Introduction

Nicotine, the psychoactive component in tobacco, is a potently addictive substance in humans and other animal species.^1,2 Indeed, smoking remains the most serious cause of preventable death in the developed world.³ Although considerable progress has been made in elucidating the neuroanatomical and neuropharmacological substrates mediating the psycho-active properties of nicotine, the precise neural substrates responsible for the rewarding and addictive properties of nicotine have not previously been identified.

Interestingly, while the prevalence of tobacco use among the general population has remained relatively stable (25–30%),⁴ the prevalence of nicotine addiction is highly elevated among schizophrenic psychiatric populations (90%).² This fact represents a serious challenge to the pharmacological treatment of the schizophrenic syndrome. While traditional neuroleptic drugs have proven highly efficacious in the treatment of psychotic symptoms associated with schizophrenia, the exceedingly high rates of nicotine addiction found in these patients represent serious co-morbid health risk factors.

Previous studies have suggested that the meso-limbic dopamine (DA) pathway, which originates from DA cell bodies in the ventral tegmental area (VTA) and projects to the nucleus accumbens (NAc), may be involved in mediating the rewarding effects of nicotine.^5,6 For example, neurotoxic lesions of the mesolimbic system or systemic administration of DA receptor antagonists reduce nicotine self-administration in rodents.^5,6 However, interpretations of this intravenous drug intake are complicated by the fact that mesolimbic DA receptor blockade increases intake of other drugs of abuse, most notably cocaine.^6,7 Furthermore, the suggestion that DA mediates a nicotinic reward signal is incongruent with clinical studies on both normal and schizophrenic human subjects that have demonstrated that DA receptor antagonists potentiate nicotine intake and smoking behaviours.^2,8 DA receptor agonists reduce nicotine intake and cigarette cravings in humans,^9,10 suggesting that activation of DA receptors may in fact counteract nicotine's reinforcing properties. Recently, the mesolimbic DA system has been implicated in the mediation of aversive motivational signalling,^11,12 suggesting that mesolimbic DA signalling may be involved in the transmission of aversive neural motivational information. Furthermore, decreased levels of mesolimbic DA receptors have been correlated with higher levels of both nicotine and amphetamine abuse.^13,14 These findings suggest that perturbations in mesolimbic DA transmission may induce a unique, drug-vulnerable phenotype for the rewarding and addictive properties of psychostimulant drugs of abuse.

Materials and methods

Surgical procedures

Male Wistar rats (Charles River, 300–350 g at the start of the experiments) were anaesthetized with sodium pentobarbitol (Somnotol) (0.8 ml/kg) and placed in a stereotaxic device. Stainless steel guide cannulae of 22 gauge (Plastics One) were bilaterally implanted at a 10° angle, 1.5 mm dorsal to central injection sites. The following stereotaxic coordinates were used. For the VTA: from bregma, AP-5.0, L2.3; from the dural surface, V-8.0. Control placements dorsal to the VTA used the following coordinates: AP-5.0, L2.3; from the dural surface V-6.5. Unilateral cannulae placements in the interpeduncular nucleus used the following coordinates: from bregma, AP –7.0, L0.0; from the dural surface, V-8.6. For the nucleus accumbens: from bregma, AP+1.8, L3.1; from the dural surface, V-6.8. Control placements dorsal to the NAc used the following coordinates: AP+1.8, L3.1; from the dural surface, V-5.8. At the conclusion of the experiments, animals were deeply anaesthetized and transcardially perfused with isotonic saline followed by 10% formalin. Brain sections were stained with cresyl violet and cannulae placements were verified with light microscopy.

Drug treatments

Nicotine-di-d-tartrate, di-hydro--erythroidine (Research Biochemicals), morphine sulphate (BDH), and -flupenthixol (Research Biochemicals) were dissolved in physiological saline (pH adjusted to 7.4). Bilateral microinfusions (0.5 l volume per infusion) were performed over 1 min and injectors were left in place for a further 1 min post-injection to ensure diffusion from the injector tip. We chose a systemic dose of -flupenthixol (0.8 mg/kg; i.p.), which has been shown to antagonize both D1 and D2 post-synaptic DA receptors¹⁵ and to block the rewarding effects of amphetamine¹⁶ and of systemic or intra-VTA opiates, in opiate-dependent and withdrawn animals.^16,17 We chose a dose of -flupenthixol (3 g (5 nmol)/0.5 l) for bilateral intra-NAc microinfusions, which has similarly been demonstrated to block the rewarding properties of both opiates and amphetamine.^18,19 Systemic -flupenthixol injections (0.8 mg/kg; i.p.) were administered 2.5 h prior to receiving intra-VTA nicotine or saline infusions. For all experiments, bilateral intra-NAc microinfusions of either -flupenthixol (3 g/0.5 l) or saline were administered 15 min prior to receiving bilateral intra-VTA nicotine or saline microinfusions.

Place conditioning

All animals were conditioned using a standard place-conditioning procedure. Conditioning took place in one of two environments, which differed in colour, texture, and smell. One environment was white with a wire mesh floor that was covered in wood chips. The other environment was black with a smooth Plexiglas floor that was wiped down with a 2% acetic acid solution before each conditioning session. Animals display no baseline preference for either of these two environments (Figure 2a). Animals received four drug-environment and four saline-environment conditioning sessions and exposure to environments was fully counterbalanced in all experiments. All neurotransmitter blocking drugs were administered before both drug and saline injections. At testing, 1 week after the end of conditioning (all animals were tested drug free), animals were placed on a narrow, neutral grey zone that separated the two test compartments. Times spent in each environment were scored separately for each animal. All data were analysed with one- or two-way ANOVA or student's t-tests where appropriate. Post hoc analyses were performed with Newman–Keuls or Bonferroni tests where appropriate.

Figure 2.

Motivational effects of nicotine in the VTA, anatomical and pharmacological control groups. Motivational effects of systemic (0.8 mg/kg; i.p.) (2.5 h post-injection) or intra-NAc (3 g/0.5 l) (15 min post-injection) -flupenthixol pretreatments. Data for this and all subsequent graphs are expressed as difference scores (time (s) spent in drug minus saline-paired environments) s.e.m.. Asterisks (*) indicate significant effects (P<0.05). (a) Animals displayed a significant preference for nicotine-paired environments at the 2, 8, 24, and 48 nmol doses (all P<0.05), and a significant aversion to the nicotine-paired environment at the lower dose of 0.008 nmol nicotine (P<0.05). Inset shows times (s) spent in white vs black conditioning environments following intra-VTA saline in both environments. Animals spent equal time in both environments (P>0.05), revealing no baseline bias for either of the two conditioning environments. (b) Animals displayed no preference for environments paired with microinfusions of nicotine 1.5 mm dorsal to the VTA (P>0.05). Similarly, unilateral microinfusions of nicotine (16 nmol) into the interpeduncular nucleus produced no preference for the nicotine-paired environment (P>0.05). (c) Motivational effects of -flupenthixol pretreatment vs saline: neither systemic (0.8 mg/kg; i.p.) nor intra-NAc -flu (3 g/0.5 l) produced any motivational effects (all P>0.05).

Full figure and legend (43K)

Conditioned taste aversion

Animals were trained to consume water on a limited-access regimen of 15 min/day for 5 consecutive days. On this regimen, animals maintain 80% of their initial body weight. Training consisted of five exposures over a 10-day period to an unsweetened 0.3% solution of either grape- or cherry-flavoured Kool-Aid™ (animals display no baseline preference for either flavour) for 15 min (all conditioning took place in the home cage), after which animals received an intra-VTA microinfusion of nicotine. On intervening days, the alternate flavour was presented for 15 min, following which animals received an intra-VTA microinfusion of nicotine or saline. After completion of training, animals were left untreated for 2 days, during which time they were given 60 min of normal water access per day. On the test day, animals were presented with both the drug and saline-paired flavours (animals were tested drug free), and amounts consumed of both flavours over 20 min were recorded. For intra-NAc -flupenthixol conditioned taste aversion (CTA) experiments, the identical protocol was used, but animals received intra-NAc -flupenthixol or saline immediately prior to exposure to the flavours, and thus 15 min prior to intra-VTA saline or nicotine. For lithium chloride (LiCl) control experiments, the identical protocol was used, but animals received LiCl (15 mg/kg; i.p.) or saline following exposure to the flavours.

Drug discrimination experiments

We used a discriminated taste avoidance paradigm to measure the discriminative properties of nicotine. Animals are trained on a limited water-access schedule as previously described. During a conditioning cycle, animals receive an injection of either drug or saline and are then given 15 min exposure to a 0.1% saccharin solution. Following this, animals are immediately injected with either a highly aversive dose of LiCl (130 mg/kg; i.p. dissolved in 3 ml saline) or the equivalent volume of saline. In the present experiments, nicotine (either intra-VTA (0.8 nmol) or systemic (0.8 mg/kg; s.c.)) preceded the aversive LiCl injection, while on the intervening day a saline injection (either intra-VTA or s.c.) preceded an injection of saline (drug exposure order was counterbalanced within groups). Animals thus learn that consumption of the saccharin solution in the presence of the nicotine cue will lead to an aversive consequence, while saccharin consumption in the presence of the saline cue will not lead to an aversive consequence (saline injection). When the drug cues have gained discriminative stimulus control over responding, animals display cycling patterns of saccharin intake, wherein saccharin intake is gradually suppressed in the presence of the LiCl-predictive (aversive) cue, and increased in the presence of the saline-predictive (non-aversive) cue.

Results

Histological analysis

Histological analysis revealed that cannulae placements were within the anatomical boundaries of the various neuroanatomical microinjection sites. Animals found to have placements outside of these boundaries were excluded from the study. Figure 1a shows a cresyl violet stained microphotograph of a representative intra-VTA cannulae placement and bilateral microinjector tip placements. Figure 1b shows a schematic of representative intra-NAc cannulae placements. Intra-NAc cannulae placements were distributed in both the core and shell regions of the NAc; however, no discernible behavioural differences were related to these placement distributions.

Figure 1.

Histological analysis of intra-VTA and intra-NAc bilateral cannulae placements. (a) Photomicrograph of a cresyl-violet-stained coronal section showing representative bilateral cannulae placements in the VTA. Arrows indicate cannulae tip placements. (b) Histological verification of bilateral cannulae placements in the nucleus accumbens (2.2, 1.7, or 1.0 mm rostral to bregma), for animals receiving 0.008–8 nmol intra-VTA nicotine, dorsal NAc placements, and intra-NAc cannulae placements for intra-VTA nicotine (0.8 nmol) CTA experiments.

Full figure and legend (205K)

Motivational effects of nicotine in the VTA

In order to examine the possible role of the VTA in nicotine's psychoactive effects, we performed discrete, bilateral microinfusions of nicotine directly in the VTA (Figure 1a). Using a fully counterbalanced, unbiased place-conditioning paradigm, we performed an extensive dose–response analysis covering a six-order-of-magnitude nicotine concentration range (a separate group of animals was used for each dose). We found that nicotine in the VTA produced biphasic motivational effects (Figure 2a). Analysis of variance revealed a significant effect of nicotine on times spent in nicotine relative to saline-paired environments (F^9,99=10.5; P<0.05). Post hoc analysis revealed that animals displayed a robust place preference for nicotine-paired environments at the 2 (n=5), 8 (n=8), 24 (n=9), and 48 (n=7) nmol doses (all P<0.05). However, animals demonstrated a significant aversion to environments paired with a lower dose of nicotine (0.008 nmol) (n=6) (P<0.05). A separate control group of animals received saline in both the white and black environments. Animals spent equal amounts of time in both environments (n=5) (t⁴=1.55; P>0.05), revealing no baseline bias for either environment. The highest intra-VTA nicotine dose tested (80 nmol) produced inconsistent conditioning effects and seizures in some animals, suggesting a lack of behavioural specificity at this concentration (Figure 2a). The behavioural effects of intra-VTA nicotine were anatomically specific to the neuroanatomical boundaries of the VTA: bilateral infusions of nicotine (8 nmol) 1.5 mm dorsal to the VTA produced no preference for either environment (n=8) (t⁷=0.8; P>0.05) (Figure 2b). Furthermore, single, midline infusions of nicotine (16 nmol) into the interpeduncular nucleus (a midline structure located caudal to the VTA that contains a dense population of nicotinic receptors) did not produce any measurable behavioural effects (n=7) (t⁶=0.87; P>0.05) (Figure 2b). These behavioural effects of intra-VTA nicotine are pharmacologically specific to VTA nicotinic acetylcholine receptors: the rewarding and aversive motivational effects of intra-VTA nicotine (0.008–48 nmol) are completely blocked by the competitive nicotinic antagonist di-hydro--erythroidine.²⁰

DA receptor blockade increases nicotine reward sensitivity

We next challenged the rewarding effects of nicotine in the VTA with a high dose of the broad-spectrum DA receptor antagonist -flupenthixol (a classical neuroleptic) either systemically (0.8 mg/kg). or with bilateral microinfusions directly into the NAc (3 g/0.5 l) (Figure 1b).

To control for any potential unconditioned motivational effects of either the systemic or intra-NAc -flupenthixol administration, the behavioural effects of both of these treatments were independently tested in separate groups of animals. Neither systemic (n=8) (t⁷=0.9; P>0.05) nor intra-NAc -flupenthixol (n=6) (t⁵=0.3; P>0.05) pretreatment alone produced any motivational effects (neither reward nor aversion) (Figure 2c). We next challenged the motivational effects of intra-VTA nicotine with either systemic or intra-NAc -flupenthixol. Separate groups of animals with bilateral VTA cannulae were used for systemic -flupenthixol pretreatment experiments while separate groups of animals with both bilateral VTA cannulae and bilateral NAc cannulae were used for intra-NAc -flupenthixol pretreatment experiments. While neither systemic nor intra-NAc neuroleptic pretreatments attenuated the rewarding effects of higher, above-reward threshold doses of intra-VTA nicotine (8–48 nmol), these same neuroleptic pretreatments potentiated the rewarding effects of intra-VTA nicotine over a three-order-of-magnitude dose range and switched the motivational valence of a lower dose of nicotine (0.008 nmol) from aversive to rewarding. For systemic neuroleptic pretreated animals, ANOVA revealed a significant interaction between nicotine dose and treatment group (F^4,75=10.3; P<0.05). Post hoc analysis revealed that the rewarding effects of higher doses of intra-VTA nicotine (8 nmol; -flupenthixol: n=8; saline: n=7) (48 nmol; -flupenthixol: n=7; saline: n=8) were not attenuated by neuroleptic pretreatment (Figure 3a) relative to saline controls. However, systemic -flupenthixol pretreated animals displayed significant preferences for the nicotine-paired environments relative to saline control animals at the 0.008 (-flupenthixol: n=7; saline: n=8), 0.08 (-flupenthixol: n=8; saline: n=7), and 0.8 (-flupenthixol: n=7; saline: n=6) nmol doses of intra-VTA nicotine (P's<0.05). The lowest dose of intra-VTA nicotine (0.0008 nmol) produced no motivational effects in either neuroleptic pretreated animals (n=7) or in saline controls (n=6, P>0.05) (Figure 3a).

Figure 3.

Effects of systemic or intra-NAc -flupenthixol pretreatment on the motivational effects of intra-VTA or systemic nicotine. (a) Systemic -flupenthixol pretreated animals displayed significant preferences for the nicotine-paired environments relative to saline control animals at the 0.008–0.8 nmol doses of nicotine (all P <0.05). The lowest dose of intra-VTA nicotine (0.0008 nmol) produced no motivational effects in either -flupenthixol pretreated animals or in saline controls (all P >0.05), while higher nicotine doses (8, 24, and 48 nmol) were not attenuated by DA receptor blockade (all P>0.05). (b) Intra-NAc -flupenthixol pretreated animals displayed significant preferences for the nicotine-paired environments relative to saline controls at the 0.008–0.8 nmol doses of intra-VTA nicotine (all P<0.05). Intra-NAc -flupenthixol pretreatment did not attenuate the rewarding effects of a higher dose of intra-VTA nicotine (8 nmol). (c) Animals displayed an aversion to the nicotine-paired environment at the higher systemic nicotine dose (0.8 mg/kg) but not the lower dose (0.1 mg/kg) (P<0.05). When the aversive effects of systemic nicotine were challenged with intra-NAc -flupenthixol, the motivational effects of nicotine (0.8 mg/kg) were switched from aversive to rewarding.

Full figure and legend (59K)

For animals challenged with intra-NAc -flupenthixol, ANOVA revealed a significant interaction between nicotine dose and treatment group (F^3,63=5.24; P<0.05). Post hoc analysis revealed that the rewarding effects of a higher dose of intra-VTA nicotine (8 nmol; -flupenthixol: n=7; saline: n=7) were not attenuated by intra-NAc DA receptor blockade relative to controls (P>0.05). However, intra-NAc -flupenthixol pretreated animals displayed significant place preferences for the nicotine-paired environments relative to saline controls at the 0.008 (-flupenthixol: n=7; saline: n=8), 0.08 (-flupenthixol: n=9; saline: n=7), and 0.8 (-flupenthixol: n=7; saline: n=6) nmol doses of nicotine (all P<0.05) (Figure 3b). A separate control group of animals receiving microinfusions of -flupenthixol 1.5 mm dorsal to the NAc (n=5) displayed no preference for environments paired with 0.8 nmol intra-VTA nicotine (t⁴=0.53; P>0.05) (data not shown). Thus, similar to systemically neuroleptic pretreated animals, intra-NAc DA receptor blockade had no effect on the rewarding effects of higher intra-VTA nicotine doses, but potentiated the rewarding effects of lower intra-VTA nicotine concentrations and switched the motivational valence of a lower intra-VTA nicotine concentration (0.008 nmol) from aversive to rewarding (Figure 3a and b).

Because both inhaled and intravenous nicotine is delivered to the CNS via the systemic circulation, we sought to confirm our initial findings with a systemically administered dose of nicotine. In separate groups of animals, we tested both a high and a low dose of systemic nicotine (0.1 and 0.8 mg/kg; s.c.) and found a robust, conditioned place aversion to environments paired with the higher dose of systemic nicotine (0.8 mg) (n=7) but not to the lower concentration (0.1 mg) (n=7) (Figure 3c). A comparison of difference scores between saline and systemic nicotine-paired environments revealed a significant difference between systemic nicotine doses (0.1 and 0.8 mg/kg; s.c.) (t¹³=2.2; P<0.05). Doses of systemic nicotine within this range produce strong activation of the mesolimbic DA system,²¹ and previous studies have reported that this identical dose of nicotine produces strong aversive effects in both the place-conditioning paradigm and in the nicotine self-administration paradigm.^1,22 In a second experiment (with separate groups of animals), we challenged the aversive effects of systemic nicotine (0.8 mg/kg) with intra-NAc -flupenthixol (as previously described). Again, intra-NAc DA receptor blockade completely reversed the aversive effects of systemic nicotine, switching the motivational valence of systemic nicotine from aversive to rewarding. Comparison of difference scores between saline pretreated animals (n=7) and -flupenthixol pretreated animals (n=7) revealed a significant difference between groups (t¹³=7.9; P<0.05) (Figure 3c).

DA signalling mediates the aversive effects of nicotine

In the CPP paradigm, we observed potent rewarding effects of higher intra-VTA nicotine concentrations (8–48 nmol) and a strong aversive effect at a lower concentration (0.008 nmol). However, middle-range concentrations (0.08–0.8 nmol) of intra-VTA nicotine produced no observable behavioural effect in the CPP paradigm (Figure 2a), suggesting a functional, dose-dependent interaction between DA and non-DA neural motivational systems. Middle-range nicotine concentrations (0.08–0.8 nmol; Figure 2a) may simultaneously activate both an aversive as well as a rewarding motivational signal. In naive animals, simultaneous activation of these separate systems (mediating nicotine reward and aversion) would effectively counteract each other at these doses, resulting in no net behavioural effect. However, in neuroleptically pretreated animals, the aversive components of the nicotine signal at low doses would be blocked, thus revealing nicotine's rewarding effects in the VTA (Figure 3a-c).

In separate groups of animals, we used a CTA paradigm to sample selectively the aversive components of the intra-VTA nicotine stimulus. We chose two doses of intra-VTA nicotine for CTA experiments: a below-reward-threshold dose of intra-VTA nicotine (0.8 nmol) (n=9) and an above-reward-threshold dose of intra-VTA nicotine (8 nmol) (n=8). In our initial study, we observed a potent CTA to the nicotine-paired flavour at the lower dose of intra-VTA nicotine (0.8 nmol) (Figure 4a). However, at the above-reward-threshold dose of intra-VTA nicotine (8 nmol), animals displayed no CTA to the nicotine-paired flavour (Figure 4a). ANOVA revealed a significant interaction between treatment group and amount of nicotine relative to saline-paired flavour consumption (F^1,35=35.3; P<0.05). Post hoc analysis revealed that animals consumed significantly less of the flavours paired with 0.8 nmol nicotine relative to saline-paired flavours (P<0.05). No difference was observed between consumption of flavours paired with 8 nmol intra-VTA nicotine relative to saline-paired flavours (P>0.05). These results are consistent with the results obtained with the CPP paradigm and suggest that nicotine produces potent aversive effects at middle-range doses that can be selectively sampled with the CTA paradigm. In contrast, higher, above-reward-threshold doses of intra-VTA nicotine (8 nmol) selectively activate a nicotine reward signal, and overshadow a nicotine aversion signal, as no CTA was observed at this higher concentration, despite the potent CPP produced by this identical concentration (Figure 2a).

Figure 4.

CTA and intra-VTA nicotine. (a) Animals displayed a potent CTA to a lower dose of intra-VTA nicotine (0.8 nmol) but not to a higher, rewarding dose of intra-VTA nicotine (8 nmol). Animals consumed significantly less of the flavours paired with 0.8 nmol nicotine relative to saline-paired flavours (P<0.05). No difference was observed between consumption of flavours paired with the higher dose of VTA nicotine (8 nmol) (P>0.05) relative to saline-paired flavours. (b) When this VTA nicotine CTA was challenged with intra-NAc -flupenthixol, aversions to the nicotine-paired flavour were completely blocked (P<0.05). In animals conditioned with LiCl (15 mg/kg; i.p.), both saline and neuroleptic pretreated animals displayed significant aversions to the LiCl-paired flavours (all P<0.05).

Full figure and legend (89K)

In separate groups of animals, we challenged the intra-VTA nicotine CTA (0.8 nmol) with intra-NAc -flupenthixol, as previously described. ANOVA revealed a significant interaction between treatment group and amount of nicotine relative to saline-paired flavour consumption (F^1,29=8.47; P<0.05). Post hoc analysis revealed that saline pretreated animals (n=8) consumed significantly less of the nicotine-paired flavour (P<0.05). However, this CTA was blocked in intra-NAc -flupenthixol pretreated animals (n=7) (P>0.05) (Figure 4b). Thus, intra-NAc DA receptor blockade completely blocked the development of the intra-VTA nicotine CTA (Figure 4b). In a control experiment with separate groups of animals, this identical dose of intra-NAc -flupenthixol had no effect on the development of a CTA to the aversive effects of a separate drug, lithium chloride (LiCl) (15 mg/kg; i.p.) (Figure 4b), indicating that intra-NAc -flupenthixol produced no general, sensory impairment. In control animals conditioned with a separate drug, LiCl, ANOVA revealed a significant interaction between treatment group and amount of LiCl vs saline-paired flavour consumption (F^1,33=114.4; P<0.05). Post hoc analysis revealed that saline pretreated (n=8) and neuroleptic pretreated animals (n=9) consumed significantly less of the LiCl relative to the saline-paired flavour (all P<0.05). Finally, intra-NAc neuroleptic had no effect on the animals' capacity to drink, as no differences in overall fluid consumption during conditioning were observed between neuroleptic (mean=18.8 ml0.9 s.e.m.) and saline (mean=18.2 ml1.1 s.e.m.) pretreated groups (data not shown).

DA receptor blockade does not influence nicotine's discriminative properties

We next performed a discriminative taste avoidance experiment²³ (in separate groups of animals) to examine whether nicotine in the VTA can serve as a discriminative 'cue' and, if so, would DA receptor blockade influence nicotine's cognitive, discriminative stimulus properties? Using systemic nicotine (0.4 mg/kg; s.c.) as the discrimination cue (a dose commonly used as a nicotine discrimination cue²⁴), animals rapidly learned to discriminate nicotine from saline after only four conditioning trials (Figure 5a), and consumed significantly less fluid under the nicotine (aversion) cue than under the saline (neutral) cue. ANOVA revealed a significant interaction between treatment (nicotine or saline) with conditioning cycles on fluid consumption (F^8,152=47.3; P<0.05). Post hoc analysis revealed that animals drank significantly less of the saccharin(0.1%)-flavoured solution in the presence of the nicotine cue as early as the fifth training cycle, and continued this pattern of discriminated responding on the sixth, seventh and eighth subsequent trials (all P<0.05). A substitution test in these same animals, at the end of training, revealed that intra-VTA nicotine (0.8 nmol) did not generalize to the systemic nicotine cue in the presence of intra-NAc -flupenthixol (Figure 5a). No difference was observed in fluid consumption between intra-VTA saline and intra-VTA nicotine (0.8 nmol) on the substitution test (t⁸=0.55; P>0.05), indicating that intra-VTA nicotine (0.8 nmol) did not generalize to the systemic nicotine cue (0.4 mg/kg), even in the presence of intra-NAc -flupenthixol (3 g).

Figure 5.

Nicotine discrimination learning in systemic or intra-VTA nicotine treated animals. (a) Animals rapidly learned to discriminate between a systemic nicotine cue (D=drug) (0.4 mg/kg; s.c.) and a systemic saline cue (S=saline). Animals drank significantly less of the saccharin(0.1%)-flavoured solution in the presence of the nicotine cue as early as the fifth training cycle, and continued this pattern of discriminated responding on the sixth, seventh and eighth subsequent trials (all P<0.05). A substitution test revealed that intra-VTA nicotine (0.8 nmol), in the presence of intra-NAc -flupenthixol, did not generalize to the systemic nicotine cue. No difference was observed in fluid consumption between intra-VTA saline and nicotine treated animals (P>0.05). (b) When intra-VTA nicotine (0.8 nmol) was used as the discrimination cue in animals receiving either intra-NAc -flupenthixol or saline, no discrimination learning was observed over training cycles (P>0.05). Further, no differences were observed in mean fluid intake over training cycles following intra-VTA nicotine or saline between intra-NAc -flupenthixol and saline pretreated groups (inset, right side) (P>0.05).

Full figure and legend (80K)

Separate groups of animals receiving intra-VTA nicotine (0.8 nmol) (n=6) or saline (n=5) as the discrimination cues did not display discrimination learning, regardless of whether having received intra-NAc saline or -flupenthixol prior to intra-VTA nicotine or saline (Figure 5b). ANOVA revealed a significant effect of treatment on total fluid consumption over trials (F^17,215=14.1; P<0.05). However, no significant interaction was observed between treatment and group on fluid consumption over training cycles (F^7,215=0.45; P>0.05). Post hoc analysis revealed that animals did not consume significantly less fluid in the presence of the intra-VTA nicotine cue relative to the saline cue during any of the training cycles (all P>0.05) (Figure 5b). No evidence of discrimination learning was evident in these groups even after nine conditioning trials, suggesting that this dose of nicotine does not serve as a discrimination cue within the VTA.

These experiments demonstrate that the VTA does not play a significant role in the discriminative effects of nicotine, nor does DA receptor blockade influence the discriminative properties of intra-VTA nicotine, congruent with previous reports.²⁵

Discussion

The results of the present study identify the VTA as a critical neural substrate mediating both the rewarding and aversive effects of nicotine. DA receptor blockade was sufficient to reverse completely the aversive effects of intra-VTA nicotine, and switch the motivational valence of nicotine from aversive to rewarding. The same intra-NAc neuroleptic pretreatment had no effects on the discriminative properties of nicotine, suggesting a functionally specific interaction with nicotine's motivational properties.

The majority of abused drugs have been reported to possess both rewarding and aversive stimulus properties.²⁶ Indeed, the bivalent motivational properties of various drug stimuli, including opiates,²⁷ amphetamine,²⁸ and nicotine,^22,29 have been extensively demonstrated in both the CTA and CPP behavioural paradigms. To our knowledge, the present results are the first demonstration that a single, discrete brain region (VTA) can serve to mediate both the rewarding and aversive stimulus properties of a specific drug stimulus. Furthermore, the present findings suggest that within the VTA, the neural substrates that mediate these bivalent motivational effects are pharmacologically dissociable: while the rewarding properties of intra-VTA nicotine are mediated through a non-DA substrate, the aversive stimulus properties of intra-VTA nicotine were dependent on mesolimbic DA transmission. An important implication of these findings is that within the VTA, there is a functional interplay between neural substrates that control relative sensitivity to the rewarding and dysphoric properties of particular stimuli. Vulnerability to nicotine's rewarding and addictive potential may therefore be critically dependent upon the balance between these systems.

Mesolimbic DA receptor blockade dramatically increased nicotine reward sensitivity over a large dose range, suggesting that neuroleptic treatment may induce a drug-vulnerable phenotype. Indeed, our findings suggest that the removal of a simultaneously active aversive process is a likely explanation for the observed potentiation in intra-VTA nicotine reward. The quantitative magnitude of intra-VTA nicotine reward in the presence of neuroleptic was virtually identical, regardless of nicotine dose, and was increased to a maximum reward level (Figure 3a and b). If DA receptor blockade can increase nicotine reward sensitivity within the VTA, one implication is that neuroleptically medicated schizophrenics would be more vulnerable to nicotine's rewarding and addictive properties.

These findings have critical implications for the understanding of dopaminergic substrates in the mediation of drug motivational effects. While previous reports have suggested that mesolimbic DA substrates exclusively mediate a nicotine reward signal,^5,6 this proposal is not congruent with findings in human subjects wherein DA receptor antagonists increase nicotine intake, while DA receptor agonists strongly attenuate nicotine intake,^2,9,10 and it is not consistent with the exceedingly high rates of nicotine addiction observed among neuroleptically medicated, schizophrenic psychiatric populations.²

In contrast, the results reported in the present study suggest that neuroleptic DA receptor blockade can induce a unique, drug-vulnerable phenotype, by increasing the sensitivity to nicotine's rewarding properties directly within the VTA. Indeed, it has recently been reported that administration of a DA D1 receptor antagonist not only increases rates of cocaine self-administration in humans but also strongly increases subjective measures of drug pleasure and craving.³⁰ Similarly, animal studies have consistently reported that DA receptor antagonists increase cocaine self-administration.^6,7,31 However, interpretations of these results have postulated that increased responding for the drug may represent a compensatory mechanism due to the attenuation of the drug's rewarding properties induced by DA receptor blockade (animals must now consume more drug to compensate for the diminished reward efficacy of the stimulus). Nevertheless, it is unlikely that such a motivational compensatory interpretation could account for the present results, given that we observed a qualitative switch in the motivational valence of nicotine, from aversive to rewarding, following both systemic and intra-NAc DA receptor blockade. More importantly, these identical doses of neuroleptic had no effect on higher concentrations of nicotine, which produced rewarding effects in both saline and neuroleptically pretreated animals. Furthermore, these identical doses of neuroleptic completely block the rewarding properties of even high doses of opiates in opiate-dependent and withdrawn animals and also of amphetamine.^16,17,18 Clearly, no opiate or amphetamine reward signals are emerging post-synaptically from DA receptor substrates at these high concentrations of neuroleptic.

Alternatively, it has been suggested that increased nicotine intake and addiction rates among neuroleptically medicated schizophrenics may represent an attempt at self-medication, in order to overcome the potentially dysphoric effects of neuroleptic medications or to ameliorate the psychotic symptoms associated with schizophrenia.² However, this interpretation is difficult to reconcile with the present findings, given that neither our systemic nor intra-NAc neuroleptic pretreatment protocols produce any unconditioned motivational effects in and of themselves. Finally, the proposal that smoking may represent a form of self-medication in schizophrenics is confounded by clinical findings that neuroleptic side effects, such as tardive dyskinesia, are actually worsened in schizophrenics who smoke,^32,33 and by reports that schizophrenics that smoke actually display more severe psychotic symptoms,³⁴ and show a poorer clinical prognosis overall.³⁵ Our results suggest that neuroleptics may influence the motivational properties of nicotine by increasing the sensitivity to nicotine's rewarding properties directly within the VTA. DA receptor blockade may increase the sensitivity of nicotine's rewarding properties by means of blocking the aversive stimulus properties of nicotine, as demonstrated in both the CPP and the CTA behavioural paradigms. Indeed, considerable evidence suggests that DA signalling plays a critical role in the mediation of various aversive motivational stimuli.³⁶ Furthermore, it has been reported that nicotine and noxious stress stimuli may functionally interact to increase mesolimbic DA release.³⁷

Considerable evidence suggests that nicotine may serve to ameliorate cognitive deficits observed in schizophrenic disorders and may improve attentional deficits induced by some neuroleptic medications.²⁴ We observed no interaction between nicotine's discriminative stimulus properties and the effects of DA receptor blockade. Indeed, nicotine failed to serve as a discriminative cue within the VTA, and could not substitute for a systemic nicotine discrimination cue. However, other neural structures besides the VTA have been implicated as important mediators of nicotine's cognitive effects,³⁸ suggesting that separate neural regions mediate the motivational and discriminative stimulus properties of nicotine.

The present findings with nicotine are the first direct evidence that separate neural systems (within a single brain area) that mediate bivalent motivational drug effects can functionally interact to influence drug reward sensitivity. The present results suggest that treatment of schizophrenic symptoms with traditional DA receptor blocking drugs may serve to potentiate sensitivity to the rewarding and addictive properties of nicotine. Interestingly, it has recently been reported that decreased levels of DA D1 and D2 receptors (both targeted by traditional neuroleptic medications) are correlated with higher levels of nicotine addiction¹⁴ and methamphetamine abuse,¹³ respectively, suggesting that abnormalities in DA transmission may indeed underlie increased drug addiction propensities. The highly elevated risk of nicotine addiction observed among neuroleptically medicated schizophrenics may, in part, be attributable to the ability of DA receptor blockade to blunt the dysphoric effects of nicotine, thereby increasing sensitivity to the dependence producing, rewarding properties of nicotine.

References

1. Rose JE, Corrigall WA. Nicotine self-administration in animals and humans: similarities and differences. Psychopharmacology 1997; 130: 28–40. | Article | PubMed | ISI | ChemPort |
2. Dalack GW, Healy DJ, Meador-Woodruff JH. Nicotine dependence in schizophrenia: clinical phenomena and laboratory findings. Am J Psychiatry 1998; 155: 1490–1501. | PubMed | ISI | ChemPort |
3. Peto R, Lopez AD, Boreham J, Thun M, Heath C. Mortality from tobacco in developed countries: indirect estimation from national vital statistics. Lancet 1992; 339: 1268–1278. | Article | PubMed | ISI | ChemPort |
4. United States Department of Health and Human Services. The Health Consequences of Smoking: Nicotine Addiction, Rep. Surgeon Gen. Office on Smoking and Health. Washington, DC: Government Printing Office, 1988.
5. Corrigall WA, Franklin KBJ, Coen KM, Clark PBS. The mesolimbic dopaminergic system is implicated in the reinforcing effects of nicotine. Psychopharmacology 1992; 107: 285–289. | Article | PubMed | ISI | ChemPort |
6. Corrigall WA, Coen KM. Selective dopamine antagonists reduce nicotine self-administration. Psychopharmacology 1991; 104: 171–176. | Article | PubMed | ISI | ChemPort |
7. Arnold JM, Roberts DCS. A critique of fixed and progressive ratio schedules used to examine the neural substrates of drug reinforcement. Pharmacol Biochem Behav 1997; 57: 441–447. | Article | PubMed | ISI | ChemPort |
8. McEvoy JP, Freudenreich O, Levin ED, Rose JE. Haloperidol increases smoking in patients with schizophrenia. Psychopharmacology 1995; 119: 124–126. | Article | PubMed | ISI | ChemPort |
9. Caskey NH, Jarvik ME, Wirshing WC. The effects of dopaminergic D2 stimulation and blockade on smoking behavior. Exp Clin Pharmacol 1999; 7: 72–78.
10. Jarvik ME, Caskey NH, Wirshing WC, Madsen DC, Iwamoto-Schaap PN, Elins JL et al. Bromocriptine reduces cigarette smoking. Addiction 2000; 95: 1173–1183. | Article |
11. Becerra L, Breiter HC, Wise R, Gonzalez G, Borsook D. Reward circuitry activation by noxious thermal stimuli. Neuron 2001; 32: 927–946. | Article | PubMed | ISI | ChemPort |
12. Pezze MA, Heidbreder CA, Feldon J, Murphy CA. Selective responding of nucleus accumbens core and shell dopamine to aversively conditioned contextual and discrete stimuli. Neuroscience 2001; 108: 91–102. | Article | PubMed |
13. Volkow ND, Chang L, Wang GJ, Fowler JS, Ding YS, Sedler M et al. Low level of brain dopamine D2 receptors in methamphetamine abusers: association with metabolism in the orbitofrontal cortex. Am J Psychiatry 2001; 158: 2015–2021. | Article | PubMed | ISI | ChemPort |
14. Dagher A, Bleicher C, Aston JAD, Gunn RN, Clarke PBS, Cumming P. Reduced dopamine D1 receptor binding in the ventral striatum of cigarette smokers. Synapse 2001; 42: 48–53. | Article | PubMed | ISI | ChemPort |
15. Creese I, Burt DR, Snyder SH. Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs. Science 1976; 192: 481–483. | Article | PubMed | ISI | ChemPort |
16. Mackey WB, van der Kooy D. Neuroleptics block the positive reinforcing effects of amphetamine but not of morphine as measured by place conditioning. Pharmacol Biochem Behav 1985; 22: 101–105. | Article | PubMed | ISI | ChemPort |
17. Nader K, van der Kooy D. Deprivation state switches the neurobiological substrates mediating opiate reward in the ventral tegmental area. J Neurosci 1997; 17: 383–390. | PubMed | ISI | ChemPort |
18. Laviolette SR, Nader K, van der Kooy D. Motivational state determines the functional role of the mesolimbic dopamine system in the mediation of opiate reward processes. Behav Brain Res 2002; 129: 17–29. | Article | PubMed | ISI | ChemPort |
19. Hiroi N, White NM. The reserpine-sensitive dopamine pool mediates (+)-amphetamine-conditioned reward in the place preference paradigm. Brain Res 1990; 510: 33–42.
20. Laviolette SR, van der Kooy D. Blockade of the 7 nicotinic receptor subunit switches the motivational valence of nicotine from rewarding to aversive in the ventral tegmental area. Behav Pharmacol 2001; 12 (Suppl 1): S57.
21. Nissell M, Nomikos GG, Svensson TH. Systemic nicotine-induced dopamine release in the rat nucleus accumbens is regulated by nicotinic receptors in the ventral tegmental area. Synapse 1994; 16: 36–44. | Article | PubMed | ISI | ChemPort |
22. Jorenby DE, Steinpreis RE, Sherman JE, Baker TB. Aversion instead of preference learning indicated by nicotine place conditioning in rats. Psychopharmacology 1990; 101: 533–538. | Article | PubMed | ISI | ChemPort |
23. Jaeger TV, van der Kooy D. Separate neural substrates mediate the motivating and discriminative properties of morphine. Behav Neurosci 1996; 110: 181–201.
24. Wiley JL, James JR, Rosecrans JA. Discriminative stimulus properties of nicotine: approaches to evaluating potential nicotinic receptor agonists and antagonists. Drug Dev Res 1996; 38: 222–230.
25. Corrigall WA, Coen KM. Dopamine mechanisms play at best a small role in the nicotine discriminative stimulus. Pharmacol Biochem Behav 1994; 3: 817–820.
26. Hunt T, Amit Z. Conditioned taste aversion induced by self-administered drugs: paradox revisited. Neurosci Biobehav Rev 1987; 11: 107–130.
27. Martin GM, Bechara A, van der Kooy D. Morphine pre-exposure attenuates the aversive properties of opiates without pre-exposure to the aversive properties. Pharmacol Biochem Behav 1988; 30: 687–692.
28. Bell SM, Thiele TE, Seeley RJ, Bernstein IL, Woods SC. Effects of food deprivation on conditioned taste aversion in rats. Pharmacol Biochem Behav 1998; 60: 459–466.
29. Shoaib M, Stolerman IP. Conditioned taste aversion in rats after intracerebral administration of nicotine. Behav Pharmacol 1995; 6: 375–385. | PubMed | ISI | ChemPort |
30. Haney M, Ward AS, Foltin RW, Fischman MW. Effects of ecopipam, a selective dopamine D1 antagonist, on smoked cocaine self-administration in humans. Psychopharmacology 2001; 155: 330–337. | Article | PubMed |
31. Koob GF, Le HT, Creese I. The D1 dopamine antagonist SCH 23390 increases cocaine self-administration in the rat. Neurosci Lett 1987; 79: 315–320. | Article | PubMed | ChemPort |
32. Menza MA, Grossman N, Van Horn M, Cody R, Forman N. Smoking and movement disorders in psychiatric patients. Biol Psychiatry 1991; 30: 109–115.
33. Yassa R, Lal S, Korpassy A, Ally J. Nicotine exposure and tardive dyskinesia. Biol Psychiatry 1987; 22: 67–72.
34. Apud JA, Egan MF, Wyatt RJ. Effects of smoking during antipsychotic withdrawal in patients with chronic schizophrenia. Schizophr Res 2000; 46: 119–127.
35. Goff DC, Henderson DC, Amico E. Cigarette smoking in schizophrenia: relationship to psychopathology and medication side effects. Am J Psychiatry 1992; 149: 1189–1194. | PubMed | ISI | ChemPort |
36. Horvitz JC. Mesolimbocortical and nigrostriatal dopamine responses to salient non-reward events. Neuroscience 2000; 96: 651–656. | Article | PubMed | ISI | ChemPort |
37. Takahashi H, Takada Y, Nagai N, Urano T, Takada A. Effects of nicotine and footshock stress on dopamine release in the striatum and nucleus accumbens. Brain Res Bull 1998; 45: 157–162. | Article | PubMed | ChemPort |
38. Anso K, Miyata H, Hironaka N, Tsuda T, Yanagita T. The discriminative effects of nicotine and their central sites in rats. Yakubutsu Seishin Kodo 1993; 13: 129–136.

Acknowledgements

This work was supported by CIHR.