QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.084954v1 314/1/455    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kuo, Y.-P. Articles by Lukas, R. J. Search for Related Content PubMed PubMed Citation Articles by Kuo, Y.-P. Articles by Lukas, R. J.

NEUROPHARMACOLOGY

Roles for Nicotinic Acetylcholine Receptor Subunit Large Cytoplasmic Loop Sequences in Receptor Expression and Function

Division of Neurobiology (Y.-P.K., L.X., J.B.E., R.J.L.) and Neurology (L.Z., J.W.), Barrow Neurological Institute, Phoenix, Arizona

Received for publication February 16, 2005
Accepted April 13, 2005.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
To evaluate possible physiological roles of the large cytoplasmic loops (C2) and neighboring transmembrane domains of nicotinic acetylcholine receptor (nAChR) subunits, we generated novel fusion constructs in which human nAChR 4, 2, or 4 subunit C2 or C2 and neighboring sequences were replaced by corresponding sequences from the mouse serotonin type 3A (5-HT_3A) receptor subunit. Following stable expression in human SH-EP1 cells, we found that extensive sequence substitutions involving third and fourth transmembrane domains and neighboring "proximal" C2 sequences (e.g., 2 H322-V335 and V449-R460) did not allow functional expression of nAChR containing chimeric subunits. However, expression of functional nAChR was achieved containing wild-type 4 subunits and chimeric 2 (2) subunits whose "nested" C2 domain sequences K336-S448 were replaced with the corresponding 5-HT_3A subunit sequences. Whereas these findings suggested indispensable roles for M3/M4 transmembrane and/or proximal C2 sequences in 42-nAChR function, nested C2 sequences in the 2 subunit are not essential for functional receptor expression. Ligand-binding analyses also revealed only subtle differences in pharmacological profiles of 42-nAChR compared with 42-nAChR. Nevertheless, there was heightened emergence of agonist-mediated self-inhibition of 42 function, greater sensitivity to functional blockade by a number of antagonists, and faster and more complete acute desensitization of 42-nAChR than for 42-nAChR. These studies are consistent with unexpected roles of nested C2 sequences in nAChR function.

Each of the 17, genetically distinct, vertebrate nAChR subunits identified to date share a common topology containing a large extracellular N-terminal domain, four transmembrane domains, a short cytoplasmic domain between the first and second transmembrane segments, a short extracellular domain between second and third transmembrane segments, a large second cytoplasmic loop (C2) situated between the third (M3) and fourth (M4) transmembrane domains, and a short C-terminal extracellular tail. The N-terminal domain contains key elements for ligand-binding/recognition (Sine, 2002), and the transmembrane domains anchor the proteins in the plasma membrane and contribute to channel kinetics and ion selectivity (Corringer et al., 2000). These structural domains are well conserved among different subunits and have been studied extensively. On the other hand, the less studied C2 domain of each subunit contains unique sequences that are distinguishing fingerprints for each subunit. These C2 domains have been suggested to play potential roles in the regulation of nAChR trafficking (Williams et al., 1998), mediation of cytoskeletal interactions (Bencherif and Lukas, 1993; Colledge and Froehner, 1997; Shoop et al., 2000), receptor assembly (Yu and Hall, 1994b), functional desensitization of nAChR (Fenster et al., 1999), and as targets of phosphorylation (Yu and Hall, 1994a; Colledge and Froehner, 1997), perhaps affecting and being affected by intracellular signaling cascades. However, aside from studies focused on homomeric 7-nAChR (Valor et al., 2002), examination of functional roles of C2 has been scant.

5-Hydroxytryptamine (serotonin) type 3 (5-HT₃) receptors also are members of the four transmembrane domain, ligandgated ion channel superfamily, and the relevant subunits share topological features of nAChR subunits. Receptors made of 5-HT_3A subunits can form functional channels as homomers, although combination with a newly identified 5-HT_3B subunit allows assembly of heteromers that have altered biophysical properties relative to 5-HT_3A receptor homomers (Brady et al., 2001). Nevertheless, recombinant heteromers display pharmacological features, large single-channel conductance, and low calcium permeability like those of native 5-HT₃ receptors (Davies et al., 1999). Chimeric subunit models constructed from partial replacement of nAChR subunit sequences with 5-HT_3A subunit sequences have been used to advantage in assessing nAChR structure-function relationships. For example, recombinant chimeric subunits containing the N-terminal domain of nAChR 7 (Eisele et al., 1993) or 8 (Cooper and Millar, 1998) subunits fused to more C-terminal regions of the 5-HT_3A receptor subunit (that include all four of the transmembrane segments, the intervening cytoplasmic and extracellular domains, and the C-terminal domain) have been successfully used in studies aimed to reveal regions of physiological significance engaged in toxin- or ligand-binding or in ion channel selectivity. Nevertheless, the ability of 5-HT_3A receptor subunits to form homomers and their topological similarities to nAChR subunits suggest that they also would be good fusion partners for studies of other domains in the subunit superfamily.

To evaluate possible physiological roles of C2 and neighboring transmembrane domains of nAChR subunits, we generated novel fusion constructs in which human nAChR 4, 2, or 4 subunit C2 sequences or both C2 and neighboring transmembrane sequences were replaced by corresponding sequences from the mouse 5HT_3A receptor subunit. We find evidence for different roles in nAChR assembly and function of M3 and M4 transmembrane domains and immediately flanking "proximal" C2 sequences (i.e., C2 domain residues just C-terminal to M3 or N-terminal to M4 domains; see Fig. 1) or for "nested" C2 sequences (i.e., C2 domain residues between the conserved proximal residues immediately flanking M3 and M4 transmembrane domains; see Fig. 1). A preliminary report of some of these findings has appeared (Kuo et al., 2002).

View larger version (57K): [in this window] [in a new window]   Fig. 1. A, strategy for construction of the chimeric nAChR 2 subunit and schematic diagram illustrating wild-type and chimeric nAChR subunit protein structures. The four putative transmembrane domains M1–M4 are displayed by cross-hatched boxes. Open boxes represent the three extracellular domains, E1, E2, and the C-terminal E3, the small C1 cytoplasmic loop, and the large C2 cytoplasmic domain located between M3 and M4 regions. The regions indicated by roman numerals and arrows above the block schematic drawing indicate protein sequence regions coded by the PCR fragments that were generated and ligated to form chimeric subunit cDNA with fragments I and III derived from the human nAChR 2 subunit sequence and fragment II derived from the murine 5-HT3A subunit sequence as described in detail under Materials and Methods. The gray area in the chimeric subunit drawing marks regions of nAChR subunits replaced by the 5-HT3A sequences. The FLAG sequence motif (LEDYKDDDK) is represented by . B, alignment for human nAChR 4, 2, 4, and 7 subunits and human or mouse 5-HT receptor 3A subunits (amino acid numberingfor all starting at the translation initiation methionine; single letter code) for sequences beginning at the start of the M3 domain and ending at the C terminus. M3 and M4 transmembrane domains are indicated, as are proximal C2 domain sequences close to M3 or M4 and the intervening nested C2 sequence between those proximal C2 sequences (italicized, boldface type above the relevant sequences). Amino acids in the nAChR 4, 2, or 4 subunit sequences indicated in boldface are among those substituted with 5-HT3A subunit sequences and characterized in the study. Amino acids in the human nAChR 7 subunit sequence in boldface or italics aid identification of regions mentioned under Discussion that correspond to those suggested to be essential or not essential, respectively, for function of rat 7-nAChR as studied by Valor et al. (2002). Note the similarities across some nAChR subunits for regions corresponding to M3, proximal C2, and M4 domains but the differences compared with 5-HT3A subunits (differences are especially marked when comparing across nAChR, 5-HT3, GABA, and ionotropic glycine receptor subunit families). The nested C2 sequences are absolutely unique to each subunit, although the high similarity and identity between human and murine 5-HT3A subunits in this region is also evident.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

Construction and Subcloning of Human nAChR/Mouse 5-HT_3A Subunit Chimeric cDNAs. Human nAChR 4 subunit cDNA was excised as a KpnI-XhoI fragment from a pcDNA3.1-zeo (Invitrogen)-based expression construct and subcloned into the pCEP4 (Invitrogen) plasmid vector, and human nAChR 2 or 4 subunit cDNAs were subcloned into pcDNA3.1-zeo or pcDNA3.1-hygro as described (Eaton et al., 2003). The mouse 5-HT_3A subunit cDNA carried by pcDNA1 vector was a gift from Dr. Philippe Séguéla (Montreal Neurological Institute, Montreal, Quebec, Canada). Similar strategies employing successive polymerase chain reactions (PCRs) were used for generation of human nAChR 4, 2, or 4/mouse 5-HT_3A subunit chimeric cDNAs. Details are provided here for construction of the nAChR 2/5-HT_3A subunit chimera studied [see Fig. 1 for the schematic illustration of the constructs (Fig. 1A) and their sequences (Fig. 1B)]. First, three partial chimeric cDNA fragments were synthesized: one coding for the N terminus to the 335th amino acid residue (counting from the initiation methionine; M1-V335) of the human nAChR 2 subunit, another coding for part of the putative C2 of the mouse 5-HT_3A subunit (amino acids R347-A448), and the third coding for amino acid residues 449 to the C terminus of the human nAChR 2 subunit (V449-K502; Fig. 1A; fragments I, II, and III). Chimeric PCR primers were designed so that the amplified partial cDNA fragments carried 18 or 17 bp of sequence overlap at the intended nAChR 2/5-HT_3A cDNA sequence junctions. Sequences of the primers are as follows: sense 5'-taatacgactcactataggg-3' (located within the pcDNA3.1-hygro T7 promoter) and chimeric antisense 5'-gtctaggaccaggtgcctcacccagggcgccatgg-3' (italicized fonts, 5-HT_3A sequence; regular font, nAChR 2 sequence) for fragment I; chimeric sense 5'-ccatggcgccctgggtgaggcacctggtcctagac-3' and chimeric antisense 5'-acgtacttccagtcctctgccacctcccgcatctc-3' for fragment II; chimeric sense 5'-gagatgcgggaggtggca-gaggactggaagtacgt-3' and antisense 5'-tagaaggcacagtcgagg-3' (3' to the multiple cloning site in pcDA3.1-hygro vector) for fragment III. Primary PCRs were carried out using 10 ng of template DNA (e.g., pcDNA3.1-hyrgo-h2), 10 pmols of sense and antisense primers, and 2.5 units of Platinum TaqDNA polymerase (Invitrogen) in 50-µl reactions for 30 cycles at 94°C for 1 min each, 55°C for 90 s, and 72°C for 90 s, followed by a 4-min extension at 72°C to generate fragments with complementary overlaps. Fragments II and III were then ligated in the secondary PCR followed by gel purification (Prep-A-Gene; Bio-Rad, Hercules, CA), and a tertiary PCR was used to fuse fragments I and II/III to complete the chimeric cDNA. The secondary and tertiary PCRs used approximately 100 ng of a mixture of the two DNA templates with one template in an approximately 10-fold molar excess over the other. For each secondary and tertiary PCR, the first five cycles of reaction (94°C for 1 min 30 s, 55°C for 90 s, and 72°C for 2 min) were carried out in the absence of PCR primers, allowing extension of ligated template fragments, followed by addition of primers selecting for the extended DNA templates in the next 30 cycles of amplification reactions. The final chimeric cDNA [2, 2(M1-V335)-5-HT_3A(R347-A448)-2(V449-K502)] was then digested with EcoRI and XbaI and subcloned into the pCDNA3.1-zeo plasmid and their sequences verified. Similar strategies were used to generate cDNAs for other chimeric subunits [2(M3-E3-FLAG), 2(M1-V294)-5-HT_3A(P306-S487)-LEDYKDDDK including the indicated C-terminal FLAG-tag; 4(M3-E3-FLAG), 4(M1-V292)-5-HT_3A(P306-S487)-LEDYKDDDK; 4(M3-E3-FLAG), 4(M1-I303)-5-HT_3A(P306-S487)-LEDYKDDDDK; 2(M3-E3), 2(M1-V294)-5-HT_3A(P306-S487); 4(M3-E3), 4(M1-V292)-5-HT_3A(P306-S487); 2(nC2-E3), 2(M1-V335)-5-HT_3A(R347-S487); illustrated in Fig. 1A], which were then digested with appropriate restriction enzymes and cloned into pCDNA3.1-zeo (2 or 4 wild-type or chimeric subunits) or pCEP4-hygro (4 wild-type or chimeric subunits) followed by sequence verification.

Model Cell Lines, Cell Culture, and Transfection. The native nAChR-null SH-EP1 cell line (Lukas et al., 1993) was used as the host for generating model cell lines stably expressing nAChR composed of wild-type or chimeric subunits. SH-EP1 cells were grown in Dulbecco's modified Eagle's medium (high glucose, bicarbonate-buffered with 1 mM sodium pyruvate and 8 mM L-glutamine) supplemented with 10% horse serum, 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.25 µg/ml amphotericin B plus 5% fetal bovine serum on 100-mm diameter plates in a humidified atmosphere containing 5% CO₂ in air at 37°C. The SH-EP1-h42-nAChR cell line expressing functional and radioligand binding human 42-nAChR was generated earlier in our laboratory (Eaton et al., 2003). For constructing SH-EP1-h42 or other transfected cell lines, low-passage (less than 15) SH-EP1 host cells were grown to approximately 50% confluence on 100-mm plates on the day of transfection and transfected with 10 µg of pCEP4 containing human wild-type 4 subunit DNA or pcDNA3.1-hygro containing chimeric 4 subunit cDNA using Superfect (QIAGEN, Valencia, CA) transfection reagent following the manufacturer's protocol. Forty-eight hours after transfection, the cells were split into four 100-mm plates, and the selection process began by supplementing the culture medium with hygromycin B (130 µg/ml; Calbiochem, San Diego, CA). Two weeks later, polyclonal SH-EP1-h4 or SH-EP1-h4(M3-E3-FLAG) cells selected by hygromycin-resistance were then pooled and used as host cells for secondary transfection with pCDNA3.1-zeo construct containing wild-type or chimeric 2 or 4 subunit cDNAs, and zeomycin and hygromycin dual-drug resistant monoclonal cells were isolated. To obtain monoclonal cells, the transfected cells were split 1:20 onto fresh 100-mm plates to start dual drug selection 2 days after the transfection. Growth of the cells was monitored until single and well isolated colonies (20–30 colonies per plate) reached the size of approximately 2 to 3 mm in diameter. The clones were then transferred to a fresh dish using cloning discs (Fisher Scientific, Tustin, CA), expanded, and then screened for function using ⁸⁶Rb⁺ efflux assays (see below). The candidate transfected cell lines identified by their ability to exhibit ⁸⁶Rb⁺ efflux in response to nicotinic agonists were further verified for their wild-type or chimeric 4 and 2 or 2 subunit RNA expression by reverse transcription (RT)-PCR, split once weekly, and maintained in low-passage (less than 26) cultures to ensure stable expression of phenotype. In some cases, ³H-labeled epibatidine (H-EBDN) binding assays (see below and Eaton et al., 2003) were also done to evaluate expression of nAChR containing wild-type or chimeric subunits as radioligand binding entities.

RNA Preparation, RT, and Polymerase Chain Reaction. To isolate total RNA from wild-type or transfected SH-EP1 cell lines, 2 ml of TRIzol reagent (Invitrogen) was added to cells growing at approximately 80% confluence in a 100-mm dish. RNA was then immediately isolated, precipitated, washed, and resuspended in RNase-free water as described by the manufacturer. Prior to the RT-PCR experiment, the RNA preparations were treated with RNase-free DNase I (Ambion, Austin, TX) to remove residual genomic DNA contamination, and the added DNaseI was inactivated at 65°C for 10 min following addition of 25 mM EDTA. For first-strand cDNA synthesis, we used 2 µg of the DNA-free total RNA, oligo(dT) primer, and the Superscript II preamplification system (Invitrogen). At the end of the RT reaction, reverse transcriptase was deactivated by incubating the reaction at 75°C for 10 min, and RNA was removed by adding 1 unit of RNaseH to the mixture followed by incubation at 37°C for 30 min. An RT-negative control was also carried out in the absence of reverse transcriptase to check for residual genomic DNA contamination in the RNA samples. Each downstream PCR was performed using 1/20 of cDNA template, 1 µl of each 10 µM sense and antisense gene-specific primers, 1 µl of 10 mM dNTP, and 2.5 units of RedTaq (Sigma-Aldrich) in a 50-µl reaction. The amplifications were carried out for 35 cycles at 95°C for 1 min, 55°C for 90 s, and 72°C for 90 s, followed by a 4-min extension at 72°C. PCR primers used in these reactions are: sense 5'-cgtattgggcgcctggtcaccag-3', antisense 5'-gtccttgcccacagccttggcagc-3' for GAPDH used as a positive control (predicted product size of 624 bp); sense 5'-gaatgtcacctccatccgcatc-3', antisense 5'-ccggca(a/g)ttgtc-(c/t)ttgaccac-3' (a human-rodent "universal" primer) for the human nAChR 4 subunit (predicted product size of 790 bp); sense 5'-cggctcccttccaaacaca-3', antisense 5'-gcaatgatggcgtggctgctgca-3' for the nAChR 2 subunit (predicted product size of 754 bp); and sense 5'-ccatggcgccctgggtgaggcacctggtcctagac-3' (chimeric primer), antisense 5'-tagaaggcacagtcgagg-3' (3' to the multiple cloning site in pcDNA-3.1hygro) for the nAChR 2 chimera subunit (predicted product size of 420 bp). RT-PCR products were electrophoretically resolved on 1% agarose gel containing ethidium bromide, and digital photography under ultraviolet illumination was used to document results.

Immunoprecipitation and Western Blotting Analysis. Preparation of solubilized membranes for immunoprecipitation began with medium removal from 10-mm dishes harboring cells at confluence, rinsing of dishes three times with sodium phosphate buffer (100 mM NaCl, 25 mM NaPO₄, pH 7.4), mechanical harvesting of cells, and centrifugation of healthy cells at 1000g for 5 min. The cell pellets were then resuspended and homogenized in ice-cold sodium phosphate buffer (200 µl per confluent plate) supplemented with Complete Protease Inhibitor cocktail (1 mini tablet per 10 ml; Roche Diagnostics, Indianapolis, IN) followed by centrifugation at 10,000g for 10 min at 4°C. After centrifugation, the supernatant was discarded and the pellet resuspended in sodium phosphate buffer supplemented with 1% Triton X-100 (200 µl per ml; supplemented with protease inhibitor cocktail) by incubating at room temperature for 30 min. The preparation was then centrifuged at 12,500g for 10 min at 4°C. The supernatant fraction containing the solubilized membrane protein was collected, and the receptor protein was immunoprecipitated using antibody H133 (sc-5591; rabbit anti-nAChR 4 subunit targeting amino acids 342–474) or antibody H92 (sc-11372; rabbit anti-nAChR 2 subunit targeting amino acids 342–433) antibodies (both from Santa Cruz Biotechnology, Santa Cruz, CA; 5 µg per 1 mg of solubilized total membrane protein) and protein G (reactive with IgG from rabbits or many other species)-agarose beads (50 µl; Calbiochem). After an overnight incubation at 4°C, the mixture was washed three times in detergent-supplemented sodium phosphate buffer, resuspended in sodium dodecyl sulfate-polyacrylamide gel electrophoresis sample buffer containing -mercaptoethanol (20 µl per sample; MP Biomedicals, Irvine, CA), heated to 95°C for 3 min, and centrifuged briefly to remove the agarose beads. The protein samples were then subjected to 10% SDS-polyacrylamide gel electrophoresis separation and transferred to nitrocellulose membrane for Western analysis. Membranes were blocked in phosphate-buffered saline containing 5% skim milk powder and incubated with primary antibodies: mAb 299 (rat anti-nAChR 4 subunit targeting the N-terminal extracellular domain; 1:1000; Sigma-Aldrich) and/or polyclonal antisera s-1724 (rabbit anti-nAChR 2 subunit targeting the N-terminal extracellular domain; 1:50; a gift from Dr. J. Patrick of the Baylor College of Medicine, Houston, TX) for 1 h. After three 5-min washes in phosphate-buffered saline containing 0.05% Tween 20, horseradish peroxidase-conjugated anti-rat and/or anti-rabbit secondary antibodies (1:1000; Santa Cruz Biotechnology) were added to the reaction. The antibody-labeled bands were visualized by using enhanced substrate (Opti-4CN; Bio-Rad). For quantification, the blots were scanned, and the signal intensity was analyzed by using Kodak 1-D gel analysis software (Kodak IBI, New Haven, CT).

Epibatidine-Binding Competition Studies. Membrane preparations for binding studies were prepared according to our previously described method (Lukas et al., 2002; Eaton et al., 2003). Briefly, transfected SH-EP1 cells were mechanically dislodged using a polypropylene policeman, pelleted by low-speed centrifugation, and resuspended in 3 ml of ice-cold 5 mM Tris (pH 7.4). The cells were then homogenized with a Polytron (45 s; Brinkmann model 10/35 with a PTA10S generator; Brinkman Instrument, Westbury, NY), and homogenized membranes were centrifuged at 45,000g for 10 min at 4°C and washed twice in 6 ml of Ringer's solution supplemented with 0.1 mg/ml sodium azide before being resuspended in the same buffer. The total membrane protein was quantified using the BCA protein assay (Pierce, Rockford, IL) according to the manufacturer's instructions. Typically, 0.24 mg of membrane protein could be obtained from a confluent 100-mm dish culture, and 1 to 3 µg of protein yielded 25 fmol of binding sites. For H-EBDN (PerkinElmer Life and Analytical Sciences, Boston, MA) binding competition assays (Eaton et al., 2003), 800-µl reactions consisting of 400 pM H-EBDN competing ligand at various concentrations and an aliquot of membrane preparation containing 25 fmol binding sites were incubated for 2 h at room temperature. Reaction mixtures were then filtered with an Inotech 1H-201-A sample processor (Inotech Biosystems, Rockville, MD) using glass fiber filters with 1.0 to 1.5 µm of retention and pretreated with 0.2% polyethylenimine. After three rinses, filters were transferred to 96-well plates and quantified for radioligand binding by scintillation counting (Wallac Microbeta Trilux 1450; PerkinElmer Life and Analytical Sciences). Data for binding, ion flux, and electrophysiological studies (see below) were plotted, analyzed, and tested for statistical significance using Prism (GraphPad Software, Inc., San Diego, CA).

Assay of nAChR Function by ⁸⁶Rb⁺ Efflux. Function properties of nAChR channels in the model cell lines were measured by using ⁸⁶Rb⁺ efflux assays with the "flip-plate" technique developed in our laboratory (Lukas et al., 2002; Eaton et al., 2003). Briefly, cells grown to confluence on two 100-mm plates were harvested by mild trypsinization, resuspended in complete medium, and seeded onto a 24-well plate. After cells had adhered overnight, medium was removed and replaced with 250 µl of complete medium supplemented with 300,000 cpm of ⁸⁶Rb⁺ (PerkinElmer Life and Analytical Sciences) per well. Following a minimum of a 4-h incubation, cells in each well were rinsed three times with 2 ml of efflux buffer (130 mM NaCl, 5.4 mM KCl, 2 mM CaCl₂, 5 mM glucose, 50 mM HEPES, pH 7.4) and introduced to drugs of choice at indicated final concentrations in efflux buffer for a 3-min incubation, all utilizing the flipplate technique. The drug solution was then "flipped" back into the efflux/drug plate and subjected to Cerenkov counting (Wallac Microbeta Trilux 1450; PerkinElmer Life and Analytical Sciences). Normalization and quality control for each experiment were accomplished using measurements of nonspecific ⁸⁶Rb⁺ efflux in samples containing efflux buffer alone (background subtraction; 0% of control) and of total ⁸⁶Rb⁺ efflux in samples containing a fully efficacious concentration of 1 mM carbamylcholine (100% of control). Specific ⁸⁶Rb⁺ efflux for each drug concentration was expressed as a percentage of specific ⁸⁶Rb⁺ efflux for 1 mM carbamylcholine.

Patch-Clamp Whole-Cell Current Recordings and Data Analysis. Conventional whole-cell current recordings combined with use of a U-tube for rapid drug application have been previously described (Wu et al., 2002). Briefly, cells plated on 35-mm culture dishes were continuously superfused with standard external solution (120 mM NaCl, 3 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 25 mM D-glucose, and 10 mM HEPES, pH 7.4, with Tris base; 2 ml/min). Glass microelectrodes (3–5 M resistance between pipette and extracellular solutions) filled with solution containing 140 mM KCl, 4 mM MgSO₄, 0.1 mM EGTA, 4 mM Na-ATP, and 10 mM HEPES with Tris base (pH 7.2) were used, and cells were voltage-clamped at holding potentials of -60 mV. To induce whole-cell current responses, 1 µM nicotine was delivered into the bath medium in close proximity to the targeted cells via the U-tube system, and ionic currents were measured using a 200B amplifier (Axon Instruments, Union City, CA). Both pipette and whole-cell current capacitance was minimized, and the series resistance was routinely compensated to 80%. Whole-cell access resistance less than 20 M was accepted. The interval between drug applications was 3 min, which was adjusted specifically to eliminate receptor functional rundown. All experiments were performed at room temperature (22 ± 1°C). Data were typically filtered at 2 kHz and acquired at 5 kHz using Pclamp8 (Axon Instruments) and were displayed and digitized on-line (Axon Instruments; Digidata 1200 series A/D board). The results were plotted using Origin 5.0 software (OriginLab Corp., Northampton, MA). The decline in whole-cell current amplitude from peak to steady-state values during agonist application was fit to a single exponential decay function [current = (peak current x e^(t/)) + steady-state current; Clampfit 8.0; Axon Instruments] using either the data from 90 to 10% of the peak current amplitude when it decays to a zero steady-state current or the data from the middle 80% of the duration of drug application. These fits allowed determination of the decay constant, [i.e., the time required for an e-fold reduction (to 37% of the peak current amplitude) in whole-cell current amplitude], even when extrapolating beyond the actual data if current amplitude did not fall during agonist exposure to 37% of peak amplitude.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Generation and Initial Characterization of SH-EP1 Cell-Based Model Cell Lines. C2 domain sequences proximal to M3 and M4 transmembrane domains are highly conserved across nAChR subunits but diverge from proximal C2 domains of analogous sequences in serotonin 5-HT_3A receptor subunits (Fig. 1B) or GABA_A or glycine receptor subunits from the same four-transmembrane domain, ligand-gated ion channel superfamily, thus distinguishing nAChR subunits as a subfamily. Perhaps proximal C2 sequences constitute radiating tunnels from the central channel suggested by Unwin's images (Miyazawa et al., 1999), through which ion flux could occur, forming additional structures that could help define channel kinetics and ion permeability. Matching of proximal C2 sequences across subunits within a closed assembly may be required for formation of functional ligand-gated ion channels. On the other hand, nAChR subunit C2 domain sequences nested between the conserved, proximal domains (Fig. 1B) are absolutely unique, nAChR subunit-specific fingerprints that nevertheless must carry signatures for interactions of nAChR with cytoplasmic or cytoskeletal proteins. These sequences may be involved in any downstream signal transduction through nAChR (Dajas-Bailador et al., 2002; Shaw et al., 2002). They also may be important in trafficking of nAChR through precursor pools and to the cell surface (Williams et al., 1998) and perhaps in recycling at the plasma membrane. In addition, post-translational modification of cytoplasmic domain residues, such as subunit phosphorylation, may help regulate subunit assembly into pentamers, trafficking of nAChR, and/or functional activation/inactivation of nAChR (Huganir and Greengard, 1990; Fenster et al., 1999; Guo and Wecker, 2002; Wecker and Rogers, 2003). However, few previous studies have investigated structure and function of nAChR subunit C2 sequences and neighboring transmembrane domains.

To critically examine the physiological role of the putative C2 domain and neighboring transmembrane sequences of nAChR subunits, we generated chimeric subunits based on the human nAChR 4, 2, or 4 subunit backgrounds, substituting selected sequences from the mouse 5-HT_3A subunit analog (Fig. 1). We reasoned that the ability of the 5-HT_3A subunit to assemble into a functional homomer and the relatively short length of its C2 domain minimized the likelihood that substitution with its sequences would hinder coassembly of chimeric subunits with other wild-type or chimeric nAChR subunits. We thought that substitutions with 5-HT_3A sequences would be more likely to produce properly folded subunits than would substitutions with, for example, green fluorescent protein sequences. We also anticipated that we could create a library of chimeras of different nAChR subunits, having in common the 5-HT_3A subunit C2 signature, allowing clearer interpretation of effects of those substitutions on nAChR expression and function.

In our initial studies, we found that nAChR chimeric 4 subunits containing 5-HT_3A subunit sequences spanning from the beginning of M3 through all of C2 and including M4 and C-terminal extracellular domains, as well as a C-terminal FLAG-tag [4(M3-E3-FLAG); Fig. 1], were unable to assemble with wild-type 2 or 4 subunits to form stably functional nAChR in appropriately transfected SH-EP1 cells (data not shown). Some specific binding of H-EBDN to presumptive 4(M3-E3-FLAG)2-nAChR in membrane preparations was observed (data not shown). Similarly, 2(M3-E3-FLAG) or 4(M3-E3-FLAG) subunits in combination with wild-type 4 subunits could not form functional nAChR, although small amounts of specific H-EBDN binding to presumptive 42(M3-E3-FLAG)-nAChR were evident (data not shown). Doubly chimeric nAChR containing 4(M3-E3-FLAG) subunits plus 2(M3-E3), 4(M3-E3), 2(M3-E3-FLAG), or 4(M3-E3-FLAG) subunits failed to form functional nAChR (data not shown). Chimeric 2(nC2-E3) subunits contained a slightly shorter stretch of 5-HT_3A subunit sequences beginning 14 amino acids C-terminal to M3 and extending through the rest of the C2, M4, and E3 domains (Fig. 1). These 2(nC2-E3) subunits were able to combine with chimeric 4(M3-E3-FLAG) subunits to form some H-EBDN binding sites, but not functional nAChR, and no nAChR-like function was observed in cells expressing chimeric 2(nC2-E3) and wild-type 4 subunits (data not shown).

A different outcome was obtained when an alternative sequence substitution scheme was used to generate a chimera based on the human nAChR 2 subunit. Rather than the substitution of most or all of the C2 sequence, the M4 (+/-M3) domain and the short C-terminal E3 tail, as was done in generation of [M3-E3(+/-FLAG)] or (nC2-E3) chimeras, the natural 2 subunit sequences in M3, M4, and E3 domains were retained along with C2 domain sequences proximal to M3 and M4, with only nested C2 sequences bounded by the proximal C2 sequences exchanged (Fig. 1). That is, the 5'- and 3'-"switch sites" of the C2 domain in the resultant 2 gene were located at sequences encoding amino acids K336-S448 (making reference to the translation initiation methionine; Fig. 1), thus, preserving the relatively conserved, proximal linkers of 14 or 12 residues situated near M3 or M4, respectively.

After transfection of SH-EP1 cells with 2 subunits along with either wild-type 4 subunits or chimeric 4(M3-E3-FLAG) subunits, candidate clonal lines, isolated by hygromycin- and zeocin-resistance (hyg^R/zeo^R), were expanded and assessed for channel function in response to nicotine using ⁸⁶Rb⁺ ion efflux assays. Cells expressing 2 and chimeric 4(M3-E3-FLAG) subunits were positive for H-EBDN binding, but not for functional responses to nicotinic agonists (data not shown). However, a clone displaying stable nicotine concentration-dependent functional responses was obtained from cells transfected with 2 and wild-type 4 subunits, and RT-PCR analysis confirmed expression of transgenes (Fig. 2). We have previously shown that SH-EP1 human epithelial cells do not express any endogenous nAChR subunits detectable by Northern analysis or any radioligand binding or functional sites corresponding to nAChR (Lukas et al., 1993). Results of the more sensitive RT-PCR approach used in the present study again confirmed the absence of nAChR 4 and 2 subunit messages in wild-type SH-EP1 cells, whereas expression of 4 and 2 subunit messages in the transfected SH-EP1 clone displaying ⁸⁶Rb⁺ ion efflux in response to nicotine was verified (Fig. 2). For reasons that were not investigated, SH-EP1-h42 cells did not grow to confluence as did SH-EP1-h42 cells and tended to lift off from the plate once the cell culture became confluent. Nevertheless, this SH-EP1-42-nAChR clone assumed the normal SH-EP1 cell morphology and replicated at the same rate as SH-EP1-42 cells generated earlier in our laboratory, having a doubling time of approximately 30 h.

View larger version (74K): [in this window] [in a new window]   Fig. 2. Confirmation of human nAChR 4 and 2 subunit transcript expression in SH-EP1-42 cells. Reverse transcription-polymerase chain reactions were executed as described under Materials and Methods using wild-type SH-EP1 cells or transfected SH-EP1-42 cells and the 4, 2, gapdh, or 2 primer sets as indicated on the image. Products of expected size [determined by reference to mass standards indicated in the left lane; MW (bp)] were generated in SH-EP1 cells for GAPDH or in SH-EP142 cells for 4 and 2 subunits. The minus-RT control was negative, and faint bands evident were not reproducibly observed.

nAChR 4 and 2 Subunit Proteins Are Expressed and Assemble Efficiently in SH-EP1-h42 Cells.

View larger version (66K): [in this window] [in a new window]   Fig. 3. Protein expression and subunit assembly of nAChR 4 and 2 or 2 subunits. Solubilized membrane protein from SH-EP1, SH-EP1-42, or SH-EP1-42 cell lines was immunoprecipitated using H133 (anti-nAChR 4 subunit; 4-IP) or H92 (anti-nAChR 2 subunit; 2-IP) antibodies and analyzed by Western analysis for immunoblots probed simultaneously with rat anti-4 mAb 299 and rabbit anti-2 polyclonal antisera s-1724. Labeled arrows on the left side of the image indicate positions for immunoreactive 4 and 2 subunits, and molecular masses (kD) of standards (middle lane) are labeled to the right of the image. Ratios of 4 versus 2 signals in both SH-EP1-42 and SH-EP1-42 cell samples immunoprecipitated with the anti-4 subunit antibody were approximately 1 as was the ratio of 4/2 subunit in SH-EP1-42 cell samples immunoprecipitated with the 2 subunit antibody targeting native 2 subunit C2 loop sequences, although that antibody failed to isolate 42 complexes from the SH-EP1-42 cell line. Faint signals from SH-EP1-42 and SH-EP1 cell samples immunoprecipitated with the anti-2 subunit antibody with apparent masses of 50 to 70 kDa likely are due to cross-reactivity with heavy chains from primary immunoprecipitating antibody carried over into the sample and were not observed in replicate studies.

42-nAChR Display Subtle Differences in Ligand-Binding Properties Compared with 42-nAChR. H-EBDN binding competition studies were executed to assess the ligand-binding properties of 42-nAChR compared with 42-nAChR (Fig. 4; Table 1). Unlabeled EBDN is the most potent ligand for blocking specific H-EBDN binding for both 42- and 42-nAChR. In addition, as previously shown for 42-nAChR (Eaton et al., 2003), all nicotinic agonists blocked H-EBDN-binding to 42-nAChR fully and with relatively high potency (Fig. 4, A and B). Generally, with the exception of cytisine, carbamylcholine, and hexamethonium (obviating systematic differences in IC₅₀ determinations between the two nAChR subtypes), ligands showed higher potency blockade of H-EBDN binding to 42-nAChR than 42-nAChR. The rank order of binding inhibition potency and IC₅₀ values for agonists acting at 42-nAChR were: 390 pM EBDN >> 10 nM cytisine > 65 nM nicotine >> 460 nM ACh 720 nM 1,1-dimethyl-4-phenyl-piperazinium (DMPP) >> 6.0 µM carbamylcholine. Antagonists tested in ligand-binding competition displayed a wider range of binding inhibition potency than agonists (Fig. 4, C–E) with rank order: 27 nM lobeline >> 280 nM suberyldicholine >> 2.3 µM dihydro--erythroidine (DHE) > 35 µM decamethonium 37 µM methyllycaconitine (MLA) 66 µM d-tubocurarine 160 µM trimethaphan > 490 µM pancuronium > 1.9 mM hexamethonium. In addition, mecamylamine did not appear to compete with H-EBDN binding and exhibited no inhibition up to 100 µM (data not shown). Generally, there were larger differences in H-EBDN binding competition potency for antagonists than agonists in comparisons between 42- and 42-nAChR. Differences (0.3 log units in log IC₅₀ values, translating into factors of two or more in molar concentrations) in inhibition potency between 42-nAChR and 42-nAChR were observed for three antagonists: lobeline, suberyldicholine, and pancuronium (Table 1), and these as well as smaller differences observed for EBDN, ACh, DHE, decamethonium, and MLA were significant at the 95% confidence limit. Pancuronium also subtly fell out of rank order established for 42-nAChR when assessed for H-EBDN binding competition properties at 42-nAChR.

View larger version (22K): [in this window] [in a new window]   Fig. 4. Drug competition for specific H-EBDN binding to sites on SH-EP1-42 and SH-EP1-42 cells. Reaction mixtures containing the indicated competing ligand at the specified concentrations (abscissa; molar log scale) were used to compete for specific binding (ordinate; % of control) of 400 pM H-EBDN to membrane preparations containing 25 fmol binding sites from SH-EP1-42 cells [dashed lines indicate curve fits to data points not displayed to enhance clarity (see Eaton et al., 2003)] or SH-EP1-42 cells [solid lines indicating curve fits to data for: A, EBDN (), nicotine (Nic; ), DMPP (); B, cytisine (Cyt; ), ACh (), carbamylcholine (Carb; ); C, suberyldicholine (Sub; ), DHE(), MLA (); D, lobeline (Lob; ), decamethonium (Deca; ), hexamethonium (Hexa; ); E, d-tubocurarine (d-TC; ), trimethaphan (Trim; ), or pancuronium (Panc; )]. Results are the averages of three separate experiments (mean ± S.E.M.).

View this table: [in this window] [in a new window]   TABLE 1 Drug competition toward specific H-EBDN binding to 42-nAChR and 42-nAChR in transfected SH-EP1 cells Radioligand binding competition assays were conducted as described under Materials and Methods. Results were fit to the logistic equation to determine log IC50 values (±S.E.M.). Overlap in the range of values at the 95% confidence interval was used as a cut-off in tests of statistical significance for any differences.

Function of 42-nAChR Assessed Using ⁸⁶Rb⁺ Efflux Assays. Function of expressed 42-nAChR was evaluated using ⁸⁶Rb⁺ efflux assays, and results were compared with findings using 42-nAChR (Figs. 5 and 6; Tables 2 and 3). Agonist concentration-response profiles did not differ dramatically at submaximally efficacious concentrations of selected ligands for 42-nAChR (Fig. 5, solid lines and filled symbols) and 42-nAChR (Fig. 5, dashed lines and open symbols). At the 95% confidence level, nicotine and DMPP had lower EC₅₀ values, and carbamylcholine had a higher EC₅₀ value when acting at 42-nAChR compared with action at 42-nAChR. However, none of these differences were more than 0.3 log units or a factor of 2, nor were they larger than between-study differences observed (compare Table 2 entries for 42-nAChR to those in Eaton et al., 2003). Nevertheless, self-inhibition of 42-nAChR function occurred at higher concentrations of EBDN, DMPP, and carbamylcholine, which did not exhibit self-inhibition in actions at 42-nAChR, and nicotine showed more self-inhibition of 42-nAChR than of 42-nAChR. Cytisine had comparable potency and comparable, submaximal efficacy without showing evidence of self-inhibition at 42- and 42-nAChR. Rank order agonist potency from the appropriate fits to agonism only or to agonism with self-inhibition profiles and corresponding EC₅₀ values (Table 2) for actions at 42-nAChR (compared with EC₅₀ values following in parentheses for actions of agonists at 42-nAChR) were: 12 nM (17 nM) EBDN >> 1.5 µM (1.0 µM) nicotine 1.8 µM (1.7 µM) ACh > 3.3 µM (3.7 µM) cytisine > 9.8 µM (5.1 µM) DMPP > 17 µM (32 µM) carbamylcholine. Note that the rank order was the same for agonists acting at 42-nAChR or 42-nAChR. In addition, self-inhibitory IC₅₀ values (Table 2) were also determined for EBDN (93 µM), DMPP (1 mM), nicotine (2.0 mM), carbamylcholine (32 mM) acting at 42-nAChR, and for nicotine (9.1 mM) acting at 42-nAChR.

View larger version (31K): [in this window] [in a new window]   Fig. 5. Functional assessment of agonist action at SH-EP1-42 and SH-EP1-42 cells. Measurements of specific 86Rb+ efflux (ordinate; percentage of 1 mM carbamylcholine control) were made in the presence of selected agonists at the indicated concentration (abscissa; molar log scale) in parallel experiments using SH-EP1-42 (solid lines fit to filled symbols) or SH-EP1-42 (dashed lines fit to open symbols) cell lines. Results are the averages of four to seven separate experiments (mean ± S.E.M.) for: A, epibatidine (Epi; , ), ACh (, ), or DMPP (, ); B, nicotine (Nic; , ), carbamylcholine (Carb; , ), or cytisine (Cyt; , ).

View larger version (29K): [in this window] [in a new window]   Fig. 6. Antagonist effects on function of 42-nAChR and 42-nAChR. Measurements of specific 86Rb+ efflux (ordinate; percentage of 1 mM carbamylcholine control) were made in the presence of 1 mM carbamylcholine and selected antagonists at the indicated concentration (abscissa; molar log scale) in parallel experiments using SH-EP1-42 (solid lines fit to filled symbols) or SH-EP1-42 (dashed lines fit to open symbols) cell lines. Results are the averages of three to seven separate experiments (mean ± S.E.M.) for: A, suberyldicholine (Sub; , ), trimethaphan (Trim; , ), lobeline (Lob; , ), MLA (, ), or DHE (, ); B, d-tubocurarine (d-TC; , ), mecamylamine (Meca; , ), or pancuronium (Panc; , ); or C, decamethonium (Deca; , ), or hexamethonium (Hexa; , ).

View this table: [in this window] [in a new window]   TABLE 2 Agonist function at 42-nAChR or 42-nAChR as assessed using 86Rb+ efflux assays in transfected SH-EP1 cells 86Rb+ efflux assays were conducted as described under Materials and Methods. Results were fit to the logistic equation to determine log EC50 and log IC50 (for self-inhibition of function) values (±S.E.M.). Overlap in the range of values at the 95% confidence interval was used as a cut-off in tests of statistical significance for any differences.

View this table: [in this window] [in a new window]   TABLE 3 Antagonist action at 42-nAChR and 42-nAChR as assessed using 86Rb+ efflux assays in transfected SH-EP1 cells 86Rb+ efflux assays were conducted as described under Materials and Methods. Results were fit to the logistic equation to determine log IC50 values (±S.E.M.). Overlap in the range of values at the 95% confidence interval was used as a cut-off in tests of statistical significance for any differences.

As is the case for actions at 42-nAChR, antagonist log concentration-response profiles showed full inhibition of 42-nAChR function stimulated by 1 mM carbamylcholine (Fig. 6; Table 3). There were insignificant differences in the abilities of suberyldicholine, trimethaphan, lobeline, MLA, or DHE to inhibit function of 42-nAChR compared with 42-nAChR (Fig. 6A; Table 3). Nevertheless, 42-nAChR, relative to 42-nAChR, displayed increased sensitivity (Table 3) to functional blockade by mecamylamine, which is a noncompetitive inhibitor of 42-nAChR function, decamethonium, which has a competitive inhibition signature at 42-nAChR, or d-tubocurarine, pancuronium, or hexamethonium, which show mixed mechanisms of functional block of 42-nAChR (Fig. 6, B and C; see also Fig. 9 in Eaton et al., 2003). Rank order antagonist potencies and IC₅₀ values for 42-nAChR were: 200 nM mecamylamine >> 2.0 µM DHE > 6.5 µM MLA > 9.1 µM hexamethonium > 18 µM lobeline 21 µM d-tubocurarine 25 µM trimethaphan > 36 µM pancuronium > 110 µM decamethonium 150 µM suberyldicholine. Furthermore, succinyldicholine at concentrations as high as 1 mM did not display antagonist activity. Relative to actions at 42-nAChR, antagonist functional potency for hexamethonium, d-tubocurarine, decamethonium, and pancuronium at 42-nAChR subtly fell out of rank order.

Accelerated Decay of Nicotine-Induced, Inward, Whole-Cell Currents in SH-EP1-42-nAChR. Patch-clamp electrophysiological recording revealed that peak whole-cell current responses to nicotinic agonists were not different for cells expressing 42- or 42-nAChR and responding to either 1 µM nicotine (Fig. 7A) or 1 mM ACh (Fig. 7B). Peak current amplitudes are 631.3 ± 142 pA for 42-nAChR and 577 ± 100 pA for 42-nAChR in response to 1 µM nicotine (Fig. 7C, left panel). However, compared with 42-nAChR-mediated currents, or to agonist-induced currents mediated by wild-type 5-HT_3A receptors (Choi et al., 2003), 42-nAChR-mediated responses exhibited faster acute desensitization (diminished inward current during the course of nicotinic agonist application) represented as a significant decrease in the current decay constant and reflected in a smaller steady-state current relative to peak current for 42-nAChR (Fig. 7, A and B). The ratios of steady-state current to peak current (I_s/I_p, as a percentage of peak current) are 41.9 ± 5.2% for 42-nAChR and 26.7 ± 4.4% for 42-nAChR (Fig. 7C, middle and right panels). Transitions from peak to steady-state currents were characterized by decay constants () for an e-fold (63%) reduction in inward current amplitude of 3508 ± 512 ms for 42-nAChR and 1721 ± 123 ms for 42-nAChR [compare with 3 s also for 5-HT_3A receptor whole-cell current responses to agonist (e.g., see Choi et al., 2003)].

View larger version (24K): [in this window] [in a new window]   Fig. 7. Whole-cell current responses in 44-nAChR and 42-nAChR. A and B, typical whole-cell current traces recorded at -70 mV holding potential in response to 4 s applications (2-s horizontal scale) of 1 µM nicotine (A, 100 pA vertical scale) or 1 mM acetylcholine (B, 500 pA vertical scale) for 42-nAChR (left) or 42-nAChR (middle) with traces superimposed on the right. Current responses desensitize faster and more completely for 42-nAChR than for 42-nAChR. C, bar graphs show no significant difference in peak current amplitudes (631.3 ± 142 pA for 42-nAChR, 577 ± 100 pA for 42-nAChR; p = 0.51; left panel) but that there are differences in ratios of steady-state current to peak current (Is/Ip as a percentage of peak current; 41.9 ± 5.2% for 42-nAChR, 26.7 ± 4.4% for 42-nAChR; p = 0.03; middle panel) and in decay constants for transition from peak to steady-state currents ( = 3508 ± 512 ms for 42-nAChR, = 1721 ± 123 ms for 42-nAChR; p = 0.009; right panel) for results obtained from eight cell measures for 42-nAChR or 12 cells for 42-nAChR in response to 1 µM nicotine.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
One of the principal findings of this study is that functional and ligand-binding nAChR are formed as a combination of human nAChR wild-type 4 subunits with chimeric 2 subunits in which nested C2 domain sequences are replaced by corresponding sequences from the 5-HT_3A receptor subunit. This indicates that native, nested C2 sequences in 2 subunits are not essential for formation of functional 42-nAChR. Radiolabeled agonist binding competition properties and functional agonist potencies varied no more than subtly between 42-nAChR and 42-nAChR, as one might expect given preservation of the E1 and E2 domains thought to compose the agonist binding site. Nevertheless, and quite unexpectedly, compared with 42-nAChR, 42-nAChR displayed novel or heightened functional self-inhibition by agonists at higher concentrations and higher sensitivity to functional blockade by several antagonists. Furthermore, 42-nAChR displayed faster and more complete acute desensitization of function. For these or any other mutagenesis or chimera study, there is a formal possibility that effects are nonspecific and due to molecular distortion imposed by sequence alterations. However, a reasonable interpretation of the current findings is that changes in sensitivity to blockade by antagonists, sensitivity to self-inhibition, and rates and extents of desensitization of nAChR correlate with changes in 2 subunit nested C2 sequences not previously implicated in nAChR function.

When more extensive substitutions were made involving 5-HT_3A subunit M3 and/or M4 transmembrane domains and their neighboring proximal C2 sequences, chimeric 4, 2, or 4 subunits are unable to form functional nAChR. When coupled with observations from previous reports concerning similar lines of study (see below), the new findings suggest that native M3, M4, and/or proximal C2 domains are essential for formation of functional 42- or 44-nAChR. However, the ability of some of the nAChR containing M3-E3 chimeric subunits to exhibit low levels of specific H-EBDN binding implies that subunit-subunit interactions allowing formation of closed assemblies to create ligand-binding interfaces does occur. Further studies, including substitutions of just proximal C2 sequences and not the transmembrane domains but also of more extensive, M3, M4, and proximal C2 sequence substitutions, are warranted to ascertain whether any inability to form functional entities is due to failure to traffic subunit assemblies to the cell surface or is due to some other functional flaw in cell surface-expressed receptors.

How do the present findings relate to other existing literature? Williams et al. (1998) noted differences, using the Xenopus oocyte heterologous expression system, in agonist functional potency and/or channel kinetics in their whole-cell current studies of wild-type chick 7-nAChR compared with nAChR composed of chimeric 7 subunits containing 3 or 5 subunit proximal-plus-nested C2 loop sequences. The close similarities between nAChR subunit proximal C2 sequences make it likely that nested C2 sequences account for the differences observed. When coupled with our findings also indicating differences in agonist effects and in whole-cell current response profiles for 42-nAChR compared with 42-nAChR, functional consequences of nested C2 domains are underscored.

Based on sequence deletion studies, Valor et al. (2002) identified regions within C2 essential for the expression of functional rat 7-nAChR. One of these corresponds to 14 residues of the human 7 subunit in C2 proximal to M3 (H319-R332; Fig. 1), and another corresponds to 11 residues in C2 proximal to M4 (human 7 E459-R469; Fig. 1). Our chimera/substitution analyses based on nAChR 4, 2, and 4 subunits are consistent with indispensable roles for these highly conserved, proximal C2 sequences in formation of functional nAChR, whether or not neighboring M3 and M4 domains also are indispensable for functional nAChR formation.

The deletion studies of Valor et al. also suggested essential roles in formation of functional rat 7-nAChR expression for the most N-terminal amino acids within the nested C2 domain (corresponding to human 7 V333-R347; Fig. 1) including a three-residue MKR signal (corresponding to human 7 M345-R347) that would not tolerate substitution. Perhaps because the human nAChR 2 subunit does not have the 7 subunit C2 domain MKR signature (having MQQ instead), substitution of the nested C2 sequence from the 5-HT_3A subunit did not compromise functional expression of 42-nAChR. Substitutions of rat nAChR 7 C339-R344 in the N-terminal, nested C2 region were tolerated in the studies of Valor et al., perhaps suggesting a role merely in maintaining a structural requirement in the region. The ability to substitute for this part of the nAChR subunit nested C2 sequence was also observed in our studies of chimeric 2 subunits.

Valor et al. reported that substitutions for rat nAChR 7 subunit nested C2 sequences (117 amino acids corresponding to human 7 P348-E454; Fig. 1) with rat nAChR 4 (226 amino acids) or 5 (45 amino acids) subunit sequences reduced heterologous expression as radioligand binding sites and fully or largely eliminated function of resultant, chimeric 7-nAChR. However, this could be due to steric incompatibilities if 7 subunit C2 sequences are substituted with sequences from subunits that do not seem to assemble as homomers. Substitution instead with green fluorescent protein (239 amino acids) or synaptosome-associated protein 25 kDa (206 amino acids) also sharply attenuated or abolished expression of functional nAChR, even though minimal substitution with an artificial, seven amino acid insert or with a natural EGM or mutated EAA or AGA sequences allowed for higher functional expression of mutant 7-nAChR expression than observed for wild-type 7-nAChR (Valor et al., 2002). However, our SH-EP1-42 cells, substituting the mouse 5-HT_3A subunit sequence for an essentially equivalent stretch of nested C2 sequence in nAChR 2 subunits, express ligand-binding and functional nAChR at levels like that seen in SH-EP1-42 cells containing wild-type subunits. Thus, the incompatibility with functional expression of the substitutions made for nested C2 sequences in the rat 7 subunit are specific to either nAChR subunit/subtype or substituting sequence, or the incompatibility can be overcome by some compensating feature of the wild-type 4 subunits expressed in our SH-EP1-42 cells. Continuing work investigating roles of nested C2 domains in 4, 2, and 7 subunits seeks to further elucidate our understanding in this area.

Several studies have been conducted addressing sequence elements required for muscle-type nAChR [(1)₂1-nAChR or 1^*-nAChR] assembly. First, the inability to form functional nAChR from chimeric 4 and/or 2 subunits containing substituted M3, M4, and proximal C2 domains is consistent with earlier studies showing that M1, M2, and M3 domains of 1 subunits are important for efficient expression of Torpedo nAChR on the cell surface in the Xenopus oocyte heterologous expression system (Tobimatsu et al., 1987), even though extracellular N-terminal domains of 1, , and subunits appear to contain information sufficient for initial specific subunit association (Verrall and Hall, 1992). However, a 17-amino acid sequence at the C terminus of the large cytoplasmic loop in the 1 subunit is essential in the late stage of receptor assembly with 1 subunits because chimeric 1 proteins containing the corresponding 1 sequence cannot properly assemble with 1 subunits (Yu and Hall, 1994b). The equivalent I434-S448 in the C-terminal part of the human 2 subunit nested C2 domain is replaced by the 5-HT_3A subunit sequence, but V449 in the beginning of the M4-proximal C2 sequence is preserved in our chimeric 2 subunit without an apparent compromise in efficiency of 42-nAChR assembly. Perhaps the sequence requirement is specific to an subunit for - subunit assembly and will be revealed in studies of roles for 4 subunits in formation of functional 42- or 44-nAChR, or it may only be relevant to formation of higher order complexes such as (1)₂1-nAChR.

Using the oocyte expression system, Morgado-Valle et al. (2001) reported a sequence motif conserved across nAChR 4, 2, 4, and other subunits RXPXTH(X)₁₄P (corresponding to human 4 R324-P344; Fig. 1), located in the C terminus of the nAChR 2 subunit C2 domain, that is necessary for function of 42-nAChR. However, our findings differ in that 42-nAChR function is evident even though human 2 P344 is replaced by an alanine in our chimeric 2 subunit. Nevertheless, note that none of the nAChR subunits capable of forming homomeric nAChR, namely 7, 8, and 9, nor 5-HT_3A subunits, contains this motif, having instead an alanine residue at the equivalent position to 2 P344 followed by a tryptophan (Fig. 1). Perhaps the AW signature present in the 2 subunit adequately substitutes for P344 in influencing subunit folding and assembly.

Although both and subunits in heteromeric, nn-nAChR can contribute to ligand binding and functional properties and coupling of drug binding to channel opening, several studies have suggested that subunits may play a dominant role in controlling the rate and extent of nAChR functional inactivation after chronic agonist exposure (Fenster et al., 1997; Kuryatov et al., 2000; Gentry et al., 2003). Results from the current study suggest, quite unexpectedly, that functional inactivation or desensitization of 42-nAChR is regulated in part by nested C2 sequences. Nevertheless, there are other recently reported indications that C2 domain determinants can influence single-channel conductance of 5-HT_3A receptors (Kelly et al., 2003) or 42-nAChR (Lambert et al., 2004), consistent with the current findings suggesting active roles of the nested cytoplasmic loop in nAChR function.

	Footnotes

 
This project, part of which was conducted in the Charlotte and Harold Simensky Neurochemistry of Alzheimer's Disease Laboratory, was funded by grants from the Arizona Disease Control Research Commission (9730 and 9615), the National Institutes of Health (NIH NS40417), Targacept, Inc., the Roberta and Gloria Wallace Foundation, and endowment and/or capitalization funds from the Men's and Women's Boards of the Barrow Neurological Foundation.

Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.105.084954.

ABBREVIATIONS: nAChR, nicotinic acetylcholine receptor(s); ACh, acetylcholine; 5-HT₃, 5-hydroxytryptamine or serotonin type 3; PCR, polymerase chain reaction; bp, base pair(s); RT, reverse transcription; H-EBDN, [³H]epibatidine; DMPP, 1,1-dimethyl-4-phenyl-piperazinium; DHE, dihydro--erythroidine; MLA, methyllycaconitine; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Address correspondence to: Dr. Yen-Ping Kuo, Department of Neurobiology, Barrow Neurological Institute, 350 West Thomas Road, Phoenix, AZ 85013. E-mail: ykuo{at}chw.edu.

	References

Top Abstract Materials and Methods Results Discussion References

 

Bencherif M and Lukas RJ (1993) Cytochalasin modulation of nicotinic cholinergic receptor expression and muscarinic receptor function in human TE671/RD cells: a possible functional role of the cytoskeleton. J Neurochem 61: 852-864.[Medline]
Brady CA, Stanford IM, Ali I, Lin L, Williams JM, Dubin AE, Hope AG, and Barnes NM (2001) Pharmacological comparison of human homomeric 5-HT3A receptors versus heteromeric 5-HT3A/3B receptors. Neuropharmacology 41: 282-284.[CrossRef][Medline]
Choi JS, Choi BH, Ahn HS, Kim MJ, Rhie DJ, Yoon SH, Min do S, Jo YH, Kim MS, Sung KW, and Hahn SJ (2003) Mechanism of block by fluoxetine of 5-hydroxytryptamine3 (5-HT3)-mediated currents in NCB-20 neuroblastoma cells. Biochem Pharmacol 66: 2125-2132.[Medline]
Colledge M and Froehner SC (1997) Tyrosine phosphorylation of nicotinic acetylcholine receptor mediates Grb2 binding. J Neurosci 17: 5038-5045.[Abstract/Free Full Text]
Cooper ST and Millar NS (1998) Host cell-specific folding of the neuronal nicotinic receptor alpha8 subunit. J Neurochem 70: 2585-2593.[Medline]
Corringer PJ, Le Novere N, and Changeux JP (2000) Nicotinic receptors at the amino acid level. Annu Rev Pharmacol Toxicol 40: 431-458.[CrossRef][Medline]
Dajas-Bailador FA, Soliakov L, and Wonnacott S (2002) Nicotine activates the extracellular signal-regulated kinase 1/2 via the alpha7 nicotinic acetylcholine receptor and protein kinase A, in SH-SY5Y cells and hippocampal neurones. J Neurochem 80: 520-530.[CrossRef][Medline]
Davies PA, Pistis M, Hanna MC, Peters JA, Lambert JJ, Hales TG, and Kirkness EF (1999) The 5-HT3B subunit is a major determinant of serotonin-receptor function. Nature (Lond) 397: 359-363.[CrossRef][Medline]
Eaton JB, Peng J-H, Schroeder KM, George AA, Fryer JD, Krishnan C, Buhlman L, Kuo Y-P, Steinlein O, and Lukas RJ (2003) Characterization of human 42-nicotinic acetylcholine receptors stably and heterologously expressed in native nicotinic receptor-null SH-EP1 human epithelial cells. Mol Pharmacol 64: 1283-1294.[Abstract/Free Full Text]
Eisele JL, Bertrand S, Galzi JL, Devillers-Thiery A, Changeux JP, and Bertrand D (1993) Chimaeric nicotinic-serotonergic receptor combines distinct ligand binding and channel specificities. Nature (Lond) 366: 479-483.[CrossRef][Medline]
Fenster CP, Beckman ML, Parker JC, Sheffield EB, Whitworth TL, Quick MW, and Lester RA (1999) Regulation of alpha4beta2 nicotinic receptor desensitization by calcium and protein kinase C. Mol Pharmacol 55: 432-443.[Abstract/Free Full Text]
Fenster CP, Rains MF, Noerager B, Quick MW, and Lester RA (1997) Influence of subunit composition on desensitization of neuronal acetylcholine receptors at low concentrations of nicotine. J Neurosci 17: 5747-5759.[Abstract/Free Full Text]
Gentry CL, Wilkins LH Jr, and Lukas RJ (2003) Effects of prolonged nicotinic ligand exposure on function of heterologously expressed, human alpha4beta2- and alpha4beta4-nicotinic acetylcholine receptors. J Pharmacol Exp Ther 304: 206-216.[Abstract/Free Full Text]
Guo X and Wecker L (2002) Identification of three cAMP-dependent protein kinase (PKA) phosphorylation sites within the major intracellular domain of neuronal nicotinic receptor 4 subunits. J Neurochem 82: 439-447.[CrossRef][Medline]
Huganir RL and Greengard P (1990) Regulation of neurotransmitter receptor desensitization by protein phosphorylation. Neuron 5: 555-567.[CrossRef][Medline]
Karlin A (2002) Emerging structure of the nicotinic acetylcholine receptors. Nat Rev Neurosci 3: 102-114.[CrossRef][Medline]
Kelly SP, Dunlop JI, Kirkness EF, Lambert JJ, and Peters J (2003) A cytoplasmic region determines single-channel conductance in 5HT3 receptors. Nature (Lond) 424: 321-324.[CrossRef][Medline]
Kuo Y-P, Wu J, Zhao L, Xu L, and Lukas RJ (2002) Characterization of human wild-type 4-chimeric 2-nicotinic acetylcholine receptors stably expressed in SH-EP1 cells. Soc Neurosci Abstr 537.13.
Kuryatov A, Olale F, Cooper J, Choi C, and Lindstrom J (2000) Human alpha6 AChR subtypes: subunit composition, assembly and pharmacological responses. Neuropharmacology 39: 2570-2590.[CrossRef][Medline]
Lambert JJ, Dunlop JI, Kelley SP, Deeb T, Hales TG, and Peters JA (2004) A determinant of single channel conductance within the large intracellular loop of a nicotinic acetylcholine receptor. Soc Neurosci Abstr 275.8.
Lindstrom J (1996) Neuronal nicotinic acetylcholine receptors, in Ion Channels (Narahashi T ed) pp 377-450, Plenum, New York.
Lukas RJ (1998) Neuronal nicotinic acetylcholine receptors, in The Nicotinic Acetylcholine Receptors: Current Views and Future Trends (Barrantes FJ ed) pp 145-173, Springer Verlag, Berlin/Heidelberg and Landes Publishing, Georgetown, TX.
Lukas RJ, Fryer JD, Eaton JB, and Gentry CL (2002) Some methods for studies of nicotinic acetylcholine receptor pharmacology, in Nicotinic Receptors and Nervous System (Levin ED ed) pp 3-27, CRC Press, Boca Raton.
Lukas RJ, Norman SA, and Lucero L (1993) Characterization of nicotinic acetylcholine receptors expressed by cells of the SH-SY5Y human neuroblastoma clonal line. Mol Cell Neurosci 4: 1-12.
Miyazawa A, Fujiyoshi Y, Stowell M, and Unwin N (1999) Nicotinic acetylcholine receptor at 4.6 angstrom resolution: transverse tunnels in the channel wall. J Mol Biol 288: 765-786.[CrossRef][Medline]
Morgado-Valle C, Garcia-Colunga J, Miledi R, and Diaz-Munoz M (2001) A motif present in the main cytoplasmic loop of nicotinic acetylcholine receptors and catalases. Proc R Soc Lond B Biol Sci 268: 967-972.[Medline]
Shaw S, Bencherif M, and Marrero MB (2002) Janus kinase 2, an early target of alpha 7 nicotinic acetylcholine receptor-mediated neuroprotection against Abeta-(1–42) amyloid. J Biol Chem 277: 44920-44924.[Abstract/Free Full Text]
Shoop RD, Yamada N, and Berg DK (2000) Cytoskeletal links of neuronal acetylcholine receptors containing alpha 7 subunits. J Neurosci 20: 4021-4029.[Abstract/Free Full Text]
Sine SM (2002) The nicotinic receptor ligand binding domain. J Neurobiol 53: 431-446.[CrossRef][Medline]
Tobimatsu T, Fujita Y, Fukuda K, Tanaka K, Mori Y, Konno T, Mishina M, and Numa S (1987) Effects of substitution of putative transmembrane segments on nicotinic acetylcholine receptor function. FEBS Lett 222: 56-62.[CrossRef][Medline]
Valor LM, Mulet J, Sala F, Sala S, Ballesta JJ, and Criado M (2002) Role of the large cytoplasmic loop of the alpha 7 neuronal nicotinic acetylcholine receptor subunit in receptor expression and function. Biochemistry 41: 7931-7938.[CrossRef][Medline]
Verrall S and Hall ZW (1992) The N-terminal domains of acetylcholine receptor subunits contain recognition signals for the initial steps of receptor assembly. Cell 68: 23-31.[CrossRef][Medline]
Wecker L and Rogers CQ (2003) Phosphorylation sites within 4 subunits of 42 neuronal nicotinic receptors: a comparison of substrate specificities for cAMP-dependent protein kinase A (PKA) and protein kinase C (PKC). Neurochem Res 28: 431-436.[CrossRef][Medline]
Williams BM, Temburni MK, Levey MS, Bertrand S, Bertrand D, and Jacob MH (1998) The long internal loop of the alpha 3 subunit targets nAChRs to subdomains within individual synapses on neurons in vivo. Nat Neurosci 1: 557-562.[CrossRef][Medline]
Wu J, Chan P, Schroeder KM, Ellsworth K, and Partridge LD (2002) 1-Methyl-4-phenylpridinium (MPP+)-induced functional run-down of GABA(A) receptor-mediated currents in acutely dissociated dopaminergic neurons. J Neurochem 83: 87-99.[CrossRef][Medline]
Yu XM and Hall ZW (1994a) The role of the cytoplasmic domains of individual subunits of the acetylcholine receptor in 43 kDa protein-induced clustering in COS cells. J Neurosci 14: 785-795.[Abstract]
Yu XM and Hall ZW (1994b) A sequence in the main cytoplasmic loop of the alpha subunit is required for assembly of mouse muscle nicotinic acetylcholine receptor. Neuron 13: 247-255.[CrossRef][Medline]

This article has been cited by other articles:

				J. E. Carland, M. A. Cooper, S. Sugiharto, H.-J. Jeong, T. M. Lewis, P. H. Barry, J. A. Peters, J. J. Lambert, and A. J. Moorhouse Characterization of the Effects of Charged Residues in the Intracellular Loop on Ion Permeation in {alpha}1 Glycine Receptor Channels J. Biol. Chem., January 23, 2009; 284(4): 2023 - 2030. [Abstract] [Full Text] [PDF]

				E. X. Albuquerque, E. F. R. Pereira, M. Alkondon, and S. W. Rogers Mammalian Nicotinic Acetylcholine Receptors: From Structure to Function Physiol Rev, January 1, 2009; 89(1): 73 - 120. [Abstract] [Full Text] [PDF]

				R. Zwart, A. L. Carbone, M. Moroni, I. Bermudez, A. J. Mogg, E. A. Folly, L. M. Broad, A. C. Williams, D. Zhang, C. Ding, et al. Sazetidine-A Is a Potent and Selective Agonist at Native and Recombinant {alpha}4{beta}2 Nicotinic Acetylcholine Receptors Mol. Pharmacol., June 1, 2008; 73(6): 1838 - 1843. [Abstract] [Full Text] [PDF]

				J. Wu, Q. Liu, K. Yu, J. Hu, Y.-P. Kuo, M. Segerberg, P. A. St John, and R. J. Lukas Roles of nicotinic acetylcholine receptor {beta} subunits in function of human {alpha}4-containing nicotinic receptors J. Physiol., October 1, 2006; 576(1): 103 - 118. [Abstract] [Full Text] [PDF]

				C. A. Briggs, E. J. Gubbins, M. J. Marks, C. B. Putman, R. Thimmapaya, M. D. Meyer, and C. S. Surowy Untranslated Region-Dependent Exclusive Expression of High-Sensitivity Subforms of {alpha}4beta2 and {alpha}3beta2 Nicotinic Acetylcholine Receptors Mol. Pharmacol., July 1, 2006; 70(1): 227 - 240. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.084954v1 314/1/455    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kuo, Y.-P. Articles by Lukas, R. J. Search for Related Content PubMed PubMed Citation Articles by Kuo, Y.-P. Articles by Lukas, R. J.