Tobacco Documents Online

I I I I I I I I I I I I I I I I I 8. THE PSYCHOPHAR?VIACOLOGICAL AND NEUROCHEMICAL CONSEQUENCES OF CHRONIC NICOTINE ADMINISTRATION David J.K. Balfour INTRODUCTION For many people their first exposure to tobacco smoke is often a relatively unpleasant experience. However, for those who go on to become regular smokers, the rewarding aspects of the smoke soon become predominant and the habitual smoker appears to develop complete tolerance to the aversive properties of tobacco smoke. Although it is clear that the rewarding properties of tobacco smoke are multi-factorial, a substantial body of evidence now supports the hypothesis that a majority of people who smoke tobacco on a regular basis, do so in order to absorb the nicotine present in the smoke and that a proportion of these regular smokers become dependent upon nicotine (Gilbert 1979; Balfour 1984, 1989; Clarke 1987). Thus, nicotine preparations. either in the form of nicotine-containing chewing gum (Fagerstrom 1988) or, more recently, nicotine skin patches (Abelin et a! 1989), have been introduced as a "treatment" for the symptoms of smoking abstinence. This approach has been criticized because it seems to involve the simple transference of the dependence from one form of nicotine presentation to another (Hughes et a! 1986; Hughes 1988). It is also clear. however, that many people become regular smokers without apparently developing a strong dependence upon any constituent in the smoke (Gilbert 1979; Balfour 1984). This presentation will, therefore, focus on some of the factors which may influence the development of dependence upon nicotine and some of the neural mechanisms which may mediate these effects of the drug. Drug dependence is classically divided into two categories-psychological and physical. Psychological dependence is characterised by a desire to continue taking the drug and is almost certainly a property of all drugs of dependence. Drugs such as the opiates and barbiturates, which in addition cause physical dependence, evoke an overt and characteristic withdrawal syndrome when the drug taking is stopped. For these compounds, avoidance of the withdrawal syndrome is thus likely to contribute significantly to the maintenance of drug-taking behaviour. There is no substantial evidence to suggest that nicotine can cause physical dependence and therefore, any 90 I

I 1 I I I I I I I I I I I I I Chronic .ti'icotine AdmrntstrQtrnn 91 dependence it does cause must be psychological in nature tBaltour 1989). Nevertheless, in humans smoking abstinence can be associated with quite distressing symptoms (Gilbert 1979; West 1988) and the rationale for using nicotine substitution during the initial stages of withdrawal is based on the hypothesis that it will attenuate these symptoms. One explanation for the appearance of symptoms in abstinent smokers is that, although nicotine does not induce physical dependence, in the sense that chronic administration is associated with changes in the control of homeostatic systems, it may influence the limbic systems in the brain which control "emotional" behav- iour. As a result, abstinence may be associated with distressing changes in mood or anxiety state. If this hypothesis is correct then it must not only explain the symptoms which appear following abstinence but also explain the fact that only a proportion of regular smokers become dependent. It seems reasonable to suggest that the development of dependence upon drugs which act on the limbic systems of the brain, will be influenced by the initial emotional state of the recipient and external factors, which may exaggerate or attenuate the psychopharmacological response to the drug. For example, in the case of nicotine, it has been suggested that the drug may exert a"tranquillising" or "anti-panic" effect and that the symptoms of abstinence may simply reflect the expression of incipient panic disorder that was controlled by the drug (Gilbert 1979; Brodsky 1985), although the possible relationship between the symptoms associated with nicotine absti- nence in incipient panic disorder has been challenged (Dilsaver 1987). Balfour (1984) has also summarised evidence from a number of laborator- ies which suggested that nicotine dependence was more likely to occur if the recipients were exposed repeatedly to aversive environmental stimuli. These animal data support the hypothesis that there may be a substantial variation in an individual's predisposition to becoming dependent upon nicotine and that this may also be influenced to a significant degree by the external stimuli to which the individual is exposed. This chapter will consider the evidence that chronic nicotine administ- ration does induce changes in brain neurochemistry which could be asso- ciated with its ability to cause dependence and will seek to relate the changes to behavioural responses which are thought to reflect nicotine abstinence. It will focus on the effects of nicotine, on the density of nicotinic cholinocxp- tors in the brain and on the regionally-selective changes in hippocampal 5- HT observed in both experimental animals and human subjects treated chronically with nicotine. CHRONIC NICOTINE AND THE DENSITY OF NICOTINIC RECEPTORS IN THE BRAIN Radioligand binding studies have identified at least two putative nicotinic  ZZ ~ CD 00 ~

I I I I I I I I I I I I I I I I 92 David J.K Balfour cholinoceptors in mammalian brain One binds to ( -)-nicotine and other ganglionic agontsts. includin¢ acet}lcholtne (ACh). with high affinny (Romano and Goldstein 1980: Benwell and Balfour 1985). The other binds to alpha-bungarotoxin with high affinity but has little affinity for nicotine (Wonnacott 1987). The precise role of each of these receptors has not. as yet, been fully established, although there is clear support for the hypothesis that the binding site with high affinity for nicotine is a central nicotinic receptor for acetylcholine (Clarke et a1 1985: Martino-Barrows and Kellar 1987). The results of behavioural studies suggest that the ability of nicotine to act as a discriminative cue in a drug discrimination paradigm, is mediated by the receptor with high affinity for nicotine (Romano et al 1981: Pratt et al 1983: Reavill er a1 1988; Goldberg et a/ 1989). There is also good evidence to suggest that the stimulant effects of nicotine on locomotor activity are also mediated by the same receptor (Clarke and Kumar 1983). Studies in a number of laboratories have shown that the chronic administration of nicotine to experimental animals can result in an increase in the density of the receptors with high affinity for nicotine when measured at postmortem using radioligand binding (Schwartz and Kellar 1983. 1985; Marks et a! 1983; Nordberg et a1 1985). Studies in our laboratory (Benwell et al 1988) have shown that the density of the receptors is also increased in brain tissue taken from human subjects who smoked tobacco. The mechanism by which this occurs is not yet fully understood, although both Schwartz and Kellar (1985) and Wonnacott (1987) have expressed the view that the up-regulation of the receptor is associated with prolonged or repeated desensitization of the receptor complex caused by the chronic exposure to the drug. This conclusion, however, is difficult to reconcile with the fact that chronic treatment with centrally-acting acetylcholinesterases inhibitor results in a decrease in the density of the receptor sites (Costa and Murphy 1983; Schwartz and Kellar 1983). It would seem reasonable to expect chronic exposure to acetylcholine to evoke the same change in receptor density whether it was induced by an anticholinesterase or chronic nicotine. Kellar et a! (1988) suggested that at 37=C. acetylcholine dissociates from the receptor complex more quickly than nicotine and therefore, the receptors responded in a manner consistent with repetitive stimulation, whereas nicotine evokes desensitization blockade. However, this explanation also seems unlikely because this difference in sensitivity does not occur at other nicotinic receptors. A recent study by De Sarno and Giacobini (1989) has shown that, in contrast to the results obtained with the organophosphates, the repeated intraventricular administration of the short-acting acetykholinesterase inhibi- tor, physostigmine, causes up-regulation of the receptor. These authors concluded that pulses of acetylcholine (or nicotine) can cause a protracted desensitization and eventually, inactivation of the receptor complex and that I

1 I I I 1 I I I I I I I I I I I I I I Chrnnrc Nirorine 4dnunicrrunon y, as a result. up-regulation of the receptor is obserned u hereas f+rolongtd exposure to constant levels do not cause inactivation of central nicotinic receptors. Interestingly, most of the experimental schedules which have been used successfully to induce up-reguiation of the receptor involve multiple dosing with nicotine. the situation seen par excellence in people who smoke cigarettes. The molecular mechanisms which underlie the increase in high affinity nicotine binding induced by chronic nicotine administration. remain to be established. Romanelli er al (1988) have suggested that the receptor exists in two affinity states and that the apparent increase in the density of the high affinity nicotine binding sites, reflects an increase in the proportion of the receptors existing in the conformation which has high affinity for nicotine at the expense of the proportion with low affinity for nicotine. Others. however. suggest that the increased density is associated with an enhanced expression of receptor protein within the neuronal membranes (Wonnacott 1987). The possible pharmacological significance of the change in receptor density also remains to be established. Perhaps the most thorough series of studies of the relationships between the density of nicotinic receptors in the brain and centrally-mediated responses to the drug. has been performed by Collins and his colleagues using mice as the experimental animal (e.g. Marks er al 1983. 1985. 1986a, b). These studies have revealed a number of significant correlations between the increase in the density of the high affinity. nicotinic receptors and the development of tolerance to the beha- vioural and physiological responses to the drug. The nicotine infusion protocol that this group use to produce up- regulation of the receptors, also often results in increased ['=SIJ-alpha- bungarotoxin binding to the putative nicotinic receptors with high affinity for the toxin but low affinity for nicotine. However, the effects on toxin binding appear to be more regionally-selective than the changes in the density of high-affinity nicotine binding sites and to be influenced to a greater extent by genetic factors (Marks et al 1983, 1986a, b). The development of tolerance to nicotine also correlates to some extent. with the changes [12SI]-alpha-bungarotoxin binding although the correlations arr less impressive than those observed with the binding site at which ['H)- nicotine is the preferred ligand. Studies using different strains of mice suggest that it is unlikely that the correlations reflect a functional relation- ship between the changes in receptor density and the development of tolerance to nicotine, since the genetic factors which appear to influence the development of nicotine tolerance, do not appear to be associated with strain differences in the susceptibility of the receptors to chronic nicotine treatment (Marks er al t986a, b).

I I I I I I I I I I I 1 I 94 David J.K. Balfour Ksir and his colleagues (1985) ha~e reported that daik iniection< of relatively small doses of nicotine (0.1 to 0.4mg kg). cause increases in the density of the nicotinic receptors with high affinity for acetylcholine and nicotine after only five days. In addition. this increase in receptor densit% correlates with the increase in locomotor activity seen in animals treated with the drug for this period of time. These authors suggested that there was a functional relationship between the increased locomotor response to the drug and the change in receptor density because when the compound was withdrawn for 21 days after a period of chronic treatment, the density of the receptors returned to control levels and the enhanced locomotor response was lost. However, other studies suggest that there is no functional relationship between the density of nicotinic receptors in the brain and the locomotor response to nicotine. For example, Benwell and Balfour (1985) found that 40 daily subcutaneous injections of nicotine (0.4mg kg) had no significant effect on the density of nicotinic receptors in rat brain, in spite of the fact that this treatment schedule is known to result in an enhanced locomotor response to the drug. In addition. a study by Collins and colleagues (1988) found that although the development of tolerance to the depressant effects of a high dose of nicotine. (twice daily injections of 1.6mg, kg subcutaneously) correlated with up-regulation of the receptors with high affinity for nicotine, the tolerance persisted for many days in nicotine- withdrawn rats whereas the increased receptor density did not. These authors concluded that. although up-regulation of the receptor may be associated with the development of tolerance to nicotine, the density of the receptors did not appear to play a role in behavioural tolerance. The possibility that the changes in receptor density evoked by chronic nicotine administration may be involved in the development of nicotine dependence has also been the subject of some speculation (Schwartz and Kellar 1985; Wonnacott 1987; Benwell et a! 1988; Balfour 1989). However, no study has yet provided unequivocal evidence that this is the case. It is generally assumed that the majority. if not all, of the psychological effects of nicotine that are sought by smokers are associated with stimulation of central nicotinic receptors (Clarke 1987). However, it is also assumed that the increase in receptor density is caused by prolonged or repeated desensitiza- tion blockade of the receptor complex. Thus. although it seems likely that the rewarding properties of the drug are mediated primarily by stimulation of nico- tinic receptors, there are no reports which directly link the effects of nicotine abstinence with either stimulation or blockade of the receptor complex. A recent study by Lapchak et a! (1989) has shown that the addition of the nicotinic agonist, N-methylcarbamylcholine (MCC) to superfused brain slices prepared from the cerebral cortex or hippocampus of rats pretreated with nicotine and producing an increase in the density of central nicodnic receptors, had no effect on the release of acetylcholine. In contrast, I

I I I I I I I I I I I I 1 I I I I Chronic Ntcntrne .admtrrtcrrarwut '~5 "Artrvated" AgonrrtlReaptor I ~ Complex Prolonged or ~ repeated exposure to agomst Inaeased density of "resensrtrsed" receptor? Agon s: Receptor "Destnsitrsed" AQonist/Receptor Complex Chronic treatment Ii Drug withdrawal Receptor Proliferation I FiYurc t Diagrammatic representation of the mechanisms which may mediate the proliferation of high•affinuy nicotine binding sites in mammalian brain, when the experiment was performed with slices prepared from drug-naive rats. the addition of MCC caused a substantial increase in acetvlcholine secretion. The authors concluded that the pretreatment had caused desensiti- zation of the presynaptic nicotinic receptors which mediated the effects of MCC on acetylcholine secretion. If nicotine treatment was suspended for 4 days prior to the experiment, there was a significant recovery in the response to MCC in the superfused slice preparation although the density of the nicotinic receptors remained elevated when compared with controls. These results provide clear support for the hypothesis that up-regulation of the receptors is associated with desensitization of the receptor complex and that for a period following withdrawal of the drug, the receptors remain in a desensitised state. Thereafter re-sensitisation occurs and after a period of time, at least 10 days in the study reported by Schwartz and Kellar (1985), the receptor density returns to control levels. This proposed sequence of events is summarised in Figure 1. Interestingly, in their study Lapchak et a! (1989) did not observe an enhanced response to the effects of MCC on tissue prepared from the nicotine-withdrawn rats suggesting that in their protocol at least, the increased receptor density was not associated with a super sensitive response to nicotinic agonists. There is evidence to suggest that abstinent smokers can ameliorate some, although certainly not all, of the effects of smoking withdrawal by using preparations such as nicotine gum, NicoretteR, which release nicotine into the systemic circulation relatively slowly (Russell 1988). However, the absorption of nicotine from gum does not mimic the pleasure and satisfaction obtained from inhaling tobacco smoke, or reduce, to any significant extent, the desire to smoke often experienced during the early stages of smoking abstinence. (See chapter by

I I I I I I I I I I I I I I I I I 96 David J.K. Balfour West. this %olume.) If up-regulation of the receptors is caused b~ dzszn~ uza- tion of the receptor complex. it seems reasonable to suggest that cheAing nicotine gum will maintain plasma nicotine levels at a concentration u hich will keep the receptor complex in the desensitised state. and that the abstinence effects which are attenuated by this type of procedure are those which result from reactivation of the receptor complex. This conclusion implies that these symptoms should also be ameliorated by drugs which antagonise the effects of nicotine and acetylcholine at these receptors and in this context, it is interesting that nicotine antagonists do appear to reduce abstinence symptoms in some abstinent smokers (Stoler- man 1986). Currently, however, there are no suitable selective antagonists available for the central nicotinic receptor and the effects of these drugs at peripheral ganglionic receptors make them unpleasant to take and therefore. unlikely to be clinically valuable for smoking cessation. It is also clear that the changes in nicotinic receptor density evoked by chronic nicotine admi- nistration and the process of desensitization which is thought to cause the up-regulation. do not, in themselves, fully account for all the behavioural changes associated with chronic exposure to nicotine. THE EFFECTS OF NICOTINE ON 5-SEROTONIN SYSTEMS IN THE BRAIN Studies completed some years ago in our laboratory, showed that the chronic administration of nicotine to rats caused a regionally-selective decrease in the concentrations of 5-hydroxytryptamine (5-HT) and its principal metabolite. 5-hydroxyindoleactic acid (5-HIAA) in the hippocam- pus (Benwell and Balfour 1979). Subsequent studies showed that this effect was associated with a decrease in 5-HT biosynthesis in synaptosomes prepared from the hippocampus of treated rats and an adaptive reduction in the density of tryptophan carrier sites in nerve terminal membranes in the hippocarrtpus (Benwell and Balfour 1982a). Synaptosomes prepared from the hippocampus of rats treated acutely with nicotine also exhibited decreased 5- HT biosynthesis but normal tryptophan uptake. These data suggest that the administration of nicotine to experimental rats causes a reduction in the turnover of 5-HT in the hippocampus and that chronic treatment evokes adaptive changes in the biosynthetic pathway which reflect a penistent reduction in the demand for 5-HT. The evidence for reduced 5-HT turnover is consistent with the hypothesis that nicotine decreases the release of 5-HT in the hippocampus although studies in other laboratories (Wolf et Q! 1985; Kuhn et al 1986) have drawn attention to the fact that changes in turnover do not necessarily reflect changes in release. More recent studies in our laboratory, using human brain tissue taken at I

I I I I I I I I I I I I I I I I Chronic ~'icottne .-tdmtnistrattnn 9" postmortem. have shown that tobacco smoking is also sssoci3ted uttn regionally-selective reductions in the concentrations of 5-HT and 5-HI.AA in the hippocampus (Benwell et al, in press). The concentration of 5-HIAA in the median raphe nuclei was also reduced in the tissue of the smokers whereas the concentrations of both 5-HT and 5-HIAA in the gyrus rectus. cerebellar cortex, and medulla oblongata appeared to be unaffected by smoking. These studies also showed that radioligand binding to 5-HT,, but not 5-HT2 receptors in the hippocampus are increased in tissue taken from subjects who had smoked tobacco. The increase in binding to the 5-HT1 receptors appeared to be due. at least in part, to an increase in the density of 5-HT,,, receptors in this region of the brain. The data from the study with human tissue is clearly very similar to those obtained with experimental rats injected with nicotine and, thus, it seems reasonable to suggest that changes observed in human brain were caused by the nicotine present in tobacco smoke. In addition, the up-regulation of the 5-HT receptors adds some support to the hypothesis that the reduction in 5- HT turnover induced by nicotine is associated with a decrease in 5-HT secretion in the hippocampus, since it seems unlikely that the density of neuronal receptors are influenced by purely intra-cellular events. Clearly, however, this will need to be confirmed using more direct measures of 5-HT release. The possible psychopharmacological significancx of the changes in hippo- campal 5-HT evoked by the administration of nicotine remain to be established. However, there is clear evidence to suggest that exposure to environmental stress enhances the desire to smoke and that many people find that the inhalation of tobacco smoke in these circumstances exerts a "tranquillising" effect (Gilbert 1979). For many years now it has been thought that the serotonergic pathways of the brain may be involved in the mechanism of action of anti-anxiety drugs, a view that has been reinforced by the discovery of putative anziolytic drugs which seem to act selectively on serotonergic neurones (Jones et al 1988; Riblet et al 1982). Much of the work in this laboratory therefore, has focused on the possibility that the effects of nicotine on hippocampal 5-HT may be involved with its effects on responses to aversive environmental stimuli. In many studies with experimental animals, plasma corticosterone levels are used as a measure of the response to an anxiogenic stimulus and we used this index initially to investigate the efferts of nicotine. Acute administration of the drug caused an increase in the plasma corticosterone concentration. However, rats readily developed tolerance to this effect and after only five daily injections of nicotine (0.4mg/kg), no increase in plasma corticosterone is observed (Benwell and Balfour 1979). In contrast, the withdrawal of nicotine, following a period of chronic treatment (19 days or more), was associated with a significant increase in plasma corticosterone, when com-

I I I I I I I I I I I I I I I 98 David J.K. Balfour pared with saline-treated control rats. Other studies have shown that the concentrations of ~-HT and 5-HIAA in the hippocampus appear to be influenced by the circulating levels of corticosterone in the plasma (Balfour and Benwell 1979). Inspection of the data from the nicotine study. however. failed to reveal any clear relationship between the changes in plasma corticosterone and the concentrations of 5-HT and 5-HIAA in the hippo- campus (Benwell and Balfour 1979). A subsequent study, (Balfour er al 1986) has shown that the selective destruction of the principal serotonergic pathways to the hippocampus had no effects on plasma corticosterone per se, but that they did attenuate the development of tolerance to the effects of nicotine on plasma corticosterone. Another series of experiments designed to investigate the possible role of hippocampal 5-HT in the mechanism by which animals habituate to an aversive environmental stimulus, showed that rats exposed acutely to an aversive environment (an elevated open platform) exhibited significantly increased levels of 5-HT in the hippocampus (Copland and Balfour 1987). Repeated daily exposure to the test environment resulted in habituation of the plasma corticosterone response and a reduction in the concentration of 5-HT in the hippocampus, habituation being associated with the appearance of a significant. positive and regionally-selective correlation between the concentrations of 5-HT in the hippocampus and plasma corticosterone (Benwell and Balfour 1982b; Copland and Balfour 1987). However, if the rats were given a nicotine injection prior to each session in the aversive environment, habituation of the corticosterone response was significantly attenuated and the correlation between the corticosterone concentration and hippocampal 5-HT was significant but negative (Benwell and Balfour 1982b). These data were taken as clear evidence that nicotine interacts with the process by which rats habituate to aversive environments. In contrast, the administration of the anxiolytic drug. diazepam. enhanced habituation to the apparatus. particularly given at high doses, and did not influence the relationship between hippocampal 5-HT and plasma corticosterone in the stress-habituated rats (Copland and Balfour 1987). The mechanisms which mediate the effects of nicotine on habituation to stressors remain to be established and there is as yet, no evidence that the drug exerts a direct effect on the 5-HT pathways which innervate the hippocampus. Nevertheless, it seems unlikely that the effects of nicotine on plasma corticosterone in both stressed and unstressed rau, are directly related to its ability to act as a motor stimulant, since a preliminary study has shown d-amphetamine does not influence plasma corticosterone levels in this paradigm (Balfour and Iyaniwura 1984a, b). There are a number of reports in the literature which suggest that smoking cessation can be associated with abstinence symptoms, although the evidence that these symptoms are related specifically to the withdrawal of nicotine is CC~ 00 U ~ ~

I I I I I I I I I I I I I I I I I Chronic Nicotine Admzntsrrarion 99 not conclusive (Fagerstrom 1988: West 1988). Nevertheless. it is widel~ assumed that nicotine withdrawal does produce some of the changes in mood which are reported to occur in abstinent smokers, so studies in a number of laboratories have used animal models to investigate the conse- quences of nicotine-withdrawal. There is no evidence to suggest that the withdrawal of nicotine from experimental animals causes any overt changes in unconditioned behaviour which could be construed as an abstinence syndrome and so nicotine differs significantly, in this respect, from drugs such as the opiates and barbiturates which cause physical dependence (Balfour 1984). However, studies with animal models have shown that withdrawal of the compound is associated with some behavioural and endocrinological changes. One of the earliest of these studies showed that the withdrawal of nicotine from rats trained on a shock avoidance schedule under the influence of the drug, evoked a marked disruption of the avoidance behaviour (Morrison, 1974). Interestingly, this withdrawal effect appeared to be influenced by the avoidance procedure used, the effect being most marked in rats trained on an unsignalled shock avoidance schedule, the most stressful form of the paradigm. Balfour (1984) suggested that these data were consistent with the hypothesis that nicotine ameliorated the disruptive effects of the stressful experience on performance and that the rats became dependent upon this property of the drug in order to maintain their avoidance behaviour. In a subsequent study it was found that, although avoidance performance correlated with the concentration of 5-HT in the hippocampus, there was no evidence to suggest that the disruption of performance observed in the withdrawn rats was, in any way, associated with the reduction in hippocam- pal 5-HT evoked by the administration of nicotine (Balfour and Morrison 1975). A more recent study by Balfour (unpublished) has shown that d- amphetamine can reverse completely the effects of nicotine-withdrawal in this model (Figure 2), which suggests that the rats may have become dependent upon some catecholaminergic effects of nicotine for the mainten- ance of avoidance behaviour. Other studies have focussed on the possibility that nicotine-withdrawal may evoke a state of heightened anxiety and have used tests designed to show that withdrawal of the compound from experimental animals is associated with responses similar to those seen in animals given an anxioge- nic drug. Benwell and Balfour (1979) reported that the withdrawal of nicotine from unstressed rats resulted in a modest but significant increase in plasma corticosterone, which is clearly consistent with the hypothesis. While studies in this laboratory have shown that the withdrawal of the anxiolytic drug, diazepam, from unstressed rats has no effect on plasma corticosterone (Copland and Balfour 1987), there is clear evidence from human studies that benzodiazepine withdrawal is anxiogenic (Tyrer 1984).