Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Ethanol acutely increases the activity of the mesolimbic dopamine (DA) pathway (Gessa et al., 1985; Imperato and DiChiara, 1986; Weiss et al., 1993; Brodie et al., 1999) and this effect has been widely implicated as one of the mechanisms by which ethanol exerts its reinforcing actions (Samson, 1986; Koob, 1992). However, in animals trained to self-administer ethanol, not only ethanol consumption (Weiss et al., 1993; Gonzales and Weiss, 1998) but also exposure to contextual stimuli associated with ethanol can activate DA transmission within the nucleus accumbens (NAcc) (Weiss et al., 1993; Katner et al., 1996; Gonzales and Weiss, 1998). These observations suggest that in addition to its presumed role in the maintenance of ethanol self-administration (i.e., the consummatory aspect of behavior motivated by ethanol), mesolimbic DA transmission may have a function in mediating behavior that brings the animal into contact with this reinforcer (i.e., the appetitive or incentive-motivational aspect of ethanol's actions).

A role of DA in the incentive motivational actions of ethanol may have important implications for understanding of mechanisms regulating the initiation of ethanol-seeking and, by extension, ethanol-craving and relapse. Alcohol-associated stimuli or events can elicit or exacerbate the desire to drink and increase self-reported withdrawal symptoms in detoxified alcoholics (Ludwig and Wikler, 1974; Kaplan et al., 1985; Cooney et al., 1997). Learned responses such as these are thought to be a critical factor contributing to the chronic relapsing nature of alcoholism (O'Brien et al., 1998). It will, therefore, be important to better understand the neuropharmacological basis for the motivating effects of alcohol-associated environmental stimuli.

A growing body of data links mesocorticolimbic DA transmission with conditioned responses associated with natural and drug rewards and in reward-related incentive learning (for reviews, see Robbins and Everitt, 1996; Beninger and Miller, 1998). Electrophysiological findings suggest a general role for ventral tegmental DA neurons in the processing of reward-predictive conditioned stimuli (Schultz et al., 1997). Neurochemical data show that anticipation of both drug and nondrug rewards increases DA release in the NAcc (Weiss et al., 1993, 2000; Katner et al., 1996; Bassareo and Di Chiara, 1997). Moreover, selective antagonism of the D1 receptor attenuates behavior elicited or maintained by stimuli associated with food or cocaine (Weissenborn et al., 1996; Ciccocioppo et al., 2001b). In light of these findings, one may predict that DA mediates the motivating effects of ethanol-related stimuli as well. However, existing data on a role of DA in the initiation of ethanol-seeking are inconclusive. Although ethanol-associated contextual stimuli can increase DA release (Weiss et al., 1993; Gonzales and Weiss, 1998; Nurmi et al., 1998), microinjections of DA agonists or antagonists into the NAcc did not affect the response latency to initiate operant responding for ethanol in rats (Samson and Hodge, 1996). On the other hand, consistent with a possible role of DA in behavioral responses elicited by ethanol cues, activation of DA transmission in the NAcc by local administration of amphetamine was shown to enhance the ability of an ethanol-paired stimulus to function as a conditioned reinforcer (Slawecki et al., 1997).

The purpose of the present study was to further investigate dopaminergic involvement in the motivating actions of ethanol-associated environmental stimuli in rats. A reinstatement model of relapse was used to test whether the initiation and maintenance of drug-seeking behavior induced by an ethanol-related contextual stimulus are reversible by treatment with DA D1 or D2 receptor antagonists. In addition, the significance of prior ethanol dependence in the effects of these DA antagonist treatments was examined. Chronic ethanol exposure alters DA function both at the presynaptic and postsynaptic levels (Liljequist and Engel, 1979; Reggiani et al., 1980; Weiss et al., 1996; Nestby et al., 1997) and these changes may modify the effects of DA antagonist treatments on ethanol-seeking behavior. Therefore, effects of D1 and D2 receptor blockade on ethanol-seeking behavior were determined in the same animals during tests conducted before and after exposure to a chronic ethanol vapor inhalation procedure.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Eighteen male Wistar rats (Charles River Co., Kingston, NY), weighing 180 to 200 g at the beginning of experiment and 400 to 650 g at the time of testing, were used. Rats were housed in groups of three per cage in a temperature- and humidity-controlled vivarium on a normal 12-h light/dark cycle (on, 6:00 AM; off, 6:00 PM). Food and water were available ad libitum except the first 3 days of operant training (see Ethanol Self-Administration). All training and experimental sessions were conducted during the light phase at the same time each day (10:00 AM-1:00 PM). All experimental procedures were carried out in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Drugs. R-(+)-SCH23390 hydrochloride and S-()-eticlopride hydrochloride (Sigma, St. Louis, MO) were dissolved in sterile physiological saline and injected at a volume of 0.3 ml/kg. Drugs were administered s.c. 30 min before behavioral test sessions.

Ethanol Vapor Inhalation Procedure. Induction of ethanol dependence by ethanol vapor inhalation was performed using procedures modified from Rogers et al. (1979). Briefly, rats were housed in groups of three in sealed Plexiglas chambers where they were continuously exposed to an air/ethanol mixture for 12 days. Ethanol vapor was created by dripping 95% ethanol onto a 2000-ml Erlenmeyer vacuum flask kept at 50°C on a warming tray. Air was blown over the bottom of the flask at a rate of 11 l/min. Ethanol vapor was independently introduced into each chamber through a stainless steel manifold. The concentration of the ethanol vapor was adjusted within the range of 22 to 27 mg/l to produce targeted blood alcohol levels (BALs) of 180 to 200 mg/100 ml. To maintain body weight, rats were provided with a supplementary diet consisting of trail mix (nuts, seeds, raisins).

Blood Alcohol Determination. Tail blood samples were collected every other day for the measurement of BALs. Blood was collected in capillary tubes and emptied into Eppendorf tubes containing evaporated heparin and kept on ice. Samples were centrifuged, and serum was decanted into fresh Eppendorf tubes. The serum was injected into an oxygen-rate alcohol analyzer (Analox Instruments, Lunenburg, MA) for blood alcohol determination.

Withdrawal Sign Ratings. Eight to 12 h after termination of ethanol vapor inhalation, withdrawal signs, including the ventromedial distal limb reflexion response, tail stiffness, and abnormal body posture were recorded using a rating scale as previously described (Macey et al., 1996). A subjective 0- to 2-point scale was used for each of these signs, with 0 representing undetectable, 1 representing moderate, and 2 representing severe withdrawal sign. An overall withdrawal severity score ranging from 0 to 6 was derived by adding rating scores for the three individual withdrawal signs.

Ethanol Self-Administration Stations. Self-administration training was conducted in standard operant conditioning chambers located inside sound-attenuating, ventilated cubicles (Coulborn Instruments, Allentown, PA). Liquid solutions were dispensed by syringe pumps into a drinking reservoir positioned 4 cm above the grid floor in the center of one side panel as previously described (Weiss et al., 1993). A retractable lever was located 4.5 cm to the right of the drinking reservoir. Depression of the lever resulted in delivery of 0.1 ml of liquid. Delivery of fluids and data collection were controlled by an IBM-compatible microcomputer.

Ethanol Self-Administration. Animals were trained to self-administer 10% ethanol in 30-min daily sessions on a fixed ratio 1 schedule of reinforcement where each response resulted in delivery of 0.1 ml of fluid. For the first 3 days of training, the rats were placed on a restriction schedule limiting water availability to 2 h/day to facilitate acquisition of operant responding maintained by a liquid reinforcer. During this time, responses at the lever resulted in delivery of a 0.2% (w/v) saccharin solution into the drinking reservoir. During all subsequent training and testing, water was again made available ad libitum in the home cages. After acquisition of saccharin-reinforced responding, rats were trained to self-administer ethanol by using a modification of the sucrose-fading procedure (Samson, 1986) that used saccharin instead of sucrose as previously described (Weiss et al., 1993). During the first 6 days of ethanol self-administration training, responses at the lever resulted in presentation of a 0.2% saccharin (w/v) solution containing 5% (w/v) ethanol. Starting on day 7, the concentration of ethanol was gradually increased from 5 to 8% and finally 10% (w/v), whereas the concentration of saccharin was slowly decreased to 0%.

Conditioning and Extinction Procedures. Once reliable ethanol-reinforced responding at the 10% (w/v) dose was established, olfactory discriminative stimuli predictive of ethanol availability versus nonreward were introduced during all subsequent ethanol self-administration and nonreward sessions. The olfactory stimuli were generated by depositing 6 drops of discrete food flavor extracts (banana or anise) into the bedding of the operant conditioning chamber 1 min before extension of the lever and remained present throughout each 30-min session. For one-half of the rats, a banana-flavored extract served as the S⁺ to signal the availability of ethanol, whereas nonreward (i.e., ethanol nonavailability) was signaled by an anise extract (S). These pairings were reversed for the other half of the animals, with anise serving as the S⁺ for ethanol and banana odor serving as the S. The bedding was changed and bedding trays were thoroughly cleaned between sessions. Daily ethanol or nonreward sessions were scheduled in an unpredictable sequence to enhance the salience of the discriminative stimuli as reliable predictors of ethanol versus nonreward. Training was continued under these conditions until rats completed a total of 20 ethanol and 20 nonreward sessions.

Reinstatement Tests. Reinstatement tests began 1 day after the final extinction session. These tests lasted 30 min and were conducted under conditions identical to those during the conditioning phase of the experiment, except that ethanol was not made available. Sessions were initiated by extension of the lever and presentation of the respective S⁺ or S. The discriminative stimulus remained present during the entire session. Responses at the lever were followed by activation of the syringe pump motor but had no other programmed consequences. Responses at the lever and response latency (that is, the time from lever presentation to initiation of first response) were recorded by a microcomputer. Before the beginning of the reinstatement test phase, the animals were randomly divided into two drug treatment groups to study the effects of the D1 antagonist SCH23390 (0, 5, 10, 50 µg/kg s.c.) and the D2 antagonist eticlopride (0, 5, 10, 50 µg/kg s.c.) on responding under both the S⁺ and S contingencies. Test sessions were conducted daily in both drug treatment groups for eight consecutive days (four S⁺ sessions and four S sessions). S⁺ and S test conditions were arranged in a random sequence to eliminate possible order effects. Three doses and vehicle of each respective drug were administered in a counterbalanced manner 30 min before the behavioral test sessions.

Data Analysis. Differences in the number of responses in the two drug treatment conditions (SCH23390 and eticlopride) and for the two discriminative stimulus conditions (S⁺ and S) were analyzed separately by one-way ANOVA. Differences among individual means were subsequently verified by Newman-Keuls post hoc tests. To evaluate differences in drug potency between the pre- and postdependence tests, data were transformed into percentage scores representing the inhibition of responding compared with vehicle effects and analyzed by mixed factorial ANOVA with subsequent analysis of simple effects. Response latency data were analyzed by mixed factorial ANOVA, followed by simple effects ANOVA to test for differences between the pre- and postdependence tests.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Conditioning and Extinction Phases. After acquisition of ethanol (10%) self-administration at the end of saccharin fading procedure, ethanol-reinforced responding increased progressively while lever responses during the nonreward sessions gradually decreased. After 40 training days (i.e., 20 ethanol and 20 nonreward sessions) all rats developed stable levels of responding for ethanol. The mean (±S.E.M.) number of responses, averaged across the final three sessions of the conditioning phase, was 24.7 ± 2.2 during ethanol sessions and 6.8 ± 1.0 during nonreward sessions (Fig. 1). The difference in responding during ethanol versus nonreward sessions was confirmed by ANOVA [F(1.81) = 52.12; p < 0.001]. In the first extinction session, rats emitted an average (±S.E.M.) of 20.2 ± 6.4 responses. During subsequent sessions, responding gradually decreased, and all rats reached the extinction criterion in 16 days.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Mean (±S.E.M.) lever-press responses per 30-min sessions in rats (n = 18) during the last 3 days of the conditioning phase and a subsequent extinction phase. Conditioning phase: responses during ethanol self-administration () and nonreward () sessions in the presence of olfactory discriminative stimuli associated with ethanol availability versus nonreward. Extinction phase: lever-press responses remained without programmed consequences during this phase. All rats reached the extinction criterion of six or fewer responses per session for three consecutive days within 16 days.

Blood Alcohol Levels and Withdrawal Signs. The mean (±S.E.M.) BAL, averaged across the last three measurements, was 190.2 ± 21.3 mg/dl (range 138.2-268.8 mg/dl). Animals showed overt ethanol withdrawal signs 8 to 12 h after removal from ethanol vapor with a mean (±S.E.M.) withdrawal severity score of 4.2 ± 0.5.

Reinstatement Tests. No differences were observed as a function of the type of olfactory stimulus (banana or anise) that served as the S⁺ or S for ethanol. Specifically, in vehicle-injected rats the mean (±S.E.M.) number of responses elicited by the respective S⁺ was 12.5 ± 3.7 (banana) and 14.3 ± 2.9 (anise) in the SCH23390 group, and 16.3 ± 3.2 (banana) and 14.3 ± 5.6 (anise) in the eticlopride group. The respective S-induced responses were 4.7 ± 1.9 (anise) and 5.1 ± 1.1 (banana) in the SCH23390 group, and 4.7 ± 2.1 (anise) and 4.1 ± 1.9 (banana) in the eticlopride group. Therefore, the data were pooled across the two olfactory conditions in both drug treatment groups. Moreover, lever responses of vehicle-injected rats under the respective condition remained undiminished over the eight daily tests (in a random order for the S⁺ or S condition). There was no significant effect of testing day on either S⁺ [F(3,51) = 1.21, N.S.] or S responses [F(3,51) = 0.98, N.S.].

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Effects of olfactory discriminative stimuli associated with ethanol availability (S+) versus nonreward (S) on responding, and modification of these effects by the D1 antagonist SCH23390. Tests lasted 30 min and were conducted in nondependent rats before a 12-day ethanol vapor inhalation procedure (predependence) and, again, beginning on day 21 after termination of ethanol vapor exposure (postdependence). Rats were treated with SCH23390 (0, 5, 10, 50 µg/kg s.c.) 30 min before reinstatement tests in a counterbalanced order. For comparison, the figure shows also the mean (±S.E.M.) number of responses during ethanol self-administration () and nonreward () sessions collapsed across the last 3 days of the conditioning phase (SA), as well as the mean (±S.E.M.) number of extinction responses at criterion (EXT). *, p < 0.05; **, p < 0.01 different from extinction baseline; +, p < 0.05; ++, p < 0.01 different from vehicle [Newman-Keuls post hoc tests after ANOVA for predependence (F(4,40) = 3.76; p < 0.05) and postdependence (F(4,40) = 13.99; p < 0.01) conditions].

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Effects of olfactory discriminative stimuli associated with ethanol availability (S+) versus nonreward (S) on responding, and modification of these effects by the D2 antagonist eticlopride. Tests lasted 30 min and were conducted in nondependent rats before a 12-day ethanol vapor inhalation procedure (predependence) and, again, beginning on day 21 after termination of ethanol vapor exposure (postdependence). Rats were treated with eticlopride (0, 5, 10, 50 µg/kg s.c.) 30 min before reinstatement tests in a counterbalanced order. For comparison, the figure shows also the mean (±S.E.M.) number of responses during ethanol self-administration () and nonreward () sessions collapsed across the last 3 days of the conditioning phase (SA), as well as the mean (±S.E.M.) number of extinction responses at criterion (EXT). *, p < 0.05; **, p < 0.01 different from extinction baseline; +, p < 0.05; ++, p < 0.01 different from vehicle [Newman-Keuls post hoc tests after ANOVA for predependence (F(4,40) = 3.59; p < 0.05) and postdependence (F(4,40) = 4.93; p < 0.01) conditions].

Effects of SCH23390 on Reinstatement Responses and Response Latency. In the predependence tests, the D1 antagonist SCH23390 attenuated the response-reinstating effect of the S⁺ [F(4,40) = 3.76, p < 0.05]. This effect increased systematically with dose and was significant at 50 µg/kg (p < 0.05). SCH23390 did not alter responding in the presence of the S (Fig. 2, predependence), which remained at extinction levels at all doses tested [F(4,40) = 0.23, N.S.].

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Dose-effect curves for SCH23390 (top) and eticlopride (bottom) in the pre- () and postdependence () tests. Drug effects are expressed as mean (±S.E.M.) percentage of inhibition of S+-induced response-reinstatement compared with the effects of vehicle injection. *, p < 0.05; **, p < 0.01 different from predependence [simple effects after ANOVA of the SCH23390 (F(1,11) = 11.94; p < 0.01) and eticlopride (F(1,13) = 6.39; p < 0.05) effects].

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Effects of the D1 antagonist SCH23390 on the latency to initiate responding in the S+ conditions during pre- and postdependence tests. *, p < 0.05; **, p < 0.01 different from vehicle; +, p < 0.05; ++, p < 0.01 different from 5 µg/kg; #, p < 0.05; ##, p < 0.01 different from predependence [simple effects after ANOVA (F(1,16) = 15.64; p < 0.01)].

Effects of Eticlopride on Reinstatement Responses and Response Latency. The D2 antagonist eticlopride attenuated the response-reinstating effect of the S⁺ in the predependence tests. This effect varied systematically with increasing doses and was significant (p < 0.05) at all doses tested (Fig. 3, predependence). Responses in the S condition remained at extinction levels, and eticlopride did not alter responding in the presence of this stimulus as a function of dose [F(4,40) = 0.29, N.S.].

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Effects of the D2 antagonist eticlopride on response latency in the S+ conditions during pre- and postdependence tests. **, p < 0.01 different from vehicle; +, p < 0.05; ++, p < 0.01 different from 5 µg/kg; §§, p < 0.01 different from 10 µg/kg.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This series of experiments yielded three major findings. First, contextual stimuli associated with the availability of ethanol reliably elicited drug-seeking behavior as measured by the recovery of extinguished responding at a previously ethanol-paired lever in the absence of further availability of the drug reinforcer. Second, the effects of these stimuli on the initiation and maintenance of ethanol-seeking were dose dependently reversible by selective D1 and D2 receptor antagonists. Finally, both DA antagonists attenuated the drug-seeking response with greater potency in rats with a history of ethanol dependence compared with their effects in the same rats tested before the induction of dependence.

Recovery of responding after extinction of the ethanol-reinforced instrumental response was observed only in the presence of the ethanol-predictive discriminative stimuli (S⁺). Responses in the presence of the stimuli associated with nonreward (S) remained at extinction levels, confirming that the initiation and maintenance of ethanol-seeking behavior were controlled selectively by the ethanol S⁺. This finding replicates earlier observations that ethanol-related contextual stimuli reliably elicit responding at a previously active lever in rats trained to self-administer ethanol under conditions of limited daily access (Katner et al., 1999; Ciccocioppo et al., 2001a). One goal of the present study was to extend this earlier work to rats with a history of ethanol dependence to determine whether the motivating effects of ethanol-related contextual stimuli are further increased after induction of ethanol dependence and withdrawal. Contrary to the expectations, the response-reinstating effects of the ethanol S⁺ measured 3 weeks after withdrawal remained identical to those recorded in the predependent state, suggesting that ethanol-related stimuli do not acquire greater incentive salience in previously dependent rats. However, to properly interpret this finding it is necessary to consider that during the training phase of the experiment, the ethanol S⁺ was present only while the animals were in the nondependent state such that conditioning of this stimulus occurred only to the positive reinforcing effects of ethanol. Recent data suggest that ethanol-related contextual stimuli can acquire stronger response-reinstating actions in previously ethanol-dependent rats when these animals are permitted to operantly self-administer ethanol after removal from a chronic ethanol vapor (Liu and Weiss, unpublished observations). Thus, the demonstration of enhanced motivating effects of ethanol-related stimuli in previously dependent rats may require the opportunity to associate ethanol self-administration with the alleviation of withdrawal symptoms (i.e., the negative reinforcing aspects of ethanol's actions), a condition that was not met in the present design.

The central objective of this study was to confirm a dopaminergic involvement in the incentive motivational aspect of ethanol's actions by testing whether the initiation and maintenance of ethanol-seeking by drug-related stimuli is sensitive to pharmacological blockade of DA neurotransmission. Consistent with this hypothesis, both the D1 antagonist SCH23390 and the D2 antagonist eticlopride increased the latency to initiate responding and reduced the total number of responses during the reinstatement tests. Both of these effects were dose-dependent and were observed in both the predependence and postdependence tests. An issue for the interpretation of these findings is whether SCH23390 or eticlopride produced nonspecific impairments in motor performance. Some information relevant to this question can be gleaned from the effects of the DA antagonists on responding in the S condition. Although the primary purpose for the use of an S was not to control for nonspecific motor effects of the DA antagonists but to confirm the selectivity of the S⁺ in controlling behavior, the data obtained in this condition suggest that neither agent produced appreciable motoric impairments. Specifically, both in the predependence and postdependence condition, SCH23390 and eticlopride did not produce any decrement in S responses at doses that produced a profound reduction in S⁺ responses. Moreover, whereas S lever responses in the postdependence condition were somewhat decreased at the 10- and 50-µg doses of SCH23390 and eticlopride, these effects did not reach statistical significance. It is clear that the DA antagonist effects on S responses are somewhat limited in providing a conclusive control for motor artifacts because of the low response rates in this condition. Nonetheless, the interpretation that the attenuation of the motivating effects of the ethanol S⁺ cannot be attributed to DA antagonist-induced motoric impairments is supported also by the literature. SCH23390 at doses similar to those used herein attenuates conditioned operant responses and conditioned locomotor activity elicited or maintained by drug cues without impairing motor performance (Weissenborn et al., 1996; Smith et al., 2000; Bevins et al., 2001). In contrast to D1 antagonists, D2-selective antagonist agents are more commonly associated with nonspecific deficits in motor performance (Smith et al., 2000). However, the dose range at which eticlopride induces motoric impairments is typically higher than that in the present study (Bardo et al., 1999; Bevins et al., 2001). In light of this literature as well as the absence of significant DA antagonist effects on responding in the S condition, it seems unlikely that performance deficits contributed importantly to the attenuation of ethanol-seeking behavior by SCH23390 and eticlopride.

The results, then, suggest that blockade of D1 or D2 receptors attenuates the conditioned incentive effects of ethanol-related contextual stimuli both in terms of their ability to elicit alcohol-seeking behavior and their efficacy in maintaining drug-seeking responses once initiated. This finding extends ample existing evidence for a role of DA in conditioned responses associated with natural and drug rewards (Robbins and Everitt, 1996; Beninger and Miller, 1998) to the motivating effects of ethanol-related stimuli. It is important, however, to consider also two nonmotivational alternative explanations for the reduction in ethanol-seeking behavior by the two DA antagonists. First, these agents may have interfered with associative processes (e.g., memory retrieval) that normally activate motivational systems, rather than disrupting these systems per se. Although this possibility cannot be ruled out on the basis of the present data, the literature suggests that D1 and D2 antagonists do not reliably interfere with the expression of learned responses, including conditioned fear (Inoue et al., 1996; Greba and Kokkinidis, 2000), conditioned taste preference (Azzara et al., 2001), and ethanol-induced conditioned place preference (Cunningham et al., 1992). Also, microinjections of a D2 antagonist (raclopride) into the basolateral amygdala, a critical substrate for associative learning and memory, do not interfere with conditioned cocaine-seeking behavior (See at al., 2001). Most importantly, it has been shown that SCH23390 can attenuate the effects of a cocaine-paired stimulus on conditioned responding while increasing conditioned responding maintained by a food-associated stimuli at an alternative lever (Weissenborn et al., 1996). Thus, interference with retrieval of learned associations does not appear likely to account for the inhibitory effects of the DA antagonists on responding elicited by the ethanol S⁺. A second, alternative explanation for these actions of SCH23390 and eticlopride involves possible "state-dependent learning" effects. According to this interpretation, the DA antagonists may have produced a change in the rats' internal stimulus state such that performance of the learned response to the ethanol S⁺ was diminished due to a decrement in stimulus generalization. This account would predict, however, that any pharmacological agent inducing a salient change in the animals' interoceptive state would produce a degradation of conditioned drug-seeking responses similar to that produced by DA antagonists herein. Contrary to this prediction, pharmacological agents can share the ability to produce salient changes in an animal's internal stimulus state (i.e., compounds with measurable discriminative stimulus properties) but, nonetheless, produce distinctly different effects on behavioral responses elicited by ethanol cues. For example, the nonspecific D1/D2 antagonist haloperidol, but not a serotonin 5-HT_2A/5-HT_2C agonist reduced responding for presentation of an ethanol-paired conditioned stimulus (Wilson et al., 2000). Similarly, a nonselective opiate antagonist (naltrexone) but not a corticotropin-releasing factor antagonist (D-Phe-corticotropin-releasing factor) blocked the response reinstatement induced by a previously ethanol-paired stimulus, although the latter agent effectively antagonized ethanol-seeking responses elicited by foot shock (Weiss and Liu, 2001). State-dependent learning effects, therefore, do not seem to provide a parsimonious explanation for the DA antagonist effects in the present study.

The results revealed a substantially greater effect on S⁺-induced responding by SCH23390 and eticlopride in the postdependence than in the predependence condition. It is unlikely that this effect reflects some erosion in the predictive or reinforcement value of the S⁺ produced by the second set of extinction sessions that preceded the postdependence tests because the response-reinstating effects of the S⁺ in vehicle-injected rats were identical in the predependence and postdependence tests. This explanation seems unlikely also in light of evidence that repeated extinction (Ciccocioppo et al., 2001b) or repeated nonreinforced testing with a cocaine S⁺ (Weiss et al., 2001) does not alter the motivating effects of this cue, and more recent data have shown that this applies to ethanol-associated discriminative stimuli as well (Ciccocioppo et al., 2001a). Alternatively, it is possible that rats with a recent history of ethanol dependence and withdrawal are more sensitive to stimuli that disrupt ongoing behavior such that the enhanced inhibitory effects of SCH23390 and eticlopride on ethanol-seeking behavior may reflect hypersensitivity to disruptive stimuli in general rather than a true increase in the potency of the DA antagonists. However, it was recently shown that in contrast to the DA antagonist effects herein, the potency of a nonselective opiate antagonist (naltrexone) to inhibit S⁺-induced ethanol-seeking was significantly reduced in postdependent rats (Weiss and Ciccocioppo, 1999). In particular, a naltrexone dose that was behaviorally effective in nondependent rats failed to reduce reinstatement responses in postdependent rats, demonstrating that drug administration by itself does not exert behaviorally disruptive effects in recently ethanol-withdrawn rats. A likely explanation for the increased antagonist potency in postdependent rats, therefore, involves changes in DA function at both the pre- and postsynaptic level known to be produced by chronic ethanol treatment. Ethanol withdrawal is associated with a deficiency in extracellular DA levels within the NAcc and decreases in the firing rate of ventral tegmental DA neurons that may outlast physical withdrawal and persist into the protracted abstinence phase (Diana et al., 1996; Weiss et al., 1996). Chronic ethanol treatments also can produce long-lasting supersensitivity (i.e., increased binding affinity) of postsynaptic DA receptors (Liljequist and Engel, 1979; Reggiani et al., 1980). Moreover, D1-dependent stimulation of cyclic AMP was shown to be enhanced 3 weeks after cessation of ethanol self-administration, and the magnitude of this effect was correlated with the amount of daily ethanol consumption (Nestby et al., 1999). Thus, long-lasting increases in DA antagonist binding and DA-dependent signal transduction coupled with a reduction in synaptic availability of DA at the time of testing may have been factors contributing to the increased DA antagonist potency in the postdependence condition.

A question that remains concerns the site of action for the effects of SCH23390 and eticlopride observed herein. Responses conditioned to drug or natural rewards typically are associated with increased DA release in the NAcc (Weiss et al., 1993; Katner et al., 1996; Gonzales and Weiss, 1998; Bassareo and Di Chiara, 1999), suggesting that the NAcc may be the neuroanatomical substrate for the DA antagonist effects. However, there is growing evidence pointing toward the basolateral amygdala and prefrontal cortical regions, both of which are innervated by ventral tegmental DA neurons as important sites for the mediation of drug-seeking behavior induced by drug cues (Ciccocioppo et al., 2001b; See et al., 2001). Thus, critical sites for the response-reinstating actions of ethanol cues will need to be explored in the future by using site-specific administration of DA antagonists and lesioning approaches.

In summary, the results confirm that ethanol-related contextual stimuli reliably elicit drug-seeking behavior and suggest that this effect depends on activation of DA neurotransmission. In addition, the results demonstrate that chronic high-dose ethanol exposure produces changes in both D1 and D2 receptor function that lead to enhanced sensitivity to the behavioral effects of antagonists for these receptors.