Here, we demonstrate that overexpression of DAT alters ethanol drinking in C57BL/6 mice. Because mesocorticolimbic DA is postulated to contribute to alcohol consumption, we transplanted neural stem cells modified to overexpress the hDAT into the NAC of C57BL/6 adult female mice. We postulated that grafted C17.hDAT cells that express high levels of the DAT should clear the extracellular DA faster than the surrounding tissue. High-speed chronoamperometric measurements showed a trend toward faster DA clearance rates in cortical C17.hDAT transplants compared with C17.mock transplants. The reduction in alcohol intake that was observed in mice with the C17.hDAT transplants in NAC is consistent with the idea that NAC DA might affect ethanol consumption.
Our ethanol-consumption data are in agreement with studies showing that D1 and D2 receptor antagonists, injected systemically or locally into the NAC, decrease alcohol consumption and operant responding to alcohol [17, 18]. Mice lacking D1 receptors also show decreased ethanol consumption in a two-bottle choice test compared with wild-type mice . Extracellular DA levels are increased in the NAC by systemic alcohol administration [20–23]. These findings are consistent with the theory that DA neurotransmission in the NAC is necessary for the rewarding properties of alcohol. However, despite numerous studies examining the involvement of the dopaminergic system in reward, the precise role for DA in drug addiction is still far from clear.
The DAT is the primary physiologic mechanism for terminating DA neurotransmission in the CNS. By inhibiting the DAT, drugs of abuse like cocaine and phencyclidine preferentially increase extracellular DA in NAC, as compared with dorsal striatum. Ethanol also elevates DA levels in the NAC. However, unlike cocaine, ethanol does this by increasing the firing of DA neurons in the VTA . Alcohol-preferring rats will self-administer ethanol into the VTA . Ethanol exerts its actions by potentiating GABAergic inhibition of inhibitory interneurons in the VTA [26, 27], leading to disinhibition of DA neurons. As with other drugs of abuse, the increased firing of VTA neurons results in enhanced release of DA in the NAC [22, 28, 29].
Some studies have reported that D1 and D2 receptor agonists, as well as pharmacologic agents that elevate DA levels, also reduce ethanol self-administration [30–32]. Conceivably, the reduced drug-seeking behavior resulting from activation of D1 or D2 receptors in these studies may be due to either a decrease in the rewarding properties of alcohol or a feedback-mediated elevation of endogenous DA, which reduces the drive for alcohol . Enhancing extracellular DA in the NAC with GBR12909, a DA-uptake inhibitor, decreased DA levels in the VTA, suggesting the presence of a negative-feedback loop in the regulation of DA release . The regulation of DAT by alcohol might also have a role in the control of alcohol intake, as supported by the observation that chronic (8-week) consumption of 15% alcohol increases DA uptake in the NAC of selectively bred high-alcohol-drinking (HAD) female rats . Although our study differs by using wild-type female mice in a shorter (27-day) experimental paradigm with a maximum alcohol concentration of 10%, both studies imply that plasticity in the reward circuitry involving DAT might be a modifier of alcohol drinking.
Recently, two alternative approaches for overexpression of DAT have been used. The first is a global one achieved by creating transgenic mice harboring a bacterial artificial chromosome containing the murine DAT gene . Although these mice display an increased locomotor response to amphetamine compared to wild-type mice, they have a reduced operant response to a natural sweet-food reward, a finding that in principle supports our observations. A second study used direct injection into rat NAC of a lentiviral construct encoding rat DAT, which led to an impulsive and risk-prone phenotype . This approach varies from ours in that lentivirus-infected neurons in the NAC can retrogradely transport DAT product into the VTA, whereas the hDAT expressed by C17.hDAT cells in our study is not transported and remains within the confines of the engrafted area.
Alcohol-preferring strains of rats, as well as the C57BL/6J mouse used here, have lower basal levels of dopaminergic function (decreased DA levels and turnover) than do non-alcohol-preferring strains [30, 37, 38]. In the C57BL/6J mice, enhancing synaptic DA decreased ethanol consumption, an effect that could be mimicked with DA-receptor agonists or blocked by DA-receptor antagonists . These results suggest that voluntary ethanol consumption is higher in these mice because of increased drive (that is, they need more ethanol to get the same reward). If this were simply the case, we might have expected that the overexpression of DAT (lowering the amount of released DA to interact with postsynaptic receptors) would increase drive and ethanol consumption. Our finding of decreased ethanol consumption after the overexpression of DAT is inconsistent with the theory that increased preference and consumption in C57BL/6 mice is due to increased drive. The alternative theory, as to why the hypodopaminergic C57BL/6 mouse is alcohol preferring, is that the ethanol-induced release of DA on top of a low baseline is perceived as a larger reward (signal-to-noise difference). However, by overexpressing DAT, we might be diminishing the actions of DA and thereby reducing the stimulation of postsynaptic DA receptors, leading to decreased consumption. Thus, our mice are not simply hypodopaminergic, but might rather have dopaminergic neurotransmission through the NAC-reward pathway selectively inhibited.
Because we have measured DA clearance by stem cell transplants within cerebral cortical slices, in future experiments, both behavior and DA clearance must be measured simultaneously in the mice to assess whether a relation exists between changes in ethanol drinking and DA uptake by the C17.hDAT stem cells transplanted into the NAC. It is intriguing to speculate that stem cells engineered to express hDAT will adjust their capacity to clear DA, based on the transplant location in the brain or the level and duration of exposure to ethanol or both. Although the current study was not specifically designed to answer this question, addressing this issue represents an important direction for future explorations.
We chose to examine female C57BL/6J mice in our study based on the reported higher preference seen in females than males of this strain of mice. Because we did not use high and aversive concentrations of ethanol for our posttransplantation preference, an increase in consumption or preference ratio would have been detectable in our study. Lesions of the NAC with 6-hydroxydopamine decrease the acquisition, but not the maintenance of alcohol preference in alcohol-preferring rats . This study suggests that preference, once established, is difficult to disrupt. It is possible that the observed effect on alcohol consumption would have been greater in magnitude if we had performed the transplants before the acquisition of alcohol preference. Alternatively, stem cells engineered to support higher dopamine clearance in vivo might also produce a larger reduction in ethanol consumption. The extent to which the posttransplantation decreases in alcohol intake might reflect changes in the reinforcing properties of ethanol should be confirmed in the future by using operant self-administration.
Studies presented here examine a very short experimental period. Future experiments will be designed to examine long-term survival of the transplanted stem cells, expression of hDAT, DA clearance, and to answer whether the effects on alcohol consumption persist. Long-term outcomes assessed over several weeks (or longer) will be therefore necessary to test our approach fully and to determine its broader usefulness.
Blood ethanol concentration (BEC) depends on alcohol intake and on the pharmacokinetics of alcohol metabolism, both of which vary between species and genders. Time of sampling is also critical for the estimation of BEC. In nocturnal animals such as mice, more than 90% of daily alcohol intake occurs during the dark . When access to alcohol is continuous, C57BL/6J females drink as much as twice the amount of 10% alcohol typically consumed by the males . In one report, male C57BL/6J mice drinking 10% ethanol in the unrestricted-access two-bottle preference test (same paradigm as the one used by us) exhibited BECs up to 150 mg/dl (in the dark) while consuming about 5 g/kg/day . In a similar experiment, which used a drinking-in-the dark paradigm with a 4-hour 20% alcohol access, a mixed group of male and female C57BL/6J mice consumed on average 7 g/kg of alcohol in 4 hours, which resulted in a BEC of 100 mg/dl . Because female mice used in our study had alcohol intakes between 15 and 18 g/kg/day, it is likely that the resulting BECs could have been even higher, particularly if measured in the dark. Because in humans, BECs around 80 mg/dl (the legal limit for driving while impaired in the United States) coincide with altered emotions, loss of control of fine motor movements, and affected driving performance , our consumption model is relevant to the human condition.
Our study demonstrates a novel strategy of using stem cell transplantation for moderating alcohol effects and intake, and our approach may provide future insights into better understanding of the mechanism of alcohol reward and addiction. This technique, when combined with the use of stem cells capable of metabolizing molecules of interest, could be of importance in studies requiring sequestration and degradation of molecular signals as well as neurotoxins. For example, glutamate neurotoxicity can be reduced by increasing expression of glutamate transporters . Similar approaches could be used to study multidrug resistance-associated transporter proteins, which are important in the context of treatment of epilepsy and brain cancers  and transport across the blood-brain barrier for delivery of neuropharmaceuticals into the CNS .