Abstract
Cariprazine {RGH-188; trans-N-[4-[2-[4-(2,3-dichlorophenyl)piperazin-1-yl]ethyl]cyclohexyl]-N′,N′-dimethylurea hydrochloride}, a novel candidate antipsychotic, demonstrated approximately 10-fold higher affinity for human D3 versus human D2L and human D2S receptors (pKi 10.07, 9.16, and 9.31, respectively). It displayed high affinity at human serotonin (5-HT) type 2B receptors (pKi 9.24) with pure antagonism. Cariprazine had lower affinity at human and rat hippocampal 5-HT1A receptors (pKi 8.59 and 8.34, respectively) and demonstrated low intrinsic efficacy. Cariprazine displayed low affinity at human 5-HT2A receptors (pKi 7.73). Moderate or low affinity for histamine H1 and 5-HT2C receptors (pKi 7.63 and 6.87, respectively) suggest cariprazine's reduced propensity for adverse events related to these receptors. Cariprazine demonstrated different functional profiles at dopamine receptors depending on the assay system. It displayed D2 and D3 antagonism in [35S]GTPγS binding assays, but stimulated inositol phosphate (IP) production (pEC50 8.50, Emax 30%) and antagonized (±)-quinpirole-induced IP accumulation (pKb 9.22) in murine cells expressing human D2L receptors. It had partial agonist activity (pEC50 8.58, Emax 71%) by inhibiting cAMP accumulation in Chinese hamster ovary cells expressing human D3 receptors and potently antagonized R(+)-2-dipropylamino-7-hydroxy-1,2,3,4-tetrahydronaphtalene HBr (7-OH-DPAT)-induced suppression of cAMP formation (pKb 9.57). In these functional assays, cariprazine showed similar (D2) or higher (D3) antagonist–partial agonist affinity and greater (3- to 10-fold) D3 versus D2 selectivity compared with aripiprazole. In in vivo turnover and biosynthesis experiments, cariprazine demonstrated D2-related partial agonist and antagonist properties, depending on actual dopaminergic tone. The antagonist–partial agonist properties of cariprazine at D3 and D2 receptors, with very high and preferential affinity to D3 receptors, make it a candidate antipsychotic with a unique pharmacological profile among known antipsychotics.
Dopamine D3 receptors, cloned in the beginning of the 1990s (Sokoloff et al., 1990), are most abundant in the mesolimbic regions (i.e., nucleus accumbens, island of Calleja) where dysregulation of neurotransmission is thought to be associated with psychosis. The discovery that most antipsychotics, in addition to binding to D2 receptors, display reasonably high affinity for D3 receptors, led to the assumption that these receptors may also be responsible for antipsychotic efficacy (Sokoloff et al., 1995). Unfortunately, the selective D3 antagonists developed so far (e.g., SB-277011, S33084) have failed to demonstrate sufficient antipsychotic-like activity in various animal models (Millan et al., 2000; Reavill et al., 2000). However, studies with D3-selective agents found that D3 receptors are very likely associated with locomotor control, cognitive behavior, and drug abuse (Joyce and Millan, 2005; Gyertyán and Sághy, 2008).
Recent human imaging studies found that approximately 60 to 75% occupancy of dopamine D2 receptors is necessary for clinical antipsychotic efficacy (Kapur and Mamo, 2003). Blocking these receptors remains one of the primary targets in schizophrenia pharmacotherapy (Seeman, 2006). However, based on the known functions, properties, and localization of D3 receptors, the hypothesis was developed that the presence of subnanomolar D3 antagonism alongside nanomolar D2 antagonism may yield an antipsychotic compound with a superior side-effect profile (e.g., reduced extrapyramidal symptoms, improvement in cognition). Indeed, representative compounds (e.g., S33138 and RG-15) developed on the basis of this idea demonstrated antipsychotic-like activity and reduced side-effect liability in animal models (Gyertyán et al., 2008; Kiss et al., 2008; Millan et al., 2008).
Several attempts have been made to use partial D2 agonists such as preclamol (Lahti et al., 1998) and terguride (Olbrich and Schanz, 1991) in the treatment of schizophrenia. However, only partial and temporary efficacy has been demonstrated for these compounds, possibly because of their higher than optimal intrinsic activity. To date, however, the optimal intrinsic activity required for a partial D2 agonist to yield the desired antipsychotic efficacy for treating the major symptoms of schizophrenia remains unclear.
Aripiprazole, a D2 partial agonist (Burris et al., 2002), received approval for the treatment of schizophrenia in the United States in 2002 (Keck and McElroy, 2003). Aripiprazole displayed subnanomolar affinity for the human dopamine D2 receptors with relatively low intrinsic activity (Tadori et al., 2009). In addition, aripiprazole showed subnanomolar–nanomolar affinity for several other receptors in vitro with antagonist (e.g., for 5-HT2B, 5-HT2A, H1) or partial agonist profile (e.g., for D3 and 5-HT1A) (Shapiro et al., 2003). Understanding the exact in vitro mechanism of action of aripiprazole was complicated by the finding that, in different signaling pathways, it was a functionally selective compound at D2 receptors (Urban et al., 2007).
Previously, we hypothesized that an effective antipsychotic agent with favorable properties (such as cognition improvement) and a beneficial side-effect profile (e.g., greatly reduced liability to induce catalepsy) would combine subnanomolar affinity for dopamine D3 receptors with nanomolar affinity for D2 receptors (Gyertyán et al., 2008; Kiss et al., 2008). From a series of compounds we selected cariprazine (former code name RGH-188; Fig. 1), a compound demonstrating subnanomolar affinity for D3 receptors and nanomolar affinity for D2 receptors with antagonist–partial agonist activity at both of these dopamine receptor subtypes. In the present study, we have elucidated the in vitro receptor binding and functional profile (i.e., antagonist/agonist properties) of cariprazine at dopamine D3/D3-related signaling pathways in vitro. In addition, we investigated its action on the cerebral dopamine turnover and biosynthesis in mouse brain under various conditions. Throughout these studies, cariprazine was compared with known D3/D2 agonists, antagonists, and first-generation and atypical antipsychotics with special regard to aripiprazole, the only D2/D3 partial agonist/antagonist antipsychotic used in the clinic.
Materials and Methods
Animals
Male NMRI mice (20–25 g), 7- to 8-week-old male (180–220 g) Hanover Wistar rats and Hartley guinea pigs (200–250 g) were obtained from Toxicoop Hungary Ltd. (Budapest, Hungary) and acclimatized at the site for at least 3 to 4 days before any experiments started. They were kept under standard conditions (temperature 21°C; relative humidity 55–65%; 12:12 h dark/light cycle) on commercial laboratory chow and tap water ad libitum. Animal maintenance and research were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All procedures using animals were approved by the local ethics committee and conformed to the rules and principles of the 86/609/EEC Directive.
Drugs
Cariprazine (RGH-188) HCl, aripiprazole HCl, S-(−)-pramipexole HCl, haloperidol, olanzapine, risperidone, L741626, SB-277011, NSD-1015, and reserpine were synthesized at Gedeon Richter Plc. or were the product of the company. Apomorphine (APM) HCl (Chemistry Department, University of Debrecen, Debrecen, Hungary), 7-OH-DPAT, 8-OH-DPAT, DOI, pyrilamine, dopamine (DA), (±)-quinpirole or (+)-quinpirole, γ-butyrolactone (GBL), and butaclamol were purchased from Sigma-Aldrich (St. Louis, MO). SB-204741 was obtained from Tocris Bioscience (Bristol, UK). (+)-PHNO was purchased from Toronto Research Chemicals Inc. (North York, ON, Canada). Radioligands, [3H]spiperone [specific activity (spec. act.), 15–16 Ci/mmol)], [3H]raclopride (spec. act., 60–80 Ci/mmol), [3H]8-hydroxy-2-(di-n-propylamino)tetralin (spec. act., 106 Ci/mmol), and [3H]ketanserine (spec. act., 88 Ci/mmol) were obtained from PerkinElmer Life and Analytical Sciences (Waltham, MA). [35S]GTPγS (spec. act.: 1000–1150 Ci/mmol) and [3H]cAMP (spec. act.: 37.2 Ci/mmol) were purchased from Amersham Radiochemicals (Little Chalfont, Buckinghamshire, UK). myo-[3H]Inositol (spec. act., 30 Ci/mmol) was purchased from American Radiolabeled Chemicals (St. Louis, MO). All other chemicals were of analytical grade and obtained from commercial sources. Cariprazine was used in base form in most experiments; use of its HCl salt is indicated separately. For in vivo experiments, drugs were dissolved in a minimum amount (∼10–20 μl) of acetic acid and further diluted with saline. They were administered orally in a volume of 10 ml/kg.
Radioligand Binding Assays Using Rat Receptors
Rats were decapitated; brains were rapidly removed and chilled in ice-cold saline; and the striatum, hippocampus, and frontal cortex of each rat were dissected on an ice-cooled surface.
Striata were homogenized by Ultra-Turrax (IKA-Werke GmbH and Co. KG, Staufen, Germany) in 50 volumes (w/v) of ice-cold buffer (50 mM Tris-HCl, 1 mM EGTA, and 5 mM MgSO4·7H2O, pH 7.4), and the homogenate was centrifuged at 40,000g for 10 min at 4°C. The pellet was suspended in the same buffer, and the homogenate was centrifuged at 40,000g for 10 min at 4°C. This procedure was repeated once. The final pellet was resuspended in 30 volumes (w/v) of the same buffer and frozen in aliquots at −70°C until use.
Hippocampi were homogenized in 40 volumes (w/v) of ice-cold 50 mM Tris-HCl buffer (pH 7.7), and the homogenate was centrifuged at 45,000g for 10 min at 4°C. The pellet was suspended in the same buffer, and the homogenate was incubated at 37°C for 10 min and centrifuged at 45,000g for 10 min at 4°C. The final pellet was resuspended in 60 volumes (w/v) of binding buffer consisting of 50 mM Tris-HCl/4 mM CaCl2/0.1% ascorbic acid/10 μM pargyline (pH 7.7) and frozen in aliquots at −70°C until use.
Frontal cortices were homogenized by a glass Potter homogenizer with Teflon pestle (B. Braun Melsungen, Melsungen, Germany) with six up-and-down strokes in 40 volumes (w/v) of ice-cold 50 mM Tris-HCl buffer (pH 7.4), and the homogenate was centrifuged at 40,000g for 15 min at 4°C. The supernatant was discarded, and the pellet was washed twice by resuspension in 40 volumes (w/v) of 50 mM Tris-HCl buffer (pH 7.4) followed by centrifugation. The final pellet was resuspended in 20 volumes (w/v) of 50 mM Tris-HCl (pH 7.4), frozen in 4-ml aliquots, and stored at −70°C until use.
Membranes of Sf9 cells expressing the human recombinant D3 receptors were purchased from PerkinElmer Life and Analytical Sciences.
Dopamine D2 (striatum), 5-HT1A (hippocampus), 5-HT2A (frontal cortex), and rat recombinant D3 binding assays were carried out under incubation conditions as summarized in Table 1. Incubations were stopped by filtration on a Whatman (Clifton, NJ) GF/B glass fiber filter presoaked in 0.05% polyethyleneimine. The filters were washed three times with 1 ml of binding buffer (striatum, Sf9 membranes expressing rat D3 receptor), two times with 5 ml of binding buffer (hippocampus), and three times with 4 ml of binding buffer (frontal cortex). Retained radioactivity was determined after the addition of 4 ml of Optiphase HiSafe (PerkinElmer Life and Analytical Sciences) using an LKB-Wallac 1409 liquid scintillation counter (LKB Instruments, Mount Waverley, Australia).
Cariprazine binding was also tested at 62 neurotransmitter receptors, five ion-channel sites, five transporter sites, and five enzymes at a test concentration of 1 μM by MDS Pharma Service (Taiwan Ltd. Pharmacology Laboratories Peitou, Taipei, Taiwan).
[35S]GTPγS Binding
Striatal and Hippocampal Membrane Preparation.
Rats were decapitated, and their brains were rapidly removed and placed on ice. Striata and hippocampi were dissected out and immediately homogenized in ice-cold buffer containing 50 mM Tris, 5 mM MgCl2, and 1 mM EDTA (pH 7.6) by a glass Dounce homogenizer. Tissue homogenates were centrifuged at 40,000g for 15 min at 4°C. Membrane pellets were resuspended in the same buffer. Hippocampal membranes were incubated for 10 min at 37°C in a shaking water bath to eliminate endogenous serotonin. Pellets were recentrifuged, and the final pellets were resuspended in ice-cold buffer (pH 7.6) containing 50 mM Tris, 100 mM NaCl, 7 mM MgCl2, 1 mM EDTA, and 1 mM dithiotreithol (DTT) to yield a concentration of 20 mg of tissue weight/ml and frozen at −70°C until use.
Human Embryonic Kidney 293-hD2 and CHO-hD3 Cell Membrane Preparation.
Cells were grown at 37°C in a sterile, humidified incubator in 5% CO2 atmosphere in Dulbecco's modified Eagle's medium (DMEM)-F12 medium (Invitrogen, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen), puromycin (Sigma-Aldrich), G418 (Calbiochem, San Diego, CA), and sodium pyruvate (Sigma-Aldrich). Cells were collected in phosphate-buffered saline–EDTA and centrifuged at 2000g for 15 min at 4°C, and the pellet was resuspended and homogenized with a glass Dounce homogenizer in 50 mM Tris (pH 7.6), 5 mM MgCl2, and 1 mM EDTA. The membrane homogenate was washed twice by centrifugation at 40,000g for 15 min at 4°C. The final pellet (80 mg protein/ml) was resuspended in 50 mM Tris (pH 7.6), 100 mM NaCl, 7 mM MgCl2, 1 mM EDTA, and 1 mM DTT, aliquoted, and stored at −70°C until use.
[35S]GTPγS Binding Assays.
These assays were done in 50 mM Tris (pH 7.4), 100 mM NaCl, 7 mM MgCl2, 1 mM EDTA, and 1 mM DTT. Assay tubes (final volume 250 μl) contained 50 μM (striatum and hippocampus) or 1 μM (D2 and D3 cell membrane) GDP, the ligand to be examined, and membrane suspension (250 μg tissue/tube for the striatum and hippocampus and 20 μg protein/tube for hD2 and hD3 membranes). Samples were preincubated for 10 min at 30°C. After the addition of 50 pM [35S]GTPγS, membranes were incubated for an additional 60 min at 30°C. Nonspecific binding was determined in the presence of 10 μM GTPγS; basal binding was determined in the presence of buffer only. The assay was terminated by rapid filtration through UniFilter GF/B (PerkinElmer Life and Analytical Sciences) using a harvester (PerkinElmer Life and Analytical Sciences), and the membranes washed four times with 1 ml of ice-cold buffer. After drying (40°C for 1 h), 40 μl of Microscint (PerkinElmer Life and Analytical Sciences) was added to the filters, and the bound radioactivity was determined by a TopCount NXT counter (PerkinElmer Life and Analytical Sciences).
Determination of IP Accumulation in A9 Cells Expressing Human D2L Receptors
The murine cell line (A9 L hDF2 S.C. 18 cells, CRL-10225) expressing human D2L dopamine receptor was purchased from the American Type Culture Collection (Manassas, VA). The cells were cotransfected with an expression plasmid coding the Gqo5 protein (pCEP-Gqo5-Ha plasmid from Molecular Devices, Sunnyvale, CA) that can activate phospholipase C-β, resulting in generation of diacylglycerol and inositol 1,4,5-trisphosphate, leading to Ca2+ release from intracellular stores. This double-transfected cell line (A9/1/49) was grown in DMEM, supplemented with 10% fetal calf serum, G418, and hygromicin B, and treated with 5 mM sodium butyrate 1 day before the experiment.
Cells were seeded on a 24-well tissue culture plate in 500 μl of medium. Fifty microliters of medium containing 0.55 μCi myo- [3H]inositol was added (final concentration 1 μCi/ml) and incubated for 18–20 h. Cells were then washed three times with buffer containing 140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 5 mM HEPES, 5 mM Na-HEPES, 20 mM glucose, and 10 mM LiCl (pH 7.4). Cells were then incubated for an additional 60 min (37°C) in medium with test compounds alone (agonist test) or alongside 1000 nM (±)-quinpirole (antagonist test). Medium was then aspirated off, cells were lysed by adding 400 μl of 0.1 M HCl/2 mM CaCl2, and supernatants were frozen at −72°C. After thawing and centrifugation at 1000g for 10 min, 200 μl of each supernatant was loaded on 250 μl of AG1-X8 (formate form) anion exchange column. Effluent was discarded, and columns were washed twice in 1.5 ml of distilled water. IPs were eluted with 2.5 ml of 1 M ammonium formate/0.1 M formic acid directly into scintillation vials, 10 ml of Optiphase HiSafe 3 (PerkinElmer Life and Analytical Sciences) was added, and the radioactivity was determined in a TriCarb 4900 scintillation counter (PerkinElmer Life and Analytical Sciences).
cAMP Accumulation in Cells Expressing Recombinant Human D3 Receptors
CHO-K1 cells expressing recombinant human D3 receptors were from Euroscreen (Brussels, Belgium). Cells were grown in DMEM-F12 medium (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen), puromycin (Sigma-Aldrich), G418 (Calbiochem), and sodium pyruvate (Sigma-Aldrich) at 37°C in a sterile, humidified incubator in 5% CO2 atmosphere. Cells were harvested by trypsinization, seeded on poly-d-lysine-coated 96-well plates at a density of 25,000 cells/well, and cultured for 2 days at 37°C. On the day of the experiment, the medium was removed, and cells were preincubated in 60 μl of Hanks' balanced salt solution for 10 min at 37°C and then subjected to test compounds (agonists, antagonists, and forskolin) in 60 μl of Hanks' balanced salt solution complemented with 3-isobutyl-1-methylxanthine (100 μM) for 20 min at 37°C. After adding 20 μl of 1 M perchloric acid to terminate the reaction, plates were frozen overnight at −20°C, thawed, and neutralized (pH 7.4) with the addition of 50 μl of ice-cold 0.5 M KOH. Samples were maintained at 4°C for 30 min and centrifuged at 700g at 4°C for 10 min. cAMP was determined by the method of Nordstedt and Fredholm (1990) using a cAMP binding protein from bovine adrenal cortex. Fifty microliters of supernatant was incubated with 0.15 pmol/well [3H]cAMP in 50 μl of distilled water and 25 μg/well cAMP-binding protein (in 200 μl of 50 mM Tris, pH 7.4) at 4°C for 130 min, and collected onto GF/B filters. Radioactivity of the samples was determined by a TopCount NXT counter (PerkinElmer Life and Analytical Sciences).
Functional Activity at Histaminergic and Serotonergic Receptors
H1 Antagonism.
The functional activity at histamine H1 receptors was determined according to Yamauchi et al. (1994). Male Hartley guinea pigs (Toxicoop) weighing 200 to 250 g were sacrificed by cervical dislocation, and the trachea was removed. Spirally cut trachea (5–6 cartilage rings) were set up in a 10-ml organ bath containing Krebs-Henseleit buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 11 mM glucose, 1.2 mM MgSO4, 25 mM NaHCO3, and 2.5 mM CaCl2, pH 7.4) maintained at 37°C and oxygenated with 95% O2/5% CO2. Resting tension was adjusted to 1.5 g, and the change in isometric tension was measured with an isometric force transducer (Hugo Sachs Elektronik-Harvard Apparatus GmbH, March-Hugstetten, Germany). To prevent modulation of contractile responses by prostaglandins and acetylcholine released from epithelium and presynaptic or postsynaptic neurons, respectively, indomethacin (10 μM) and atropine (0.2 μM) were added for every experiment. During equilibration (90 min), the bath solution was changed every 15 min. After equilibration, histamine (3 μM) was repeatedly applied in 45-min intervals so that after the contraction reached a plateau the bathing solution was exchanged. After testing three contractions, different concentrations of the test substances were added, and the application of histamine was repeated after 30 min.
5-HT2A Antagonism.
A total of 20,000 CHO cells expressing human 5-HT2A receptor (5-HT2A-CHO) cells (Euroscreen) were seeded on a 24-well tissue culture plate in 500 μl of medium. After 100 μl of medium containing 0.6 μCi of myo-[3H]inositol (American Radiolabeled Chemicals) (final concentration 1 μCi/ml) was added, cells were incubated for 18 to 20 h. After loading with myo-[3H]inositol, cells were washed three times with IP buffer (140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 5 mM HEPES, 5 mM Na-HEPES, 20 mM glucose, and 10 mM LiCl, pH 7.4).
Cells were treated with different concentrations of test compounds for 20 min at 37°C, and 5-HT2A receptors were then stimulated by 30 nM DOI for 60 min at 37°C. Determination of IP formation was as described above.
5-HT2B Antagonism.
The functional activity at 5-HT2B receptors was determined in rat isolated fundus strips (Baxter et al., (1994). Male Wistar rats weighing 250 to 350 g (Toxicoop) were killed by cervical dislocation, and longitudinal strips (2–2.5 × 20 mm) were dissected from the greater curvature of the fundus and mounted in a 10-ml tissue bath containing oxygenated (95% O2/5% CO2) Tyrode's solution (136.9 mM NaCl, 2.7 mM KCl, 0.4 mM NaH2PO4, 1.0 MgCl2, 5.6 mM glucose, 11.9 mM NaHCO3, 1.8 mM CaCl2, and 3 μM indomethacin) at 37°C. After preincubation for 2 h, with a starting preload of 2 g, the baseline tension was recorded with an isometric force transducer (Hugo Sachs Elektronik-Harvard Apparatus GmbH) coupled to a Multi-Pen recorder (Rikadenki, Freiburg, Germany). Strips were exposed to 5-HT (100 nM ≅ EC80) three times. Serotonin was applied until a maximum contraction was achieved and then washed out after the tension started to decrease. After the third contraction and washout, the strips were incubated with the test substance (100 nM) for 35 min, and the 5-HT application was repeated. The determination of inhibitory effect of different concentrations of the test compounds was similar to the protocol of the pilot experiment except that 5-HT (100 nM) was repeatedly applied in 35-min intervals, while increasing concentrations of test substances were applied after each 5-HT washout cycle.
Determination of Dopamine, Serotonin, and Metabolites in Mouse Brain Regions
Dopamine, 3,4-dihydroxyphenyl acetic acid (DOPAC), homovanillic acid (HVA), serotonin (5-HT), and 5-hydroxyindolyl acetic acid (5-HIAA) were determined by high-pressure liquid chromatography coupled with electrochemical detection as described (Kiss et al., 2008) and expressed in pmol/g tissue. Ratios of (DOPAC+HVA)/DA and 5-HIAA/5-HT, respectively, were calculated as an index of DA and 5-HT turnover, respectively. Ratios from vehicle-treated animals were used as controls, and drug effects were normalized to percentage of control value.
Determination of DOPA Accumulation
Accumulation of 3,4-dihydroxyphenylalanine (DOPA) after treatment with the aromatic amino acid decarboxylase inhibitor NSD-1015 was used for the estimation of the in vivo dopamine biosynthesis rate. In these experiments, drug- or vehicle-treated animals received NSD-1015 intraperitoneally 30 min before decapitation. DOPA levels were determined from dissected brain regions by high-pressure liquid chromatography coupled with electrochemical detection (Kiss et al., 2008).
Data Analysis
IC50 values were calculated by nonlinear least-squares regression analysis (MDS Pharma, Data Analysis Toolbox; MDL Information Systems, San Leandro, CA). Some rat receptor binding assays and evaluation of functional activity were carried out at the Gedeon Richter Laboratory. Receptor binding isotherms and concentration or dose-response curves were analyzed by nonlinear regression using Origin 6.0 (OriginLab Corp., Northampton, MA) or Prism 4.0 software (GraphPad Software Inc., San Diego, CA); both programs gave identical results. From the IC50 values, inhibition constants (Ki) were calculated by the Cheng and Prusoff (1973) equation; affinity (Kb) values were calculated according to the formula described by Craig (1993). In turnover and biosynthesis experiments, comparison between control and treatment groups was made by Tukey-Kramer multiple comparison test.
Results
In Vitro Receptor Binding Profile of Cariprazine.
The affinity of cariprazine for receptors, channels, and transporters and its effect on certain enzymes are summarized in Table 2. Cariprazine showed picomolar affinity for human dopamine D3 receptors and subnanomolar affinity for both D2L and D2S subtypes of human D2 receptors, and human 5-HT2B receptors. Cariprazine had nanomolar affinity for human 5-HT1A receptors, moderate affinity for human 5-HT2A, histamine H1, and σ1 receptors, and it had low affinity for all tested adrenergic receptors. Cariprazine preference for rat D3 versus D2 receptors was similar to human receptors, although the overall affinity for rat receptors was approximately 10-fold lower than for human receptors. This difference was not observed for binding to human and rat 5-HT1A receptors.
The receptor binding profile of cariprazine differed from aripiprazole, which was assayed alongside cariprazine (Table 2) Aripiprazole displayed highest affinity for both human dopamine D2 subtypes and human 5-HT2B, followed by the human histamine H1, dopamine D3, and human and rat 5-HT1A receptors. Furthermore, aripiprazole bound more potently than cariprazine to rat striatal dopamine D2, human and rat 5-HT2A, 5-HT2C, and adrenergic receptors.
Cariprazine Antagonizes Native Rat and Recombinant Human D2 and D3 Receptors: [35S]GTPγS Binding Results.
Dopamine, pramipexole, quinpirole, APM, 7-OH-DPAT, and (+)PHNO all stimulated [35S]GTPγS binding in rat striatal, human embryonic kidney (HEK)293-D2, and CHO-D3 cell membranes consistent with their agonist properties. Cariprazine and aripiprazole did not show stimulation of [35S]GTPγS binding (Fig. 2A, C, and E; Table 3), and both compounds, similar to other antipsychotics (e.g., haloperidol, olanzapine, and risperidone), inhibited dopamine-stimulated [35S]GTPγS binding in these preparations (Fig. 2 B, D, and F; Table 3). Among the compounds tested, cariprazine displayed the highest antagonist potency for CHO-D3 receptors. Aripiprazole, followed by cariprazine, was the most potent antagonist at HEK-D2 cells. Comparison of the antagonist potency (i.e., pKb) values obtained in these experiments indicated that the reported D3 receptor-selective compound SB-277011 had the highest (100-fold) selectivity toward D3 versus D2 receptors, followed by cariprazine (2.4-fold). Aripiprazole and L741626, a selective D2 antagonist, were the most D2- versus D3-selective compounds in these tests. In the rat striatal membrane preparation, cariprazine showed the highest antagonist potency, followed by aripiprazole, haloperidol, L741626, and risperidone.
Antagonist–Partial Agonist Activity of Cariprazine at hD2 and hD3 Receptors: IP and cAMP Formation.
In murine A9 cells expressing human D2 receptors cotransfected with Gqo5 protein, dopamine, pramipexole, and 7-OH-DPAT fully stimulated IP formation. Cariprazine proved to be partial agonist in this assay; it stimulated IP formation with a high potency (pEC50 8.50) with relatively low efficacy (Emax 30%). Aripiprazole showed similar efficacy (maximum 34%), but its potency (pEC50 7.66) at these human D2 receptors was approximately 7-fold less than cariprazine (Fig. 3A; Table 4). IP formation stimulated by (±)-quinpirole was concentration-dependently and most effectively inhibited by haloperidol (pKb 10.47) followed by cariprazine and aripiprazole. Although aripiprazole displayed higher binding affinity for human D2 receptors (Table 2), it had less antagonist activity (pKb 8.52) than cariprazine (pKb 9.22) in this cell-based functional assay system (Fig. 3B; Table 4).
In vitro functional activity of cariprazine at dopamine D3 receptors was tested by measuring cAMP production in CHO cells transfected with human D3 receptors. Dopamine and the D3/D2 agonists 7-OH-DPAT, quinpirole, and pramipexole all demonstrated full agonist activity (i.e., they inhibited forskolin-stimulated cAMP production; Fig. 3C; Table 4). Cariprazine proved to be a potent partial agonist with relatively high intrinsic activity (Emax 70.9%). Aripiprazole demonstrated partial agonist properties with similar intrinsic efficacy to cariprazine but with an 11-fold lower EC50 value (Fig. 3C; Table 4). As expected from partial agonists, cariprazine and aripiprazole also displayed antagonist properties and partially reversed 7-OH-DPAT-induced inhibition of forskolin-stimulated cAMP accumulation in D3-CHO cells by up to 27 and 24% for cariprazine and aripiprazole, respectively. Pure antagonists, such as haloperidol, fully antagonized the effect of 7-OH-DPAT (Fig. 3D; Table 4). Among the tested compounds, cariprazine displayed the highest antagonist potency in reversing 7-OH-DPAT response, approximately 60-fold higher than aripiprazole.
Functional Activity at Serotonin and Histamine H1 Receptors.
In the [35S]GTPγS binding assay using rat hippocampal membrane preparation both cariprazine and aripiprazole showed partial agonism at 5-HT1A receptors with somewhat higher potency than the 5-HT1A full agonist 8-OH-DPAT (Fig. 4A).
In the in vitro functional assays using CHO cells expressing human 5-HT2A receptors neither cariprazine nor aripiprazole stimulated IP formation, indicating negligible intrinsic activity of both compounds at these receptors. Cariprazine showed weak 5-HT2A receptor antagonist activity, inhibiting the DOI-induced IP formation. Consistent with the in vitro binding results, aripiprazole had approximately 7-fold higher potency than cariprazine in this assay (Fig. 4B).
Both cariprazine and aripiprazole showed high affinity for human 5-HT2B receptors (Table 2). Their functional activity was assayed by using isolated rat stomach fundus, a tissue known to express high levels of 5-HT2B receptors (Fig. 4C). Neither drug caused contraction in this preparation when applied at 100 nM final concentration (not shown). However, both cariprazine and aripiprazole, similar to the selective 5-HT2B antagonist SB-204741, concentration dependently antagonized contractions induced by 100 nM serotonin.
The functional activity of cariprazine at histamine H1 receptors was compared with aripiprazole by using isolated guinea pig trachea preparation (Fig. 4D). Histamine produced half-maximal contraction (EC50) at 1.45 μM, whereas neither cariprazine nor aripiprazole caused contraction in this preparation (not shown). Pyrilamine, a prototypical H1 antagonist, potently antagonized the histamine-induced contractions, whereas both cariprazine and aripiprazole were less potent antagonists.
Effects on Cerebral Dopamine and Serotonin Turnover.
Cariprazine dose dependently increased the DA turnover [i.e., (DOPAC+HVA)/DA ratio] in both mouse striatum and olfactory tubercles. The elevated levels of fronto-cortical DOPAC are also consistent with the D2 receptor antagonist property of cariprazine. Conversely, cariprazine at the higher dose range (i.e., at 3 and 10 mg/kg) appeared to reduce 5-HT turnover (i.e., 5-HIAA/5-HT ratio) (Fig. 5).
The increase in DA turnover evoked by cariprazine treatment (1 mg/kg p.o.) persisted for at least 8 h; at this time, increases of ∼75 and 125% were noted in striatum and olfactory tubercles, respectively. Similar increases in DOPAC levels were found in the frontal cortex (∼130% above control), but DOPAC levels returned to baseline in 2 h. Serotonin turnover was moderately but statistically significantly, reduced (from 15 to 32%) in all three regions at time points of 2, 4, and 8 h (data not shown).
Haloperidol and the atypical antipsychotics risperidone and olanzapine caused a maximal 3- to 4-fold increase in dopamine turnover in the mouse striatum and olfactory tubercles. In contrast, cariprazine and aripiprazole produced lower maximal enhancement of dopamine turnover in these two regions. Furthermore, the latter two drugs produced consistently higher rates of dopamine turnover in the olfactory tubercles than in the striatum (Fig. 6).
Cariprazine Has Dopamine Partial Agonist and Antagonist Properties in the GBL Model In Vivo.
Treatment with GBL induces cessation of impulse flow in the nigrostriatal tract accompanied by increased dopamine biosynthesis in the terminal region (as determined by DOPA accumulation after inhibition of aromatic amino acid decarboxylase by NSD-1015). This method allows for the assessment of activity of presynaptic biosynthesis- and release-modulating dopamine D2 receptors in the striatum (Walters and Roth 1976). The dopamine D2 full agonist APM dose-dependently reduced the GBL-induced enhancement of DOPA accumulation with almost complete blockade at 3 mg/kg s.c. (data not shown). Cariprazine and aripiprazole given before GBL dose dependently but only partially reduced the GBL-induced DOPA accumulation in mouse striatum (Fig. 7A). Significant reduction in GBL-induced increase of DOPA accumulation was achieved by 3 and 10 mg/kg p.o. cariprazine and 10 mg/kg p.o. aripiprazole. Conversely, oral cariprazine given before APM and GBL dose-dependently and fully antagonized the effect of APM-evoked reduction of GBL-induced enhancement of DOPA accumulation, whereas partial antagonism was achieved by aripiprazole even at relatively high doses (i.e., at 10 and 30 mg/kg) (Fig. 7B).
Cariprazine Is a Partial Dopamine Receptor Agonist in the Reserpine Model.
Reserpine treatment, via disruption of vesicular storing mechanisms, leads to depletion of striatal dopamine and causes an increase in dopamine biosynthesis (Hjorth et al., 1988). Reserpine (1 mg/kg s.c.) given to mice 18 h before decapitation enhanced dopamine biosynthesis in the mouse striatum by approximately 90 to 130% above control as measured by DOPA accumulation rate after NSD-1015 treatment. Lower increases (approximately 50–80% above control) of DOPA formation were found in the olfactory tubercles (not shown).
The potency and efficacy of cariprazine for reducing striatal DOPA accumulation in reserpinized mice was compared with drugs that have different functional interactions with dopamine D2 receptors (Fig. 8).
APM, a full dopamine D2 receptor agonist, dose-dependently reduced DOPA accumulation and, at maximal doses, fully blocked DOPA accumulation to 100% below the nonreserpinized control levels. In contrast, dopamine D2 pure antagonists such as the typical antipsychotic haloperidol and the atypical antipsychotics risperidone and olanzapine did not change the reserpine-induced enhancement of striatal DA biosynthesis.
Cariprazine potently and dose-dependently, but only partially, reduced the DA biosynthesis in the striatum of reserpine-treated mice. It is noteworthy that cariprazine administration adjusted DOPA accumulation to the control levels with no further decrease, reaching a plateau beginning at 0.3 mg/kg and extending to 10 mg/kg, the highest dose tested.
Aripiprazole, like cariprazine, dose-dependently and partially inhibited the reserpine-induced increase of DA biosynthesis in the striatum. However, unlike cariprazine, aripiprazole at doses above 1 mg/kg reduced DA biosynthesis below that of nonreserpine controls. At the highest dose tested (30 mg/kg), response to aripiprazole produced a response that was 80% of the maximum effect elicited by the full agonist APM.
Discussion
In Vitro Receptor Profile and Functional Activity of Cariprazine.
Cariprazine demonstrated picomolar affinity for human dopamine D3 and subnanomolar affinity for rat D3 receptors with 6-, 8-, 7-, and 31-fold selectivity against human D2S, D2L, 5-HT2B, and 5-HT1A receptors, respectively, and more than 100-fold selectivity against other tested receptors (Table 2). Cariprazine has a markedly different receptor profile and D3 selectivity than aripiprazole, which demonstrated highest affinity for D2 receptors with much less selectivity against other receptors (Shapiro et al., 2003). Very high affinity for and selectivity toward the dopamine D3 receptor distinguishes cariprazine from compounds like bifeprunox (D2 partial agonist) (Marquis et al., 2005), which was reported to possess antipsychotic-like properties.
Dopamine D2 and D3 receptor agonists [e.g., dopamine, pramipexole, apomorphine, 7-OH-DPAT, quinpirole, (+)PHNO] fully activated [35S]GTPγS binding in membrane preparation from rat striatum, HEK293-D2, and CHO-D3 cells. In these tests, cariprazine and aripiprazole did not show G-protein activation at concentrations up to 10 μM. Lack of agonist efficacy of aripiprazole at D2 receptors using [35S]GTPγS binding is consistent with findings reported by Lin et al. (2006). However, like compounds with D2 and D3 antagonist properties (e.g., haloperidol, L741626, SB-277011, olanzapine, risperidone), cariprazine and aripiprazole antagonized dopamine-stimulated [35S]GTPγS binding. Similar antagonist potency and D2 or D3 selectivity have been reported for aripiprazole (Shapiro et al., 2003), SB-277011 (Reavill et al., 2000), and L741626 (Millan et al., 2000). Among the compounds tested, SB-277011 displayed the highest D3 antagonist selectivity (approximately 100-fold), followed by cariprazine (2.4-fold); however, cariprazine was 27 times more potent at D3 receptors. Consistent with published data, aripiprazole demonstrated the highest D2 potency and selectivity (Burris et al., 2002; Shapiro et al., 2003; Tadori et al., 2005, 2008). Antagonist potencies from the [35S]GTPγS binding assays are consistent with receptor binding affinities for cariprazine and aripiprazole.
In mouse A9 cells expressing human D2L receptors (cotransfected with Gqo5 protein), dopamine, pramipexole, 7-OH-DPAT, and quinpirole concentration-dependently stimulated IP formation. In these cells, pramipexole, 7-OH-DPAT, and dopamine fully activated IP formation, whereas quinpirole was less efficacious. Shapiro et al. (2003) also found lower intrinsic efficacy for quinpirole than for dopamine in MES-235 cells expressing D2L receptors by measuring outward K+ currents. Both cariprazine and aripiprazole demonstrated partial agonist activity with relatively low intrinsic efficacy (Emax 30 and 34%, respectively) and high or medium potency (pEC50 8.50 and 7.66, respectively). Potency and efficacy values obtained for aripiprazole were slightly different from those reported earlier (Burris et al., 2002; Tadori et al., 2005); differences were likely caused by dissimilarities in cellular systems and assay methodologies. Our system contains a transfected “promiscuous” Gqo5 protein, which brings about artificial stochiometry and may explain different results obtained with cariprazine and aripiprazole in native tissues. Both cariprazine and aripiprazole concentration dependently antagonized (±)-quinpirole-induced IP accumulation with high affinity, consistent with results from [35S]GTPγS binding assays.
In CHO cells expressing human D3 receptors, cariprazine displayed partial agonist activity, inhibiting forskolin-induced cAMP accumulation with high potency and intrinsic efficacy (Emax) of 71%. Aripiprazole showed similar intrinsic efficacy, but 10-fold lower potency. Tadori et al. (2008) also found aripiprazole to be a partial agonist at human D3 receptors but with somewhat lower efficacy and potency. Consistent with the hD3 binding data, cariprazine demonstrated high antagonist potency for D3 receptors in this functional assay. Moreover, cariprazine showed approximately 3-fold higher antagonist selectivity for D3 versus D2 receptors, whereas the reverse was true for aripiprazole.
Results from in vitro receptor binding and functional experiments clearly demonstrate high and preferential affinity of cariprazine for dopamine D3 versus D2 receptors and underline that functional activity of compounds like cariprazine (and aripiprazole) largely depends on the assay systems (Urban et al., 2007).
Cariprazine, like aripiprazole, displayed nanomolar affinity for human and rat 5-HT1A receptors; both compounds demonstrated relatively low intrinsic efficacy (Emax 38.6%) in the [35S]GTPγS binding assay using rat hippocampal membrane preparation. At high doses, both aripiprazole and cariprazine reduced serotonin turnover rate in the striatum, olfactory tubercle, and prefrontal-frontal cortex of mouse brain (data not shown), assumedly by partial agonism at 5-HT1A autoreceptors (Fig. 5). These changes, however, were apparent only at doses 10- and 30-fold higher than the ED50 for antipsychotic efficacy (e.g., inhibition of apomorphine-induced climbing) in mice (I. Gyertyán et al., unpublished data).
Reduced cataleptogenic properties of some newly developed antipsychotics are partly attributed to partial agonist activity at 5-HT1A receptors (Bardin et al., 2006). Neurochemical data reported here strongly suggest that, at lower doses, the in vivo partial agonist activity of cariprazine at 5-HT1A receptors may contribute minimally to the antipsychotic-like and side-effect profile; at higher doses, 5-HT1A receptor partial agonism and its contribution to the favorable side-effect profile (i.e., lack of extrapyramidal symptoms) of cariprazine cannot be excluded.
Affinity for 5-HT2A receptors is considered an important component of atypicality of second-generation antipsychotics (Meltzer et al., 2003). Cariprazine displayed approximately 10- to 60-fold lower affinity for h5-HT2A receptors in vitro compared with marketed atypicals including risperidone, olanzapine, clozapine, and aripiprazole (Shahid et al., 2009), and it demonstrated weak antagonist activity (pKb 6.85). Human 5-HT2A affinity and antagonist potency of aripiprazole was 10- and 6-fold higher (pKi 8.75 and pKb 7.72, respectively) than cariprazine. Considering that aripiprazole at clinically effective doses produced 54 to 60% occupancy of 5-HT2A receptors in human positron-emission tomography studies (Mamo et al., 2007), it is anticipated that the clinical effects of cariprazine caused by 5-HT2A receptor activity would be considerably less.
Cariprazine displayed high affinity for 5-HT2B receptors and behaved as a pure antagonist. Consistent with data from Shapiro et al. (2003), aripiprazole also demonstrated high affinity for 5-HT2B receptors. The atypical antipsychotics olanzapine, clozapine, and risperidone have also been reported to display relatively high affinity for 5-HT2B receptors (Wainscott et al., 1996). Whether the 5-HT2B receptor antagonism of cariprazine (and aripiprazole), in addition to its D3, D2, and 5-HT1A affinities, contributes to its antipsychotic-like activity and side-effect profile is presently unclear.
Cariprazine demonstrated approximately 10-fold lower affinity for histamine H1 and 5-HT2C receptors than aripiprazole; in the guinea pig trachea preparation, cariprazine showed H1 antagonist activity. Therefore, it is anticipated that cariprazine may have low propensity for causing sedation and body weight gain, side effects commonly associated with high H1 and 5-HT2C antagonist activity (Kroeze et al., 2003).
In Vivo Dopamine Antagonist–Partial Agonist Properties of Cariprazine.
Cariprazine and aripiprazole, at doses within the in vivo antipsychotic-like efficacy range (I. Gyertyán et al., unpublished data), moderately enhanced dopamine turnover (and biosynthesis) in mouse striatum, olfactory tubercles, and the cortical area, showing antagonist activity at “normosensitive” D2 receptors. Maximal dopamine turnover increase produced by cariprazine in both regions was approximately 2- to 3-fold lower than that achieved by risperidone, olanzapine, and haloperidol. Both cariprazine and aripiprazole, unlike risperidone, olanzapine, or haloperidol, produced greater enhancement of dopamine turnover (and biosynthesis) in mouse olfactory tubercle (i.e., limbic region) compared with striatum. The greater olfactorial versus striatal dopamine turnover enhancing action of aripiprazole in this experiment is consistent with results reported by Nakai et al. (2003). Altogether, these data indicate that cariprazine, similar to aripiprazole, has lower potential for inhibiting dopamine neurotransmission in the striatum relative to the limbic region compared with other antipsychotics (e.g., risperidone, olanzapine, haloperidol), suggesting that cariprazine may have low propensity for causing extrapyramidal symptoms.
The moderate turnover (and biosynthesis)-enhancing actions of cariprazine and aripiprazole resembled those produced by the D2 partial agonists SDZ-208-211, SDZ 208-212, and terguride (Svensson et al., 1991). Both cariprazine and aripiprazole proved to be low-efficacy partial agonists at presynaptic autoreceptors in the GBL model. Conversely, cariprazine fully, whereas aripiprazole only partially, antagonized the biosynthesis-reducing effect of the D2 full agonist apomorphine in GBL-treated animals. In addition, in reserpine-treated mice, cariprazine partially inhibited increased striatal dopamine synthesis, with reversal to the control level (at 0.3 mg/kg and higher), whereas aripiprazole at the highest dose tested caused nearly full reversal. These differences suggest that cariprazine has greater in vivo D2 antagonist activity than aripiprazole.
Cariprazine demonstrated high affinity for dopamine D3 and D2 receptors in binding experiments, with approximately 10-fold selectivity for the D3 versus D2 subtype. In this regard, cariprazine characteristically differs from aripiprazole, the prototype D2 partial agonist antipsychotic, and other clinically used typical and atypical antipsychotics with D2 or D2/5-HT2A antagonism. Cariprazine displayed lower 5-HT2A, 5-HT2C, and H1 receptor affinity relative to marketed antipsychotics. In functional assays, it possessed both antagonist and partial agonist properties at dopamine D3 and D2 receptors depending on the assay/signaling system and showed high D3 antagonist potency. In in vivo neurochemical experiments, cariprazine, similar to aripiprazole, displayed D2-related partial agonist properties in conditions with low dopaminergic tone and behaved as an antagonist in conditions with high dopaminergic tone, demonstrating higher D2 antagonist efficacy than aripiprazole. Potent antagonist–partial agonist activity at D2 and D3 dopamine receptors, with high affinity for D3 subtypes, are properties that may render cariprazine as a potential antipsychotic candidate with a distinct profile. Positive results have been reported in phase II bipolar mania and schizophrenia trials; clinical development of cariprazine is ongoing.
Acknowledgments
We thank K. Domján, Cs. Tóvári, K. Berkó, E. Varga-Tilly, and A. Vasadi for expert technical assistance; Dr. Peter Werner of the Forest Research Institute for critical reading and comments on this manuscript; and Dr. Adam Ruth of the Prescott Medical Communications Group for editorial support.
Footnotes
- Received August 13, 2009.
- Accepted January 19, 2010.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
doi:10.1124/jpet.109.160432.
-
ABBREVIATIONS:
- Cariprazine (RGH-188)
- trans-N-[4-[2-[4-(2,3-dichlorophenyl)piperazin-1-yl]ethyl]cyclohexyl]-N′,N′-dimethylurea hydrochloride
- 5-HT
- serotonin
- 5-HIAA
- 5-hydroxyindolyl acetic acid
- hD2
- human D2
- hD3
- human D3
- DA
- dopamine
- DOPA
- 3,4-dihydroxyphenylalanine
- DOPAC
- 3,4-dihydroxyphenyl acetic acid
- GBL
- γ-butyrolactone
- HVA
- homovanillic acid
- L741626
- 3-[[4-(4-chlorophenyl)-4-hydroxypiperidin-1-yl]methyl-1H-indole
- SB-277011
- trans-N-[4-[2-(6-cyano-1,2,3,4-tetrahydro-2-isoquinolinyl)ethyl]cyclohexyl]quinoline-4-carboxamide
- NSD-1015
- 3-hydroxy-benzylhydrazine HCl
- 7-OH-DPAT
- R(+)-2-dipropylamino-7-hydroxy-1,2,3,4-tetrahydronaphtalene HBr
- 8-OH-DPAT
- R(+)-2-dipropylamino-8-hydroxy-1,2,3,4-tetrahydronaphtalene HBr
- DOI
- (−)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane HCl
- IP
- inositol phosphate
- DMEM
- Dulbecco's modified Eagle's medium
- CHO
- Chinese hamster ovary
- HEK
- human embryonic kidney
- APM
- apomorphine
- DTT
- dithiotreithol
- S33084
- (3aR,9bS)-N-[4-(8-cyano-1,3a,4,9b-tetrahydro-3H-benzopyrano[3,4-c]pyrrole-2-yl)-butyl]-(4-phenyl) benzamide
- S33138
- N-[4-[2-[(3aS,9bR)-8-cyano-1,3a,4,9b-tetrahydro[1]-benzopyrano[3,4-c]pyrrol-2(3H)-yl)ethyl]phenacetamide
- RG-15
- trans-N-{4-[2-[4-(3-cyano-5-trifluoromethyl-phenyl)-piperazine-1-yl]-ethyl]-cyclohexyl}-3-pyridinesulfonic amide dihydrochloride
- (+)-PHNO
- trans-1a,2,3,4a,5,6-hexahydro-9-hydroxy-4-propyl-4H-naphtho[1,2-b]-1,4-oxazine
- SDZ-208-911
- N-[(8a)-2,6-dimethylergoline-8-yl]-2,2-diethylpopanamide
- SDZ-208-912
- N-[(8a)-2-chloro-6-methylergoline-8-yl]-2,2-diethylpropanamide
- MK-912
- (2S,12bS)-1′,3′-dimethylspiro[1,3,4,5′,6,6′,7,12b-octahydro-2H-benzo[b]furo[2,3-a]quinolizine-2,4′-pyrimidin]-2′-one
- RTI-55
- (−)-3β-(4-iodophenyl)tropane-2β-carboxylic acid methyl ester
- SB-204741
- N-(1-methyl-1H-indolyl-5-yl)-N′′-(3-methyl-5-isothiazolyl)urea
- 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.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics