In the present study, all three amphetamines tested effectively depleted tissue concentrations of both dopamine and serotonin. Some previous studies in the mouse indicate that PCA selectively induces depletions of serotonin (Sanders-Bush et al. 1975
; Steranka et al. 1977
; Steranka and Sanders-Bush 1980
). However, more recent studies indicate that the hyperthermic effects of PCA are mediated by dopaminergic action (Sugimoto et al. 2001
), and dosing regimens of PCA can lead to depletion of markers for (Itzhak et al. 2004
) and tissue concentrations of (Saadat et al. 2006a
) both serotonin and dopamine. Studies of the persistent effects of MDMA and METH in the mouse have mostly shown selective dopaminergic effects (O'Callaghan and Miller 1994
; Stone et al. 1987
); however, serotonergic effects have also been reported (Hirata et al. 1995
; Renoir et al. 2008
). It is unclear why these discrepancies across studies arise but they may be related to experimental conditions such as environmental factors, drug dosage, dose scheduling, or strain differences. Nevertheless, the consistent pattern across these previous studies and the present report is that, although each derivative can produce selective effects, there are conditions under which all three derivatives can affect either neurotransmitter in the mouse.
Although all three amphetamines were capable of depleting dopamine and serotonin levels in some of the brain regions examined, their effects were not identical, and they exhibited different patterns of effects on the regional concentrations of the major metabolites of these neurotransmitters. In this regard, PCA appears to have a more robust capacity for depleting serotonin than either MDMA or METH as it significantly decreased serotonin levels in all 6 brain regions examined, whereas MDMA and METH significantly decreased serotonin levels in only the posterior striatum and the cingulate. The dopamine depleting effects of these compounds also exhibited some specificity. Specifically, although all three derivates significantly decreased dopamine levels in the posterior striatum, only METH and PCA significantly decreased dopamine levels in the anterior striatum. Moreover, while all three derivatives engendered qualitatively similar depletions of the dopamine metabolite DOPAC in the posterior striatum, only METH and PCA significantly depleted DOPAC levels in the anterior striatum or HVA levels in the posterior striatum. Similarly, only METH and PCA significantly depleted tissue levels of the serotonin metabolite 5-HIAA in the frontal cortex or in the cingulate. As relatively little is known about the pharmacodynamic effects of PCA, it is difficult to know what pharmacological targets are mediating these differences in neurochemical effects. Nevertheless, the differential effects on neurochemistry of the amphetamines utilized in this study do not appear to be related to potency or pharmacokinetic differences as an effect scaling procedure (Fantegrossi et al. 2008
; Wang et al. 2004
) was used to control for these variables. As such, we propose that it is possible that potency differences at the monoamine transporters are not responsible for the differential neurochemical effects of these amphetamines, and determining the pharmacological effects of these compounds at other targets may help to explain their differential effects on neurochemistry.
The differential capacities of MDMA, METH, and PCA to impair PA performance may be related to their differential effects on neurochemistry. In this regard, exposure to METH and PCA impaired PA performance and engendered a specific set of neurochemical depletions, whereas exposure to MDMA neither engendered some of these neurochemical depletions nor impaired PA performance. For example, while MDMA depleted tissue content of dopamine in only the posterior striatum, METH and PCA depleted tissue content of dopamine in both the anterior and the posterior striatum. Importantly, not only did within subject correlation analysis show that anterior striatal dopamine levels predict PA performance, but stepwise multiple linear regression analysis showed that the level of dopamine in the anterior striatum was the strongest predictor of PA performance among all of the regional neurochemicals that were selectively depleted by both METH and PCA. Consistent with previous reports that PA behavior is independent of tissue levels of serotonin (Barrionuevo et al. 2000
; Myhrer 2003
; Santucci et al. 1996
), serotonin levels were not selectively depleted by METH and PCA in any brain region. However, the tissue concentrations of the serotonin metabolite 5-HIAA in the cingulate were depleted by only METH and PCA, and addition of the tissue concentrations of 5-HIAA in the cingulate significantly enhanced the capacity of the multi regression model to predict PA performance. This work supports previous findings that PA may be mediated via dopaminergic mechanisms (Adriani et al. 1998
), extends those findings by showing that there may be subregional specificity in the striatum underlying this learning and memory process, and indicates that other dopamine dependent behaviors may be impaired by amphetamine derivative exposure. Moreover, this work indicates that serotonergic systems may also have a modulatory role in PA behavior and that PA behavior may be influenced by interactions between the cingulate and the striatum.
The finding that exposure to METH and PCA impairs PA performance is consistent with the notion that exposure to these drugs impairs learning and memory. However, because we administered each amphetamine derivative prior to PA training and testing, some alternative interpretations of our results are that the METH- and PCA-treated animals were differentially sensitive to the training stimulus or that METH and PCA exposure engendered anxiogenic- or anxiolytic-like effects. Indeed, exposure to METH and PCA did alter the initial latencies of the mice to cross prior to training. Moreover, this design allowed us to potentially assess the effects of the drugs examined on overall learning and memory, but it did not allow us to determine whether any impairments observed represent selective effects on learning, selective effects on memory, or combined effects on both. Future studies should be designed to compare the effects of METH and PCA treatments when they are administered before the training session, shortly after the training session, or before the retention test because these experiments would begin to elucidate whether METH- and PCA-induced behavioral and cognitive deficits represent selective deficits in learning and memory.
Although it is difficult to make direct comparisons between cognitive processes in humans and laboratory animals, the results of this study appear to have relevance for human addicts of amphetamines. Some of the cognitive deficits in METH addicts that have the strongest support are deficits in information processing speed, attention, learning, memory, reaction times, and executive functions (Kalechstein et al. 2003
; Simon et al. 2010
). Indeed, a meta-analysis of the literature showed that the largest effects sizes in METH addicts were for deficits in executive functions, learning, and memory (Scott et al. 2007
). Although PA may have little relevance for deficits in executive functions, it provides preclinical assessments of deficits in learning and memory. Given the consistency between the effects reported in this study and the deficits that have been described in human METH addicts, the continued use of the PA assay may allow us to determine the factors that influence some of the specific deficits that occur in this clinical population and allow us to study treatments that may reverse these deficits. It is important to note, however, that deficits in learning and memory have also been reported in MDMA abusers (Kalechstein et al. 2007
), and the results of the present study are not consistent with those findings. This lack of consistency may be a result of the complexity of the terms learning and memory, as these concepts encompass a broad range of processes, and the possibility that each amphetamine derivative may engender discrete deficits. In this regard, it has been show the MDMA does impair conditioned place aversion induced by lithium chloride (Achat-Mendes et al. 2005
), a learning process that is similar but not identical to PA. Similar to the present study, future studies should continue to compare and contrast the discrete neurochemical and cognitive deficits engendered by different amphetamines as this may advance our understanding of the neurobiology of learning and memory and our understanding of the cognitive consequences of exposure to these amphetamines.
In summary, the present study demonstrates that changes in tissue concentrations of dopamine in the anterior striatum strongly predict deficits in PA behavior in the mouse. METH and PCA significantly decreased dopamine in this brain region and concomitantly impaired PA behavior, whereas MDMA did not. Similarly, only METH and PCA significantly decreased 5-HIAA concentrations in the cingulate and depletion of this metabolite of serotonin also was predictive of PA performance. Differences in potency or pharmacokinetics do not appear to account for the differences between the neurochemical and cognitive consequences of exposure to MDMA and exposure to either METH or PCA, as the dosing regimens utilized for each compound were effectively matched using an effect scaling procedure. These studies demonstrate that certain amphetamines impair PA performance in mice and that these impairments may be attributable to specific neurochemical depletions.