αConotoxin Arenatus IB[V11L,V16D] Is a Potent and Selective Antagonist at Rat and Human Native α7 Nicotinic Acetylcholine Receptors

The α7 subtype of nicotinic acetylcholine receptors (nAChRs) is currently attracting considerable interest for its roles in brain function, including synaptic plasticity (McKay et al., 2007) and attentional aspects of cognitive performance (Levin et al., 2006). Consequently, α7 nAChRs are therapeutic targets for cognitive impairments in Alzheimer's disease, schizophrenia, and attention deficit hyperactivity disorder (Mazurov et al., 2006; Chiamulera and Fumagalli, 2007; Wilens and Decker, 2007). α7 nAChRs also have therapeutic potential in several other processes, including pain, inflammation, and wound healing (Decker and Meyer, 1999; de Jonge and Ulloa, 2007). Selective antagonists are essential research tools for elucidating and clarifying the molecular and cellular mechanisms that under-pin these effects.

Research into α7 nAChRs has profited from the availability of two useful antagonists: αbungarotoxin (αBgt) and methyllycaconitine (MLA). αBgt, from the venom of the Elapidae snake Bungarus multicinctus, is a large polypeptide with slow binding kinetics. This confers high-affinity binding with very slow dissociation, an advantage for some studies. It can also be a limitation in functional assays because increased incubation times and/or concentrations are required to achieve blockade. Moreover, the slowly reversible block limits recovery in experiments where repeated agonist application is desirable, and the large size of the toxin can compromise its access to α7 nAChRs in intact preparations.

The norditerpenoid MLA from Delphinium and Consolida sp. is also a competitive antagonist of α7 nAChRs. It binds to α7 nAChRs with nanomolar affinity (K_i ∼ 1 nM; Davies et al., 1999). In addition, it exhibits moderately high affinity for α6β2-containing nAChRs (K_i, 33 nM; Mogg et al., 2002), which compromises its utility for identifying α7 nAChRs in areas where the α6 subunit is expressed, notably catecholaminergic and visual pathways. Furthermore, both αBgt and MLA interact with high affinity with nAChRs containing α9 and α10 subunits (Elgoyhen et al., 2001; Baker et al., 2004). The potential colocalization of α7, α9, and α10 subunits in sympathetic neurones (Lips et al., 2006) and non-neuronal cells (Bschleipfer et al., 2007) demands more discriminating antagonists.

The α-conotoxins, small peptides from the venom of Conus sp., can exhibit exquisite specificity for nAChR subtypes (McIntosh et al., 1999; Nicke et al., 2004). In a recent study, a naturally occurring αconotoxin from C. arenatus was modified by changing two amino acid residues to generate a novel toxin, α-CtxArIB[V11L,V16D] (Fig. 1A) (Whiteaker et al., 2007). This toxin displayed high potency and specificity for discriminating α7 nAChRs (K_i, 7 nM) from all other nAChR subtypes tested in binding assays. Functionally, α-CtxArIB-[V11L,V16D] specifically blocked currents evoked by stimulation of rat homomeric α7 nAChRs expressed in Xenopus oocytes, whereas it was without effect on heteromeric nAChRs (including α9α10 nAChR) in the same expression system (Whiteaker et al., 2007). However, the subunit composition of native nAChRs can be more diverse and more complex than that of recombinant receptors. Here, we have assessed the ability of α-CtxArIB[V11L,V16D] to antagonize responses to native nAChRs.

Clinical studies demand investigation of human tissues. The exquisite specificity of α-conotoxins for particular nAChR subtypes (Nicke et al., 2004), coupled with subtle species differences that can affect pharmacological selectivity (Young et al., 2007), requires caution when extrapolating from rodent to human tissues. Therefore, we have exploited both PC12 and SH-SY5Y cell lines that are representative of rat and human peripheral autonomic neurones, respectively. Both cell lines express α3, α5, α7, β2, and β4 nAChR subunits (Lukas et al., 1993; Henderson et al., 1994; Peng et al., 1994; Wang et al., 1996; Blumenthal et al., 1997) and have functional α7 and non-α7 nAChRs that can be discriminated using subtype-selective ligands (Dickinson et al., 2007). Additional nAChR subtypes are expressed in the brain, and their heterogeneity has been particularly well defined with respect to nAChR-evoked [³H]dopamine release from rodent striatal preparations. Several β2-containing heteromeric nAChRs (α4β2, α4α5β2, α4α6β2β3, α6β2β3, α6β2) reside on dopamine terminals (Grady et al., 2007). In addition, α7 nAChRs assigned to neighboring glutamate afferents can also modulate striatal [³H]dopamine release (Kaiser and Wonnacott, 2000; Barik and Wonnacott, 2006). We show that in rat striatal minces and in the rat and human cell lines, α-CtxArIB[V11L,V16D] selectively inhibited responses attributable to α7 nAChRs, whereas it had no effect on non-α7 nAChR responses. These results establish the utility of this new antagonist for interrogating the actions of rat and human α7 nAChRs in vitro.

Materials and Methods

Materials. [7,8-³H]Dopamine (43 Ci/mmol) and ¹²⁵I-αbungarotoxin (263 Ci/mmol) were purchased from GE Healthcare (Chalfont St. Giles, UK). The PC12 cell line was a gift from D. Berg (University of California, San Diego, CA). SH-SY5Y cells were obtained from European Collection of Cell Cultures (Porton Down, Wiltshire, UK). 5-I-A85380 and PNU-120596 were purchased from Tocris Bioscience (Bristol, UK). αBgt, fluo-3 AM, and Pluronic F127 were purchased from Invitrogen (Carlsbad, CA). (-)-Nicotine hydrogen tartrate, pargyline hydrochloride, choline tartrate, ascorbic acid, nomifensine maleate, mecamylamine, RPMI 1640 medium, horse serum, fetal bovine serum, l-glutamine, penicillin, and streptomycin were obtained from Sigma Chemical (Poole, Dorset, UK). AR-R17779 was provided by GlaxoSmithKline (Uxbridge, Middlesex, UK). Optiplates and Optiphase Supermix were obtained from PerkinElmer Life and Analytical Sciences (Waltham, MA). Other standard chemicals of analytical grade were attained from Sigma Chemical. Male Sprague-Dawley rats (approximately 250 g) were from Charles River (Margate, Kent, UK).

Nicotine tartrate, choline, and KCl solutions were prepared freshly for each experiment by dissolving in assay buffer and correcting the pH to 7.4 if necessary. Other drugs were diluted from stocks made up in water, with the exception of PNU-120596 (100 mM stock in dimethylsulfoxide).

Radioligand Binding. Competition binding experiments for ¹²⁵I-αBgt and [³H]epibatidine binding sites were carried out on rat brain P2 membranes as described previously (Davies et al., 1999; Whiteaker et al., 2000). For ¹²⁵I-αBgt, membranes (250 μg) were incubated in a total volume of 1 ml in 10 mM phosphate buffer, pH 7.4, containing 0.1% bovine serum albumin (BSA) and 1 nM ¹²⁵I-αBgt in the presence and absence of increasing concentrations of α-CtxArIB[V11L,V16D] or (-)-nicotine for 4 h at 37°C, followed by chilling to 4°C. Samples were filtered through a double thickness of Gelman A/E filters, presoaked overnight in 0.3% polyethylenimine (bottom filter) or 4% milk powder (top filter), using a Brandel Cell Harvester (Brandel Inc., Gaithersburg, MD). For [³H]epibatidine, membranes (250 μg) were incubated in a total volume of 2 ml in 10 mM phosphate buffer, containing 0.2 nM [³H]epibatidine in the presence and absence of increasing concentrations of α-CtxArIB-[V11L,V16D] or (-)-nicotine for 2 h at room temperature. Samples were filtered through a double thickness of Gelman A/E filters, presoaked overnight in 0.3% polyethylenimine, using a Brandel Cell Harvester. Nonspecific binding was determined in the presence of 1 mM nicotine. Filters were subsequently washed five times using cold phosphate-buffered saline and counted for radioactivity in a Packard 1600 Tricarb scintillation counter (Packard Instrument Company, Meriden, CT).

Cell Culture. PC12 cells were cultured as described by Dickinson et al. (2007). Cells were grown in a humidified incubator under 7% CO₂ at 37°C in RPMI medium supplemented with 5% fetal bovine serum, 10% heat-inactivated donor horse serum, 2 mM l-glutamine, and 50 U/ml penicillin/streptomycin. Cells were plated at a density of 2.5 × 10⁴ cells/cm² onto 25-mm glass coverslips in six-well plates or into 96-well plates, both coated with poly-d-lysine (5 μg/ml). Assays were conducted 24 h later with confluent cultures.

Human neuroblastoma SH-SY5Y cells were cultured as described previously (Ridley et al., 2001). Cultures were maintained in 175-cm² tissue culture flasks in Dulbecco's modified Eagle's medium/F12, supplemented with 15% fetal bovine serum, 2 mM l-glutamine, 1% nonessential amino acids, 190 U/ml penicillin, and 0.2 mg/ml streptomycin, until confluent. Cells were plated (1:5 dilution) into 96-well plates, and experiments were performed 72 h later with confluent cultures.

Calcium Fluorometry (1) Cell Populations. Increases in intracellular Ca²⁺ in PC12 and SH-SY5Y cells were monitored by measuring changes in fluorescence in cells loaded with fluo-3, as described previously (Dickinson et al., 2007). In brief, cells cultured in 96-well plates were washed twice with Tyrode's salt solution (TSS; 137.0 mM NaCl, 2.7 mM KCl, 1.0 mM MgCl₂, 2.5 mM CaCl₂, 0.2 mM NaH₂PO₄, 12.0 mM NaHCO₃, 5.5 mM glucose, pH 7.4, supplemented with 0.1% BSA to prevent α-CtxArIB[V11L,V16D] binding to plastic or glass) and incubated with fluo-3 AM (10 μM) and 0.02% Pluronic F127 for 1 h at room temperature (22°C) in the dark. After washing the cells twice with TSS, cells were preincubated with 80 μl of TSS, with or without antagonist (10 min for αCtxArIB[V11L,V16D], mecamylamine, or DHβE; 20 min for αBgt). Where used, PNU-120596 was added 1 min before the end of preincubation. Basal fluorescence (excitation, 485 nm; emission, 538 nm) was recorded for 5 s in a Fluoroskan Ascent fluorescent plate reader (Thermo Fisher Scientific, Waltham, MA). Then, nAChR agonist, KCl, or vehicle (20 μl) was added, and the change in fluorescence was recorded for an additional 20 s. Under all conditions examined, increases in fluorescence reached a maximal level within 5 s, and this response was sustained over the remainder of the time course of the experiment. To normalize fluo-3 signals, the maximal and minimal fluorescences from each well were determined by the addition of 0.3% Triton X-100 (F_max) followed by 100 mM MnCl₂ (F_min). Drug-evoked responses were expressed as a percentage of the corresponding (F_max-F_min) value. None of the drugs or vehicles used significantly altered the basal level of fluorescence.

Calcium Fluorometry (2) Time Course Experiments. Changes in intracellular Ca²⁺ in individual PC12 cells loaded with fura-2 were monitored using dynamic video imaging (Concord System; PerkinElmer Life and Analytical Sciences). PC12 cells cultured on coverslips were washed twice with Ca²⁺ buffer (in 140.0 mM NaCl, 5.0 mM KCl, 1.0 mM MgCl₂, 1.8 mM CaCl₂, 10.0 mM glucose, 5.0 mM HEPES, pH 7.4; supplemented with 0.1% BSA) and incubated with fura-2 AM (5 μM) and 0.02% Pluronic F127 for 1 h at room temperature (22°C) in the dark. Coverslips were placed in a temperature-controlled microscope chamber maintained at 37°C (Harvard Apparatus Inc., Holliston, MA) and continuously perfused with buffer (5 ml/min). Agonist with or without α-CtxArIB[V11L,V16D] was applied via the perfusion system for 1 min, followed by washout before reapplication of agonist alone. Fura-2 was excited at 340 and 380 nm using a SpectroMaster I (PerkinElmer Life and Analytical Sciences), and emissions at 510 nm were detected with an intensified Ultrapix PDCI low-light level charge-coupled device camera (PerkinElmer Life and Analytical Sciences). Data were analyzed using Ultraview software (PerkinElmer Life and Analytical Sciences) and expressed as a ratio of F₃₄₀/F₃₈₀. Each cell that responded to agonist was analyzed separately, and these data were pooled for each experiment.

cRNA Preparation and Injection. Human α7 nAChR subunit cDNA was subcloned into the RNA expression vector pBluescript II SK^- (Stratagene, La Jolla, CA), and the resulting construct (from J. Garrett, Cognetix Inc., Salt Lake City, UT) was used to make cRNA. Capped cRNA was made using the mMessage mMachine in vitro transcription kit (Ambion, Austin, TX) following linearization of the plasmid with NotI and transcription under the control of a T7 promotor. The cRNA was purified using the QIAGEN RNeasy kit (QIAGEN, Valencia, CA). The concentration of cRNA was determined by the absorbance at 260 nm. Fifty nanoliters (608 ng/μl) of cRNA was injected into each Xenopus oocyte with a Drummond microdispenser (Drummond Scientific, Broomall, PA), as described previously (Cartier et al., 1996), and incubated at 17°C. Oocytes were injected within 1 day of harvesting, and recordings were made 3 to 6 days postinjection.

Voltage-Clamp Recording. Oocytes were voltage-clamped and exposed to ACh and peptide as described previously (Cartier et al., 1996). In brief, the oocyte chamber consisting of a cylindrical well (∼30 μl in volume) was gravity perfused at a rate of ∼2 ml/min with ND-96 buffer (96.0 mM NaCl, 2.0 mM KCl, 1.8 mM CaCl₂, 1.0 mM MgCl₂, 5 mM HEPES, pH 7.5) containing 0.1 mg/ml BSA. Oocytes were subjected once a minute to a 1-s pulse of 200 μM ACh. Once a stable baseline was achieved, either ND-96 alone or ND-96 containing varying concentrations of α-CtxArIB[V11L;V16D] was perfusion-applied, during which 1-s pulses of 200 μM ACh were applied every minute until a constant level of block was achieved. For experiments involving PNU-120596, the ACh concentration was reduced to 20 μM, and PNU-120596 was added to the ND-96 at a final concentration of 10 μM.

Data Analysis. For the baseline response, at least three ACh-induced responses were averaged. To determine the percentage block induced by toxin, after a steady-state block had been achieved, ACh-induced responses were averaged, and the value was divided by the pretoxin baseline value to yield a percentage response. The concentration-response data were fit to the equation, y = 100/(1 + 10((LogEC₅₀ - Log[Toxin]) × n_H)), where n_H is the Hill coefficient, by nonlinear regression analysis using GraphPad Prism (GraphPad Software Inc., San Diego, CA). Each data point represents the mean ± S.E.M. from at least four oocytes.

[³H]Dopamine Release. A 96-well filter plate assay was used to monitor [³H]dopamine release from striatal minces (Puttfarcken et al., 2000; Barik and Wonnacott, 2006). For each experiment, two rats were killed by cervical dislocation, brains were rapidly removed, and striata were dissected and transferred to ice-cold Krebs buffer (KB; 118.0 mM NaCl, 2.4 mM KCl, 2.4 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄.7H₂O, 25 mM NaHCO₃, 10 mM d-glucose, 1 mM ascorbic acid, 0.01 mM pargyline, gassed with 95% O₂ and 5% CO₂ for at least 1 h at 37°C, pH 7.4). Tissue was chopped using a McIlwain tissue chopper to give 150-μm prisms. After two washes with warm KB, striatal minces were incubated for 30 min with 50 nM [³H]dopamine in 5 ml of KB at 37°C. After five washes in KB containing 0.5 μM nomifensine, minces (100 μl/well) were loaded into a 96-well filter plate (model MABVN1250; Millipore Corporation, Billerica, MA) and preincubated for 10 min with buffer (70 μl) in the presence or absence of antagonist or modulator. Buffer was removed by filtration and replaced with fresh buffer (70 μl) containing agonist and/or antagonist. After an additional 5 min at 37°C, buffer containing released [³H]dopamine was collected by filtration into a 96-well Packard Optiplate (PerkinElmer Life and Analytical Sciences). Filters were removed to a separate Optiplate containing 70 μl water/well. Radioactivity was determined using a Microbeta liquid scintillation counter (Wallac 1450 Microbeta Trilux; PerkinElmer Life and Analytical Sciences), counting efficiency 30%. The amount of [³H]-dopamine released was expressed as a percentage of total radioactivity present in the slices before stimulation (i.e., tritium remaining in the tissue + amount of tritium released). Each condition was examined in six replicates and repeated in at least three independent experiments.

Statistics. Statistical significance (p < 0.05) was determined using the Student's t test, carried out using SigmaPlot version 8.0 (Systat Software, Inc., San Jose, CA).

Results

α-CtxArIB[V11L,V16D] (Fig. 1A) potently displaced ¹²⁵I-α-bungarotoxin binding to rat P2 membranes, with a K_i value of 4.0 ± 0.6 nM (Fig. 1B), confirming that α-CtxArIB-[V11L,V16D] interacts competitively at α7 nAChRs, as described previously (Whiteaker et al., 2007). α-CtxArIB-[V11L,V16D] did not compete for [³H]epibatidine (0.2 nM) binding to rat brain membranes (Fig. 1C). At this concentration, [³H]epibatidine will not label α7 nAChRs (K_i ∼ 0.2 μM) (Davies et al., 1999; see Marks et al., 2006). Therefore, this result indicates that over the concentration range of 0.1 nM to 1 μM α-CtxArIB[V11L,V16D] does not interact with non-α7 nAChRs. α-CtxArIB[V11L,V16D] was then tested for functional potency at native nAChRs.

Effects of α-CtxArIB[V11L,V16D] on nAChR-Mediated Ca²⁺Responses in PC12 Cells. PC12 cells express rat α7 and non-α7 nAChRs (Blumenthal et al., 1997), and we have recently characterized the differential coupling of these nAChR subtypes to increases in intracellular Ca²⁺ (Dickinson et al., 2007). Here, we first used choline to increase Ca²⁺ by selectively activating α7 nAChRs. Choline (1, 3 mM) elicited very small increases in fluorescence in PC12 cells loaded with fluo-3 AM, but these responses were substantially amplified in the presence of 10 μM PNU-120596, an α7-selective positive allosteric modulator (Fig. 2, A and B) (Hurst et al., 2005). PNU-120596 alone produced no response (data not shown). αBgt (100 nM; 20-min preincubation) reduced responses to 1 and 3 mM choline plus PNU-120596 by 90.9 ± 0.9 and 80.6 ± 1.8%, respectively (Fig. 2B).

In initial experiments, cells were preincubated with 100 nM α-CtxArIB[V11L,V16D] for 10 min before stimulation with choline (1 and 3 mM) and 10 μM PNU-120596. This concentration of α-CtxArIB[V11L,V16D] significantly inhibited responses, by 62.0 ± 5.4 and 49.9 ± 6.2%, respectively (Fig. 2B). α-CtxArIB[V11L,V16D] was then tested over the concentration range of 10 to 300 nM against 1 mM choline plus PNU-120596. α-CtxArIB[V11L,V16D] blocked α7 nAChR-mediated responses in a concentration-dependent manner with an IC₅₀ of 88.0 ± 17.4 nM (Fig. 2C). Maximal inhibition of 83.6% was observed with 300 nM α-CtxArIB-[V11L,V16D]. Higher concentrations were not tested to conserve the toxin.

To assess the specificity of α-CtxArIB[V11L,V16D] for α7 nAChR, its effects on responses induced by a nonselective nAChR agonist (nicotine), an agonist selective for β2-containing nAChRs (5-I-A85380) and a general depolarizing agent (KCl) were examined (Dickinson et al., 2007). Nicotine (100 μM), 5-I-A85380 (30 μM), and KCl (60 mM) elicited increases in fluorescence of 26.7 ± 2.1, 14.8 ± 1.7, and 44.2 ± 3.7%, respectively (mean ± S.E.M., n = 4; Fig. 2D). α-CtxArIB-[V11L,V16D], tested at the highest concentration used previously (300 nM; Fig. 2C), failed to reduce the responses to any of these stimuli. To determine whether 100 μM nicotine activated α7 nAChRs in addition to non-α7 nAChRs, it was applied in the presence of PNU-120596 (10 μM). Under this condition, responses were modestly enhanced by 53.6% (from 27-41% of the maximal response, in the absence and presence of PNU-120596, respectively). This is consistent with the response to 100 μM nicotine, including a small α7 nAChR-mediated component. In the presence of α-CtxArIB[V11L,V16D], the PNU-120596-enhanced response was not observed (Fig. 2D).

The time course of recovery from blockade by 300 nM α-CtxArIB[V11L,V16D] was assessed in PC12 cells loaded with fura-2 AM and monitored using dynamic video imaging. The response to 3 mM choline plus 10 μM PNU-120596 (60 s) was blocked by 300 nM α-CtxArIB[V11L,V16D] by 87.1 ± 1.6% [mean ± S.E.M. from 34 agonist-responsive cells (40.5% of total number of cells monitored) in three separate experiments]. After removal of the toxin from the perfusing buffer, responses to choline plus PNU-120596 slowly recovered, with complete recovery after 15 min of washout (Fig. 2E).

Blockade of Human α7 nAChR-Mediated Responses by α-CtxArIB[V11L,V16D]. To determine whether α-CtxArIB-[V11L,V16D] is similarly effective at human α7 nAChRs, the toxin was examined for blockade of human α7 nAChRs heterologously expressed in Xenopus laevis oocytes. Responses to 200 μM ACh by oocytes transfected with human α7 nAChR were blocked by α-CtxArIB[V11L,V16D], with an IC₅₀ value of 3.4 ± 0.037 (2.9-4.11) nM and a Hill slope of 1.20 ± 0.12 (1.5-1.0) (Fig. 3A). A maximal block of 100% was obtained with 100 nM toxin. Repeating this experiment with 20 μM ACh plus 10 μM PNU-120596 resulted in a slight rightward shift of the concentration-response curve for inhibition by α-CtxArIB[V11L,V16D]; the IC₅₀ value was increased 2-fold (6.7 nM; Fig. 3A). This indicates that amplification of α7 nAChR responses by PNU-120596 does not significantly perturb the sensitivity to toxin. Kinetic analysis shows that α-CtxArIB[V11L,V16D] blocks α7 nAChR currents evoked by 200 μM ACh within 60 s of exposure, with responses beginning to recover 180 s after removal of the toxin (Fig. 3B).

SH-SY5Y cells resemble human fetal sympathetic neurones grown in primary cultures and express a similar range of nAChR subunits to the PC12 cell line (Lukas et al., 1993; Peng et al., 1994). Changes in intracellular Ca²⁺ in SH-SY5Y cells loaded with fluo-3 were determined to assess the effects of α-CtxArIB[V11L,V16D] on native human α7 nAChRs. In agreement with the data from PC12 cells (Fig. 2), 3 mM choline alone was without effect, but in the presence of 10 μM PNU-120596, it evoked an increase in intracellular Ca²⁺ in SH-SY5Y cells that was inhibited by αBgt by 93.4 ± 1.0%, whereas DHβE had no significant effect (Fig. 4A). α-CtxArIB[V11L,V16D] produced a concentration-dependent inhibition of responses to choline plus PNU-120596, with an IC₅₀ value of 74.0 ± 1.8 nM. Another α7 nAChR-selective agonist, AR-R17779 (Jensen et al., 2005; Fig. 4B), was examined for comparison with choline. This agonist was also ineffective in the absence of PNU-120596; responses in the presence of the allosteric modulator were blocked by αBgt by 95.4 ± 1.8%, but not by DHβE. The concentration dependence of inhibition by α-CtxArIB[V11L,V16D] resulted in an IC₅₀ value of 51 ± 1.0 nM; maximal inhibition by 1 μM α-CtxArIB[V11L,V16D] was 95.9%.

Effects of α-CtxArIB[V11L,V16D] on Nicotine- and Choline-Evoked [³H]Dopamine Release from Rat Striatal Minces. The ability α-CtxArIB[V11L,V16D] to inhibit nAChR-evoked release of radiolabeled dopamine in vitro was investigated using a high-throughput 96-well filter plate assay. BSA (0.01%) was included in all conditions to prevent nonspecific adsorption of α-CtxArIB[V11L,V16D]; BSA had no effect on basal release (Fig. 5). Nicotine (10 μM) significantly increased [³H]dopamine release (10.5 ± 2.1%) compared with buffer alone, and this increase was abolished in the presence of 10 μM mecamylamine. However, α-CtxArIB[V11L,V16D] (300 nM) had no effect on either basal (data not shown) or nicotine-evoked [³H]dopamine release (Fig. 5A).

Choline was used to specifically target α7 nAChRs capable of modulating [³H]dopamine release (Fig. 5B). Choline (3 mM) increased fractional release by 6.1 ± 0.9%, and this was further increased in the presence of 10 μM PNU-120596 to 11.9 ± 1.2%. α-CtxArIB[V11L,V16D] (300 nM) inhibited responses to choline and choline plus PNU-120596 by 81.4 ± 1.2 and 77.9 ± 1.0%, respectively (Fig. 5B).

Discussion

The α-conotoxin ArIB analog α-CtxArIB[V11L,V16D] potently displaced [¹²⁵I]α-bungarotoxin binding to rat brain membranes and inhibited functional rat and human α7 nAChR responses in cell lines, transfected oocytes, and rat striatal minces. In contrast, α-CtxArIB[V11L,V16D] did not inhibit [³H]epibatidine binding or functional responses evoked by non-α7 nAChR agonists or KCl. These results confirm the specificity of α-CtxArIB[V11L,V16D] for α7 nAChR over other native nAChRs.

There is increasing evidence for a role for α7 nAChRs in both neural and non-neural signaling and plasticity (de Jonge and Ulloa, 2007; McKay et al., 2007). α7 nAChRs are a therapeutic target for psychiatric and neurological disorders with cognitive or attentional impairments and other non-neurological conditions (Mazurov et al., 2006; Chiamulera and Fumagalli, 2007). Investigation of the physiological and pathological roles of α7 nAChRs requires selective ligands. Synthetic efforts in this area have generated several novel α7 nAChR-selective agonists (Hajós et al., 2005; Jensen et al., 2005; Mazurov et al., 2006; Biton et al., 2007) and allosteric modulators (Hurst et al., 2005; Grønlien et al., 2007; Timmermann et al., 2007). However, antagonism of α7 nAChR responses relies on αBgt and MLA, whose utility is constrained by slow kinetics and lack of specificity, respectively. The development of α-CtxArIB[V11L,V16D] and demonstration that it displays high specificity for blockade of α7 nAChRs versus other recombinant nAChRs (including muscle and α9 nAChRs) represents a potentially important advance for α7 nAChR research (Whiteaker et al., 2007). Here, we have established that this specificity for α7 nAChRs is preserved with respect to functional native nAChRs and encompasses human and rat receptors.

Choline was employed to elicit responses from native α7 nAChRs. Choline is a weak α7 nAChR-selective agonist (Alkondon et al., 1997) and a noncompetitive inhibitor of heteromeric nAChRs, including the α3β4 subtype (Alkondon and Albuquerque, 2006). In PC12 and SH-SY5Y cells, the small responses to choline were substantially enhanced by PNU-120596, an allosteric modulator that selectively prolongs agonist-evoked α7 nAChR currents by reducing desensitization (Hurst et al., 2005; Grønlien et al., 2007). This is in agreement with our previous observations using another α7 nAChR-selective agonist, compound A, in PC12 cells (Dickinson et al., 2007). Likewise, the structurally unrelated α7 nAChR agonist AR-R17779 (Jensen et al., 2005) provoked negligible responses in SH-SY5Y cells when applied alone, but in the presence of PNU-120596, significant increases in intracellular Ca²⁺ resulted. Blockade of the amplified responses by αBgt confirms the specific activation of α7 nAChRs by choline and AR-R17779. α-CtxArIB[V11L,V16D] effectively inhibited choline- and AR-R1779-evoked increases in fluorescence, consistent with its antagonism of α7 nAChRs.

In addition to its potent and selective interaction with rat α7 nAChRs (this study; Whiteaker et al., 2007), α-CtxArIB-[V11L,V16D] inhibited recombinant human α7 nAChRs expressed in X. laevis oocytes or native human α7 nAChRs in SH-SY5Y cells. The time course of blockade and recovery indicates full inhibition within 2 min, with a slower recovery taking 15 to 20 min of washout to regain full agonist responsiveness. This rate of recovery is similar to that observed for rat α7 nAChRs in PC12 cells. Although the IC₅₀ values for inhibition of whole-cell currents recorded from oocytes expressing recombinant receptors (∼3 nM) were in close agreement for the two species, the IC₅₀ values for inhibition of α7 nAChR-evoked increases in intracellular Ca²⁺ in both rat and human cell lines were approximately 20-fold higher (50-80 nM). This is not attributable to differences between agonists used because choline and AR-R17779 resulted in comparable IC₅₀ values in SH-SY5Y cells. Although the IC₅₀ value for a competitive antagonist is influenced by agonist concentration, all the agonists were applied at a concentration approximating to their EC₅₀ value, nor was it because of using PNU-120596 to elicit measurable responses from the cell lines because the IC₅₀ value derived from oocytes was not significantly increased in the presence of the allosteric modulator, in agreement with the lack of effect of PNU-120596 on the ACh inactivation curve (Grønlien et al., 2007). Therefore, the consistent discrepancy between recombinant and native preparations is likely to reflect methodological differences between the assays. For example, the apparent potency of competitive antagonists is increased when the agonist duration is very short (Wyllie and Chen, 2007); oocyte recordings used 1-s agonist applications, compared with 20 s for the calcium fluorometry assays.

Brain α7 nAChRs that modulate striatal [³H]dopamine release from rat striatal minces were also substantially inhibited by 300 nM α-CtxArIB[V11L,V16D]. Although α7 nAChRs are generally accepted to be absent from dopaminergic terminals in the striatum, inhibition of the αBgt-sensitive portion of release by glutamate receptor antagonists supports the localization of α7 nAChRs on glutamatergic afferents (Kaiser and Wonnacott, 2000). This indirect action is discernible in striatal slices or minces that preserve some local anatomical connections and can be elicited with choline (Barik and Wonnacott, 2006). In the present experiments, PNU-120596 potentiated responses to 1 mM choline, and α-CtxArIB[V11L,V16D] was equally effective at blocking responses to choline alone and PNU-120596-potentiated responses.

The specificity of α-CtxArIB[V11L,V16D] for α7 nAChRs over other native nAChR subtypes is demonstrated by its failure to inhibit Ca²⁺ increases elicited by nicotine, 5-I-A85380, or KCl in PC12 cells or nicotine-evoked dopamine release from striatal slices. The inability of α-CtxArIB-[V11L,V16D] to displace [³H]epibatidine binding to rat brain membranes confirms that it does not interact with the agonist binding site of non-α7 nAChRs, in agreement with previous binding assays on mouse brain preparations (Whiteaker et al., 2007). In PC12 cells, non-α7 nAChRs comprise α3β2- and α3β4-containing subtypes (Henderson et al., 1994); the lack of any significant blockade by 300 nM α-CtxArIB[V11L,V16D] is consistent with the evaluation of recombinant α3β2 nAChRs in X. laevis oocytes (Whiteaker et al., 2007). Nicotine-evoked [³H]dopamine release is mediated by α4β2- and α6β2-containing nAChRs (Grady et al., 2007). The lack of effect of α-CtxArIB[V11L,V16D] on [³H]dopamine release evoked by 10 μM nicotine contrasts with the modest inhibition of this response by 30 to 100 nM MLA (Mogg et al., 2002). Moreover, MLA selectively inhibits αconotoxin MII-sensitive nAChR responses on dopaminergic cell bodies (Klink et al., 2001), and this predilection for α6β2-containing nAChRs confounds the use of MLA as an α7 nAChR-selective antagonist in tissue preparations where α6β2^* nAChRs might be present. The present evidence for the insensitivity to α-CtxArIB[V11L,V16D] of α4β2- and α6β2-containing nAChRs that mediate striatal dopamine release is supported by its very weak effect on recombinant α6/α3β2β3 nAChRs (IC₅₀ > 20 μM; Whiteaker et al., 2007), giving α-CtxArIB-[V11L,V16D] a clear advantage over MLA in this respect. Together with its lack of effect at α9α10 nAChRs and α1 muscle-type nAChRs (see Whiteaker et al., 2007), α-CtxArIB-[V11L,V16D] exhibits greater selectivity for α7 nAChRs than either MLA or αBgt.

These findings extend the previous characterization of α-CtxArIB[V11L,V16D] (Whiteaker et al., 2007) and establishes that it is a specific inhibitor of both rat and human native α7 nAChRs. It exhibits relatively fast blockade and is slowly reversible over several minutes, offering advantages over the very slow kinetics of αBgt inhibition. Its high degree of selectivity, with no effect on α6β2-containing nAChRs or α9 nAChRs, is superior to that of MLA. With effective blockade achieved by submicromolar concentrations, α-CtxArIB-[V11L,V16D] will be a useful addition to the portfolio of nicotinic ligands.