CHIRALITY 13:523–532 (2001)
Resolution, Configurational Assignment, and Enantiopharmacology at Glutamate Receptors of 2-Amino-3-(3-carboxy-5-methyl-4-isoxazolyl)propionic Acid (ACPA) and Demethyl-ACPA TOMMY N. JOHANSEN, TINE B. STENSBØL, BIRGITTE NIELSEN, STINE B. VOGENSEN, KARLA FRYDENVANG, FRANK A. SLØK, HANS BRÄUNER-OSBORNE, ULF MADSEN, AND POVL KROGSGAARD-LARSEN* NeuroScience PharmaBiotec Research Center, Department of Medicinal Chemistry, The Royal Danish School of Pharmacy, Copenhagen, Denmark Dedicated to Professor Koji Nakanishi on the occasion of his 75th birthday
ABSTRACT We have previously described (RS)-2-amino-3-(3-carboxy-5-methyl-4isoxazolyl)propionic acid (ACPA) as a potent agonist at the (RS)-2-amino-3-(3-hydroxy5-methyl-4-isoxazolyl)propionic acid (AMPA) receptor subtype of (S)-glutamic acid (Glu) receptors. We now report the chromatographic resolution of ACPA and (RS)-2-amino-3(3-carboxy-4-isoxazolyl)propionic acid (demethyl-ACPA) using a Sumichiral OA-5000 column. The configuration of the enantiomers of both compounds have been assigned based on X-ray crystallographic analyses, supported by circular dichroism spectra and elution orders on chiral HPLC columns. Furthermore, the enantiopharmacology of ACPA and demethyl-ACPA was investigated using radioligand binding and cortical wedge electrophysiological assay systems and cloned metabotropic Glu receptors. (S)ACPA showed high affinity in AMPA binding (IC50 = 0.025 µM), low affinity in kainic acid binding (IC50 = 3.6 µM), and potent AMPA receptor agonist activity on cortical neurons (EC50 = 0.25 µM), whereas (R)-ACPA was essentially inactive. Like (S)-ACPA, (S)demethyl-ACPA displayed high AMPA receptor affinity (IC50 = 0.039 µM), but was found to be a relatively weak AMPA receptor agonist (EC50 = 12 µM). The stereoselectivity observed for demethyl-ACPA was high when based on AMPA receptor affinity (eudismic ratio = 250), but low when based on electrophysiological activity (eudismic ratio = 10). (R)-Demethyl-ACPA also possessed a weak NMDA receptor antagonist activity (IC50 = 220 µM). Among the enantiomers tested, only (S)-demethyl-ACPA showed activity at metabotropic receptors, being a weak antagonist at the mGlu2 receptor subtype (KB = 148 µM). Chirality 13:523–532, 2001. © 2001 Wiley-Liss, Inc. KEY WORDS: chiral HPLC; X-ray crystallography; AMPA receptor agonism; metabotropic receptor antagonism (S)-Glutamic acid (Glu) is the main excitatory neurotransmitter in the mammalian central nervous system (CNS). Receptors activated by Glu consist of a family of ionotropic Glu (iGlu) receptors and a family of G-proteincoupled metabotropic Glu (mGlu) receptors. The family of iGlu receptors, which mediates fast synaptic transmission through ion channels, is subdivided into three classes of receptors, N-methyl- D -aspartic acid (NMDA), (RS)-2amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA), and kainic acid (KA) receptors.1,2 Eight mGlu receptor subtypes have been cloned and, based on pharmacology, signal transduction pathway, and sequence homology, these receptor subtypes have been divided into three groups: mGlu1,5, mGlu2,3, and mGlu4,6,7,8.3–5 It is generally agreed that iGlu as well as mGlu receptors play important roles in the healthy and in the diseased CNS and that both receptor families are potential drug targets.1–3 © 2001 Wiley-Liss, Inc.
In order to identify potent and selective Glu receptor ligands as pharmacological tools a number of AMPA analogs with alkyl, aryl, and heteroaryl substituents in the 5-position of the isoxazole ring have been synthesized and pharmacologically characterized.6–8 With the exception of the demethylated analog of AMPA (Fig. 1), which shows relatively high AMPA receptor affinity but weak AMPA receptor agonist potency, a good correlation between affinity and agonist potency has been observed for these compounds.6 Replacing the 3-isoxazolol moiety of AMPA by a
Contract grant sponsors: The Danish Medical Research Council, the Lundbeck Foundation, and the Alfred Benzon Foundation. *Correspondence to: Professor Povl Krogsgaard-Larsen, Department of Medicinal Chemistry, The Royal Danish School of Pharmacy, 2 Universitetsparken, DK-2100 Copenhagen, Denmark. E-mail:
[email protected] Received for publication 27 October 2000; Accepted 15 January 2001
524
JOHANSEN ET AL.
tometer. Optical rotations were measured in thermostated cuvettes on a Perkin-Elmer 241 polarimeter. Circular dichroism (CD) spectra were recorded in 0.1 M HCl in 1.0 cm cuvettes at room temperature on a Jasco J-720 spectropolarimeter. Elemental analyses were performed at Microanalytical Laboratory, Department of Physical Chemistry, University of Vienna, Austria, or by Mr. P. Hansen, Department of General and Organic Chemistry, University of Copenhagen, Denmark, and were within ±0.4% of the theoretical values. [3H]-(RS)-3-(2-Carboxy-4-piperazinyl)propyl-1phosphonic acid ([3H]CPP), [3H]AMPA, and [3H]KA were purchased from New England Nuclear (Boston, MA, USA). All other chemicals were obtained through standard commercial sources. Chiral Liquid Chromatography
Fig. 1. Structures of Glu, AMPA, demethyl-AMPA, and the enantiomers of ACPA and demethyl-ACPA.
3-carboxyisoxazole unit afforded the Glu homolog (RS)-2amino-3-(3-carboxy-5-methyl-4-isoxazolyl)propionic acid (ACPA), which has been reported to be a selective AMPA receptor agonist more potent than AMPA on cortical neurons as well as on Xenopus oocytes expressing cloned AMPA receptors.9,10 Based on binding experiments and electrophysiological studies, ACPA appears to bind to and activate AMPA receptors in a manner different from that of AMPA.10,11 In continuation of these studies, we now report on the resolution and the configurational assignment of the enantiomers of ACPA and of the demethylated analog of ACPA, demethyl-ACPA (Fig. 1). Furthermore, the enantiopharmacology of these pairs of enantiomers at native iGlu receptors and at cloned mGlu receptors are reported. MATERIALS AND METHODS
Melting points were determined in capillary tubes and are uncorrected. Column chromatography was performed on Merck silica gel 60 (70–230 mesh). 1H NMR spectra were recorded either on a Varian Gemini-2000 BB (300 MHz) NMR spectrometer in D2O using acetonitrile (␦ 2.06 ppm) as internal standard, or on a Bruker AC-200F (200 MHz) NMR spectrometer in CDCl3 or in D2O using TMS (␦ 0 ppm) or 1,4-dioxane (␦ 3.75 ppm) as internal standards. IR spectra were recorded from KBr discs on a Perkin-Elmer (Norwalk, CT) 781 grating infrared spectropho-
Preparative chromatography was performed with an HPLC system consisting of either a Jasco Model 880 pump or a Waters Model 590 pump, a Rheodyne Model 7125 injector fitted with a 5.0 ml sample loop, and a Waters Model 481 detector set at 225 nm connected to a Hitachi D-2000 Chromato-Integrator. The temperature was controlled by an LKB 2155 HPLC column oven. The preparative resolutions were performed on a Sumichiral OA-5000 column (250 × 10 mm i.d.) connected to a Sumichiral OA-5000 guard column (10 × 4.0 mm i.d.) (Sumika Chemical Analysis Service, Japan). The column was eluted at 3.0 ml/min with an aqueous solution of ammonium acetate (10 mM), adjusted to pH 4.7 using acetic acid, and containing copper(II) acetate (0.1 mM) and 2-propanol (5% v/v). The temperature was kept at 50°C and 38°C for the resolution of ACPA and demethyl-ACPA, respectively. Removal of copper ions was performed on an XK-16/20 column (Pharmacia, Uppsala, Sweden), packed with Chelating Sepharose Fast Flow (Pharmacia) (capacity: 0.7–1.0 mmol copper ions) and eluted with water at 1.0 ml/min. Ion-exchange was performed on the same type of column, packed with S-Sepharose (Pharmacia) (ion capacity: 6.8 mmol) and eluted at 3.0 ml/min first with acetic acid (2 M) and then with water. The analytical determinations of the enantiomeric excess (ee) of the isolated enantiomers of ACPA and demethyl-ACPA were performed on a Waters HPLC system consisting of an M510 pump, a U6K injector, and an M991 photodiode array detector. The Sumichiral OA-5000 column (150 × 4.6 mm i.d.) equipped with a Sumichiral OA5000 guard column (10 × 4.0 mm i.d.) (both from Sumika Chemical Analysis Service, Japan) was eluted using a flow rate of 1.0 ml/min at room temperature with a solution of ammonium acetate (50 mM), adjusted to pH 4.7 using acetic acid and containing copper(II) acetate (0.1 mM) and 2-propanol (5% v/v). The ee values were calculated from peak areas at 225 nm and 220 nm for the ACPA and the demethyl-ACPA enantiomers, respectively. Chemistry and Chromatographic Resolutions Ethyl 4-bromomethyl-3-isoxazolecarboxylate (2). A mixture of ethyl 3-isoxazolecarboxylate12 (1) (11.7 g, 83 mmol), 1,3,5-trioxane (9.7 g, 107 mmol), and 63% aqueous HBr
ACPA AND DEMETHYL-ACPA ENANTIOMERS
(100 ml) was stirred in a sealed flask at 55°C for 14 days. After cooling, the mixture was poured into ethanol (250 ml), evaporated at 60°C, and reevaporated three times from a mixture of ethanol (100 ml) and toluene (400 ml). The residue was dissolved in CH2Cl2 (200 ml) and washed with water (200 ml). The organic phase was dried (MgSO4), filtered, and evaporated. Column chromatography [toluene/ethyl acetate (30:1)] afforded 2 (2.4 g containing 25 mol% 1 based on the 1H NMR spectrum, corresponding to a yield of 9% of the theoretical value); 1H NMR (200 MHz, CDCl3): ␦ 8.66 (1H, s), 4.58 (2H, s), 4.48 (2H, q, J = 7.1 Hz), 1.39 (3H, t, J = 7.1 Hz). The mixture was used without further purification. Methyl 2-acetamido-2-methoxycarbonyl-3-(3ethoxycarbonyl-4-isoxazolyl)propionate (3). NaH (404
mg, 60% suspension in mineral oil, 10.1 mmol) was suspended in dry DMF (30 ml). A solution of dimethyl acetamidomalonate (1.9 g, 10.0 mmol) dissolved in dry DMF (20 ml) was added at 0°C over a period of 30 min followed by the addition of a solution of 2 (2.4 g, corresponding to 7.7 mmol) in dry DMF (10 ml) over a period of 1 h. Stirring was continued for 1 h at 0°C and for 12 h at room temperature. After addition of acetic acid (1 ml) and evaporation, the residue was dissolved in CH2Cl2 (125 ml) and washed with 1 M NaOH (3 × 50 ml) and with water (50 ml). The organic phase was dried (MgSO4), filtered, and evaporated and the residue was recrystallized (ether/ petroleum ether) affording 3 (1.7 g, 64%); mp 108–111°C; 1 H NMR (200 MHz, CDCl3): ␦ 8.30 (1H, s), 6.65 (1H, s), 4.42 (2H, q, J = 7.1 Hz), 3.80 (6H, s and 2H, s), 2.02 (3H, s), 1.42 (3H, t, J = 7.1 Hz). Anal. C14H18N2O8: C, H, N. (RS)-2-Amino-3-(3-carboxy-4-isoxazolyl)propionic acid (demethyl-ACPA). A solution of 3 (800 mg, 2.34
mmol) in hydrochloric acid (150 ml, 2 M) was refluxed for 20 h, evaporated, and reevaporated twice from water (2 × 50 ml). The residue was dissolved in ethanol and propylene oxide (5.0 ml, 0.072 mol) was added. The precipitate was collected and recrystallized (water) giving demethyl-ACPA (150 mg, 31%); mp 221–223°C; 1H NMR (200 MHz, D2O): ␦ 8.30 (1H, s), 4.15 (1H, dd, J = 5.4 and 6.9 Hz), 3.32 (1H, dd, J = 5.4 and 15.5 Hz), 3.18 (1H, dd, J = 6.9 and 15.5 Hz);. Anal. C7H8N2O5 ⭈ 0.25 H2O: C, H, N. Evaporation of the mother liquor and recrystallization (water) afforded an additional 43 mg of demethyl-ACPA (total yield: 40%). (−)- and (+)-2-Amino-3-(3-carboxy-4-isoxazolyl)propionic acid [(−)- and (+)- demethyl-ACPA]. Racemic de-
methyl-ACPA (200 mg, 1.00 mmol) dissolved in water (3 mg/ml) was passed through a Millex HV filter (45 µm, Millipore, Bedford, MA) and resolved on a Sumichiral OA5000 column in up to 10 mg per injection. In each run the two enantiomers were collected separately. The combined fractions of the first eluting enantiomer having negative optical rotation at 589 nm were pooled, evaporated to dryness, and reevaporated three times from water. The resulting clear blue oil was dissolved in water and copper ions were removed by passing the solution, in six injections, through Chelating Sepharose. Appropriate
525
fractions were pooled, evaporated, and reevaporated three times from water. In order to remove traces of ammonium ions the residue was dissolved in acetic acid (2 M) and subjected to ion-exchange chromatography. Appropriate fractions were pooled, evaporated, reevaporated three times from water, and dried in vacuo over P2O5. Recrystallization of the white crystalline residue from water gave (−)-demethyl-ACPA (78.1 mg, 78%); mp >250°C; ee >99.9%; 25 1 [␣]25 589 = −4.8°, [␣]365 = +5.8° (c = 0.40, 0.1 M HCl); H NMR (300 MHz, D2O): ␦ 8.62 (1H, s), 4.18 (1H, dd, J = 5.4 and 6.9 Hz), 3.35 (1H, dd, J = 5.4 and 15.0 Hz), 3.22 (1H, dd, J = 6.9 and 15.1 Hz); IR: 3600–2500 (several bands, m-s), 1735 (br, m), 1670 (br, m), 1610 (br, m), 1527 (s) cm−1; ⌬ (215 nm) = +0.13 m2/mol. Recrystallization of the mother liquor furnished an additional 7.4 mg (−)-demethyl-ACPA; mp >250°C; ee = 99.8% having an IR spectrum identical with that of the first batch (total yield: 86%). All of the fractions of the second enantiomer having a positive optical rotation at 589 nm were processed as described for the first enantiomer. The white crystalline residue was recrystallized from water to give (+)-demethylACPA (77.8 mg, 78%); mp >250°C; ee = 99.9% [␣]25 589 = +4.1°, [␣]25 365 = −6.4° (c = 0.39, 0.1 M HCl); ⌬ (215 nm) = −0.12 m2/mol. IR spectrum was identical with that of (−)demethyl-ACPA. Anal. C7H8N2O5: C, H. N. Recrystallization of the mother liquor furnished an additional 4.0 mg (+)-demethyl-ACPA; ee = 99.3%. Melting point and IR spectrum were identical with those of the first batch of (+)demethyl-ACPA (total yield: 82%). (+)- and (−)-2-Amino-3-(3-carboxy-5-methyl-4isoxazolyl)propionic acid [(+)- and (−)-ACPA]. Racemic
ACPA9 (214 mg, 1.00 mmol) was dissolved in water (3 mg/ml), passed through a Millex HV filter (45 µm, Millipore), and resolved on a Sumichiral OA-5000 column in up to 4.5 mg per injection. In each run the two enantiomers were collected separately. All of the fractions of the first enantiomer having a positive optical rotation at 589 nm were pooled and reduced in vacuo to a blue, almost clear solution (20–30 ml). A very small amount of blue insoluble material was, after heating, filtered off, and in order to remove copper ions the filtrate was passed through Chelating Sepharose in eight injections. Appropriate fractions were pooled, evaporated, and reevaporated three times from water and then lyophilized. In order to remove traces of ammonium ions, the colorless oil was dissolved in acetic acid (2 M) and, in six injections, passed through an S-Sepharose ion-exchange column. Appropriate fractions were pooled, evaporated, and reevaporated three times from water and dried in vacuo over P2O5. Recrystallization of the white crystalline residue from water gave (+)-ACPA (74.8 mg, 70%); mp 211–212°C (dec.); ee 25 > 99.9%; [␣]25 589 = +9.4°, [␣]365 = +50.9° (c = 0.33, 0.1 M HCl); 2 ⌬ (213 nm) = +0.13 m /mol; 1H NMR (300 MHz, D2O): ␦ 4.05 (1H, t, J = 6.2 Hz), 3.23 (1H, dd, J = 6.2 and 15.2 Hz), 3.11 (1H, dd, J = 6.3 and 15.3 Hz), 2.38 (3H, s); IR: 3700– 2800 (several bands, m-s), 1690 (m), 1620 (br, s), 1550 (m) cm−1. Recrystallization of the mother liquor furnished an additional 12.8 mg (+)-ACPA; mp 210–211°C (dec.); ee =
526
JOHANSEN ET AL.
99.9%. The IR spectrum was identical with that of the first batch (total yield: 82%). The combined fractions of the second enantiomer were processed as described above for the first enantiomer. Recrystallization from water gave (−)-ACPA (66.7 mg, 62%); 25 mp 212–213°C (dec.); ee = 99.9%; [␣]25 589 = −10.7°, [␣]365 = 2 −52.2° (c = 0.34, 0.1 M HCl); ⌬ (213 nm) = −0.13 m /mol. Anal. C8H10N2O5: C, H, N. The IR spectrum was identical with that of (+)-ACPA. The mother liquor furnished, after recrystallization from water, an additional 18.5 mg (−)ACPA; ee = 99.7%. Melting point and IR spectrum were identical with those of the first batch (total yield: 80%). X-ray Crystallographic Analyses X-ray crystallographic analysis of (−)-(R)-ACPA. Colorless single crystals were obtained from a solution in water. Crystal data: [C8H10N2O5], Mr = 214.18, orthorhombic, space group P212121 (No. 19), a = 5.734(1) Å, b = 6.1676(9) Å, c = 25.563(6) Å, V = 904.1(3) Å3, Z = 4, Dc = 1.574 Mg/m3, F(000) = 448, µ(Cu K␣) = 1.147 mm−1, T = 122.0 (5) K, crystal dimensions = 0.10 × 0.10 × 0.27 mm.
Diffraction data were collected on an Enraf-Nonius CAD-4 diffractometer using graphite monochromated Cu K␣ radiation ( = 1.54184 Å).13 Intensities were collected using the /2 scan mode. Unit cell dimensions were determined by least-squares refinement of 25 reflections ( range 39.37–40.56°).13 The reflections were measured in the range −7 ⱕ h ⱕ 7, −7 ⱕ k ⱕ 7, −31 ⱕ l ⱕ 32, (3.46° < < 74.94°). Data were reduced using the programs of Blessing (DREADD).14,15 The intensities of five standard reflections were monitored every 104 sec (decay 3.5%, corrected). Absorption correction was applied using the program ABSORB (Tmin 0.823; Tmax = 0.938).16 A total of 7,605 reflections were averaged according to the point group symmetry 222 resulting in 1,863 unique reflections (Rint = 0.0400 on F2o). The structure was solved by the direct method using the program SHELXS9717,18 and refined using the program SHELXL97.19 Full matrix least-squares refinement on F2 was performed, minimizing ∑w(F2o − F2c)2, with anisotropic displacement parameters for the nonhydrogen atoms. The positions of the hydrogen atoms were located on intermediate difference electron density maps and refined with fixed isotropic displacement parameters. Correction for extinction was applied (coefficient: 0.0067(7)). The refinement (168 parameters, 1,863 reflections) with the molecule having the R-configuration converged at RF = 0.0242, wRF = 0.0643 for 1,821 reflections with Fo > 4(Fo); w = 1/[2(F2o) + (0.0142P)2 + 0.1774P], where P = (F2o + 2F2c)/3; S = 1.099. In the final difference Fourier map, maximum and minimum electron densities were 0.211 and −0.173 eÅ−3, respectively. Refinement of the Flack absolute structure factor x in the final refinement gave x = −0.02(16).19,20 The large standard deviation of x is due to the atom types present in the compound (C, H, N, and O), which produce minor anomalous scattering. Complex atomic scattering factors for neutral atoms were as incorporated in SHELXL97.19,21 2
X-ray crystallographic analysis of (−)-(S)-demethylACPA. Colorless single crystals were obtained by the
hanging drop vapor diffusion method at 20°C: a mixture of (−)-demethyl-ACPA (2.7 µl, 0.03 M) and trifluoromethanesulfonic acid (1.7 µl, 0.05 M) both in water was equilibrated against an aqueous reservoir solution of NaCl (0.5 ml, 1 M). Crystal data: [C7H8N2O5], Mr = 200.15, orthorhombic, space group P212121 (No. 19), a = 6.7876(7) Å, b = 7.888(2) Å, c = 14.753(2) Å, V = 789.9(2) Å3, Z = 4, Dc = 1.683 Mg/m3, F(000) = 416, µ(Cu K␣) = 1.266 mm−1, T = 122.0 (5) K, crystal dimensions = 0.10 × 0.20 × 0.40 mm. Diffraction data were collected on an Enraf-Nonius CAD-4 diffractometer using graphite monochromated Cu K␣ radiation ( = 1.54184 Å).13 Intensities were collected using the /2 scan mode. Unit cell dimensions were determined by least-squares refinement of 20 reflections ( range 39.22–40.59°).13 The reflections were measured in the range −8 ⱕ h ⱕ 8, −9 ⱕ k ⱕ 9, −18 ⱕ l ⱕ 18, (6.00° < < 74.89°). Data were reduced using the programs of Blessing (DREADD).14,15 The intensities of five standard reflections were monitored every 104 sec (decay 5.4%, corrected). Absorption correction was applied using the program ABSORB (Tmin = 0.664; Tmax = 0.894).16 A total of 6,391 reflections were averaged according to the point group symmetry 222 resulting in 1,614 unique reflections (Rint = 0.0387 on F2o). Structure solution and refinement were performed as described above for (−)-(R)-ACPA. Correction for extinction was applied (coefficient: 0.0184(12)). The refinement (153 parameters, 1,614 reflections) with the molecule having the S-configuration converged at RF = 0.0239, wRF = 0.0657 for 1,609 reflections with Fo > 4(Fo); w = 1/[2(F2o) + (0.0376P)2 + 0.2308P], where P = (F2o + 2F2c)/3; S = 1.068. In the final difference Fourier map, maximum and minimum electron densities were 0.308 and −0.165 eÅ−3, respectively. Refinement of the Flack absolute structure factor x in the final refinement gave x = 0.00(17).19,20 The large standard deviation of x is due to the atom types present in the compound (C, H, N, and O), which produce minor anomalous scattering. Complex atomic scattering factors for neutral atoms were as incorporated in SHELXL97.19,21 2
In Vitro Pharmacology Receptor binding assays and in vitro electrophysiology.
Affinity for NMDA, AMPA, and KA receptor sites was determined using the ligands [ 3 H]CPP 2 2 (5 nM), [3H]AMPA23 (5 nM), and [3H]KA24 (5 nM), respectively, with the modifications previously described.25 A previously described rat cortical wedge preparation26 in a slightly modified version27 was used for the electrophysiological characterization. Agonists were applied for 90 sec, whereas antagonists were applied for 90 sec followed by a 90-sec coapplication of agonist and antagonist. Cell culture. The Chinese hamster ovary (CHO) cell lines stably expressing mGlu1␣, mGlu2, or mGlu4a receptors have previously been described.28–30 They were maintained at 37°C in a humidified 5% CO2 incubator in Dul-
ACPA AND DEMETHYL-ACPA ENANTIOMERS
527
becco’s Modified Eagle Medium (DMEM) containing a reduced concentration of (S)-glutamine (2 mM) and were supplemented with 1% (S)-proline, penicillin (100 U/ml), streptomycin (100 mg/ml) and 10% dialyzed fetal calf serum (all GIBCO, Paisley, Scotland). Two days before assay 1.8 × 106 cells were divided into 48-well plates (mGlu1␣) or 96-well plates (mGlu2 and mGlu4a). Second-messenger assays. PI hydrolysis was measured as described previously.31,32 Briefly, the cells were labeled with [3H]inositol (2 µCi/ml) 24 h prior to the assay. For agonist assays, the cells were incubated with ligand dissolved in phosphate-buffered saline (PBS)-LiCl for 20 min, and agonist activity was determined by measurement of the level of [3H]-labeled mono-, bis-, and tris-inositol phosphates by ion-exchange chromatography. For antagonist assays, the cells were preincubated with the ligand dissolved in PBS-LiCl for 20 min prior to incubation with ligand and 30 µM Glu for an additional 20 min. The antagonist activity was then determined as the inhibitory effect of the Glu mediated response. The assay of cyclic AMP formation was performed as described previously.31,32 Briefly, the cells were incubated for 10 min in PBS containing the ligand and 10 µM forskolin and 1 mM 3-isobutyl-1methylxanthine (IBMX) (both Sigma Chemical, St. Louis, MO). The agonist activity was then determined as the inhibitory effect of the forskolin-induced cyclic AMP formation. For antagonist assay, the cells were preincubated with ligand dissolved in PBS containing 1 mM IBMX for 20 min prior to a 10-min incubation in PBS containing the ligand, 30 µM Glu, 10 µM forskolin, and 1 mM IBMX. Compounds showing antagonist activity were further analyzed by performing concentration–response curves of Glu in the absence or presence of a fixed concentration of the antagonist. The antagonist potency was calculated from the Gaddum equation KB = [B]/(DR − 1),33 where the dose-ratio (DR) is the ratio of the EC50 values of Glu in the presence and in the absence of a fixed antagonist concentration [B]. RESULTS Chemistry and Resolution
The synthesis of racemic demethyl-ACPA is shown in Scheme 1. Ethyl 3-isoxazolecarboxylate (1) was prepared
Fig. 2. Chromatograms showing enantioseparation of 4.5 mg ACPA (A) and of 10 mg demethyl-ACPA (B) on a Sumichiral OA-5000 column (250 × 10 mm). The signs of the optical rotation at 589 nm are indicated.
using a 1,3-dipolar cycloaddition reaction as described12 starting from vinyl acetate and ethyl 2-chloro-2-hydroxyiminoacetate.34 Despite extensive optimization of the following bromomethylation reaction on 1, it was not possible to obtain complete conversion of the starting material without extensive decomposition. After a quick purification on silica gel, compound 2 was isolated as a 3:1-mixture with starting material 1, in a yield corresponding to 9% of the theoretical value. Using the sodium salt of dimethyl acetamidomalonate, compound 2 was subsequently converted to the protected amino acid 3, which finally was deprotected to demethyl-ACPA, isolated in the zwitterionic form. Preparative resolution of demethyl-ACPA and ACPA was carried out by liquid chromatography on a Sumichiral OA5000 chiral ligand-exchange column35 (Fig. 2). In both cases, satisfactory enantioseparations were obtained using an ammonium acetate/acetic acid buffer containing 2-propanol and complexing copper ions as mobile phase. After removal of the mobile phase, the individual enantiomers were obtained in good yields and with high stereochemical purity (ee ⱖ 99.3%), as determined on an analytical Sumichiral OA-5000 column. Configurational Assignment
Scheme 1 Synthesis of demethyl-ACPA.
In order to obtain crystals suitable for an absolute configurational assignment of (−)-demethyl-ACPA, attempts were made to prepare salts containing atoms with large anomalous dispersion effects or components with known absolute stereochemistry. These attempts were, however, unsuccessful. Structure solution of the crystal obtained in the presence of trifluoro-methanesulfonic acid (Fig. 3A) revealed that the crystal was only composed of the zwitterion of (−)-demethyl-ACPA. In order to improve the configurational assignment of the individual isomers, a single
528
JOHANSEN ET AL.
Fig. 4. CD-spectra of the enantiomers of ACPA (A) and of demethylACPA (B) (c = 0.35 and 0.44 mM, respectively, in 0.1 M HCl).
Fig. 3. Single crystal of (−)-(S)-demethyl-ACPA used in the X-ray crystallographic analysis (A) (crystal dimension: 0.1 × 0.2 × 0.4 mm). Perspective drawings36 of the molecules of (−)-(R)-ACPA (B) and of (−)-(S)demethyl-ACPA (C). Displacement ellipsoids enclose 50% probability. Hydrogen atoms are represented by spheres of arbitrary size.
crystal of zwitterionic (−)-ACPA was also analyzed by X-ray crystallography.* Perspective drawings36 of the molecular structures of the two compounds with atomic labeling are depicted in Figure 3B, C. For (−)-ACPA, the Flack absolute *Crystallographic data for the structures reported in this article have been deposited with the Cambridge Crystallographic Data Centre [(−)-ACPA, CCDC# 155313; (−)-demethyl-ACPA, CCDC# 155314]. Copies of the data can be obtained, free of charge, on application to the Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, United Kingdom (fax: 44-1223-336-033 or e-mail:
[email protected]).
structure parameter19,20 for the R-configuration was calculated to x = −0.02(16) and for the S-configuration to x = 1.02(16). For (−)-demethyl-ACPA, the Flack absolute structure parameter19,20 for the S-configuration was calculated to x = 0.00(17) and for the R-configuration to x = 0.96(16). These results strongly suggest that (−)-ACPA possesses the R-configuration and that (−)-demethyl-ACPA possesses the S-configuration. However, unequivocal statements are not possible due to the high SD of the Flack parameters. To strengthen the configurational assignments obtained from the X-ray crystallographic analyses, CD spectra were recorded for both pairs of enantiomers in acidic solution. Although the first eluting enantiomer of ACPA and of demethyl-ACPA on the Sumichiral OA-5000 column have opposite signs of optical rotation at 589 nm (Fig. 2), the two enantiomers have very similar CD spectra (Fig. 4). (+)ACPA and (−)-demethyl-ACPA both display positive Cotton effects at around 215 nm, which is in agreement with naturally occurring S-␣-amino acids and structurally similar 3-isoxazolol-containing S-␣-amino acids with known absolute stereochemistry.8,25,37,38
ACPA AND DEMETHYL-ACPA ENANTIOMERS
529
TABLE 1. Membrane receptor binding and rat cortical wedge electrophysiological data [3H]AMPA
[3H]KA
[3H]CPP
IC50 (µM) a
AMPA DemethylAMPAb ACPAc (S)-ACPA (R)-ACPA DemethylACPA (S)-DemethylACPA (R)-DemethylACPA
Cortical wedge EC50 (µM)
0.04 ± 0.014 0.27
>100 >100
>100 >100
3.5 ± 0.2 900
0.020 ± 0.012 0.025 ± 0.004 80 ± 11 0.10 ± 0.01
6.3 ± 1.6 3.6 ± 0.7 >100 14 ± 2
>100 >100 >100 15 ± 4
1.0 ± 0.1 0.25 ± 0.02 367 ± 55 23 ± 2
0.039 ± 0.007
14 ± 2
>100
12 ± 1
9.4 ± 3.1
>100
4.5 ± 1.1
108 ± 16d
Values are expressed as mean ± SEM of 3–5 individual experiments. a Madsen et al.27 b Sløk et al.6 c Madsen and Wong.9 d Plus NMDA receptor antagonism (IC50 = 220 ± 36 µM).
Studies of the elution orders of the ACPA and demethylACPA enantiomers on three different chiral HPLCcolumns provided additional supporting evidence for the assignment of (−)-ACPA and (−)-demethyl-ACPA as having R- and S-configuration, respectively. As expected for S-␣amino acids, both (+)-ACPA and (−)-demethyl-ACPA were the first eluting enantiomers on the Sumichiral OA-5000, on the (R)-proline, and on the Crownpak CR(−) columns.35,39,40 In conclusion, all data obtained from the CD and chiral HPLC studies support the configurational assignments based on X-ray crystallographic analyses. In Vitro Pharmacology
The enantiomers of ACPA and demethyl-ACPA were characterized in iGlu receptor binding assays using [3H]CPP,22 [3H]AMPA,23 and [3H]KA24 as radioligands (Table 1). The binding affinities previously reported for racemic ACPA in [3H]AMPA and [3H]KA binding assays9 were shown to reside in the S-enantiomer, (S)-ACPA having high affinity for [3H]AMPA binding sites (IC50 = 0.025 µM) and markedly lower affinity for [3H]KA binding sites (IC50 = 3.6 µM). (R)-ACPA was essentially inactive in all three binding assays, and the low affinity of (R)-ACPA observed in the [3H]AMPA binding assay (IC50 = 80 µM) is likely caused not by (R)-ACPA itself (ee = 99.9%) but rather by the small contamination of the S-enantiomer. DemethylACPA turned out to have a more complex binding profile with affinity in all three binding assays. The affinity in [3H]AMPA and [3H]KA binding were shown to reside primarily in the S-enantiomer (IC50 = 0.039 µM and 14 µM, respectively), whereas the affinity in [3H]CPP binding exclusively originated from the R-enantiomer (IC50 = 4.5 µM). However, in the case of (R)-demethyl-ACPA (ee = 99.9%) the observed affinity in [3H]AMPA binding (IC50 = 9.4 µM) can only be partially explained by the presence of a stereochemical impurity.
Fig. 5. Trace from the rat cortical wedge preparation illustrating antagonism of NMDA-induced currents elicited by (R)-demethyl-ACPA.
In the rat cortical wedge preparation,26 (S)-ACPA induced a potent agonist response (EC50 = 0.25 µM) (Table 1). In accordance with the receptor binding data, (R)-ACPA (ee = 99.9%) showed an agonist response which could be partly blocked (60%) by the AMPA receptor antagonist 2,3dihydroxy-6-nitro-7-sulfamoyl-benzo(F)-quinoxaline (NBQX)41 (5 µM) and fully blocked (100%) by NBQX (20 µM). This response probably reflects the agonist effect induced by a contamination of (S)-ACPA (0.05%). As compared to (S)- and (R)-ACPA, the demethylated analogs showed a more complex electrophysiological profile. Both of these enantiomers induced agonist responses, (S)demethyl-ACPA being the more potent agonist (EC50 = 12 µM) as compared to the R-enantiomer (EC50 = 108 µM). The responses elicited by (S)-demethyl-ACPA (15 µM) and by (R)-demethyl-ACPA (150 µM) could be partly blocked (70%) by NBQX (5 µM) and fully blocked (100%) by NBQX (20 µM). However, the stereoselectivity observed for the demethyl-ACPA enantiomers based on electrophysiological data (eudismic ratio of approximately 10) was markedly lower than that observed for [3H]AMPA binding affinity (eudismic ratio of approximately 250). Besides being an agonist, (R)-demethyl-ACPA was also able to block NMDA (10 µM) responses in the cortical wedge preparation (IC50 = 220 µM). As depicted in Figure 5, NBQX (25 µM) did not affect the response elicited by NMDA (10 µM), but completely blocked the response induced by (R)-demethylACPA (200 µM). When co-applying NBQX and (R)demethyl-ACPA together with NMDA, less than half of the NMDA response remained, thus demonstrating the antagonist properties of (R)-demethyl-ACPA. A control NMDA response showed that the NMDA receptor block induced by (R)-demethyl-ACPA was reversible. In order to investigate the pharmacological profile at mGlu receptors (Table 2) all compounds were initially tested for agonist and antagonist activity at 1 mM concentration on CHO cells stably expressing mGlu1␣, mGlu2, or mGlu4a receptors representing the three groups of receptors. Of the compounds tested in this study, only demethylACPA and (S)-demethyl-ACPA were active, showing antagonist activities at the mGlu2 receptor subtype. Antagonist potencies were determined by rightward-shifting the concentration–response curve of Glu by 1 mM demethylACPA or (S)-demethyl-ACPA, as exemplified for (S)demethyl-ACPA in Figure 6. Using the Gaddum equation, KB values of 477 µM and 148 µM were determined for
530
JOHANSEN ET AL.
TABLE 2. Antagonist activities at cloned mGlu receptors expressed in CHO cellsa KB (µM)b
AMPA Demethyl-AMPA ACPA (S)-ACPA (R)-ACPA Demethyl-ACPA (S)-Demethyl-ACPA (R)-Demethyl-ACPA
mGlu1␣
mGlu2
mGlu4a
>1,000 >1,000 >1,000 >1,000 >1,000 >1,000 >1,000 >1,000
>1,000 >1,000 >1,000 >1,000 >1,000 477 ± 94 148 ± 0 >1,000
>1,000 >1,000 >1,000 >1,000 >1,000 >1,000 >1,000 >1,000
a All compounds were inactive at the three receptor subtypes when tested for agonist activity at 1 mM concentration. b KB values ± SEM of at least two independent experiments.
demethyl-ACPA and (S)-demethyl-ACPA, respectively (Table 2). DISCUSSION
In recent years, a large number of Glu analogs have been reported where the structure of Glu has been modified using bioisosteric principles in order to identify potent and subtype-selective iGlu as well as mGlu receptor ligands as tools for the studies of receptor activation, inhibition, and modulation.1,2,5 In this area of research we have previously reported the isoxazole-containing homolog of Glu, ACPA, as a very potent AMPA receptor agonist, which interacts with AMPA receptors in a manner different from that of AMPA.9–11 Therefore, we decided to resolve and to study the enantiopharmacology of ACPA at iGlu and mGlu receptors. Electrophysiological characterization in cortical neurons revealed that the potent agonist activity of ACPA resides exclusively in the S-enantiomer (EC50 = 0.25 µM) (Table 1).
Fig. 6. Concentration–response curves for Glu in the absence (䊉) or presence (䊏) of 1 mM (S)-demethyl-ACPA at mGlu2-expressing CHO cells. Cells were preincubated with PBS containing 1 mM IBMX for 20 min and then incubated in the same buffer containing 10 µM forskolin and varying concentrations of Glu for 10 min. To antagonist assays were added a fixed concentration of 1 mM (S)-demethyl-ACPA. Data are the means (±SD) of a representative experiment performed in triplicate.
A similar high degree of stereoselectivity is typically seen for compounds acting at AMPA receptors, including Glu and AMPA and analogs and homologs of AMPA.8,25,38,42 The dual affinities observed for ACPA towards AMPA and KA receptors could not be segregated by the resolution, as the S-enantiomer of ACPA was found to be responsible for both receptor affinities. The demethylated analog of ACPA, demethyl-ACPA, turned out to possess a more complex pharmacological profile as compared to the parent compound. In addition to an antagonist effect at cloned mGlu2 receptors, demethylACPA showed affinity for all three classes of iGlu receptors and possessed AMPA receptor agonist activity as well as NMDA receptor antagonist effect (Tables 1, 2, Fig. 5). Compared to the very high degree of stereoselectivity observed for ACPA, demethyl-ACPA showed a relatively high eudismic ratio of 250 calculated for AMPA binding affinity, but only a low eudismic ratio of 10 concerning functional data. Furthermore, (S)-demethyl-ACPA was almost equipotent with (S)-ACPA in the [3H]AMPA binding assay, but showed a 50-fold lower EC50 value than (S)-ACPA in the cortical wedge preparation. Based on the present data, this discrepancy for (S)-demethyl-ACPA cannot be explained. However, a plausible explanation might be interaction(s) of (S)-demethyl-ACPA with excitatory amino acid transport systems present in the cortical tissue used in the electrophysiological studies, reducing the observed response of (S)-demethyl-ACPA in this test system. A similar and an even more pronounced discrepancy between AMPA receptor affinity and potency in cortical neurons has previously been reported6 for the structurally similar 5-unsubstituted isoxazole amino acid, demethyl-AMPA (Table 1). A comprehensive project on these aspects is in progress. In addition to its AMPA receptor agonist activity, (R)demethyl-ACPA, which has a carbon backbone one atom longer than that of Glu, turned out to be an NMDA receptor antagonist in the cortical wedge preparation. This antagonist effect of (R)-demethyl-ACPA is not surprising, considering a variety of homologs of Glu of which the Renantiomers have been shown to be NMDA antagonists.1 It is, however, interesting to notice that (R)-ACPA which has a methyl substituent in the 5-position of the isoxazole ring, does not exhibit this NMDA receptor antagonist effect. (S)-Demethyl-ACPA was the only enantiomer investigated which showed mGlu receptor interaction, being a relatively weak antagonist at the mGlu2 receptor subtype (KB = 148 µM) (Table 2). The majority of known selective mGlu2 antagonists do, however, have a Glu-like backbone chain length,43 whereas only a few examples of selective mGlu2 receptor antagonists with longer chain length are known.1,5 The difference in the pharmacology observed for ACPA, which is a highly stereoselective AMPA receptor agonist, and for demethyl-ACPA, which is a mixed Glu receptor ligand, apparently is related to the presence of the substituent in the 5-position of the isoxazole ring. The methyl groups of (S)-ACPA and of (S)-AMPA apparently direct the molecular flexibility of both molecules towards conformations that are almost exclusively recognized by the agonist
ACPA AND DEMETHYL-ACPA ENANTIOMERS
531
conformation(s) of the AMPA receptors. Alternatively, the steric hindrance introduced by these methyl groups prevents interactions with other Glu receptors.
12. Micetich RG. Studies in isoxazole chemistry. II. Isoxazoles from the ⌬2-isoxazolin-5-ols and their acetates. Can J Chem 1970;48:467–476.
ACKNOWLEDGMENTS
14. Blessing RH. Data reduction and error analysis for accurate single crystal diffraction intensities. Cryst Rev 1987;1:3–58.
The technical assistance of Mrs. K. Jørgensen and Mr. F. Hansen, Department of Chemistry, University of Copenhagen, with the recording of CD spectra and collection of X-ray data, and of Ms. H. Petersen and Mrs. A. Kristensen with the pharmacological characterization is gratefully acknowledged.
15. Blessing RH. DREADD—data reduction and error analysis for singlecrystal diffractiometer data. J Appl Cryst 1989;22:396–397.
Supporting Information Available
18. Sheldrick GM. Phase annealing in SHELX-90: direct methods for larger structures. Acta Crystallogr 1990;A46:467–473.
Tables for the compounds (−)-(R)-2-amino-3-(3-carboxy5-methyl-4-isoxazolyl)propionic acid [(−)-(R)-ACPA] and (−)-(S)-2-amino-3-(3-carboxy-4-isoxazolyl)propionic acid [(−)-(S)-demethyl-ACPA] listing crystal data and structure refinements (1 page), final atomic coordinates and equivalent isotropic displacement parameters (1 page), bond lengths and bond angles (2 pages), anisotropic displacement parameters for nonhydrogen atoms (1 page), atomic coordinates and fixed isotropic displacement parameters for hydrogen atoms (1 page), torsion angles (1 page), hydrogen bond dimensions (1 page), and lists of structure factors (9 pages) are available upon request to the authors.
19. Sheldrick GM. SHELXL-97. Program for crystal structure refinement. Go ¨ttingen, Germany: University of Go ¨ttingen; 1997.
LITERATURE CITED 1. Bra¨uner-Osborne H, Egebjerg J, Nielsen EØ, Madsen U, KrogsgaardLarsen P. Ligands for glutamate receptors: design and therapeutic prospects. J Med Chem 2000;43:2609–2645. 2. Bleakman D, Lodge D. Neuropharmacology of AMPA and kainate receptors. Neuropharmacology 1998;37:1187–1204. 3. Kno ¨pfel T, Kuhn R, Allgeier H. Metabotropic glutamate receptors: novel targets for drug development. J Med Chem 1995;38:1417–1426. 4. Conn PJ, Pin J-P. Pharmacology and functions of metabotropic glutamate receptors. Annu Rev Pharmacol Toxicol 1997;37:205–237. 5. Schoepp DD, Jane DE, Monn JA. Pharmacological agents acting at subtypes of metabotropic glutamate receptors. Neuropharmacology 1999;38:1431–1476. 6. Sløk FA, Ebert B, Lang Y, Krogsgaard-Larsen P, Lenz SM, Madsen U. Excitatory amino-acid receptor agonists. Synthesis and pharmacology of analogues of 2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid. Eur J Med Chem 1997;32:329–338.
13. Enraf-Nonius. CAD-4 Software; v. 5.0. Delft, The Netherlands: EnrafNonius; 1989.
16. DeTitta GT. ABSORB: an absorption correction programme for crystals enclosed in capillaries with trapped mother liquor. J Appl Cryst 1985;18:75–79. 17. Sheldrick GM. SHELXS-97. Program for the solution of crystal structures. Go ¨ttingen, Germany: University of Go ¨ttingen; 1997.
20. Flack HD. On enantiomorph-polarity estimation. Acta Cryst 1983;A39: 876–881. 21. Wilson AJC, ed. International tables for crystallography. Dordrecht, The Netherlands: Kluwer Academc Publishers; 1995;Vol C: Tables 4.2.6.8 and 6.1.1.4. 22. Murphy DE, Schneider J, Boehm C, Lehmann J, Williams K. Binding of [3H]3-(2-carboxypiperazine-4-yl)propyl-1-phosphonic acid to rat membranes. A selective high affinity ligand for N-methyl-D-aspartate receptors. J Pharmacol Exp Ther 1987;240:778–783. 23. Honore´ T, Nielsen M. Complex structure of quisqualate-sensitive glutamate receptors in rat cortex. Neurosci Lett 1985;54:27–32. 24. Braitman DJ, Coyle JT. Inhibition of [3H]kainic acid receptor binding by divalent cations correlates with ion affinity for the calcium channel. Neuropharmacology 1987;26:1247–1251. 25. Stensbøl TB, Borre L, Johansen TN, Egebjerg J, Madsen U, Ebert B, Krogsgaard-Larsen P. Resolution, absolute stereochemistry and molecular pharmacology of the enantiomers of ATPA. Eur J Pharmacol 1999;380:153–162. 26. Harrison NL, Simmonds MA. Quantitative studies on some antagonists of N-methyl-D-aspartate in slices of rat cerebral cortex. Br J Pharmacol 1985;84:381–391. 27. Madsen U, Frølund B, Lund TM, Ebert B, Krogsgaard-Larsen P. Design, synthesis and pharmacology of model compounds for indirect elucidation of the topography of AMPA receptor sites. Eur J Med Chem 1993;28:791–800. 28. Aramori I, Nakanishi S. Signal transduction and pharmacological characteristics of a metabotropic glutamate receptor, mGluR1, in transfected CHO cells. Neuron 1992;8:757–765. 29. Tanabe Y, Masu M, Ishii T, Shigemoto R, Nakanishi S. A family of metabotropic glutamate receptors. Neuron 1992;8:169–179.
7. Bang-Andersen B, Lenz SM, Skjærbæk N, Søby KK, Hansen HO, Ebert B, Bøgesø KP, Krogsgaard-Larsen P. Heteroaryl analogues of AMPA. Synthesis and quantitative structure-activity relationships. J Med Chem 1997;40:2831–2842.
30. Tanabe Y, Nomura A, Masu M, Shigemoto R, Mizuno N, Nakanishi S. Signal transduction, pharmacological properties and expression patterns of two rat metabotropic glutamate receptors, mGluR3 and mGluR4. J Neurosci 1993;13:1372–1378.
8. Falch E, Brehm L, Mikkelsen I, Johansen TN, Skjærbæk N, Nielsen B, Stensbøl TB, Ebert B, Krogsgaard-Larsen P. Heteroaryl analogues of AMPA. 2. Synthesis, absolute stereochemistry, photochemistry, and structure-activity relationships. J Med Chem 1998;41:2513–2523.
31. Hayashi Y, Tanabe Y, Aramori I, Masu M, Shimamoto K, Ohfune Y, Nakanishi S. Agonist analysis of 2-(carboxycyclopropyl)glycine isomers for cloned metabotropic glutamate receptor subtypes expressed in Chinese hamster ovary cells. Br J Pharmacol 1992;107:539–543.
9. Madsen U, Wong EHF. Heterocyclic excitatory amino acids. Synthesis and biological activity of novel analogues of AMPA. J Med Chem 1992; 35:107–111.
32. Hayashi Y, Sekiyama N, Nakanishi S, Jane DE, Sunter DC, Birse EF, Udvarhelyi PM, Watkins JC. Analysis of agonist and antagonist activities of phenylglycine derivatives for different cloned metabotropic glutamate receptor subtypes. J Neurosci 1994;14:3370–3377.
10. Wahl P, Madsen U, Banke T, Krogsgaard-Larsen P, Schousboe A. Different characteristics of AMPA receptor agonists acting at AMPA receptors expressed in Xenopus oocytes. Eur J Pharmacol 1996;308: 211–218. 11. Stensbøl TB, Sløk FA, Trometer J, Hurt S, Ebert B, Kjøller C, Egebjerg J, Madsen U, Diemer NH, Krogsgaard-Larsen P. Characterization of a new AMPA receptor radioligand, [3H]2-amino-3-(3-carboxy-5-methyl-4isoxazolyl)propionic acid. Eur J Pharmacol 1999;373:251–262.
33. Lazareno S, Birdsall NJM. Estimation of antagonist Kb from inhibition curves in functional experiments: alternatives to the Cheng-Prusoff equation. Trends Pharmacol Sci 1993;14:237–239. 34. Skinner GS. Deaminization. III. Evidence of the existence of aliphatic diazonium salts from the formation of chloro-oximino compounds. J Am Chem Soc 1924;46:731–741. ˆ i N, Kitahara, H, Kira R. Direct separation of enantiomers by high35. O
532
36.
37.
38.
39.
JOHANSEN ET AL.
performance liquid chromatography on a new chiral ligand-exchange phase. J Chromatogr 1992;592:291–296. Johnson CK. ORTEP II Report ORNL-5138. A Fortran thermal-ellipsoid plot program for crystal structure illustrations. Oak Ridge, TN: Oak Ridge National Laboratory; 1976. Legrand M, Rougier MJ. Application of the optical activity to stereochemical determinations. In: Kaagan HB, editor. Stereochemistry. Fundamentals and methods, vol. 2. Determination of configurations by dipole moment, CD or ORD. Stuttgart: Georg Thieme Publishers; 1997. p 33–183. Johansen TN, Ebert B, Bra¨uner-Osborne H, Didriksen M, Christensen IT, Søby KK, Madsen U, Krogsgaard-Larsen P, Brehm L. Excitatory amino acid receptor ligands: resolution, absolute stereochemistry, and enantiopharmacology of 2-amino-3-(4-butyl-3-hydroxyisoxazole-5yl)propionic acid. J Med Chem 1998;41:930–939. Gu ¨ bitz G. Direct separation of enantiomers by high performance li-
gand exchange chromatography on chemically bonded chiral phases. J Liq Chromatogr 1986;9:519–535. 40. Shinbo T, Yamagushi T, Nishimura K, Sugiura M. Chromatographic separation of racemic amino acids by use of chiral crown ether-coated reversed-phase packings. J Chromatogr 1987;405:145–153. 41. Sheardown MJ, Nielsen EØ, Hansen AJ, Jacobsen P, Honore´ T. 2,3Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline: a neuroprotectant for cerebral ischemia. Science 1990;247:571–574. 42. Hansen JJ. Stereoselectivity of excitatory amino acid receptors: enzymatic resolution of AMPA receptor agonists. Curr Top Med Chem 1993;1:377–385. 43. Kingston AE, Ornstein PL, Wright RA, Johnson BG, Mayne NG, Burnett JP, Belagaje R, Wu S, Schoepp DD. LY341495 is a nanomolar potent and selective antagonist of group II metabotropic glutamate receptors. Neuropharmacology 1998;37:1–12.
