We define the pharmacological and pharmacokinetic profiles of a novel α2C-adrenoceptor agonist, compound A [N-[3,4-dihydro-4-(1H-imidazol-4-ylmethyl)-2H-1,4-benzoxazin-6-yl]-N-ethyl-N′-methylurea]. This compound has high affinity (Ki) for the human α2C-adrenoceptor (Ki = 12 nM), and 190- to 260-fold selectivity over the α2A- and α2B-adrenoceptor subtypes. In cell-based functional assays, compound A produced good agonist (EC50 = 166 nM) and efficacy (Emax = 64%) responses at the α2C-adrenoceptor, much lower potency and efficacy at the α2A-adrenoceptor (EC50 = 1525 nM; Emax = 8%) and α2B-adrenoceptor (EC50 = 5814 nM; Emax = 21%) subtypes, and low or no affinity and functional activity at the α1A-, α1B-, and α1D-adrenoceptor subtypes. In the human saphenous vein postjunctional α2C-adrenoceptor bioassay, compound A functions as a potent agonist (pD2 = 6.3). In a real-time contraction bioassay of pig nasal mucosa, compound A preferentially constricted the veins (EC50 = 108 nM), and the magnitude of arteriolar contraction reached only 50% of the maximum venular responses. Compound A exhibited no effect on locomotor activity, sedation, and body temperature in mice (up to 100 mg/kg) and did not cause hypertension and mydriasis (30 mg/kg) in conscious rats. Compound A is orally bioavailable (24%) with good plasma exposure. This compound is a substrate for the efflux P-glycoprotein transporter, resulting in very low central nervous system (CNS) penetration. In summary, compound A is a highly selective, orally active, and non-CNS-penetrating α2C-adrenoceptor agonist with desirable in vitro and in vivo pharmacological properties suitable for the treatment of nasal congestion.
Heterogeneity between α2-adrenoceptors was recognized approximately 30 years ago (Timmermans and van Zwieten, 1982), and the α2-adrenoceptors have since been divided into three subtypes (Bylund et al., 1994), first the α2Aand α2B subtypes (Bylund et al., 1988) and later the α2C subtype (Blaxall et al., 1991), based on pharmacological analysis and molecular cloning evidence. All three α2-adrenoceptor subtypes are highly homologous and well conserved across mammalian species and share common signal transduction pathways and properties (MacDonald et al., 1997). The three human α2-adrenoceptors belong to the seven-transmembrane G protein-coupled receptors and share ∼75% sequence identity in their membrane-spanning domains (Kurose and Lefkowitz, 1994). All three subtypes are coupled to the Gi/o signaling system, inhibiting activity of adenylate cyclase, inhibiting the opening of voltage-gated Ca2+ channels, and activating K+ channels (MacDonald et al., 1997).
The α2-adrenoceptor subtypes are differently distributed in cells and tissues and possess different physiological functions (Civantos Calzada and Aleixandre de Artiñano, 2001). Understanding the function of each specific α2-adrenoceptor subtype has been difficult because of the lack of sufficient subtype-selective adrenoceptor agonists and antagonists with other adrenoceptors (α1- and β-adrenoceptors) and with respect to other α2-adrenoceptor subtypes. Therefore, a strong need exists to develop subtype-selective α2-adrenoceptor compounds to identify the pharmacological properties attributable to each specific α2-adrenoceptor subtype. Among the α2-adrenoceptor subtypes, it has been suggested that the α2C-adrenoceptor was implicated as a potential target for the treatment of nasal congestion (Stafford-Smith et al., 2007) and other indications such as migraine (Willems et al., 2003), chronic heart failure (Brede et al., 2002), neuropathic pain (Graham et al., 2000), and schizophrenia (Sallinen et al., 1998). However, because of the lack of ligands that recognize only the α2C-adrenoceptor subtype, studies on the distribution and the physiological relevance of this subtype and their therapeutics have been limited.
The most commonly used nasal decongestant drugs are the α-adrenoceptor agonist sympathomimetic agents, which are effective because of their vasoconstrictor action (Empey and Medder, 1981), which opposes mucosal engorgement in the highly vascular nasal mucosa. Decongestants activate postjunctional α-adrenoceptors on precapillary (resistance) and postcapillary (capacitance) blood vessels in the nasal mucosa (Johnson and Hricik, 1993) and cause vasoconstriction. Such vasoconstriction decreases blood flow through the nasal mucosa and results in shrinkage of this tissue and therefore reduces nasal congestion of the mucosa. However, as a consequence of intrinsic nonselectivity of α-adrenoceptor agonists (phenylpropanolamine, oxymetazoline) or targeting of the α1-adrenoceptor only (phenylephrine), such agents can produce significant side effects such as hypertension, insomnia, anorexia, and/or rebound congestion (rhinitis medicamentosa).
The pathophysiology of nasal congestion is predominantly associated with swelling of the venous sinusoids and collecting veins, and venous constriction of these vessels leads to nasal decongestion (Stafford-Smith et al., 2007). An α2-adrenoceptor agonist, by preferentially constricting venous capacitance blood vessels over arterial resistance blood vessels in nasal mucosa (Corboz et al., 2005, 2008), should thus avoid the hypertensive liability of α1-adrenoceptor agonists. It is known that α2-adrenoceptor agonists produce decongestion in humans, and in the 1960s, the α2-adrenoceptor agonist clonidine was initially developed for the treatment of nasal congestion (Stahle, 2000), as was Tinazoline (Nagarajan et al., 1981), and more recently, an α2-adrenoceptor agonist demonstrated significant relief of nasal congestion in human subjects with allergic rhinitis (Berkowitz et al., 2005). However, adverse events such as hypotension, bradycardia, somnolence, nausea, and dry mouth, likely caused by activation of the central α2A-adrenoceptor subtype, were reported. Nonetheless, findings such as localization of α2C-adrenoceptors in venous sinusoids (Stafford-Smith et al., 2007) and the lack of hemodynamic effects of α2C-adrenoceptors (Link et al., 1996) suggest that vascular α2C-adrenoceptors represent an attractive target for nasal congestion. First, in situ hybridization studies showed that in the periphery α2C-adrenoceptors are abundantly expressed in the nasal venous sinusoids and arteriovenous anastamoses (Stafford-Smith et al., 2007). Second, it was demonstrated through the use of knockout mice that sympathetic vascular tone seems predominantly regulated by the α2A- and α2B-adrenoceptor subtypes, but not by the α2C-adrenoceptor subtype (MacDonald et al., 1997). Therefore, taken together, these findings support the hypothesis that selective α2C-adrenoceptor agonists will have the potential to display nasal decongestant activity without hypertensive liabilities associated with standard α-adrenoceptor decongestants.
Therefore, to avoid the side effects associated with α1-adrenoceptors or nonselective α2-adrenoceptor agonists, we targeted the development of an α2C-subtype-selective adrenoceptor agonist. It is anticipated that the development of a selective α2C-adrenoceptor agonist will represent a significant improvement in the treatment of nasal congestion because of its selective venoconstrictor action, minimal CNS and systemic cardiovascular side-effect profiles, and no potential abuse liability inherent with pseudoephedrine and its conversion to methamphetamine. In the present study, we characterized the preclinical in vitro and in vivo pharmacological and pharmacokinetic activities of compound A [N-[3,4-dihydro-4-(1H-imidazol-4-ylmethyl)-2H-1,4-benzoxazin-6-yl]-N-ethyl-N′-methylurea] (McCormick et al., 2007), a novel, selective, and potent α2C-adrenoceptor agonist for the treatment of nasal congestion with potentially improved side-effect profiles over current marketed nasal decongestants (pseudopehedrine and phenylephrine).
Materials and Methods
Compound A was supplied by the Department of Chemistry, Merck Research Laboratories (Kenilworth, NJ) (Fig. 1).
Chinese hamster ovary (CHO)-K1 cell lines stably expressing the human recombinant human α2A-, α2B-, and α2C-adrenoceptors were purchased from Euroscreen (Brussels, Belgium). The Swiss-Prot accession numbers for the recombinant human α2A- and α2C-adrenoceptor cDNA clones used for the cell transfections are P08913 and P18825, respectively. The GenBank accession number for the recombinant human α2B-adrenoceptor cDNA clone used for cell transfection is M34041. [3H]5-bromo-6-[2-imidazolin-2-yl-amino]-quinoxaline (UK14304) (74.7 Ci/mmol), [35S]GTPγS (1259 Ci/mmol), and Basic FlashPlate were purchased from PerkinElmer Life and Analytical Sciences (Waltham, MA). GF/C unifilter plates and Microscint 20 were purchased from PerkinElmer Life and Analytical Sciences. Bovine serum albumin (BSA), EDTA, EGTA, GTPγS, GTP, MgCl2, and Tris-HCl were purchased from Sigma-Aldrich (St. Louis, MO). Ham's F12 medium, penicillin, streptomycin, and Hanks' balanced salt solution (HBSS) were purchased from Invitrogen (Carlsbad, CA). Fetal bovine serum (FBS) was obtained from Summit Biotechnology (Fort Collins, CO), and G418 [(2R,3S, 4R,5R,6S)-5-amino-6-[(1R,2S,3S,4R,6S)-4,6-diamino-3-[(2R, 3R,4R,5R)-3,5-dihydroxy-5-methyl-4-methylaminooxan-2-yl]oxy-2-hydroxycyclohexyl]oxy-2-(1-hydroxyethyl)oxane-3,4-diol] was from Gemini Bioproducts (Calabasas, CA).
The α2-adrenoceptor agonist clonidine, the α1-adrenoceptor agonist phenylephrine, the mixed α1/α2-adrenoceptor agonists norepinephrine and oxymetazoline, the α2-adrenoceptor antagonist yohimbine, and the α1-adrenoceptor antagonist prazosin were obtained from Sigma/RBI (Natick, MA), and the α2-adrenoceptor agonist medetomidine was synthesized by the Chemical Research Department, Merck Research Laboratories. The P-glycoprotein (P-gp) transport inhibitor Elacridar was purchased from Research Chemicals Incorporation (Toronto, ON, Canada). Compound A, clonidine, phenylephrine, norepinephrine, oxymetazoline, yohimbine, and prazosin were prepared as concentrated stocks in deionized water before dilution to final concentration in the human saphenous vein functional bioassay buffer and in the pig nasal mucosa bioassay buffer. Compound A, clonidine, and phenylephrine were prepared with 0.4% methyl cellulose and dosed orally with a volume of 5 ml/kg in the in vivo rat assay. Compound A and metedomidine were prepared with 20% hydroxy-propyl β-cyclodextrin (HPBCD) and dosed orally with a volume of 10 ml/kg in the in vivo mouse assay.
Animal Care and Use
These studies were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act in a program accredited by the Association for Assessment and Accreditation of Laboratory Animal Care.
Receptor Binding Studies.
CHO-K1 cells stably expressing the recombinant human α2A-, α2B-, α2C-, α1A-, and α1B-adrenoceptors (Euroscreen) and l-fibroblasts for α1D-adrenoceptors were grown in complete Ham's F12 media containing 10% FBS, 50 IU/ml penicillin, 50 μg/ml streptomycin, and 400 μg/ml G418 antibiotic solution. Membranes from each cell type were prepared by homogenizing the cells in buffer containing 15 mM Tris-HCl, pH 7.5, 2 mM MgCl2, 0.3 mM EDTA, and 1 mM EGTA, followed by two consecutive centrifugation steps at 40,000g for 25 min separated by a wash in the same buffer. Membranes were resuspended in buffer containing 7.5 mM Tris-HCl, pH 7.5, 12.5 mM MgCl2, 0.3 mM EDTA, 1 mM EGTA, and 250 mM sucrose. Protein was quantitated using the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA).
To assay α2-adrenoceptor binding, 25 μl of [3H]UK14304 (1–2.5 nM depending on the Kd of the receptor subtype) was added to wells containing binding buffer (50 mM Tris-HCl, pH 7, 1 mM EDTA, 12.5 mM MgCl2, 0.2% BSA, and either 0.1% dimethyl sulfoxide or various concentrations of test compounds). Nonspecific binding was defined in the presence of α2-adrenoceptor antagonist yohimbine (10 μM). For α1-receptor subtype, 25 μl of [3H]prazosin (0.3–1 nM depending on subtype) was the ligand, and nonspecific binding was defined in the presence of the α1-adrenoceptor agonist phentolamine (10 μM). Binding was initiated by the addition of 150 μl of membrane (10–20 μg). After 1 h at room temperature, the reaction was stopped by rapid filtration through Whatman (Clifton, NJ) GF/C unifilter plates that had been presoaked with 0.3% polyethylenimine. Filtration was accomplished using a Packard FilterMate Harvester (PerkinElmer Life and Analytical Sciences), and filters were washed five times with 0.5 ml of cold 50 mM Tris-HCl, pH 7.4. After drying, bound radioactivity was determined by liquid scintillation counting (Packard Top Count; PerkinElmer Life and Analytical Sciences) with Microscint 20 at 50 μl/well. IC50 values (concentration of test compound that inhibits 50% of the α-adrenoceptor agonist response) were determined from competition binding curves, and Ki values were determined according to Cheng and Prusoff (1973) using the program Prism (GraphPad Software Inc., San Diego, CA).
Agonist-Induced [35S]GTPγS Functional Assay.
Membranes from CHO-K1 cells expressing the α2A-, a2B-, and a2C-adrenoceptors were prepared as described above. The final [35S]GTPγS assay buffer was identical to the binding buffer described that lacked NaCl but included 1 μM GDP. Each reaction was set up in quadruplate wells by adding the reagents in the following order to Basic FlashPlate microplates: membranes (20 μg of protein/well in 160-μl assay buffer), 20-μl serial dilutions of compound A or 1 μM cold GTPγS (nonspecific binding), and 20 μl of 0.1 nM [35S]GTPγS for a total volume of 200 μl per well. After 30 min at room temperature, including 2 min of slow shaking on a titer plate shaker, the plates were centrifuged for 5 min at 2500 rpm at 4°C in a tabletop Sorvall centrifuge (Thermo Fisher Scientific, Waltham, MA) and counted immediately with Packard Top-Count. The percentage increase over basal binding of [35S]GTPγS was calculated as follows: 100 × [[mean total sample cpm − mean basal cpm]/mean basal cpm]. Basal cpm was defined as the mean cpm in the absence of compound A minus the mean nonspecific binding cpm. Half-maximal effective concentrations, EC50 (the concentration of compound A required to give 50% of its own maximal stimulation) were calculated using nonlinear regression with Prism. The maximal increase over basal binding of [35S]GTPγS (Emax) achieved for each drug is expressed as a percentage of the maximal UK14304 response tested in the same experiment.
FLIPR Assay (Calcium Flux).
α1A, α1B, and α1D cells were harvested using 5 mM EDTA prepared in Dulbecco's phosphate-buffered saline without Ca2+ and Mg2+ for FLIPR assay. Cells were cultured at 15,000 cells/well overnight in 96-well black wall with clear-bottom plates coated with 50 μl of 1 mg/ml solution of poly-d-lysine. Cells were loaded using Ca2+ plus assay kit from Molecular Devices (Sunnyvale, CA). Dye from assay kit was dissolved in HBSS with Ca2+ and Mg2+ with 20 mM HEPES and 2.5 mM probenecid. Cell loading buffer for FLIPR assay was HBSS (with Ca2+ and Mg2+), 20 mM HEPES, and 2.5 mM probenicid. Compound was prepared at 6× in HBSS (with Ca2+ and Mg2+), 20 mM HEPES, 2.5 mM probenicid, and 0.5% BSA. Medium was removed from cells cultured overnight with 100 μl of dye prepared in loading buffer and incubated for 1 h at 37°C. The plate was then transferred to FLIPR assay, and compound was added to the cells in 20 μl at a speed of 15 μl/s. Data were captured every second for the first 60 s followed by 2 s for 60 s. Data were analyzed using Prism.
CHO-K1 cells expressing the α2A-, α2B-, and α2C-adrenoceptors were grown according to the protocol provided by Euroscreen, with the exception that fungizone was not included in the culture medium. Cells were cultured as a monolayer in tissue culture flasks in complete Ham's F12 culture medium at 37°C with 5% αCO2 and were recultured every 2 to 3 days; cells were harvested from the culture flasks by using Dulbecco's phosphate-buffered saline (without Ca/Mg2+) containing 5 mM EDTA.
CHO-K1 cells expressing the α1A- and α1B-adrenoceptors were maintained in Dulbecco's modified Eagle medium high glucose, 10% FBS, 5000 IU penicillin, and 5000 μg/ml streptomycin and l-glutamine as a monolayer in tissue culture flasks at 37°C with 5% CO2, and cells were recultured every 2 to 3 days. CHO-K1 cells expressing the α1A-adrenoceptors were maintained in F12K medium with G418 (400 μg/ml).
Human Saphenous Vein Assay
Human saphenous veins were obtained from coronary artery bypass graft surgery patients and were procured from the Morristown Memorial Hospital (Morristown, NJ) and the Hackensack University Medical Center Institute for Biomedical Research (Hackensack, NJ). Human saphenous veins were received at the Merck Research Laboratory 24 to 72 h after removal. On the day of arrival, human saphenous veins were either used fresh or cryopreserved for use at a later date.
Functional α-Adrenoceptor Bioassay.
The human saphenous vein functional assay has been described previously (Rizzo et al., 2001). In brief, stainless-steel tissue hooks and 2-0 silk were used to anchor the human saphenous veins ring segments in 25.0-ml organ baths (Q-bath; Radnoti Glass Technology, Inc., Monrovia, CA) and attach them to Grass FT-03 force transducers (Astro-Med, West Warick, RI). Isometric tension was continuously recorded using a model K2G physiograph (Astro-Med). The organ baths were filled with a pH 7.4 Krebs-style buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO4, 1.2 mM KH2PO4, 24.9 mM NaHCO3, 11.1 mM glucose, and 2.55 mM CaCl2) maintained at 37°C and continuously aerated with 95% O2 and 5% CO2 gas. Rings were placed under 1.0-g initial resting tension and equilibrated for 1 h. Tissues were tested for responsiveness with the nonselective α-adrenoceptor agonist norepinephrine (100 μM) and washed three times during the equilibration period. During the experimental portions of the procedure, contractions to rising cumulative concentrations of the selective α2C-adrenoceptor agonist compound A were observed in the absence or presence of the α1-adrenoceptor antagonist prazosin and in the absence or presence of the α2-adrenoceptor antagonist yohimbine. Antagonist equilibration time was 1 h before the agonist challenge. Upon completion of the experiment, a KCl (80 mM)-induced contraction was performed.
Additional studies were performed to determine whether compound A undergoes desensitization after repeated challenge. Four successive administrations of compound A were tested in the human saphenous vein assay at two high concentrations (3 and 10 μM).
Blood Vessel Constriction Measurement in Pig Nasal Mucosa Explants
Blood vessel constriction in porcine nasal mucosa explants was measured as described previously (Lieber et al., 2010). Porcine nasal turbinates from domestic pigs (male and female, 110–250 kg) were provided by a local abattoir, Animal Parts (Scotch Plains, NJ). Nasal mucosa was removed from turbinates and cut into strips (0.5 × 1.5 cm). Mucosal strips were fixed in 6% agarose (Thermo Fisher Scientific) in Krebs buffer containing 118 mM NaCl, 2.55 mM CaCl2, 4.7 mM KCl, 1.2 mM MgSO4, 1.2 mM KH2PO4, 24.9 mM NaHCO3, and 11.1 mM glucose, pH 7.4 at 37°C. Tissues in the agarose were then cooled at 4°C for approximately 20 min (until agarose was firm). Once firm, the agarose-fixed mucosa tissues were cut into slices of 100 to 300 μm in Krebs buffer at 4°C, using the Krumdieck Tissue Slicer (Alabama Research and Development, Munford, AL). Tissue slices were isolated from the surrounding agarose and cultured overnight in six-well tissue culture dishes in Clonetics SmGM2 medium (Lonza Walkersville, Inc., Walkersville, MD) containing 1% penicillin/streptomycin (Lonza Walkersville, Inc.) at 37°C in humidified air containing 5% CO2.
The next day, nasal mucosa slices were equilibrated for 15 min at 37°C in Krebs buffer before recording. Blood vessel constriction in response to compound A was evaluated. Cross-sectional images of arteries and veins were recorded before and after addition of cumulative concentrations of compound A (0.01–100 μM) under a Zeiss Axiovert 100 microscope (Carl Zeiss Inc., Thornwood, NY). At the end of the experiments, maximal blood vessel constriction in response to the nonselective α-adrenoceptor agonist norepinephrine (1 mM) was recorded. In another set of experiments, blood vessel constriction in response to the standard decongestant oxymetazoline (100 μM) was recorded. Blood vessel constriction was expressed as percentage of area decrease from baseline. Cross-section areas of veins or arteries were measured using the computer software ImageJ (National Institutes of Health, Bethesda, MD).
In Vivo Assessment of Blood Pressure and Mydriasis Activities in the Conscious Rat
Male Sprague-Dawley rats (300–325 g) equipped with carotid cannulas for blood pressure measurements were used in this study. The surgical procedure (cannulation of the carotid artery) was performed at the Charles River Laboratory (Raleigh, NC). Fasted and unanesthetized animals were orally dosed either with a single bolus of the α1-adrenoceptor agonist phenylephrine (3–20 mg/kg), the α2-adrenoceptor agonist clonidine (0.1–3 mg/kg), the selective α2C-adrenoceptor agonist compound A (30 mg/kg), or the vehicle 0.4% methyl cellulose in a total volume of 5 ml/kg body weight. Measurement of pupil diameter and blood pressure were taken before oral administration of the bolus, and then every half hour the first hour and every hour the next 3 h. Rats were placed in the experimental conditioning unit (not a restrainer) for 4 h. Animals were conditioned to periods of confinement in the unit by introducing them to the unit the day before experiment. The day of the experiment, animals were taken from the unit only for measurement of pupil diameter performed with a ruler (0.1-mm graduation) and a binocular microscope. All observations were done under green light at a constant intensity to minimize the light reflex pupil constriction and add contrast to the iris (Gherezghiher and Koss, 1979). Blood pressure was continuously monitored via the indwelling carotid artery catheter. A terminal measurement was taken at 4 h after dosing.
Mouse Locomotor Activity and Body Temperature
Locomotor Activity Measurement.
Studies were conducted in male C57BL/6 mice (Charles River Laboratories, Inc., Wilmington, MA). After oral administration of the vehicle (20% HPBCD) in distilled water, the selective α2C-adrenoceptor agonist compound A (3–100 mg/kg), or the α2-adrenoceptor agonist medetomidine (0.3 mg/kg), mice were placed into one of eight clear Plexiglas locomotor activity (LMA) chambers (25 × 25 × 40 cm; length × width × height) placed within eight activity monitor stations, and locomotor activity was assessed for 60 min using an automated photobeam system (Tru Scan system; Coulbourn Instruments, Allentown, PA). Total locomotor distance, an indicator of ambulatory and exploratory activities, was the dependent measure. Horizontal (distance traveled) and vertical (number of rears) activities were recorded during the 60-min test.
Core Body Temperature Measurement.
Core body temperature (°C) was recorded with a thermometer probe inserted in the rectum. The body temperature was recorded twice, immediately before and 60 min after the vehicle (20% HPBCD) or drug administrations, i.e., immediately after the locomotor activity test.
Pharmacokinetic Profile in the Rat and Evaluation of the P-gp Inhibitor Elacridar (GF120918)
Male Sprague-Dawley rats (n = 3) were dosed orally and intravenously at 5 mg/kg using a 20% (w/v) HPBCD suspension of the selective α2C-adrenoceptor agonist compound A as the micronized crystalline dihydrochloride salt. Blood samples were collected into tubes containing heparin at 0.25, 0.5, 1, 2, 4, 6, 8, and 24 h after dose. Blood samples were chilled on ice and the plasma was separated, isolated, and placed into sample tubes. A brain/plasma study was conducted at 10 mg/kg where the animals were sacrificed at various time points and both brain and blood samples were collected. A brain P-gp inhibition study was conducted by dosing compound A (3 mg/kg) alone or in combination with the P-gp transport inhibitor Elacridar (N-(4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-phenyl)-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide; GF120918) (5 mg/kg) (Toronto Research Chemicals Inc., North York, ON, Canada) intravenously to two rats. Two hours after the intravenous administration, brain and plasma samples were obtained and analyzed for compound A. All samples were stored at −20°C until assayed by high-performance liquid chromatography (HPLC)-atmospheric pressure ionization/tandem mass spectrometry (MS/MS). The limit of quantitation for brain samples was 20 ng/g, and the limit of quantitation for plasma samples was 5 ng/ml.
Sample Preparation for Measuring Compound A in Plasma.
LC-MS/MS assays were used to analyze compound A. In a typical run, a standard curve of compound A was generated in heparanized plasma by spiking aliquots with the appropriate compound A stock solution to give final concentrations within the range of 1 to 5000 ng/ml plasma. Samples and standards were prepared for analysis by removing a 10-μl aliquot of each plasma sample and precipitating the proteins with 60 μl of 100% acetonitrile containing an internal standard. The samples were vortexed and centrifuged. Approximately 50 μl of the supernatant was removed and placed in a 96-well plate for HPLC-atmospheric pressure ionization/MS/MS analysis. Duplicate calibration standard sets were included: one set was placed before and the second after the test samples. The same procedure was used to assay brain samples after they were homogenized with water (1:4: brain/water).
LC-MS/MS Analysis of Plasma Samples.
Typical sample analysis was conducted using a Shimadzu HPLC system consisting of a solvent delivery system (Shimadzu USA Manufacturing Inc., Columbia, MS) coupled with a Quantum mass spectrometer (Thermo Fisher Scientific). Analyte separation was typically achieved by using a Capcell Pak (Phenomenex Inc., Torrance, CA) C18 2.0 × 35 mm, 5-μm particle size analytical column. A gradient mobile phase consisting of solvent A (20:80 methanol/water containing 0.01 M ammonium acetate) and solvent B (0.01 M ammonium acetate in methanol containing 0.6 ml/l of 10% v/v acetic acid) was used. For a typical sample analysis, the total run time for each sample was approximately 1.7 min; compound A and an internal standard eluted at 0.8 and 0.7 min, respectively. Compound A and the internal standard were monitored in the positive ion electron spin ionization/MS/MS mode with a Quantum (San Jose, CA) mass spectrometer. The MS/MS fragmentations selected to monitor compound A and the internal standard were the transitions from m/z 316 to 236 and from m/z 504 to 306, respectively. Quantification was performed using Excalibur 1.4 (Quantum) based on duplicate standard curves. For each analytical run accepted, the accuracy of at least three-fourths of the individual calibration standards was within ± 27.5% of the respective nominal values, and the correlation between response and concentrations (r2) was ≥0.94. In addition, at least two-thirds of the quality-control samples fell within ± 25% of their respective nominal values.
All results are expressed as S.E.M. For the in vitro studies, segments from the same saphenous vein were used to study the effects of only one treatment, whereas arteries and veins from the same pig nasal mucosa received similar or different treatments. Number of observations (n) indicates 1) the number of segments, from at least three human saphenous veins and 2) the number of arteries and veins from five pigs in each group, except the oxymetazoline group (three animals). For the in vivo studies, each animal was used to study the effects of only one treatment with 1) at least five animals in the blood pressure and mydriasis studies, except for experiments with clonidine treatments (three to four animals) and 2) 12 animals in the LMA and hypothermia studies.
The Kruskal-Wallis nonparametric test (multiple-group comparison) was used for the comparison between vehicle- and drug-treated tissues or animals. When the overall comparison showed a significant difference between groups, the Kruskal Wallis test was followed by the Mann-Whitney U test (two-group comparisons) for pairwise comparison. Student's t test was used to compare pig arteries and veins responses to compound A in the real-time tissue contractility assay, and one-way analysis of variance with post hoc Dunnett's multiple comparison test was performed for the mouse locomotor activity and body temperature experiments. Probability (P) < 0.05 was accepted as the level of statistical significance.
In Vitro Pharmacology Profile
Binding/Cell-Based Activity Profile.
The affinity of compound A for human α2C-adrenoceptors was determined by competition binding analysis with the radioligand agonist [3H]UK14304 in membranes prepared from CHO cells expressing the human wild-type α2C-adrenoceptor, and the mean affinity constant (Ki) of compound A was 12 nM (Table 1 and Fig. 2A). Compound A stimulated [35S]GTPγS binding to the human wild-type α2C-adrenoceptor expressed in the CHO cells, and the mean functional EC50 in the GTPγS assay was 166 nM with a mean efficacy of 64% when expressed as Emax, relative to the full agonist UK14304 defined as 100% (Table 1 and Fig. 2B). Indeed, agonist-induced α2-adrenoceptor coupling to heterotrimeric GTP-binding protein (G proteins) promotes the release of GDP from the α-subunit of G proteins followed by the binding and subsequent hydrolysis of GTP. This G-protein activation can be measured in the membranes of cells expressing α2-adrenoceptors by the binding of [35S]GTPγS, a radiolabeled poorly hydrolysable GTP analog (Umland et al., 2001).
The affinity and activity of compound A were evaluated at the related α2A- and α2B-adrenoceptors, using membranes prepared from CHO cells expressing each of these receptors. The mean Ki values of compound A for α2A- and α2B-adrenoceptor were 2284 and 3076 nM, respectively (Table 1), representing 190- and 256-fold lower affinity than at the α2C-adrenoceptor. The mean functional EC50 and efficacy Emax of compound A were 1525 nM and 8%, respectively, at α2A-adrenoceptor and 5814 nM and 21%, respectively, at α2B-adrenoceptor. By comparison, the affinities and potencies for the nonselective α2-adrenoceptor agonist UK14304 were: Ki = 2.2 nM and EC50 = 41 nM for the α2C-adrenoceptor subtype (versus 12 and 167 nM for compound A), Ki = 0.3 nM and EC50 = 1.0 nM for the α2A-adrenoceptor subtype (versus 2.3 and 1.5 μM for compound A), and Ki = 12 nM and EC50 = 112 nM for the α2B-adrenoceptor subtype (versus 3.1 and 5.8 μM for compound A). Affinities and potencies for both compounds were in the nanomolar range for the α2C-adrenoceptor subtype. However, affinities and potencies for UK14304 were in the nanomolar range for the α2A- and α2B-adrenoceptor subtypes, whereas they were in the micromolar range for compound A. In addition, compound A showed good efficacy at the α2C-adrenoceptor subtype (64%) and marginal efficacy at the α2A- and α2B-adrenoceptor subtypes (8 and 21%, respectively). Compound A was also evaluated at the related human α1A-, α1B-, and α1D-adrenoceptors in CHO cells (α1A, α1B) and l-fibroblasts (α1D). The α1-adrenoceptor affinity of compound A was determined by competition binding analysis with the radioligand antagonist [3H]prazosin. Functional activity was assessed by the ability of the compound to stimulate intracellular calcium. Compound A demonstrated negligible to no affinity and/or efficacy at the α1A, α1B, and α1D subtypes (Table 1). Thus, compound A had a high binding affinity, a good potency, and a high selectivity for the α2C-adrenoceptor subtype.
To further assess the selectivity of compound A, the compound was tested against a panel of 69 additional G protein-coupled receptors, enzymes, and other biologically relevant proteins [adenosine A2B (h), A3 (h), adrenergic β3 (h), angiotensin AT1 (h), AT2 (h), benzodiazepine (peripheral), bombesin BB (nonselective), bradykinin B2 (h), calcitonin gene-related peptide (h), cholecystokinin CCKA (h) (CCK1) complement C5a (h), corticotrophin releasing factor CRF1 (h), dopamine D3 (h), D5 (h), endothelin ETA (h), ETB (h), N-formyl-l-methionyl-l-leucyl-l-phenylalanine, GABA (nonselective), GABAB (1b) (h), galanin GAL1 (h), glucagon (h), N-methyl-d-aspartate, glycine (strychnine sensitive), epidermal growth factor (h), platelet-derived growth factor, tumor necrosis factor-α (h), histamine H1 (h), H2 (h), leukotriene B4 (h) (BLT1), leukotriene D4 (h) (CysLT1), melanocortin MC4 (h) melatonin, ML1, monoamine oxidase-A, monoamine oxidase-B, motilin (h), neurotensin NT1 (h) (NTS1), N (neuronal) (α-BCTX sensitive), pituitary adenylyl cyclase-activating protein (PAC1) (h), platelet-activating factor (h), thromboxane A2/prostaglandin H2 (h) (TP), purinergic P2X, 5-hydroxytryptamine (5-HT) 2A (h), 5-HT2B (h), 5-HT2C (h), 5-HT3 (h), 5-HT5A (5-ht5A), 5-HT6 (h), 5-HT7 (h), somatostatin sst (nonselective), glucocorticoid (h), estrogen (h) (estrogen receptor), progesterone (h) (progesterone receptor), androgen (h) (AR), thyrotropin-releasing hormone, vasoactive intestinal peptide VIP1 (h) (VPAC1), Ca2+ channel (L, DHP site), Ca2+ channel (L, diltiazem site), Ca2+ channel (L, verapamil site), Ca2+ channel (N), K+ ATP channel, K+ V channel, SK+ Ca channel, Na+ channel (site 2), Cl− channel, norepinephrine transporter (h), dopamine transporter (h), GABA transporter, choline transporter (h), 5-HT transporter (h)] at Cerep (Celle L'Evescault, France). In addition, compound A was tested against 30 targets in the human G protein-coupled receptor counterscreen (adenosine A1, A2A, cannabinoid CB1, CB2, chemokine CXCR3, dopamine D1, D2, histamine H1, H3, Mas-related G-protein coupled receptors MRGX1, muscarinic M1 M2, M3, M4, M5, neurokinin NK2, opioid κ, δ, μ, nociceptin NOP1, neuropeptide Y5, protein kinase R PKR1, PKR2, 5-hydroxytryptamine 5HT-1A, proteinase PAR1, vasopressin V1a, V1b, V2, oxytocin, and vanilloid receptor 1 VR1N) at Merck Research Laboratory. No significant activity was observed in these assays up to a concentration of 10 μM (data not shown).
Human Saphenous Vein.
The postjunctional α2C-adrenoceptor-mediated contractility was characterized in human saphenous vein. We showed previously that the α2C-adrenoceptor is preferentially expressed at the mRNA level (Umland et al., 2001) and is the predominant postjunctional α2-adrenoceptor subtype in this preparation (Rizzo et al., 2001). Compound A produced a concentration-dependent contraction (pD2 = 6.28 ± 0.05) that was significantly blocked by the α2-adrenoceptor antagonist yohimbine (100 nM) but not by the α1-adrenoceptor antagonist prazosin (30 nM) (Fig. 3). The concentrations of 100 nM for yohimbine and 30 nM for prazosin were chosen from the study of Rizzo et al. (2001) using the same preparation. Those authors demonstrated that the human saphenous vein contractions to the α2-adrenoceptor agonist BHT-920 (5,6,7,8-tetrahydro-6-(2-propen-1-yl)-4H-thiazolo[4,5-d]azepin-2-amine dihydrochloride) were competitively blocked by yohimbine (pA2 = 8.7) but not by prazosin (30 nM). In contrast, prazosin blocked the α1-adrenoceptor agonist phenylephrine-induced contractions (pKb = 7.9).
Four successive administrations of compound A, at two different concentrations, did not promote desensitization in the human saphenous vein, a smooth muscle bioassay with a prominent postjunctional α2C-adrenoceptor population (Rizzo et al., 2001). The contractions induced by compound A were 0.013 ± 0.006, 0.023 ± 0.005, 0.052 ± 0.016, and 0.042 ± 0.017 g at 3 μM and 0.040 ± 0.020, 0.039 ± 0.014, 0.072 ± 0.026, and 0.055 ± 0.024 g at 10 μM.
Differential Contractility in Arteries and Veins of Pig Nasal Mucosa.
The effects of compound A on nasal mucosa veins (capacitance vessels) and arteries (resistance vessels) were evaluated independently in pig nasal mucosa explants. Real-time cross-section pictures of vessels were taken before and 20 min after incubation with compound A. Contraction was evaluated as the percentage of decrease of the cross-sectional luminal area. Compound A (10 nM to 0.1 mM) contracted veins in a concentration-dependent manner, with an EC50, defined as the concentration causing 50% of the maximal response to the nonselective α-adrenoceptor agonist norepinephrine, of 108 nM (n = 19). Compound A had a much weaker contractile effect on nasal mucosal arteries at high concentrations (n = 12, EC50 >10 μM; Fig. 4), and the magnitude of arteriolar contraction reached only 50% of the maximum venular responses. These results suggest that compound A preferentially constricted capacitance vessels, the vascular component predominantly involved in nasal congestion. In contrast, the positive comparator oxymetazoline (100 μM) did not show a differential contractile response in arteries and veins (Fig. 4). In all of the tissues tested, the nonselective α-adrenoceptor agonist norepinephrine (1 mM) produced maximal contractility of both arteries and veins.
In Vivo Pharmacology Profile
In Vivo Assessment of Blood Pressure and Mydriasis Activity in the Conscious Rat.
Blood pressure effect.
Compound A, administered orally at 30 mg/kg in rat, produced a small decrease in blood pressure compared with the control group (0.4% methylcellulose) (Fig. 5A), whereas the α1-adrenoceptor agonist phenylephrine produced maximum hypertensive effects (24 and 53 mm Hg) at 10 and 20 mg/kg, respectively (Figs. 5B and 6A) 30 min after administration.
The plasma concentrations were 19.3 μM for compound A at 30 mg/kg and 0.3 and 0.7 μM for phenylephrine at 10 and 20 mg/kg, respectively (Fig. 6B). Although the plasma concentration of phenylephrine was 28- and 64-fold less than the compound A concentration, significant hypertensive effects were observed with phenylephrine (Figs. 5B and 6A).
Compound A (30 mg/kg) had no effect on mydriasis (pupil dilation) (Fig. 7A). In contrast, a significant increase in pupil diameter was observed with the α2-adrenoceptor agonist clonidine (0.3–3 mg/kg) (Fig. 7B). Clonidine at 3 mg/kg produced significant mydriasis effect at all time points. The α1-adrenoceptor agonist phenylephrine did not affect mydriasis (data not shown).
Effect of Compound A on Mouse Locomotor Activity and Body Temperature.
Compound A was tested in a mouse LMA assay to determine any effect on motor behavior with emphasis on sedative-like effects. Compound A had no effect up to a dose of 100 mg/kg on locomotor activity, measured as the total distance traveled during the test or vertical exploration measured as the number of rears (Fig. 8). In comparison, the α2-adrenoceptor agonist medetomidine at a dose of 0.3 mg/kg greatly reduced locomotor activity. Furthermore, core body temperature was measured to determine whether compound A produced hypothermic effects. Compound A had no effect on body temperature when administered up to a dose of 100 mg/kg (Fig. 9). In contrast, medetomidine at 0.3 mg/kg caused profound hypothermia. It is noteworthy that a dose of 100 mg/kg did not produce any effect on locomotor activity and body temperature although the plasma concentration was 75 μM at 1 h and 47 μM at 2 h after the administration of compound A.
Pharmacokinetic Profile in the Rat
The plasma concentration versus time data for compound A was subjected to pharmacokinetic analysis by noncompartmental methods. For each animal, the following pharmacokinetic parameters were determined: maximum plasma concentration (Cmax), time of maximum plasma concentration (Tmax), and area under the plasma concentration versus time curve. The mean Cmax, Tmax, and area under the curve values were 0.8 μM, 0.7 h, and 2 μM · h, respectively after a 5 mg/kg oral dose. In addition, rats showed high plasma concentrations (7.75–7.86 μM) of compound A (10 mg/kg) at 0.25 and 0.5 h after administration, whereas brain concentrations were only 2.2 and 2.5%, respectively, of the plasma concentrations at each of the time points, indicating that compound A has poor brain penetration. The oral bioavailability of compound A is 24%.
The poor brain penetration of compound A suggested that compound A may be a substrate for the P-gp transporter. This was confirmed by using the P-gp inhibitor Elacridar (Polli et al., 1999). In the absence of the P-gp inhibitor, rat brain levels of compound A were below the level of detection (Table 2). Although coadministration of a P-gp inhibitor Elacridar increased plasma concentrations (∼3-fold), brain levels were much more increased (>5-fold), assuming brain levels in control rats were at or below the limit of quantitation (20 ng/g).
In this study, we describe the pharmacological characterization of a novel selective α2C-adrenoceptor agonist, compound A, as a potential nasal decongestant with improved side-effect profiles compared with standard nonselective sympathomimetic decongestants (pseudoephedrine and phenylephrine). This compound has high affinity (Ki = 12 nM), good potency (EC50 = 166 nM), and high selectivity for the human α2C-adrenoceptor over the other α-adrenoceptors subtypes (α2A, α2B, α1A, α1B, and α1D). Affinities and potencies for compound A are in the nanomolar range for the α2C-adrenoceptor subtype and in the micromolar range for the α2A- and α2B-adrenoceptor subtypes, whereas the affinities and potencies for the nonselective α2-adrenoceptor agonist UK14304 are in the nanomolar range for all α2-adrenoceptor subtypes. Compound A also displays functional agonist activity in a human saphenous vein smooth muscle contractility bioassay with a prominent postjunctional α2C-adrenoceptor population (Rizzo et al., 2001) and preferentially constricts the nasal veins in a real-time pig nasal mucosa blood vessel contraction bioassay. Compound A does not affect blood pressure or mydriasis (increase in pupil diameter) in the in vivo rat assay up to 30 mg/kg and has no effects on locomotor activity, sedation, and body temperature in the in vivo mouse model up to 100 mg/kg. Finally, compound A at 5 mg/kg showed good oral bioavailability (24%) and low brain penetration in rat and is a substrate for the efflux P-gp transporter.
Engorgement of the large venous sinusoids and collecting veins is a prominent aspect of the pathophysiology of nasal congestion, one of the symptoms of rhinitis. Indeed, nasal decongestion is triggered by an active constriction of the venous sinusoids (Kristiansen et al., 1993), and a decrease in nasal cavity pressure after α-adrenoceptor agonist administration reflects shrinkage of nasal capacitance vessels (Berridge and Roach, 1986). Several other studies in different species (Lacroix, 1989; Corboz et al., 2003, 2007, 2008; Wang and Lung, 2003), including human (Johannssen et al., 1997; Corboz et al., 2005), indicated a predominance of the α2-adrenenoceptor mechanism in the regulation of capacitance vessels. In addition, α2-adrenoceptor proteins are expressed in human nasal mucosa (Corboz et al., 2005) and levels of α2-adrenoceptor protein are higher than α1-adrenoceptor protein in nasal mucosa of nonallergic and allergic patients (van Megen et al., 1991). However, the α2-adrenenoceptor subtype involved in the regulation of capacitance vessels was not defined because all of these previous pharmacological studies have used either agonists or antagonists with selectivity for the α2-adrenoceptor but with marginal or no subtype selectivity.
Stafford-Smith et al. (2007) demonstrated that the α2C-adrenoceptors were preferentially localized to the venous and sinusoidal vasculature of the nasal mucosa, suggesting that, by specifically targeting this component of the nasal circulation, an α2C-adrenoceptor agonist should reduce venous engorgement, resulting in decongestion. In the present study, we demonstrated that the selective α2C-adrenoceptor agonist compound A preferentially constricted the veins of pig nasal mucosa, e.g., the vascular component responsible for mucosal engorgement and nasal congestion. This last observation confirms and extends those of previous pharmacological studies. Indeed, we reported previously that the α2C-adrenoceptor subtype is preferentially found in pig nasal mucosa (Corboz et al., 2003), whereas Stafford-Smith et al. (2007) demonstrated by in situ hybridization that the α2C-adrenoceptor mRNA is the only α-adrenoceptor mRNA present in human nasal venous sinusoids and arteriovenous anastomoses. Although these pig models (organ chamber and real-time tissue contractility assays) have limitations with regard to translation of results to human, it is worthy of note that pig and human share several similarities such as the anatomy, morphology, physiology, histology, and biochemistry (Hare, 1975; Pound and Houpt, 1978; Corboz et al., 2007) and similar distribution in terms of location and cellular organization (Corboz et al., 2007). Moreover, the pig α2-adrenoceptor subtypes are pharmacologically more related to those of human than to those of rodents (Wikberg-Matsson et al., 1995). This involvement of the α2C-adrenoceptor in regulation of basal nasal patency was confirmed by Mingo et al. (2010).
In addition to acting selectively on large venous sinusoids and collecting veins and to causing decongestion, a selective α2C-adrenoceptor agonist should not affect blood pressure. In normal mice, intravenous injection of the nonselective α2-adrenoceptor agonist dexmedetomidine leads to a biphasic blood pressure response: an initial increase, attributed to the α2B-adrenoceptor subtype, which is gradually reversed to hypotension associated with a severe bradycardia and mediated centrally by the α2A-adrenoceptor subtype (MacDonald et al., 1997). Mice with a deletion of the α2C-adrenoceptor gene showed no differences from wild-type mice in the hypertensive, hypotensive, and bradycardic responses to dexmedetomidine (MacDonald et al., 1997), indicating that the α2C-adrenoceptor subtype is not involved in the control of blood pressure homeostasis. Oral administrations of compound A in the in vivo rat preparation did not cause increase in blood pressure up to 30 mg/kg (with a plasma concentration of 19 μM), whereas phenylephrine produced significant hypertensive effects (24 and 53 mm Hg with plasma concentrations of 0.3 and 0.7 μM at 10 and 20 mg/kg, respectively). Phenylephrine (α1-adrenoceptor agonist) and the standard decongestants, phenylpropanolamine, pseudoephedrine, and oxymetazoline (mixed α1- and α2-adrenoceptor agonists) also caused pronounced increase in blood pressure at doses that attenuated nasal congestion in the cat congestion model (Erickson et al., 2001; McLeod et al., 2001).
α2A-Adrenoceptors are widely distributed throughout the body and are the most prominent α2-adrenoceptor subtype in the CNS. In addition to mediating hypotension, central activation of α2A-adrenoceptors affects sedation, hypothermia (Hunter et al., 1997), and mydriasis (Heal et al., 1995). No such adverse CNS effects were observed with compound A in several rodent models. This could be explained by a combination of low CNS penetration of compound A and its high degree of α2C-adrenoceptor subtype selectivity. Very low brain concentrations (62 ng/g) of compound A were found in rats after oral administration (10 mg/kg) despite having high plasma concentrations (2480 ng/ml or 7.9 μM at 0.5 h), suggesting efflux. The efflux mechanism was determined to be mediated through the P-gp transporter by coadministering compound A with the P-gp transport inhibitor Elacridar (Table 2). In addition, compound A had no effect on locomotor activity, sedation, and body temperature in the mouse at doses up to 100 mg/kg (with plasma concentrations of 75 μM), whereas the α2-adrenoceptor agonist medetomidine (0.3 mg/kg) caused a significant reduction in locomotor activity and a profound sedation and hypothermia. In the in vivo rat assay, compound A (up to 300 mg/kg) had no mydriatic effect (data not shown), whereas the α2-adrenoceptor agonist clonidine produced a dose-dependent increase (0.3–3 mg/kg) in pupil diameter. Taken together, these findings, such as absence of effects on blood pressure, locomotion, sedation, body temperature, and mydriasis, confirmed the low affinity and activity of compound A for the α2A-adrenoceptor subtype observed in binding and functional GTPγs assays (Table 1). Although the low brain concentration of compound A helps minimize adverse CNS effects, the α2C-adrenoceptor subtype selectivity also plays an important role. However, structurally related compounds with α2C-adrenoceptor subtype selectivity and greater brain penetration than compound A also lack adverse CNS effects in our animal models.
The absence of desensitization is an additional positive attribute for the α2C-adrenoceptor subtype. Indeed, α2-adrenoceptor-mediated physiological responses undergo agonist-promoted desensitization in various tissues (Insel and Motulsky, 1988), which involved the α2A- and α2B-adrenoceptor subtypes but not the α2C-adrenoceptor subtype (Kurose and Lefkowitz, 1994). We confirm this absence of desensitization in the human saphenous vein assay, a prominent α2C-adrenoceptor assay, in which four successive administrations of compound A at two high concentrations (3 and 10 μM) did not affect the contractions.
In summary, compound A is a highly selective and peripherally active α2C-adrenoceptor agonist with low CNS penetration. The compound displays desirable in vitro and in vivo pharmacological properties with no adverse cardiovascular and CNS side effects and an acceptable pharmacokinetic profile. A selective α2C-adrenoceptor agonist, by reducing nasal mucosa venous capacitance tone, will decrease mucosa swelling and improve patency of nasal cavity. In addition to a therapeutic use in nasal congestion, compound A provides a powerful tool to explore the various physiological roles of peripheral α2C-adrenoceptors. α2C-Adrenoceptor agonists might be potentially attractive targets for a variety of conditions such as migraine (Willems et al., 2003), chronic heart failure (Brede et al., 2002), neuropathic pain (Graham et al., 2000), and schizophrenia (Sallinen et al., 1998). New selective α2C-adrenoceptor drugs may therefore represent novel therapeutic strategy in the treatment of these diseases.
Participated in research design: Corboz, McCormick, Umland, Jia, McLeod, Varty, Aslanian, Palamanda, Nomeir, Korfmacher, Hunter, and Hey.
Conducted experiments: Rivelli, Wan, Shah, Lieber, Morgan, and Palamanda.
Contributed new reagents or analytic tools: McCormick, Wu, Boyce, and Aslanian.
Performed data analysis: Corboz, Umland, Varty, Jia, and Hey.
Wrote or contributed to the writing of the manuscript: Corboz, McCormick, Umland, Jia, Varty, Palamanda, Korfmacher, and Anthes.
Other: Feng formulated the samples for in vivo studies.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- compound A
- bovine serum albumin
- Chinese hamster ovary
- central nervous system
- fetal bovine serum
- fluorometric imaging plate reader
- Hanks' balanced salt solution
- hydroxy-propyl β-cyclodextrin
- high-performance liquid chromatography
- tandem mass spectrometry
- locomotor activity
- N-(4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-phenyl)-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide
- 5,6,7,8-tetrahydro-6-(2-propen-1-yl)-4H-thiazolo[4,5-d]azepin-2-amine dihydrochloride.
- Received October 26, 2010.
- Accepted January 12, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics