The present study demonstrates that there are significant adaptations in the striatal glutamatergic system in response to chronic cocaine self-administration, though these appear to be restricted to the more dorsal territories of the striatum. We chose to focus on the striatum in this study based on three facts: 1) the striatum plays an essential role in the reinforcing effects of cocaine [39
], 2) group II mGluRs attenuate the reinforcing effects of cocaine [18
] and 3) group II mGluRs are localized in the striatum [14
]. The finding of significantly higher densities of group II mGluRs in the caudate nucleus as a result of chronic cocaine self-administration indicates the presence of considerable plasticity within the dorsal striatum. In contrast, no differences in receptor density were observed between the groups in either the putamen or nucleus accumbens. These data suggest, therefore, that elements of the glutamate system within the dorsal striatum may be more sensitive, compared with the ventral striatum, to the effects of chronic cocaine exposure in the non-human primate.
The localization of [3
H]LY341495 binding sites within the non-human primate striatum agrees with previous autoradiographic and immunohistochemical labeling studies [14
]. A novel finding of the present study was the identification of a medial-lateral gradient (high to low) in the binding density of group II mGluRs. The highest levels were found in the medial caudate nucleus and lowest in the lateral portions of the caudate, putamen and nucleus accumbens core and shell. Given the topographical organization of glutamatergic projections to different regions of the striatum [41
], the differential distribution of group II mGluRs within the medial and lateral portions of the striatum may have functional implications for how glutamate regulates the striatum.
Group II mGluRs in the striatum are thought to be localized primarily pre-synaptically [5
], where they can modulate the release of neurotransmitters [44
]. As exposure to cocaine increases, there is hypothesized to be a gradual and escalating involvement of the glutamatergic system [2
]. Indeed, previous data in rodents have shown that acute administration of cocaine does not induce the release of glutamate [3
], whereas following chronic self-administration, levels of extracellular glutamate are progressively increased with each dose of cocaine [3
]. One mechanism explaining our results, then, may be a compensatory increase in the levels of group II mGluRs autoreceptors in response to these sustained elevations in extracellular glutamate concentration. One of the predicted consequences of these up-regulated group II mGluR autoreceptors may be lower basal levels of glutamate, which have been reported to occur in cocaine-exposed rodents [3
] and humans [51
]. In agreement with our data, Weiss and colleagues reported functional increases in group II mGluRs in a number of brain areas such as the prefrontal cortex, central nucleus of the amygdala, and hippocampus following escalation of cocaine self-administration in rats [52
]. Thus, future studies will be aimed at investigating how group II mGluRs are dysregulated in response to cocaine self-administration in these and other brain areas.
One of the interesting outcomes from the present study was the lack of effects in the ventral striatum (incorporating both ventral putamen and nucleus accumbens). A recent investigation in rodents also failed to observe any impact on the function of group II mGluRs in the nucleus accumbens following chronic cocaine self-administration [52
]. Although the initial effects of cocaine have been shown to be primarily located in limbic-related areas, such as the ventral striatum [31
], with longer exposure the effects of cocaine expand to incorporate more dorsal regions [53
]. Given the extended duration of cocaine self-administration in the present study, therefore, it is possible that group II mGluRs may have been transiently affected in the ventral striatum earlier in the course of exposure. Thus, the compartmentalization of effects and relative sensitivity of the glutamate system in the dorsal striatum following chronic exposure to cocaine self-administration is in accordance with these studies.
In contrast to the data reported here, data from rodent studies have reported decreased function, or lower levels, of group II mGluRs following repeated cocaine exposure [55
]. There are, however, important differences between these studies. First, there are considerable species differences in the glutamate system [57
], as well as in the response of the brain to cocaine [60
]. Second, the monkeys in our study self-administered cocaine, while the rodents in the other studies received cocaine non-contingently. There is evidence of behavioral and neurochemical differences in the effects of cocaine delivered contingently versus non-contingently [62
]. Third, the total intake of cocaine received by the animals between the studies was very different (180–200 mg/kg in the rodent studies, 900 mg/kg in the present study) and fourth, these previous studies involved a period of abstinence (up to 3 weeks) before measuring either the density or functionality of group II mGluRs. As it relates to the latter point, future studies are needed to investigate the role of cocaine abstinence following chronic cocaine self-administration to determine how malleable group II mGluR densities are in response to long-term cocaine exposure and abstinence in non-human primate brains.