Abstract
Chronic nicotine administration in animal models evokes a dose-dependent increase in brain nicotinic receptor numbers. Genetically determined variability in nicotinic receptor number in different mouse strains has also been reported, which is thought to affect sensitivity to nicotine, as well as the development of tolerance. Humans self-administer nicotine principally in the form of cigarettes and other tobacco products. The present study compared [3H]nicotine binding in human postmortem brain from thalamus and hippocampus of nonsmoking subjects, subjects who had variable life-long smoking histories and subjects who had quit smoking. A significant increase was seen in [3H]nicotine binding in both hippocampus and thalamus of subjects with life-long smoking histories. In the hippocampus, this change resulted from a change in total receptor number (B max), with no change in receptor affinity (K d). There was also a positive correlation between the degree of smoking, as measured by the average reported packs smoked per day, and the number of nicotine binding sites found in both the hippocampus and thalamus, showing that humans exhibit a dose-dependent increase in brain nicotinic receptor binding. Receptor levels in these brain regions after smoking cessation were at or below those found in the control population, which indicated that smoking-induced changes are reversible after cessation of nicotine treatment. These results suggest that increases in nicotinic receptor levels in the human brain may underlie nicotine tolerance and addiction in smokers.
Chronic treatment with agonists for most neurotransmitter receptor systems results in a decrease in receptor number (Creese and Sibley, 1981). However, it has been demonstrated that chronic nicotine treatment in mice (Bhat et al., 1991; Collins et al., 1989; Marks et al., 1983, 1985, 1986a, 1989a; Wonnacott, 1990) and rats (Floreset al., 1992; Ksir et al., 1987; Nordberget al., 1989; Schwartz and Kellar, 1983, 1985) will elicit a dose-dependent increase in brain [3H]nicotine and [125I]αBTX binding sites (Marks et al., 1983, 1986b). High-affinity nicotine binding is up-regulated at a lower dose than that required to elicit changes in the [125I]αBTX site; however, at an adequate dose, the number of [125I]αBTX sites increases more rapidly (Miner and Collins, 1989). The up-regulation of [3H]nicotine binding is not permanent, with binding levels returning to control values within 7 to 10 days in mouse (Markset al., 1985) and in 15 to 20 days in rat (Collins et al., 1990) after cessation of nicotine treatment. The levels of [125I]αBTX binding return to control levels within 2 to 4 days after nicotine treatment is discontinued in both rat and mouse (Collins et al., 1990; Miner and Collins, 1989).
A genetically determined variability in the number of brain nicotinic receptors has been reported in mice (Collins and Marks, 1989, 1991;Marks et al., 1989a). Analysis of nicotinic receptor binding in regionally dissected brain samples obtained from 19 different inbred mouse strains demonstrated that the number of both [3H]nicotine and [125I]αBTX binding sites varied among the various mouse strains (Marks et al., 1989a). The variation in receptor density was significantly correlated with differences among the strains in response to an acute challenge of nicotine (Marks et al., 1989b), and could reflect differences in the development of tolerance to chronic nicotine treatment.
A genetic component may also exist for the sensitivity and tolerance of humans to cigarette smoking (Collins, 1990a). Studies of monozygotic and dizygotic twins, discordant for smoking, suggest that there is a strong genetic component in humans for both initiation and persistence of smoking. The concordance rate for smoking was higher among monozygotic twins than it was among dizygotic twins (Carmelli et al., 1992). There also appears to be a genetic effect on smoking persistence, because monozygotic twins were more likely to be concordant for smoking cessation (Heath and Martin, 1993; Heathet al., 1993, 1995).
In contrast to what is known about the regulation of neuronal nicotinic receptor levels in rodents, regulation of the nicotinic receptor family is poorly understood in human brain. Although it has been reported that human tobacco users may show an increase in brain [3H]nicotine binding (Benwell et al., 1988), it is unknown whether receptor number increases with extent and persistence of smoking, or if binding returns to control levels when a person ceases to smoke. In the present study, the levels of [3H]nicotine binding were examined in human postmortem hippocampus and thalamus for changes in nicotinic receptor levels in relation to smoking history. To examine the regulatory effect and the dose dependency of tobacco use on nicotinic receptor numbers in humans, nicotine binding was examined in tissue samples from nonsmoking subjects, subjects who had variable life-long smoking histories and subjects who had quit smoking.
Materials and Methods
Human postmortem brain collection and storage.
Human brains were collected at autopsy. Hospital and autopsy records were reviewed to determine age, sex, race, cause of death, mental illness, cigarette, alcohol and drug use. Family members and physicians were also interviewed to detail the smoking history of the subject. After the brain was weighed and examined for gross pathology, it was divided sagittally and one hemisphere was preserved in formalin for neuropathological analysis. The other hemisphere was sliced coronally into 1-cm slices, and regions of interest were dissected in approximately 1-g blocks, frozen in dry ice snow and packaged for storage at −75οC (Leonard et al., 1993). The patient histories and tissue characteristics, including mental health status and alcohol abuse, smoking history, age of the patient, cause of death, postmortem interval and storage time are listed in table1. Subjects included in this study had no history of chronic psychotic disorders. Binding studies were performed in two separate experiments. The initial experiment included only hippocampal tissue (n = 9) on which Scatchard analyses were performed. The second experiment included both hippocampus (n = 32, total n = 41) and thalamic tissue (n = 32), on which single-point determinations of nicotine binding were performed. Of these subjects, 11 where nonsmokers, 21 were life-long smokers and 9 subjects were smokers who had quit at least 2 months before death.
Tissue preparation.
Dissected regions of human postmortem hippocampus and thalamus were weighed and homogenized in 10 volumes of ice-cold Krebs-Ringer HEPES buffer (118 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl2, 1.2 mM MgSO4, 20 mM HEPES, pH 7.5) in a Potter-Elvehjem homogenizer with a motor-driven Teflon pestle. Membranes were prepared by the method of Romano and Goldstein (1980), as described previously (Marks et al., 1983). Three additional centrifugation steps after resuspension and rehomogenization in 0.1× Krebs-Ringer HEPES were included to provide thorough washing of the membranes. After the final wash, pellets were resuspended in 0.1× Krebs-Ringer HEPES buffer (1 ml/g original wet weight), aliquoted, and assayed for protein content (BCA assay, Pierce, Rockford, IL). Membranes were stored at −75°C until analyzed for [3H]nicotine binding.
l-[3H]Nicotine binding.
[3H]Nicotine binding ([N-methyl-3H]nicotine, specific activity 60.0 Ci/mmol; Amersham Corp., Arlington Heights, IL; repurified by the method of Rommet al. (1990) to reduce nonspecific binding of the labeled ligand) was measured at 4°C by a modification of the method of Romano and Goldstein (1980), as previously described (Marks et al., 1986a). The assay was modified to allow filtration of samples in an Inotech filtration apparatus (Inotech Biosystems, Lansing, MI). Incubations were conducted in an incubation volume of 100 μl in 96-well polystyrene culture dishes. Samples containing 100 to 200 μg of protein were incubated at 4°C for 90 min in Krebs-Ringer HEPES containing 200 mM Tris. The binding reaction was terminated by filtration in an Inotech apparatus equipped with a 96-well head. Two glass fiber filters [top filter: GB100 (Micro Filtration Systems, Dublin, CA); bottom filter: Type A/E (Gelman Sciences, Ann Arbor, MI)], previously soaked in buffer containing 0.5% polyethylenimine, were used to trap protein. The use of two filters provided results that did not differ from those obtained using Millipore filter flasks with #30 glass fiber filters (Schleicher and Schuell, Keene, NH). Blanks were established by including 10 μM unlabeled nicotine in the incubation. Single-point assays were done at a [3H]nicotine concentration of 5 nM. Total binding was analyzed in triplicate and nonspecific binding in duplicate. In selected hippocampal samples, saturation curves were constructed by use of six concentrations of [3H]nicotine (1–29 nM) to determine whether smoking history affected the receptor affinity for nicotine. After filtration of the samples, the glass fiber filters were placed in polypropylene vials with 2.5 ml of scintillation fluid (Budget Solve, Research Products International, Mt. Prospect, IL), the filters mixed by shaking for 30–60 min and radioactivity determined (Packard 1600CA Liquid Scintillation Spectrometer; counting efficiency, 53%).
Data analysis.
Specific binding was calculated (fmoles/mg protein) and the data were analyzed by comparing the nicotine levels in adult nonsmokers, smokers up to the time of death and smokers who had quit at least 2 months before death. All data were statistically analyzed by ANOVA (Crunch statistical software, Oakland, CA) followed by ad hoc specific contrasts to identify the source of the variance (Scheff 233 test for multiple comparisons; Keppel, 1982).
Results
Postmortem brain tissue samples.
Patient histories and tissue characteristics, including mental health status, alcohol abuse, age of patient, cause of death, postmortem interval and storage time are listed in table 1, grouped by smoking history. The means and standard errors for each category are listed for each of the groups. Neuropathological analysis revealed that tissue samples used in this study were free of any neuropathological disorders. No statistical difference (all P > .15) was found among the three experimental groups compared for age of the subjects at death (average age, 55.8 ± 2.4 years; P = .43), postmortem interval (average PMI, 14.1 ± 1.1 hours; P = .24) or tissue storage (average time in storage, 859 ± 74 days; P = .23). In addition, no significant correlations were found between any of these parameters and [3H]nicotine binding (fig. 1; all P > .05). On average, smokers who had quit smoked fewer packs per day than those subjects with life-long smoking histories (smokers, 1.62 ± 0.16 packs/day; smokers who quit, 0.87 ± 0.15 packs/day, P < .009).
l-[3H]Nicotine binding.
Results and group means of single-point [3H]nicotine binding data ([3H]nicotine concentration of 5 nM) for both the hippocampus and thalamus are listed in table 1. Scatchard analyses were performed with a selected number of subjects in hippocampal tissue (n = 9), and results for three subjects with different smoking histories are shown in figure2. The inset graph shows data for sample SL010, a moderate smoker, which indicate that [3H]nicotine binding was saturable in human postmortem brain. Regression analysis of the data for these subjects fit a straight line, consistent with that expected for a single binding site that uses repurified [3H]nicotine (Romm et al., 1990). As shown in figure 2, there was an increase in the B max for [3H]nicotine binding with increasing degree of smoking (figure 2; SL007, heavy smoker, B max = 5.5 fmol/mg protein; SL010, moderate smoker, B max = 26.8 fmol/mg protein; SB154, nonsmoker, B max = 46 fmol/mg protein). Parallelism of the lines shows that the slopes are nearly identical, which indicates that smoking history did not correlate with any change in the affinity of nicotine for its receptor in hippocampus. The mean K d, which included members from nonsmokers, smokers and smokers who had quit, was 2.79 ± 0.19. (Individual data are listed in the legend of fig. 2; group means: nonsmoker, 2.4; smokers, 2.58 ± 0.08; smokers who quit, 3.27 ± 0.48). No difference was found in receptor affinity (K d) between these groups (F(1,6) = 3.46, P = .11), or in the correlation between receptor affinity and the number of packs smoked per day at the time of death (r = −0.38, P = .31). Although binding affinity was not specifically examined in thalamic tissue, previous results in mice (Marks et al., 1985, 1989a) and human postmortem brain (Benwell et al., 1988) have shown that the receptor affinity for nicotine was similar throughout the brain and was not changed as a result of nicotine treatment or smoking history.
A significant difference was found in mean [3H]nicotine binding between nonsmokers, life-long smokers and smokers who had quit in both hippocampus (F(2,38) = 16.58, P < .001) and thalamus (F(2,29) = 7.46, P < .003). As shown in the scatterplot in figure 3 (top), [3H]nicotine binding increased in the hippocampi of smokers compared with both nonsmokers (P < .0004) and smokers who had quit (P < .0001). Although not statistically different, smokers who had quit had a lower mean [3H]nicotine binding level than nonsmokers (P = .76). In thalamus, [3H]nicotine binding levels were approximately three times higher than that found in hippocampus. In thalamus, smokers also had a greater level of [3H]nicotine binding compared with nonsmokers (P < .012), and smokers who had quit (P < .018). Smokers who had quit also had a lower mean [3H]nicotine binding level in the thalamus than nonsmokers (P = .96). A high correlation was found between the [3H]nicotine binding in hippocampus and thalamus within all subjects (r = 0.743, P < .0001). When individual groups were examined, no correlation was found between nicotinic levels in the hippocampus and thalamus of nonsmokers (r = 0.04, P < .91) or smokers who had quit (r = 0.46, P < .35). However, there was a significant correlation in these measures in smokers (r = 0.72, P < .002), which suggests that the regulatory mechanisms that increase nicotinic receptor numbers in brains of subjects who smoke may be similar in both of these tissues.
The differences in [3H]nicotine binding were examined further with regard to the degree of smoking. In figure4, [3H]nicotine binding is correlated with two measures of smoking history. Packs per day was defined as the reported number of packs smoked per day at the time of death. This measure provided information on daily nicotine intake immediately before death. The other measure, pack years, was defined as the average number of packs per day while smoking, multiplied by the number of years smoked. Pack years provides a general measure of lifetime smoking, but not daily nicotine intake at the time of death. Because smokers who had quit had [3H]nicotine binding levels that were similar to those found in the control population, the correlational analysis was performed using only the nonsmokers and smokers at the time of death, to examine the direct effect of nicotine intake on [3H]nicotine binding. As shown in figure 4, receptor number showed an increase with both measures of smoking history; however, the significance of the effect depends on which measure of smoking history was used. The correlational analysis using pack years demonstrated a modest correlation only in the hippocampus (r = 0.43, P < .015), and not in thalamic tissue (r = 0.29, P = .16). The correlational analysis using packs per day found a positive correlation in both the hippocampus (r = 0.67, P < .0001) and thalamus (r = 0.44, P < .026) with the [3H]nicotine binding levels increasing with the number of packs smoked. This indicates there is a dose-dependent increase in [3H]nicotine binding levels in the brains of human smokers.
Discussion
Previous studies in rodents have shown that chronic nicotine treatment induces an increase in high-affinity nicotinic receptor binding, which persists for up to 20 days after treatment is suspended (Collins et al., 1990; Marks et al., 1985). It has also been shown that a similar increase in [3H]nicotine binding to high-affinity receptors occurs in human postmortem cortex, cerebellum and hippocampus of smokers, when compared to non-smokers (Benwell et al., 1988). The present study examined the effects of both nicotine intake and smoking cessation on the high-affinity nicotinic receptors in human postmortem hippocampus and thalamus. Nicotine binding was increased in the hippocampi of smokers, as well as in the thalamus, a brain region with high [3H]nicotine binding levels (Clarke et al., 1985; Rubboli et al., 1994). In the hippocampus, the changes in nicotinic receptor binding resulted from a change in total receptor numbers (B max) and not from a change in receptor affinity (K d) (Benwellet al., 1988; Marks et al., 1985, 1989a). There was no apparent effect of age on receptor numbers, nor a significant change in receptor numbers as a result of tissue handling parameters such as PMI or storage time, which suggests that nicotinic receptors are relatively stable under the conditions used in our assay system (Benwell and Balfour, 1985). Nicotinic receptor numbers were also found to be positively correlated with the level of nicotine intake, defined as the number of cigarettes smoked per day. Smokers who had quit smoking at least 2 months before death had levels of [3H]nicotine binding which were comparable to levels found in nonsmoking subjects.
The effect of the degree of smoking and daily nicotine intake on the level of [3H]nicotine binding was examined by comparing two measures of smoking with nicotinic receptor levels. Packs smoked per day was used as a measure to examine the influences of nicotine dose dependency, and pack years was used to examine the effect of the overall amount smoked during the subject’s lifetime. Although [3H]nicotine binding levels increased with smoking history in both brain regions, use of the number of packs smoked per day for the correlational analysis provided a more robust effect than did pack years. This would indicate that the increase in [3H]nicotine binding levels in humans was dose-dependent and most affected by the daily nicotine intake before death, rather than the overall amount smoked during a lifetime. In addition, it was found that only subjects that smoked had a significant correlation between the [3H]nicotine binding in hippocampus and thalamus. This confirms the dependence of nicotine intake on the observed increases in [3H]nicotine binding levels in smokers and suggests that the regulatory mechanisms that increase nicotinic receptor numbers in smokers may be similar in both of these tissues.
Although several studies have examined the effect of chronic and acute nicotine exposure in rodents, the exact nature of the increase in nicotine binding has yet to be fully understood. The α4/β2 containing receptors have been reported to account for >90% of the nicotinic receptors in rat brain (Lindstrom et al., 1990;Whiting and Lindstrom, 1987), and western blotting has shown that the increase in nicotinic receptors in rats was caused by an increased expression of the α4/β2 containing receptors, although other receptor subtypes were not examined (Flores et al., 1992). It has been shown in various strains of mice that nicotine-induced increases in nicotinic receptor numbers do not increase to the same degree in all brain regions, or even within a brain region (Collinset al., 1989; Marks et al., 1992). For example, mouse thalamus was one brain area that showed regional variations in the nicotine-induced increases in [3H]nicotine binding, whereas hippocampal [3H]nicotine binding was consistently increased throughout the structure (Collins et al., 1989;Marks et al., 1992). Heterogeneous up-regulation of nicotinic receptor expression within human thalamus may account for the reduced correlation of smoking with [3H]nicotine binding, when compared with that in human hippocampus. A detailed autoradiographic analysis of [3H]nicotine binding will be required to examine regional variations in human brain.
The increase in nicotinic receptor numbers in rodents is not caused by an increase in mRNA levels (Marks et al., 1992). Whether this is true in human brain remains to be determined. However, the lack of an effect on nicotinic receptor transcription in mice suggests that nicotine-induced increases in nicotinic receptor levels could result from changes in post-translational processing, or as shown in tissue culture, a decrease in receptor turnover (Peng et al., 1994). It has been hypothesized that the increase in nicotinic receptor number and the decreased rate of receptor turnover may be related to nicotinic receptor channel desensitization, which appears to reflect the conformational state of the receptor channel (Marks et al., 1983; Peng et al., 1994; Schwartz and Keller, 1985). Once the nicotinic receptor channels are desensitized and rendered inactive, additional receptors would be recruited to maintain the nicotinic response of the neuron, which results in an overall increase in nicotine binding (Bencherif et al., 1995).
It is interesting to note that smokers who had quit at least 2 months before death had nicotinic receptor binding levels that were similar to those found in nonsmokers. This suggests that nicotine-induced up-regulation of receptor numbers is a temporary effect, similar to that found in rodents (Collins et al., 1990; Marks et al., 1985). It is also possible that these subjects may have had a lower basal nicotinic receptor level before smoking initiation. Such individuals might be less sensitive to the rewarding effects of nicotine and would, therefore, be less likely to persist in a smoking behavior than a person with higher nicotinic receptor levels (Eysenck, 1983; Collins, 1990a). Conversely, it could be speculated that humans, who continue to smoke after initial experimentation with cigarette use, might have had either a higher basal nicotinic receptor level or an increase in the rate of the observed nicotinic receptor up-regulation. In either case, these subjects would be more sensitive to the rewarding effects of nicotine, and more likely to persist in smoking behaviors (Collins, 1990b). This hypothesis suggests an underlying heterogeneity in expression of the nicotinic acetylcholine receptor family in human brain, which is supported by the observed variability in [3H]nicotine binding levels within the various groups examined in these studies (see fig. 3). Regional variations in basal nicotinic receptor levels in mouse brain have also been reported among different mouse strains (Marks et al., 1989a).
Several studies in laboratory animals suggest that genetic factors may be involved in regulating both neuronal nicotinic receptor expression and behavioral and physiological response to nicotine (Collins et al., 1989). Epidemiological studies of smoking history in humans also supports the involvement of a genetic component in the propensity for nicotine sensitivity, use and withdrawal (Carmelli et al., 1992; Collins, 1990a; Heath and Martin, 1993; Health et al., 1993). Continued use of tobacco products in humans results in nicotine tolerance (Collins, 1990b). The dose-dependent increase in nicotinic receptor numbers in human subjects that persists in smokers may, at least in part, influence the development of nicotine tolerance and addiction. Studies are currently underway to examine the effect of smoking history on the mRNA levels for specific nicotinic receptor subunits to investigate the mechanism involved in nicotinic receptor up-regulation in humans.
Footnotes
-
Send reprint requests to: Dr. Sherry Leonard, Department of Psychiatry, Box C268–71, University of Colorado Health Sciences Center, 4200 E. 9th Avenue, Denver, CO 80262.
-
↵1 This work was supported by USPHS Grants, DA09457 and VA Medical Research Service to S.L., DA03194 and DA00197 to A.C.C. and AA11164 to C.R.B.
- Abbreviations:
- αBTX
- α-bungarotoxin
- HEPES
- N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid
- PMI
- postmortem interval
- Received October 15, 1996.
- Accepted March 6, 1997.
- The American Society for Pharmacology and Experimental Therapeutics