Abstract
Subtypes of neuronal nicotinic acetylcholine receptors (nAChRs) are differentially sensitive to up-regulation by chronic nicotine exposure in vitro. To determine whether this occurs in animals, rats were implanted with minipumps containing saline ± nicotine (6.0 mg/kg/rat/day) for 14 days. Autoradiography with [125I]epibatidine using 3-(2(S)-azetidinylmethoxy)pyridine dihydrochloride (A-85380) or cytisine as selective competitors allowed quantitative measurement in 33 regions of 3 families of nAChR binding, with properties of α4β2, α3β4, and α3/α6β2. Chronic nicotine exposure caused increases of 20 to 100% for α4β2-like binding in most regions surveyed. However, binding to this subtype was not increased in some regions, including habenulopeduncular structures, certain thalamic nuclei, and several brainstem regions. In 9 of 33 regions, including catecholaminergic areas and visual structures, α3/α6β2-like binding represented >10% of total binding. Binding to this subtype was up-regulated by nicotine in only two of these nine regions: the nucleus accumbens and superior colliculus. α3β4-Like binding represented >10% of total in 15 of the 33 regions surveyed. Binding to this subtype was increased by nicotine in only 1 of these 15 regions, and actually decreased in subiculum and cerebellum. These studies yielded two principal findings. First, chronic nicotine exposure selectively up-regulates α4β2-like binding, with relatively little effect on α3/α6β2-like and α3β4-like binding in vivo. Second, up-regulation by chronic nicotine exposure shows considerable regional variation. Differential subtype sensitivity to chronic nicotine exposure may contribute to altered pharmacological response in individuals who smoke or use nicotine replacement therapy.
The actions of nicotine in brain are mediated by neuronal nicotinic acetylcholine receptors (nAChRs). These receptors are ligand-gated ion channels comprising combinations of α and β subunit proteins in a pentameric structure. Expression of at least seven α and three β subunits has been detected in mammalian brain. Despite the large number of subunits, pharmacological studies have so far found evidence for only a limited number of actual subtypes. Initially, two classes were identified by receptor binding methods: one selectively labeled by [125I]α-bungarotoxin, thought to consist largely of α7 homomers, and another class with higher affinity for acetylcholine, nicotine, and cytisine. The introduction of radiolabeled epibatidine revealed the presence of additional subclasses of nAChR not labeled by these other ligands. It is now recognized that there are several different subtypes of these so-called high-affinity nAChRs, based on selective ligands and molecular studies. The major form is believed to be α4β2 (Flores et al., 1992); multiple forms may exist based on variable stoichiometry or the inclusion of additional subunits such as α5, although these variations have only minimal effects on ligand binding properties. A second form of high-affinity nAChR closely resembles ganglionic nicotinic receptors, believed to be α3β4. This subtype may also exist in several variations, with additional subunits or variable stoichiometry. A third subclass of high-affinity nAChR consists of a heterogeneous mix of subunits: α6 (and/or α3) paired with β2 subunits, with the possible addition of β3 or α4 subunits in some cases (Le Novere et al., 1996; Lena et al., 1999; Quik et al., 2000; Klink et al., 2001; Champtiaux et al., 2002; Whiteaker et al., 2002; Marubio et al., 2003).
The frog skin alkaloid (±)-epibatidine is a very high-affinity agonist at most nAChRs; it has much lower affinity at α7 and muscle-type nAChRs (Gerzanich et al., 1995). We have used both [3H]epibatidine and [125I]epibatidine to demonstrate distribution of high-affinity nAChR in rat brain with autoradiography (Perry and Kellar, 1995; Dávila-García et al., 1997). More recently, we have devised an autoradiographic method using [125I]epibatidine in the presence of the competing ligands A-85380 or cytisine to selectively measure subtypes with properties of α4β2, α3β4, and α3/α6β2 (Perry et al., 2002). Knowledge of the affinities of each ligand at different combinations of subunits expressed in mammalian cell lines allowed quantification of the populations of each of these three classes in different brain regions. As expected, α4β2-like binding predominated in most forebrain regions; α3/α6β2 receptors were not the predominant subtype in any region, although substantial levels were seen in many optic-related structures, most catecholaminergic regions and several other regions as well. The α3β4 subtype had a surprisingly widespread distribution: this subtype was high in subiculum and several brainstem nuclei, and was the predominant subtype detected in the habenulopeduncular system, area postrema, cerebellum, and the pineal gland (Perry et al., 2002).
Chronic exposure to nicotine is well known to increase nicotinic receptor binding in animal models (Marks et al., 1983; Schwartz and Kellar, 1983; Sanderson et al., 1993). Up-regulation is also seen in human smokers (Benwell et al., 1988; Breese et al., 1997; Perry et al., 1999). Up-regulation does not appear to be the result of altered transcription (Marks et al., 1992); other mechanisms, including a decrease in protein turnover or altered subunit assembly, are believed to be responsible (Peng et al., 1994; Wang et al., 1998; Harkness and Millar, 2002). Evidence from in vitro studies strongly indicates that nAChR subtypes are differentially sensitive to up-regulation by nicotine exposure. For example, α4β2 receptors are readily up-regulated following chronic nicotine exposure, and return to control levels slowly (Peng et al., 1994; Fenster et al., 1999; Harkness and Millar, 2002). However, α3-containing receptors are usually less readily up-regulated, and appear to recover more rapidly from whatever up-regulation does occur (Wang et al., 1998; Meyer et al., 2001). In vivo studies using immunoprecipitation have demonstrated up-regulation of α4β2 nAChR in rat forebrain (Flores et al., 1992). In retina and a variety of peripheral tissues, α3-containing receptors do not show significant up-regulation after chronic nicotine exposure (Flores et al., 1997; Dávila-García et al., 2003). However, the response of α3- and/or α6-containing receptors to chronic nicotine in mammalian brain has not been reported, in part due to technical limitations of existing methods, and also the fact that expression of these subtypes in brain is largely limited to relatively small regions.
Our recently developed autoradiographic method can overcome these limitations by combining quantitative measurement of three separate nAChR subtype classes with the anatomical precision of autoradiography. We have used this method to measure the responses of three high-affinity subtypes of nAChR to chronic nicotine infusion in the rat brain. These results are of potential value in understanding the altered neurobiological responses that accompany chronic exposure to nicotine, and the mechanisms of nicotine tolerance and addiction.
Materials and Methods
Osmotic minipumps (Alzet model 2002; Durect Corporation, Cupertino, CA) were filled with sterile saline or with nicotine hydrogen tartrate (Sigma-Aldrich, St. Louis, MO) in saline, at concentrations in the range of 450 to 500 mg/ml. These pumps deliver solution at a rate of 12 μl/day for a duration of 14 to 17 days; the nicotine concentrations were calculated to achieve a dose of 6 mg/kg/day (free base equivalent). This dosing regimen has been shown to achieve blood nicotine levels in rats (Trauth et al., 2000) approximating those seen in typical human smokers (Rose et al., 1999). Male Sprague-Dawley rats (225–275 g) were anesthetized lightly with ketamine (44 mg/kg i.p.) plus rompun (5 mg/kg i.p.), and the minipumps inserted into a subcutaneous pocket via a small incision made over the shoulders. The wound was closed with clips and the area swabbed with antiseptic; after recovery from anesthetic, animals were returned to individual cages. At 14 days after minipump implantation, animals were sacrificed by decapitation without anesthesia and brains rapidly removed and frozen on dry ice. Frozen brain sections (20 μm) were thaw-mounted onto microscope slides (SuperFrost Plus, Fisher Scientific Co., Pittsburgh, PA) and stored at –80°C until use. Animal use and procedures were approved by the George Washington University Medical Center Institutional Animal Care and Use Committee.
[125I]Epibatidine (725 Ci/mmol) was a generous gift from Dr. John Musachio (Musachio et al., 1997). A-85380 and cytisine were generous gifts from Dr. Kenneth J. Kellar. Autoradiography was done using an adaptation of published methods (Perry et al., 2002). Brain sections were preincubated 10 min in Tris-HCl buffer (pH 7.4) containing 120 NaCl, 5 mM KCl, 2.5 mM CaCl2, and 1 mM MgCl2, then incubated with [125I]epibatidine (0.165 nM) for 120 min at room temperature. Incubations of adjacent sections included unlabeled A-85380 (1 nM) or cytisine (50 nM), or 300 μM(–)-nicotine hydrogen tartrate for nonspecific binding. After incubation, sections were rinsed twice for 10 min each in ice-cold buffer, then dipped briefly in distilled water. After air drying, the sections were apposed to Kodak BioMax MR-1 film (Amersham Biosciences Inc., Piscataway, NJ) along with [125I]standards (Amersham Biosciences Inc.) for 4 days, followed by a re-exposure of some sections for 6 h. Quantitative densitometric analysis of binding was done using the Loats Inquiry digital densitometry system (Loats Associates, Westminster, MD). Binding was expressed as femtomoles per milligram by comparison to standard curves obtained by the images from the [125I] standards. Regions were identified by comparison with the rat brain atlas of Paxinos and Watson (1998). Under these conditions, nonspecific binding in adjacent sections was not distinguishable from film background and therefore total binding was identical to specific binding. Because the working exposure range of the autoradiographic film is smaller than the range of binding densities of high-affinity neuronal nicotinic receptors, regions with exceptionally high receptor densities can easily be “overexposed.” Therefore, quantitation for several such regions (including medial habenula, fasciculus retroflexus, interpeduncular nucleus, and superior colliculus) was done using the 6-h exposure; all others were determined from the 4-day exposure. Note that although a large number (33) of regions were assessed, this was not an exhaustive survey of binding in all brain regions.
Determination of subtype-selective binding was done according to a previously published method (Perry et al., 2002). Briefly, this method assumes that total [125I]epibatidine binding consists of binding to a combination of up to three separate classes of nAChR receptor subtypes. Based on analysis of binding patterns of epibatidine, A-85380, and cytisine to different combinations of subunits expressed in vitro (Perry et al., 2002), it is possible to divide [125I]epibatidine binding into α4β2-like, α3/α6β2-like, and α3β4-like binding using the differential affinities of the three ligands to receptors in heterologous expression systems. While [125I]epibatidine binds with high affinity to all three subtypes (Kd values: α4β2, 0.17 nM; α3β2, 0.10 nM; α3β4, 1.1 nM), A-85380 has high affinity for only the α4β2 and α3/α6β2 subtypes (Ki values: α4β2, 0.11 nM; α3β2, 0.16 nM; α3β4, 59 nM) and cytisine is relatively selective for the α4β2 subtype (Ki values: α4β2, 1.7 nM; α3β2, 47 nM; α3β4, 240 nM). α2 nAChRs were not considered in this analysis due to the very limited expression of this subunit (Wada et al., 1989).
For every region, we made three separate measurements of [125I]epibatidine binding in adjacent tissue sections: X1 = binding with no competitors added; X2 = binding in the presence of 1 nM A-85380; X3 = binding in the presence of 50 nM cytisine. These concentrations were chosen to yield a partial, but not complete, blockade of α4β2 and α3/α6β2 receptors (by A-85380) and α3β2 and α3/α6β2 receptors (by cytisine). Our previous publication describing this method (Perry et al., 2002) used higher concentrations of competitors, which provided striking visual images but in some cases left only a small amount of [125I]epibatidine bound, decreasing measurement precision. The lower concentrations used here do not completely inhibit binding, allowing more accurate measurements and not affecting the mathematical calculations. Every measurement of [125I]epibatidine binding consists of some combination of binding to as many as three separate nicotinic receptor subtypes, A, B, and C: these are the unknowns. One can calculate the amount of [125I]epibatidine bound to any receptor (i.e., receptor A) by multiplying the fractional occupancy of [125I]epibatidine at that receptor (i.e., fA) times the number (population) of receptors: fA × A. The fractional occupancy of the three receptors of interest by a known concentration of [125I]epibatidine can be readily calculated using affinities of the three ligands for each subtype determined from cell binding experiments. These terms can all be related in the following three equations, which state simply that for any region, the radioligand binding (X) measured under any condition (1, 2, 3) is a sum of the occupancy of the three receptor subtypes (A, B, C) present in that region:
These three equations with three unknowns (A–C), when solved, allowed us to calculate the population of up to three subtypes in any given region.
Although the theoretical basis for this method of subtype calculations is sound, there are several limitations. This method identifies different classes of nAChR based on ligand binding. As discussed above, each of these “subtypes” may include multiple subunit combinations. Therefore we use the terminology “α4β2-like binding” to acknowledge this heterogeneity. Also, the method is based on the assumption that all [125I]epibatidine binding can be accounted for by these three subtypes. Although this is a reasonable assumption for most regions, it may not hold for every area: for example, while α2 subunits are not commonly found in the mammalian central nervous system, they are expressed in the interpeduncular nucleus (Wada et al., 1989). Cell binding data for the α6β2 subtype is not available, so identification of this subclass is based on similarities to the α3β2 subtype. The calculations depend on data obtained using rat subunits expressed in human embryonic kidney 293 cells; differences in post-translational modification between this expression system and rat central nervous system could alter binding characteristics. Finally, measurements in regions with low overall expression, or low expression of a subtype, are likely to be less reliable; we have restricted our analysis of subtypes to regions in which that subtype composed ≥10% of the total binding.
Means for total binding and for binding to the three subtype classes were determined for saline- and nicotine-treated animals (nine animals per group) and compared by t test without correction for multiple comparisons. Statistical significance is expressed as *p < 0.05; **p < 0.01; ***p < 0.001.
Results
Total binding of [125I]epibatidine to three different pairs of comparable coronal sections from saline- and nicotine-treated rats is shown in Fig. 1. The sections from the nicotine-treated animals are noticeably darker in most regions than the corresponding saline-treated animals, indicative of widespread increases in total [125I]epibatidine binding. Binding was quantified by densitometry in multiple regions across the nine animals from each treatment group. As noted under Materials and Methods, quantification in regions with especially dense binding (i.e., interpeduncular nucleus, fasciculus retroflexus) was done using separate films exposed for shorter times, to rule out film saturation artifacts. Results in Table 1 show a 73-fold range in binding densities among the regions assessed. In the great majority of these regions (23 of 33), chronic exposure to nicotine significantly increased binding, in some cases to levels almost three times those seen in saline-treated animals. The greatest increases in total binding were detected in forebrain regions, including cerebral cortex, amygdala, dentate gyrus, and nucleus accumbens. Increased total [125I]epibatidine binding was not seen in the habenulopeduncular regions, subiculum, posterior thalamic nuclear group, dorsal lateral geniculate nuclei, subiculum, nucleus of the solitary tract, area postrema, or pineal gland.
To determine subtype-selective binding, adjacent sections were exposed to [125I]epibatidine either alone or in the presence of the unlabeled competitors A-85380 (1 nM, which competes against β2-containing receptors) or cytisine (50 nM, which competes against α4β2, and to a lesser extent α3/α6β2, receptors). An example of this competition is shown in Fig. 2. [125I]Epibatidine binding was reduced by these competitors in most regions; however, binding in some select regions was relatively unaffected. For example, binding to the fasciculus retroflexus, an area highly enriched in α3β4-like nAChR, is largely unaffected by either competitor. Binding to the superficial gray layer of the superior colliculus, a region enriched in α3/α6β2-like receptors, is only partially blocked by cytisine. We have recently developed a strategy to use results from such differential binding competition to quantify the relative proportions of three different classes of nAChR binding: α4β2-like, α3β4-like, and α3/α6β2-like (Perry et al., 2002). Table 2 shows the percentage of each of these three classes of binding that was calculated to be present in different regions from saline-treated animals. The results confirm the dominance of the α4β2-like nAChR binding in most forebrain regions. Binding to α6/α3β2-like receptors composed >10% of the total in 9 of the 33 regions surveyed, including caudate putamen, two geniculate nuclei, several optic structures, nucleus of the solitary tract, and spinal trigeminal nucleus. As previously reported (Perry et al., 2002), α3β4-like binding was the dominant form detected in the habenulopeduncular structures, and also in cerebellum, the nucleus of the solitary tract, dorsal motor nucleus of the vagus, spinal trigeminal nucleus, area postrema, and pineal gland. Substantial amounts of this subtype were also detected in subiculum and ventral tegmental areas.
We then determined the effect of chronic nicotine exposure on the binding to the three nAChR subtypes. The results for α4β2-like binding are shown in Table 3. Overall, binding was significantly increased in 24 of 32 regions surveyed. The regional pattern of binding increases was very similar to that of total binding, which is not surprising given that this subtype represents the majority in most regions. Two regions (subiculum and cerebellum) showed increased α4β2-like binding despite showing no significant increase in total binding. Note that each of these regions included a substantial portion of α3β4-like binding. In one region (dorsal motor nucleus of the vagus), total binding increased, without a statistically significant increase in α4β2 binding.
We calculated the effect of chronic nicotine on α6/α3β2-like binding in those regions where this subtype represented >10% of the total, plus the medial habenula and interpeduncular nucleus, each of which has very high levels of total binding; results are shown in Table 4. (Because total binding was very low in the amygdala, this region was not included in this analysis.) Of these 10 regions, α6/α3β2-like binding was significantly increased in only two: nucleus accumbens and superior colliculus. The effect of chronic nicotine on α3β4-like binding is shown in Table 5. Only one region examined showed a significant increase in this subtype, the dorsal motor nucleus of the vagus. Surprisingly, decreases in α3β4-like binding were seen in several regions, reaching statistical significance in the cerebellum and subiculum.
Discussion
Up-regulation of α4β2 nAChR following chronic exposure to nicotine in rodents was first observed 20 years ago (Marks et al., 1983; Schwartz and Kellar, 1983). α3β4-Like binding in adrenal gland (Flores et al., 1997; Dávila-García et al., 2003), trigeminal ganglia (Ulrich et al., 1997), superior cervical ganglia, and pineal gland (Dávila-García et al., 2003) is not up-regulated by chronic nicotine infusion. However, there has been no direct demonstration of the response to chronic nicotine of high-affinity nAChRs in brain other than α4β2. The current results demonstrate that α3β4-like and α6/α3-like binding in rat brain is highly resistant to up-regulation compared with α4β2-like binding.
Several factors may contribute to differences in sensitivity to up-regulation between subtypes of nAChR. The rank order of nicotine affinity in vitro is α4β2 > α3β2 > α3β4 (Parker et al., 1998), similar to the observed sensitivity to up-regulation. This raises the possibility that brain levels of nicotine may not have reached levels high enough to occupy the lower sensitivity subtypes. By linear extrapolation from published data for brain nicotine levels in rat following minipump infusion (Ghosheh et al., 2001), we estimate that our treatment (6 mg/kg/day for 14 days) achieved brain levels of nicotine in the range of 1 to 2 μM. This concentration is severalfold higher than published values for the α3β4 affinity (Parker et al., 1998), suggesting that this dose was sufficient to achieve continuous occupation of even the lowest sensitivity subtype.
Substitution of β4 for β2 can have profound effects on pharmacology, channel properties, time course of desensitization, and sensitivity to up-regulation in vitro (Parker et al., 1998; Wang et al., 1998). Experiments using heterologous expression systems have found that α3β4 receptors are resistant to up-regulation by chronic nicotine compared with α3β2 receptors (Wang et al., 1998). Others, however, have demonstrated clear up-regulation of α3β4 receptors in vitro (Meyer et al., 2001), although this subtype recovered from the up-regulation very rapidly (Wang et al., 1998; Meyer et al., 2001). Agonist efficacy can also differ among subtypes. Nicotine is a full agonist at α4β2 receptors but only a partial agonist at α3β2 receptors (Olale et al., 1997; Chavez-Noriega et al., 2000), which could contribute to a differential response between these two subtypes. However, nicotine is a full agonist at α3β4 receptors, the most resistant subtype to up-regulation.
Up-regulation of nAChR is not thought to be caused by increased mRNA expression. Possible mechanisms include a decreased rate of subunit degradation (Marks et al., 1992; Peng et al., 1994; Ke et al., 1998), alone or in concert with other mechanisms, including post-translational modifications such as phosphorylation, enhanced subunit assembly, and altered subcellular distribution (Ke et al., 1998; Wang et al., 1998). Under normal circumstances, most subunit proteins existed in an unassembled form within the endoplasmic reticulum: chronic nicotine enhances both assembly of subunits, and insertion into the plasma membrane (Harkness and Millar, 2002). Differences in subtype susceptibility to these various mechanisms may contribute to differences in subtype sensitivity to up-regulation. In a possibly related finding, chronic nicotine has been shown to alter stoichiometry within a subtype, causing an apparent shift from (α4)3(β2)2 to (α4)2(β2)3 in human embryonic kidney cells (Nelson et al., 2003).
Besides differences in sensitivity of nAChR subtypes to up-regulation, this study highlights an interesting regional variability in up-regulation within a subtype. For instance, α3/α6β2 binding was unaffected in most regions, but showed significant up-regulation in two regions. The increase of 46% in the superior colliculus, a structure innervated by the optic tract, contrasts with the large (although nonsignificant) decrease in α6/α3β2-like binding in the tract itself. A similar regimen of chronic nicotine did not alter [125I]epibatidine binding in rat retina (Dávila-García et al., 2003).
The precise subunit identity of α3/α6β2-like binding is uncertain. Initial studies identified this class as α3β2, based on inhibition by α-conotoxin MII (Cartier et al., 1996; Whiteaker et al., 2000). Evidence now indicates that in most brain regions, the nAChR identified by this toxin contain α6, perhaps along with α3 subunits in some regions (Quik et al., 2001; Champtiaux et al., 2002; Marubio et al., 2003). Most are likely to be paired with β2 subunits, and in some cases also β3 or α4 subunits (Lena et al., 1999; Klink et al., 2001; Champtiaux et al., 2002; Whiteaker et al., 2002; Marubio et al., 2003). In analogy to mouse nAChRs, a likely pattern is that rat α6-containing receptors are found in catecholaminergic regions (i.e., caudate putamen, nucleus accumbens), and probably also in visual structures (optic tract and nuclei and superior colliculus) (Le Novere et al., 1996; Lena et al., 1999; Klink et al., 2001; Whiteaker et al., 2002), whereas the labeling in habenulopeduncular structures may be due at least in part to α3β2 receptors (Whiteaker et al., 2002). Thus one possible explanation for the regional differences in sensitivity to up-regulation is α6 versus α3 heterogeneity. No increases of α3/α6β2 binding were detected in the habenulopeduncular structures, suggesting that the α3β2 subtype in brain may not be up-regulated. Not all of the differences in sensitivity can be explained by this, because in several other regions also presumed to contain α6 receptors (i.e., caudate putamen, geniculate nuclei), no evidence for up-regulation was seen. These differences could also be caused by further subunit differences, such as additional subunit heterogeneity. We are not aware of previous reports for regulation of α6* receptors by chronic nicotine in vivo.
Significant regional heterogeneity of response was also seen for α4β2, with binding in nicotine-treated animals ranging from 57% to 258% of control animals. Homogenate binding studies have consistently reported regional differences in response to nicotine treatment (Sanderson et al., 1993). In general, the highest increases have been reported in cortex and hippocampus, with smaller or no increases detected in thalamus, midbrain, and striatum (Marks et al., 1992; Sanderson et al., 1993; Flores et al., 1997; Ulrich et al., 1997). The increased anatomical precision compared with homogenate binding reveals a more complex picture of binding. The thalamus, for example, showed great heterogeneity: total and α4β2-like binding increased in the laterodorsal and ventro-medial thalamic nuclei but not in the posterior nuclear group; and in the medial and ventral lateral geniculate nuclei but not the dorsal lateral geniculate. Therefore, reports of up-regulation or no up-regulation in thalamus can be misleading, given the large heterogeneity of responses in this area. Our results in rat brain are largely compatible with autoradiographic analysis of [3H]nicotine binding in mouse brain after chronic infusion of nicotine (Pauly et al., 1991, 1996): thalamic nuclei showed a great variability of sensitivity to up-regulation, while cerebral cortical regions and most forebrain structures were uniformly sensitive (Pauly et al., 1991). Some differences between species were also seen. For example, we found high levels of binding but no up-regulation in medial habenula, whereas in mouse this region had only modest levels of binding, but was readily up-regulated (Pauly et al., 1991). Besides the species difference, the latter study used [3H]nicotine, which labels only a subset of nAChR labeled by [125I]epibatidine.
One possible explanation for different regional up-regulation responses would be if regions were exposed to different nicotine concentration. Nicotine readily penetrates into brain, and with continued infusion brain levels continue to increase, achieving higher concentrations than in blood (Ghosheh et al., 2001), arguing against limited access of nicotine. Furthermore, it is difficult to see how differences in access could explain different responses in adjacent regions such as medial geniculate α4β2 binding in nicotine animals (161% of control) and dorsal lateral geniculate (97% of control); α4β2 receptors can also differ in their stoichiometry, the inclusion of additional subtypes such as α5, or their localization on neurons (i.e., somatic versus perisynaptic versus synaptic). Receptors in these different neuronal sites may be programmed for response to different types of signaling, including synaptic release (large but rapidly changing concentrations of acetylcholine), or “volume transmission” (smaller and gradually changing concentrations of acetylcholine). Such differences might be reflected in different patterns of response to continuous nonphysiological stimulation by exogenous agonists.
These results demonstrate that chronic exposure to nicotine has two important effects on high-affinity nAChRs. First, it selectively up-regulates α4β2-like receptors, with little effect on α3/α6β2 and α3β4-like subtypes. Thus, these detailed anatomical studies confirm the earlier results in rat cerebral cortex (Flores et al., 1997). Second, the α4β2 subtype exhibits dramatically different sensitivity to up-regulation across different brain regions. Additionally, this is the first report that exposure to chronic nicotine can decrease binding to nAChR, in this case to the α3β4 subtype.
Millions of people take nicotine daily in the form of tobacco products, and an increasingly large number also take nicotine chronically in the form of subcutaneous patches, nicotine gum, or nicotine inhalers. The present results suggest that the pharmacological response of these individuals to nicotine is likely to be significantly altered as a result of a shift in nAChR number across subtypes and across brain regions.
Acknowledgments
We thank Nikhil Mull for technical assistance and Dr. Ken Kellar for valuable discussions.
Footnotes
-
DOI: 10.1124/jpet.103.056408.
-
ABBREVIATIONS: nAChR, neuronal nicotinic acetylcholine receptors; A-85380, 3-(2(S)-azetidinylmethoxy)pyridine dihydrochloride.
- Received July 2, 2003.
- Accepted September 9, 2003.
- The American Society for Pharmacology and Experimental Therapeutics