A Highly Conserved Cytoplasmic Cysteine Residue in the α4 Nicotinic Acetylcholine Receptor Is Palmitoylated and Regulates Protein Expression*

Background: The mechanisms underlying nicotinic acetylcholine receptor (nAChR) trafficking are unclear. Results: Cysteine mutations within cytoplasmic loops of the α4 nAChR subunit change surface and total receptor expression, and a cysteine in the first loop is palmitoylated. Conclusion: α4 nAChR intracellular cysteines influence receptor stability and trafficking. Significance: Identifying the determinants of nAChR trafficking will provide insight into nAChR biology. Nicotinic acetylcholine receptor (nAChR) cell surface expression levels are modulated during nicotine dependence and multiple disorders of the nervous system, but the mechanisms underlying nAChR trafficking remain unclear. To determine the role of cysteine residues, including their palmitoylation, on neuronal α4 nAChR subunit maturation and cell surface trafficking, the cysteines in the two intracellular regions of the receptor were replaced with serines using site-directed mutagenesis. Palmitoylation is a post-translational modification that regulates membrane receptor trafficking and function. Metabolic labeling with [3H]palmitate determined that the cysteine in the cytoplasmic loop between transmembrane domains 1 and 2 (M1–M2) is palmitoylated. When this cysteine is mutated to a serine, producing a depalmitoylated α4 nAChR, total protein expression decreases, but surface expression increases compared with wild-type α4 levels, as determined by Western blotting and enzyme-linked immunoassays, respectively. The cysteines in the M3-M4 cytoplasmic loop do not appear to be palmitoylated, but replacing all of the cysteines in the loop with serines increases total and cell surface expression. When all of the intracellular cysteines in both loops are mutated to serines, there is no change in total expression, but there is an increase in surface expression. Calcium accumulation assays and high affinity binding for [3H]epibatidine determined that all mutants retain functional activity. Thus, our results identify a novel palmitoylation site on cysteine 273 in the M1-M2 loop of the α4 nAChR and determine that cysteines in both intracellular loops are regulatory factors in total and cell surface protein expression of the α4β2 nAChR.

Nicotinic acetylcholine receptor (nAChR) cell surface expression levels are modulated during nicotine dependence and multiple disorders of the nervous system, but the mechanisms underlying nAChR trafficking remain unclear. To determine the role of cysteine residues, including their palmitoylation, on neuronal ␣4 nAChR subunit maturation and cell surface trafficking, the cysteines in the two intracellular regions of the receptor were replaced with serines using site-directed mutagenesis. Palmitoylation is a post-translational modification that regulates membrane receptor trafficking and function. Metabolic labeling with [ 3 H]palmitate determined that the cysteine in the cytoplasmic loop between transmembrane domains 1 and 2 (M1-M2) is palmitoylated. When this cysteine is mutated to a serine, producing a depalmitoylated ␣4 nAChR, total protein expression decreases, but surface expression increases compared with wild-type ␣4 levels, as determined by Western blotting and enzyme-linked immunoassays, respectively. The cysteines in the M3-M4 cytoplasmic loop do not appear to be palmitoylated, but replacing all of the cysteines in the loop with serines increases total and cell surface expression. When all of the intracellular cysteines in both loops are mutated to serines, there is no change in total expression, but there is an increase in surface expression. Calcium accumulation assays and high affinity binding for [ 3 H]epibatidine determined that all mutants retain functional activity. Thus, our results identify a novel palmitoylation site on cysteine 273 in the M1-M2 loop of the ␣4 nAChR and determine that cysteines in both intracellular loops are regulatory factors in total and cell surface protein expression of the ␣4␤2 nAChR.
Neuronal nicotinic acetylcholine receptors (nAChRs) 3 are part of the Cys-loop family of ligand-gated cationic channels. They are assembled from different combinations of ␣2-␣10 and ␤2-␤4 subunits to form pentameric receptors (1). The subunit structure contains four transmembrane domains giving rise to two cytoplasmic loops, a small linker region between the first and second transmembrane domains (M1-M2) and a large loop between the third and fourth transmembrane domains (M3-M4) varying in size between the subunits, with the ␣4 nAChR subunit containing the largest loop. The M3-M4 intracellular loop of the rat ␣4 nAChR subunit contains specific leucine residues that are critical for trafficking the receptor to the cell surface (2) and also contains a motif necessary for receptor targeting to axons (3). However, the mechanisms underlying trafficking and targeting remain unclear.
Protein S-palmitoylation is a reversible, covalent attachment of the 16-carbon fatty acid, palmitate, to cysteine residues on proteins by thioester bond formation. It is a dynamic posttranslational modification that plays a role in the trafficking of many soluble and membrane proteins, including nAChRs, through the secretory pathway to the cell surface, localization in specific membrane domains at the surface, and internalization, recycling, and/or degradation (for review, see . In neurons, palmitoylation modulates growth cone formation and neurite outgrowth (8), presynaptic vesicle fusion (9), and postsynaptic signal transduction (10).
The muscle-type nAChR was the first receptor type found to be palmitoylated on both the ␣ and ␤ nAChR subunits (11). More recently, the neuronal ␣7 and ␣4␤2 nAChRs have been shown to be palmitoylated (12,13), but the specific palmitoyla-tion sites have not been identified. We exchanged all of the cysteines with serines in the intracellular loops of the ␣4 nAChR subunit to determine the role of cysteine modifications, including palmitoylation, on the expression and trafficking of the ␣4 nAChR subunit. Within the M1-M2 region, there is a single cysteine residue, whereas the M3-M4 loop contains 11 cysteine residues. The cysteine in the M1-M2 loop is palmitoylated and is conserved in the ␣2, ␣3, ␣4, ␣6, ␤2, and ␤4 nAChR subunits. The homologous cysteine previously has been associated with use-dependent inactivation of ␣3-containing nAChRs in hyperglycemic and oxidative conditions (14). When this cysteine is mutated to a serine, creating a depalmitoylated mutant, total protein expression is decreased, whereas cell surface expression increases. The cysteine to serine mutations in the M3-M4 cytoplasmic loop lead to increased total and cell surface protein expression. When all of the cysteines are mutated to serines in both cytoplasmic domains, there is no change in total protein expression, but surface expression is increased. Functional activation, ligand-binding affinity, and targeting to presynaptic terminals of the mutant ␣4␤2 nAChRs do not vary from wild-type (WT) ␣4 nAChR subunits. Our results suggest that cysteine residues modulate both total and surface protein expression of the ␣4 nAChR subunit.

EXPERIMENTAL PROCEDURES
Constructs-All constructs were made by PCR using appropriate pairs of forward and reverse synthetic oligonucleotide primers (Invitrogen) and PfuTurbo DNA polymerase (Agilent Technologies, Santa Clara, CA). The rat ␣4 nAChR subunit containing an N-terminal FLAG tag and the untagged ␤2 nAChR subunit cDNAs were cloned into the mammalian cell expression vector, C-terminal p3xFLAG-CMV vector (Sigma). Mutagenesis of the ␣4 nAChR subunit was performed using the QuikChange site-directed mutagenesis kit (Agilent Technologies). Three mutant constructs were generated in the ␣4 nAChR: 1) a cysteine to serine mutation within the M1-M2 loop (␣4C273S), 2) 11 mutations within the M3-M4 loop (␣4⌬C), and 3) mutation of all cysteines within the M1-M2 and M3-M4 loops (␣4⌬⌬C). For the ␣4⌬⌬C mutant, the ␣4⌬C plasmid backbone was used with the ␣4C273S primers to generate a mutant lacking all of the intracellular cysteines (see supplemental "Experimental Procedures" for primer pairs).
tsA 201 Cell Culture-Human tsA 201 cells, a derivative of the HEK cell line 293, were cultured in DMEM (Invitrogen) supplemented with 10% FBS (HyClone, Thermo Scientific, Logan, UT) and 20 g/ml gentamicin (Invitrogen) as described previously (2). tsA 201 cells were transfected using Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen). For all experiments, tsA 201 cells were plated on day 1 and transfected on day 2, medium was exchanged on day 3, moved to 30°C to promote assembly of ␣4␤2 nAChRs on day 4, and used for experiments on day 5.
Primary Hippocampal Neuron Culture-Cultures were prepared and maintained as described previously (15). Neurons were transfected at 7-10 days in vitro (DIV) using the Clontech CalPhos Mammalian Transfection Kit (BD Biosciences) as described (16).
Metabolic Labeling of nAChRs with [ 3 H]Palmitate-tsA 201 cells were transfected with ␣4, ␣4⌬C, ␣4⌬⌬C, or ␣4C273S. On the day of the experiment, medium was exchanged for starvation (DMEM containing 1% fatty acid-free BSA), incubated at 37°C for 30 min, and then metabolically labeled in starvation medium containing 1 millicurie/ml [ 3 H]palmitate (Perkin-Elmer Life Science) for 4 h at 37°C. The cells were washed in PBS and lysed in solubilization buffer (50 mM NaCl, 5 mM Tris, pH 8.0, 5 mM EDTA, 5 mM EGTA, 2% Triton X-100) with aprotinin, leupeptin, and pepstatin (all at 5 g/ml), 1 mM PMSF, 1 mM benzamidine, and 1ϫ phosphatase inhibitor cocktails 1 and 2 (Sigma), and solubilized for 2 h at 4°C. The lysates were spun for 15 min at 16,000 ϫ g, and the supernatants were transferred to new tubes. The ␣4 FLAG-tagged subunits were captured by immunoprecipitation (IP) using the FLAG M2 beads (Sigma) as described previously (2). For the hydroxylamine experiment, FLAG M2 beads were treated with 1 M hydroxylamine (Sigma) or control 1 M Tris-HCl, pH 7.4 for 1 h. 10% of the IP eluates (5 l) were run on a 7.5% SDS-PAGE, transferred to PVDF, and blotted with anti-FLAG or anti-C-20 antibodies. The remaining 90% of IP eluates were run on a parallel blot that was stained with Coomassie Brilliant Blue (Bio-Rad Laboratories), destained (50% methanol, 10% acetic acid), incubated with Amplify (GE Healthcare) for 30 min, dried under vacuum at 80°C for 2 h, exposed to film (Kodak BioMax MS) at Ϫ80°C for 21 days and then developed.
Pulldowns from Transfected tsA 201 Cells-Transfected tsA 201 cells were solubilized in 2% Triton X-100 buffer for 2 h at 4°C and captured by IP using FLAG M2 beads as described previously (2).
Immunoblotting-Following separation using SDS-PAGE, proteins were transferred onto PVDF membrane and incubated with diluted Abs in PBS containing 3% nonfat milk powder. The binding of the primary Abs to proteins was detected using appropriate secondary Abs as described previously (2). Blots were scanned and analyzed in ImageJ software. ␣4 nAChR subunit expression was normalized to GAPDH (n ϭ 5).
Immunostaining and Imaging-Transfected tsA 201 cells were washed and fixed in Hanks' balanced salt solution (with Ca 2ϩ and Mg 2ϩ ) supplemented with 4% paraformaldehyde and 4% sucrose, pH 7.4 (15 min, room temperature), blocked with Hanks' balanced salt solution containing 3% normal goat serum and 3% BSA (for surface immunocytochemistry) or blocked with Hanks' balanced salt solution containing 3% normal goat serum, 3% BSA, and 0.2% Triton X-100 (for total immunocytochemistry) (30 min at room temperature), and incubated with the appropriate primary (2 h, 4°C) and secondary (90 min at room temperature) antibodies. Coverslips were mounted onto slides with ProLong Gold Antifade Reagent (Invitrogen).
For the mixed neuron/tsA 201 cell assays, the cultures were processed as described previously (15). They were fixed, blocked, and incubated with the appropriate primary (2 h, 4°C) and secondary (90 min, room temperature) antibodies.
Cells and cocultures were visualized as described (15). Single-plane fluorescence images were captured using a Hamamatsu EM camera, and images were processed using the Slide Book software (version 4.2). Figures were processed with Adobe Photoshop CS5 Extended (Adobe Systems, Inc., San Jose, CA).
Enzyme-linked Immunoassay for Quantitating Cell Surface ␣4␤2 nAChRs-Transfected tsA 201 cells were plated in 12-well dishes and analyzed for cell surface expression of ␣4␤2 nAChRs using the methodology described previously (2). Surface expression was detected with a rat mAb against the ␤2 nAChR subunit (mAb 295) whose epitope is located in the extracellular domains of the subunit. The nonspecific background binding of antibodies to cells was determined by expressing the ␤2 nAChR subunit alone. To inhibit palmitoylation, medium was replaced with fresh medium containing either 2-bromopalmitate (2-BP, 120 M, Sigma) or an equivalent volume of DMSO as a vehicle control on day 4.
Fluorescence Calcium Accumulation Assays-tsA 201 cells were transiently transfected with plasmid DNA for the ␤2 nAChR subunit and either WT ␣4, ␣4⌬C, ␣4⌬⌬C, or ␣4C273S subunits. After 48 h at 37°C, the cells were plated onto 96-well plates at a density of 1 ϫ 10 5 cells per well, and moved to 30°C overnight. The cells were washed twice with HEPES buffered Krebs-Saline solution (155 mM NaCl, 4.6 mM KCl, 1.2 mM MgSO 4 , 1.8 mM CaCl 2 , 6 mM glucose, 20 mM HEPES, pH 7.2), loaded with calcium 5 NW dye (Molecular Devices, Sunnydale, CA), and incubated for 1 h at room temperature in the dark. Fluorescence was measured at 0.7-s intervals for 2 min per well using a fluorescence plate reader (FLEXstation, Molecular Devices) at an excitation of 494 nm and an emission of 525 nm. Fluorescence was recorded for 60 s before stimulation and 60 s after stimulation with various concentrations of epibatidine. Functional responses were quantified by first calculating the net fluorescence change (the difference between control, nonstimulated groups, and agonist-treated groups). Results were expressed as a percentage of maximal agonist group responses. Curve fitting was generated using Prism software (GraphPad Software, Inc., San Diego, CA).
Ligand Binding Studies-[ 3 H]Epibatidine binding to ␣4␤2 nAChRs was done directly in wells as described previously (2). For inhibition experiments, cells were incubated with 400 pM [ 3 H]epibatidine in the presence or absence of 10, 30, 100, 300, or 1000 nM nicotine overnight at 4°C. The bound epibatidine was eluted in 0.5% SDS, 0.5 M NaOH for 1 h. The eluates were mixed with liquid scintillation fluid, and the amount of radioactivity present was measured in a LS6500 (Beckman Coulter, Brea, CA). Background binding was determined using untransfected cells and was subtracted from the amount of radioactivity in eluates. The data shown were normalized to the amount bound in the absence of nicotine.
Calculations and Statistics-For cell surface binding assays and total protein analyses, the difference between groups was analyzed statistically by paired t test in Excel. For calcium accumulation and ligand binding assays, curves were fitted using Prism 5 software. Half-maximal effective (EC 50 ) values were derived from the sigmoidal dose-response equation, Y ϭ 100/ (1 ϩ 10((logEC 50 Ϫ X)⅐HS)), and half-maximal inhibitory (IC 50 ) values were derived from the sigmoidal dose-response equation, Y ϭ 100/(1 ϩ 10((logIC 50 Ϫ X)⅐HS)), where Y is the percentage of the maximal effect at a given concentration (X), and HS is the Hill slope (HS ϭ 1 was used for these calculations).
and radioligandNM ϭ 0.40 nM and HotKdNM ϭ 0.015 nM (17). The EC 50 and K i values were compared between the groups by one-way ANOVA. All results were expressed as means Ϯ S.E.

RESULTS
␣4 nAChR Is Palmitoylated at Cysteine 273 in M1-M2 Intracellular Loop-To determine the functional roles of cysteines within the ␣4 nAChR subunit, we used site-directed mutagenesis to replace the cysteine residues with serine residues in the ␣4 nAChR cytoplasmic loops. We made three mutant constructs: ␣4C273S, in which the single cysteine in the M1-M2 loop was replaced with a serine; ␣4⌬C, in which all 11 cysteines in the M3-M4 loop were replaced with serines; and ␣4⌬⌬C, in which all 12 cysteines in both loops were replaced with serines ( Fig. 1A).
Our initial hypothesis was that the M3-M4 loop contained one or more palmitoylation sites, so tsA 201 cells were transiently transfected with N-terminally FLAG-tagged WT ␣4 or the ␣4⌬C cysteine mutant nAChR subunits in the absence or presence of the untagged ␤2 nAChR subunit and metabolically labeled with [ 3 H]palmitate, captured by IP with FLAG M2 beads, separated by SDS-PAGE, dried, and exposed to film (Fig.  1B). Both the ␣4 and ␤2 nAChR subunits were labeled metabolically with [ 3 H]palmitate (Fig. 1B, top autoradiograph, lanes  1 and 2), as observed previously (13). The ␣4 nAChR subunits are labeled in the absence or presence of the ␤2 nAChR subunits (Fig. 1B). Surprisingly, cells transfected with the ␣4⌬C did not incorporate a detectably reduced amount of [ 3 H]palmitate compared with the ␣4 nAChR subunit, suggesting that none of the cysteines in the M3-M4 loop were palmitoylated (Fig. 1B,  lanes 3 and 4). One-tenth of the IP eluate was run on a parallel SDS-PAGE and immunoblotted for IP capture efficiency with anti-FLAG antibody to detect the ␣4 nAChR subunit and with anti-C20 antibody to detect the ␤2 nAChR subunit (Fig. 1B, bottom blots).
Because [ 3 H]palmitate incorporation was similar between the WT ␣4 and the ␣4⌬C mutant nAChR subunit, the ␣4, ␣4⌬C, ␣4⌬⌬C, and ␣4C273S nAChR subunits were transfected and metabolically labeled with [ 3 H]palmitate. The ␣4 and ␣4⌬C nAChR subunits were labeled (Fig. 1C, lanes 1 and 4), but the ␣4⌬⌬C and the ␣4C273S mutant did not incorporate [ 3 H]palmitate (Fig. 1C, lanes 2 and 3), indicating that Cys 273 is the only cysteine in the ␣4 nAChR subunit that is palmitoylated. To determine whether [ 3 H]palmitate was linked to the ␣4 and ␤2 nAChR subunits by hydroxylamine-sensitive thioester bonds, an IP sample from ␣4␤2 lysate was run in parallel with an IP sample that was treated with hydroxylamine (Fig. 1D). The [ 3 H]palmitate was absent from the sample treated with hydroxylamine, indicating that labeling was due to palmitoylation of cysteine residues.
Immunocytochemistry further verified the changes in ␣4␤2 nAChR surface and total protein expression. tsA 201 cells were transfected transiently with the same plasmid combinations used in Fig. 2, A and B, and processed for immunocytochemistry. Single plane images were collected using identical settings. Cells were fixed and processed with or without permeabilization to qualitatively examine total or cell surface expression, respectively. The expression patterns observed by immunocytochemistry were comparable with those observed by cell surface and Western assays. All three of the mutants appeared to have increased surface immunostaining intensities when compared with the wt ␣4␤2 nAChR (Fig. 2C, top panels). When cells were fixed and permeabilized with 0.2% Triton X-100 and imaged to detect total ␣4␤2 nAChR expression, ␣4⌬C had increased staining intensity, ␣4⌬⌬C had comparable intensity, and ␣4C273S had decreased total intensity compared with WT ␣4 (Fig. 2C, bottom panels). Taken together, these data show that mutating the cysteines in the M3-M4 loop increase both total and cell surface expression, whereas mutating the cysteine in the M1-M2 loop decreases total protein expression but increases surface expression compared with WT ␣4␤2 nAChR levels.
If palmitoylation of ␣4 was the only variable in the experiment with 2-BP and if ␣4C273S affected expression only through an effect on palmitoylation, then the expected result with complete block of palmitoylation would be that 1) the expression ratio of wt ␣4␤2 nAChR would equal the expression FIGURE 2. Removal of cysteines from the ␣4 nAChR cytoplasmic loops increases the surface expression of the receptor. A, tsA 201 cells were transfected with untagged ␤2 and N-terminally FLAG-tagged ␣4, ␣4⌬C, ␣4⌬⌬C, or ␣4C273S nAChR subunits. Cell surface expression of ␣4␤2 was measured using an enzyme-linked immunoassay using an mAb against the extracellular domain of the ␤2 nAChR subunit (mAb 295). The absorbance data were normalized, and the WT ␣4␤2 nAChR was set to 100%, and data were expressed as means Ϯ S.E. The ␣4⌬C␤2, ␣4⌬⌬C␤2 and ␣4C273S␤2 all had significantly higher surface expression than ␣4␤2. *, p Ͻ 0.05; **, p Ͻ 0.001 (n ϭ 6). B, transiently transfected tsA 201 cells were lysed in 750 l of 2% Triton X-100 buffer. Soluble fractions (20 l) were analyzed by SDS-PAGE and Western blotting. Blots were cut horizontally so that the same blot could be probed with either mouse monoclonal anti-FLAG antibody to detect the ␣4 nAChR, goat polyclonal anti-␤2 antibody to detect the ␤2 nAChR or rabbit polyclonal anti-GAPDH antibody. Lysate from untransfected (UT) cells was run as a negative control. Molecular mass is in kDa along the left side of the blot. C, single-plane confocal images of transiently transfected tsA 201 cells were captured with a spinning disc confocal microscope. Top panel, to detect ␣4␤2 nAChR expression on the cell surface, fixed, non-permeabilized cells were immunolabeled with mAb 295 followed by Alexa Fluor 594-conjugated goat anti-rat secondary antibody. Bottom panel, cells in a parallel set of wells were fixed, permeabilized, and immunolabeled with mAb 295 and Alexa Fluor 594-conjugated goat anti-rat secondary antibody to detect total ␣4␤2 nAChR expression. All of the images were taken for the same exposure time with the same camera settings and processed identically. Scale bar, 10 M.

␣4 nAChR Subunit Palmitoylation
JUNE 29, 2012 • VOLUME 287 • NUMBER 27 ratio of ␣4⌬C because 2-BP would depalmitoylate both these constructs. In addition, 2) ␣4C273S and ␣4⌬⌬C should not decrease (i.e. the expression ratios ϭ 1) because they are already unable to be palmitoylated. Finally, 3) the ␣4C273S untreated mutant would equal the 2-BP-treated WT ␣4␤2 nAChR because removing the palmitoyl group in the WT ␣4 nAChR subunit should make the two constructs equivalent. On the other hand, if palmitoylation of the ␣4 nAChR was the only variable in the experiment with 2-BP and if ␣4C273S intrinsically affected expression independent of palmitoylation, then 4) the untreated ␣4C273S and ␣4⌬⌬C in the absence of 2-BP would be significantly different from the surface expression for ␣4⌬C and WT ␣4␤2 nAChR in the presence of 2-BP because an effect other than palmitoylation is modulating the expression of the constructs.
These data do not meet expectation 1 (WT ␣4␤2 and ␣4⌬C are significantly different, p ϭ 0.013, n ϭ 3), expectation 2 (both ␣4⌬⌬C and ␣4C273S significantly decrease when treated with 2-BP), or expectation 3 (WT ␣4␤2 decreases 76.1%, whereas ␣4C273S increases to 120%), but they do meet expectation 4 (␣4C273S and ␣4⌬⌬C increase in the absence of 2-BP, whereas ␣4⌬C and WT ␣4␤2 nAChR decrease in the presence of 2-BP). Thus, these data suggest that palmitoylation of ␣4C273S is not the only variable affecting protein expression and that the effect of ␣4C273S may be independent of palmitoylation. Total protein expression was not affected by treatment with 2-BP, as determined by analysis of parallel samples by Western blotting (Fig. 3B).
Overall, these data show that 2-BP down-regulates ␣4␤2 nAChR cell surface expression, consistent with previous ligand binding data (13). These findings suggest that ␣4Cys 273 is important for cell surface expression but do not allow us to conclude that palmitoylation of ␣4Cys 273 is responsible for the change in protein expression.
Cysteine Modifications Do Not Significantly Alter ␣4␤2 nAChR EC 50 or Ligand Binding Affinity-To determine whether the cysteine-to-serine amino acid replacement affects the EC 50 of activation of the receptor, we compared the activation of WT to mutant receptors using a fluorescent assay that measures changes in intracellular calcium. The concentration response effects of epibatidine on WT or cysteine mutants show only relatively small differences in EC 50 values (Fig. 4). The EC 50 for the WT ␣4, ␣4⌬C, ␣4⌬⌬C, and ␣4C273S are 25.2, 16.2, 23.7, and 20.2 nM, respectively (Table 1). These EC 50 values are well within the published range for rat ␣4␤2 nAChR functional data (18). There are no significant differences in EC 50 values between the WT and mutant receptors, as determined by one-way ANOVA analysis.
We also compared the relative binding affinities of the receptors for nicotine in a competitive inhibition assay (Fig. 5). The K i values were determined using a published K d of 0.015 nM (17). No significant differences in the K i values between ␣4 and ␣4-mutant containing receptors were seen (determined by one-way ANOVA), indicating that the intracellular cysteines do not affect ligand-binding affinity of surface receptors (Table  1). Thus, changes in the loop via cysteine mutations do not appear to significantly affect the EC 50 or ligand-binding affinity of the ␣4␤2 nAChR.

DISCUSSION
Palmitoylation is a post-translational modification that plays a significant role in the trafficking and function of many neuronal receptors. Previously, it has been reported that nAChRs are palmitoylated (13), but the specific palmitoylation sites have not been determined. Our study is the first to identify the site of palmitoylation on the ␣4 nAChR subunit as the sole intracellu-lar cysteine in the M1-M2 cytoplasmic loop. Additionally, we found that this mutant decreases total protein levels but increases expression levels at the cell surface compared with the WT ␣4 nAChR. Surface expression also was increased in the mutant lacking all of the cysteines in the M3-M4 loop, but no palmitoylation sites were identified in this loop. Additionally, intracellular cysteines do not affect ␣4␤2 nAChR ligand-binding affinity or localization to presynaptic terminals. Taken together, our findings suggest that cysteine residues, especially Cys 273 , play a prominent role in receptor trafficking to the cell surface rather than in synaptic targeting or function.
Our study demonstrates that the ␣4 nAChR subunit undergoes palmitoylation and identifies the specific amino acid at position 273. This cysteine is located near the distal end of the short (about six residues) M1-M2 loop in certain nAChRs and is highly conserved across species and subunits, including ␣2, ␣3, ␣4, ␣6, ␤2, and ␤4 nAChR subunits and the 5HT3B subunit. Proposed physical mechanisms by which palmitoylation could exert its effects in membrane proteins include increasing the hydrophobic length of a transmembrane domain, associating with the lipid bilayer, and tilting of a transmembrane domain (4). The position of ␣4Cys 273 at the distal end of the M1-M2 loop based on homology modeling with a Cys loop channel FIGURE 5. Cysteine mutations in the ␣4 nAChR subunit do not affect ␣4␤2 nAChR ligand binding affinity. Cultured tsA 201 cells expressing the various ␣4 constructs plus ␤2 were fixed with paraformaldehyde, permeabilized, and incubated with 400 pM epibatidine in the presence or absence of 10, 30, 100, 300, or 1000 nM nicotine overnight at 4°C. After several washes, the amount of bound epibatidine was quantified by scintillation counting. The data shown was normalized to the amount bound in the absence of nicotine and represent the mean Ϯ S.E., n ϭ 6 for ␣4; n ϭ 5 for ␣4⌬C; n ϭ 3 for ␣4⌬⌬C and ␣4C273S. Curves were fitted to a one-site competitive binding using a K d ϭ 0.015 nM, using the equation Y ϭ 100/(1 ϩ 10((log IC 50 Ϫ X)⅐HS), where Y is the percentage of the maximal effect at a given concentration (X), and HS is the Hill Slope (Hill slope ϭ 1).  from Caenorhabditis elegans (19) is closer to hydrophilic spaces of the cytoplasm and the ion permeation path than to the hydrophobic core of the membrane bilayer (Fig. 7). Access by the hydrophobic palmitoyl group on ␣4Cys 273 to a preferred hydrophobic environment of the membrane bilayer or within the transmembrane domains of the nAChR subunits likely would distort the closely packed transmembrane domains. The position of ␣4Cys 273 adjacent to the start of M2 suggests that more than one of the proposed physical mechanisms of palmitoylation effects might be relevant for ␣4␤2 nAChRs.
The C273S mutation leads to an increase in surface expression, but when the cells are treated with 2-BP, there is a decrease in ␣4␤2 nAChR cell surface expression. Why is there a discrepancy between the depalmitoylated mutant and using a palmitoylation inhibitor? 2-BP is a general palmitoylation inhibitor, so it inhibits the palmitoylation of many proteins in the cell beyond ␣4 nAChR subunits. A reasonable conclusion from the failure of expectations 1, 2, and 3 is that palmitoylation of ␣4Cys 273 was not the only variable in the experiments with 2-BP, making the results impossible to interpret strictly in terms of palmitoylation of ␣4Cys 273 . Additionally, 2-BP inhibits palmitoylation by an unknown mechanism of action and has been shown to have pleiotropic effects on other cellular functions including cellular metabolism (20). This generalized effect of 2-BP might impair the function of other proteins that traffic nAChRs to the cell surface. Alternatively, Cys 273 may regulate cell surface stability independent of palmitoylation. Therefore, we are unable to say unequivocally that palmitoylation is necessary for the surface expression of ␣4␤2 nAChRs.
It is interesting to note that the cysteine that is palmitoylated in our study has been suggested to undergo reactive oxygen species-regulated use-dependent inactivation in the ␣3␤4 nAChR (14). The authors propose that use-dependent inactivation of the ␣3␤4 nAChR occurs after the receptors have been repeatedly exposed to agonist over time due to mild elevations in reactive oxygen species (14,21,22). Receptor use can also enhance the depalmitoylation of proteins (23)(24)(25)(26)(27). Although we observed similar EC 50 and ligand-binding affinities between WT and mutant receptors, we have not examined whether usedependent inactivation and depalmitoylation of the M1-M2 cysteine residue are linked. Future studies are necessary to address this question.
Our work provides two possible models for the cell surface trafficking of the ␣4␤2 nAChR. Compared with the wt ␣4 nAChR subunit, the ␣4⌬C mutant has increased surface and total protein expression, suggesting that the increased surface expression is due to increased overall protein expression. The ␣4C273S mutant has decreased total protein expression compared with WT, but it still increases receptor expression at the cell membrane. ␣4⌬⌬C has increased surface expression but no change in total protein expression. These findings imply that the mutation of Cys 273 to a serine either allows the receptor to pass through the secretory pathway more efficiently or impairs the internalization of the receptor, resulting in accumulation at the cell membrane. Further studies are necessary to determine which of these conclusions is correct, or if there is another explanation.
Changes in nAChRs expression levels are associated with many neurological diseases and disabilities, including nicotine addiction, autism, Alzheimer disease, Parkinson disease, schizophrenia, and diabetic neuropathies. Here, we have discovered a novel site of palmitoylation on the ␣4␤2 nAChR and determined that cysteine residues modulate nAChR trafficking to the cell surface. Our findings add to the literature on nAChR trafficking and may lead to new therapeutic strategies for altering nAChR cell surface expression in nAChR-associated diseases in the future.