- Email: [email protected]
6J mice
Neuropharmacology 39 (2000) 2840–2847 www.elsevier.com/locate/neuropharm
Antagonism of the discriminative and aversive stimulus properties of nicotine in C57BL/6J mice J. Gommans, I.P. Stolerman *, M. Shoaib Section of Behavioural Pharmacology, Institute of Psychiatry, King’s College London, De Crespigny Park, London SE5 8AF, UK Accepted 18 July 2000
Abstract Mice of the C56BL/6J strain were trained to discriminate between nicotine (1.2 mg/kg) and saline in a two-lever drug discrimination procedure under a tandem variable-interval 60 s fixed-ratio 10 schedule of food reinforcement. Mice of the same strain were trained in conditioned taste aversion (CTA) experiments where drinking a saccharin or saline solution was paired with injection of nicotine or vehicle. During testing with both flavours presented simultaneously, a reduction in the intake of the nicotine-paired solution indicated CTA. The nicotine discrimination was acquired successfully and nicotine yielded a steep dose–response curve. The competitive nicotinic antagonist dihydro-β-erythroidine (DHβE, 0.6–3.0 mg/kg) shifted the dose–response for the discriminative stimulus effect of nicotine to the right; the α7 nicotinic receptor antagonist methyllycaconitine (MLA, 1.0–10 mg/kg) had no effect. The mice showed strong CTA to 2.0 mg/kg of nicotine and marginally to 0.6 and 1.2 mg/kg of nicotine. DHβE (3.0–5.6 mg/kg) attenuated the CTA while MLA (1.0–10 mg/kg) had no effect. These studies show that nicotine has discriminative and aversive stimulus properties in C57BL/6J mice and that the effects are mediated primarily by receptors sensitive to DHβE; there was no evidence for the involvement of α7 nicotinic receptors. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Nicotine; Drug discrimination; Conditioned taste aversion; Mice; Dihydro-β-erythroidine; Methyllycaconitine
1. Introduction Drug discrimination and conditioned taste aversion (CTA) techniques have been used extensively to study the stimulus properties of nicotine. The discriminative stimulus effects of nicotine serve as a behavioural assay to characterise the neuropharmacological mechanisms underlying nicotine’s behavioural effects. Most nicotine discrimination studies have been carried out in rats, with some in humans but surprisingly, experiments in mice have been reported only recently (Stolerman et al., 1999; Varvel et al., 1999). Both DBA and C56BL/6 strains could be trained to discriminate between nicotine and saline and the discrimination was dose-dependent (Stolerman et al., 1999; Varvel et al., 1999). The nicotine stimulus could be completely antagonised by the nonselective nicotinic receptor antagonist mecamylamine
* Corresponding author. Tel.: +44-020-7848-0370; fax: +44-0207848 0579. E-mail address: [email protected] (I.P. Stolerman).
(Stolerman et al., 1999; Varvel et al., 1999), but not by the muscarinic receptor antagonist scopolamine (Varvel et al., 1999). The aversive effects of nicotine as measured in a conditioned taste aversion might be involved in the regulation of nicotine intake by setting an upper limit to the amount taken (other interpretations for taste aversions produced by positively reinforcing drugs have been offered (Grigson, 1997; Stolerman and D’Mello, 1981). Etscorn (1980) and Risinger and Brown (1996) reported conditioned taste aversion experiments with nicotine in mice. In contrast to some other strains, with a one-stimulus procedure in which only the nicotine-paired flavoured solution was presented during testing, C57BL/6 mice did not show a CTA for nicotine in doses up to 2.0 mg/kg SC (Risinger and Brown, 1996). The present study aims to characterise further the discriminative and aversive stimulus effects of nicotine in mice. It is necessary to establish robust measures for nicotine in mice because some techniques for identifying genetic factors in susceptibility to drugs rely upon the use of recombinant inbred mice (Crabbe et al., 1999);
0028-3908/00/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 3 0 - 1
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
similarly, mice but not rats with targeted deletion of nicotinic receptor subunits have been constructed, on a C57BL/6 background (Picciotto et al., 1995). In preparation for such studies, C57BL/6 mice have been trained to discriminate 1.2 mg/kg of nicotine in a twolever discrimination procedure and two specific nicotine receptor antagonists have been tested. Grote and Brown (1971) showed that a conditioned taste aversion procedure with a two-stimulus test, in which animals could choose between two flavours of which one had previously been paired with cyclophosphamide, was more sensitive than a conditioned taste aversion procedure with a one-stimulus test. We have therefore used a twostimulus CTA procedure to investigate the aversive stimulus effects of nicotine and the effects of nicotine antagonists. The nicotine receptor has a pentameric architecture, consisting of combinations of α-, or α- plus β-subunits (Galzi and Changeux (1995), and in the brain there are at least eight different α-receptor subunits (α2–α9) and three different β-receptor subunits (β2–β4). For the present experiments antagonism studies were carried out with methyllycaconitine (MLA) and dihydro-β-erythroidine (DHβE). Studies using Xenopus oocytes suggested that MLA was an α7-receptor antagonist (Wonnacott et al., 1993) and that DHβE was a competitive nicotinic receptor antagonist at several subtypes of receptor. These included α4β2, α4β4 and α3β2 receptors, and to a lesser extent at α2β2 and α2β4 receptors (Harvey et al., 1996). DHβE has been shown to antagonise the nicotine discriminative stimulus and taste aversion effects in rats (Stolerman et al., 1997; Shoaib et al., 2000) and some of the behavioural effects of nicotine in mice (Damaj et al., 1995; Decker et al., 1995). In contrast, MLA has not been found to block nicotine discrimination in rats or its antinociceptive effect in mice (Brioni et al., 1996; Damaj et al., 1998). It was hypothesised that differential effects of these antagonists would help to reveal the diversity of subtypes of nicotinic receptors that may mediate the behavioural effects of nicotine.
2. Materials and methods 2.1. Animals For drug discrimination, 22 male C57BL/6J mice (Harlan Olac, Bicester, UK) were trained. They were placed on a restricted diet to maintain their body weight at approximately 85% of their free feeding weight; water was freely available. All animals were housed individually in a room at a controlled temperature (21°C) with a 12-h light–dark cycle (light from 07:30 to 19:30 h). For conditioned taste aversion a total of 88 C57BL/6J mice (Harlan Olac) was used. Food was freely available but water was restricted as below. All animals were
2841
housed individually in a room at a controlled temperature (21°C) with a 12-h light–dark cycle (light from 07:30 to 19:30 h). All studies complied with local ethical requirements and were carried out in accordance with the Animals (Scientific Procedures) Act, 1986, under licence from the UK Home Office. 2.2. Apparatus For drug discrimination, six standard operant chambers for mice (MED Associates, Georgia, VT) contained in ventilated, sound-insulated housings were used. Each chamber was equipped with two ultra-sensitive levers, one on each side of a recess in which a dispenser could deliver 25 mg pellets of food. All studies of conditioned taste aversion were carried out in the home cages. Drinking cylinders calibrated to 0.1 ml were attached directly to the cage. 2.3. Procedure for drug discrimination The procedure was based on that of Stolerman et al. (1999). Initially, mice were trained to consume food pellets delivered automatically under a variable-time 30 s schedule, until all food delivered was consumed by the end of a 15 min session. During this stage no levers were present in the chamber. Subsequently, one lever was inserted into the chamber and training to lever-press began. Mice were trained first under a continuous reinforcement schedule, then under a fixed-ratio schedule that increased progressively to FR-10. After this response was acquired, the original lever was removed and a single lever was inserted on the opposite site of the chamber and mice were again shaped to perform under the FR-10 schedule of food presentation. After animals acquired stable FR-10 responding a variableinterval (VI) component was introduced. Under this tandem schedule of reinforcement, the tenth lever press was reinforced after a randomly determined interval of time. Initially the mean value of the VI was 15 s, but after five sessions it was increased to 30 s (range 7–53 s). After this schedule was in operation for several sessions, discrimination training started. Both levers were present but responses on only one lever were reinforced, depending on the drug condition. Animals were injected 10 min prior to the start of sessions with either nicotine (1.2 mg/kg, SC) or saline. For half the animals, presses on the left lever were reinforced after nicotine and on the right lever after saline; for the other half of the animals, presses on the right lever were reinforced after nicotine and on the left lever after saline. Each mouse had a unique random order of nicotine and saline sessions except that there were never more than three nicotine or three saline sessions in succession. After several weeks
2842
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
of training the duration of the VI component of the schedule was increased to 60 s. After the mice acquired the nicotine discrimination as shown by 80% accuracy of lever selection in a block of 10 consecutive sessions, generalisation testing took place twice weekly. Mice were placed in the chambers for 5 min with both levers present but no reinforcers available. Mice that reached a criterion of 80% accuracy during initial tests with saline and the training dose of nicotine were used for subsequent experiments. For nicotine dose–response determinations, different doses of nicotine or saline were injected 10 min prior to testing. For antagonism tests, the antagonists were injected 30 min and nicotine 10 min prior to testing. Tests with different drug treatments were in random order. 2.4. Procedure for conditioned taste aversion The procedure was adapted from those described previously for rats (D’Mello et al., 1977; Kumar et al., 1983). Mice were trained to drink tap water for 30 min twice daily, once in the morning (between 9:00 and 10:30 h) and once in the afternoon (between 4:00 and 5:00 h). When the amount consumed was stable, conditioning began. Mice were presented in the morning session with either a sweet solution (sodium saccharin, 0.1%) or a saline solution (sodium chloride, 0.9%) on every second day. After drinking the solutions the mice were injected immediately with either nicotine or vehicle. There were a total of four conditioning sessions, two with each flavour (hence, two sessions with nicotine and two with vehicle). To balance out effects due to inherent differences in palatability of the two flavours, for half the animals nicotine was paired with the saccharin flavour and vehicle with the saline flavour, and for the other half nicotine was paired with saline and vehicle with saccharin. The two solutions were presented alternately and to balance out side preferences each flavour was presented once at the left side of the cage and once at the right side. During the afternoon drinking sessions and on intervening days, mice could drink normal tap water. Two days after the last conditioning session the mice were presented with both solutions simultaneously and were free to choose what to drink. On the next day, this test was repeated and the left–right position of the two flavours was reversed, to counterbalance possible side preferences. No injections were given after these two test sessions. Three different groups of C57BL/6J mice (n=8) were conditioned with different doses of nicotine (0.6, 1.2 and 2.0 mg/kg, SC). Antagonism tests were performed with DHβE (1.0, 3.0 and 5.6 mg/kg) and MLA (1.0, 3.0 and 10.0 mg/kg) in different groups of mice (n=8). The antagonists were injected at the same time as nicotine or vehicle, i.e. immediately after drinking sessions. To
avoid possible conditioning with the antagonists, they were administered after both nicotine and vehicle conditioning sessions. 2.5. Data analysis For drug discrimination, the number of responses made on each lever during extinction tests was recorded for the whole 5-min test session. The percentage of total responses that were made on the drug-appropriate lever served as a measure for stimulus generalisation; the total number of responses served as a measure for response rate. Data were analysed with repeated-measure analysis of variance, with arcsine transformations to normalise the distribution of percentage scores (Winer, 1970). For the antagonism studies ED50 values were calculated (by interpolation) as the dose of nicotine associated with 50% drug-appropriate responding. For conditioned taste aversion, the amounts consumed of the nicotine and vehicle paired flavours during the conditioning and two test sessions were measured. The mean amounts consumed of the drug-paired flavour as a percentage of total fluid intake were calculated. These percentage scores were subjected to arcsine transformations to normalise their distributions (Winer, 1970) and subsequently analysed with a one sample t-test to determine whether means differed significantly from 50%. A mean score of 50% would indicate that nicotine induced neither conditioned taste aversion nor conditioned taste preference; scores below 50% indicated CTA. This approach to quantifying the magnitude of CTA and to evaluation of significance has been used in previous studies (e.g. D’Mello et al., 1977; Kumar et al., 1983; Reavill et al., 1986). 2.6. Drugs Nicotine hydrogen-(+)-tartrate was dissolved in isotonic saline. The pH was adjusted to 7 with 0.5 N NaOH. Dihydro-β-erythroidine hydrobromide and methyllycaconitine citrate (both from Sigma-Aldrich, Poole, Dorset, UK) were dissolved in saline. All doses were calculated as those of the base. For drug discrimination, nicotine was injected 10 min prior to testing or training, DHβE and MLA were administered 30 min prior to testing; for conditioned taste aversion all drugs were administered immediately after the drinking sessions ended. All injections were given subcutaneous in a volume of 1 ml/100 g. 3. Results 3.1. Drug discrimination A total of 19 of the 22 mice trained reached the 80% criteria for accuracy and were then used for dose–
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
response tests with nicotine. These mice were divided into two groups for the purpose of testing two antagonists later. As can be seen from Fig. 1, both groups showed similar dose-dependent nicotine generalisation gradients (group to be tested with DHβE, F(4,36)=6.58, p⬍0.001, n=10; MLA group, F(4,32)=13.2, p⬍0.001, n=9). Nicotine also reduced the number of responses in both groups (F(4,36)=29.6, p⬍0.001 and F(4,32)=14.9, p⬍0.001, respectively). Fig. 2 shows the results of antagonism tests with DHβE and MLA. There was a clear difference in effect of the two antagonists: DHβE shifted the dose–response curve of nicotine to the right (F(2,18)=103.9, p⬍0.001, based on data for nicotine doses of 0.6–1.6 mg/kg), and with a 3.0 mg/kg dose of DHβE there was an almost complete block of the 1.2 mg/kg training dose of nicotine. MLA did not shift the nicotine dose–response curve (F(2,16)⬍1). The ED50 for nicotine alone was 0.6 mg/kg, whereas for nicotine with DHβE in doses of 0.6 mg/kg and 3.0 mg/kg the ED50s were 0.9 and 1.7 mg/kg, respectively. Thus, 3.0 mg/kg of DHβE shifted the dose– response curve for nicotine to the right by a factor of 2.8. Nicotine reduced response rates in both groups (F(5,45)=5.55, p⬍0.001) and F(4,32)=12.7, p⬍0.001);
2843
there was, however, no effect of either antagonist (DHβE: F(2,18)=3.54; MLA: F(2,16)⬍1). 3.2. Conditioned taste aversion Fig. 3 shows that there was a conditioned taste aversion to 0.6 mg/kg (34.4% consumption of drug-paired flavoured solutions, t(7)=2.7, p⬍0.05) and in particular to 2.0 mg/kg of nicotine (17.9%, t(7)=5.8, p⬍0.001). A dose of 1.2 mg/kg of nicotine failed to induce taste aversion (27.8%, t(7)=2.3). The intake of the saline-paired flavoured solutions was reduced on trial 3 because both flavoured solutions were presented simultaneously and mice could satisfy their fluid requirements by drinking some of each (there were no differences between the groups in total fluids consumed). The dose of 2.0 mg/kg of nicotine was considered to show a robust CTA, and therefore this dose was used for subsequent antagonism studies in which nicotine was tested in combination with different doses of DHβE or MLA. Saline or a 1.0 mg/kg dose of DHβBE had no effect on nicotine-induced CTA (nicotine alone, 12.3%, t(7)=5.4, p⬍0.001; nicotine plus DHβBE 1.0 mg/kg, 12.0%, t(7)=6.8, p⬍0.001). Fig. 4 shows that doses of 3.0 and 5.6 mg/kg of DHβBE antagonised nicotineinduced CTA, i.e. data were not significantly different from 50% (DHβBE 3.0 mg/kg, 30.6%, t(7)=1.9; DHβBE 5.6 mg/kg, 36.1%, t(7)=1.6). In trial 3, mice pretreated with 5.6 mg/kg of DHβBE consumed a total of 1.12 ml of the two flavoured solutions, as compared with 1.06 ml for the mice pretreated with saline. The reduced intake of the saline-paired flavoured solutions on trial 3 in mice pretreated with DHβBE can therefore be attributed to the concurrent intake of the nicotine-paired solutions. MLA had no effect on nicotine-induced CTA (Fig. 5). Nicotine showed strong CTA when combined with each dose of MLA (minimum t(7)=6.4, p⬍0.001 at each dose).
4. Discussion
Fig. 1. Dose–response curves for nicotine in two groups of mice trained to discriminate nicotine (1.2 mg/kg, SC) from saline. Upper panel: responses on the drug-appropriate lever as a percentage of total responses on both levers (mean±s.e.m.). Lower panel: total number of responses. Group 1 (䊏, n=10) was used subsequently for antagonism studies with DHβE; group 2 (쐌, n=9) was used for studies with MLA.
The present experiments confirm earlier reports that nicotine can produce dose-related discriminative stimulus effects in C57BL/6 mice (Stolerman et al., 1999; Varvel et al., 1999). The results also show that DHβE, but not MLA, blocked the discriminative stimulus effects of nicotine in mice. These findings are in agreement with previous studies in rats (Brioni et al., 1996; Shoaib et al., 2000; Stolerman et al., 1997), and support the view that α7 receptors are not involved in the discriminative stimulus effects of nicotine. Increasing the dose of nicotine reversed the block produced by DHβE. A 3.0 mg/kg dose of DHβE completely antagonised 0.6 and 1.2 mg/kg of nicotine but larger doses of nicotine (1.6 and 2.0 mg/kg) in combination with DHβE induced high
2844
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Fig. 2. Dose–response curves for the discriminative stimulus effect of nicotine in antagonism studies with DHβE (왖, 0.6 mg/kg and 䊏, 3.0 mg/kg, n=10), and MLA (䉬, 1.0 mg/kg, and 왖, 10 mg/kg, n=9). Vehicle controls are also shown (쐌). Upper panel: responses on the drugappropriate lever as a percentage of total responses on both levers (mean±s.e.m.). Lower panel: total number of responses.
Fig. 3. Nicotine-induced conditioned taste aversion in three groups of mice (n=8). Trials 1 and 2 were conditioning sessions in which flavoured solutions were paired with nicotine and saline; trial 3 was a two-stimulus test in which nicotine and saline-paired flavoured solutions were presented simultaneously (there were no drug injections). Nicotine (2.0 mg/kg) resulted in a strong taste aversion, nicotine 0.6 and 1.2 mg/kg had weaker effects (쐌, saline-paired flavour; 䊏, nicotine-paired flavour).
levels of nicotine-lever responding. This might indicate a competitive mechanism and has been reported before in nicotine discrimination in rats (Stolerman et al., 1997). In rats the nicotine dose–response curve was shifted to the right by a factor of 9.4, whereas in the present study the shift was much smaller (2.8). The present experiments also show for the first time
that nicotine can induce a conditioned taste aversion in C56BL/6 mice, in contrast to previous experiments (Risinger and Brown, 1996). There are major differences between these and the previous experiments. The first difference is the use of a two-stimulus test procedure in the present experiments versus a one-stimulus procedure by Risinger and Brown (1996). The two-stimulus pro-
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
2845
Fig. 4. Effect of DHβE on nicotine-induced conditioned taste aversion in four groups of mice (n=8). Nicotine (2.0 mg/kg) administered in combination with DHβE (0.0–1.0 mg/kg) resulted in taste aversion. Larger doses of DHβE (3.0 and 5.6 mg/kg) antagonised the taste aversion (쐌, saline-paired flavour; 䊏, nicotine-paired flavour). Other details as for Fig. 3.
cedure was reported to be more sensitive (Grote and Brown, 1971). In the present experiments, nicotine was administered at the end of a 30 min drinking period as compared with a 60 min period in Risinger and Brown (1996). As most fluid intake occurs early in the drinking periods, the interval between exposure to the flavours and drug effect was shorter in the present experiments. A further advantage of the present two-flavour procedure is that animals serve as their own controls and there is no need for an extra group of animals that does not receive nicotine. The experiments also show that DHβE could antagonise nicotine-induced CTA, as has been previously reported for rats (Shoaib et al., 2000), but MLA had no effect on nicotine induced CTA. This argues for a role of α4β2, α4β4 and/or α3β2-receptors and possibly α2β2 or α2β4 receptors (but not for α7 receptors) in nicotineinduced CTA. Other injection-test intervals and routes of administration should also be examined. However, Turek et al. (1995) reported, in rats 6.2 µmol/kg (IP) of MLA (equivalent to 5.4 mg/kg), produced maximal brain concentrations of MLA after 30 min. For the present experiment both smaller and larger doses of MLA were used, administered 30 min before testing.
The results show some congruence between drug discrimination and CTA with respect to the involvement of specific nicotinic receptors. However, due to the lack of more selective antagonists, the present data do not make it possible to differentiate between receptor subtypes in a more conclusive manner. In the present study, the response rates were reduced by nicotine, but this effect was not antagonised by DHβE. In rats, DHβE also failed to antagonise the locomotor depressant and the operant response rate-reducing effect of nicotine (Stolerman et al., 1997). In mice there is conflicting evidence for antagonism by DHβE of effects of nicotine on motoric responses. In ICR mice DHβE antagonised nicotine-induced motor impairment in a rotarod test and partially antagonised nicotinereduced locomotor activity (Damaj et al., 1995). In CD1 mice, however, DHβE failed to antagonise nicotinereduced locomotor activity (Decker et al., 1995). Strain differences, different tests, and different doses of DHβE used might account for these different results and make it difficult to generalise between studies. One can conclude that DHβE antagonises nicotine-affected motoric behaviours under specific experimental circumstances. In conclusion, nicotine induces both discriminative
2846
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Fig. 5. Lack of effect of MLA on nicotine-induced conditioned taste aversion in four groups of mice (n=8). Nicotine (2.0 mg/kg) in combination with MLA (0.0–10.0 mg/kg) resulted in taste aversions (쐌, saline-paired flavour; 䊏, nicotine-paired flavour). Other details as for Fig. 3.
stimulus and a conditioned taste aversion effects in C57BL/6 mice; these effects are mediated by receptors sensitive to DHβE, but there is no evidence yet for a role of receptors that contain the α7 subunit. The availability of genetically manipulated mice is one reason for establishing these behavioural methods. There are at the moment mice that lack the genes coding for several subunits of nicotinic receptors, including α3 (Xu et al., 1999), α4 (Marubio et al., 1999), α7 (Orr-Urtreger et al., 1997), α9 (Vetter et al., 1999) and β2 (Picciotto et al., 1995). Future studies with these knock-out mice might show more precisely the role of the different nicotinic receptor subunits.
Acknowledgements This work was supported by an EU-TMR and a MRC research grant.
References Brioni, J.D., Kim, D.J.B., O’Neill, A.B., 1996. Nicotine cue: lack of effect of the α7 nicotinic antagonist methyllycaconitine. European Journal of Pharmacology 301, 1–5.
Crabbe, J.C., Phillips, T.J., Buck, K.J., Cunningham, C.L., Belknap, J.K., 1999. Identifying genes for alcohol and drug sensitivity: recent progress and future directions. Trends in Neurosciences 22, 173–179. Damaj, M.I., Fei-Yin, M., Dukat, M., Glassco, W., Glennon, R.A., Martin, B.R., 1998. Antinociceptive responses to nicotinic acetylcholine receptor ligands after systemic and intrathecal administration in mice. Journal of Pharmacology and Experimental Therapeutics 284, 1058–1065. Damaj, M.I., Welch, S.P., Martin, B.R., 1995. In vivo pharmacological effects of dihydro-β-erythroidine, a nicotinic antagonist, in mice. Psychopharmacology 117, 67–73. Decker, M.W., Anderson, D.J., Brioni, J.D., Donnelly-Roberts, D.L., Kang, C.-H., O’Neill, A.B., Piattoni-Kaplan, M., Swanson, S., Sullivan, J.P., 1995. Erysodine, a competitive antagonist at neuronal nicotinic acetylcholine receptors. European Journal of Pharmacology 280, 79–89. D’Mello, G.D., Stolerman, I.P., Booth, D.A., Pilcher, C.W.T., 1977. Factors influencing flavour aversions conditioned with amphetamine in rats. Pharmacology Biochemistry and Behavior 7, 185–190. Etscorn, F., 1980. Sucrose aversions in mice as a result of injected nicotine or passive tobacco smoke inhalation. Bulletin of the Psychonomic Society 15, 54–56. Galzi, J.-L., Changeux, J.-P., 1995. Neuronal nicotinic receptors: molecular organization and regulations. Neuropharmacology 34, 563–582. Grigson, P.S., 1997. Conditioned taste aversions and drugs of abuse: a reinterpretation. Behavioural Neuroscience 111, 129–136. Grote, F.W., Brown, R.T., 1971. Conditioned taste aversions: twostimulus tests are more sensitive than one-stimulus tests. Behavioural Research Methods and Instrumentation 3, 311–312.
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Harvey, S.C., Maddox, F.N., Luetje, C.W., 1996. Multiple determinants of dihydro-β-erythroidine sensitivity on rat neuronal nicotinic receptor α subunits. Journal of Neurochemistry 67, 1953–1959. Kumar, R., Pratt, J.A., Stolerman, I.P., 1983. Characteristics of conditioned taste aversion produced by nicotine in rats. British Journal of Pharmacology 79, 245–253. Marubio, L.M., Arroyo-Jimenez, M.D.M., Cordero-Erausquin, M., Le´na, C., Le Nove`re, N., D’Exaerde, A.D.K., Huchet, M., Damaj, M.I., Changeux, J.-P., 1999. Reduced antinociception in mice lacking neuronal nicotinic receptor subunits. Nature 398, 805–810. Orr-Urtreger, A., Go¨ldner, F.M., Saeki, M., Lorenzo, I., Goldberg, L., De Biasi, M., Dani, J.A., Patrick, J.W., Beaudet, A.L., 1997. Mice deficient in the α7 neuronal nicotinic acetylcholine receptor lack αbungarotoxin binding sites and hippocampal fast nicotinic currents. Journal of Neuroscience 17, 9165–9171. Picciotto, M.R., Zoli, M., Le´na, C., Bessis, A., Lallemand, Y., Le Nove`re, N., Vincent, P., Pich, E.M., Bruˆlet, P., Changeux, J.-P., 1995. Abnormal avoidance learning in mice lacking functional highaffinity nicotine receptor in the brain. Nature 374, 65–67. Reavill, C., Stolerman, I.P., Kumar, R., Garcha, H.S., 1986. Chlorisondamine blocks acquisition of the conditioned taste aversion produced by (–)-nicotine. Neuropharmacology 25, 1067–1069. Risinger, F.O., Brown, M.M., 1996. Genetic differences in nicotineinduced conditioned taste aversion. Life Sciences 58, 223–229. Shoaib, M., Zubaran, C., Stolerman, I.P., 2000. Antagonism of stimulus properties of nicotine by dihydro-β-erythroidine (DHβE) in rats. Psychopharmacology 149, 140–146. Stolerman, I.P., Chandler, C.J., Garcha, H.S., Newton, J.M., 1997. Selective antagonism of behavioural effects of nicotine by dihydroβ-erythroidine in rats. Psychopharmacology 129, 390–397. Stolerman, I.P., D’Mello, G.D., 1981. Oral self administration and the
2847
relevance of conditioned taste aversions. In:. Thompson, T., Dews, P.B., McKim, W.A. (Eds.), Advances in Behavioral Pharmacology, vol. 3. Academic Press, New York, pp. 169–214. Stolerman, I.P., Naylor, C., Elmer, G.I., Goldberg, S.R., 1999. Discrimination and self-administration of nicotine by inbred strains of mice. Psychopharmacology 141, 297–306. Turek, J.W., Kang, C.-H., Campbell, J.E., Arneric, S.P., Sullivan, J.P., 1995. A sensitive technique for the detection of the α7 neuronal nicotinic acetylcholine receptor antagonist, methyllycaconitine, in rat plasma and brain. Journal of Neuroscience Methods 61, 113– 118. Varvel, S.A., James, J.R., Bowen, S., Rosecrans, J.A., Karan, L.D., 1999. Discriminative stimulus (DS) properties of nicotine in the C57BL/6 mouse. Pharmacology Biochemistry and Behavior 63, 27–32. Vetter, D.E., Liberman, M.C., Mann, J., Barhanin, J., Boulter, J., Brown, M.C., Saffiote-Kolman, J., Heinemann, S.F., Elgoyhen, A.B., 1999. Role of α9 nicotinic Ach receptor subunits in the development and function of cochlear efferent innervation. Neuron 23, 93–103. Winer, B.J., 1970. Statistical Principles in Experimental Design, second ed. McGraw-Hill, London. Wonnacott, S., Albuquerque, E.X., Bertrand, D., 1993. Methyllycaconitine: a new probe that discriminates between nicotinic acetylcholine receptor subclasses. Methods in Neurosciences 12, 263–275. Xu, W., Gelber, S., Orr-Urtreger, A., Armstrong, D., Lewis, R.A., Ou, C.-N., Patrick, J., Role, L., De Biasi, M., Beaudet, A.L., 1999. Megacystis, mydriasis, and ion channel defect in mice lacking the α3 neuronal nicotinic acetylcholine receptor. In: Proceedings of the National Academy of Sciences of the United States of America 96., pp. 5746–5751.
Antagonism of the discriminative and aversive stimulus properties of nicotine in C57BL/6J mice J. Gommans, I.P. Stolerman *, M. Shoaib Section of Behavioural Pharmacology, Institute of Psychiatry, King’s College London, De Crespigny Park, London SE5 8AF, UK Accepted 18 July 2000
Abstract Mice of the C56BL/6J strain were trained to discriminate between nicotine (1.2 mg/kg) and saline in a two-lever drug discrimination procedure under a tandem variable-interval 60 s fixed-ratio 10 schedule of food reinforcement. Mice of the same strain were trained in conditioned taste aversion (CTA) experiments where drinking a saccharin or saline solution was paired with injection of nicotine or vehicle. During testing with both flavours presented simultaneously, a reduction in the intake of the nicotine-paired solution indicated CTA. The nicotine discrimination was acquired successfully and nicotine yielded a steep dose–response curve. The competitive nicotinic antagonist dihydro-β-erythroidine (DHβE, 0.6–3.0 mg/kg) shifted the dose–response for the discriminative stimulus effect of nicotine to the right; the α7 nicotinic receptor antagonist methyllycaconitine (MLA, 1.0–10 mg/kg) had no effect. The mice showed strong CTA to 2.0 mg/kg of nicotine and marginally to 0.6 and 1.2 mg/kg of nicotine. DHβE (3.0–5.6 mg/kg) attenuated the CTA while MLA (1.0–10 mg/kg) had no effect. These studies show that nicotine has discriminative and aversive stimulus properties in C57BL/6J mice and that the effects are mediated primarily by receptors sensitive to DHβE; there was no evidence for the involvement of α7 nicotinic receptors. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Nicotine; Drug discrimination; Conditioned taste aversion; Mice; Dihydro-β-erythroidine; Methyllycaconitine
1. Introduction Drug discrimination and conditioned taste aversion (CTA) techniques have been used extensively to study the stimulus properties of nicotine. The discriminative stimulus effects of nicotine serve as a behavioural assay to characterise the neuropharmacological mechanisms underlying nicotine’s behavioural effects. Most nicotine discrimination studies have been carried out in rats, with some in humans but surprisingly, experiments in mice have been reported only recently (Stolerman et al., 1999; Varvel et al., 1999). Both DBA and C56BL/6 strains could be trained to discriminate between nicotine and saline and the discrimination was dose-dependent (Stolerman et al., 1999; Varvel et al., 1999). The nicotine stimulus could be completely antagonised by the nonselective nicotinic receptor antagonist mecamylamine
* Corresponding author. Tel.: +44-020-7848-0370; fax: +44-0207848 0579. E-mail address: [email protected] (I.P. Stolerman).
(Stolerman et al., 1999; Varvel et al., 1999), but not by the muscarinic receptor antagonist scopolamine (Varvel et al., 1999). The aversive effects of nicotine as measured in a conditioned taste aversion might be involved in the regulation of nicotine intake by setting an upper limit to the amount taken (other interpretations for taste aversions produced by positively reinforcing drugs have been offered (Grigson, 1997; Stolerman and D’Mello, 1981). Etscorn (1980) and Risinger and Brown (1996) reported conditioned taste aversion experiments with nicotine in mice. In contrast to some other strains, with a one-stimulus procedure in which only the nicotine-paired flavoured solution was presented during testing, C57BL/6 mice did not show a CTA for nicotine in doses up to 2.0 mg/kg SC (Risinger and Brown, 1996). The present study aims to characterise further the discriminative and aversive stimulus effects of nicotine in mice. It is necessary to establish robust measures for nicotine in mice because some techniques for identifying genetic factors in susceptibility to drugs rely upon the use of recombinant inbred mice (Crabbe et al., 1999);
0028-3908/00/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 3 0 - 1
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
similarly, mice but not rats with targeted deletion of nicotinic receptor subunits have been constructed, on a C57BL/6 background (Picciotto et al., 1995). In preparation for such studies, C57BL/6 mice have been trained to discriminate 1.2 mg/kg of nicotine in a twolever discrimination procedure and two specific nicotine receptor antagonists have been tested. Grote and Brown (1971) showed that a conditioned taste aversion procedure with a two-stimulus test, in which animals could choose between two flavours of which one had previously been paired with cyclophosphamide, was more sensitive than a conditioned taste aversion procedure with a one-stimulus test. We have therefore used a twostimulus CTA procedure to investigate the aversive stimulus effects of nicotine and the effects of nicotine antagonists. The nicotine receptor has a pentameric architecture, consisting of combinations of α-, or α- plus β-subunits (Galzi and Changeux (1995), and in the brain there are at least eight different α-receptor subunits (α2–α9) and three different β-receptor subunits (β2–β4). For the present experiments antagonism studies were carried out with methyllycaconitine (MLA) and dihydro-β-erythroidine (DHβE). Studies using Xenopus oocytes suggested that MLA was an α7-receptor antagonist (Wonnacott et al., 1993) and that DHβE was a competitive nicotinic receptor antagonist at several subtypes of receptor. These included α4β2, α4β4 and α3β2 receptors, and to a lesser extent at α2β2 and α2β4 receptors (Harvey et al., 1996). DHβE has been shown to antagonise the nicotine discriminative stimulus and taste aversion effects in rats (Stolerman et al., 1997; Shoaib et al., 2000) and some of the behavioural effects of nicotine in mice (Damaj et al., 1995; Decker et al., 1995). In contrast, MLA has not been found to block nicotine discrimination in rats or its antinociceptive effect in mice (Brioni et al., 1996; Damaj et al., 1998). It was hypothesised that differential effects of these antagonists would help to reveal the diversity of subtypes of nicotinic receptors that may mediate the behavioural effects of nicotine.
2. Materials and methods 2.1. Animals For drug discrimination, 22 male C57BL/6J mice (Harlan Olac, Bicester, UK) were trained. They were placed on a restricted diet to maintain their body weight at approximately 85% of their free feeding weight; water was freely available. All animals were housed individually in a room at a controlled temperature (21°C) with a 12-h light–dark cycle (light from 07:30 to 19:30 h). For conditioned taste aversion a total of 88 C57BL/6J mice (Harlan Olac) was used. Food was freely available but water was restricted as below. All animals were
2841
housed individually in a room at a controlled temperature (21°C) with a 12-h light–dark cycle (light from 07:30 to 19:30 h). All studies complied with local ethical requirements and were carried out in accordance with the Animals (Scientific Procedures) Act, 1986, under licence from the UK Home Office. 2.2. Apparatus For drug discrimination, six standard operant chambers for mice (MED Associates, Georgia, VT) contained in ventilated, sound-insulated housings were used. Each chamber was equipped with two ultra-sensitive levers, one on each side of a recess in which a dispenser could deliver 25 mg pellets of food. All studies of conditioned taste aversion were carried out in the home cages. Drinking cylinders calibrated to 0.1 ml were attached directly to the cage. 2.3. Procedure for drug discrimination The procedure was based on that of Stolerman et al. (1999). Initially, mice were trained to consume food pellets delivered automatically under a variable-time 30 s schedule, until all food delivered was consumed by the end of a 15 min session. During this stage no levers were present in the chamber. Subsequently, one lever was inserted into the chamber and training to lever-press began. Mice were trained first under a continuous reinforcement schedule, then under a fixed-ratio schedule that increased progressively to FR-10. After this response was acquired, the original lever was removed and a single lever was inserted on the opposite site of the chamber and mice were again shaped to perform under the FR-10 schedule of food presentation. After animals acquired stable FR-10 responding a variableinterval (VI) component was introduced. Under this tandem schedule of reinforcement, the tenth lever press was reinforced after a randomly determined interval of time. Initially the mean value of the VI was 15 s, but after five sessions it was increased to 30 s (range 7–53 s). After this schedule was in operation for several sessions, discrimination training started. Both levers were present but responses on only one lever were reinforced, depending on the drug condition. Animals were injected 10 min prior to the start of sessions with either nicotine (1.2 mg/kg, SC) or saline. For half the animals, presses on the left lever were reinforced after nicotine and on the right lever after saline; for the other half of the animals, presses on the right lever were reinforced after nicotine and on the left lever after saline. Each mouse had a unique random order of nicotine and saline sessions except that there were never more than three nicotine or three saline sessions in succession. After several weeks
2842
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
of training the duration of the VI component of the schedule was increased to 60 s. After the mice acquired the nicotine discrimination as shown by 80% accuracy of lever selection in a block of 10 consecutive sessions, generalisation testing took place twice weekly. Mice were placed in the chambers for 5 min with both levers present but no reinforcers available. Mice that reached a criterion of 80% accuracy during initial tests with saline and the training dose of nicotine were used for subsequent experiments. For nicotine dose–response determinations, different doses of nicotine or saline were injected 10 min prior to testing. For antagonism tests, the antagonists were injected 30 min and nicotine 10 min prior to testing. Tests with different drug treatments were in random order. 2.4. Procedure for conditioned taste aversion The procedure was adapted from those described previously for rats (D’Mello et al., 1977; Kumar et al., 1983). Mice were trained to drink tap water for 30 min twice daily, once in the morning (between 9:00 and 10:30 h) and once in the afternoon (between 4:00 and 5:00 h). When the amount consumed was stable, conditioning began. Mice were presented in the morning session with either a sweet solution (sodium saccharin, 0.1%) or a saline solution (sodium chloride, 0.9%) on every second day. After drinking the solutions the mice were injected immediately with either nicotine or vehicle. There were a total of four conditioning sessions, two with each flavour (hence, two sessions with nicotine and two with vehicle). To balance out effects due to inherent differences in palatability of the two flavours, for half the animals nicotine was paired with the saccharin flavour and vehicle with the saline flavour, and for the other half nicotine was paired with saline and vehicle with saccharin. The two solutions were presented alternately and to balance out side preferences each flavour was presented once at the left side of the cage and once at the right side. During the afternoon drinking sessions and on intervening days, mice could drink normal tap water. Two days after the last conditioning session the mice were presented with both solutions simultaneously and were free to choose what to drink. On the next day, this test was repeated and the left–right position of the two flavours was reversed, to counterbalance possible side preferences. No injections were given after these two test sessions. Three different groups of C57BL/6J mice (n=8) were conditioned with different doses of nicotine (0.6, 1.2 and 2.0 mg/kg, SC). Antagonism tests were performed with DHβE (1.0, 3.0 and 5.6 mg/kg) and MLA (1.0, 3.0 and 10.0 mg/kg) in different groups of mice (n=8). The antagonists were injected at the same time as nicotine or vehicle, i.e. immediately after drinking sessions. To
avoid possible conditioning with the antagonists, they were administered after both nicotine and vehicle conditioning sessions. 2.5. Data analysis For drug discrimination, the number of responses made on each lever during extinction tests was recorded for the whole 5-min test session. The percentage of total responses that were made on the drug-appropriate lever served as a measure for stimulus generalisation; the total number of responses served as a measure for response rate. Data were analysed with repeated-measure analysis of variance, with arcsine transformations to normalise the distribution of percentage scores (Winer, 1970). For the antagonism studies ED50 values were calculated (by interpolation) as the dose of nicotine associated with 50% drug-appropriate responding. For conditioned taste aversion, the amounts consumed of the nicotine and vehicle paired flavours during the conditioning and two test sessions were measured. The mean amounts consumed of the drug-paired flavour as a percentage of total fluid intake were calculated. These percentage scores were subjected to arcsine transformations to normalise their distributions (Winer, 1970) and subsequently analysed with a one sample t-test to determine whether means differed significantly from 50%. A mean score of 50% would indicate that nicotine induced neither conditioned taste aversion nor conditioned taste preference; scores below 50% indicated CTA. This approach to quantifying the magnitude of CTA and to evaluation of significance has been used in previous studies (e.g. D’Mello et al., 1977; Kumar et al., 1983; Reavill et al., 1986). 2.6. Drugs Nicotine hydrogen-(+)-tartrate was dissolved in isotonic saline. The pH was adjusted to 7 with 0.5 N NaOH. Dihydro-β-erythroidine hydrobromide and methyllycaconitine citrate (both from Sigma-Aldrich, Poole, Dorset, UK) were dissolved in saline. All doses were calculated as those of the base. For drug discrimination, nicotine was injected 10 min prior to testing or training, DHβE and MLA were administered 30 min prior to testing; for conditioned taste aversion all drugs were administered immediately after the drinking sessions ended. All injections were given subcutaneous in a volume of 1 ml/100 g. 3. Results 3.1. Drug discrimination A total of 19 of the 22 mice trained reached the 80% criteria for accuracy and were then used for dose–
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
response tests with nicotine. These mice were divided into two groups for the purpose of testing two antagonists later. As can be seen from Fig. 1, both groups showed similar dose-dependent nicotine generalisation gradients (group to be tested with DHβE, F(4,36)=6.58, p⬍0.001, n=10; MLA group, F(4,32)=13.2, p⬍0.001, n=9). Nicotine also reduced the number of responses in both groups (F(4,36)=29.6, p⬍0.001 and F(4,32)=14.9, p⬍0.001, respectively). Fig. 2 shows the results of antagonism tests with DHβE and MLA. There was a clear difference in effect of the two antagonists: DHβE shifted the dose–response curve of nicotine to the right (F(2,18)=103.9, p⬍0.001, based on data for nicotine doses of 0.6–1.6 mg/kg), and with a 3.0 mg/kg dose of DHβE there was an almost complete block of the 1.2 mg/kg training dose of nicotine. MLA did not shift the nicotine dose–response curve (F(2,16)⬍1). The ED50 for nicotine alone was 0.6 mg/kg, whereas for nicotine with DHβE in doses of 0.6 mg/kg and 3.0 mg/kg the ED50s were 0.9 and 1.7 mg/kg, respectively. Thus, 3.0 mg/kg of DHβE shifted the dose– response curve for nicotine to the right by a factor of 2.8. Nicotine reduced response rates in both groups (F(5,45)=5.55, p⬍0.001) and F(4,32)=12.7, p⬍0.001);
2843
there was, however, no effect of either antagonist (DHβE: F(2,18)=3.54; MLA: F(2,16)⬍1). 3.2. Conditioned taste aversion Fig. 3 shows that there was a conditioned taste aversion to 0.6 mg/kg (34.4% consumption of drug-paired flavoured solutions, t(7)=2.7, p⬍0.05) and in particular to 2.0 mg/kg of nicotine (17.9%, t(7)=5.8, p⬍0.001). A dose of 1.2 mg/kg of nicotine failed to induce taste aversion (27.8%, t(7)=2.3). The intake of the saline-paired flavoured solutions was reduced on trial 3 because both flavoured solutions were presented simultaneously and mice could satisfy their fluid requirements by drinking some of each (there were no differences between the groups in total fluids consumed). The dose of 2.0 mg/kg of nicotine was considered to show a robust CTA, and therefore this dose was used for subsequent antagonism studies in which nicotine was tested in combination with different doses of DHβE or MLA. Saline or a 1.0 mg/kg dose of DHβBE had no effect on nicotine-induced CTA (nicotine alone, 12.3%, t(7)=5.4, p⬍0.001; nicotine plus DHβBE 1.0 mg/kg, 12.0%, t(7)=6.8, p⬍0.001). Fig. 4 shows that doses of 3.0 and 5.6 mg/kg of DHβBE antagonised nicotineinduced CTA, i.e. data were not significantly different from 50% (DHβBE 3.0 mg/kg, 30.6%, t(7)=1.9; DHβBE 5.6 mg/kg, 36.1%, t(7)=1.6). In trial 3, mice pretreated with 5.6 mg/kg of DHβBE consumed a total of 1.12 ml of the two flavoured solutions, as compared with 1.06 ml for the mice pretreated with saline. The reduced intake of the saline-paired flavoured solutions on trial 3 in mice pretreated with DHβBE can therefore be attributed to the concurrent intake of the nicotine-paired solutions. MLA had no effect on nicotine-induced CTA (Fig. 5). Nicotine showed strong CTA when combined with each dose of MLA (minimum t(7)=6.4, p⬍0.001 at each dose).
4. Discussion
Fig. 1. Dose–response curves for nicotine in two groups of mice trained to discriminate nicotine (1.2 mg/kg, SC) from saline. Upper panel: responses on the drug-appropriate lever as a percentage of total responses on both levers (mean±s.e.m.). Lower panel: total number of responses. Group 1 (䊏, n=10) was used subsequently for antagonism studies with DHβE; group 2 (쐌, n=9) was used for studies with MLA.
The present experiments confirm earlier reports that nicotine can produce dose-related discriminative stimulus effects in C57BL/6 mice (Stolerman et al., 1999; Varvel et al., 1999). The results also show that DHβE, but not MLA, blocked the discriminative stimulus effects of nicotine in mice. These findings are in agreement with previous studies in rats (Brioni et al., 1996; Shoaib et al., 2000; Stolerman et al., 1997), and support the view that α7 receptors are not involved in the discriminative stimulus effects of nicotine. Increasing the dose of nicotine reversed the block produced by DHβE. A 3.0 mg/kg dose of DHβE completely antagonised 0.6 and 1.2 mg/kg of nicotine but larger doses of nicotine (1.6 and 2.0 mg/kg) in combination with DHβE induced high
2844
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Fig. 2. Dose–response curves for the discriminative stimulus effect of nicotine in antagonism studies with DHβE (왖, 0.6 mg/kg and 䊏, 3.0 mg/kg, n=10), and MLA (䉬, 1.0 mg/kg, and 왖, 10 mg/kg, n=9). Vehicle controls are also shown (쐌). Upper panel: responses on the drugappropriate lever as a percentage of total responses on both levers (mean±s.e.m.). Lower panel: total number of responses.
Fig. 3. Nicotine-induced conditioned taste aversion in three groups of mice (n=8). Trials 1 and 2 were conditioning sessions in which flavoured solutions were paired with nicotine and saline; trial 3 was a two-stimulus test in which nicotine and saline-paired flavoured solutions were presented simultaneously (there were no drug injections). Nicotine (2.0 mg/kg) resulted in a strong taste aversion, nicotine 0.6 and 1.2 mg/kg had weaker effects (쐌, saline-paired flavour; 䊏, nicotine-paired flavour).
levels of nicotine-lever responding. This might indicate a competitive mechanism and has been reported before in nicotine discrimination in rats (Stolerman et al., 1997). In rats the nicotine dose–response curve was shifted to the right by a factor of 9.4, whereas in the present study the shift was much smaller (2.8). The present experiments also show for the first time
that nicotine can induce a conditioned taste aversion in C56BL/6 mice, in contrast to previous experiments (Risinger and Brown, 1996). There are major differences between these and the previous experiments. The first difference is the use of a two-stimulus test procedure in the present experiments versus a one-stimulus procedure by Risinger and Brown (1996). The two-stimulus pro-
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
2845
Fig. 4. Effect of DHβE on nicotine-induced conditioned taste aversion in four groups of mice (n=8). Nicotine (2.0 mg/kg) administered in combination with DHβE (0.0–1.0 mg/kg) resulted in taste aversion. Larger doses of DHβE (3.0 and 5.6 mg/kg) antagonised the taste aversion (쐌, saline-paired flavour; 䊏, nicotine-paired flavour). Other details as for Fig. 3.
cedure was reported to be more sensitive (Grote and Brown, 1971). In the present experiments, nicotine was administered at the end of a 30 min drinking period as compared with a 60 min period in Risinger and Brown (1996). As most fluid intake occurs early in the drinking periods, the interval between exposure to the flavours and drug effect was shorter in the present experiments. A further advantage of the present two-flavour procedure is that animals serve as their own controls and there is no need for an extra group of animals that does not receive nicotine. The experiments also show that DHβE could antagonise nicotine-induced CTA, as has been previously reported for rats (Shoaib et al., 2000), but MLA had no effect on nicotine induced CTA. This argues for a role of α4β2, α4β4 and/or α3β2-receptors and possibly α2β2 or α2β4 receptors (but not for α7 receptors) in nicotineinduced CTA. Other injection-test intervals and routes of administration should also be examined. However, Turek et al. (1995) reported, in rats 6.2 µmol/kg (IP) of MLA (equivalent to 5.4 mg/kg), produced maximal brain concentrations of MLA after 30 min. For the present experiment both smaller and larger doses of MLA were used, administered 30 min before testing.
The results show some congruence between drug discrimination and CTA with respect to the involvement of specific nicotinic receptors. However, due to the lack of more selective antagonists, the present data do not make it possible to differentiate between receptor subtypes in a more conclusive manner. In the present study, the response rates were reduced by nicotine, but this effect was not antagonised by DHβE. In rats, DHβE also failed to antagonise the locomotor depressant and the operant response rate-reducing effect of nicotine (Stolerman et al., 1997). In mice there is conflicting evidence for antagonism by DHβE of effects of nicotine on motoric responses. In ICR mice DHβE antagonised nicotine-induced motor impairment in a rotarod test and partially antagonised nicotinereduced locomotor activity (Damaj et al., 1995). In CD1 mice, however, DHβE failed to antagonise nicotinereduced locomotor activity (Decker et al., 1995). Strain differences, different tests, and different doses of DHβE used might account for these different results and make it difficult to generalise between studies. One can conclude that DHβE antagonises nicotine-affected motoric behaviours under specific experimental circumstances. In conclusion, nicotine induces both discriminative
2846
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Fig. 5. Lack of effect of MLA on nicotine-induced conditioned taste aversion in four groups of mice (n=8). Nicotine (2.0 mg/kg) in combination with MLA (0.0–10.0 mg/kg) resulted in taste aversions (쐌, saline-paired flavour; 䊏, nicotine-paired flavour). Other details as for Fig. 3.
stimulus and a conditioned taste aversion effects in C57BL/6 mice; these effects are mediated by receptors sensitive to DHβE, but there is no evidence yet for a role of receptors that contain the α7 subunit. The availability of genetically manipulated mice is one reason for establishing these behavioural methods. There are at the moment mice that lack the genes coding for several subunits of nicotinic receptors, including α3 (Xu et al., 1999), α4 (Marubio et al., 1999), α7 (Orr-Urtreger et al., 1997), α9 (Vetter et al., 1999) and β2 (Picciotto et al., 1995). Future studies with these knock-out mice might show more precisely the role of the different nicotinic receptor subunits.
Acknowledgements This work was supported by an EU-TMR and a MRC research grant.
References Brioni, J.D., Kim, D.J.B., O’Neill, A.B., 1996. Nicotine cue: lack of effect of the α7 nicotinic antagonist methyllycaconitine. European Journal of Pharmacology 301, 1–5.
Crabbe, J.C., Phillips, T.J., Buck, K.J., Cunningham, C.L., Belknap, J.K., 1999. Identifying genes for alcohol and drug sensitivity: recent progress and future directions. Trends in Neurosciences 22, 173–179. Damaj, M.I., Fei-Yin, M., Dukat, M., Glassco, W., Glennon, R.A., Martin, B.R., 1998. Antinociceptive responses to nicotinic acetylcholine receptor ligands after systemic and intrathecal administration in mice. Journal of Pharmacology and Experimental Therapeutics 284, 1058–1065. Damaj, M.I., Welch, S.P., Martin, B.R., 1995. In vivo pharmacological effects of dihydro-β-erythroidine, a nicotinic antagonist, in mice. Psychopharmacology 117, 67–73. Decker, M.W., Anderson, D.J., Brioni, J.D., Donnelly-Roberts, D.L., Kang, C.-H., O’Neill, A.B., Piattoni-Kaplan, M., Swanson, S., Sullivan, J.P., 1995. Erysodine, a competitive antagonist at neuronal nicotinic acetylcholine receptors. European Journal of Pharmacology 280, 79–89. D’Mello, G.D., Stolerman, I.P., Booth, D.A., Pilcher, C.W.T., 1977. Factors influencing flavour aversions conditioned with amphetamine in rats. Pharmacology Biochemistry and Behavior 7, 185–190. Etscorn, F., 1980. Sucrose aversions in mice as a result of injected nicotine or passive tobacco smoke inhalation. Bulletin of the Psychonomic Society 15, 54–56. Galzi, J.-L., Changeux, J.-P., 1995. Neuronal nicotinic receptors: molecular organization and regulations. Neuropharmacology 34, 563–582. Grigson, P.S., 1997. Conditioned taste aversions and drugs of abuse: a reinterpretation. Behavioural Neuroscience 111, 129–136. Grote, F.W., Brown, R.T., 1971. Conditioned taste aversions: twostimulus tests are more sensitive than one-stimulus tests. Behavioural Research Methods and Instrumentation 3, 311–312.
J. Gommans et al. / Neuropharmacology 39 (2000) 2840–2847
Harvey, S.C., Maddox, F.N., Luetje, C.W., 1996. Multiple determinants of dihydro-β-erythroidine sensitivity on rat neuronal nicotinic receptor α subunits. Journal of Neurochemistry 67, 1953–1959. Kumar, R., Pratt, J.A., Stolerman, I.P., 1983. Characteristics of conditioned taste aversion produced by nicotine in rats. British Journal of Pharmacology 79, 245–253. Marubio, L.M., Arroyo-Jimenez, M.D.M., Cordero-Erausquin, M., Le´na, C., Le Nove`re, N., D’Exaerde, A.D.K., Huchet, M., Damaj, M.I., Changeux, J.-P., 1999. Reduced antinociception in mice lacking neuronal nicotinic receptor subunits. Nature 398, 805–810. Orr-Urtreger, A., Go¨ldner, F.M., Saeki, M., Lorenzo, I., Goldberg, L., De Biasi, M., Dani, J.A., Patrick, J.W., Beaudet, A.L., 1997. Mice deficient in the α7 neuronal nicotinic acetylcholine receptor lack αbungarotoxin binding sites and hippocampal fast nicotinic currents. Journal of Neuroscience 17, 9165–9171. Picciotto, M.R., Zoli, M., Le´na, C., Bessis, A., Lallemand, Y., Le Nove`re, N., Vincent, P., Pich, E.M., Bruˆlet, P., Changeux, J.-P., 1995. Abnormal avoidance learning in mice lacking functional highaffinity nicotine receptor in the brain. Nature 374, 65–67. Reavill, C., Stolerman, I.P., Kumar, R., Garcha, H.S., 1986. Chlorisondamine blocks acquisition of the conditioned taste aversion produced by (–)-nicotine. Neuropharmacology 25, 1067–1069. Risinger, F.O., Brown, M.M., 1996. Genetic differences in nicotineinduced conditioned taste aversion. Life Sciences 58, 223–229. Shoaib, M., Zubaran, C., Stolerman, I.P., 2000. Antagonism of stimulus properties of nicotine by dihydro-β-erythroidine (DHβE) in rats. Psychopharmacology 149, 140–146. Stolerman, I.P., Chandler, C.J., Garcha, H.S., Newton, J.M., 1997. Selective antagonism of behavioural effects of nicotine by dihydroβ-erythroidine in rats. Psychopharmacology 129, 390–397. Stolerman, I.P., D’Mello, G.D., 1981. Oral self administration and the
2847
relevance of conditioned taste aversions. In:. Thompson, T., Dews, P.B., McKim, W.A. (Eds.), Advances in Behavioral Pharmacology, vol. 3. Academic Press, New York, pp. 169–214. Stolerman, I.P., Naylor, C., Elmer, G.I., Goldberg, S.R., 1999. Discrimination and self-administration of nicotine by inbred strains of mice. Psychopharmacology 141, 297–306. Turek, J.W., Kang, C.-H., Campbell, J.E., Arneric, S.P., Sullivan, J.P., 1995. A sensitive technique for the detection of the α7 neuronal nicotinic acetylcholine receptor antagonist, methyllycaconitine, in rat plasma and brain. Journal of Neuroscience Methods 61, 113– 118. Varvel, S.A., James, J.R., Bowen, S., Rosecrans, J.A., Karan, L.D., 1999. Discriminative stimulus (DS) properties of nicotine in the C57BL/6 mouse. Pharmacology Biochemistry and Behavior 63, 27–32. Vetter, D.E., Liberman, M.C., Mann, J., Barhanin, J., Boulter, J., Brown, M.C., Saffiote-Kolman, J., Heinemann, S.F., Elgoyhen, A.B., 1999. Role of α9 nicotinic Ach receptor subunits in the development and function of cochlear efferent innervation. Neuron 23, 93–103. Winer, B.J., 1970. Statistical Principles in Experimental Design, second ed. McGraw-Hill, London. Wonnacott, S., Albuquerque, E.X., Bertrand, D., 1993. Methyllycaconitine: a new probe that discriminates between nicotinic acetylcholine receptor subclasses. Methods in Neurosciences 12, 263–275. Xu, W., Gelber, S., Orr-Urtreger, A., Armstrong, D., Lewis, R.A., Ou, C.-N., Patrick, J., Role, L., De Biasi, M., Beaudet, A.L., 1999. Megacystis, mydriasis, and ion channel defect in mice lacking the α3 neuronal nicotinic acetylcholine receptor. In: Proceedings of the National Academy of Sciences of the United States of America 96., pp. 5746–5751.