the acquisition of responding, factors that may contribute to the frequency of responding maintained by midazolam were investigated including multiple doses and food deprivation. In this regard, varying the dose of midazolam resulted in a characteristic inverted U-shaped dose-response function, whereas food deprivation appears to enhance rates of responding. Lastly, following extinction of responding maintained by midazolam, reinstatement tests were conducted with non-contingent injections of benzodiazepines and response-dependent presentations of drug-paired cues. Responding previously maintained by midazolam self-administration recurred during reinstatements tests, but only when the drug-paired cues were present. These projects have developed and are refining a rodent model of benzodiazepine self-administration that may prove useful in studying the effects of environmental and pharmacological factors related to benzodiazepine use, abuse, and the relapse of benzodiazepine-maintained behavior.

The ?-aminobutyric acid type A (GABAA) receptors are pentamers constituted from structurally-distinct proteins, with each protein family comprised of different subunits (i.e., a, , and ? subunit families). Benzodiazepines bind to a site on the GABAA receptor located at the interface of the a and ? subunit, with a1, a2, and a3 subtypes likely involved in self-administration of benzodiazepines. TP003, a compound selective for a3 subunit-containing GABAA receptors, was self-administered under a progressive-ratio schedule of reinforcement in rhesus monkeys experienced in benzodiazepine self-administration but not by those experienced in cocaine self-administration. However, compounds selective for a1 subunit-containing GABAA receptors were self-administered by benzodiazepine- and cocaine-experienced monkeys, indicating that GABAA subtype and past drug experience combine to determine the reinforcing effects of benzodiazepines. The finding that TP003 was not self-administered in cocaine-experienced monkeys may be due to this compound being either a neutral stimulus without reinforcing effects or a punishing stimulus. Using a choice procedure in cocaine-experienced monkeys, we are evaluating the ability of selective compounds to enhance or attenuate cocaine choice. Based upon current views on neural circuitry of benzodiazepine reinforcement, we predict that TP003 will function primarily as a punisher, although initial findings do not support this idea.

Neuroactive steroids and benzodiazepines share behavioral effects, such as anxiolysis and self-administration, which may be enhanced differentially when these drugs are combined. Under a progressive-ratio (PR) schedule of reinforcement, combinations of the benzodiazepine, triazolam, and neuroactive steroid, pregnanolone, resulted in infra-additive reinforcing effects in male rhesus monkeys. It is unknown, however, if these findings extend to female subjects as well as other benzodiazepine-neuroactive steroid combinations. We are currently re-examining self-administration of triazolam and pregnanolone in female monkeys and including the clinically-relevant benzodiazepine, clonazepam, and a neuroactive steroid currently in clinical trials, ganaxolone. We are also using a behavioral economics approach to assess the extent to which combining the drugs alters reinforcing effectiveness relative to either drug alone. For these studies, male rhesus monkeys were trained to self-administer midazolam under a fixed-ratio (FR) 5 schedule of reinforcement. Both clonazepam and ganaxolone functioned as reinforcers under these conditions, and we are determining demand curves by systematically increasing the FR value for the drugs alone or combined. Using PR and behavioral economic approaches to assess the reinforcing effects of benzodiazepine and neuroactive steroid combinations may shed light on their potential clinical utility as anxiolytic treatments with reduced abuse liability.