Address correspondence and reprint requests to Dr Isabel Bermudez, School of Biological and Molecular Sciences, Oxford Brookes University, Gipsy Lane Oxford OX3 0BP, UK. E-mail: firstname.lastname@example.org
α4 and β2 nicotinic acetylcholine (nACh) receptor subunits expressed heterologously in Xenopus oocytes assemble into a mixture of receptors with high and low agonist sensitivity whose relative abundance is influenced by the heteropentamer subunit ratio. We have found that inhibition of protein kinase A by KT5720 decreased maximal [3H]cytisine binding and acetylcholine (ACh)-induced current responses, and increased the relative proportion of α4β2 receptors with high agonist sensitivity. Mutation of serine 467, a putative protein kinase A substrate in a chaperone protein binding motif within the large cytoplasmic domain of the α4 subunit, to alanine or asparate decreased or increased, respectively, maximal [3H]cytisine binding and ACh response amplitude. Expression of α4S467A mutant subunits decreased steady levels of α4 and the relative proportion of α4β2 receptors with low agonist sensitivity, whilst expression of α4S467D increased steady levels of α4 and α4β2 receptors with low agonist sensitivity. Difopein, an inhibitor of chaperone 14-3-3 proteins, decreased [3H]cytisine binding and ACh responses and increased the proportion of α4β2 with high sensitivity to activation by ACh. Thus, post-translational modification affecting steady-state levels of α4 subunits provides a possible means for physiologically relevant, chaperone-mediated variation in the relative proportion of high and low agonist sensitivity α4β2 nACh receptors.
Neuronal nicotinic acetylcholine (nACh) receptors are ligand-gated ion channels expressed in the central and peripheral nervous systems that respond to the neurotransmitter, acetylcholine (ACh). In vertebrates, neuronal nACh receptor subtypes assemble from combinations of 9 α (α2-α10) and three β (β2-β4) subunits (Gotti and Clementi 2004). The α4β2 is the most abundant nACh receptor subtype in the brain where it forms the high-affinity binding site for nicotine (Picciotto et al. 2001). Like most brain nACh receptors, α4β2 nACh receptors appear to be localized primarily presynaptically where they facilitate the release of various neurotransmitters (Wonnacott 1997; Dani 2001). The α4β2 nACh receptor has been implicated in learning and memory, nociception and nicotine dependence (Picciotto et al. 2001; Tapper et al. 2004; Maskos et al. 2005). Mutations in both the α4 (CHRNA4) and β2 (CHRNB2) subunit genes cause autosomal nocturnal frontal lobe epilepsy (Weiland et al. 2000). Losses in numbers of α4β2 nACh receptor have been documented in age-related degenerative diseases such as Alzheimer's and Parkinson's diseases, suggesting possible pathological involvement (Zanardi et al. 2002).
There is little insight into endogenous mechanisms that may influence the expression of α4β2 nACh in neurones. However, recent studies have demonstrated that interactions between α4 nAChR subunits, protein kinase A (PKA), and the chaperone protein 14-3-3 increase steady state levels of unassembled and assembled α4 subunits (Jeancloss et al. 2001) by masking of a dibasic retention signal within the large cytoplasmic domain of α4 (O'Kelly et al. 2002). These findings suggest that neurons may be able to regulate the type of α4β2 nACh expressed by modulating the relative abundance of nACh receptor subunits in the ER, where ligand-gated ion channels assemble. To address this issue, we have investigated how mutations of the 14-3-3 binding motif in the α4 subunit affect expression and the relative abundance of high- and low-sensitivity α4β2 nACh receptors. We show that maximal [3H]cytisine binding and maximal amplitude of ACh-mediated current responses are increased by PKA–driven interactions between α4 subunit and 14-3-3 proteins. The interplay between PKA, α4 subunits and 14-3-3 proteins also changes the relative proportions of high and low agonist sensitivity α4β2 nACh receptors through a process that appears to be largely due to an alteration in the relative abundance of α4 subunits.
Materials and Methods
Xenopus laevis were purchased from Blades Biological (Kent, UK). [3H]cytisine (30.4 mCi/mmol) was obtained from Perkin Elmer Life Sciences (Boston, MA, USA). Ecoscint H scintillation cocktail was from National Diagnostics, UK. Antibodies AChRα4 A-20 and 14-3-3 β K-19, and blocking peptides AChRα4 A-20 p andr 14-3-3 K-19 p were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The PKA inhibitor KT5720 was from Tocris Bioscience, UK. All other chemicals were from Sigma Chemical Co. (St Louis, MO, USA).
Expression of α4β2 nACh receptors in Xenopus oocytes
Xenopus toads were kept and killed in accordance with the Animals (Scientific Procedures) Act, 1986, UK. Stage V and VI Xenopus oocytes were prepared as previously described (Houlihan et al. 2001). Human α4 and β2 nAChR cDNAs were provided by Professor Jon Lindstrom (University of Pennsylvania, PA, USA) and subcloned into the pcDNA 3.1(hygro-) expression vector (Invitrogen, Carlsbad, CA, USA) for nuclear injection in oocytes. In order to minimise differences in the translation of the two cDNAs that may be brought about by uncoding regions (UTRs) flanking the coding regions, subcloning was performed by PCR using primers annealing at the start and end of the cDNAs. Using these primers, Xba I and Not I restriction sites were introduced into the beginning and end, respectively, of both subunit cDNAs for subsequent cloning into the linker region of the expression vector. For transfections, the cDNAs were dissolved in distilled water at approximately equal concentrations of 1 µg/mL (spectrophotometric plus agarose gel electrophoresis determinations). Mixtures of these solutions at 1 : 1 α4:β2 subunit cDNA ratios were injected into the nuclei of oocytes in a volume of 18.4 nL/oocyte, using a variable volume automatic oocyte microinjector (Drummond Broomall, PA, USA). Transfections were carried out 24 h after oocyte removal from ovaries. The total amount of nACh receptor subunit expression vector injected per oocyte was kept constant (3 ng). After injection, the oocytes were incubated at 19°C in a modified Barth's solution containing 88 mm NaCl, 1 mm KCl, 2.4 mm NaHCO3, 0.3 mm Ca(NO3)2, 0.41 CaCl2, 0.82 mm MgSO4, 15 mm Hepes, and 50 mg/L neomycin (pH 7.6 with NaOH). Biochemical and electrophysiological experiments were performed between 2 and 5 days after injection. Where indicated, Barth's solution was supplemented with 100 nm KT5720 to decrease PKA activity.
Site directed mutagenesis
S441 (as labeled by Jeancloss et al. 2001) within the large cytoplasmic domain of the rat α4 subunit mediates the effects of PKA and 13-3-3 on expression of rat α4β2 nACh receptors (Jeancloss et al. 2001). S441 is equivalent to S467 within the large intracellular domain of the human α4 subunit. We did not find PKA phosphorylation sites on the human β2 subunit. Accordingly, we mutated S467 in α4 using the QuickChangeTM. site-directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA) and oligonucleotides containing the desired nucleotide substitution. The full-length sequences of mutant cDNAs were verified (Biochemistry Department, University of Oxford, Oxford).
Whole-cell currents from wild type or mutant human α4β2 nACh receptors were recorded using standard two-electrode voltage clamp techniques (GeneClamp 500, Axon Instruments, USA) with agarose-cushioned electrodes filled with 3 m KCl. Oocytes were continually supplied with fresh Ringer solution (in mM: 115 NaCl, 2.5 KCl, 1.8 CaCl2, 10 Hepes, pH 7.2) in a 60 µL bath, using a gravity-driven perfusion system at a rate of 10 mL/min. ACh was applied by gravity perfusion using a manually activated valve and impaled oocytes were held at a membrane potential of − 60 mV. Between each successive drug application, the cell was superfused with Ringer solution for 4 min to allow drug clearance and recovery from receptor desensitisation.
[3H]cytisine binding assays
[3H]cytisine binding assays were carried out on membrane homogenates prepared from oocytes transfected with equal amounts of α4 and β2 cDNAs. Oocytes (20–120, depending on expression levels or drug treatment) were homogenised using a glass homogeniser in 10 mL of a buffer containing 340 mm sucrose, 1 mm EGTA, 1 mm EDTA, and 0.1 mm of serine protease inhibitor phenylmethylsulfonylfluoride (PMSF). Homogenates were centrifuged at 4°C at 1000 g for 10 min. The supernatant was removed for further centrifugation, avoiding both the surface lipid layer and the pellet that contained pigment granules, both of which increased non-specific binding. The supernatant was centrifuged for 1 h at 100 000 g. The resulting pellet was resuspended in binding saline (10–30 µL per oocyte depending on levels of expression or drug treatment) of the following composition: 144 mm NaCl, 1.5 mm KCl, 2 mm CaCl2, 1 mm MgSO4, 20 mm HEPES, 1% BSA (w/v), pH 7.5 plus 1 mm EGTA, 1 mm EDTA, 0.1 mm PMSF, supplemented with 10 µg/mL each of apronitin, leupeptin and pepstatin A to protect the receptors. [3H]cytisine binding assays were performed by adding 5 nm[3H]cytisine to the resuspended membranes for 75 min at 4°C, and 10 µm nicotine was used to define non-specific binding. For equilibrium saturation binding assays, the concentration of [3H]cytisine was varied from 0.2 to 5 nm. Bound and free fractions were separated by rapid vacuum filtration through Whatman GF/B filters pre-soaked in binding saline with 0.1% polyethelyeneimine. Radioactivity was determined by liquid scintillation using Ecoscint H.
Western blot assays
Western blot assays were carried out on total oocyte membrane homogenates prepared from 3-day post-injection oocytes. Four batches of oocytes (50 oocytes per batch) were used. Three batches of oocytes were injected with α4 (wild-type or mutant) subunit alone. Oocytes were homogenised in 10 volumes of ice-cold homogenisation buffer [0.3 m sucrose and 20 mm Tris-HCl (pH 7.6) containing 5 mm EGTA, 5 mm EDTA, 1 µm pepstatin, 1 mg/mL leupeptin, 0.1 mm PMSF]. The homogenates were centrifuged at 1000 g for 10 min at 4°C, and the supernatant was re-centrifuged at 100 000 g for 30 min at 4°C. Pellets were then resuspended in 100 µL of 4% SDS loading buffer, and then 10 µL aliquots were separated using SDS-PAGE electrophoresis (Novex, 12% Tris-glycine gels; Invitrogen, UK). The proteins were subsequently transferred onto nitrocellulase membranes (Optitran BA-S83, Schleider & Schuell, Germany) by electroblotting (overnight transfer at 4°C with 20 mV). Membranes were blocked for 2 h at room temperature with 5% non-fat dry milk in phosphate buffered saline (PBS) and 0.1% Tween 20 (Sigma-Aldrich Co., UK). Membranes were then incubated overnight at 4°C with 0.5 µg/mL of primary antibody (AChRα4 A-20 or 14-3-3 β K-19) and then incubated for 2 h at room temperature with secondary antibody (anti-goat or anti-rabbit IgG alkaline phosphatase, as appropriate) as recommended by the supplier (Sigma-Aldrich Co.). Bound antibodies were visualised using BCIP/NBT solution (Sigma-Aldrich Co.). The identity of the bands was determined based on sensitivity to blockade of immunoreactivity in the presence of blocking peptides (AChRα4 A-20 P or 14-3-3 K-19 P (1 µg/mL) and molecular weight of the immunoreactive protein (α4, ≈70 kDa and 14-3-3, ≈ 28 kDa).
Role of 14-3-3 chaperone in regulating expression and proportion of high- and low-affinity α4β2 receptors
To assess roles of 14-3-3 protein in regulating cell surface expression of α4β2 receptors, a specific 14-3-3 inhibitor, difopein, was used. Difopein binds to all 14-3-3 isoforms with high affinity and specificity without having to be phosphorylated and effectively dissociates 14-3-3 from its binding partners (Master and Fu 2001). As difopein is not cell permeable, difopein fluorescent protein expression vector (pSCM138, provided by Dr H. Fu, Emory University School of Medicine, Atlanta, USA) was injected into oocytes with or without α4 and β2 subunit cDNAs. Prior to the binding experiments, 10 oocytes per batch of oocytes injected were checked for expression of difopein using fluorescent microscopy, and if at least 7 oocytes tested positive for difopein-fluorescence, the batch of oocytes was used for binding and electrophysiological studies. Experiments were carried out 48 h post-injection. Because 14-3-3 proteins regulate apoptosis, experiments were carried out only on oocytes that showed no signs of cell damage.
Concentration-response curves were obtained by normalizing the current responses to varying concentrations of ACh to the maximal response observed to a saturating concentration of ACh (2 mm) in each oocyte. Fits to full concentration-response curves for individual oocytes were made independently using Prism 4 (Graphpad Software, CA, USA) and then averaged in order to compare significant differences between groups. Concentration-response data for wild or mutant α4β2 receptors were fit using one and two component sigmoidal dose–response equations as previously reported (Houlihan et al. 2001), Y = I/ImaxBottom + (I/ImaxTop– I/ImaxBottom)/(1 + 10^((LogEC50– X)*nHill)) and Y = I/ImaxBottom + [(I/ImaxTop– I/ImaxBottom)*Fraction/(1 + 10^((LogEC50− 1–X)*nHill-1))] + [I/ImaxTop – I/ImaxBottom)*(1-Fraction)/(1 + 10^((LogEC50−2–X)*nHill_2))], respectively, where X is the logarithm of the agonist concentration and Y is the peak amplitude of the current response. An F-test determined whether the one-site or two-site model best fit the data; the simpler one-component model was preferred unless the extra sum-of-squares F-test had a value of p less than 0.05.
Binding and concentration-response curve data were pooled from at least two different batches of oocytes. Tests of significance were determined using the Student's t- test, and p-values less than 0.05 were considered significant.
Inhibition of PKA decreases maximal [3H]cytisine binding and the relative proportion of α4β2 nACh receptors with high sensitivity to activation by ACh
ACh produced robust inward currents in voltage-clamped oocytes injected with α4 and β2 subunit cDNAs in ratios of 1 : 1 (Fig. 1a). The concentration-response data for ACh were best fit to a two-component sigmoidal equation (p = 0.001, F-test, n = 10), indicating the presence of two types of α4β2 nACh receptors that differed in their sensitivity to activation by ACh. The estimated EC50 values [and 95% confidence intervals] and nHill values (± SEM) were 0.67 [0.51–0.85]µM and 1.5 ± 0.2 for the receptor type with higher sensitivity to ACh and 81 [73–90]µM and 0.92 ± 0.04 for the receptor type with lower sensitivity to ACh. The higher sensitivity component represented 20 ± 2% of the receptors contributing to the ACh concentration-response curve. These values are comparable to previously reported data concerning ACh effects on high and low agonist sensitivity α4β2 nACh receptors expressed in oocytes (Zwart and Vijverberg 1998; Covernton and Connolly 2000; Houlihan et al. 2001) or mammalian cells (Buisson and Bertrand 2001; Nelson et al. 2003; Kuryatov et al. 2005; Vallejo et al. 2005). Also, the ACh sensitivity of α4β2 receptor types characterized in this study was comparable, respectively, to that of high ACh sensitivity receptors having a stoichiometry of (α4)2(β2)3 and of lower sensitivity receptors having an (α4)3(β2)2 stoichiometry in stable transfected HEK cells (Nelson et al. 2003). Equilibrium binding of [3H]cytisine to membrane homogenates prepared from oocytes cotransfected with equal amounts of α4 and β2 cDNAs revealed only a single component (p = 0.001, F-test, n = 3), in accord with [3H]nicotine binding data obtained from mammalian cell lines expressing both types of α4β2 receptors (Kuryatov et al., 2005). Calculations yielded (± SEM) a KD value of 0.49 ± 0.77 nm and a Bmax value of 77 ± 7 fmol/oocyte.
Before starting our mutagenesis study of the 14-3-3 protein binding – PKA consensus sequence within the large cytoplasmic domain of the human α4 subunit, we first investigated how inhibition of PKA activity may affect maximal [3H]cytisine binding and the relative proportion of high and low ACh sensitivity α4β2 nACh receptors. For these studies, transfected oocytes were kept in Barth's saline supplemented with 100 nm of the PKA inhibitor KT5720 for 48 h and then used for [3H]cytisine binding assays or electrophysiological recordings. As shown in Fig. 2(a), exposure to KT5720 significantly decreased maximal binding of [3H]cytisine by 66% (p = 0.0041; Student's t-test; n = 3) and the maximal amplitude of the current responses elicited by ACh by 37% (p = 0.048, Student's t-test; n = 10). KT5720 also increased significantly the population of the α4β2 nACh receptors more sensitive to activation by ACh to 33 ± 2% of the total (Fig. 2b) (p = 0.0001, Student t-test, n = 6), but this increase did not eliminate the biphasic nature of the ACh concentration-response data, which was best described by a two-component sigmoidal model (p = 0.0034, F-test; n = 6). The estimated EC50 values were 0.66 [0.52–0.83]µM and 68 [59–78]µM, which were comparable to the respective EC50 values of α4β2 receptors expressed under control experimental conditions.
Mutation of S467 within the large cytoplasmic domain of the human α4 subunit mimics effects of PKA inhibition on nACh receptor numbers and functional properties
The preceding studies showed that inhibition of PKA decreases maximal binding of [3H]cytisine to α4β2 nACh receptors and increases the proportion of α4β2 receptors with higher sensitivity to activation to ACh. Substitution of S467 by A in the α4 subunit produced similar effects. S467 is a substrate for PKA phosphorylation (Pacheco et al. 2003; Wecker and Rogers 2003) and is located in a 14-3-3 protein binding motif within the large intracellular loop of the α4 subunit (RSLSV; PKA target underlined). Maximal [3H]cytisine binding and maximal amplitudes of ACh current responses were significantly reduced by 70% (p = 0.0042, Student's t-test; n = 3) and 40% (p = 0.023, student's t-test, n = 12), respectively, in mutant α4S467Aβ2 receptors (data shown in Table 1). The ACh concentration response curve data for α4S467Aβ2 were best fit to a two-component sigmoidal equation (p = 0.0027; F-test; n = 12) (Fig. 3a). High- and low-sensitivity EC50 values did not differ significantly between α4S467Aβ2 and wild α4β2 receptors (data summarized in Table 1), but the relative proportion of receptors with higher ACh sensitivity increased significantly to 36 ± 3 of the total (p < 0.0001, Student's t-test; n = 12). In contrast, when S467 was replaced by D to mimic a constitutively phosphorylated site, maximal [3H]cytisine binding and maximal ACh responses were increased significantly by 80% (p = 0.0071, Student's t-test, n = 3) and 60% (p = 0.0067, Student's t-test, n = 11), respectively (Table 1). Furthermore, the ACh concentration-response curve data for α4S467Dβ2 receptors could only be fit adequately to a single-component sigmoidal equation (Fig. 3b), and the estimated EC50 value was not significantly different from that of the α4β2 receptors with lower sensitivity to activation by ACh (Table 1). Thus, the inability to phosphorylate α4S467 increased (but did not shift completely) the proportion of α4β2 receptors with higher sensitivity to ACh, and the mimicry of α4S467 phosphorylation eliminated the small amount of expression of α4β2 receptors with higher agonist sensitivity.
Table 1. Influences of nAChR α4 subunit mutations, protein kinase A-targeted drug treatments and an inhibitor of 14-3-3 protein on maximal [3H]cytisine binding and sensitivity of α4β2 nACh receptors to activation by ACh
[3H]Cytisine binding Bmax ± SEM (fmol/oocyte)
EC50 (95% CI) µM
ACh Imax± SEM (µA)
ACh concentration-response curves were derived from whole-cell current responses in oocytes expressing the indicated receptors or treatments described in Materials and Methods and illustrated in Figs 1–4. Parameters provided were determined from fits to one component or two component (with high and low ACh sensitivity components indicated; HS, LS, respectively) sigmoidal equations. Values for the amplitude of responses elicited by 2 mM ACh represent the average ± SEM of the amplitude responses of 3–12 oocytes. Bmax values represent the average ± SEM of three binding assays carried out in triplicate. Asterisks indicate that values were significantly different (p < 0.05) from wild-type receptors.
77 ± 7
5.2 ± 0.4
α4β2 + KT5720
26 ± 5*
3.3 ± 0.8*
23 ± 6*
3.1 ± 0.7*
139 ± 10*
8.3 ± 0.9*
α4β2 + difopein
21 ± 9*
2.7 ± 0.3*
14-3-3 proteins are implicated in the effects of PKA on expression of high- and low-affinity α4β2 receptors
To assess whether the effects of PKA on human α4β2 receptors involved 14-3-3 proteins, a specific inhibitor of 14-3-3 proteins, difopein fluorescent protein, was coexpressed with α4 and β2 subunits. As shown in Fig. 4(a), immunoblotting with anti-14-3-3 antibody K-19 that cross-reacts with multiple isoforms of 14-3-3 confirmed previous findings that oocytes express 14-3-3 proteins (Duckworth et al. 2002). After two days of transfection, difopein significantly decreased maximal [3H]cytisine binding by 73% (p = 0.008, Student's t-test; n = 3) and maximal amplitude of ACh responses by 48% (p = 0.003; Student's t-test; n = 4) (Table 1). These effects were comparable to those of K5720 treatment or the S467A mutation. Inhibition of 14-3-3 proteins by difopein also increased significantly the component of the ACh concentration response curve with higher ACh sensitivity to 28 ± 1% of the total (p = 0.006, Student's t-test, n = 4) without altering the biphasic nature of the concentration-response curve (p = 0.003; F-test, n = 4) (Fig. 4b). These results suggest that 14-3-3 proteins are involved in the effects of PKA on radioligand binding and function of α4β2 nACh receptors and on the relative proportions of α4β2 nACh receptors with high and low sensitivity to ACh.
Previous immunoblotting studies that examined heterologous expression of rat α4 subunit levels in transfected tsA 201 cells have shown that 14-3-3 protein activity increases steady-state levels of rat α4 subunit (Jeancloss et al. 2001). The same approach was used here to examine whether changes in the steady-levels of α4 subunit may underlie the effects of PKA and 14-3-3 protein on the relative proportion of high and low agonist sensitivity α4β2 nACh receptors. Immunoblotting with anti-AChR α4 subunit antibody A-20 raised against a recombinant protein corresponding to amino acids 500–600 of the α4 subunit, showed that steady-state levels of α4S467 A were reduced markedly in comparison to wild-type or α4S467D subunits, whilst steady-state levels α4S467D subunits were increased relative to the steady-levels of α4 or α4S467 A subunits, even when equal amounts of the respective cDNAs were injected and equal amounts of samples prepared in an identical manner were immunoblotted (Fig. 5).
We have shown that the relative proportion of high and low agonist sensitivity α4β2 nACh receptors can be modulated by 14-3-3 protein in combination with PKA-dependent phosphorylation of the α4 subunit. The proportion of α4β2 receptors with higher agonist sensitivity is increased in the presence of the PKA inhibitor KT5720 or when wild-type β2 subunits are coexpressed with α4S467 A subunits incapable of being phosphorylated at that site by PKA. Treatment with the 14-3-3 protein inhibitor difopein increases the relative proportion of high agonist sensitivity α4β2 nACh receptors to the same levels as KT5720 treatment or α4S467A expression. Collectively, these findings point to convergence on interactions of PKA-phosphorylated α4 subunits with the 14-3-3 protein chaperone in elevation of the relative proportion of α4β2 nACh receptors with lower sensitivity to activation by ACh. While PKA (Gopalakrishnan et al. 1996) and 14-3-3 protein (Jeancloss et al. 2001) have been shown previously to increase the number of α4β2 nACh receptors, which this study confirmed, to our knowledge this the first time that it has been shown that 14-3-3 increases the relative proportion of human α4β2 nACh receptors with low ACh sensitivity by a mechanism that requires PKA phosphorylation of S467 within the large cytoplasmic domain of 4 subunit.
There are limits on the magnitude of effects on nACh receptor numbers controlled via α4 subunit phosphorylation, PKA activity, and protein 14–3−3 interactions. There is a finite and significant level of radioligand binding and function of α4β2 receptors in the presence of KT5720 or difopein or of α4S467 A β2 receptors. This shows that there is a basal level of PKA-14-3-3-independent expression of α4β2 nACh receptors and that intracellular processes such as PKA activation (Jeancloss et al. 2001; this study) increase receptor number above this level. Basal expression of human α4β2 nACh receptors occurs despite the presence of a dibasic ER retention signal (R345-R346) within the large cytoplasmic domain of α4 that constitutes a binding site for the ER-retaining chaperone β-COP. In certain multimeric proteins such as KCNK (O'Kelly et al. 2002) and KATP potassium channels (Yuan et al. 2003) β-COP works in concert with 14-3-3 chaperones to allow ER exit of only properly assembled proteins (O'Kelly et al. 2002; Yuan et al. 2003). Thus, binding of 14-3-3 protein to KCNK (O'Kelly et al. 2002) and KATP potassium channels (Yuan et al. 2003) interferes with the binding of β-COP to the ER dibasic retention motifs, and this is crucial for overriding ER retention and for the forward transport of the multimeric protein complex. Perhaps the dibasic retention signal in α4 is hindered during assembly of α4 and β2 subunits as occurs during assembly of muscle nACh receptor subunits (Keller et al. 2001). Thus, PKA–driven interactions of 14-3-3 protein with α4 subunits seem to be a molecular device to modulate expression of α4β2 nACh receptor with lower agonist sensitivity.
S467 in human α4 is equivalent to S441 in rat α4, and both are within a PKA consensus sequence, RSLSV (PKA target underlined) that becomes a binding motif for 14-3-3 protein upon phosphorylation (Ma and January 2002; Dougherty and Morrison 2004). In the case of rat α4 subunit, PKA-driven binding of 14-3-3 to RSLSV increases steady-levels of α4 subunit and α4β2 nACh receptors. Although we did not investigate directly binding of 14-3-3 proteins to human α4 subunits, collectively our findings make a strong case for a functional relationship between PKA, human α4 subunit and 14-3-3 that boosts steady-levels of α4 subunit and α4β2 receptors with lower ACh sensitivity. It remains unclear whether and how generally this mechanism applies to regulation of the expression of other nACh receptors. Other nACh receptor subunits do possess serine residues that are putative PKA phosphorylation sites, but none of them possess the 14-3-3 binding motif, RSLSV, found in the α4 subunit.
Under conditions of subunit balance, α4β2 nACh receptors are predominantly in a low agonist sensitivity form, which is locked in by the α4S467D mutation and reduced in abundance but not eliminated in the α4S467 A mutant. Such subtle changes in receptor form may have escaped detection in previous studies investigating 14–3−3 interactions with rat nAChR α4 subunits (Jeancloss et al. 2001), perhaps due to species-specific differences in other factors that control levels and forms of α4β2 nACh receptor expression. How do 14-3-3 protein and PKA increase the relative proportion of α4β2 nACh receptors with lower sensitivity to activation by ACh? Our Western blotting studies showed that 14-3-3 in a PKA-driven process increases steady-levels of α4 subunits. Previous studies of α4β2 nACh receptors have shown that increasing the amount of transfected β2 cDNA in HEK cells increases expression of the stoichiometry (α4)2(β2)3, which has higher sensitivity to activation by ACh (Nelson et al. 2003). By analogy, even though we did not determine the stoichiometry of the high and low ACh sensitivity α4β2 receptors, it is tempting to speculate that the PKA-14-3-3-dependent increase in steady-levels of α4 subunit promotes assembly of the (α4)3(β2)2 stoichiometry, which has lower sensitivity to ACh. However, we cannot preclude the possibility that PKA-14–3−3 interactions with the α4 subunit alter the functional state of α4β2 receptors, as has been proposed for the effects of chronic exposure to nicotine on radioligand binding and ACh sensitivity of α4β2 receptors (Vallejo et al. 2005). However, because we did observe that increases in maximal [3H]cytisine binding and amplitude of ACh current responses were accompanied by increased steady-state levels of α4 subunit, it seems more plausible that 14-3-3 protein in combination with PKA promotes greater incorporation of α4 subunits into receptor complexes. Additional studies, including the determination of the stoichiometry of the high and low agonist sensitivity α4β2 receptors expressed in Xenopus oocytes, will be necessary to discriminate between these two possibilities.
In summary, supra-basal levels of expression of α4β2 nACh receptors under balanced conditions of subunit availability are sensitive to α4 subunit phosphorylation, PKA activity, and interactions with 14-3-3 chaperone proteins, with the chaperoning process sitting atop the hierarchy, but not as actively in the absence of phosphorylation of α4 subunits. nACh receptor α4 subunit phosphorylation state (as implied through point mutagenesis studies) influences the equilibrium between high and low ACh sensitivity α4β2 nACh receptors. There are subtle but perhaps physiologically significant effects of α4S467 phosphorylation (implied by mutation studies) on high : low ACh sensitivity α4β2 nACh receptor functional forms. Thus, regulation of the proportion of high and low agonist sensitivity forms of α4β2 nACh receptor can be influenced by chaperone interactions, which in turn could be affected by factors such as nicotine exposure. The behavioral relevance of this plasticity and its expression in neurons remain to be elucidated, but it is noteworthy that a mouse Chrna4 A529T polymorphism that alters the ratio of high to low agonist sensitivity α4β2 nACh receptors (Kim et al. 2003) is associated with mouse strain differences in sensitivity to acute effects of nicotine (Stitzel et al. 2001; Tritto et al. 2002).
We thank John Connolly and Patrick Covernton for helpful discussion on concentration response data analysis. This work was supported by QR Oxford Brookes University Research funding. RE was funded by a BBSRC-Eli Lilly Case PhD studentship. Funding from National Institutes of Health grants NS40417 and DA015389, from Arizona Disease Control Research Commission grants 9730 and 9615, and by endowment and/or capitalization funds from the Men's and Women's Boards of the Barrow Neurological Foundation also is acknowledged (RJL).