We have employed a relatively simple design in some of our studies to evaluate the behavioral and neurobiological variables regulating the acquisition and expression of cocaine-induced conditioned effects. Basically, three groups of rats are employed in this paradigm. In its simplest form, on d 1 the first groups of rats (PAIRED) is injected with cocaine (30-40 mg/kg ip) prior to placement in locomotor activity chambers (scented with peppermint) for 30 min. One hour following return to their home cages, these rats are injected with saline. The second group (UNPAIRED) is treated in a similar fashion, but receives saline prior to placement in the locomotor activity chamber and cocaine in the home cage. The third group (CONTROL) receives saline in both environments. On d 2, all rats are challenged with either saline or 10 mg/kg of cocaine prior to placement in the locomotor activity chamber. In some of our studies, the training sessions were increased to 3, 5, or 7 d when it was necessary to increase the persistence and strength of conditioning.
We have shown significant conditioned effects of cocaine using this design, which is reflected by dramatic increases in locomotor output in the PAIRED group on the test day relative to the other two groups. Evidence from the laboratory (16) indicates that such increases in locomotor activity in the PAIRED group is established through associative learning mechanisms and does not simply reflect the inability to habituate to the environment under the influence of the drug, as has been suggested by some.
One of our main interests has been to define the neurobiological substrates underlying the conditioned effects of cocaine. We have found, for example, that dopamine (DA)-depleting lesions of the nucleus accumbens and amygdala prevented cocaine-induced conditioning after one training session. The amygdala lesions, however, were not effective in preventing conditioning when more extensive training was employed. The dopaminergic components of the amygdala appear to play a more subtle role in the formation of cocaine-conditioned behavior than those in the nucleus accumbens. Dopamine-depleting lesions of the frontal cortex and striatum were not effective in preventing the conditioned locomotor effects of cocaine. Neurotoxin-induced lesions of the raphe and locus ceruleus were equally ineffective. Radio-frequency lesions of the dorsal and ventral hippocampus, as well as the cerebellum, also had little effect on the establishment of cocaine-induced conditioning after one day of training. Such findings strongly suggested that DA function in the nucleus accumbens and, to a lesser degree, in the amygdala, is necessary for the formation of cocaine-conditioned behaviors.
It has been suggested that different neurobiological processes are involved in the acquisition and expression of psychomotor stimulant-induced conditioned increases in motor behavior. With lesions made prior to training, it is not possible to determine whether the deficit seen is related to disruptions in the acquisition process or to the expression of the behavior. Because DA is involved in the stimulatory and appetitive properties of psychomotor stimulants, it is not surprising that the blockade of DA function would lead to decreases in the acquisition of conditioning. For example, neuroleptics coadministered with either amphetamine (17) or cocaine (18,19) have been found to prevent the development of conditioned locomotor behaviors. More recently, we have found that D1 and D2 DA receptor antagonists are equally effective in preventing the formation of cocaine-induced conditioning (20). Likewise, conditioning in the 1-d design was found only following administration of a combination of D1 and D2 agonists during training, and not when either was administered separately. This would suggest that concurrent D1 and D2 DA receptor occupation is necessary for conditioning to occur.
There are a variety of mechanisms by which DA antagonists could disrupt the acquisition of cocaine conditioning (20). It is most likely, however, that the ability of these drugs to decrease or prevent conditioning to psychomotor stimulants is related to their ability to attenuate the unconditioned effects of the drugs, which are critical in forming the conditioned association. We have suggested that the ability of DA blockers to prevent conditioning is related to their ability to decrease the motivational significance of the unconditioned stimulus (e.g., cocaine). It is well established that the strength of conditioning is directly related to the intensity of the unconditioned stimulus in other conditioning paradigms (16).
Although mixed D1-D2 and selective D1 and D2 antagonists have been found to prevent the establishment of conditioning to cues associated with cocaine, they have been reported to be relatively ineffective in preventing expression once established. Early studies by Beninger and Hahn (21) and Beninger and Herz (18) found that pimozide did not eliminate the behavioral differential between cocaine-conditioned animals and their controls. We have recently extended these findings by demonstrating that neither D1 nor D2 antagonists are effective in altering the differential in performance between the conditioned and unconditioned rats during the test phase (20).
On the surface, these findings, together with the ability of DA antagonists to block the acquisition of conditioned behaviors, appear to suggest that although intact DA function is critical for the development of conditioning to cocaine-associated cues, it is not necessary for the expression of the conditioned response once established. It is possible that nondopaminergic pathways acquire the ability to elicit such conditioned reactions. The second alternative is that DA is involved in the expression of the conditioned behavior and that the differential in activity seen between the conditioned and unconditioned groups is determined and maintained by increased activity of mesolimbic DA pathways in the former group, despite similar partial blockades of DA receptors in all experimental groups.
Using the 1-d conditioning paradigm described previously, we have evaluated the ability of stimuli associated with cocaine to increase mesolimbic DA functions (22). Microdialysis procedures revealed significant increases in extracellular DA in the nucleus accumbens in the PAIRED group relative to the control groups during test day. Kalivas and Duffy (23) also have reported increases in mesolimbic DA elicited by stimuli associated with cocaine. More recent studies in our laboratory have not found such conditioned increases in DA overflow in either the amygdala or striatum, suggesting some regional specificity in the effects of conditioned stimuli on DA function. Lesion studies also appear to support these findings. DA-depleting lesion studies of the nucleus accumbens made immediately after 7 d of conditioning were able to prevent the expression of the conditioned response when rats were tested 7 d postoperatively. Gold et al. (24) have reported similar findings.
Was this article helpful?