Partial agonist therapies for the treatment of nicotine addiction and dependence depend on both agonistic and antagonistic effects of the ligands, and side effects associated with other nAChRs greatly limit the efficacy of nicotinic partial agonists. We evaluated the in vitro pharmacological properties of four partial agonists, two current smoking cessation drugs, varenicline and cytisine, and two novel bispidine compounds, BPC and BMSP, by using defined nAChR subtypes expressed in Xenopus laevis oocytes and human embryonic kidney 293 cells. Similar to varenicline and cytisine, BPC and BMSP are partial agonists of α4β2 nAChRs, although BMSP produced very little activation of these receptors. Unlike varenicline and cytisine, BPC and BMSP showed desired low activity. BPC produced mecamylamine-sensitive steady-state activation of α4* receptors that was not evident with BMSP. We evaluated the modulation of α4*- and α7-mediated responses in rat lateral geniculate nucleus (LGN) neurons and hippocampal stratum radiatum (SR) interneurons, respectively. The LGN neurons were sensitive to a very low concentration of varenicline, and the SR interneuron responses were also sensitive to varenicline at a submicromolar concentration. Although 300 nM BPC strongly inhibited the ACh-evoked responses of LGN neurons, it did not inhibit the α7 currents of SR interneurons. Similar results were observed with 300 nM BMSP. Additionally, the bispidine compounds were efficacious in the mouse tail suspension test, demonstrating that they affect receptors in the brain when delivered systemically. Our data indicate that BPC and BMSP are promising α4β2* partial agonists for pharmacotherapeutics.
Partial agonist-based therapies are attractive for indications where there is a natural or induced imbalance in a neural regulatory system, because partial agonists can function as activity buffers, preventing overactivation by exogenous or endogenous agents and providing basal activity on their own in the absence of other stimuli. Both of these forms of action may be important for management of drug addiction, dependence, and mood symptoms associated with nicotine and cigarette smoking.
There is a rich variety of nicotinic acetylcholine receptor (nAChR) subtypes in the brain and periphery (for review, see Millar and Gotti, 2009). Receptors are pentamers, and there are ten α-subunits in vertebrates and seven non-alpha subunits (β1, β2, β3, β4, γ, δ, and ε), of which α2−α10 and β2−β4 are expressed in neurons and the others in muscle. Studies of knockout mice (Picciotto and Kenny, 2013) have implicated receptors containing α4 and β2 subunits in the reinforcing/addicting effects of nicotine. Receptors containing these subunits at varying ratios and sometimes also incorporating α5 subunits constitute the majority of the high-affinity nicotine-binding receptors in brain. Partial agonists at (α4β2*) nAChRs may be useful not only for managing nicotine reward and dependence but also for managing the mood and depression symptoms that are comorbid with smoking and smoking cessation (Mineur and Picciotto, 2010).
The most abundant types of α4 subunit-containing receptors (collectively designated as α4* receptors) include one configuration classified as a low sensitivity (LS) type with three α4 subunits and two β2 subunits [α4(3)β2(2)], which responds to ACh and nicotine with progressively larger responses across a relatively wide range of concentrations. A second configuration is classified as a high sensitivity (HS) subtype with two α4 subunits and three β2 subunits [α4(2)β2(3)], which responds to low concentrations of ACh and nicotine and with maximal currents limited at high concentrations. A third type of HS α4* receptor has two α4 subunits and two β2 subunits and an α5 subunit [α4(2)β2(2)α5]. The generation of nAChR subunit concatamers (Zhou et al., 2003; Kuryatov and Lindstrom, 2011; Stokes and Papke, 2012) has allowed for heterologous expression of receptors with defined subunit composition, an approach we have taken advantage of in this study.
Another class of receptors that has been associated with nicotine reward contains the α6, β2, and other subunits (such as α4β2α6β2β3). The α6 and β3 subunits are highly expressed in catecholaminergic neurons, particularly in the ventral tegmentum and substantia nigra. In ex vivo preparations, blocking these α6* receptors with highly selective toxins decreases nicotine-evoked dopamine release (Grady et al., 2002), and blockade or knockout of these nAChRs in vivo decreases nicotine self-administration (Pons et al., 2008; Brunzell et al., 2010; Gotti et al., 2010).
In the pursuit of drugs to manage nicotine dependence, three classes of receptors may be hypothesized to be “off-target”: receptors of the neuromuscular junction, receptors in autonomic ganglia (α3β4*), and the homopentamers of α7 subunits. Although muscle-type receptors are unequivocally off target, receptors containing α3, β4, and α5 subunits are also found in limited areas of the brain, where they have been implicated in aversive effects of high nicotine dosages. Polymorphisms in the α3-β4-α5 gene cluster have also been linked to smoking behavior. Activation of α7 receptors in the nucleus accumbens may decrease the motivation to self-administer nicotine (Brunzell and McIntosh, 2012) and has been implicated in the drive of people with schizophrenia to self-medicate by smoking (Leonard et al., 2007). Deficiencies in α7 function have been implicated in schizophrenia, and individuals with schizophrenia smoke at very high rates. These observations suggest that α7- and α3-containing receptors should not necessarily be considered off target but rather they may need to be targeted differently than α4* and α6* receptors. For people at risk for mental illness, it is likely that the function of α7 receptors should be especially spared.
The two drugs currently used as smoking cessation aids, cytisine and varenicline, have significant activity at α7 and α3* receptors, which may limit their utility and generate side effects. We report two novel compounds, BPC and BMSP (Fig. 1A), which modulate the brain nAChRs most strongly implicated in addiction, with reduced likelihood of perturbing the function of other nAChR such as the ganglionic α3β4 subtypes or brain α7 receptors. The extension of our studies with these agents will therefore be useful to ultimately evaluate which nAChR subtypes are the best targets for treating nicotine dependence. The development and characterization of such agents for smoking cessation aids may also have crossover value for other indications such as augmentation therapy for depression (Mineur and Picciotto, 2010).
Materials and Methods
Solvents and reagents were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO). Cell culture supplies were purchased from Life Technologies (Grand Island NY). Acetylcholine chloride (ACh), atropine, N-2,3,3-tetramethylbicyclo[2.2.1]heptan-2-amine hydrochloride (mecamylamine), dihydro-β-erythroidine hydrobromide, methyllycaconitine citrate, (−)-nicotine, and cytisine were purchased from Sigma-Aldrich. Varenicline tartrate was purchased from Tocris/R&D Systems, Inc. (Minneapolis, MN). BPC and BMSP were synthesized and provided by Dr. D. Guendisch. Fresh ACh stock solutions were made each day of experimentation. Stock solutions of BPC and BMSP were prepared in dimethyl sulfoxide and stored at −20°C. Working solutions were prepared freshly each day at the desired concentration from the stored stock.
Heterologous Expression of nAChRs in Xenopus laevis Oocytes.
Mouse muscle nAChR α1, β1, and δ clones used for receptor expression in X. laevis oocytes were obtained from Dr. J. Boulter (University of California, Los Angeles, CA), and the mouse ε clone was provided by Dr. P. Gardner (University of Massachusetts Medical School, Worcester, MA). Human nAChR clones and concatamers were obtained from Dr. J. Lindstrom (University of Pennsylvania, Philadelphia, PA). The human resistance-to-cholinesterase 3 (RIC-3) clone, obtained from Dr. M. Treinin (Hebrew University, Jerusalem, Israel), was coinjected with α7 to improve the level and speed of α7 receptor expression without affecting the pharmacological properties of the receptors (Halevi et al., 2003). Subsequent to linearization and purification of the plasmid cDNAs, cRNAs were prepared using the mMessage mMachine in vitro RNA transfection kits (Ambion/Life Technologies, Austin, TX).
Oocytes were surgically removed from mature female X. laevis frogs (Nasco, Ft. Atkinson, WI) and injected with appropriate nAChR subunit cRNAs as described previously (Papke and Stokes, 2010). Frogs were maintained in the Animal Care Service facility of the University of Florida, and all procedures were approved by the University of Florida Institutional Animal Care and Use Committee. In brief, the frog was first anesthetized for 15–20 minutes in 1.5 liter frog-tank water containing 1 g of ethyl 3-aminobenzoate methanesulfonate buffered with sodium bicarbonate. The harvested oocytes were treated with 1.25 mg/ml collagenase (Worthington Biochemicals, Freehold, NJ) for 2 hours at room temperature in a calcium-free Barth’s solution (containing in mM: 88 NaCl, 1 KCl, 2.38 NaHCO3, 0.82 MgSO4, 15 HEPES, and 12 mg/l tetracycline, pH 7.6) to remove the follicular layers. Stage V oocytes were subsequently isolated and injected with 50 nl of 5–20 ng of nAChR subunit cRNA. Recordings were carried out 1–7 days after injection.
Two-Electrode Voltage Clamp Electrophysiology.
Experiments were conducted using OpusXpress 6000A (Molecular Devices, Union City, CA) (Papke and Stokes, 2010). Both the voltage and current electrodes were filled with 3 M KCl. Oocytes were voltage-clamped at –60 mV. The oocytes were bath-perfused with Ringer’s solution (containing in mM: 115 NaCl, 2.5 KCl, 1.8 CaCl2, 10 HEPES, and 0.001 atropine, pH 7.2) at 2 ml/min for α7 receptors and at 4 ml/min for other subtypes. To evaluate the effects of different partial agonists on ACh-evoked responses of various nAChR subtypes expressed in oocytes, baseline conditions were defined by two initial applications of ACh made before coapplications of experimental compounds with the control ACh. The agonist solutions were applied from a 96-well plate via disposable tips, and the test partial agonists were either coapplied with ACh by the OpusXpress pipette delivery system for acute coapplication experiments or introduced into the bath using the OpusXpress system to switch the running buffer for bath application experiments. For the concentration-response study, drug applications alternated between ACh controls and experimental compounds. Unless otherwise indicated, drug applications were 12 seconds in duration followed by a 181-second washout period for α7 receptors and 6 seconds with a 241-second washout for other subtypes. A typical recording for each oocyte contained two initial control applications of ACh, an experimental compound application, and then a follow up control application of ACh to determine the desensitization or rundown of the receptors. The control ACh concentrations were 30 μM for α1β1εδ, 60 μM for α7, 100 μM for α3β4, 10 μM for (α4)2(β2)3 (the HS form), 100 μM for (α4)3(β2)2 (the LS form), 10 μM for α4β2α5, and 30 μM for β3α4β2α6β2.
Data were collected at 50 Hz, filtered at 20 Hz, analyzed by Clampfit 9.2 (Molecular Devices) and Excel 2003 (Microsoft, Redmond, WA), and normalized to the averaged peak current or net charge response of the two initial ACh controls (Papke and Porter Papke, 2002). Data were expressed as means ± S.E.M. from at least four oocytes for each experiment and plotted by KaleidaGraph 3.6.2 (Synergy Software, Reading, PA).
Whole-Cell Patch-Clamp Electrophysiology.
The A7R3HC10 cells stably expressing human α7 and human RIC-3 were generated from low passage number human embryonic kidney 293 cells obtained from American Type Culture Collection (Manassas, VA) (Williams et al., 2012). The A7R3HC10 cells were routinely cultured in Dulbecco’s modified Eagle's medium supplemented with 10% fetal bovine serum and 0.45 mg/ml Geneticin selective antibiotic (G418 sulfate) and 0.015 mg/ml hygromycin at 37°C with 5% CO2. For normal passaging, cells were dissociated with 1 mM ethylenediamine tetraacetic acid (EDTA) in calcium- and magnesium-free Hanks’ balanced saline solution (Life Technologies) to avoid nonselective damage to the α7 nAChRs expressed on the cell surface. Cells with 8–12 passages after stable transfection were used for whole-cell patch-clamp recordings.
The 12-mm glass coverslips (Thermo Fisher Scientific) were coated with 0.1 mg/ml poly-d-lysine (Sigma-Aldrich) at 37°C for 5 minutes. A7R3HC10 cells were plated onto the coverslips 1–4 days before recording. Whole-cell recordings were performed at room temperature using an Axopatch 200B amplifier (Molecular Devices). Briefly, cells were bathed in an external solution containing (in mM) 165 NaCl, 5 KCl, 2 CaCl2, 10 glucose, 5 HEPES, and 0.001 atropine, pH 7.35. Patch pipettes (3–5 MΩ) were pulled from borosilicate glass (o.d. of 1.5 mm and i.d. of 0.86 mm; Sutter Instruments, Novato, CA) using a Flaming/Brown micropipette puller (Model P-97; Sutter Instruments) and filled with an internal solution containing (in mM) 120 CsCl, 2 MgCl2, 10 EGTA, 10 HEPES, and 5 MgATP, pH 7.35. Cells were voltage-clamped at −70 mV and perfused with external solution at a flow rate of 4 ml/min. Local application of 1 mM ACh was made every 60 seconds using single-barrel pipettes attached to a Picospritzer III pressure system (General Valve Corp., Fairfield, NJ) with Teflon tubing (10–20 psi for 3 seconds). The drug pipette was positioned approximately 10–15 μm from the cell and loaded with 1.5 mM ACh because of the 1.5-fold dilution factor inherent in the picospritzer drug delivery (Williams et al., 2012). Five baseline responses induced by ACh alone were initially recorded, followed by a 20-minute bath application of specific agonist or partial agonist. To minimize the mixture of control solution (external solution alone) and drug solution (external solution containing specific agent), a Valve Driver II fluid control system (General Valve) was used to rapidly switch the running buffer. Recordings were filtered to 5 kHz and digitized at 20 kHz with a DigiData 1322A digitizer (Molecular Devices) using Clampex 9.2 (Molecular Devices). The access resistance, input resistance, and whole-cell capacitance were monitored throughout the experiment by a 10-millisecond/10-mV pulse before each response. Data were analyzed with Clampfit 10.3 (Molecular Devices) and Excel 2003. Cells with access resistance >40 MΩ or holding current >200 pA were excluded from analysis. Responses were measured as peak currents. Data were plotted by KaleidaGraph and represented as means ± S.E.M. of 4–7 cells.
Rat Brain Slice Recording.
Preparation of rat brain slices and whole-cell patch-clamp recordings were carried out as described previously (López-Hernández et al., 2009). All procedures involving animals were approved by the University of Florida Institutional Animal Care and Use Committee and were in accord with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Young male Sprague-Dawley (SD) rats were anesthetized with isoflurane (Patterson Veterinary Supply, Inc., Devons, MA) and swiftly decapitated. For whole-cell recordings of lateral geniculate nucleus (LGN) neurons, male SD rats of postnatal day 22–31 were used, while for interneurons of the stratum radiatum (SR), male SD rats of postnatal day 22–29 were used. Transverse (300 μm) whole-brain slices were prepared using a vibratome (Pelco, Redding, CA) and a high Mg2+/low Ca2+ ice-cold artificial cerebral spinal fluid (ACSF) containing (in mM) 124 NaCl, 2.5 KCl, 1.2 NaH2PO4, 2.5 MgSO4, 1 CaCl2, 10 d-glucose, and 25.9 NaHCO3, saturated with 95% O2–5% CO2. Slices were incubated at 30°C for 30 minutes and then left at room temperature until they were transferred to a submerged chamber (Warner Instruments, Hamden, CT) for recording. During experiments, slices were perfused at a rate of 2 ml/min with normal ACSF containing (in mM) 126 NaCl, 3 KCl, 1.2 NaH2PO4, 1.5 MgSO4, 2.4 CaCl2, 11 d-glucose, 25.9 NaHCO3, and 0.004 atropine, saturated with 95% O2–5% CO2 at 30°C. The LGN neurons and the SR interneurons were visualized with infrared differential interference contrast microscopy using an E600FN microscope (Nikon, Tokyo, Japan).
Patch-clamp recording pipettes and single-barrel drug application pipettes were pulled from borosilicate glass with an o.d. and i.d. of 1.5 and 0.86 mm, respectively (Sutter Instruments, Novato, CA). The recording pipettes were filled with an internal solution of (in mM) 125 K-gluconate, 1 KCl, 0.1 CaCl2, 2 MgCl2, 1 EGTA, 2 MgATP, 0.3 Na3GTP, and 10 HEPES, pH 7.35. Neurons were held at −70 mV, and a −10 mV/10 millisecond test pulse was applied before each response to determine access resistance, input resistance, and whole-cell capacitance. Local somatic applications of ACh (1 mM pipette concentration) were made every 30 seconds using single-barrel glass pipettes attached to a picospritzer (General Valve) with Teflon tubing (14–18 psi for 20 milliseconds). The single-barrel drug application pipettes were usually placed within 10–15 μm of the cell soma. In the coapplication experiments for each neuron, five baseline responses evoked by ACh were recorded, followed by responses evoked by applications of ACh in combination with bath application of the agonist, partial agonist, or antagonist. These agents were introduced into the ACSF using a syringe pump (KD Scientific, Holliston, MA) loaded with a 50-fold concentrated stock solution diluted to the final concentration in the perfusion line prior to entering the recording chamber (at a pump rate of 2.4 ml/h). Evoked responses were then recorded for 18–20 minutes. It should be noted that pressure application from a drug pipette containing 1 mM ACh delivered an effective concentration of approximately 30 μM to the soma of hippocampal SR interneurons in SD rat brain slices; thus a dilution factor of 30 would be expected for each ACh pulse pressure-ejected in our system (López-Hernández et al., 2007). Signals were recorded using a MultiClamp 700A amplifier (Molecular Devices), digitized using a Digidata 1322A (Molecular Devices), sampled at 20 kHz using Clampex 9.2, and filtered at 6 KHz. Data analysis was done with Clampfit 10.3 (Molecular Devices), Excel 2003, and Kaleidagraph 3.0.2. Cells with access resistances >60 MΩ or those requiring holding currents >300 pA for LGN neurons and >200 pA for SR interneurons were not included in the final analyses. Data are represented as means ± S.E.M. of 7–13 neurons.
Tail Suspension Test in Mice.
Adult C57BL/6 male mice (20–30 g, 3 to 5 months of age) were used for the experiments. The animals were housed in a room with controlled temperature (24 ± 1°C) and a 12-hour light/12-hour dark cycle. All animals were acclimatized to the laboratory environment for at least 48 hours before the experiments. Food and water were available ad libitum. The experimental protocols were approved by the Animal Care and Use Committee at Yale University. On the day of testing, mice were transferred to the testing room at least 30 minutes prior to the first test. The tail suspension test was carried out as previously described (Mineur et al., 2007). BPC and BMSP were administered intraperitoneally 30 minutes before the test at doses of 2 and 5 mg/kg, respectively, dissolved in phosphate-buffered saline (pH 7.4) and injected at a volume of 10 ml/kg. Mice were gently suspended by a small paperclip taped to the tip of their tail, about 4/5 the distance from the base. The total time spent immobile during the 6-minute testing period was recorded. Mice were returned to the holding room after every animal was tested.
Pharmacological Activity at nAChR Subtypes Expressed in X. laevis Oocytes.
We evaluated the in vitro pharmacological properties of BPC and BMSP (Fig. 1) compared with the current smoking cessation drugs using concatamers of human nAChR subtypes along with an excess of single free subunits in X. laevis oocytes, allowing for expression of defined subunit compositions (Zhou et al., 2003). All four agents showed relatively little activity for mouse muscle nAChR, evoking no more than 5% of the ACh response when applied alone (Fig. 2A) and producing relatively little antagonism of ACh-evoked responses when either coapplied with ACh (Fig. 2B) or preincubated with the receptors prior to ACh application (Fig. 2C), although BPC did produce a small inhibition at the highest concentration tested.
The agonist activity was evaluated for both HS and LS forms of human α4β2 nAChR and other nAChR subtypes (Fig. 3). Like varenicline and cytisine, BPC is a partial agonist of α4β2 receptors, especially of the HS subtype. However, BMSP produced very little activation of these α4β2 receptors. BPC had a level of partial agonist activity for human α4β2α5 receptors that was comparable to cytisine, whereas BMSP was least efficacious among the four agents on this receptor as well as α4β2α6β2β3 receptors. Compared with the cytisine-based molecules, the two bispidine compounds showed the reduced activity for the two alternative nAChR subtypes, α7 and α3β4.
Although the agonist activity of BMSP was too low to characterize effectively, BPC evoked significant currents when applied alone to α4* nAChR. Therefore we conducted full concentration-response studies, shown in Fig. 4. BPC was most efficacious for HS α4β2 [α4(2)β2(3)] receptors and, not surprisingly, least potent for LS α4β2 [α4(3)β2(2)] nAChRs (Table 1).
One of the important functions for partial agonists used as smoking cessation agents is through their ability to modulate the intrinsic responsiveness of the α4β2* receptors to nicotine and ACh. Therefore we tested the ability of these agents to decrease agonist (ACh)-evoked responses. Note that similar effects can be obtained with nicotine as an agonist. However, because nicotine is sequestered and concentrated in the oocytes, we relied on the more readily reversible effects of ACh to probe for receptor function.
Of the neuronal nAChR subtypes tested, HS α4β2 receptors were the most sensitive to all four agents as antagonists in acute coapplication experiments with ACh (Fig. 5), although they were least sensitive to BPC compared with the other three. All four agents inhibited α4β2α5 receptors, whereas α6* receptors were most sensitive to the current smoking cessation drugs using this simple coapplication protocol. Because of their intrinsic agonist activity, cytisine and varenicline caused an additive activation, rather than inhibition of α3β4 and α7 receptors, especially in the case of varenicline on α3β4 nAChR. These nAChR subtypes have been hypothesized to be off-target for smoking cessation therapies, and the bispidine compounds had relatively little effect on them. Note that because varenicline and cytisine differ significantly in their effects on human and rat α3β4 receptors (Stokes and Papke, 2012), we also tested BPC and BMSP on rat α3β4 receptors expressed in X. laevis oocytes. Neither drug produced significant inhibition of responses evoked by 100 µM ACh when coapplied at 10 µM, and likewise, neither compound produced significant responses when applied alone at 100 µM to cells expressing rat α3β4 receptors (data not shown).
Acute coapplication of nicotinic agents is not a very good model for investigating drug-receptor interactions in vivo, because therapeutic agents will be present before and during the endogenous release of ACh or the rapid delivery of nicotine to the brain via cigarette smoking. Therefore, we also tested the modulation of ACh-evoked responses when the drugs were preincubated with the receptors for 5 minutes (Fig. 6). With this protocol, several types of α4* receptors were sensitive to all four agents, with the greatest effects obtained on the HS α4β2 receptors. There was also a rather selective inhibition of α7-mediated ACh responses with varenicline.
Although preincubations better emulate in vivo drug interactions than do coapplications, a better model still is to use bath applications of the drugs at very low concentrations and observe the longer term perturbation of receptor function. We previously applied this approach to characterize varenicline and cytisine (Papke et al., 2011). Figure 7 shows that bath application of BMSP at submicromolar concentrations effectively downregulated the responses of all α4* receptors, other than those also including the α6 subunit. All of the α4* receptors were also functionally downregulated by bath application of 100 nM BPC (Fig. 8A).
At the end of bath application experiments, we routinely applied 100 µM mecamylamine to determine if there were steady-state currents that produced an apparent shift in baseline, as was previously reported for nicotine, varenicline, and cytisine (Papke et al., 2011). As shown in Fig. 8B, 100 nM BPC strongly decreased the size of the ACh-evoked responses and also generated a steady-state current revealed by mecamylamine application. This steady-state current was not observed with BMSP. For HS α4β2, α4β2α5, and α6* receptors, bath application of 100 nM BPC produced steady-state activation that was approximately 2–3% the size of the peak current responses to control applications of ACh.
Modulation of ACh-Evoked Responses in Cells Stably Expressing α7 and RIC-3.
We evaluated the effects of bath-applied nicotine and the α4* partial agonists on the modulation of ACh-evoked responses of A7R3HC10 cells which stably express human α7 nAChR (Williams et al., 2012). As shown in Fig. 9, 300 nM nicotine produced ∼50% downregulation in the ACh-evoked responses, which was greater than the effects observed when BPC or BMSP were bath-applied at 300 nM. In contrast, as expected from the oocyte data, α7-mediated responses were very sensitive to bath applications of varenicline, so that 50 nM varenicline produced ∼70% reduction in the ACh-evoked responses. It required higher than 1 µM of cytisine in the bath to produce a comparable downregulation of ACh evoked responses (1.7 µM cytisine produced 78 ± 3% inhibition after 15 minutes, data not shown).
Modulation of ACh-Evoked Responses of Native nAChR Subtypes in Brain Slices.
To investigate and compare these agents on native nAChR subtypes in the brain, we focused on two neuronal types: the primary neurons in the lateral geniculate nucleus (LGN), which we previously showed to express primarily α4* receptors (Papke and Thinschmidt, 2009), and the α7-expressing interneurons of the hippocampal stratum radiatum (SR) (López-Hernández et al., 2009). In the absence of modulation by bath-applied drugs, the ACh-evoked responses in LGN neurons were stable, or in the case of the SR interneurons, showed a small run-up (Supplemental Fig. 1A), as is sometimes seen in in vitro experiments (Papke et al., 2011). As expected, the responses of the LGN neurons were sensitive to bath application of dihydro-β-erythroidine and the SR interneurons to the α7-selective antagonist methyllycaconitine (Supplemental Fig. 1B).
The responses of LGN neurons were sensitive to a very low concentration of nicotine, whereas the SR interneuron responses were insensitive to nicotine at a submicromolar concentration (Fig. 10A). In contrast to the relatively selective effects of nicotine on LGN neurons, LGN neurons and SR interneurons were both sensitive to varenicline at submicromolar concentrations (Fig. 10B). Both types of neurons were approximately 10-fold less sensitive to bath applications of cytisine (Fig. 10C).
Although 300 nM BPC strongly inhibited the ACh-evoked responses of the LGN neurons, it produced no effects on the α7-mediated responses of SR interneurons (Fig. 11A). Similar results were obtained with BMSP, although as expected from the oocyte data the downregulation of α4β2 responses by 300 nM BMSP was less than observed with 300 nM BPC (Fig. 11B).
In Vivo Activity.
BPC and BMSP were evaluated for their efficacy in the tail suspension test in mice to determine whether they could alter behaviors associated with smoking cessation in vivo. This test is commonly used to evaluate potential antidepressant drugs, and cytisine was previously shown to reduce immobility in this assay (Mineur et al., 2007), whereas reports also suggest antidepressant-like effects of varenicline (Rollema et al., 2009). Both BPC and BMSP reduced immobility in this behavioral model at the concentrations tested (Fig. 12). The efficacy of the compounds in the mouse tail suspension test was similar to that of other nicotinic compounds tested previously and was equivalent to the effects of classic antidepressant medications on immobility time, although BPC and BMSP were somewhat less potent than cytisine or varenicline. The behavioral effects of BPC and BMSP in the tail suspension test also suggest that both agents are able to cross the blood-brain barrier and enter the brain.
For the purpose of developing α4β2* ligands with increased selectivity, cytisine was structurally simplified down to its bispidine (3,7-diazabicyclo[3.3.1]nonane) skeleton. The bispidine scaffold, which can form cation-pi/HB interactions with nAChRs, displays weak inhibition of α4β2* (Ki = 600 nM) and α4β2α6β2β3 nAChR subtypes and is an agonist at α7 nAChRs with negligible effects on α3β4* and muscle nAChRs (Tomassoli et al., 2011). It served as a synthetic starting point for new compound libraries, applying a hybrid (Meunier, 2008) and a template approach, where a second pharmacophoric element, a hydrogen bond acceptor motif, was introduced. These strategies led to compounds with improved α4* selectivity. The hybrid approach design, where two active pyridine-based nAChR ligands were overlapped at a hydrogen bond acceptor motif, generated BPC, which is a relatively efficacious α4* agonist. In contrast to the hybrid BPC, BMSP, with very low α4* efficacy, displays a carbonyl oxygen as its hydrogen bond acceptor functionality.
Our utilization of human nAChR subtypes makes our data relevant to therapeutics. By extending our studies to the native receptor subtypes in rodent brain, we both confirm the in vitro pharmacology and validate the drugs for further study in animal models. Although cytisine and varenicline have similar activity for human and rat α4* and α7 receptors (Papke and Heinemann, 1994; Papke and Porter Papke, 2002; Mihalak et al., 2006, Papke et al., 2011), we previously showed that they differ in their activity for human and rat ganglionic-type α3β4 receptors (Stokes and Papke, 2012), for which they produce significant activation but with differences in potency and efficacy, indicating a significant likelihood for side effects in humans, especially for varenicline. In contrast, our data show that BPC and BMSP have minimal activity for human and rat α3β4 receptors.
Varenicline and cytisine are sometimes referred to as α4β2-selective partial agonists; however, we confirm that both compounds have significant efficacy at α3β4 and α7. The high levels of α7 modulation shown by varenicline may be of particular concern because this agent has been associated with adverse neuropsychiatric events (McClure et al., 2009; Moore et al., 2011). Cardiovascular events that have occurred with varenicline (Ware et al., 2013) might be attributable to effects on ganglionic α3β4 receptors. Our data show that BPC and BMSP come much closer to selectively regulating the activity of α4*. BPC is especially effective at targeting the HS form of α4* receptors, which may be important in smokers (Lester et al., 2009).
Although α7 and α3β4* nAChR subtypes should not be dismissed as nonrelevant targets for the management of smoking cessation, the fact that these receptors can induce both positive and negative effects related to addictive behavior make them problematic targets for smoking cessation strategy. Conversely, the demonstrated ability of β2* nAChR blockade to blunt nicotine addiction, prevent relapse, and induce positive mood effects makes our focus highly relevant. It may not be possible to make a clear separation between on-target and off-target receptors for the treatment of nicotine dependence. Although numerous lines of evidence support α4* and α6* receptors as underlying the reinforcing effects of nicotine, special considerations must be given to other nAChR subtypes as well. Although primarily found in autonomic ganglia, there are α3β4 receptors in the brain, most highly concentrated in the medial habenula, and some coassembled with α5. The α5-containing receptors have been implicated in establishing aversive effects of high nicotine doses (Frahm et al., 2011), and, independent of α5, α3β4 receptors have also been implicated in nicotine reward and withdrawal (Jackson et al., 2013). Likewise, although homomeric α7 receptors in the brain are generally considered off-target for smoking therapies, these receptors contribute to circuits associated with reward (Mansvelder et al., 2002; Brunzell and McIntosh, 2012). The question then is, not only whether subtypes such as α7- and α5-containing receptors should be targeted but whether they should be targeted differently from α4* and α6* receptors. The ideal drug might downregulate the function of α4* and α6* receptors and enhance the function of α5-containing receptors and thereby increase the aversive effects of nicotine. Unfortunately, no such ideal drug exists, so at this point we must build up from the limited efficacy of the existing therapeutic drugs. Other preclinical drug development programs continue to work up from the cytisine scaffold (Mineur et al., 2009; Tasso et al., 2009; Sala et al., 2013). In some assays, the cytisine dimer (Sala et al., 2013) shows a similar progression in selectivity as the bispidine compounds we report, but the cytisine dimer has significant intrinsic activity for stimulating dopamine release, which may not be ideal.
In addition to effects on behaviors related to drug reward, decreasing activity of β2* nAChRs has repeatedly been shown to be antidepressant-like. The bispidines tested, like various cytisine derivatives, have positive effects in tests of antidepressant efficacy. However, comparisons of in vitro properties and in vivo behavioral effects of novel nicotinic compounds must be made carefully, because in vivo effects are limited by other properties, including brain penetration and clearance. For instance, although 5-bromo-cytisine has high affinity and is a weak partial agonist of α4β2 nAChRs, it is not effective in behavioral models of antidepressant efficacy when injected peripherally, but is effective when infused directly into the ventricles. Overall, the advancements toward the profile of an “ideal drug” based on the cytisine scaffold have been relatively small and incremental. Our basic starting point was the simpler bispidine scaffold; we were hoping that we could achieve a cleaner separation of α4 partial agonism from activity at α3β4 and α7, and to a large degree our data support the strength of that approach.
Although both BPC and BMSP have improved selectivity profiles over varenicline and cytisine, there are interesting differences between the two agents. Specifically, BPC has much greater intrinsic agonist activity than does BMSP, so comparison of the two agents may indicate whether symptoms such as dysphoria during withdrawal can be managed best with an agent that provides a baseline of stimulation in the absence of the drug. Such activity may contribute to the ability of a medication to maintain abstinence following smoking cessation.
Although decreasing activity of β2* nAChRs has repeatedly been shown to produce antidepressant-like effects in mice, several clinical trials have suggested that blockade of nAChRs in conjunction with administration of antiserotonergic antidepressants may augment the efficacy of antidepressant therapies in humans (Philip et al., 2009). However, a large clinical trial of a mecamylamine isomer was not successful, suggesting that the use of truly selective partial agonists with some intrinsic efficacy at specific nAChR subtypes, such as BPC, may be a more successful strategy than complete nicotinic blockade. Such agents may have better compliance, fewer side effects, and may be particularly effective for treating the mood symptoms that occur during smoking cessation (Moore et al., 2011).
A number of studies have shown that α6* nAChRs are involved in nicotine self-administration in rodents (Pons et al., 2008; Brunzell et al., 2010; Gotti et al., 2010). The current smoking cessation drugs cytisine and varenicline both have higher efficacy at α4β2α6β2β3 nAChRs than do the bispidine compounds, but it is not known whether this is important for their therapeutic efficacy. Because blockade of nAChRs containing the α4 subunit is sufficient to block nicotine reward (McGranahan et al., 2011), it is not clear whether activity at both α4* and α6* nAChRs will be necessary for an effective medication or whether only one of the two is essential. The identification of BMSP, which is highly specific for α4* nAChRs, will be extremely useful for answering this question.
Just as sparing α7 receptor function may be an important consideration in a smoking cessation therapy, equal consideration should be given to the self-medication needs of smokers suffering from, or at risk of, mental illness. The incidence of smoking is twice as high in the mentally ill as in the normal population and four times higher than normal in schizophrenia. In the United States, 30% of all smokers suffer from mental illness (Mackowick et al., 2012). Although varenicline may be especially bad for such individuals because of its potent suppression of α7 activation, a bispidine-type drug lacking any α7 effects may not fulfill the secondary drive for nicotine self-delivery that is not simply involved with dopamine-mediated reward.
In conclusion, we characterized two novel agents that offer potentially important alternative approaches for the experimental investigation and management of nicotine dependence and withdrawal. In vitro experiments with nAChR receptors and these agents indicate that they have a higher degree of selectivity for α4* and α6* nAChR than existing therapeutic agents and therefore will be useful in testing whether modulating the activity of these receptors will be sufficient to manage nicotine dependence with potentially fewer side effects. Their in vitro activity profiles correctly predicted their activity on native nAChR in brain slices, and, importantly, they were effective at sites in the mouse brain when given systemically. Moreover, the efficacy of these agents in the tail suspension test promotes their potential utility for managing the mood symptoms that occur during smoking cessation and nicotine withdrawal.
The authors thank Dr. Nicole A. Horenstein for many helpful discussions. OpusXpress experiments were conducted by Shehd Abdullah Abbas Al Rubaiy, Matthew R. Kimbrell, Lu Wenchi Corrie, and Sarhadelia Pinheiro.
Participated in research design: Peng, Mineur, Picciotto, Papke.
Conducted experiments: Peng, Tian, Mineur.
Contributed new reagents or analytic tools: Guendisch, Eibl, Tomassoli.
Performed data analysis: Peng, Stokes, Mineur, Papke.
Wrote or contributed to the writing of the manuscript: Peng, Stokes, Guendisch, Picciotto, Papke.
- Received May 29, 2013.
- Accepted August 19, 2013.
This work was supported by the National Institutes of Health National Institute of General Medical Sciences [Grant R01-GM57481]; by the National Institutes of Health National Institute of Mental Health [Grant MH077681] (Y.S.M. and M.R.P.); by the National Institutes of Health National Center for Research Resources [Grant 5P20RR016467] (to C.E., I.T., and D.G.); and by James and Esther King Biomedical Research KG12.
- artificial cerebral spinal fluid
- N-2,3,3-tetramethylbicyclo[2.2.1]heptan-2-amine hydrochloride
- high sensitivity
- lateral geniculate nucleus
- low sensitivity
- nicotinic acetylcholine receptor
- human resistance-to-cholinesterase 3
- stratum radiatum
- Copyright © 2013 by The American Society for Pharmacology and Experimental Therapeutics