Even though decades of research have shown that the familiarity of a physical location where addicts use cocaine or heroin can markedly affect their drug response—or even their vulnerability to an overdose—the mechanisms of this process remain poorly understood. According to Bruce Hope, a researcher at the US National Institute on Drug Abuse, this is due in part to belated recognition by molecular and cellular biologists of the effects of external conditions on animals used in drug studies.

“We didn't appreciate environment,” he says. “We'd always assumed that the mechanisms we saw in the home cage are going to be the same ones operating in a novel environment, and it turns out not to be so.” In fact, several studies from recent years have revealed neuronal subpopulations that are specifically activated only when drugs are consumed in an environment previously associated with such behavior.

Hope was interested in studying the neurons involved in these circuits, but found himself limited by existing neurobiological techniques designed to examine brain tissue as homogenous cell clusters rather than sparsely distributed, specialized cellular circuits. To counter this limitation, his team devised a new approach for disabling only those cells involved in environment-specific response to cocaine intake.

They noted that rats injected with repeated doses of cocaine in a particular environment (for example, a square, flat-floored chamber) exhibit increased locomotor activity—behavior that is enhanced by additional treatments in the same environment, but not when the rat is treated with cocaine in an unfamiliar environment (for example, a round chamber with woodchips). Upon examining Fos (c-fos) expression in these rats—a general indicator of neuronal activity—they found that ∼2–3% of the neurons in the nucleus accumbens, a brain region involved in 'reward-seeking' behavior, are active in these environmentally sensitized rats.

To selectively target these cells, the researchers used transgenic rats in which the c-fos promoter was linked to the gene encoding β-galactosidase. By treating these rats with Daun02, a compound that is converted by this enzyme into an action potential–blocking drug, they specifically inhibited neurons engaged in active signaling. When they infused these transgenic rats with Daun02 after initial cocaine treatment, the rats lost their sensitized response, with subsequent cocaine doses inducing movement equivalent to that observed in rats receiving their first dose, regardless of whether it was given in familiar or unfamiliar surroundings.

Notably, this treatment did not appear to exert any nonspecific neurological effects, nor did it affect the rats' capacity to respond to cocaine or engage in normal locomotor activity. “It's a unique set of neurons that were only activated by the repeated presence of both the drug and the environment together,” says Hope. “This means that it could be a mechanism for how the learned association is produced between those two factors.”

His team is now investigating characteristics of the neurons affected in these experiments, but their initial findings suggest that Daun02 inactivation could offer a general strategy for precisely manipulating neural circuitry underlying other complex behavioral processes. Accordingly, Hope is now collaborating with National Institute on Drug Abuse colleague Yavin Shaham to apply their method to 'cue-induced reinstatement', the process by which certain environmental cues can restore seeking behavior in animals previously weaned off of drugs—in this case, heroin. “The parallel might be exposing an addict to drug paraphernalia, and all of a sudden they crave heroin,” says Hope. “This is what that's supposed to model.”