Resistance to Inhibitors of Cholinesterase 3 (Ric-3) Expression Promotes Selective Protein Associations with the Human α7-Nicotinic Acetylcholine Receptor Interactome

The α7-nicotinic acetylcholine receptor (α7-nAChR) is a ligand-gated ion channel widely expressed in vertebrates and is associated with numerous physiological functions. As transmembrane ion channels, α7-nAChRs need to be expressed on the surface of the plasma membrane to function. The receptor has been reported to associate with proteins involved with receptor biogenesis, modulation of receptor properties, as well as intracellular signaling cascades and some of these associated proteins may affect surface expression of α7-nAChRs. The putative chaperone resistance to inhibitors of cholinesterase 3 (Ric-3) has been reported to interact with, and enhance the surface expression of, α7-nAChRs. In this study, we identified proteins that associate with α7-nAChRs when Ric-3 is expressed. Using α-bungarotoxin (α-bgtx), we isolated and compared α7-nAChR-associated proteins from two stably transfected, human tumor-derived cell lines: SH-EP1-hα7 expressing human α7-nAChRs and the same cell line further transfected to express Ric-3, SH-EP1-hα7-Ric-3. Mass spectrometric analysis of peptides identified thirty-nine proteins that are associated with α7-nAChRs only when Ric-3 was expressed. Significantly, and consistent with reports of Ric-3 function in the literature, several of the identified proteins are involved in biological processes that may affect nAChR surface expression such as post-translational processing of proteins, protein trafficking, and protein transport. Additionally, proteins affecting the cell cycle, the cytoskeleton, stress responses, as well as cyclic AMP- and inositol triphosphate-dependent signaling cascades were identified. These results illuminate how α-bgtx may be used to isolate and identify α7-nAChRs as well as how the expression of chaperones such as Ric-3 can influence proteins associating with α7-nAChRs. These associating proteins may alter activities of α7-nAChRs to expand their functionally-relevant repertoire as well as to affect biogenesis and membrane trafficking of α7-nAChRs.

to investigate the effect of Ric-3 expression on the α7-nAChR interactome. A comparison of α7-nAChR associated proteins in both cell lines allows for the identification of receptor-protein interactions that occur with Ric-3 co-expression. Ric-3-mediated α7-nAChR associated proteins may interact with the receptor during and after direct interaction of Ric-3 with α7-nAChRs. During direct interaction with α7-nAChRs, Ric-3 may recruit other proteins to the receptor complex to facilitate surface expression. After the dissociation of Ric-3, proteins may associate with mature α7-nAChRs as a result of Ric-3-mediated surface expression. The comparison of α7-nAChR complexes from SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cells provides a method of identifying associated proteins, including those that may be essential for Ric-3-mediated enhancement of α7-nAChR surface expression.

Materials and Methods
Preparation of α-bgtx-Sepharose affinity beads Cyanogen bromide-activated Sepharose beads 4B (Sigma-Aldrich, St. Louis, MO) (1 g) were hydrated in 5 mL cold 1mM HCl for 30 minutes and washed with 500 mL 1 mM HCl over a coarse glass filter. The beads were then added to 7.5 mL coupling buffer (0.25 M NaHCO 3 , 0.5 M NaCl, pH 8.3) and subsequently centrifuged at 4°C for 5 minutes at 1,500 x g. The supernatant was discarded, and the pellets were resuspended in 7.5 mL coupling buffer containing 4 mg of α-bgtx (Life Technologies, Eugene, OR). Bead/ligand mixtures were incubated with gentle agitation at 4°C for 18 hours. The beads were subsequently pelleted and resuspended in 7.5 ml of 0.2 M glycine in 80% coupling buffer, 20% ultrapure water and gently agitated overnight at 4°C to block unreacted groups on the beads. The beads were then washed several times over a course glass filter, first with 100 mL of 0.1 M NaHCO 3 , 0.5 M NaCl, pH 8.0, then 100 mL of 0.1 M NaCH 3 CO 2 , 0.5 M NaCl, pH 4.0, again with 100mL of 0.1 M NaHCO 3 , 0.5 M NaCl, pH 8.0, 100 mL coupling buffer, and lastly twice with 100 mL Tris-buffered saline (TBS: 150 mM NaCl, 50 mM Tris, pH 7.5). Washed beads were resuspended in TBS for storage at 4°C. Prior to use, α-bgtx-affinity beads were uniformly resuspended into a slurry and were centrifuged at 4°C for 5 minutes at 1,500 x g. Pelleted beads were resuspended to make a 50/50 slurry with homogenization buffer (100 mM NaCl, 25 mM NaH 2 PO 4 , pH 7.4) before use.
were isolated following centrifugation at 10,000 x g for 10 minutes at 4°C. Membrane pellets were then homogenized in solubilization buffer (100 mM NaCl, 25 mM NaH 2 PO 4 , 1% Triton X-100, pH 7.4) with 40 strokes of a Potter-Elvehjem glass homogenizer and incubated for 30 minutes at 4°C with agitation to solubilize membrane-bound proteins. Following centrifugation at 12,600 x g for 10 minutes at 4°C, the solubilized membrane extract was recovered in the supernatant. All buffers used to isolate the solubilized membrane extract were supplemented with protease inhibitors (Roche Applied Science, Indianapolis, IN). Protein content of solubilized membrane extracts was determined using a BCA assay (Pierce).

Ric-3 immunoblotting
Detergent solubilized receptor preparations (12.5 μg protein per lane) of SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cell lines were used for immunoblotting. Samples were incubated at 60°C for 1 hour with 47.6 mM TCEP and 1x NuPAGE sample buffer (Life Technologies, Eugene, OR), then alkylated in 76.3 mM iodoacetamide at room temperature for 1 hour in the dark. Proteins were separated by SDS-PAGE and transferred at 100 V for 90 minutes to a nitrocellulose membrane (Thermo Fisher Scientific Inc., Waltham, MA). The membrane was blocked in 5% nonfat milk in TBST buffer (150 mM NaCl, 10 mM Tris, 0.05% Tween-20, pH 8) for 1 hour at room temperature and then was incubated with anti-Ric-3 antibodies (ab112911, Abcam, Cambridge, MA) diluted 1:500 in 3% milk TBST buffer overnight at 4°C. After washing with TBST, the membrane was incubated with peroxidase conjugated mouse anti-rabbit secondary antibody (211-032-171, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1:50,000 in 3% milk TBST buffer. The membrane was then washed three times with TBST and twice with TBS (150 mM NaCl, 10 mM Tris, pH 8) before being incubated for 5 minutes in SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific Inc., Waltham, MA). Reactive bands were visualized on film after a 3 minute exposure. Ric-3 antibodies were subsequently stripped from blots using Restore Western Blot Stripping Buffer (Thermo Fisher Scientific Inc., Waltham, MA) and probed a second time with anti-GAPDH antibodies diluted 1:1000 in 3% milk TBST buffer overnight at 4°C (14C10, Cell Signaling, Danvers, MA). Following the incubation with anti-GAPDH antibodies, the protocol was the same as described above. Bands were visualized on film after a 30 second exposure.

α7-nAChR and associated protein complex isolation
Immediately following the isolation of solubilized membrane extracts, a volume containing 3 mg of solubilized protein was incubated with 200 μl of the 50/50 α-bgtx-affinity bead/homogenization buffer slurry for 4 hours at 4°C with gentle agitation. Control samples were solubilized receptor preparations treated with 5 μM MLA (Sigma-Aldrich, St. Louis, MO) during the affinity-immobilization incubation. Following the incubation, α-bgtx-affinity beads and bound protein were transferred to Pierce Spin Cups (Thermo Fisher Scientific Inc., Waltham, MA) and washed several times with solubilization buffer. After washing, the total affinity-immobilized α7-nAChR content was measured using a 125 I-α-bgtx radioligand binding assay or the isolated proteins were eluted for mass spectrometric analysis (Fig 1).

Radioligand binding assays
The use of α-bgtx to affinity immobilize α7-nAChRs and concurrently detect them is possible because α7-nAChRs contain multiple α-bgtx binding sites [31]. Affinity-immobilized α7-nAChR content was determined by incubating the membrane protein-α-bgtx-affinity bead complex with 5 nM 125 I-α-bgtx (Perkin Elmer, Boston, MA) for 1 hour at room temperature. Non-specific binding was determined by the inclusion of 1 μM unlabeled α-bgtx before addition of 125 I-α-bgtx. Following incubation with 125 I-α-bgtx, beads were washed three times with solubilization buffer and measured using a Wallac 1275 Minigamma gamma counter.

Sample preparation, precipitation, and in-solution trypsin digestion
Affinity beads with bound α7-nAChRs and associated proteins were washed three times with solubilization buffer followed by a single high salt solubilization buffer wash (2M NaCl, 25 mM NaH 2 PO 4 , 1% Triton X-100, pH 7.4) to reduce inclusion of non-specific proteins. Immobilized proteins were specifically eluted from the α-bgtx-affinity beads by incubation with 100 μl 1 M carbachol (Sigma-Aldrich, St. Louis, MO) in 20 mM HEPES, pH 8.0 for 30 minutes with agitation every 5 minutes at room temperature. α-Bgtx-affinity beads were allowed to sediment and the eluted proteins in the supernatant were removed and stored at -80°C until preparation for mass spectrometry analysis. Protein content was determined using a Pierce BCA Protein Assay Kit. To prepare for mass spectrometric analysis, samples were thawed and disulfide/sulfhydryl residues were reduced with 47 mM TCEP in 20 mM HEPES, pH 8.0 for 1 hour at 60°C. Samples were alkylated with 83 mM iodoacetamide in 20 mM HEPES, pH 8.0 for 1 hour in the dark at room temperature. Samples were then concentrated and purified via precipitation using a BioRad ReadyPrep 2-D Cleanup Kit (BioRad, Hercules, CA). Precipitated protein was resuspended in 50 mM ammonium bicarbonate, pH 7.8 supplemented with 100 ng trypsin (Promega, Madison, WI) and digested overnight in-solution at 37°C.

Liquid chromatography & mass spectrometry of protein digests
Tryptic digests were analyzed at the Brown University (Providence, RI) NSF-EPSCoR Proteomics Core Facility with an Agilent 1200 (Agilent Technologies) high performance liquid chromatography (HPLC) in-line with a Q Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific Inc., Waltham, MA). Separation of peptides was achieved using a 12 cm Monitor C18 (Column Engineering) reversed-phase column with an internal diameter of 75 μm and integrated 4 μm electrospray ionization tip (self-pack PicoTip, New Objective). Peptides were eluted during a 50 minute linear gradient of 100% solvent A (0.1 M acetic acid in water), 0% solvent B (0.1 M acetic acid in acetonitrile) to 60% solvent A, 40% solvent B) at a flow rate of 200 nl/min and introduced into the mass spectrometer via electrospray ionization (ESI) for analysis (data-dependent mode with 30-second dynamic exclusion with one MS scan followed by nine MS/MS scans). Peak lists of MS/MS spectra were created using msconvert.exe (v. 2.2.3300) available in the ProteoWizard tool [32]. Data were bioinformatically matched against a concatenated target-decoy (sequence-reversed) Homo sapiens database (Uniprot, April 2013) using the Mascot algorithm (Matrix Science, Boston, MA). Database searches used the following parameters: Up to two missed trypsin cleaves allowed, 7 ppm MS tolerance, 20 ppm MS/MS tolerance, fixed carbamidomethyl modification, and variable methionine oxidation modification. Mascot search DAT files were loaded into ProteoIQ (Premier Biosoft) for further analysis. Proteins were filtered using a minimum peptide length of 6 amino acids, 1% protein false-discovery rate (FDR) and 90% group probability of correct identity assignment using the PROVALT and ProteinProphet algorithm respectively, presence in 2 or more independent replicates, and 0% probability in controls [33][34][35]. Protein probabilities represent the probability of correct assignment of all observed peptides in a protein group to the identified protein. Both the PROVALT and ProteinProphet algorithm are integrated into ProteoIQ. Only Top and Co-Top identifications, i.e. identifications which include all peptide data in a protein group, were used. Each cell line was analyzed with five biological replicates (Fig 1). Identified proteins were categorized by their reported Gene Ontology (GO) biological process terms using Database for Annotation, Visualization and Integrated Discovery (DAVID) [36]. If an identified protein did not have a GO term for associated biological processes, Protein ANalysis THrough Evolutionary Relationships (PANTHER) was used for classification [37,38]. If neither classification system had an entry for an identified protein, the protein was classified as unattributed.

α7-nAChR associated proteins
Carbachol elutions of α-bgtx-affinity immobilized α7-nAChRs from both SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cell preparations were analyzed using a Q Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific Inc., Waltham, MA). We set the following a priori inclusion criteria parameters to identify proteins: 1% protein FDR, 90% group probability of correct identity assignment, and the presence in two or more of five independent biological replicates with 0% probability of correct identity assignment in controls as determined by the ProteinProphet algorithm. Using these criteria, the α7-nAChR was identified in all SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cell replicates with 100% and 98% probability, respectively, by way of the peptide FPDGQIWKPDILLYNSADER ( Table 1). The identified α7-nAChR peptide was not identified as a peptide from the reported sequence of the CHRFAM7A protein product (Accession number Q494W8). Ric-3 was detected in SH-EP1--hα7-Ric-3 cell samples and met all inclusion criteria, but it was associated with a borderline probability score of 88% (compared to the preset criteria of 90%). This may reflect the fact that Ric-3 is only transiently associated with α7-nAChRs. Not all α7-nAChRs will be interacting with Ric-3 at the time of α-bgtx-affinity bead isolation. The 2 M NaCl washes for all samples were also analyzed to confirm that α7-nAChRs were not eluted during the washing of material bound to α-bgtx-affinity beads. Neither α7-nAChR peptides nor Ric-3 peptides were identified in the 2 M NaCl bead wash from SH-EP1-hα7-Ric-3 cell samples. Identification of the α7-nAChR in SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cell samples confirms that α7-nAChRs were eluted from the α-bgtx-affinity beads by the cholinergic agonist carbachol. Neither Detergent solubilized membrane extracts were incubated with α-bgtx-affinity beads for 4 hours at 4°C. Protein-α-bgtx-affinity bead complexes were incubated with 5 nM 125 I-α-bgtx for 1 hour at room temperature. Non-specific binding was determined in controls by the inclusion of 1 μM unlabeled α-bgtx to preparations prior to the addition of 125 I-α-bgtx. Following incubation with 125 I-α-bgtx, beads were washed three times with solubilization buffer and measured. Comparable 125 I-α-bgtx binding activity of protein-α-bgtx-affinity bead complexes isolated from SH-EP1-hα7-Ric-3 (56 ± 15 fmol/mg, in blue) and SH-EP1-hα7 (49 ± 9 fmol/mg, in green) was observed (Student's t test, p = 0.40) while SH-EP1 preparations (in purple) did not show α-bgtx binding activity. No 125 I-α-bgtx binding to protein-α-bgtx-affinity bead complexes was observed in samples treated with 5 μM MLA confirming α7-nAChR specificity (Student t test, p < 0.05). SH-EP1-hα7-Ric-3 and SH-EP1-hα7 125 I-bgtx binding activity was analyzed with five independent biological replicates. MLA treated samples and SH-EP1 125 I-α-bgtx binding activity were analyzed with three independent biological replicates.
α7-nAChR nor Ric-3 peptides were identified in carbachol-eluted samples prepared from SH-EP1 cells, which lack expression of both proteins.
A peptide corresponding to α7-nAChR subunits was identified in α-bgtx-enriched samples of both SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cell lines. Data analysis was performed using Pro-teoIQ version 2.7 Protein inclusion criteria include 1% protein FDR, minimum peptide length of six amino acids 90% probability, identification in 2 or more of 5 replicates (i.e., Data sets), and 0% probability in controls. FDRs were determined using the PROVALT algorithm and probabilities were determined with the ProteinProphet algorithm through ProteoIQ analysis. Only Top and Co-Top identifications were considered.
Proteins identified in our analysis of the α7-nAChR interactome are most likely components of large protein complexes and may either be associating directly with the receptor or with another member of the complex. Comparison of carbachol-eluted proteins from SH-EP1-hα7-Ric-3 and SH-EP1-hα7 identified thirty-nine Ric-3-promoted α7-nAChR associated proteins ( Table 2). Fourteen of the thirty-nine proteins identified as Ric-3-mediated have previously been reported as associated with a cellular process known to affect protein expression ( Table 2, category labeled blue). These fourteen Ric-3-mediated associated proteins may be directly or indirectly recruited by Ric-3 to facilitate receptor assembly and targeting. In addition to proteins associated with protein expression, seven proteins are associated with protein turnover, four with signaling, and fourteen associated with other processes ( Table 2). In total, seven of the thirty-nine proteins have functions previously shown to affect nAChRs ( Table 3). Six of the thirty-nine proteins listed in Table 2 were identified with a single unique peptide in two or more replicates and are summarized in Table 4. Peptide-level detail for all thirty-nine proteins is provided in S1 Table.  Receptor-α-bgtx bead complexes were eluted with carbachol from α-bgtx-affinity resin, reduced, alkylated, precipitated and digested with trypsin in solution. Tryptic peptides were then analyzed using a Q Exactive mass spectrometer. Thirty nine Ric-3 stimulated proteins were identified through comparison of carbachol-eluted proteins from SH-EP1-hα7-Ric-3 and SH-EP1-hα7 α-bgtx-affinity immobilized samples. Each condition was analyzed with five replicates (i.e., Data sets). Data analysis was performed using ProteoIQ version 2.7. Protein inclusion criteria included 1% protein FDR, minimum peptide length of six amino acids, 90% probability, identification in 2 or more of 5 replicates, and 0% probability in controls. FDRs were determined using the PROVALT algorithm and probabilities were determined with the ProteinProphet algorithm through ProteoIQ analysis. Only Top and Co-Top identifications were considered. Biological processes are listed as determined by DAVID analysis of Gene Ontology (GO) terms. Biological process GO terms for six proteins are not available. Biological processes for three of these proteins were available through PANTHER analysis (denoted by "(P)") and the remaining three are listed as unattributed. Each protein listed is categorized as potentially involved with surface expression, protein turnover, signaling, or associated with biological processes not included in the previous categories as "Other proteins".
Seven of the thirty-nine proteins identified as Ric-3-mediated have reported functions which have been previously shown to affect nAChRs. The cited literature reporting the relationships between each of the listed proteins and nAChRs is categorized as either indicating a direct association to the listed proteins specifically, or linked by a previously associated class of proteins (e.g., tyrosine phosphatases), or by both. Two proteins, cAMP-dependent protein kinase type I-alpha regulatory subunit and inositol 1,4,5-trisphosphate receptor (IP 3 R) type 1 are listed as both since the literature does not always distinguish either specific proteins in the PKA complex nor which IP 3 R type is being discussed. These proteins are further separated by whether their functions have been associated with α7-nAChRs only, other nAChR subtypes but not α7-nAChRs or with α7 and other nAChR subtypes.
Six proteins from Table 2 were identified by one unique peptide. Mass to charge ratios (m/z), charges (z), peptide sequence, and Mascot ion scores are listed for each single-peptide-based identification. For analysis, replicates were assigned replicate numbers one through five. Each identification is listed separately for each replicate number in which the single-peptide was observed.
Ninety-seven proteins were uniquely isolated on α-bgtx-affinity beads from SH-EP1-α7 cells that were not identified in preparations from SH-EP1-hα7-Ric-3 cells. These proteins represent possible protein associations with α7-nAChRs that form in the absence of Ric-3 expression (S2 Table). A total of 625 proteins that met the inclusion criteria were identified in both cell lines (S3 Table). These proteins common to both cell lines may represent general α7-nAChR interacting proteins or non-specific interactions with α-bgtx-affinity beads. Analysis of the cellular compartment GO terms for proteins unique to SH-EP1-hα7-Ric-3 samples and those unique to SH-EP1-α7 samples suggests a difference in cellular distribution of the receptors between the two cell lines (S4 Table). The reported Ric-3-mediated interactome consists of proteins associated with the cytosol, intracellular membranes, and the ER. Many of the identified Ric-3-mediated proteins are reported to be localized in the ER, which agrees with previous reports that Ric-3 is a chaperone predominantly expressed in the ER. In comparison, none of the proteins identified as unique in SH-EP1-α7 samples have been reported to be localized in the ER.

Discussion
To identify the Ric-3-mediated α7-nAChR interactome, specific α-bgtx-binding proteins were isolated from cells stably expressing the receptor alone or the receptor and Ric-3 using α-bgtx affinity beads and specifically eluted using a cholinergic agonist. Eluted proteins were digested with trypsin and the resulting peptides were analyzed with mass spectrometry. Analysis of peptide fragmentation spectrum was used to identify the proteins that associated with α7-nAChRs in samples isolated from cells expressing, or not expressing Ric-3. Identified in this study were thirty-nine proteins whose association with α7-nAChR was mediated by co-expression of Ric-3.

Cellular model
The two cell lines utilized, SH-EP1-hα7-Ric-3 and SH-EP1-hα7, heterologously express human α7-nAChRs and differentially express the chaperone Ric-3. These cell lines were chosen to study the Ric-3-mediated α7-nAChR interactome for several reasons. First, the use of these two transfected cell lines provides a level of control for selective expression of the two proteins of interest that would be more difficult to achieve using endogenous expression models. Second, these two cell lines are a reliable source of α7-nAChR and Ric-3 expression. Previously, fifty-five α7-nAChR interacting proteins were identified by tandem mass spectrometry by comparison of α-bgtx affinity immobilized protein from α7-nAChR wild type and α7-nAChR knockout mouse brain tissue [29]. However, α7-nAChR peptides were not identified by tandem mass spectrometry in this study. Although the α7-nAChR was identified in the study presented here, none of the fifty-five α7-nAChR interacting proteins identified in the previous study were identified. In addition to the important distinction that we identified the α7-nAChR while the previous study did not, there are several differences between the present study and the previous study that may account for the disparity between the two identified interactomes. Substantial modifications were made to the α-bgtx-affinity immobilization protocol and mass spectrometry sample preparation in order to maximize isolation and detection of α7-nAChRs. The model system in the investigation presented here is also human in origin and used clonal cells of a single morphology as compared to the heterogeneity of the cell types found in of the murine brain. The work shown here investigates a more focused, Ric-3-mediated α7-nAChR interactome, rather than a general α7-nAChR interactome, which was the aim of the previous study.

Ric-3-mediated α7-nAChR associated proteins
The role of the molecular chaperone Ric-3 in α7-nAChR expression has been investigated by a number of different methods in multiple models and previous reports have demonstrated an increase in cell surface expression of α7-nAChRs in cells also expressing Ric-3 [9,23,26,39]. Human cells lines were used to identify α7-nAChR protein-associations that appear with coexpression of Ric-3. Of a total of thirty-nine identified members of the Ric-3-mediated α7-nAChR interactome, fourteen proteins have been previously reported to be associated with a process known to affect protein expression. Of the remaining proteins, five are associated with signal transduction/intracellular signaling, seven with protein catabolism and/or autophagy, and fourteen that do not have a reported connection to α7-nAChR surface expression, signaling, protein catabolism or autophagy ( Table 2). The fourteen proteins associated with protein expression as well as the seven proteins associated with protein catabolism and/or autophagy may represent receptor-protein interactions contributing to the life-cycle of α7-nAChRs (Fig 4). Ric-3 was identified by mass spectrometry with 88% probability (versus the a priori 90% inclusion criteria) and met all other inclusion criteria. The probability of correct identification of Ric-3 may fall outside the preset inclusion criteria due to its transient interaction (interacting intracellularly) with α7-nAChR. That transient interaction of Ric-3 nevertheless may lead to the interactions with the α7-nAChR protein complex identified in this study. Endoplasmic reticulum localized associated proteins Evidence suggests that two of the Ric-3-mediated α7-nAChR-associated proteins are involved in early stages of protein expression in the ER. First, translocon-associated protein subunit gamma is a TRAP protein which interacts with SEC61 and is involved in protein translocation in the ER [40,41]. Second, dolichol-phosphate mannosyltransferase, is an enzyme that may be Ric-3 Mediated Proteins of the α7-nAChR Interactome involved in N-glycosylation (Table 3). N-glycosylation and subsequent glucose trimming is an important regulatory step in protein expression in the ER, and α7-nAChRs have been shown to be glycosylated [3,[42][43][44][45]. Further investigation is required to deduce if dolichol-phosphate mannosyltransferase may be involved with α7-nAChR N-glycosylation. In addition, seven proteins associated with protein folding and receptor assembly were identified: calnexin; calreticulin; peptidyl-prolyl cis-trans isomerase A; DnaJ homolog subfamily B member 11; hypoxia upregulated protein 1; t-complex protein 1 subunit epsilon; and reticulocalbin-3. Calnexin and calreticulin are two ER chaperones which bind to unfolded or misfolded proteins and are central to a cycle of repeated folding and unfolding [45]. The calnexin/calreticulin cycle is a well-studied ER mechanism for achieving proper protein folding and receptor assembly. The calnexin/calreticulin cycle has also been identified previously as important for muscle nAChR localization (Table 3) [45][46][47][48]. However, the interaction of both chaperones with α7-nAChRs has not been previously reported. In addition to the two chaperones, a number of other proteins have been shown to have a role in the calnexin/calreticulin cycle. Peptidyl-proyl cis-trans isomerases such as peptidyl-prolyl cis-trans isomerase A may also contribute to the calnexin/calreticulin cycle and have been shown to enhance α7-nAChR folding in the ER (Table 3) [45,49]. Moreover, BiP, another chaperone associated with protein expression, has been previously shown to associate with α subunits of the muscle type nAChR [50][51][52]. BiP is a member of a large ER protein complex, and while BiP itself was not identified as a α7-nAChR-associated protein in this study, two other members of the BiP complex were identified: DnaJ homolog subfamily B member 11 and hypoxia up-regulated protein 1 [53]. The identification of DnaJ homolog subfamily B member 11 and hypoxia up-regulated protein 1 as proteins in complex with α7-nAChR suggests the possible involvement of the BiP complex in facilitating protein folding in the ER. The interaction of muscle-type nAChR subunits with BiP is short lived [48]. If the interaction with α7 subunits is similarly short lived, BiP itself would not be identified in this study. T-complex protein 1 subunit epsilon is a member of the BBS/CCT complex which facilitates protein folding through a complex mechanism of trapping unfolded proteins that undergo a series of ATP hydrolysis-driven confirmation changes to induce proper folding [54]. CCT complexes have been associated previously with a myriad of proteins but not with nicotinic subunits [55]. Additionally, reticulocalbin-3 is a calcium binding protein localized to the ER and has been shown to facilitate maturation of certain proteins. Based on its identification in the current study, reticulocalbin-3 may have a similar function in the biosynthesis of α7-nAChRs [56,57].

Associated proteins localized in the Golgi complex
Following proper folding and receptor assembly, nicotinic receptors are transported to the Golgi complex before being transported to the cell surface. Once at the plasma membrane, receptors may undergo endocytosis to be recycled to the Golgi complex, recycled back to the plasma membrane, or be degraded. Three proteins that were identified as regulated through Ric-3 in SH-EP1-hα7-Ric-3 cells are associated with protein trafficking. Gamma-adducin is a membrane-cytoskeleton-associated protein that promotes protein exit from the Golgi complex by remodeling the actin network surrounding the Golgi complex. Optineurin is a protein vital to the maintenance of Golgi complex structure in addition to being implicated in trafficking from the Golgi complex to the plasma membrane [45]. ADP-ribosylation factor 4 is associated with recycling proteins from endosomes to the trans-Golgi network [58].

Identified kinase and phosphatase associated proteins
Both kinase and phosphatase activity has been implicated in nAChR up-regulation [6,59,60]. One kinase subunit and one phosphatase were identified: cAMP-dependent protein kinase type I-alpha regulatory subunit and tyrosine-protein phosphatase non-receptor type 14. Tyrosine dephosphorylation has been shown to increase α7-nAChR surface expression in oocytes by promoting exocytosis of intracellular receptor pools (Table 3) [59,61]. Conversely, tyrosine phosphatase activity has been shown to promote muscle-type nAChR turnover, emphasizing how nAChR subtypes may respond differently to the same modification [6,62]. Kinase activity of cAMP-dependent protein kinase (PKA) has been shown to increase α7-nAChR surface expression in neonatal rat sympathetic neurons as well as in human embryonal kidney cells [63,64]. PKA enzymes are comprised of four subunits, two catalytic and two regulatory [65]. The cAMP-dependent protein kinase type I-alpha regulatory subunit has previously been shown to colocalize with cholinergic markers [66]. Activation of α7-nAChRs has also been shown to stimulate PKA activity (Table 3) [67]. The identification of cAMP-dependent protein kinase type I-alpha regulatory subunit, coupled with these previous observations suggest that PKA activity may be linked to α7-nAChRs through the association of one of the enzyme's regulatory subunits. PKA activity in turn may have a diverse effect through other pathways leading to numerous biological processes, such as enhancement of synaptic efficiency and nicotinestimulated long term potentiation [68][69][70]. Determining whether the effects of kinases and phosphatases are through direct phosphorylation or dephosphorylation of nAChRs or effects upon a member of the nAChR interactome requires additional study. There may also be a temporal component with phosphorylation or dephosphorylation occurring at different stages of nAChR biogenesis.

Protein turnover, autophagy and apoptosis related proteins
Many of the mechanisms and pathways that are utilized in receptor turnover may overlap with other mechanisms such as autophagy [71]. Seven of the Ric-3-mediated α7-nAChR-associated proteins identified have been reported to play a role in receptor turnover, apoptosis or autophagy: nuclear receptor coactivator 4, autophagy-related protein 9A, ubiquitin-like modifieractivating enzyme 1, LIM domain only protein 7, calcium-binding and coiled-coil domain-containing protein 2, KN motif and ankyrin repeat domain-containing protein 2, and tax1-binding protein 1 [72][73][74][75][76][77][78][79][80][81]. Several mechanisms may regulate the association of autophagy with α7-nAChR only when Ric-3 is expressed. The associated proteins could also be involved in other pathways related to autophagy, such as protein catabolism [71]. In theory, with Ric-3 coexpression, more α7-nAChRs reach the surface of the cell and are subject to mechanisms regulating receptor turnover. In cells in which dramatically fewer α7-nAChRs reach the cell surface (i.e., cells in which Ric-3 is not co-expressed), the proteins involved with such turnover functions would be diminished as well.

Signal transduction and intracellular signaling associated proteins
In addition to the surface expression-related proteins described above, Ric-3 co-expression appears to enhance association of α7-nAChRs with proteins involved in signal transduction and intracellular signaling. These include: Inositol 1, 4, 5-trisphosphate receptor type 1; cell cycle progression protein 1; Rho guanine nucleotide exchange factor 17; and angiopoietinrelated protein 2. These interactions are of interest because Ric-3-mediated co-expression may promote subsequent signaling cascades. Inositol 1, 4, 5-trisphosphate receptor (IP 3 R) type 1 is associated with intracellular Ca 2+ release and signaling. Nicotine stimulation of α7-nAChRs has been shown to lead to Ca 2+ flux through IP 3 R type 1 and through activation of phospholipase C (PLC) ( Table 3) [82][83][84][85]. In addition to the effect of α7-nAChRs on the activity of IP 3 Rs, it has also been shown that α7-nAChRs colocalize with IP 3 Rs in PC12 cells [85]. IP 3 Rs have also been shown to colocalize with muscle-type nAChRs at the neuromuscular junction in rat skeletal muscle [86]. Expanding on what was shown previously, the association of IP 3 Rs with α7-nAChR identified in this study reflects a direct interaction between the IP 3 R and the α7-nAChR interactome. Cell cycle progression protein 1, Rho guanine nucleotide exchange factor 17, and angiopoietin-related protein 2 are associated with RhoA GTPases and may be involved in a number of processes [87][88][89]. Angiopoietin-related protein 2, for example, has been linked to cellular processes such as angiogenesis and cell motility, and members of the small RhoA GTPase family participate in the endocytosis of muscle-type nAChRs [90][91][92][93][94][95].

Additional identified associated proteins
Fourteen proteins were identified that are not currently linked to α7-nAChR surface expression, signal transduction/intracellular signaling, or protein catabolism and/or autophagy. These fourteen proteins are associated with the cytoskeletal organization; developmental processes; ion transport; nucleobase, nucleoside, nucleotide, and nucleic acid metabolic processes; biosynthetic processes; response to stress, or do not currently have a well-defined associated biological process. Several of the additional proteins identified are associated with the cytoskeleton and developmental processes (i.e., SUN domain-containing protein 2; keratin, type I cytoskeletal 15; keratin, type II cuticular Hb4; keratin, and type II cytoskeletal 75) and may be involved in the subcellular localization of nAChRs. Keratin however is often considered to be non-specific contaminant in mass spectrometry investigations and these proteins would have to be investigated further to confirm specificity to α7-nAChRs. Ferritin light chain is a subunit of the protein ferritin, which is involved in the transport of iron [96]. Ferritin light chain was shown previously to be enriched in autophagosomal fractions from cancer cell lines as was calcium-binding and coiled-coil domain-containing protein 2, optineurin, autophagy-related protein 9A, all of which were also identified in this study [76]. Several identified proteins were associated with nucleobase, nucleoside, nucleotide, and nucleic acid metabolic processes: 5'nucleotidase, FAD synthase, and TRMT1-like protein. Of these three proteins, 5'-nucleotidase is of interest as it is a marker for types of lipid rafts and during hypoxia is involved with nAChR-simulated adenosine production [97,98]. The biological process of Erythrocyte band 7 integral membrane protein was characterized by DAVID as protein complex assembly though this attribution refers to the proteins ability to form homo-oligomers [99]. Erythrocyte band 7 integral membrane protein is of particular interest due to its previous association with lipid rafts and possible regulation of ion channel activity [100]. The protein LYRIC is a marker found in numerous cancer cell lines [101,102]. Peroxidasin homolog is an extracellular matrix component that may be associated with reactive oxygen species metabolism [103,104]. There is currently no literature reporting on the biological processes of BTB/POZ domain-containing protein 2, RNA-binding protein 33, and uncharacterized protein "TPM1" (Uniprot Accession F5H7S3). These proteins represent a population with an assortment of different biological functions that warrant further investigation to discern the functionality of their relationship with α7-nAChR.

Conclusions
Receptor-protein interactions are dynamic and dependent upon many factors. Identifying α7-nAChR-associating proteins as described in this study captures a snapshot of possible interactions under standard tissue culture conditions. A single peptide of the human α7-nAChR subunit was detected in all SH-EP1-hα7-Ric-3 and SH-EP1-hα7 samples. This reproducibly identified single α7-nAChR subunit peptide would be ideal for absolute quantitation using mass spectrometry that may be of interest for future studies investigating α7-nAChR expression.
SH-EP1-hα7-Ric-3 and SH-EP1-hα7 cells are ideal for identifying Ric-3-mediated α7-nAChR-associating proteins though it is possible that in this model, interactions are present that would not occur endogenously in native cells. It is therefore important to develop appropriate methodologies to continue these investigations in models that endogenously express α7-nAChRs, such as SH-SY5Y cells [27]. Additional protein interactions with α7-nAChRs and other nAChR subtypes have been reported by other groups that were not identified in this study [5,105]. Our inability to detect these previously identified α7-nAChR-associated proteins may reflect the ability of some proteins to compete with α-bgtx binding, and thus be affected during the α-bgtx affinity bead incubation. For example, the three-fingered toxin family member Lynx1 (aka SLURP2) has been shown to interact competitively with α-bgtx for binding to α7-nAChRs and was not identified in this study [106,107].
This investigation expands upon our previous work from a murine model to a human model of α7-nAChR-associating proteins. The work described here is an example of how αbgtx-affinity may be harnessed as a tool for proteomic investigations of α7-nAChRs. Here we investigated receptor-protein interactions mediated by the differential expression of the Ric-3 chaperone. This approach can be applied to any protein to investigate possible alterations on the α7-nAChR interactome. Furthermore, this approach reproducibly identified a tryptic peptide of the α7-nAChR subunit. This peptide was identified in all SH-EP1-hα7-Ric-3 and SH-EP1-hα7 samples and was not observed in SH-EP1 samples. The size and reproducibility of this peptide could be used for absolute quantitation of α7-nAChRs by mass spectrometry using a heavy-labeled variant of the peptide. The study reported here presents a unique investigation of the role of Ric-3 in modification of the proteins associating with α7-nAChR. Identifying these proteins as members of the α7-nAChR macromolecular complex provides vital insight for understanding α7-nAChR surface expression and may assist in the identification of future therapeutic targets.
Supporting Information S1 Fig. Uncropped Fig 3 Ric-3 and GAPDH immunoblots. Immunoblots for investigation of Ric-3 and GAPDH immunoreactivity in SH-EP1-hα7-Ric-3 (A) and SH-EP1-hα7 (B) solubilized membrane extracts. Blots are initially probed for Ric-3 immunoreactivity, stripped, and subsequently probed again for GAPDH immunoreactivity to confirm consistent gel loading. (TIF) S1 Table. Characterization of each peptide identified to infer the identity of the 39 Ric-3-promoted α7-nAChR-associated proteins reported in Table 2. "Group number" lists all proteins, 1 through 39, in the order in which they were grouped by ProteoIQ during analysis. Only Top and Co-Top protein identifications, i.e. only proteins identifications that can account for all peptide information within a protein group, were analyzed. For all 39 identified proteins, all Top and Co-Top identifications were either different isoform entries for protein products of the same gene or alternative database entries. Uniprot accession numbers, protein names, and gene names are provided for each Top and Co-Top entry. Also described per Top and Co-Top entry are probability score, protein score, the number of unique peptides identified, and sequence coverage. The observed mass to charge (m/z), charge (z), ion score, peptide sequence, and possible modifications are listed for every peptide identified for both Top and Co-Top entries. For analysis, replicate samples were assigned replicate numbers 1 through 5. Each peptide identification is listed separately for each replicate number in which the peptide was observed. (XLSX) S2 Table. Ontological grouping of α7-nAChR associated proteins independent of Ric-3 expression. Receptor protein complexes were eluted from α-bgtx-affinity beads complexes using carbachol, reduced, alkylated, precipitated and digested with trypsin in solution. Tryptic fragments were analyzed using a Q Exactive mass spectrometer. Ninety-seven proteins were identified as interacting with α7-nAChR only in the absence of Ric-3 expression by comparing α-bgtx isolated proteins from SH-EP1-hα7 samples SH-EP1-hα7-Ric-3 samples. Each condition was analyzed with five replicates. Data analysis was performed using ProteoIQ version 2.7 Protein inclusion criteria include 1% protein FDR, minimum peptide length of six amino acids 90% probability, identification in 2 or more of 5 replicates, and 0% probability in controls. FDRs were determined using the PROVALT algorithm and probabilities were determined with the ProteinProphet algorithm through ProteoIQ analysis. Only Top and Co-Top identifications were considered. Biological processes are listed as determined by DAVID analysis of Gene Ontology (GO) terms. Biological process GO terms for six proteins are not available. (XLSX) S3 Table. Proteins identified in α-bgtx-affinity enriched samples from SH-EP1-hα7-Ric-3 and SH-EP1-hα7. 625 identified proteins meeting the established inclusion criteria were identified in both SH-EP1-hα7-Ric-3 and SH-EP1-hα7 samples. These proteins represent both possible α7-nAChR interactions regardless of Ric-3 expression as well as non-specific binding to α-bgtx-affinity beads. Further investigation is required to distinguish which identifications are non-specific. Data analysis was performed using ProteoIQ version 2.7 Protein inclusion criteria include 1% protein FDR, minimum peptide length of six amino acids 90% probability, identification in 2 or more of 5 replicates, and 0% probability in controls. FDRs were determined using the PROVALT algorithm and probabilities were determined with the ProteinProphet algorithm through ProteoIQ analysis. Only Top and Co-Top identifications were considered. (XLSX) S4 Table. Cellular compartment GO terms associated with identified proteins. GO terms for cellular localization of identified SH-EP1-hα7-Ric-3 unique (A) and SH-EP1-hα7 unique (B) proteins. A total of 82% of the 39 (32 total) SH-EP1-hα7-Ric-3 unique, Ric-3-mediated proteins and 83% of the 97 (82 total) SH-EP1-hα7 unique proteins have GO terms that identifies the cellular compartment where the proteins have been reported to be localized. Proteins are identified by Uniprot accession numbers. The number of proteins associated with each compartment is listed as "Protein count" and the proportion of proteins classified into each compartment are listed as a percent of the total attributed proteins. (XLSX)