Neuronal nicotinic α7 acetylcholine receptors (α7nAChRs) are expressed primarily in the brain and are implicated in modulating many cognitive functions (e.g., attention, working and episodic memory). Not surprisingly, much effort has been committed to the development of molecules acting at α7nAChRs as potential therapies for a variety of central nervous system diseases (e.g., Alzheimer's). N-[(3S)-1-azabicyclo[2.2.2]oct-3-yl]-1H-indazole-3-carboxamide hydrochloride (RG3487) binds potently to the human α7nAChR (Ki = 6 nM), in which it acts as a partial agonist (63–69% of acetylcholine) as assessed by whole-cell patch-clamp recordings in both oocytes and QM7 cell lines. RG3487 activates human α7nAChRs with an EC50 of 0.8 μM (oocytes) and 7.7 μM (QM7 cells). RG3487 also exhibits antagonist properties at the serotonin 3 receptor [IC50 = 2.8 nM (oocytes), 32.7 nM (N1E-115 cells)]. In vivo, RG3487 improved object recognition memory in rats after acute [minimally effective dose (MED) 1.0 mg/kg p.o.] or repeated (10 day) administration at brain and plasma concentrations in the low-nanomolar range. Spatial learning deficits in age-impaired rats were reversed after RG3487 administration (MED: 0.03 mg/kg i.p.) as evaluated in the Morris water maze task. In the prepulse inhibition (PPI) of startle model of sensorimotor gating, RG3487 improved apomorphine-induced deficits in PPI performance (MED: 0.03 mg/kg i.p.) and reversed phencyclidine-induced impairments in an attentional set-shifting model of executive function (MED: ≤0.03 mg/kg i.p.). Cumulative evidence from these studies indicates RG3487 is a novel and potent α7nAChR partial agonist that improves cognitive performance and sensorimotor gating.
Nicotine is capable of improving cognitive processes, including attention, working memory, and long-term memory, across a variety of species (Newhouse et al., 1997; Rezvani and Levin, 2001; Pichat et al., 2007). The α7 and α4β2 nicotinic acetylcholine receptor (nAChR) subtypes are the most prominent in the central nervous system and are believed to mediate the procognitive properties of nicotine. Moreover, nAChRs have been implicated in disease states characterized by cognitive impairments, including Alzheimer's, schizophrenia, and attention-deficit hyperactivity disorder, suggesting a potential role of the nicotinic system in disease pathology (Rezvani and Levin, 2001). Thus, much attention has been devoted to exploring the potential for novel therapies directed toward nAChRs (Bencherif and Schmitt, 2002).
nAChRs belong to the superfamily of ligand-gated receptors that include GABAA, glycine, and serotonin-3 (5-HT3) receptors (Gotti et al., 2006; Dani and Bertrand, 2007). The structural and functional diversity of the nAChRs derives from 10 α and four β subunits that combine to form three main heteropentameric receptors (α4β2, α6α4β2β3, and α6β2β3), and one homopentameric receptor (α7). The α7nAChR is the only known homopentameric receptor in the mammalian brain, with five identical ACh binding sites located at the interface between each subunit. In addition, evidence has demonstrated that the α7 subunit can combine with the β2 subunit in vitro and in vivo (Liu et al., 2009).
In particular, the α7nAChR is an attractive target for cognitive impairing disorders in that this receptor subtype is highly expressed in key brain regions involved with cognitive processing (e.g., hippocampus, cortex) with limited peripheral expression (Levin and Rezvani, 2000; Gotti et al., 2006), and activation of the α7nAChR has been shown to improve cognitive function. For example, N-[(3S)-1-azabicyclo[2.2.2]oct-3-yl]-1H-indazole-3-carboxamide hydrochloride (RG3487) has been reported to improve sustained attention in rodents after acute administration (Rezvani et al., 2009). The α7nAChR is characterized primarily by its high permeability to calcium ions and its rapid desensitization upon agonist binding (Séguéla et al., 1993). Localization of the α7nAChR is both presynaptic and postsynaptic, and this receptor subtype is involved in numerous processes including modulation of neurotransmitter release, regulation of postsynaptic excitability, long-term potentiation, and cognitive function. Nicotine has been shown to enhance excitatory synaptic neurotransmission through activation of presynaptic α7nAChRs (McGehee et al., 1995), and these data have been corroborated by the use of selective α7nAChR agonists [e.g., (4-bromophenyl)1,4-diazabicyclo[3.2.2]nonane-4-carboxylate (SSR-180711)] (Pichat et al., 2007). In addition, activation of α7nAChRs by 2-methyl-5-(6-phenyl-pyridazin-3-yl)-octahydro-pyrrolo[3,4-c]pyrrole (A-582941) has been shown to enhance extracellular signal-regulated kinase 1/2 and cAMP response element binding phosphorylation, both important postsynaptic mediators of long-term memory (Bitner et al., 2007), which may contribute to the procognitive properties of the α7nAChR-selective agonists.
The α7nAChRs, rather than high-affinity nicotinic or muscarinic cholinergic receptors, are central to the function of auditory sensory gating, in which schizophrenic patients show marked deficits (Martin et al., 2004). Activation of α7nAChRs has also been shown to increase GABAergic neurotransmission, which is hypothesized to restore sensory gating deficits associated with schizophrenia (Hajós, 2005). In addition, abnormal P50 suppression has been linked to genetic markers at the locus of the α7nAChR subunit gene on chromosome 15q13-14, and α7nAChR expression is reduced in postmortem brain tissue from schizophrenic patients. In agreement with these hypotheses, administration of α7nAChR agents (e.g., tropisetron) and the nonselective nicotinic agonist nicotine reverse the P50 auditory gating deficit observed in schizophrenic patients (Simosky et al., 2002). Evidence has emerged that α7nAChR agonists may be therapeutic for cognitive deficits in schizophrenia. Administration of the α7nAChR partial agonist 3(2,4-dimethoxybenzylidene)anabaseine (GTS-21) to nonsmoking schizophrenic patients significantly improved cognitive performance in the Repeatable Battery for Assessment of Neuropsychological Status test, and it normalized the P50 auditory evoked potential (Olincy et al., 2006). However, GTS-21, as well as RG3487, did not reverse cognitive impairments in schizophrenics in tests using the MATRICS scale (Freedman et al., 2008; Umbricht et al., 2009). Thus, it remains to be elucidated whether α7nAChR agonists can improve cognitive deficits in these patients.
Overall, there is a large unmet medical need for safe and efficacious treatments for cognitive disorders (e.g., Alzheimer's). The studies in this article describe the pharmacological characterization of RG3487 (formerly known as MEM3454), a novel partial agonist of the α7nAChR with 5-HT3 receptor antagonist properties currently being evaluated for the treatment of cognitive impairments in human disease.
Materials and Methods
Receptor Rat Brain Membrane Preparations.
Rat brains (Pel-Freez Biologicals, Rogers, AR) were placed into ice-cold 0.32 M sucrose containing protease inhibitor (Protease Inhibitor Cocktail Tablets; Roche Diagnostics, Indianapolis, IN) (one tablet per 50 ml). The tissue was homogenized and centrifuged for 10 min at 1,000g at 4°C. The supernatant was centrifuged at 20,000g at 4°C. The pellet was then homogenized in ice-cold water for 10 s and centrifuged for 30 min at 39,000g at 4°C. The last two steps were repeated twice, and the remaining pellet was resuspended in binding buffer [200 mM Tris-HCl, 20 mM HEPES, pH 7.5, 144 mM NaCl, 1.5 mM KCl, 1 mM MgSO4, 2 mM CaCl2, 0.1% (w/v) BSA]. The protein concentration was adjusted to approximately 2.5 mg/ml. The membrane preparation was stored at −80°C.
Receptor Saturation and Competition Binding Studies.
Saturation binding for α7nAChR was carried out by incubating 200 μg of membrane protein with increasing concentrations of the selective α7nAChR antagonist, [3H]methyllycaconitine (MLA; 0.02–20 nM), in a final volume of 200 μl of binding buffer for 2 h at room temperature. Competition binding assays were performed by incubating 200 μg of membrane proteins with 1 nM to 10 μM RG3487 dissolved in buffer [200 mM Tris-HCl, 20 mM HEPES, pH 7.5, 144 mM NaCl, 1.5 mM KCl, 1 mM MgSO4, 2 mM CaCl2, 0.1% (w/v) BSA] and 2 nM [3H]MLA for 2 h at room temperature. The nonspecific binding was defined by 10 μM MLA.
Saturation binding for the α4β2nAChR was carried out by incubating 300 μg of membrane protein with increasing concentrations of the α4β2nAChR partial agonist [3H]cytisine (0.02–30 nM) in a final volume of 1000 μl of binding buffer for 75 min at 4°C. Competition binding assays were performed by incubating 300 μg of membrane proteins with 1 nM to 100 μM RG3487 dissolved in buffer (50 mM Tris-HCl, pH 7.4, 120 mM NaCl, 5.0 mM KCl, 2 mM MgCl2, 2 mM CaCl2) and 1 nM [3H]cytisine for 75 min at 4°C. The nonspecific binding was defined by 1 mM nicotine.
Receptor binding for the 5-HT3R was assessed by using human recombinant 5-HT3 receptors expressed in human embryonic kidney 293 cells using standard transfection methods. RG3487 was tested to displace [3H]BRL-43694 (granisetron; 2 nM), a 5-HT3R selective antagonist, in rat brain cortical membranes for 60 min at room temperature. Competition binding assays were performed by incubating 2 μg of membrane protein with 0.001 to 100 μM RG3487 dissolved in buffer (50 mM Tris-HCl, pH 7.5, 5 mM MgCl2, 1 mM EDTA) and 2 nM [3H]BRL-43694 for 60 min at room temperature. Tropanyl 3,5-dichlorobenzoate (MDL72222), a 5-HT3R selective antagonist, was used to assess nonspecific binding.
For all binding experiments, the binding reaction was terminated by vacuum filtration onto Filtermat A filter presoaked with 0.3% polyethyleneimine using a Tomtec (Hamden, CA) harvester. After the filter was washed with buffer (50 mM Tris-HCl, pH 7.5) and dried in a microwave oven, the radioactivity was counted with a Trilux (PerkinElmer Life and Analytical Sciences, Waltham, MA) Microbeta counter. All raw data were analyzed with Prism (GraphPad Software Inc., San Diego, CA) using the three-parameter logistic equation.
Neuroblastoma cells (SK-N-SH, ATCC, HTB-11; American Type Culture Collection, Manassas, VA) expressing α7nAChR, α3β2nAChR, α3β4nAChR, and other nAChRs were used in the FLIPR assay (Molecular Devices Sunnyvale, CA) to test compound cross-reactivity. Cells were seeded overnight in 100 μl of Dulbecco's modified Eagle's medium, supplemented with 10% fetal bovine serum (FBS). On the second day, the cells were incubated with 1× Hank's buffered salt solution, 20 mM HEPES, 0.1% BSA, 2.5 mM probenecid, and 0.01% Pluronic acid F-127, pH 7.4 buffer in the presence of Fluo-3/AM (Molecular Probes, Carlsbad, CA) at 37°C for 60 min. After three washes to remove extracellular Fluo-3/AM, the plates were filled with 100 μl of buffer and then placed onto a FLIPR. Compound addition was performed by the FLIPR pipetting system. Fluorescence was monitored (λex = 488 nM, λEM = 540 nM) for 3 min immediately after the compound addition. The relative fluorescence unit was measured as peak fluorescence intensity minus basal fluorescence intensity.
For FLIPR studies with the human rhabdomyosarcoma muscle cells (ATCC) expressing α1β1γδ nAChR the cells were seeded overnight in 100 μl of Dulbecco's modified Eagle's medium, supplemented with 10% fetal bovine serum. On the second day, the cells were incubated with Dulbecco's modified Eagle's medium + 10% FBS, 20 mM HEPES, 0.1% BSA, 2.5 mM probenecid, and 0.01% Pluronic F-127, pH 7.4 buffer in the presence of 4 μM Fluo-3/AM (Molecular Probes) at 37°C for 60 min in a 50-ml VWR (West Chester, PA) centrifuge tube. After three washes with buffer (1× Hank's buffered salt solution, 20 mM HEPES, 0.1% BSA, 2.5 mM probenecid Pluronic F-127, pH 7.4) to remove the Fluo-3/AM, the cells were seeded into 96-well black wall and clear-bottom plates (Corning Costar; Corning Life Sciences, Lowell, MA) at a density of 100,000 cells per well in 100 μl of the same buffer, and then placed onto the FLIPR (Molecular Devices). Compound additions were performed by using the FLIPR pipetting system. The fluorescence was monitored (λex = 488 nM, λEM = 540 nM) for 3 min immediately after the compound addition. The relative fluorescence unit was assessed as stated above.
RG3487 receptor binding studies conducted at neurotransmitter receptors, ion channels, and enzymes were conducted at Cerep (Redmond, WA) using standard protocols and procedures described on their website (www.cerep.com).
Whole-Cell Patch-Clamp Recordings Using Dynaflow.
Whole-cell recordings were made at room temperature using fire-polished borosilicate recording pipettes (∼2–5 MΩ; World Precision Instruments, Inc., Sarasota, FL) filled with the following solution: 140 mM CsCl, 4 mM NaCl, 10 mM HEPES, 2 mM CaCl2, 1 mM MgCl2, and 5 mM EGTA, pH 7.3. Evoked currents were acquired by using pClamp 9.0 software (Molecular Devices) with an Axon Multiclamp 700A amplifier (Molecular Devices), low-pass filtered at 2 kHz, and digitized with a Digidata 1322 digitizer (Molecular Devices) at 5 kHz. The Dynaflow chip (Cellectricon, Inc., Gaithersburg, MD) was loaded with 80 μl of drug solution per well and grounded, and the recording area was filled with ∼2 ml of extracellular solution (Ringer's solution: 140 mM NaCl, 5 mM KCl, 10 mM HEPES, 2 mM CaCl2, 1 mM MgCl2, 10 mM glucose, and 1 μM atropine, pH 7.3). Cells were dissociated from the growth plate by gently washing the surface with 1 ml of extracellular solution and then transferred to the Dynaflow chip. Drug delivery was driven by pressure from a syringe pump (12 μl/min) attached to the Dynaflow chip.
For α7nAChR recordings, a QM7 cell line stably expressing the human α7nAChR was used. QM7 cells were cultured in medium 199, 10% FBS, 0.26% tryptose phosphate, and 1% DMSO. During the recordings, cells were held at −70 mV, and the holding potential was stepped to −90 mV during drug application (1 s) by using the Dynaflow system. The responses were analyzed by evaluating both peak response and total net charge (i.e., integration of the area under the curve). RG3487 agonism was assessed by calculating evoked currents as a percentage of the cell's maximal response to acetylcholine (1–3 mM).
For 5HT3R recordings, a N1E-115 cell line stably expressing human 5HT3Rs was used to record native 5HT3R currents. Inward 5HT3R currents were evoked by applying 5-HT (10 μM) for 1 to 2 s to the cell by using the Dynaflow system. Drug antagonism was assessed by coapplication with 5-HT. Drugs were prepared from frozen DMSO (≤ 0.1%) for RG3487 or diH2O for ACh and 5-HT stock solutions immediately before use.
All experiments were carried out on human nAChRs expressed in Xenopus oocytes by using the method of cDNA expression. Xenopus oocytes were prepared and injected using standard procedures as described previously (Hogg et al., 2008). In brief, ovaries were harvested from Xenopus laevis females that were deeply anesthetized. A small piece of ovary was isolated for immediate preparation, while the remaining part was placed at 4°C in a sterile Barth solution containing 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO3, 10 mM HEPES, 0.82 mM MgSO4 · 7H2O, 0.33 mM Ca(NO3)2 · 4H2O, and 0.41 mM CaCl2 · 6H2O, at pH 7.4, and supplemented with 20 μg/ml of kanamycine, 100 unit/ml penicillin, and 100 μg/ml streptomycin. All recordings were performed at 18°C, and cells were superfused with OR2 medium containing 82.5 mM NaCl, 2.5 mM KCl, 5 mM HEPES, 1.8 mM CaCl2 · 2H2O, and 1 mM MgCl2 · 6H2O, pH 7.4, and 1 μM atropine was added to prevent possible activation of endogenous muscarinic receptors.
Injections of cDNAs encoding for the human α7, α3, α4, α5, α6, β2, or 5HT3A,B were performed in at least 100 oocytes by using a proprietary automated injection device (Hogg et al., 2008), and receptor expression was examined at least 2 days later. Oocytes were penetrated with two electrodes, and their membrane potential was maintained at −80 mV throughout the experiment using standard voltage-clamp procedures.
Currents evoked by ACh or other agonists were recorded by using an automated process equipped with a standard two-electrode voltage-clamp configuration. Unless indicated, cells were held at −80 mV. Data were captured and analyzed by using a HiQScreen proprietary data acquisition and analysis software running under Matlab (The Mathworks Inc., Natick MA).
ACh was prepared as a concentrated stock solution (10−1 M) in water and then diluted in the recording medium to obtain the desired test concentration. Nicotine was diluted in water as stock solution (10−1 M), kept frozen, and diluted immediately before the experiment at the desired concentration. Compounds were prepared as stock solution (10−1M) in DMSO and then diluted in the recording medium to obtain the desired test concentration. Residual DMSO did not exceed the concentration of 1%, a concentration that has no effect on Xenopus oocyte function.
Novel object recognition (NOR) and locomotor activity (LMA) tasks used male Sprague-Dawley adult rats (Hilltop Labs, Scottsdale, PA). The Morris water maze (MWM) test used young adult (3–6 months) and aged (22–24 months) male Fischer F344 rats obtained from Hilltop Labs. For attentional set-shifting male Long-Evans rats (Harlan, Indianapolis, IN) were used. For prepulse inhibition of startle male albino Wistar adult rats (Harlan) were used. All animals were maintained in temperature-controlled rooms, and all procedures were in compliance with local institutional animal care and use committee and Association for Assessment and Accreditation of Laboratory Animal Care guidelines.
RG3487 was dissolved in saline (0.9% sodium chloride) or deionized water (diH20) and was administered intraperitoneally 30 to 60 min before testing, except for the object recognition test, in which it was administered orally 60 min before training, or intraperitoneally immediately after training (consolidation experiment). Ondansetron was dissolved in diH20 and administered orally 60 min before testing. Methyllycaconitine was dissolved in diH20 and administered intraperitoneally 60 min before training. Apomorphine, dissolved in saline (0.9% sodium chloride), was administered subcutaneously 10 min before testing. Haloperidol was dissolved in saline and administered intraperitoneally 30 min before testing. Phencyclidine (PCP) HCl (5.0 mg/ml) was prepared in saline and administered twice daily (8:00 AM and 8:00 PM) intraperitoneally for 7 days. The test doses and pretreatment times of the compounds were selected based on the pharmacokinetic properties of the molecules, previously published scientific literature, and/or personal experience of the investigators. All compounds were obtained from Sigma-Aldrich (St. Louis, MO) except for RG3487, which was synthesized in-house (Roche/Memory).
Plasma Concentration Analysis.
RG3487 plasma concentrations were determined by using a validated assay. In brief, RG3487 and the internal standard, [13C6]-RG3487, were extracted by protein precipitation into organic acetonitrile containing [13C6]-RG3487 using 0.20 ml of plasma. An aliquot of this extract was injected into a high-performance liquid chromatography system coupled to a tandem mass spectrometer. The analytes were separated by reverse-phase chromatography and detected using the selected reaction monitoring mode of tandem mass spectrometry. Quantitation of plasma sample concentration was carried out by the comparison of the ratio of RG3487/[13C6]-RG3487 response against a calibration curve in the range of 0.250 to 512 ng/ml for RG3487 in plasma.
Novel Object Recognition.
Before training, rats were acclimated to laboratory conditions (e.g., daily handling, weighing) and habituated to the test environment (e.g., placed in an opaque plastic training/testing chamber, 78.7 × 39.4 × 31.8 cm with bedding on the floor, inside dimly lit (4 lux) room for 10 min). During training, rats were placed in the chamber containing two identical objects [either a metal tower (23 cm in height) or a metal conical cone with a chrome top (23 cm in height, 7.5 cm in diameter)] for a 15-min session. The objects were placed 14 cm from the sides of the two short walls and 18 cm from the sides of the long walls of the chamber; distance between the two objects was 25 cm. After a retention delay interval of 48 h, animals were tested for recognition memory. In the test session, one object identical to that used in training (familiar) and one new (novel) object were placed in the chamber, and the animal was allowed to explore for 5 min. The familiar and novel objects and chamber positions of objects were randomly assigned, and objects were cleaned between sessions with a 50% ethanol solution to eliminate olfactory cues. Object exploration was scored when the animal directed its nose to the object at a distance of ≤2 cm and/or touched it with its nose; climbing or sitting on the object was not considered object exploration behavior. The primary behavioral measure was percentage of time spent investigating the novel object time (seconds) [i.e., novel time/(novel + familiar time × 100].
Age-Impaired Morris Water Maze.
The Morris water maze test assesses spatial learning and memory. In brief, animals were placed in a 1.5-m-diameter circular tank filled with 25°C water and had three training sessions each day for 5 days using spatial cues around the room to locate a hidden escape platform (12 cm diameter) submerged 2 cm under the water. Maximum swim time was 120 s, and the intertrial interval was approximately 20 min. The start location for each trial was one of seven different sites, each of which was used once per day. Based on mean escape latency for all trials on days 3 to 5, aged rats (22–24 months old) were classified as either impaired (AI) or unimpaired (AU). The scores of AU rats were ≤0.5 standard deviations from the mean of the response from the young group, whereas the AI score was ≥2.0 standard deviations higher than the response from the young group.
On training day 6, drug or vehicle was administered to AI animals 30 min before the beginning of training. The same protocol was followed for days 7 and 8, with the last trial on day 8 being a probe trial. On training day 9, all aged rats trained in the visible platform version of the task to assess visual acuity. In this paradigm the platform remained in the same position as during hidden platform training, but it was elevated 1 cm above the surface of the water and a black cylinder (10 cm high, 2.5 cm diameter) was placed in the center of the platform. The animals received four trials of 60-s duration, with a 30-s intertrial interval. If an animal failed to swim to the platform on any trial, it was classified as sensory/motor impaired and was removed from subsequent analysis.
Attentional Set Shifting.
The set-shifting task adapted for rodents uses olfactory and tactile stimuli, and experimental methods have been reported in detail previously (Rodefer et al., 2008). In brief, rats were trained to dig in small bowls filled with sand to retrieve food rewards (Honey Nut Cheerios; General Mills, Minneapolis, MN) using sensory cues (i.e., odor, digging media).
Initially, a simple discrimination (SD) between either two odors or two digging media was presented, followed by a compound discrimination (CD) with the same positive stimulus as the initial SD. In the CD, a new dimension was introduced, but it was not a reliable predictor of the location of the food reward. An intradimensional shift (IDS) task was then presented; the IDS task was a compound discrimination in which the specific stimuli within both relevant and irrelevant dimensions were changed, but the relevant dimension (either odor or medium) remained the same. The IDS task was then reversed, so that what was formerly the negative stimulus was changed to be the positive stimulus, with the irrelevant dimension still not predictive of the location of the reward. Then, the rats were presented with an extradimensional shift (EDS) task in which the formerly irrelevant dimension became the relevant one, whereas the originally relevant dimension no longer held predictive value. Finally, the EDS task was reversed such that the formerly negative stimulus became the positively reinforced stimulus. EDS direction (i.e., odor to medium or medium to odor) was counterbalanced across subjects and was without effect on any of the behavioral variables (p > 0.05), so it was not considered in the presentation of results.
After subchronic PCP injections, rats experienced a washout period of 10 days before beginning habituation and training to the set-shifting procedure. On the day of testing, rats were administered a dose of RG3487 or vehicle 30 min before the test session.
Prepulse Inhibition of Startle.
The prepulse inhibition of startle paradigm has been described in detail previously (Hohnadel et al., 2007). In brief, trials consisted of a prepulse (20-ms burst of white noise with intensities of 75, 80, or 85 dB) followed 100 ms later by a startle stimulus (120 dB, 20-ms white noise). Animals were acclimated to test conditions for 2 days before the study began. On the day of drug testing, the rats were placed in the startle chamber (San Diego Instruments, San Diego, CA) and received 12 startle trials, 12 no-stimulus trials, and 12 trials of each of the prepulse/startle trials presented pseudo-randomly for a total of 60 trials. The intertrial interval ranged from 10 to 30 s, and the total session lasted approximately 25 to 30 min. The startle trials consisted of single 120-dB white noise bursts lasting 20 ms. During the no-stimulus trial, no startle noise was presented, but the movement of the rat was scored and represented a control trial for detecting differences in overall activity. Basal startle amplitude was determined as the mean amplitude of the 12 startle trials. Prepulse inhibition was calculated according to the formula 100 − 100% × (PPx/P120), in which PPx is the mean of the 12 prepulse inhibition trials (i.e., for each individual prepulse level), and p120 is the basal startle amplitude. The average level of PPI was also calculated (average of the responses to pp75, pp80, or pp85) and analyzed separately.
LMA was measured by using activity chambers (42 × 42 × 30 cm; AccuScan Instruments Inc., Columbus, OH) equipped with infrared photo sensors to measure horizontal and vertical activity. Twenty-four hours before testing, animals were brought to the laboratory and acclimated before being weighed, injected with saline (intraperitoneally), and individually placed in the clear Plexiglas open-field chamber located in a dimly lit (40 lux) room for 60 min of habituation. Approximately 24 h later, animals were brought back to the laboratory and acclimated before administration of vehicle or RG3487 (0.1–10 mg/kg i.p.) in which LMA was recorded for 60 min. The primary behavioral measure was total distance traveled in the chamber.
For most paradigms (NOR, PPI, MWM, LMA), data were analyzed by a one-way (between group) analysis of variance followed by post hoc comparisons to identify individual differences. All comparisons were made with an experimental type I error rate (α) set at 0.05. The attentional set-shifting data analyses were based on previously described methods (Rodefer et al., 2008). In brief, set-shifting performance across all six phases of task was examined (SD, CD, IDS, IDS-reversal, EDS, and EDS-reversal) between animals treated with subchronic PCP or saline using a series of a priori planned contrasts using independent samples t tests. All group variances were evaluated by using Levene's test for equality of variances. After performing the planned contrasts, performance in the EDS phase with an analysis of variance, using corrected post hoc analyses tested mean differences.
RG3487 is composed of N-[(3S)-1-azabicyclo[2.2.2]oct-3-yl]-1H-indazole-3-carboxamide hydrochloride (C15H19ClN4O) with a molecular weight of 306.8 g/mol (as the HCl salt form). The chemical structure is shown in Fig. 1.
RG3487 Receptor Binding Studies.
RG3487 displaced binding of the α7nAChR antagonist [3H]MLA (0.02–20 nM), with a Ki value of 6 nM (n = 3) in rat membrane preparations. In addition, the selectivity of RG3487 was investigated at a concentration of 10 μM at binding sites for 60 different membrane and soluble receptors, ion channels, and monoamine transporters (Table 1). RG3487 inhibited binding by less than 50% in all cases, except at the 5-HT3R. In a binding assay using human recombinant 5-HT3Rs expressed in human embryonic kidney 293 cells, RG3487 was tested to displace [3H]BRL-43694, a 5-HT3R selective antagonist. In this study, RG3487 showed high affinity to 5-HT3Rs with a Ki value of 1.2 nM.
The effects of RG3487 in various in vitro cell biology assays were also investigated. RG3487 was profiled at a concentration of 10 μM against a broad panel of enzymes (30 enzymes). The results showed less than 20% inhibition for all enzymes tested (Table 1).
RG3487 Acts as a Partial Agonist at α7nAChRs.
The ability of RG3487 to activate α7nAChRs was evaluated by using the whole-cell patch-clamp electrophysiological recordings in oocytes stably expressing the human α7nAChR. In this study, RG3487 (0.1–300 μM) evoked a maximum peak current (Imax) that was 63% of the current evoked by ACh (1280 μM), suggesting that RG3487 is a partial agonist at the α7nAChR (Fig. 2a). The EC50 of RG3487 at the human α7nAChR expressed in oocytes was 0.8 ± 0.15 μM with a Hill coefficient of 1.4 ± 0.08 (n = 4). Using net charge analysis, the EC50 of RG3487 was 0.18 ± 0.018 μM with a Hill coefficient of 1.5 ± 0.32 (n = 3; not shown). The evoked current decreased at concentrations above 10 μM RG3487.
To confirm that RG3487 acts directly as an agonist of the α7nAChRs, a competition experiment was performed with the specific antagonist MLA. Preincubation (5 min) with 10 nM MLA fully inhibited the RG3487-evoked currents, confirming that these responses arise from the activation of α7nAChRs (not shown).
In addition to the activation of the α7nAChRs expressed in oocytes, we investigated the agonist activity of this compound by using whole-cell patch-clamp recordings of human α7nAChR stably expressed in QM7 cell lines. A range of RG3487 (0.1–100 μM) concentrations was tested, and the Imax was 69% of the maximal response to ACh (3 mM) (Fig. 2b). The EC50 of RG3487 in QM7 cells was 7.7 ± 1.7 μM (peak current, n = 5) compared with ACh, which exhibited an EC50 = 268 ± 22 μM (peak current, n = 6) with a Hill coefficient of 1.24 ± 0.09. When measured by net charge, the EC50 of RG3487 was 0.152 ± 0.041 μM (n = 5) with a Hill coefficient of 1.6 ± 0.67 and Imax 53% of ACh. For ACh, the net charge EC50 = 22 ± 1.4 μM (n = 6) and the Hill coefficient was 2.5 ± 0.34.
Selectivity of RG3487 for α7nAChRs Versus Other nAChR Subtypes.
To evaluate the properties of RG3487 at non-α7nAChR subtypes, oocytes expressing the desired receptor combination were first challenged with a reference ACh (30 μM) test pulse to assess evoked current amplitude, and then were challenged with increasing concentrations of RG3487. In oocytes expressing the human α4 and β2 nACh subunits, RG3487 (0.1–1000 μM) did not evoke any measurable currents. Likewise, RG3487 tested at 10 and 100 μM evoked no detectable currents at the following nACh receptors: α4β2α5, α4β2α6β3, α4β2α6, α3β2, and α2β2 (not shown).
In addition, RG3487 was assessed for potential cross-reactivity by using a FLIPR functional assay at α3β2nAChR and α3β4nAChR expressed in SK-N-SH cell lines and α1β1γδnAChR expressed in human rhabdomyosarcoma muscle cell lines. Application of RG3487, at concentrations of up to 100 μM, did not trigger any calcium mobilization signals, indicating that RG3487 does not have cross-activity at these nAChRs subtypes as expressed in these cells.
RG3487 Is a Potent Antagonist at 5-HT3R.
Using Xenopus oocytes stably expressing the human 5-HT3A,B receptor, the functional effects of RG3487 were assessed by whole-cell patch-clamp electrophysiology. Currents evoked by 10 μM 5-HT were measured after incubation with RG3487 (0.1–300 nM). RG3487 dose-dependently inhibited 5-HT-evoked current when assessed in this system with an IC50 = 2.84 ± 0.81 nM and a Hill coefficient of 0.83 ± 0.06 nM (n = 4) (Fig. 3a). The current responses were normalized to the 5-HT Imax.
Further in vitro investigation of the interaction of RG3487 with 5-HT3Rs was conducted in N1E-115 cell lines using whole-cell patch-clamp recordings in the Dynaflow system. 5-HT (10 μM) reliably evoked currents from N1E-115 cells with an EC50 of 5.5 ± 0.18 μM and a Hill coefficient of 2.68 ± 0.14 (Fig. 3b) that were abolished with the coapplication of the selective 5-HT3R antagonist ondansetron (100 nM; data not shown). When coapplied with 5-HT (10 μM), RG3487 inhibited the 5-HT3R-mediated current with an IC50 of 32.7 ± 0.9 nM, and a Hill slope equal to 1.3 ± 0.4. RG3487 (0.001–1 μM) did not exhibit any agonist properties when applied alone (data not shown).
Desensitization Properties of RG3487 at α7nAChR and α4β2nAChR.
One of the key features of nAChRs is their strong desensitization upon exposure to an agonist. To explore the effects of RG3487 on desensitization, Xenopus oocytes expressing the α7nAChR were exposed to sustained (45 s) increasing concentrations of RG3487 (0.003–1 μM) and then stimulated with a standard pulse of ACh (100 μM). In this study, RG3487 desensitized the α7nAChRs in a dose-dependent manner (IC50 = 40 ± 4.1 nM; Hill coefficient 2.7 ± 0.08) (Fig. 4). Note that for the low concentration of the RG3487 an increase of the peak current is observed and a reduction of the current becomes apparent for concentrations more than 10 nM.
Using a similar protocol, investigation into the desensitization effects of RG3487 on heteromeric α4β2nAChRs was also conducted at 100 nM. Two control responses evoked by a brief ACh (30 μM) test pulse were recorded, and three responses were successively recorded in the presence of RG3487. In addition, a control response was measured at the end to evaluate receptor recovery. The results from this experiment revealed that at concentrations up to 100 nM RG3487 did not alter the α4β2nAChR responses (data not shown).
Plasma Concentrations of RG3487 in Rats.
After oral administration of RG3487 (0.3–10 mg/kg) in fasted male Sprague-Dawley rats, the mean Cmax was typically observed between 0.5 and 4 h (Table 2). In addition, to assess brain penetration of the compound, RG3487 was administered orally at 10 mg/kg to rats, and animals were sacrificed at 1 h after dose. Brains were removed and homogenized. Concentrations of RG3487 measured 1 h after administration were 43.8 ± 11.8 ng/ml. Dividing by the plasma concentration at 1 h, brain-to-plasma ratios were 0.13 ± 0.01.
RG3487 Selectively Enhances Object Recognition Memory.
To investigate the effects of RG3487 on episodic memory, rats were tested in the novel object recognition test. Administration of RG3487 (0.1–10 mg/kg p.o.) before training significantly increased object recognition memory at the 48-h retention interval in a dose-related manner (F5, 37 = 9.667; p < 0.05) (MED = 1.0 mg/kg p.o.) (Fig. 5a). Similar effects were observed after intraperitoneal administration of RG3487 (0.01–1.0 mg/kg) (F3, 22 = 4.634; p < 0.05) (data not shown).
Pretraining administration of RG3487 suggests that its procognitive effects may reflect improvements in the acquisition and/or consolidation of the recognition memory. To investigate the effect on consolidation specifically, administration of RG3487 directly after object recognition training was explored. In this paradigm, RG3487 (0.1 and 1.0 mg/kg i.p.) significantly enhanced the percentage of time exploring the novel object when tested at the 48-h retention delay interval compared with vehicle-treated controls (F3,25 = 7.404; p < 0.05) (Fig. 5b).
To confirm that the RG3487-induced improvement in recognition memory was mediated through the α7nAChR, animals were treated with the selective α7nAChR antagonist MLA (0.3125, 1.25, and 5 mg/kg i.p.) in combination with 3.0 mg/kg p.o. of RG3487. MLA (1.25 and 5 mg/kg) significantly antagonized the procognitive effects of RG3487 (F4,28 = 6.241; p < 0.05) (Fig. 5c). Because RG3487 is also a potent antagonist at the 5-HT3 receptor, we investigated the effects of the selective 5-HT3R antagonist ondansetron (0.1, 1, and 10 mg/kg p.o.) in the novel object recognition model. Ondansetron did not show any improvement in recognition memory at the 48-h retention delay (F3,21 = 0.168; p > 0.05) (Fig. 5d), suggesting that the cognitive enhancing effect of RG3487 in the object recognition model was caused by activation of the α7nAChR.
It was of interest to determine whether the procognitive effects of RG3487 could be observed after repeated administration. To this end, RG3487 (3 mg/kg i.p.) was administered once daily for 10 days to determine its effects on object recognition memory. Repeated administration of RG3487 significantly enhanced the percentage of time animals spent investigating the novel object at the 48-h delay interval (t17 = −5.680; p < 0.05) (Fig. 5e) similar to the results from the acute administration study, suggesting that pharmacological tolerance to repeated daily dosing did not occur with RG3487.
Improvement in Spatial Memory by RG3487 in Age-Impaired Animals.
In the Morris water maze test, age-impaired animals were identified and administered either RG3487 (0.01–10 mg/kg i.p.) or vehicle before each of 3 additional training days. RG3487 treatment in rats significantly improved the aged-impaired performance deficits compared with vehicle-treated age-impaired animals (F6,473 = 5.977; p < 0.05) (Fig. 6). Post hoc analyses revealed a significant reduction in swim latencies at 0.03–0.3 mg/kg doses of RG3487, whereas higher doses (1 and 10 mg/kg) were inactive. In age-impaired rats, the RG3487-mediated improvements in spatial navigation were reduced as the dose of the compound increased (rats: ≥ 1 mg/kg), yielding a U-shaped dose-response curve. There were no statistically significant differences in swim speeds or visual acuity in age-impaired animals treated with RG3487 versus vehicle for either species (data not shown). These results demonstrate that RG3487 effectively attenuates age-related impairments in spatial learning and memory rats.
RG3487 Improves Executive Function after Subchronic PCP Administration.
In this study, subchronic PCP treatment produced a significant impairment in number of trials to criterion only in the EDS phase (t16 = 3.79; p < 0.05). There were no suggestions of PCP-induced impairment on trials to criterion in SD, CD, IDS, IDS-reversal, or EDS-reversal (all p > 0.05) consistent with previous data (Rodefer et al., 2008). Animals that received an acute injection of RG3487 (0.03–1 mg/kg i.p.) before testing in the set-shifting paradigm exhibited a treatment-dependent improvement in performance in the EDS (F4,40 = 11.59, p < 0.05). Dunnett's post hoc analyses indicated that significant effects of RG3487 were observed at all doses tested (p < 0.05) compared with the saline-treated PCP rats. Thus, acute administration of RG3487 attenuated the PCP-induced deficit on EDS performance in the set-shifting task (Fig. 7).
RG3487 Reverses Apomorphine-Induced Deficits in Sensorimotor Gating.
In this study, the prepulse stimuli used (75, 80, and 85 dB) clearly inhibited the startle response to a 120-dB auditory stimulus in a fashion that depended on the prepulse intensity (i.e., the greater the decibel level of the prepulse, the greater the inhibition of the startle response). Moreover, apomorphine (0.5 mg/kg s.c.) significantly diminished PPI, which was significantly antagonized by the positive control haloperidol (0.3 mg/kg i.p.). Administration of RG3487 (0.01–1.0 mg/kg i.p.) attenuated the apomorphine-induced deficit in PPI in young male rats in a dose-dependent manner (F6,90 = 6.71; p < 0.05) (Fig. 8).
RG3487 Does Not Alter Locomotor Activity.
RG3487 was also assessed for its effects on locomotor activity and rearing behavior in Sprague-Dawley rats in a 60-min session. At the doses tested, RG3487 (0.1–10 mg/kg i.p.) did not significantly alter total distance traveled (F3,14 = 1.029; p > 0.05) or rearing behavior (F3,14 = 0.237; p > 0.05) (not shown).
Enhancing cognitive performance via activation of the α7nAChR represents a promising new approach to treating diseases such as Alzheimer's. The present studies were conducted to characterize the in vitro and in vivo pharmacological properties of the novel compound RG3487. In particular, RG3487 selectively activated the α7nAChR without affecting other nicotinic receptors (i.e., >100-fold selectivity versus other nicotinic subtypes). In addition, RG3487 did not show any appreciable binding affinity for >90 other G protein-coupled receptors, ion channels, monoamine transporters, or enzymes, except for the 5-HT3R at which it acted as an antagonist. When tested in vivo, RG3487 demonstrated consistent procognitive properties after acute and repeated administration in healthy and impaired animal models. RG3487 also improved sensorimotor gating deficits in rats. These data suggest RG3487 has a pharmacological profile ideal for investigating the role of the α7nAChR in cognitive disorders.
RG3487 exhibited low-nanomolar binding affinity at the α7nAChR in which it acted as a partial agonist compared with ACh in both oocytes and QM7 cells. Moreover, RG3487-mediated current activation in oocytes could be abolished by the selective α7nAChR antagonist MLA, confirming its agonist activity at the α7nAChR. A difference in peak current EC50 values was noted between the oocytes and the QM7 cells, which may be caused by differing experimental conditions and/or differences in the physiological and pharmacological properties of the α7nAChR expressed, a finding that has been reported previously with other α7nAChR agonists (Biton et al., 2007). In addition, it is interesting to highlight the differences between peak current-derived EC50 values that exhibited micromolar potencies compared with the left-shifted EC50 values derived from net charge analysis in the nanomolar range between the oocyte and cell line systems. It has been shown that EC50 values for α7nAChR agonists derived from peak current amplitudes may underestimate agonist potency, owing to the rapid desensitizing properties of the receptor (Papke and Porter Papke, 2002). Thus, presumed differences in the rate of drug addition in our two recording systems may have introduced variability, contributing to the difference between the peak-derived EC50 values in oocytes and cell lines.
One of the key characteristics of the α7nAChR is rapid desensitization after agonist binding (Revah et al., 1991; Séguéla et al., 1993). Sustained exposure of Xenopus oocytes to increasing concentrations of RG3487 progressively prevented ACh from activating α7nAChR, but not α4β2nAChR-mediated currents, consistent with the idea that RG3487 desensitized α7nAChRs in a concentration-dependent manner. Blockade of ACh-evoked currents after sustained exposure of RG3487 occurred at concentrations approximately 1000-fold less than those necessary to activate the receptor in oocytes. Critically, this low concentration range of RG3487 approximates the efficacious brain and plasma levels of compound identified in rat after oral administration. It is noteworthy that a pulse of ACh applied after sustained exposure of 3 and 10 nM RG3487 potentiated the ACh-evoked current, which may reflect enhanced cooperativity of agonist binding at these low concentrations, as has been reported previously for the α4β2nAChR (Smulders et al., 2005).
The 5-HT3R belongs to the superfamily of ligand-gated ion channels that includes the nAChRs (Maricq et al., 1991). Significant sequence homology exists between 5-HT3Rs and α7nAChRs including the ligand binding domain, and cross-reactivity of certain compounds (e.g., tropisetron) has been reported previously (Macor et al., 2001). Subsequently, we determined that RG3487 acts as a potent 5-HT3R antagonist, because it inhibited 5-HT-induced current in both Xenopus oocytes and N1E-115 cell lines expressing 5-HT3Rs and did not evoke currents when applied alone.
In an effort to expand on the in vitro characterization of RG3487, the compound was further assessed in multiple in vivo models of cognition. RG3487 produced dose-related improvements in episodic memory function in both young and age-impaired animal models as measured in the NOR and Morris water maze tasks, respectively. The observed procognitive effects of RG3487 were mediated by the α7nAChR in that the improvement in recognition memory could be blocked by administration of MLA. These data suggest that acute activation of the α7nAChR is sufficient to produce procognitive effects and is similar to previous reports with other selective α7nAChR agonists (Bitner et al., 2007; Pichat et al., 2007). In addition, the RG3487-mediated improvement in episodic memory is in alignment with previously published data showing that this compound improves sustained attention after acute administration (Rezvani et al., 2009).
We also investigated the potential procognitive properties of 5-HT3R inhibition using the NOR model, because of the high-affinity binding and functional antagonist properties of RG3487 at this receptor. Antagonism of the 5-HT3R has been shown to improve cognitive performance in some animal paradigms (e.g., pharmacologically induced hypocholinergic models) (Hodges et al., 1996), but not in all in vivo experimental systems tested (Pitsikas and Borsini, 1997). Using the selective, competitive 5-HT3R antagonist ondansetron, no improvement in recognition memory was observed in the present study. These data combined with MLA blockade of the cognitive-enhancing effects of RG3487, as well as the reported procognitive properties of selective α7nAChR agonists themselves (e.g., A-582941, SSR-180711) (Bitner et al., 2007; Pichat et al., 2007), suggest that α7nAChR activation and not 5-HT3R antagonism mediates the procognitive properties of RG3487.
In addition to the acute nootropic effects of RG3487, no evidence of tachyphylaxis was observed after chronic (i.e., 10 day) administration of this compound as tested in the NOR model. Although rapid desensitization of the α7nAChR is a key characteristic of this ligand-gated ion channel (Revah et al., 1991; Séguéla et al., 1993), it seems that a balance between receptor activation and desensitization may occur with RG3487 that results in a net procognitive effect after repeated dosing. This effect may be caused by pharmacokinetic properties of RG3487, as well as the low concentrations of this molecule that are sufficient to activate a subset of α7nChRs at any given moment allowing time for other previously activated receptors to recover from desensitization. Overall, these data make it tempting to speculate that transient inactivation of the α7nAChR after RG3487 binding does not translate into a functional limitation for this drug target.
The administration of RG3487 either before training or immediately after training improved long-term memory in the NOR model, indicating that RG3487 can influence both acquisition and consolidation phases of memory formation. One mechanism underlying this effect may be caused by enhanced neurotransmission through activation of α7nAChRs localized presynaptically (Radcliffe and Dani, 1998). To this end, selective activation of α7nAChRs in primary neuronal culture systems and in vivo has been shown to enhance extracellular concentrations of neurotransmitters with recognized involvement in cognitive processes (e.g., glutamate, acetylcholine, dopamine) (Radcliffe and Dani, 1998; Biton et al., 2007; Pichat et al., 2007).
The procognitive effects of RG3487 described may also be mediated by postsynaptic activation of α7nAChRs. Not only have α7nAChRs been identified postsynaptically, but their high Ca2+ permeability suggests that metabotropic Ca2+-mediated second messenger signaling could underlie its cognitive enhancing properties (Berg and Conroy, 2002). In support of this concept, activation of α7nAChRs can trigger Ca2+-induced Ca2+ release from internal stores (Dickinson et al., 2007), which has been linked to transcriptional regulation and synaptic plasticity (Berg and Conroy, 2002). To this end, Bitner et al. (2007) report activation of the extracellular signal-regulated kinase 1/2 pathway and subsequent downstream phosphorylation of cAMP response element binding protein, a critical mediator of synaptic plasticity after administration of the selective α7nAChR agonist A-582941.
Both Alzheimer's and schizophrenic patients exhibit alterations in the expression of the α7nAChR (Freedman et al., 1995; Court et al., 1999), which may contribute to the cognitive impairments observed in these diseases. In addition, genetic polymorphisms in the promoter region of the α7nAChR gene on chromosome 15 have been linked to sensory gating deficits in schizophrenic patients (Leonard et al., 1996). When tested using the PPI of startle model of sensorimotor gating, RG3487 improved the ability of the animal to inhibit responding after apomorphine-induced impairments. These data compare favorably with standard antipsychotic agents (e.g., risperidone, haloperidol) in the PPI model and are in agreement with some, but not all, results reported for other α7nAChR agonists in this model (Stevens et al., 1998; Schreiber et al., 2002).
In some of the in vivo assays we used to test RG3487 (i.e., PPI, MWM), a U-shaped dose response was observed, which has been reported previously for nAChR agonists (Picciotto, 2003; Feuerbach et al., 2009). Notably in the oocyte system, a decrease in RG3487-evoked currents was observed at concentrations above 10 μM, which may be caused by rapid receptor desensitization, a well recognized property of the α7nAChR (Revah et al., 1991; Séguéla et al., 1993), or open channel blockade (Colquhoun et al., 1979). These in vitro findings may provide a mechanistic explanation for the inverted U-shaped dose responses that were observed in vivo.
In addition to improving sensorimotor gating deficits, RG3487 improved executive function deficits induced by subchronic PCP administration using an attentional set-shifting paradigm, which is considered an analogous task to the Wisconsin Card Sort Test in humans (Rodefer et al., 2008). Schizophrenic patients, as well as animals that receive a subchronic regimen of PCP, show a specific disruption in the ability to acquire the extradimensional set-shift discrimination that may be attributable to N-methyl-d-aspartate receptor hypofunction. Typical and atypical antipsychotic medications, which are useful in treating the positive symptoms of schizophrenia, do little to improve cognitive dysfunction in this patient population and in the extradimensional set shifting (Rodefer et al., 2008). It remains to be determined whether agonists directed at the α7nAChR will improve executive function in schizophrenic patients, because the clinical data to date have shown moderate to no effect on cognitive parameters in this patient population (Olincy et al., 2006; Freedman et al., 2008; Umbricht et al., 2009).
The development of α7nAChR agonists as potential therapeutic agents for cognitive impairment in diseases such as Alzheimer's is supported by data showing improvement in attention, learning, and memory after activation of this receptor subtype. RG3487 is a selective α7nAChR partial agonist that exhibits procognitive effects in healthy and age-impaired animals across multiple cognitive domains. In addition, RG3487 effectively reverses pharmacologically induced sensorimotor and executive function impairments, two paradigms postulated to model aspects of schizophrenia. These preclinical findings suggest that RG3487 may be a promising therapeutic agent for treating cognitive deficits in human disease.
Participated in research design: Wallace, Callahan, Tehim, Bertrand, Tombaugh, Wang, Xie, Rowe, Terry, Rodefer, Herbert, Murray, Santarelli, and Lowe.
Conducted experiments: Callahan, Bertrand, Tombaugh, Wang, Xie, Rowe, Ong, Graham, Terry, Rodefer, Herbert, and Murray.
Contributed new reagents or analytic tools: Wang, Xie, Ong, and Murray.
Performed data analysis: Wallace, Callahan, Bertrand, Tombaugh, Wang, Xie, Rowe, Terry, Rodefer, Murray, and Lowe.
Wrote or contributed to the writing of the manuscript: Wallace, Callahan, Bertrand, Tombaugh, Wang, Rowe, Terry, Rodefer, Porter, and Lowe.
We thank Sonia Bertrand, Sofya Dragan, Shuangdan Sun, and Daguang Wang for technical contributions.
This work was supported by F. Hoffmann-La Roche, Ltd. and Memory Pharmaceuticals, Inc.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- nicotinic acetylcholine receptor
- novel object recognition
- Morris water maze
- prepulse inhibition
- minimally effective dose
- serotonin 3
- 5-HT3 receptor
- N-[(3S)-1-azabicyclo[2.2.2]oct-3-yl]-1H-indazole-3-carboxamide hydrochloride
- bovine serum albumin
- fetal bovine serum
- dimethyl sulfoxide
- deionized water
- locomotor activity
- tropanyl 3,5-dichlorobenzoate
- age impaired
- age unimpaired
- simple discrimination
- compound discrimination
- intradimensional shift
- extradimensional shift.
- Received July 15, 2010.
- Accepted October 18, 2010.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics