Nicotine increases alcohol self-administration and reinstates alcohol seeking in rats
- First Online:
- Cite this article as:
- Lê, A.D., Wang, A., Harding, S. et al. Psychopharmacology (2003) 168: 216. doi:10.1007/s00213-002-1330-9
- 354 Views
Rationale and objective
Alcohol and tobacco are often co-abused in humans and previous studies found that nicotine increases alcohol consumption in rats. Here, we studied whether nicotine would reinstate alcohol-taking behavior in drug-free rats and whether this effect would be enhanced by prior exposure to nicotine during alcohol self-administration training.
Rats were trained to press a lever for alcohol (12% w/v, 1 h/day), and following stable alcohol intake groups of rats (n=11–12) were given daily vehicle or nicotine (0.2, 0.4 or 0.8 mg/kg, SC) injections just prior to the self-administration sessions for 10 days. Rats were then given 6 days of alcohol self-administration in the absence of nicotine and an additional 5–10 drug-free days during which lever presses were not reinforced (extinction). Subsequently, rats were tested for reinstatement of alcohol seeking following exposure to priming injections of vehicle or nicotine (0.4 mg/kg, SC).
Nicotine increased alcohol self-administration in a dose- and time-dependent manner over the 10-day period. Nicotine also reinstated alcohol seeking after extinction of the alcohol-reinforced behavior, and this effect was strongly enhanced by prior nicotine exposure.
The present data extend previous studies on the effect of nicotine on alcohol self-administration, and further indicate that nicotine is an effective stimulus for reinstatement of alcohol seeking during drug-free periods.
Alcohol and nicotine are the two most frequently used drugs and they are often taken together. In humans, epidemiological studies have found high correlations between alcohol consumption and tobacco use, and smoking prevalence in alcoholics is approximately three times higher than in the general population (Istvan and Matarazzo 1984; Sobell et al. 1990). In laboratory rats, several studies reported that repeated treatment with nicotine increases home-cage consumption (Potthoff et al. 1983; Blomqvist et al. 1996; Smith et al. 1999; Olausson et al. 2001) and operant self-administration (Lê et al. 2000a; Clark et al. 2001) of alcohol. In contrast, mecamylamine, a nicotinic receptor antagonist, decreases home-cage consumption (Blomqvist et al. 1996) and operant self-administration (Lê et al. 2000a) of alcohol.
The preclinical studies described above concentrated on the effect of nicotine on alcohol consumption and operant self-administration. At present, however, no studies have examined whether exposure to nicotine induces resumption of alcohol seeking (relapse) during drug-free periods in an animal model. In recent years, several laboratories have used a reinstatement model to study factors involved in relapse to alcohol seeking (Lê and Shaham 2002; Weiss and Porrino 2002). In this model, the effect of exposure to drug or non-drug stimuli on reinstatement of drug seeking is examined after training for drug self-administration and subsequent extinction of the drug-reinforced behavior (de Wit and Stewart 1981; Stewart and de Wit 1987; Shalev et al. 2002). In the present report, we used this reinstatement model to study whether nicotine would reinstate alcohol seeking after extinction of the alcohol-reinforced behavior. We also studied the effect of prior exposure to nicotine during alcohol self-administration training on nicotine-induced reinstatement of alcohol seeking during a drug-free period.
Materials and methods
Male Long-Evans rats (Charles River, Montreal, Quebec, Canada) weighing 175–200 g at the start of the experiment were used. Rats were housed singly under 12:12 h light-dark cycle (light on from 7:00 a.m. to 7:00 p.m). Food and water were freely available and rats were allowed 1–2 weeks to acclimate to the housing conditions prior to training for alcohol drinking in a two-bottle choice limited-access procedure. The experimental procedures followed the "Principles of laboratory animal care" (NIH publication no. 85-23, 1996) and were approved by the local animal care committee.
The self-administration chambers were constructed by the institution's technical staff. Each chamber was equipped with two levers, symmetrically centered on the side panel. Responding on one lever (an active lever) activated the infusion pump (Razel Scientific, Stamford, Conn., USA). Presses on the other lever (an inactive lever) were recorded, but did not activate the pump. Activation of the infusion pump resulted in the delivery of 0.19 ml of a 12% alcohol w/v solution in a liquid drop receptacle located between the two levers over a period of 5 s. During the infusion, a stimulus light above the active lever was turned on and lever presses were not reinforced (a timeout period of 5 s).
Rats were trained to self-administer alcohol according to methods described previously (Lê et al. 1998, 1999; 2000b, 2002). Briefly, rats were initially provided with access to alcohol solutions and water in modified Richter tubes for 30 min/day in drinking cages. Alcohol solutions were provided in escalating concentrations: 3% for the first 5 days, 6% for the next 5 days and 12% for the last 10 days. Eight rats that consumed less than 0.4 g/kg ethanol during the last 10-day period were excluded. Subsequently, operant self-administration of alcohol was initiated on a fixed ratio-1 schedule of reinforcement (FR1, each lever press is reinforced) for 10 days (1 h/day). The requirement for alcohol delivery was then increased to an FR2 schedule for five sessions. Subsequently, the schedule requirement was increased to an FR3 schedule until the rats had 3 days of stable alcohol taking (<20% deviation from the mean). Rats were then injected for 4 days (sessions 17–20 under the FR3 schedule) with saline (0.5 ml/kg, SC, 15 min prior to the daily self-administration sessions) in order to habituate them to the injection procedure.
Forty-six rats that demonstrated stable alcohol self-administration during the saline-pretreatment phase were used. Rats were divided into four groups (n=11-12), matched for their alcohol self-administration behavior and were assigned to either the saline or the nicotine (0.2, 0.4 or 0.8 mg/kg, SC) conditions. Nicotine (Sigma, St Louis, Mo., USA) was dissolved in saline (pH adjusted to 7.0) and doses are expressed as the free base. Daily injections of saline or nicotine were given for ten daily sessions. Nicotine or saline injections were given 15 min before the daily sessions. Drug treatment was then terminated and all rats received saline injections 15 min prior to the alcohol self-administration sessions for 6 additional days prior to the start of extinction phase. During the extinction phase, the conditions were the same as in training, with the exception that presses on the active lever were not reinforced with alcohol. Extinction sessions were performed for 5–10 days until the rats reached an extinction criterion of 15 presses or less on the previously active lever per hour (two rats that failed to reach the extinction criterion following 10 days of extinction were excluded). Rats were injected with saline 15 min prior to the start of the extinction sessions. On the day after reaching the extinction criterion, rats were tested for the effect of the middle dose of nicotine (0.4 mg/kg, SC) on reinstatement of alcohol seeking under extinction conditions.
Data from the self-administration phase were analyzed with repeated measures ANOVA, using a between-subjects factor of Nicotine pretreatment (saline, low, medium or high dose) and a within-subjects factor of Session (ten sessions of alcohol self-administration). Data from the tests for reinstatement were analyzed with repeated measures ANOVA, using the between-subjects factor of Nicotine pretreatment (saline, low or high dose during the alcohol self-administration phase) and the within-subjects factor of Nicotine priming (saline or 0.4 mg/kg nicotine). The last day of extinction served as the baseline, saline condition, against which the effect of nicotine priming on lever-pressing behavior was compared. Significant effects (P values <0.05) were followed by post-hoc (Newman-Keuls) tests.
Effect of nicotine on alcohol self-administration
Effect of nicotine on reinstatement of alcohol-seeking
We studied the effect of nicotine on alcohol self-administration and reinstatement of alcohol seeking after extinction of the drug-reinforced behavior in Long-Evans rats. We found that repeated exposure to the low and medium doses of nicotine had a modest effect on alcohol self-administration, while exposure to the high dose of nicotine resulted in time-dependent increases in alcohol self-administration, leading to intake levels that were more than twice of those of the saline-treated group on days 7–10 of treatment (Fig. 1a). These data extend those from our previous study with Wistar rats in which we found a modest effect of nicotine on operant alcohol self-administration (about 50% increases with the 0.8 mg/kg dose) (Lê et al. 2000a). More important, our data provide the first demonstration that nicotine can reinstate alcohol seeking in a rat model of drug relapse (Stewart 2000; Shaham et al. 2003) and that this effect on reinstatement is strongly enhanced by prior exposure to nicotine during alcohol self-administration training (Fig. 2a). Interestingly, this enhancement of nicotine-induced reinstatement also was observed in rats pre-exposed to the low and moderate nicotine doses, which had a modest effect on alcohol self-administration. In the sections below, we discuss (1) methodological considerations, and (2) neuronal mechanisms that may be involved in the effect of nicotine on alcohol-taking behavior.
Several methodological issues should be considered in the interpretation of the present data. First, repeated nicotine exposure is known to induce locomotor sensitization (Morrison and Stephenson 1972; Clarke and Kumar 1983a, 1983b; Kelsey et al. 2002). Thus, the time-dependent increases in alcohol self-administration following repeated nicotine exposure and the enhancement of nicotine-induced reinstatement of alcohol seeking by prior nicotine exposure may be due to non-selective increases in locomotor activity. This possibility, however, is unlikely because nicotine injections during the self-administration or the reinstatement phases had no effect on responses on the inactive lever, which was not associated with alcohol self-administration during training. In addition, there is no evidence from studies on the effect of nicotine on lever pressing maintained by food under fixed-ratio schedules of reinforcement that repeated nicotine administration increases rate of responding above baseline levels (Goldberg et al. 1989; Lau et al. 1994; Shoaib et al. 1997).
Second, the increase in alcohol self-administration and reinstatement of alcohol seeking by nicotine may be due to the "anxiogenic" effects of nicotine (File et al. 1998; Irvine et al. 2001). This possibility, however, is unlikely because several studies reported that exposure to "anxiogenic" agents such as Ro15-4513 or pentylenetrazol do not increase alcohol consumption (Petry 1995; Buczek et al. 1997, 1998). In addition, our attempts to mimic the effect of footshock stress on reinstatement of heroin or alcohol seeking (see Lê et al. 1998; Shaham et al. 2000) with "anxiogenic" compounds such as FG-7142 or pentylenetetrazole were not successful (unpublished data).
Third, it is possible that the effect of nicotine on alcohol self-administration may be due to the pharmacokinetic effects of nicotine on alcohol availability. This possibility, however, is unlikely. A study by Hisaoka and Levy (1985) indicates that chronic nicotine exposure does not alter alcohol pharmacokinetics.
Fourth, prior exposure to nicotine may enhance nicotine-induced reinstatement of alcohol seeking because during self-administration training nicotine has become a discriminative alcohol cue that sets the occasion for alcohol availability. While this possibility cannot be ruled out by the present data, it is relatively unlikely. Nicotine injections were only given for 10 days after the alcohol self-administration behavior has been established, and the groups exposed to nicotine also were exposed to saline injections for 10 days during the alcohol self-administration phase (4 days prior to and 6 days after nicotine treatment). In addition, while discriminative alcohol cues can reinstate alcohol seeking following explicit discrimination training (Weiss and Porrino 2002), the magnitude of responding following exposure to these cues during tests for reinstatement (Katner and Weiss 1999; Katner et al. 1999) is much lower than the effect of nicotine on reinstatement observed here.
Finally, only one dose of nicotine was used during the test for reinstatement of alcohol seeking. Thus, the greater magnitude of responding following nicotine priming in the nicotine-treated rats may be due to a shift to the right in the dose-response curve of these rats, resulting in increased responding for a dose located on the descending limb of the curve (see Yokel 1987). We argue, however, that this possibility is relatively unlikely. The priming nicotine dose (0.4 mg/kg) increased alcohol self-administration (Fig. 1) and there is a general agreement in the alcohol field that increased alcohol intake by pharmacological or environmental manipulations is most likely due to increases (rather than decreases) in the reinforcing effects of alcohol (Spanagel and Holter 2000; Lê and Shaham 2002). Thus, to the degree that the mechanisms underlying alcohol reinforcement and reinstatement by pharmacological agents overlap (Lê and Shaham 2002), it is not likely that increased responding in the nicotine-treated rats following priming injections of nicotine is a due to a shift to the right in the dose-response curve of these rats.
Mechanisms underlying the effect of nicotine on alcohol-taking behavior
The mechanisms underlying the robust effect of priming injections of nicotine on reinstatement of alcohol seeking in rats previously treated with nicotine are not known, but it is likely that activation of the mesolimbic dopamine (DA) reward system (Wise and Rompre 1989; Di Chiara 1995) is involved in this effect. Systemic injections of nicotine increase DA utilization in the nucleus accumbens (NAc) (Clarke et al. 1988; Di Chiara and Imperato 1988) and other mesolimbic terminal regions (Di Chiara 2000). In addition, repeated nicotine injections result in augmented release of DA in the NAc (Balfour et al. 1998; Shim et al. 2001). The mesolimbic DA neurons also are critically involved in reinstatement of heroin and cocaine seeking induced by priming injections of the self-administered drugs (Stewart 1984; Self and Nestler 1998; De Vries et al. 1999; Stewart 2000; Shalev et al. 2002).
It is also possible that nicotine-induced activation of the mesolimbic DA system is involved in the increase in alcohol self-administration induced by repeated nicotine injections. Both nicotine (see above) and alcohol (Di Chiara and Imperato 1988; Weiss et al. 1993) increase DA release in the NAc. In addition, there is evidence that alcohol-induced DA release in the NAc is mediated by activation of nicotinic receptors in the ventral tegmental area (VTA) (Blomqvist et al. 1997; Soderpalm et al. 2000), the cell body region of the mesolimbic DA system (Fallon and Moore 1978). Soderpalm et al. (2000) also reported that blockade of nicotinic receptors in the VTA by mecamylamine decreases alcohol consumption and that repeated treatment with nicotine increases alcohol-induced DA release in the NAc. However, there is also evidence that may not be consistent with the hypothesis that mesolimbic DA neurons contribute to the effect of nicotine on alcohol self-administration. Specifically, several studies have found that DA receptor antagonists (Brown et al. 1982; Goodwin et al. 1996) or 6-OHDA lesions of mesolimbic DA neurons (Rassnick et al. 1993) have no effect on alcohol consumption, but see Rassnick et al. (1992) and Bono et al. (1996). Thus, studies on the effect of systemic and intracranial injections of DA receptor antagonists on nicotine-induced increases in alcohol self-administration and reinstatement of alcohol seeking are needed in order to support (or refute) the hypothesis that activation of the mesolimbic DA system is involved in these effects.
The present data and the recent data of Clark et al. (2001) on nicotine-induced increases in responding during extinction, a behavior controlled by the reward-associated cues (Pavlov 1927), may have clinical implications for the treatment of relapse to alcohol in humans. The pertinent clinical issue is whether alcoholics undergoing treatment for the prevention of relapse to alcohol should also receive concurrent treatment for smoking cessation (Kozlowski and Ferrence 1990; Sobell et al. 1990; Hughes 1993). Thus, to the degree that data from studies using a reinstatement model can be generalized to humans (Shaham et al. 2003), the data from the present study and from Clark et al.'s (2001) study suggest that treating nicotine addiction in recovering alcoholics might lead to lower rates of relapse to alcohol.
The experiments were supported by funds from the Ontario Mental Health Foundation and the NIAAA/NIH. We thank Dr. Paul Clarke for helpful comments.