PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Biochim Biophys Acta. Author manuscript; available in PMC 2010 September 1.
Published in final edited form as:
PMCID: PMC2744590
NIHMSID: NIHMS127101

Photoaffinity Labeling the Agonist Binding Domain of α4β4 and α4β2 Neuronal Nicotinic Acetylcholine Receptors with [125I]Epibatidine and 5[125I]A-85380

Abstract

The development of nicotinic acetylcholine receptor (nAChR) agonists, particularly those that discriminate between neuronal nAChR subtypes, hold promise as potential therapeutic agents for many neurological diseases and disorders. To this end, we photoaffinity labeled human α4β2 and rat α4β4 nAChRs affinity-purified from stably transfected HEK-293 cells, with the agonists [125I]epibatidine and 5[125I]A-85380. Our results show that both agonists photoincorporated into the β4 subunit with little or no labeling of the β2 and α4 subunits respectively. [125I]epibatidine labeling in the β4 subunit was mapped to two overlapping proteolytic fragments that begin at β4V102 and contain Loop E (β4I109-P120) of the agonist binding site. We were unable to identify labeled amino acid(s) in Loop E by protein sequencing, but we were able to demonstrate that β4Q117 in Loop E is the principal site of [125I]epibatidine labeling. This was accomplished by substituting residues in the β2 subunit with the β4 homologs and finding [125I]epibatidine labeling in β4 and β2F119Q subunits with little, if any, labeling in α4, β2, or β2S113R subunits. Finally, functional studies established that the β2F119/β4Q117 position is an important determinant of the receptor subtype-selectivity of the agonist 5I-A-85380, affecting both binding affinity and channel activation.

Keywords: Cys-loop receptor, neuronal nicotinic receptor, HEK-293 cell, affinity-purification, photoaffinity labeling, protein sequencing

1. Introduction

Neuronal nicotinic acetylcholine receptors (nAChR) comprise a diverse group of cation-selective ligand-gated ion channels that are widely distributed in neuronal and non-neuronal cells [1]. Within the nervous system, nAChRs influence a variety of neuronal functions by mediating the action of the endogenous neurotransmitter acetylcholine (ACh) and modulating the release of ACh and a panel of other neurotransmitters [2]. Therefore, drugs that target neuronal nAChRs have therapeutic potential in many pathophysiological conditions including Alzheimer's disease and nicotine addiction [3]. The neuronal nAChRs are pentameric membrane proteins formed by the assembly of five nAChR subunits (e.g. homomeric α7 and heteromeric α4β2 nAChRs). The agonist binding sites (ABS) of nAChRs are located within the N-terminal extracellular domain at the interface of adjacent subunits and different nAChR subunit combinations form ABS with distinct physical and pharmacological properties [4,5]. This diversity of neuronal nAChR subtypes, which is typical for many brain regions, makes it necessary to develop subtype-selective ligands to pharmacologically target a specific nAChR subtype without invoking side effects associated with binding at other nAChRs. The high degree of sequence homology among nAChR subunits (38-68%) allows the use of common structural templates such as the 4Å structure of Torpedo nAChR [6] and the X-ray structure of the acetylcholine binding protein (AChBP; [7,8] to study the ABS of neuronal nAChRs. But, to develop subtype-selective ligands, it is necessary to develop more refined models that reflect the subtle structural differences among the ABS of different neuronal nAChR subtypes.

In heteromeric neuronal nAChRs, the principal components of the ABS (amino acids in Loops A-C) are carried by α subunits and the complementary components (amino acids in Loops D-F) are carried by the β2 or β4 subunit. Amino acids residues that contribute to the ABS from Loops A-C are conserved among α subunits except α5 which does not form an ABS [9]. In the complementary component of the ABS, amino acid residues in Loops D, E and F, except the tryptophan residue in Loop D, are not conserved among Torpedo (muscle-type) γ and δ and neuronal β2 and β4 nAChR subunits.

Regardless of the α subunit subtype, the presence of either the β2 or β4 subunit influences the pharmacological properties of the ABS. Epibatidine and its analogue, iodo-epibatidine, bind with high affinity (~pM) to heteromeric neuronal nAChRs and with much lower affinity to homomeric neuronal and muscle nAChRs [3]. [3H]epibatidine (and [125I]epibatidine) bind with higher affinity to β2- compared to β4-containing receptors (α2β2/α2β4, ~8 fold difference; α3β2/α3β4, ~22 fold; α4β2/α4β4, ~3 fold; [10,11]. 5I-A-85380, a more recently introduced agonist, shows very high discrimination between α4β2 and α4β4 receptors. It is much less potent at inhibiting binding of [3H]epibatidine to the α4β4 receptor (IC50 24 nM) than the α4β2 receptor (IC50 0.06 nM; [10]).

While the structural determinants that are responsible for these pharmacological differences are not completely understood, amino acid residues within Loop E are implicated in conferring agonist binding selectivity between β2- and β4-containing nAChRs. Substitution of amino acids β4-(106-120) with β2-(104-120) confer β2-like agonist binding properties [12]. Moreover, azidoepibatidine, a photoreactive analogue of epibatidine, labels M116 within Loop E (as well as C195 within Loop C) of Aplysia AChBP [13].

In this report, we extended these studies by employing [125I]epibatidine and 5[125I]A-85380 as photoaffinity probes of the agonist binding domain of α4β2 and α4β4 nAChRs. Starting with nAChRs affinity-purified from stably transfected HEK-293 cells, our data show that both [125I]epibatidine and 5[125I]A-85380 selectively photoincorporate into the β4 subunit (with little or no labeling in β2 and α4 subunits respectively) and for [125I]epibatidine specific (agonist inhibitable) labeling mapped to Loop E (β4I109-P120) of the complementary component of the agonist binding site. We were unable to directly determine the sites of labeling in the β4 subunit by protein sequencing. However, we introduced point mutations in Loop E at β2S113 and β2F119 (equivalent to M116 in AChBP) to the corresponding residues in the β4 subunit (designated hereafter β2S113R and β2F119Q, respectively). For wild-type and mutant nAChRs transiently expressed in HEK-293 cells, we found labeling in the β4 and β2F119Q subunits, indicating that β4Q117 is the principal site of [125I]epibatidine labeling in the α4β4 nAChR. Furthermore, when we tested the effect of these mutations on agonist binding and channel activation, we found that β2F119 but not β2S113 plays a significant role in determining the pharmacological profile of the ABS of the α4β2 nAChR.

2. Material and methods

2.1 Materials

[125I]epibatidine and 5[125I]A-85380 (2200 Ci/mmol) were obtained from Perkin Elmer Life Sciences, Inc. (Boston, MA) and stored in 95% ethanol at 4°C. Acetylcholine chloride (ACh), carbamylcholine chloride (Carb), and bromoacetylcholine bromide were from Sigma-Aldrich (St. Louis, MO). 5-iodo-A-85380 dihydrochloride (5I-A-85380) was from Tocris, Ellisville, MO. Glutamyl endopeptidase Glu-C (V8 protease) was from MP Biochemicals and trypsin (TPCK-treated) from Worthington. Peptide-N-Glycosidase F (PNGase F; 5 units/μL), endo-β-N-acetylglycosaminidase (Endo H; 3 mU/μl), and protease inhibitor cocktail set III were from Calbiochem (LaJolla, CA. Sodium cholate and CHAPS were from USB Corporation (Cleveland, OH). Affi-Gel 10 was from Bio Rad (Hercules, CA).

2.2 Constructs, transient transfection and stable cell line

These techniques have been published [14-16]. Human α4 and β2 constructs were kindly provided by Dr. Jon M. Lindstrom (University of Pennsylvania) (α4 accession number NM000744, β2 accession number NM 000748). Subunit cDNAs were transferred to pcDNA3 (Invitrogen, San Diego CA) for expression. Mutations were made using the QuikChange kit (Stratagene, La Jolla CA), and all mutated subunits were sequenced over the full length of the subunit to confirm that only the desired mutation had been produced. HEK 293 cells for transient transfection were plated in 150 mm tissue culture dishes and transfected when at about 75% confluence. Transient transfections were performed using Effectene according to the manufacturer's instructions (Qiagen, Valencia CA), with a mixture of equal amounts of α4 and β2 cDNA. Cells were washed with culture medium 24 hours after transfection and harvested 2 to 3 days later.

The HEK 293 cell line that stably expresses hα4β2 nAChRs (HEK-hα4β2 hereafter) was prepared by growing the transfected cells in the presence of G418 (450 μg/ml) and immunoselection with mAb 270 as described [16]. HEK293 cells stably expressing rat α4β4 nAChRs (HEK-α4β4 hereafter) were kindly provided by Drs. Yingxian Xiao and Kenneth J. Kellar (Georgetown University; [10]). HEK-hα4β2 and HEK-rα4β4 cells were grown at 37°C in a humidified incubator at 5 % CO2, in 150 mm tissue culture dishes and were maintained in DMEM/Ham's F-12 (Mediatech, Inc.), supplemented with 10% fetal bovine serum, 100 units/mL penicillin G, 100 μg/mL streptomycin and 450 μg/mL geneticin (G418) as a selection agent.

2.3 Membrane preparation and affinity purification

For membrane preparation, transiently transfected HEK 293, HEK-hα4β2 and HEK-rα4β4 cells were homogenized in vesicle dialysis buffer (VDB: 100 mM NaCl, 0.1 mM EDTA, 0.02% NaN3, 10 mM MOPS, pH 7.5) in the presence of 0.2 μL/mL protease inhibitor cocktail set III (Calbiochem). Membrane fractions were isolated by centrifugation (39,000g for 1 h) then resuspended in VDB containing 0.2 μL/mL protease inhibitor cocktail III and stored at -80°C. For affinity purification, HEK-hα4β2 and HEK-rα4β4 cell membranes in VDB containing 0.2 μL/mL protease inhibitor cocktail III (1 mg/ml protein) were solubilized with 1% CHAPS, centrifuged (91,500g for 1 h) to pellet insoluble material, and then dialyzed for 5 h against 1% cholate. The hα4β2 and rα4β4 nAChRs (typically from ~600 dishes; ~2g HEK-293 cell membrane protein) were affinity-purified on a bromoacetylcholine bromide-derivatized Affi-Gel 10 column and reconstituted into asolectin/total lipid extract from porcine brain (Avanti Polar Lipids) lipid vesicles as previously described [17]. The purity of hα4β2 and rα4β4 nAChRs was >50% as estimated by densiometric quantification of Coomassie Blue stained gels and by [3H]nicotine binding (~ 4nmol binding sites/mg of protein). Densiometric analysis of stained gels containing α4β2 and α4β4 AChRs was also consistent with a pentameric subunit stoichiometry of (α4)2(β2/β4)3 [17].

2.4 [125I]Epibatidine and 5[125I]A-85380 Photolabeling

nAChR containing HEK cell membranes (~ 500 μg) and affinity-purified hα4β2 and rα4β4 nAChR membranes (~30 μg) were incubated for 1 h with ~ 1 nM [125I]epibatidine or 5[125I]A-85380 (2200 Ci/mmol) in the absence or presence of 100 μM nicotine (or 100 nM epibatidine) then irradiated in glass text tubes, for 30 or 15 min respectively with a 254 nm hand-held UV lamp (Spectroline EN-280L). The labeled membranes were then pelleted by centrifugation (39,000g for 1 h), resuspended in electrophoresis sample buffer (12.5 mM Tris-HCL, 2% SDS, 8% sucrose, 1% glycerol, 0.01% bromophenol blue, pH 6.8), and resolved on 8% polyacrylamide/0.33 bisacrylamide gels [18]. After staining with Coomassie Blue R-250 (0.25% (w/v) in 45% methanol, 10% acetic acid, 45% H2O) and destaining to visualize bands, gels were dried and exposed to Kodak X-OMAT LS film with an intensifying screen at -80°C (overnight to 4 week exposure). The bands corresponding to the β4 subunit and a lower molecular weight fragment of β4 (β4′) were excised from stained 8% polyacrylamide gels, soaked in overlay buffer (5% sucrose, 125 mM Tris-HCl, 0.1% SDS, pH 6.8) for 30 min, transferred to 15% polyacrylamide mapping gels and digested in-gel with 10 μg V8 protease [19]. After electrophoresis, the mapping gels were processed for autoradiography as described above.

2.5 Isolation and sequencing of [125I]epibatidine-labeled peptides

[125I]epibatidine-labeled β4 subunit was isolated from a preparative scale labeling (~1 mg affinity-purified rα4β4 nAChR at 0.3 mg/ml labeled with 12-18 nM [125I]epibatidine) and subjected to in-gel digestion with V8 protease. Based on the autoradiograph of the mapping gel, labeled bands were excised and the polypeptides were retrieved by passive diffusion into 25 mL of elution buffer (0.1M NH4HCO3, 0.1% (w/v) SDS, 1% β-mercaptoethanol, pH 7.8) and concentrated in Centriprep-10 and Centricon-3 concentrators (10/3 kDa cutoff, Amicon) to a final volume of ~150 μl.

For sequence analysis, the eluted peptides were further purified by reversed-phase HPLC on a Shimadzu LC-10A binary HPLC system, using a Brownlee Aquapore C4 column and a nonlinear elution gradient described in the legend for Fig. 2. Fractions were collected every 2.5 min at a flow rate of 0.2 mL/min and 125I cpm associated with each fraction was determined by γ-counting. Peak 125I cpm containing fractions were loaded onto PVDF filters using Prosorb® Sample Preparation Cartridges (Applied Biosystems #401959) as recommended by manufacturer and sequenced on an Applied Biosystems PROCISE™ 492 protein sequencer configured to utilize 1/6 of each cycle of Edman Degradation for amino acid detection and the other 5/6 for 125I cpm counting.

Figure 2
Mapping the sites of [125I]epibatidine (Panel A) and 5[125I]A-85380 (Panel B) photoincorporation in the β4 subunit and β4′ subunit fragment. The bands corresponding to the β4 subunit and the β4′ subunit ...

For digestion with PNGase F and Endo H, eluted peptides were acetone precipitated (>85% acetone at -20°C overnight) to remove SDS, resuspended in 100 μL of Tris buffer (15 mM Tris, 0.1% SDS; pH 8.3) and incubated at room temperature for 2 d with 25 U of PNGase or 15 mU of Endo H. The digests were then fractionated on small pore (16.5%T, 6%C) Tricine SDS-PAGE gels [20]. After electrophoresis, Tricine gels were processed for autoradiography as described above for 8% gels.

2.6 [3H]Epibatidine saturation and competition binding assays

Saturation binding of [3H]epibatidine (45.1 Ci/mmol; Perkin Elmer Life Sciences) and the effect of 5I-A-85380 on [3H]epibatidine binding to hα4β2, rα4β4, hα4β2F119Q and hα4β2S113R nAChR HEK-293 cell membranes were determined using a centrifugation assay. For saturation [3H]epibatidine binding, membranes at 0.33 mg protein/mL VDB were incubated for 5 h at room temperature with increasing concentrations of [3H]epibatidine (final concentrations 2-60 nM). For competition binding, membranes at 0.25 mg protein/mL were incubated for 5 h at room temperature with [3H]epibatidine (5 nM), in the absence or the presence of increasing concentrations of 5I-A-85380 (final concentrations 0.1 nM -10 μM). For both saturation and competition binding, bound and free [3H]epibatidine were separated by centrifugation (39,000g for 1 h) then quantified by liquid scintillation counting. For saturation [3H]epibatidine binding, total, nonspecific, and specific 3H cpm were converted to fmol of bound [3H]epibatidine/mg of protein, and free 3H cpm was converted to nM [3H]epibatidine. For competition binding, bound [3H]epibatidine was normalized as a percentage of total binding in the absence of 5I-A-85380. Non-specific binding was determined in the presence of 150 μM epibatidine. Curve fitting and parameter estimations (including calculation of standard deviation) were performed using Graph Pad Prism v5.0 software (San Diego, CA).

2.7 Oocyte expression and electrophysiological recordings

Voltage clamp electrophysiological studies were made of receptors expressed in Xenopus oocytes [15,21]. cRNA was synthesized by linearizing the plasmid with Xho1, followed by cRNA synthesis using mMessage mMachine T7 kit (Ambion, Austin TX). The cRNA was dissolved in RNAase-free water, and 10 to 12 ng (based on the OD at 260 nm) in a volume of 18 nL was injected into each oocyte. The subunits were injected at a 1:8 ratio (α4:β2). Oocytes were voltage clamped with a two-electrode clamp (Warner Instruments, Hamden CT). The bath solution contained (mM) NaCl, 96; KCl, 2; MgCl2, 1; BaCl2, 1.8; HEPES, 10; pH adjusted to 7.5 with NaOH, and recordings were made at room temperature (approximately 18-21 °C). Analog data were filtered at 20 Hz with a Bessel low pass filter and digitized at 50 ms intervals with using pClamp (Molecular Dynamics, Mountain View CA). Data were subsequently analyzed using pClamp, Excel (Microsoft, Redmond WA) and SigmaPlot (Systat Software, San Jose CA). Concentration-response curves were fit with the equation Y(X) = Ymax {Xn / (Xn + EC50n)}, where Ymax is the maximal response, X is the concentration of agonist, EC50 is the concentration producing a half-maximal response and n is the Hill coefficient. ACh was prepared as a 1 M stock solution and 5I-A-85380 was prepared at 50 mM in ion exchange-purified distilled water and stored in aliquots at −20°C. All solutions applied were made freshly from stock on the day of an experiment.

3. Results

3.1 [125I]Epibatidine and 5[125I]A-85380 labeling of α4β4 and α4β2 nAChRs

Photoaffinity labeling studies were carried out with human α4β2 and rat α4β4 neuronal nAChRs affinity-purified from stably transfected HEK-293 cells. Following a 1 h equilibration with 1 nM [125I]epibatidine or 5[125I]A-85380, nAChR samples (30 μg in 500 μL VDB) were irradiated for 30 or 15 min (at 254 nm) respectively and the pattern of incorporation assessed by SDS-PAGE followed by autoradiography. As seen in the autoradiographs of 8% acrylamide gels (Fig. 1), for α4β4 nAChRs both [125I]epibatidine (Fig. 1A) and 5[125I]A-85380 (Fig. 1B) incorporated selectively into the β4 subunit, including a lower molecular weight band subsequently identified as a proteolytic fragment of the β4 subunit (β4′), with no apparent labeling of the α4 subunit. The specificity of labeling was addressed by co-application of 100 μM nicotine (Fig. 1, +lanes), which significantly reduced the extent of [125I]epibatidine labeling (Fig. 1A; >50% reduction) in both the β4 subunit and the β4′ subunit fragment as determined by γ-counting of excised gel bands. In subsequent labeling experiments, a 15 min preincubation of α4β4 nAChR membranes with either 100 μM nicotine or 100 nM epibatidine (Fig. 1B, + lane) prior to the addition of radiolabeled agonist resulted in a complete elimination of labeling.

Figure 1
Photoincorporation of [125I]epibatidine (Panel A) and 5[125I]A-85380 (Panel B) into α4β4 nAChRs. ~0.3 mg of affinity-purified rα4β4 nAChRs were co-equilibrated for 1 h with 6 μCi [125I]epibatidine or 5[ ...

In contrast to α4β4 nAChRs, for α4β2 nAChRs very little if any photoincorporation is apparent in either the α4 or β2 subunit for either [125I]epibatidine or 5[125I]A-85380 (Supplementary Fig. S1). This result was rather surprising given the pM binding affinity of epibatidine and 5I-A-85380 for α4β2 nAChRs and the high degree of sequence conservation between the β2 and β4 subunits (64%; [22]) and immediately suggests that labeling differences between α4β2 and α4β4 nAChRs primarily reflect differences in the photoreactivity of a limited number of amino acid residues.

The bands corresponding to the β4 subunit and β4′ fragment were next isolated from 8% gels and digested in-gel with V8 protease. [125I]epibatidine and 5[125I]A-85380 photoincorporation was detected in two subunit fragments with apparent molecular masses of 6.4 KDa (β4V8-6.4; β4′V8-6.4,) and 14.1 KDa (β4V8-14.1; β4′V8-14.1; Figure 2). In each, addition of 100 nM epibatidine during nAChR labeling eliminated incorporation in each subunit fragment (Figure 2, + lanes) consistent with labeling that is restricted to the agonist binding domain.

3.2 Mapping the sites of [125I]epibatidine labeling in the α4β4 nAChR

While the cost of 5[125I]A-85380 (Perkin Elmer) made preparative scale labeling experiments prohibitively expensive, to identify [125I]epibatidine-labeled fragments, β4V8-6.4, β4V8-14.1 (from β4 and β4′) were isolated from a preparative scale labeling (1 mg α4β4 nAChR; 13 nM [125I]epibatidine) and further purified by reversed-phase HPLC (Figures 2C and 2D). HPLC fractions containing peak 125I cpm were pooled and subjected to automated Edman degradation (10 sequencing cycles). The primary peptide detected for both β4V8-6.4 and β4V8-14.1 fragments began at β4V102 [residue and pmol amount detected in each sequencing cycle for β4V8-6.4 are V(30)/S(5)/V(20)/Y(10)/T(6.5)/N(11)/V(10)/I(11)/V(9)/R(5) and for β4V8-14 are V(12.6)/S(2.6)/V(2.3)/Y(1.4)/T(1.2)/N(1.9)/V(1.8)/I(1.1)/V(2)/R(0.4)]. As shown in Figure 3 (the first ten amino acids detected by protein sequencing are underlined) generation of these two peptides results from V8 protease cleavage at β4E101. Based on the apparent molecular mass of each fragment (6.4 and 14 KDa), each of these two fragments extends well beyond β4V102 and contains the amino acid sequence of Loop E (β4-I109 – P120) of the complementary component of the agonist binding sites. While the primary peptide detected in sequencing of β4V8-6.4 and β4V8-14.1 began at β4V102, it is possible that the radioactivity ([125I]epibatidine labeling) is associated with a peptide sequence that was present at much lower abundance and therefore was not detected during sequencing. To further confirm the association between the radioactivity (i.e. [125I]epibatidine labeling) and the proteolytic fragments containing Loop E (β4V8-6.4 and β4V8-14.1), we took advantage of the fact that β4V102 is located 12 and 40 amino acids before two potential sites for N-glycosylation at β4N113 and β4N141, respectively (Bolded and Italized in Figure 3). We therefore examined the effects of Endo H, which hydrolyzes asparagine-linked oligomannose but not complex oligosaccharides, and PNGase F, which hydrolyzes asparagine-linked high mannose as well as complex oligosaccharides, on the electrophoretic mobility of β4/β4′V8-6.4 (V102-E131) and β4/β4′V8-14.1 (V102-E175; Figure 4). The autoradiographic bands for Endo H- and PNGase F-treated β4V8-6.4 migrated with an apparent molecular mass of 4 KDa (2.4 KDa shift) consistent with the removal of a oligomannose (simple sugar) at β4N113. The bands for Endo H- and PNGase F-treated β4V8-14.1 migrated with an apparent molecular mass of 7.4 KDa (6.7 KDa shift) consistent with the removal of two simple sugar moieties (~3.3 KDa each) at β4N113 and β4N141. While these results are consistent with the site of [125I]epibatidine photoincorporation residing in Loop E of the agonist binding site, alternative explanation are possible. For example, V8 cleavage at β4E47 could generate fragments of 5.4 KDa (with glycosylation ~7 KDa) and 10 KDa (~14 KDa) that include glycosylation at β4N68 and β4N113 and labeling within Loop D (e.g. W55, K57). However, a peptide sequence beginning at β4D48 was not detected in sequencing of β4V8-6.4 and β4V8-14.1 material and the apparent molecular mass and deglycosylation shift of β4V8-6.4 are not consistent with the results presented in Fig. 4, leaving [125I]epibatidine photoincorporation within Loop E as the simplest explanation of the data.

Figure 3
Amino acid sequence alignment of Loop E from different nAChRs. The approximate limits of Loop E are indicated, key residues are bolded, potential sites of N-linked glycosylation are bolded and italized, and amino acids in the β4 subunit identified ...
Figure 4
The effect of deglycosylation on the electrophoretic mobility of β4V8-6.4 and β4V8-14.1 fragments. [125I]epibatidine-labeled β4V8-6.4 and β4V8-14.1 fragments (e.g. Figure 2) were digested at room temperature for 2 d with ...

When the [125I]epibatidine-labeled β4V8-6.4 fragment was subjected to amino acid sequencing on a configuration that allows quantification of 125I cpm release associated with each cycle of Edman degradation, the release profile reflected a wash out of 125I cpm from the sequencing filter (Supplementary Figure S2). This result suggests that the labeling of one or more amino acids was not stable during amino acid sequencing. To further test this, a Loop E-containing peptide was isolated from [125I]epibatidine-labeled Torpedo γ subunit and subjected to amino acid sequencing under the same conditions (Supplementary Figure S3). The 125I cpm profile was similar to that for β4V8-6.4 establishing that the UV-induced [125I]epibatidine amino acid adduct(s) are not stable under the acidic sequencing conditions, limiting our ability to directly determine the site of [125I]epibatidine photolabeling in Loop E through direct sequencing of labeled peptide. As detailed in the following sections we therefore explored other means by which to identify the site(s) of [125I]epibatidine photoincorporation in the α4β4 AChR.

3.2 Effect of β2S113R and β2F119Q mutations on receptor expression and epibatidine and 5I-A-85380 binding affinity

As stated previously, [125I]epibatidine/5[125I]A-85380 labeling of the β4 subunit (for α4β4 nAChRs) and the absence of labeling in α4β2 nAChRs is rather surprising given the pM binding affinity of epibatidine and 5I-A-85380 for α4β2 nAChRs and the high degree of sequence conservation between the β2 and β4 subunits (64%; [22]) and immediately suggests that labeling differences between α4β2 and α4β4 nAChRs primarily reflects differences in the photoreactivity of a limited number of amino acid residues within Loop E of the agonist binding domain. Amino acids within Loop E have been previously implicated in the differences of binding affinities between α–γ and α–δ ABS of the mouse nAChR [23] and between β2- and β4-containing nAChRs [12]. Amino acid sequence alignment of β2 and β4 nAChR subunits (Figure 3) and the fact that azidoepibatidine labeled M116 within Loop E of the Aplaysia AChBP [13] suggest that β4Q117 (equivalent to M116 in AChBP) and possibly β4R111 (this non conserved residue is located on the same face of β-strand as β4Q117) are likely targets for [125I]epibatidine and 5[125I]A-85380 labeling and the corresponding residues in the β2 subunit (β2F119 and β2S113) may be important selectivity determinants of β2-containing receptors. To test this, we constructed hα4β2 receptors containing point mutations at β2S113 and β2F119, in which the residue at each position was substituted to the corresponding residue in the rat β4 subunit (β2F119Q and β2S113R, respectively, hereafter).

When α4β2F119Q and α4β2S113R receptors were transiently expressed in HEK 293 cells, they displayed similar expression levels (~2-4 pmol [3H]epibatidine binding sites/mg protein) and [3H]epibatidine binding affinities (400-700 pM) to those of the wild-type (WT) hα4β2 (Figure 5A-C). Competition binding experiments (Figure 5E) further established that the α4β2 selective agonist 5I-A-85380 ([10]; Figure 2D) displaced [3H]epibatidine binding to α4β2 and α4β2S113R nAChRs with identical potency (IC50 11.43 ±0.89 and 11.18 ±0.68 nM respectively), but with ~4- and 200-fold reduced potency for hα4β2F119Q and rα4β4 nAChRs (IC50 40.0 ±1.9 nM and 2.23 ± 0.13 μM respectively).

Figure 5
Panels A-C, [3H]Epibatidine binding to WT α4β2 (A), α4β2S113R (B), and α4β2F119Q (C) nAChRs. Membranes at 0.33 mg protein/mL were incubated for 5 h at room temperature with increasing concentrations of [ ...

3.3 Effect of β2S113R and β2F119Q mutations on [125I]epibatidine photolabeling

Membranes from HEK 293 cells transiently expressing α4β2F119Q, α4β2S113R or WT α4β2 receptors were photolabeled with [125I]epibatidine under equivalent conditions (that is approximately the same amount, 1.5 pmol, of receptor was present in each labeling sample; Figure 6). While the level of [125I]epibatidine labeling is low (3 week exposure), not surprising given the small amount of nAChR (1.5 pmol) present in each sample, we did not detect any 125I cpm associated with the α4 subunit, consistent with previous labeling results with affinity-purified α4β2 and α4β4 nAChRs. There is however clear [125I]epibatidine photoincorporation into the β2F119Q subunit but not the β2S113R or WT β2 subunit. Furthermore, preincubation of α4β2 nAChR membranes with 100 nM epibatidine prior to the addition of [125I]epibatidine completely eliminated labeling in the β2F119Q subunit (Figure 6, + lanes). These results are consistent with the absence of labeling in the β2 subunit (for WT α4β2 nAChRs) resulting from a lack of [125I]epibatidine (and likely 5[125I]A-85380) photoreactivity with β2F119 and with β4Q117 in Loop E being the principal site of [125I]epibatidine labeling in the α4β4 nAChR.

Figure 6
[125I]epibatidine photolabeling of WT and mutant α4β2 nAChRs. Membranes isolated from HEK 293 cells transiently expressing α4β2F119Q, α4β2S113R or WT α4β2 receptors (~1.5 pmol [3 ...

3.4 Effect of β2S113R and β2F119Q mutations on α4β2 nAChR activation

The pharmacological properties of receptors containing mutated β2 subunits were examined by expressing receptors in Xenopus oocytes. Subunit cRNAs were injected at a ratio of 1:8:α4:β2, which results in a preferred subunit stoichiometry of (α4)2(β2)3 [24,25]. Oocytes expressing receptors with α4 paired with any of the β2 constructs respond to applications of ACh (Figure 7A). Activation by ACh is similar for receptors containing wild-type β2, β2S113R or β2F119Q subunits (EC50 values respectively 3.7 ± 0.3 μM (5), 8.8 ± 2.1 μM (5) and 6.7 ± 0.7 μM (5); mean ± SE for fits to data for (N) oocytes; see Figure 7B).

Figure 7
Pharmacological properties of mutated β2 subunits expressed in Xenopus oocytes. Wild-type and mutated β2 subunits were expressed with α4 in Xenopus oocytes. Panel A shows responses of oocytes to 1 mM ACh (solid traces) or 10 μM ...

We then tested the agonist 5I-A-85380. Previous physiological studies of 5I-A-85380 have found that this drug is a potent and efficacious agonist at nicotinic (α4)2(β2)3 receptors [25], with an EC50 about 10 nM and a maximal response larger than that for ACh itself. Our data for receptors containing wild-type β2 are similar (Figure 7C). For receptors containing wild-type β2, the maximal response to 5I-A-85380 is 1.6 ± 0.1 (5) times the response to 1 mM ACh and the EC50 is 23 ± 2 nM (5). Activation of receptors containing the β2S113R subunit is indistinguishable from wild-type (Figure 7C; maximal response ratio 1.7 ± 0.1 (7) and EC50 30 ± 7 nM (7)). However, the presence of the β2F119Q mutation both reduces the maximal relative response and shifts the EC50 to a higher concentration of 5I-A-85380 (Figure 7C; maximal response ratio 1.0 ± 0.0 (5) and EC50 338 ± 77 nM (5)). Both the maximal response ratio and the EC50 differ significantly from values for wild-type or β2S113R subunits (P < 0.02 for response ratio and P < 0.001 for EC50, ANOVA with Bonferroni correction).

4. Discussion

4.1 Photoincorporation of [125I]epibatidine and 5[125I]A-85380 into the agonist binding domain of nAChRs

In this report we introduced the agonists [125I]epibatidine and 5[125I]A-85380 as photoaffinity probes of the agonist binding domain of neuronal nAChRs. It has been well established that iodoepibatidine and 5I-A-85380 bind with high affinity (0.062-24 nM; [10]) to α4β2 and α4β4 nAChRs. We show here with α4β2 and α4β4 nAChRs affinity-purified from stably transfected HEK-293 cells, that irradiation at 254 nm results in selective photoincorporation of [125I]epibatidine and 5[125I]A-85380 into the β4 subunit with little to no labeling of the α4 and β2 subunits. We further show that this labeling is highly specific and is localized to the agonist binding domain in that preincubation with nicotine or epibatidine completely eliminates labeling for either radiolabeled agonist (Fig. 1B; Fig. 2). Selective photoincorporation of [125I]epibatidine and 5[125I]A-85380 into the β4 subunit (no labeling of the α4 and β2 subunits) contrasts with the selective labeling of the α4 subunit (α4β2 AChRs) by azidoepibatidine [13]. On the other hand, azidoepibatidine labeled residues on both the principal (Y195, Loop C) and complementary (M116, Loop E) face of the AChBP [13] and therefore labeling at the subunit level alone is not a predictor of AChR-subtype ligand selectivity. Moreover, the effect that the iodo group (Fig. 5D) may exert on the orientation of the ligand in the binding site and/or the photochemistry must also be considered as well as additional factors such as the photoreactivities of individual amino acids.

Site(s) of [125I]epibatidine labeling in the β4 subunit were next mapped to two overlapping proteolytic fragments that begin at β4V102 and contain the amino acid sequence of Loop E (β4I109 - P120) of the complementary component of the agonist binding site. We were not able to identify amino acids in Loop E that reacted with [125I]epibatidine by protein sequencing as the labeling was not stable under the acidic conditions of automated Edman degradation. We were however able to provide further confirmation that the radioactivity ([125I]epibatidine) was associated with Loop E, that is there are one or more sites of labeling contained within Loop E, by: 1) demonstrating that the electrophoretic mobility of labeled bands which contain Loop E, β4/β4′V8-6.4 (V102-E131) and β4/β4′V8-14.1 (V102-E175), were shifted by endoglycosidase treatment in a manner consistent with the removal of asparagine-linked simple sugar moieties at β4N141 and/or β4N113 (Figure 4); 2) [125I]epibatidine labeling within the Torpedo (muscle-type) nAChR also mapped to a proteolytic fragment containing Loop E of the γ-subunit (Supplementary Fig. S3).

Several lines of evidence led us to then examine β4Q117 and β4R111 in Loop E as likely site(s) of [125I]epibatidine labeling: 1) labeling in the β4 subunit (for α4β4 nAChRs) and the lack of labeling in the β2 subunit (for α4β2 nAChRs) cannot be explained by a lack of binding given the pM binding affinity of iodoepibatidine for α4β2 and α4β4 nAChRs [10]; 2) the high degree of sequence conservation between β2 and β4 subunits (64%; [22]) and in particular the amino acid sequence alignment of Loop E of β2 and β4 subunits (Fig. 3), immediately suggests that labeling differences between α4β2 and α4β4 nAChRs primarily reflect differences in the photoreactivity of a limited number of amino acid residues within Loop E (i.e. positions that differ in β2/β4: β4I109, β4R111, β4S112, β4N113, β4Q117); 3) that azidoepibatidine labeled M116 (equivalent to β4Q117) within Loop E of the Aplaysia AChBP [13] and β4R111 is located on the same face of the β-sheet as β4Q117. To then (indirectly) determine whether or not β4Q117 and/or β4R111 are site(s) of [125I]epibatidine labeling in α4β4 nAChRs, we constructed hα4β2 receptors containing point mutations at β2S113 and β2F119, in which the residue at each position was substituted to the corresponding residue in the rat β4 subunit (β2F119Q and β2S113R, respectively, hereafter). We found that replacement of β2F119, but not β2S113, with the corresponding β4 residues is sufficient to result in [125I]epibatidine photoincorporation similar to that seen in WT α4β4 receptors (Figure 6). These results indicate that the lack of [125I]epibatidine labeling of the WT β2 subunit results from a lack of photoreactivity towards β2F119 and provides indirect evidence that β4Q117 is the major site of [125I]epibatidine photolabeling in α4β4 nAChRs. While additional studies are needed to further map the sites of 5[125I]A-85380 with the β4 subunit, that the receptor subtype labeling was identical to that of [125I]epibatidine and that labeling in the β4 subunit mapped to the same proteolytic fragments β4/β4′V8-6.4 and β4/β4′V8-14.1 (Fig. 2B), strongly suggests that β4Q117 in Loop E is also the principal site of 5[125I]A-85380 labeling. These results are again also consistent with labeling of M116 (equivalent to β4Q117) in the AChBP by azidoepibatidine [13]. The photoreactive azido group of azidoepibatidine is located at the 5 position of the chlorinated pyridine ring (Scheme 1; [13]) and labeling of M116 in the AChBP is consistent with the crystal structure of the epibatidine bound form of AChBP [8]. While it is not possible to predict the precise site(s) of photoreactivity in either [125I]epibatidine or 5[125I]A-85380, substituted pyridines, such as the iodinated pyridine rings of both of these compounds (Fig. 5D), are known to undergo complex photoaddition reactions [26-28]. Therefore photoreaction of the iodinated pyridine ring for either radiolabeled agonist with β4Q117 is fully consistent with these published findings.

4.2 The residue at position β2F119/β4Q117 is an important determinant of nAChR subtype-selectivity for the agonist 5I-A-85380

Having established that β4Q117 in Loop E is the principal site of [125I]epibatidine (and likely 5[125I]A-85380) photoinsertion, the next question is what role does the residue at this position play in the pharmacological properties of the nAChR. While substitutions at β2F119Q and β2S113R had no significant effect on the binding affinity of [3H]epibatidine to nAChRs (Fig. 5A), the β2F119Q substitution resulted in a 4-fold reduction in the potency of 5I-A-85380 for inhibiting the binding of [3H]epibatidine to α4β2F119Q nAChRs (Fig. 5E). These results establish that position β2F119 is important for conferring α4β2 binding selectivity for the agonist 5I-A-85380. However, the 4-fold shift in 5I-A-85380 inhibition potency conferred by the β2F119Q substitution, contrasts with the nearly 200-fold difference in potency for α4β2 vs. α4β4 nAChRs (Fig. 5E). Clearly additional residues in Loops D and E (and perhaps elsewhere) contribute to the receptor subtype-selectivity of 5I-A-85380 and may in fact provide a greater contribution to selectivity than the β2F119/β4Q117 position.

Functional studies have also contributed to our understanding of the role the β2F119 position. The concentration-response curves (Figure 7) indicate the β2F119Q mutation shifts (14-fold) the EC50 to a higher concentration and reduces the maximal relative response to 5I-A-85380, with no effect on the response to ACh. Such a reduction would be expected to cause a shift of the overall relationship to higher concentration, so the increase in EC50 might not directly reflect a reduced affinity for binding to the resting (closed channel) form of the receptor. Similarly, the increase in the IC50 for inhibition of [3H]epibatidine binding (Figure 5E) might reflect either reduced affinity or a reduced ability to cause conformational changes after binding (in this case possibly to the desensitized form of the receptor). The functional data suggest that a major consequence of the β2F119Q mutation might be a destabilization of 5I-A-85380 in the binding pocket after the conformational changes associated with gating. In contrast, the mutation has no significant effect on activation by the less bulky ACh, likely because ACh has less extensive interactions with this region of the binding pocket on the (-) side (complementary face) of the interface [29,30].

Finally, a very recent study has provided evidence that the high affinity of α4β2 nAChRs for ACh and nicotine (and likely for other agonists) results from an enhanced cation-π interaction with Trp149 in loop B [31]. Taken together, these results indicate that in general the α4 subunit confers high-affinity agonist binding but that loop E and in particular β2F119 is an important determinant of the receptor subtype-selectivity of the agonist 5I-A-85380 and likely for other agonists as well.

Supplementary Material

01

Supplementary Figure S1: Photoincorporation of [125I]epibatidine (Panel A) and 5[125I]A-85380 (Panel B) into α4β2 nAChRs. Affinity-purified hα4β2 (~0.3 mg) were photolabeled with [125I]epibatidine or 5[125I]A-85380 (~ 1 nM, 6 μCi) in the absence (-lanes) and presence (+ lanes) of 100 μM nicotine or 100 nM epibatidine respectively then irradiated in glass test tubes at 254 nm for 30 or 15 min respectively. For comparison, affinity-purified α4β2 nAChRs (0.1 mg) were equilibrated for 1 h with [125I]TID (0.4 μM; 10 μCi) then irradiated at 365 nm for 7 min. The polypeptides were resolved by SDS-PAGE. Shown are autoradiographs of 8 % gels containing α4β2 nAChRs photolabeled with [125I]TID (Panel A, lane 1; 2 day exposure), [125I]epibatidine (Panel A, lanes 2 and 3; 4 week exposure), 5[125I]A-85380 (Panel B, lanes 1 and 2; 3 week exposure). The electrophoretic mobility of α4 and β2 subunits and the proteolytic fragment β2′ are indicated.

Supplementary Figure S2: Amino acid sequencing of [125I]epibatidine –labeled β4 subunit. 1 mg of affinity-purified α4β2 nAChR membranes were photolabeled with [125I]epibatidine (28 nM; 440 uCi) and the β4 subunit band was retrieved from the 8% gel and digested in solution with V8 protease. The digest was then fractionated by reversed-phase HPLC and the peak 125I cpm fractions were loaded on PVDF filter using Prosorb® Sample Preparation Cartridges (Applied Biosystems #401959). The filter was treated with Biobrene and sequenced on an Applied Biosystems PROCISE™ 492 protein sequencer. 17% of the eluate of each Edman degradation cycle was used for amino acid identification/quantification, and the other 83% was collected for 125I cpm counting. While the β4V102 peptide was detected (legend at top of figure), the presence of V8 protease peptides precluded an accurate determination of the amount (pmol) of β4 residues present in each cycle of Edman degradation.

Supplementary Figure S3: [125I]epibatidine labeling of Torpedo γ subunit. 3 mg of affinity-purified Torpedo nAChR membranes were photolabeled with [125I]epibatidine (10 nM; 220 uCi) and the γ subunit band was excised from a 8% SDS-PAGE gel and digested in-gel with V8 protease. Panel A, autoradiograph of 15% SDS-PAGE gel (1 week exposure) showing [125I]epibatidine photoincorporation in a γ subunit proteolytic fragment with an apparent molecular mass of 9 KDa (γV8-9K). Panel B, reversed-phase HPLC fractionation of peptides eluted from the labeled band γV8-9K. Panel C, 125I (●) and PTH-amino acids (□) released during sequence analysis of the 125I cpm peak from HPLC fractionation of γV8-9K. Fractions 25-27 were loaded on PVDF filter using Prosorb® Sample Preparation Cartridges (Applied Biosystems #401959) and the filter was treated with Biobrene as recommended by manufacturer. Amino acid sequencing was performed on an Applied Biosystems PROCISE™ 492 protein sequencer. 17% of the eluate of each Edman degradation cycle was used for amino acid determination, and the other 83% was collected for 125I cpm counting. For each peptide detected, the amount of amino acid (f(x), in pmol) in cycle x, determined from the peak height(x) - peak height(x-1), was fit to the equation f(x) = I0Rx to determine the initial amount of the peptide (I0) and the sequencing repetitive yield (R).

Acknowledgments

This research was supported in part by American Heart Association South Central Affiliate Grant-In-Aid 0755029Y (M.P.B.), the South Plains Foundation (M.P.B.), and by United States Public Health Services grant NS-22356 (J.H.S.). JHS is the Russell and Mary Shelden Professor of Anesthesiology. We would like to thank Drs. Jonathan B. Cohen and David C. Chiara (Department of Neurobiology, Harvard Medical School) for performing the amino acid sequencing experiments and for their valuable contributions during the writing of this manuscript. We also thank Sarah Hiyari for technical assistance.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. Gotti C, Clementi CF. Neuronal nicotinic receptors: From structure to pathology. Progress in Neurobiology. 2004;74:363–396. [PubMed]
2. Hogg RC, Raggenbass M, Bertrand D. Nicotinic acetylcholine receptors: From structure to brain function. Rev Physiol Biochem Pharmacol. 2003;147:1–46. [PubMed]
3. Jensen AA, Frolund B, Liljefors T, Krogsgaard-Larsen P. Neuronal Nicotinic Acetylcholine Receptors: Structural Revelations, Target Identifications, and Therapeutic Inspirations. J Med Chem. 2005;48:4705–4745. [PubMed]
4. Corringer PJ, LeNovere N, Changeux JP. Nicotinic receptors at the amino acid level. Annu Rev Pharmacol Toxicol. 2000;40:431–458. [PubMed]
5. Corringer PJ, Galzi JL, Eiselé JL, Bertrand S, Changeux JP, Bertrand D, Gotti C, Clementi F. Neuronal nicotinic receptors: From structure to pathology. Progress in Neurobiology. 2004;74:363–396. [PubMed]
6. Unwin N. Refined structure of the nicotinic acetylcholine receptor at 4Å resolution. J Mol Biol. 2005;346:967–989. [PubMed]
7. Brejc K, van Dijk WJ, Klaassen RV, Schuurmans M, van Der Oost J, Smit AB, Sixma TK. Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature. 2001;411:269–76. [PubMed]
8. Hansen SB, Sulzenbacher G, Huxford T, Marchot P, Taylor P, Bourne Y. Structures of Aplysia AChBP complexes with nicotinic agonists and antagonists reveal distinctive binding interfaces and conformations. EMBO J. 2005;24:3635–46. [PubMed]
9. LeNovere N, Grutter T, Changeux JP. Models of the extracellular domain of the nicotinic receptors and of agonist- and Ca2+-binding sites. Proc Natl Acad Sci USA. 2002;99:3210–5. [PubMed]
10. Xiao Y, Kellar KJ. The comparative pharmacology and up-regulation of rat neuronal nicotinic receptor subtype binding sites stably expressed in transfected mammalian cells. J Pharmacol Exp Ther. 2004;310:98–107. [PubMed]
11. Parker MJ, Beck A, Luetje CW. Neuronal nicotinic receptor beta2 and beta4 subunits confer large differences in agonist binding affinity. Mol Pharmacol. 1998;54:1132–1139. [PubMed]
12. Cohen BN, Figl A, Quick MW, Labarca C, Davidson N, Lester HA. Regions of beta 2 and beta 4 responsible for differences between the steady state dose-response relationships of the alpha 3 beta 2 and alpha 3 beta 4 neuronal nicotinic receptors. J Gen Physiol. 1995;105:745–64. [PMC free article] [PubMed]
13. Tomizawa M, Maltby D, Medzihradszky KF, Zhang N, Durkin KA, Presley J, Talley TT, Taylor P, Burlingame AL, Casida JE. Defining nicotinic agonist binding surfaces through photoaffinity labeling. Biochemistry. 2007;46:8798–8806. [PMC free article] [PubMed]
14. Kishi M, Steinbach JH. Role of the agonist binding site in up-regulation of neuronal nicotinic alpha4beta2 receptors. Mol Pharmacol. 2006;70:2037–2044. [PubMed]
15. Paradiso K, Zhang J, Steinbach JH. The C terminus of the human nicotinic alpha4beta2 receptor forms a binding site required for potentiation by an estrogenic steroid. J Neurosci. 2001;21:6561–6568. [PubMed]
16. Zhang J, Steinbach JH. Cytisine binds with similar affinity to nicotinic α4β2 receptors on the cell surface and in homogenates. Brain Research. 2002;959:98–102. [PubMed]
17. Hamouda AK, Sanghvi M, Chiara DC, Cohen JB, Blanton MP. Identifying the lipid-protein interface of the alpha4beta2 neuronal nicotinic acetylcholine receptor: hydrophobic photolabeling studies with 3-(trifluoromethyl)-3-(m-[125I]iodophenyl) diazirine. Biochemistry. 2007;46:13837–13846. [PMC free article] [PubMed]
18. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–685. [PubMed]
19. Cleveland DW, Fischer SG, Kirschner MW, Laemmli UK. Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel electrophoresis. J Biol Chem. 1977;252:1102–1106. [PubMed]
20. Schagger H, von Jagow G. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem. 1987;166:368–379. [PubMed]
21. Li W, Jin X, Covey DF, Steinbach JH. Neuroactive steroids and human recombinant rho1 GABAC receptors. J Pharmacol Exp Ther. 2007;323:236–247. [PubMed]
22. Boulter J, O'Shea-Greenfield A, Duvoisin RM, Connolly J, Wada E, Jensen A, Gardner PD, Ballivet M, Deneris E, McKinnon D, Heinemann S, Patrick J. α3, α5, and α4: Three members of the rat neuronal nicotinic acetylcholine receptor-related gene family form a gene cluster. J Biol Chem. 1990;265:4472–4482. [PubMed]
23. Prince RJ, Sine SM. Epibatidine binds with unique site and state selectivity to muscle nicotinic acetylcholine receptors. J Biol Chem. 1998;273:7843–7849. [PubMed]
24. Nelson ME, Kuryatov A, Choi CH, Zhou Y, Lindstrom J. Alternate stoichiometries of alpha4beta2 nicotinic acetylcholine receptors. Mol Pharmacol. 2003;63:332–341. [PubMed]
25. Zwart R, Broad LM, Xi Q LM, Moroni M, Bermudez I, Sher E. 5-I A- 85380 and TC-2559 differentially activate heterologously expressed alpha4beta2 nicotinic receptors. Eur J Pharmacol. 2006;539:10–17. [PubMed]
26. Caplain S, Castellano A, Catteau JP, Lablache-Combier A. Etudes photochimiques-VI: Mechansime de la photosubstitution de la pyridine en solution. Tetrahedron. 1971;27:3541–3553.
27. Whitten DG. Photoreduction and photoaddition reactions of heterocyclic compounds. In: Buchardt O, editor. Photochemistry of Heterocyclic Compounds. New York: Wiley; 1976. pp. 524–573.
28. Singh SP, Stenberg VI, Parmar SS. Photochemistry of alkaloids. Chem Rev. 1980;80:269–282.
29. Williamson PTF, Verhoeven A, Miller KW, Meier BH, Watts A. The conformation of acetylcholine at its target site in the membrane-embedded nicotinic acetylcholine receptor. Proc Natl Acad Sci USA. 2007;104:18031–18036. [PubMed]
30. Celie PHN, van Rossum-Fikkert SE, van Dijk WJ, Brejc K, Smit AB, Sixma TK. Nicotine and carbamylcholine binding to nicotinic acetylcholine receptors as studied in AChBP crystal structures. Neuron. 2004;41:907–914. [PubMed]
31. Xiu X, Puskar NL, Shanata JP, Lester HA, Dougherty DA. Nicotine binding to brain receptors requires a strong cation-π interaction. Nature. 2009;458(7237):534–537. [PMC free article] [PubMed]