Abstract
In the present study we used the nicotinic ligand 5-iodo-A-85380 [5-iodo-3(2(S)-azetidinylmethoxy)pyridine], which selectively binds to β2-containing nicotinic acetylcholine receptors, to elucidate the nicotinic receptor subtypes affected by nigrostriatal damage in the monkey. Autoradiographic studies in control monkeys showed that 5-[125I]A-85380 ([125I]A-85380) binds throughout the brain with the characteristics of a nicotinic receptor ligand. Competition experiments with cytisine and nicotine yielded Ki values of ∼1 and 10 nM, respectively, with complete inhibition of [125I]A-85380 binding at a 10−6 M concentration of these ligands. In contrast, α-conotoxin MII blocked radioligand binding in the striatum by 30% at the highest concentrations, suggesting that a subset of striatal [125I]A-85380 sites are α-conotoxin MII-sensitive. Monkeys treated with the nigrostriatal neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine showed a selective decrease in striatal [125I]A-85380 sites, with a 42% reduction in the caudate and putamen of animals with moderate nigrostriatal lesioning and a 53% decline in the striatum of severely lesioned animals. Our previous work had demonstrated that there were two populations of nicotinic receptors eliminated after nigrostriatal damage, an α-conotoxin MII-sensitive and an α-conotoxin MII- resistant subtype. Analysis of both striatal [125I]A-85380 and [125I]epibatidine competition studies extend our earlier studies by demonstrating that the α-conotoxin MII-sensitive sites eliminated after moderate nigrostriatal lesioning appear to be composed of two nicotinic receptor subtypes. The data may be important for potential therapeutic approaches because they suggest that there are at least three populations of nicotinic receptors in monkey striatum, of which two are selectively vulnerable to nigrostriatal damage, while the third is more resistant.
Extensive evidence now indicates that there are significant declines in central nervous system nicotinic acetylcholine receptors (nAChRs) in Parkinson's disease, a movement disorder characterized by a loss of nigrostriatal dopaminergic neurons (Gotti et al., 1997; Lang and Lozano, 1998; Court et al., 2000; Ball, 2001). Because nAChR deficits may result in decreased activity of neuronal pathways normally involved in the regulation of motor control, there is the potential that receptor agonists may provide a useful therapeutic approach to treat the symptoms of Parkinson's disease (Kelton et al., 2000). In addition, nicotinic agonists may have potential as neuroprotective agents against nigrostriatal degeneration. Epidemiological studies show a very consistent inverse correlation between tobacco use and Parkinson's disease (Morens et al., 1995; Quik and Jeyarasasingam, 2000). Although the active agent in tobacco products that mediates this effect remains to be determined, the finding that nicotine stimulates striatal dopamine release (MacDermott et al., 1999) and is neuroprotective in culture and in animal models (Quik and Jeyarasasingam, 2000) suggests it may be responsible for this apparent neuroprotection.
Nicotine exerts its effects on biological systems by interacting with ligand-gated channels composed of five subunits (Wonnacott, 1997;Changeux et al., 1998; Jones et al., 1999; Lukas et al., 1999). Nine nAChR subunits have been isolated from mammalian brain, including the α subunits (α2–α7), which contain the cysteines necessary for acetylcholine binding, and the β subunits (β2–β4), which are structural subunits contributing to the ligand-binding affinity of the receptor. Although receptors containing the α7 subunit appear to form homopentamers in the brain, the other α subunits form receptors in combination with other α subunits or with the β subunits (Luetje and Patrick, 1991; Parker et al., 1998). Since these latter receptors have two ligand-binding sites at α/β subunit interfaces (Sargent, 1993), there is the potential for numerous distinct receptor subtypes. Although knowledge of the subunit combinations present in mammalian brain is critical for a clear understanding of their role in function, their identification has been difficult because of the limited number of nAChR ligands with selectivity for specific α/β and/or α/α subunit interfaces.
One ligand that has recently proved useful in this regard is α-conotoxin MII, a cone snail toxin that appears to interact with α3* and α6* nicotinic receptors (Grady et al., 2001; Champtiaux et al., 2002; Whiteaker et al., 2002). Initial studies had suggested that α-conotoxin MII identified receptors with an α3β2 interface (Cartier et al., 1996; Kulak et al., 1997; McIntosh et al., 1999). However, more recent work now indicates that the receptors labeled by this toxin contain the α6 rather than α3 subunit. [125I]α-Conotoxin MII-sensitive receptors are completely eliminated in the brains of α6 (−/−) mice, suggesting that all the toxin sites contain this subunit. However, the presence of some α3-dependent [125I]α-conotoxin MII binding in the interpeduncular nucleus in α3 (−/−) mice would suggest the existence of a population of sites containing both the α3 and the α6 subunit (Champtiaux et al., 2002; Whiteaker et al., 2002). Interestingly, in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated monkey, which represents one of the best models for Parkinson's disease, there appears to be a selective vulnerability of α6* nicotinic receptors to nigrostriatal damage; this nAChR subtype is the first to be decreased after lesioning, with declines in other nAChR receptor populations only after more severe degeneration (Quik et al., 2001; Kulak et al., 2002a).
To extend our knowledge concerning the nAChR subtypes in monkey striatum and the effect of nigrostriatal damage on receptor expression, we performed quantitative autoradiography in control and parkinsonian monkey brain using 5-[125I]iodo-A-85380 ([125I]A-85380), a ligand with potential for in vivo imaging of nAChRs. Initial studies had suggested that [125I]A-85380 was selective for α4β2* nAChR subtypes (Sullivan et al., 1996; Villemange et al., 1998; Mukhin et al., 2000; Fujita et al., 2002), although more recent work indicates it may also bind to α6* nAChRs (Kulak et al., 2002b). In the present study we examine the effect of α-conotoxin MII on binding of [125I]A-85380 to β2*-containing nicotinic receptors and extend previous work elucidating the nAChR subtypes affected by nigrostriatal damage in the primate striatum.
Materials and Methods
Materials.
5-[125I]Iodo-A-85380 (5-[125I]iodo-3(2(S)-azetidinylmethoxy)pyridine, 1200–2200 Ci/mmol) was prepared as described (Musachio et al., 1998;Kulak et al., 2002b), and [125I]RTI-121 [3β-(4-[125I]iodophenyl)tropane-2β-carboxylic acid isopropyl ester; 2200 Ci/mmol] was purchased from PerkinElmer Life Sciences (Boston, MA). α-Conotoxin MII was synthesized according to the method of Cartier et al. (1996). Nicotine hydrogen tartrate and cytisine were obtained from Sigma-Aldrich (St. Louis, MO).
Animals.
Twenty adult, drug-free squirrel monkeys (Saimiri sciureus; Osage Research Primates, Osage Beach, MO) were used for these studies as previously described (Quik et al., 2001;Kulak et al., 2002a). Briefly, animals were housed individually on a 13-h light/11-h dark cycle, fed once daily, and had free access to water. MPTP treatment and behavioral assessments were performed as detailed earlier (Quik et al., 2001; Kulak et al., 2002a). Monkeys were killed 4 weeks after the final MPTP injection in accordance with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association. Ketamine hydrochloride (15–20 mg/kg i.m.) was administered to sedate the animals, followed by injection of 0.22 ml/kg i.v. euthanasia solution (390 mg of sodium pentobarbital and 50 mg of phenytoin sodium/ml).
Tissue Preparation.
The brains were removed, cut into 6-mm-thick blocks, quick frozen in isopentane, and kept at −80°C until use. Brain sections, 20 μm thick, were prepared at −20°C using a Leica cryostat. They were thaw-mounted onto poly(l-lysine)-coated slides, dried, and stored at −80°C. The squirrel monkey atlas (Emmers and Akert, 1963) was used to identify the different brain regions using Nissl-stained tissue sections. Level assignments indicate the distance (in millimeters) anterior to the interaural line; for example, level A15.0 is 15 mm anterior to the interaural line.
[125I]A-85380 Binding.
Binding was done as previously described (Mukhin et al., 2000). Twenty-micrometer brain sections were thawed and incubated at room temperature for 60 min in buffer (50 mM Tris, pH 7.0, 120 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 1.0 mM MgCl2) plus [125I]A-85380. Nonspecific binding was defined in the presence of 10 μM nicotine and was similar to the film blank. Inhibition curves were conducted in the presence of competing ligand with no additional changes to the protocol. Following incubation, the sections were washed twice for 5 min in buffer at 4°C and once for 10 s in ice-cold deionized H2O. After drying at room temperature, slides were exposed for 1 to 2 days to Kodak MR film (PerkinElmer Life Sciences), simultaneously with known125I standards (Amersham).
[125I]RTI-121 Binding.
To measure the dopamine transporter, [125I]RTI-121 binding was performed as previously described (Quik et al., 2001; Kulak et al., 2002a). Sections were exposed twice for 15 min to preincubation buffer (50 mM Tris-HCl, pH 7.4, 120 mM NaCl, 5 mM KCl). Incubation was then done for 2 h with 50 pM [125I]RTI-121 in preincubation buffer plus 0.025% bovine serum albumin and 1 μM fluoxetine. Sections were washed four times for 15 min in ice-cold preincubation buffer and once for 10 s in ice-cold doubly distilled H2O, and dried. They were exposed for 2 days to Hyperfilm βmax film (Amersham Biosciences, Piscataway, NJ). The dopamine uptake inhibitor nomifensine (100 μM) was used to define nonspecific binding.
Data Analysis and Quantitation.
Quantitative differences in radioligand binding were determined by computer-assisted densitometry (ImageQuant; Molecular Dynamics, Sunnyvale, CA). Absorbance values of autoradiographic film images were corrected for background and converted to femtomoles per milligram of tissue by comparison with curves generated from known radioisotope standards exposed to film with the sections. Absorbance values for tissue sections and standards were within the linear range of the film. For the caudate and putamen, density was measured at levels A15.0, A13.5, and A12.5, and the values were averaged to obtain the amount of radioligand binding in each region per monkey.
The Kd,Bmax, andKi values were determined using GraphPad Prism (GraphPad Software Inc., San Diego, CA). Saturation and inhibition curves were fit to both one- and two-site models and statistically compared to determine best fit (GraphPad Prism). All values are expressed as the mean ± S.E.M. for the indicatedn. For statistical analysis, one-way ANOVA followed by Newman-Keuls multiple comparison was used with p < 0.05 considered significant (GraphPad Prism).
Results
Saturable Binding of [125I]A-85380 in Monkey Brain.
Tissue sections from control monkey brains were incubated with 5 to 750 pM [125I]A-85380 (Fig.1). [125I]A-85380 binding plateaued with Bmax values of 22.0 ± 1.8 fmol/mg for the caudate, 21.70 ± 2.4 fmol/mg for the putamen, and 15.8 ± 2.7 fmol/mg for the frontal cortex.Kd values were 305.0 ± 72.1 pM for the caudate, 274.4 ± 70.0 pM for the putamen, and 107.9 ± 12.5 pM for the frontal cortex (n = 3). The data fit best to a single-site model, suggesting that [125I]A-85380 binds with a similar affinity to one or more populations of nAChRs in monkey caudate, putamen, and frontal cortex.
Widespread Distribution of [125I]A-85380 Sites in Control Monkey Brain.
[125I]A-85380 autoradiography showed that binding sites were present throughout the monkey brain (Fig. 2), with the greatest expression in the interpeduncular nucleus, medial habenula, and thalamus (Table 1). In the basal ganglia, the highest density was observed in the caudate, putamen, and substantia nigra, with moderate binding in the nucleus accumbens and low levels of expression in the globus pallidus. This distribution is similar to that observed with [125I]epibatidine (Kulak et al., 2002a), although there were differences in signal intensity in a given brain region with the two radioligands. For example, binding of [125I]A-85380 is similar for the caudate, putamen, and substantia nigra, whereas the substantia nigra exhibits more [125I]epibatidine binding than the striatum; there is also higher [125I]A-85380 binding in the basal ganglia than in the hippocampus, with the reverse situation for [125I]epibatidine binding (Kulak et al., 2002a).
[125I]A-85380 Binding Is Inhibited by nAChR Ligands.
In control animals, nicotine and cytisine dose dependently inhibited [125I]A-85380 binding that was complete at 10–6 M in all regions investigated (Fig. 3, Table2). In contrast, α-conotoxin MII (1 μM) blocked only 30% of the [125I]A-85380 sites in the caudate and putamen, and none of the [125I]A-85380 sites in the frontal cortex. Our previous studies (Kulak et al., 2002a) had shown that α-conotoxin MII inhibited [125I]epibatidine by 50% in the caudate and putamen. To determine whether the percentage inhibitions of [125I]A85380 and [125I]epibatidine by conotoxin MII were different or not, competition experiments were performed simultaneously using sequential tissue sections from the same animals (control, moderate, and severe). The percentage inhibition of [125I]A-85380 by α-conotoxin MII was statistically significantly less than the percentage inhibition of [125I]epibatidine sites in control animals, with values of 28.7 ± 4.3 and 45.9 ± 4.9%, respectively (see also Fig. 6).
Nigrostriatal Damage Selectively Decreases [125I]A-85380 Binding in the Caudate and Putamen.
Lesioning of the dopaminergic nigrostriatal system with MPTP resulted in a decline in [125I]A-85380 binding in the caudate and putamen (Fig. 4). This appeared to be due to a decline in the maximal number of sites (Bmax values), with declines to 11.3 ± 3.9 and 11.7 ± 3.6 fmol/mg in the lesioned caudate and putamen, respectively (see Fig. 1). This effect was selective for the striatum, with no change in theBmax value for the frontal cortex (18.1 ± 3.6 fmol/mg) as compared with control.Kd values were similar in control and lesioned brain sections (caudate, 284.7 ± 113.2 pM; putamen, 299.0 ± 121.0 pM; frontal cortex, 149.2 ± 25.2 pM).
As an approach to evaluate the effects of varying degrees of nigrostriatal damage on [125I]A-85380 binding sites, MPTP-treated monkeys were separated into two groups as previously described (Quik et al., 2001; Kulak et al., 2002). This includes moderately lesioned animals with striatal dopamine transporter levels ∼25% of control and severely lesioned animals with transporter levels ≤10% of control (Quik et al., 2001; Kulak et al., 2002a). After nigrostriatal lesioning, [125I]A-85380 binding (250 pM) decreased by 42 and 55% in the caudate of moderately and severely lesioned animals, respectively, with similar results in the putamen (Fig. 4).
MPTP Treatment Effects Inhibition of [125I]A-85380 Binding by nAChR Ligands.
Although α-conotoxin MII inhibited [125I]A-85380 binding in striatal sections from control monkeys (Fig. 5A), the conotoxin-sensitive component of [125I]A-85380 binding was completely eliminated in sections from moderately and severely lesioned animals. Interestingly, despite the finding that inhibition of [125I]A-85380 by cytisine or nicotine was monophasic in control striatal sections, there was a significant decrease in the Ki value for both these ligands after nigrostriatal damage (Table 2). These results suggest that [125I]A-85380 binds to multiple nicotinic receptor subtypes that cannot be distinguished in competition assays because of their similar affinities. Complete inhibition of [125I]A-85380 binding was observed at the higher concentrations (10−6 M) of nicotine and cytisine (Fig. 5).
Discussion
The present results are the first to describe the effects of nigrostriatal damage on the binding of the novel radioligand [125I]A-85380 to nAChRs in nonhuman primates. Our data to characterize [125I]A-85380 sites in control brain tissue show that it represents an excellent ligand for autoradiographic studies, with little if any nonspecific binding. The overall distribution of [125I]A-85380 sites shares resemblances with that of [125I]epibatidine (Kulak et al., 2002a), although there were differences in binding in various brain areas, an observation suggesting that not all [125I]epibatidine sites contain the β2 subunit. For example, the increased [125I]epibatidine binding in the substantia nigra compared with the striatum may be due to the presence of β3* and/or β4* nAChRs, as the mRNA for these subunits is highly expressed in this region (Quik et al., 2000a,b).
[125I]A-85380 appears to interact with a site in monkey brain with the properties of a nicotinic receptor ligand since binding was completely inhibited by nicotine and cytisine. The potential subunit composition of [125I]A-85380 binding sites in control monkey brain was investigated using the α6-selective nAChR ligand α-conotoxin MII (Champtiaux et al., 2002;Whiteaker et al., 2002). The toxin inhibited a maximum of 30% of [125I]A-85380 binding in the caudate and putamen, with no effect in the frontal cortex. This percentage inhibition of [125I]A-85380 binding by α-conotoxin MII was significantly less than that for [125I]epibatidine binding (50%) in experiments conducted in parallel with the same monkeys. This was somewhat unexpected since [125I]A-85380 is a ligand with greater nAChR subunit specificity than [125I]epibatidine, and one might anticipate that it binds to a subset of [125I]epibatidine binding sites, resulting in a greater, not smaller, inhibition. These findings are consistent with the presence of two populations of α-conotoxin MII-sensitive nicotinic receptors (Fig.6). With respect to their composition, it has recently been demonstrated that the α-conotoxin MII-sensitive nAChRs in rodent striatum contain α6 and β2 subunits (Grady et al., 2001; Champtiaux et al., 2002; Whiteaker et al., 2002). If these observations in rodents can be extrapolated to the monkey, they suggest that one of the α-conotoxin MII-sensitive populations in monkey striatum contains the α6 and β2 subunits.
Our data with MPTP-treated monkeys demonstrate significant changes in striatal nAChR expression compared with control tissue. After moderate degeneration, there is ∼50% decrease in [125I]A-85380 binding sites in monkey striatum, which is similar in magnitude to the decline we had previously identified in striatal [125I]epibatidine sites (Kulak et al., 2002a). Our earlier results had also shown that [125I]epibatidine binding in control striatum is inhibited 50% by α-conotoxin MII and suggested that all the sites lost after lesioning were α-conotoxin MII-sensitive nAChRs (Kulak et al., 2002a). Unexpectedly, the present data in control animals showed that α-conotoxin MII maximally inhibited [125I]A-85380 binding by only 30%, indicating that only a portion of the 50% [125I]A-85380 sites lost after MPTP treatment are α-conotoxin MII-sensitive nAChRs. One explanation consistent with these observations is that there are two populations of α-conotoxin MII-sensitive receptors in the striatum that are both decreased after moderate nigrostriatal damage.
With respect to their potential subunit arrangement, one of these nAChR populations may consist of receptors containing an α6β2 interface. The second population may be composed of β2* receptors with two nonequivalent ligand-binding surfaces, neither of which would include a α6β2 interface. However, the β2 subunit may form an interface with an α subunit other than α6, since it binds [125I]A-85380 and [125I]epibatidine but not α-conotoxin MII. The other interface may involve an α6 but not a β2 subunit since it binds [125I]epibatidine and α-conotoxin MII but not [125I]A-85380, and may be of an α6/β4, or α6/β3 configuration. The rationale for this latter possibility stems from the observation that there is a very intense and selective localization of β3 mRNA in the substantia nigra (Le Novere et al., 1996; Han et al., 2000; Quik et al., 2000a). Moreover, there is a decrease in striatal [125I]α-conotoxin MII binding in β3 knockout mice, suggesting that at least one of the α-conotoxin MII-sensitive nAChR populations includes a β3 subunit (Whiteaker et al., 1998; M. Quik, unpublished observation).
Our data also suggest the presence of a third nAChR population that forms 50% of the striatal nAChRs that bind [125I]A-85380 and [125I]epibatidine. These nAChRs do not bind α-conotoxin MII and are decreased only with severe nigrostriatal lesioning. The ligand affinity is consistent with a population of nAChRs that contain at least one β2, but not an α6, subunit at the ligand binding interface.
A question that arises is the cellular localization of the different nicotinic receptor subtypes in striatum. Our lesion studies with the selective dopaminergic neurotoxin MPTP show that declines in α-conotoxin MII-sensitive receptors mirror those of the dopamine transporter, an observation suggesting that the majority of these sites are located presynaptically on dopaminergic afferents from the substantia nigra. However, α-conotoxin MII-sensitive receptors comprise only ∼50% of the total nicotinic receptor population as measured using radiolabeled epibatidine. Although the precise localization of the remaining sites awaits further study, they may be located on striatal GABAergic cell bodies, cholinergic interneurons, incoming glutamatergic terminals from the cortex, serotonergic afferents from the raphe nucleus, and/or other neurotransmitter inputs to the striatum (Schwartz et al., 1984; Smith and Kieval, 2000; Zhou et al., 2001).
Recent studies suggest that the 123I-labeled 5-iodo-A-85380 ([123I]A-85380) represents a promising candidate for in vivo imaging of nicotinic receptors in human brain, potentially providing a diagnostic tool for cholinergic-related disorders such as Parkinson's and Alzheimer's diseases (Vaupel et al., 2001; Wong et al., 2001; Fujita et al., 2002). The present experiments indicate that [125I]A-85380 binds to multiple β2* nAChRs, at least in the nonhuman primate striatum. Since results from nonhuman primates often model those in humans, our findings would suggest caution in data interpretation when using [123I]A-85380.
In summary, the present results show that the β2-selective nicotinic receptor ligand [125I]A-85380 interacts specifically with multiple nAChR populations in monkey striatum and that the various β2* nAChR subtypes are differentially affected by nigrostriatal damage. Lesion studies, combined with competition analyses of [125I]A-85380 and [125I]epibatidine binding in the presence of α-conotoxin MII (Kulak et al., 2002a), extend our earlier data and suggest that there are at least three nAChR populations in monkey striatum. This includes an α6* receptor population, which is particularly susceptible to nigrostriatal damage and may include two subtypes, with and without an α6/β2 interface. The other population of β2-containing nAChR sites is insensitive to α-conotoxin MII and less vulnerable to nigrostriatal degeneration.
Footnotes
-
This work was supported by the California Tobacco Related Disease Research Program Grants 7RT-0015 and 11RT-0216 and by National Institutes of Health Grants CA77349, MH53631, and DA12242.
-
DOI: 10.1124/jpet.102.039347
- Abbreviations:
- nAChR
- nicotinic acetylcholine receptor
- [125I]A-85380
- 5-[125I]iodo-3(2(S)-azetidinylmethoxy)pyridine
- MPTP
- 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
- [125I]RTI-121
- 3β-(4-[125I]iodophenyl)tropane-2β-carboxylic acid isopropyl ester
- ∗
- nicotinic receptors containing the indicated α and/or β subunit and possibly also additional undefined subunits
- Received May 21, 2002.
- Accepted July 3, 2002.
- The American Society for Pharmacology and Experimental Therapeutics