Abstract
The in vitro and in vivo pharmacological profile of SB 223412 [(S)-(−)-N-(α-ethylbenzyl)-3-hydroxy-2-phenylquinoline-4-carboxamide], a novel human NK-3 (hNK-3) receptor antagonist, is described. SB 223412 demonstrated enantioselective affinity for inhibition of [125I][MePhe7]neurokinin B (NKB) binding to membranes of CHO cells expressing the hNK-3 receptor (CHO hNK-3). SB 223412, the (S)-isomer, (K i = 1.0 nM), has similar affinity as the natural ligand, NKB (K i = 0.8 nM) and another nonpeptide NK-3 receptor antagonist, SR 142801 (K i = 1.2 nM). SB 223412 was selective for hNK-3 receptors compared with hNK-1 (>10,000-fold selective) and hNK-2 receptors (>140-fold selective), and selectivity was further demonstrated by its lack of effect, in concentrations up to 1 or 10 μM, in >60 receptor, enzyme and ion channel assays. SB 223412 enantioselectively inhibited the NKB-induced Ca++mobilization in HEK 293 cells stably expressing the hNK-3 receptor. SB 223412 (10–1,000 nM) produced concentration-dependent rightward shifts in NKB-induced Ca++ mobilization concentration-response curves with a K b value of 3 nM. In addition, SB 223412 antagonized senktide-induced contraction in the isolated rabbit iris sphincter muscle (K b = 1.6 nM). In mice, oral administration of SB 223412 produced dose-dependent inhibition of behavioral responses induced by the NK-3 receptor-selective agonist, senktide (ED50 = 12.2 mg/kg). Pharmacokinetic evaluation of SB 223412 in rat and dog indicated low plasma clearance, oral bioavailability and high and sustained plasma concentrations after 4 to 8 mg/kg oral dosages. The preclinical profile of SB 223412 (high affinity, selectivity, reversibility and oral activity) suggests that it will be a useful tool compound to define the physiological and pathophysiological roles of NK-3 receptors.
The mammalian tachykinins, also known as neurokinins, are a family of small peptides that share the common carboxyl-terminal region Phe-X-Gly-Leu-Met-NH2; the main members are substance P, NKA and NKB (Maggio, 1988; Maggi et al., 1993). The tachykinins are differentially distributed in both the central and peripheral nervous systems, with a prominent location in the peripheral endings of capsaicin-sensitive primary afferent neurons (unmyelinated C-fibers) that innervate many sites, notably the airways, gastrointestinal and urinary tracts and skin (Holzer, 1988; Maggi, 1996; Maggi et al., 1995; Otsuka and Yoshioka, 1993).
The diverse biological effects of the tachykinins are mediatedvia three known human tachykinin receptor subtypes, designated NK-1, NK-2 and NK-3, which are members of the superfamily of G protein-coupled, seven-transmembrane-spanning receptors (Maggio, 1988; Maggi et al., 1993; Nakanishi, 1991; Regoli et al., 1988). The three human tachykinin receptors have been cloned and expressed (Buell et al., 1992; Gerard et al., 1990, 1991; Huang et al., 1992). The endogenous tachykinin ligands interact with all tachykinin receptors, although there is a defined agonist rank order of potency (e.g., for NK-3, it is NKB > NKA ≥ substance P) such that substance P, NKA and NKB have the highest affinities for the NK-1, NK-2 and NK-3 receptors, respectively.
Several potent and selective, nonpeptide receptor antagonists for the NK-1 and NK-2 receptors have been identified recently (Desai et al., 1992; Emonds-Alt et al., 1992; McLean et al., 1993; Snider et al., 1991), and these compounds have assisted significantly in the investigation and clarification of the pathophysiological roles of these receptors and the potential therapeutic utility of NK-1 and NK-2 receptor antagonists (Ishizukaet al., 1995; Lowe and Snider, 1993; Mantyh et al., 1994; Walsh et al., 1995). Much less is known about the biology and pathophysiological significance of the NK-3 receptor, in large part because of the lack of potent and selective antagonists. Recently, “peptoid” and peptide-derived NK-3 receptor antagonists were identified (Boden et al., 1994, 1995), and SR 142801, (S)-(+)-N-{{3-[1-benzoyl-3-(3,4dichlorophenyl)piperidin-3-yl]prop-1-yl}-4-phenylpiperidin4-yl}-N-methylacetamide, was reported as the first potent and selective, nonpeptide NK-3 receptor antagonist (Emonds-Alt et al., 1995; Oury-Donatet al., 1995). In this study, the pharmacological and pharmacokinetic profile of a novel NK-3 receptor antagonist, SB 223412, (S)-(−)-N(α-ethylbenzyl)-3-hydroxy-2-phenylquinoline-4-car-boxamide (fig. 1), is described. SB 223412 is a member of a new class of potent, competitive and selective nonpeptide NK-3 receptor antagonists that are based on the 2-phenylquinoline backbone (Giardinaet al., 1996).
Experimental Procedures
Materials.
[125I], [MePhe7]NKB (specific activity, 2200 Ci/mmol), [125I]NKA (specific activity, 2200 Ci/mmol) and [3H]substance P (specific activity, 34 Ci/mmol) were obtained from New England Nuclear Research Products (Boston, MA). The tachykinin peptides NKA, NKB, substance P and [MePhe7]NKB were purchased from Peninsula Laboratories (Belmont, CA), and senktide [succinyl-[Asp9-MePhe8]SP(6–13)] from California Peptide Research (Napa, CA). SR 142801, SB223412 isomers and racemate, CP 99994, [(+)-(2S,3S)-cis-(2-methoxybenzylamino)-2-phenylpiperidine dihydrochloride] were synthesized in the Department of Medicinal Chemistry, SmithKline Beecham Pharmaceuticals, Milan, Italy.
Receptor cloning and expression.
Human cDNAs for the NK-1, NK-2 and NK-3 tachykinin receptors, with sequences identical to published reports, were isolated from human placenta poly(A)+ RNA using reverse transcriptase-PCR technology and site-directed mutagenesis. Oligonucleotide primers [5′-CTAGCTTCGAA ATG GATAACGTCCTCCCGGT-3′ and 5′-AAAGGCCCTGTGGC CTA GGAGAGCACATTGG-3′ for human NK1 receptor cDNA (Gerard et al., 1991), 5′-GCAGCC ATG GGGACCTGTGACATTGTGACT-3′ and 5′-AACACTGCCACATTGGGA TCA AATTTCAAC-3′ for human NK2 receptor cDNA (Gerard et al., 1990) and 5′-GGCG ATG GCCATCCTCCCAGCAGCAGAAACCT-3′ and 5′-TACCTCAGGAAATGGAA T TA AGAATATTC-3′ for human NK3 receptor cDNA (Buell et al., 1992; Huanget al., 1992) (the translation start and stop codons areunderlined)] were synthesized and used for PCR using the human placenta cDNA as template. The individual fragments were subcloned into the mammalian expression vector, pCDN (Aiyar et al., 1994), and the resulting constructs were completely sequenced to confirm their identity and orientation. CHO stable cell lines for the pCDN-NK1, pCDN-NK2 and pCDN-NK3 expression vectors were obtained by electroporation followed by clonal selection using G418. Stable cell lines of these same vectors were also generated in HEK 293 cells using calcium phosphate precipitation for DNA transfection. The permanently transfected cell lines were obtained by selection with G418. The CHO and HEK 293 stable cell lines were screened for high-level receptor expression by ligand binding assays on whole cells. From this screen, the clonal cell line producing the highest number of receptors per cell was chosen for each receptor.
Radioligand binding assays.
Receptor binding assays were performed with crude membranes from CHO cells expressing the NK-1, NK-2 and NK-3 receptors. The cells were cultured at 37°C in a humidified incubator under 5% CO2/95% air in 1017 SO3 (proprietary in-house formulation) media containing nucleosides plus geneticin (400 mg/liter). The cells were harvested by centrifugation at 600 ×g for 10 min. The cell pellet was resuspended in hypotonic buffer (10 mM Tris, pH 7.4, 1.0 mM EDTA, 10 μg/ml soybean trypsin inhibitor, 100 μg/ml bacitracin, 100 μM benzamidine and 10 μM phenylmethylsulfonyl fluoride) and then rapidly frozen and thawed (three times), followed by Dounce homogenization for preparation of crude membranes. For NK-3 receptor competition binding studies, [125I][MePhe7]NKB binding to CHO hNK-3 membranes was performed according to Sadowski et al. (1993). Briefly, membranes (∼15 μg of protein) were incubated with 0.15 nM [125I][MePhe7]NKB in a total of 150 μl of 50 mM Tris, pH 7.4, 4 mM MnCl2, 1 μM phosphoramidon and 0.1% ovalbumin, with or without various concentrations of antagonist, for 90 min at 25°C. Incubations were stopped by rapid filtration with a Brandell tissue harvester (Gaithersburg, MD) through Whatman GF/C filters that were presoaked for 60 min in 0.5% BSA. Membranes were washed with 10 ml of ice-cold 20 mM Tris, pH 7.4, containing 0.1% BSA and then placed into vials with 10 ml of Beckman Ready Safe and counted in a Beckman LS 6000 (Fullerton, CA, USA) liquid scintillation counter. Concentration-response curves for each compound were run using duplicate samples in at least three independent experiments. Specific binding was determined by subtracting total binding from nonspecific binding, which was assessed as the binding in the presence of 0.5 μM cold [MePhe7]NKB. Percent inhibition of specific binding was determined for each concentration of compound and the IC50 value, defined as the concentration required to inhibit 50% of the specific binding, obtained from concentration-response curves. Values presented are the apparent inhibition constant (Ki ), which was calculated from the IC50 according to Cheng and Prusoff (1984).
NK-3 binding assays were also performed using brain tissue from male Hartley guinea pigs (450–650 g, Hazelton Research Animals, Denver, PA) and Sprague-Dawley rats (250–350 g, Charles River Breeding Laboratories, Kingston, NY). Crude membranes were prepared from brain cortex tissue through homogenization and centrifugation. [125I][MePhe7]NKB binding to the membranes was done as described above for the CHO hNK-3 membranes using ∼50 μg of membrane protein.
For NK-2 competition binding studies, [125I]NKA binding to membranes of CHO cells stably expressing the NK-2 receptor (CHO hNK-2) was performed essentially as described by Aharony et al. (1992). Cells were grown and membranes were prepared as described above for the hNK-3 binding assay. The assay buffer was the same as used for hNK-3 binding assay with a total volume of 150 μl, ∼10 μg of membrane protein and 0.15 nM [125I]NKA, which was incubated for 90 min at 25°C, with various concentrations of antagonist. Nonspecific binding was determined in the presence of 0.5 μM cold NKA. Filtration was through Whatman GF/C filters soaked for 30 min in 0.1% polyethylenimine, and membranes were washed as described above. The Ki value was determined as described for the NK-3 receptor assay.
Competition binding studies for the NK-1 receptor were performed on membranes of CHO cells stably expressing the human NK-1 receptor (CHO hNK-1) essentially as described by Payan et al. (1986). The assay volume was 300 μl, and the assay buffer (25 mM Tris, pH 7.4, 2.0 mM CaCl2, 2.0 mM MgCl2, 1 μM phosphoramidon and 0.1% ovalbumin) contained various concentrations of antagonist and 1.0 nM [3H]substance P. Membranes were incubated for 45 min at 25°C, and nonspecific binding was determined in presence of 1 μM cold substance P. Whatman filters were presoaked with BSA and membranes were washed as described for the hNK-3 binding assay. The Ki value was determined as described for the NK-3 receptor assay.
Calcium mobilization assay.
The cellular functional assay used to assess agonist/antagonist activity of test compounds was NKB-induced Ca++ mobilization in HEK 293 cells stably expressing the hNK-3 receptor (HEK 293 hNK-3). Cells were grown to ∼80% confluency in T-150 flasks and washed with phosphate-buffered saline. Cells were knocked loose from the flasks, suspended at 106 cells/ml in KRH (118 mM NaCl, 4.6 mM KCl, 25 mM NaHCO3, 1 mM KH2PO4 and 11 mM glucose) containing 50 mM HEPES, pH 7.4, 1 mM CaCl2, 1 mM MgCl2, 0.1% BSA and 2 μM Fura-2/AM and incubated for 45 min at 37°C. Cells were centrifuged at 200 × g for 3 min and resuspended in the same buffer without Fura-2/AM, incubated for 15 min at 37°C to complete the hydrolysis of intracellular Fura-2/AM and then centrifuged as before. Cells (5 × 105cells/ml) were resuspended in cold KRH with 50 mM HEPES, pH 7.4, 1 mM CaCl2, 1 mM MgCl2 and 0.1% gelatin and maintained on ice until assayed. For antagonist studies, aliquots (2 ml) of cells were prewarmed at 37°C for 5 min in 3-ml plastic cuvettes, and fluorescence was measured with a fluorometer (Johnson Foundation Biomedical Group, Philadelphia, PA) with magnetic stirring and temperature maintained at 37°C. Excitation was set at 340 nm, and emission was set at 510 nm. Various concentrations of antagonists or vehicle were added, and fluorescence was monitored for ∼15 sec to ensure that there was no change in base-line fluorescence, followed by the addition of 1 nM NKB. An exception was SR 142801, which required pretreatment to obtain maximal inhibitory activity; therefore, all Ca++ studies with this compound had a 5-min pretreatment at 37°C. Maximal Ca++ levels attained after agonist stimulation was calculated as described by Grynkiewicz et al. (1985). The percentage of maximal NKB-induced Ca++mobilization was determined for each concentration of antagonist and the IC50, defined as the concentration of test compound that inhibits 50% of the maximal 1 nM NKB response, obtained from the concentration-response curve (five to seven concentrations of antagonists). Values presented are the mean ± S.E.M. IC50 value of at least three individual experiments.
Senktide-induced contraction in isolated rabbit iris sphincter muscle.
Because the isolated rabbit iris sphincter muscle preparation contains functional NK-3 receptors (Hall et al., 1993; Medhurst et al., 1997), the effects of SB 223412 were investigated in this tissue. Iris sphincter muscle strips were prepared from male New Zealand White rabbits (2–3 kg, Charles River, Margate, UK) that were killed with intravenous pentobarbitone. Tissues were placed in 50-ml organ baths containing Krebs-Henseleit solution (118 mM NaCl, 5.4 mM KCl, 25 mM NaHCO3, 1 mM NaH2PO4·2H20, 2.5 mM CaCl2, 0.7 mM MgSO4·7H2O and 11 mM glucose) for the isometric measurement of tension as previously described (Medhurst et al., 1996). After a reference contractile response to 10 μM carbachol was obtained, experiments were conducted in the presence of 1 μM CP 99994 and 1 μM atropine. Tissues were then exposed to SB 223412 (10 nM) or vehicle (DMSO) for 120 min before cumulative concentration-effect curves to senktide were determined. Responses to senktide were expressed as a percentage of the carbachol-induced contraction. The dissociation constant,Kb , for the antagonist-NK-3 receptor complex was calculated from the equation:Kb = [B]/CR − 1, where CR is the concentration ratio of agonist used in the presence and absence of antagonist B.
Senktide-induced behavioral activity.
Male BALB/c inbred mice (six mice per group) from Charles River Breeding Laboratories (Raleigh, NC) were orally administered various concentrations of SB 223412, prepared in 50% PEG-400/1% methylcellulose, or vehicle alone. Thirty min later the mice were challenged with senktide (1.0 mg/kg s.c.), head twitches and/or tail whips were counted over 10 min and the mean for the group was determined. The ED50 was calculated from the concentration-response curve using regression analysis software.
Pharmacokinetic studies in rat and dog.
Bioavailability evaluations were carried out in rat and dog using crossover experimental designs. Indwelling femoral vein (for drug infusion) and artery catheters (for blood sampling) were placed in male Sprague-Dawley rats (300–400 g, n = 3) under ketamine/xylazine anesthesia 1 week before the studies. Blood samples were collected at various times over 24 hr after dosing. Plasma was prepared by centrifugation and stored at −30°C until analysis. Concentrations of SB 223412 in plasma were measured by quantitative LC/MS/MS analysis. Positive ion multiple reaction monitoring was used for MS/MS detection of SB 223412 and an internal standard. The molecular ion of SB 223412 ([MH]+, m/z 383) was selected by the first quadrupole filter, bombarded with argon in the second quadrupole to generate fragment ions, one of which (m/z 248) was selectively monitored in the third quadrupole and detected by an electron multiplier. The assay had a lower limit of quantification of 10 ng/ml using 50 μl of plasma. Systemic plasma clearance was determined after the intravenous infusion of 3 mg/kg. The oral bioavailability of SB 223412 in solution (8 mg/kg at 2 mg/ml in PEG-400) administered to the same rats (fasted) 5 days later was calculated using noncompartmental pharmacokinetic methods (Rowland and Tozer, 1995).
For studies in dogs, catheters were placed in a saphenous vein for intravenous infusion and a cephalic vein for blood sampling of male beagle dogs (13–15 kg, n = 3) on each study day. Systemic plasma clearance was determined after the intravenous infusion of 1 mg/kg. The oral bioavailability of SB 223412 in solution (4 mg/kg at 15 mg/ml in PEG-400) administered in gelatin capsules to the same dogs 1 week later was calculated using noncompartmental pharmacokinetic methods (Rowland and Tozer, 1995). The dogs were fasted overnight before each dose administration and were allowed access to food ∼4 hr after dosing.
Central nervous system penetration studies were performed by intravenous infusion of SB 223412 to rats (n = 3) for 6 hr at 1 mg/kg/hr to approach steady state conditions. Blood samples were collected at 30-min intervals during the final 2 hr of the infusion. Immediately on completion of the infusion, the animals were killed, and the brain tissue was harvested and then homogenized in saline. Plasma and brain tissue homogenate samples were stored at −30°C until analysis. Concentrations of SB 223412 in plasma and brain tissue homogenate at the end of the infusion were determined by LC/MS/MS analysis as described above.
Results
In Vitro Activity
Binding studies.
Binding of [125I][MePhe7]NKB to membranes prepared from suspended CHO cells stably expressing the human NK-3 receptor (CHO hNK-3 cells) is saturable, specific and of high affinity. The apparent dissociation constant (Kd ) was 0.61 ± 0.10 nM, and the maximum number of binding sites (B max) was 1006 ± 71 fmol/mg of protein (n = 3). Competition of [125I][MePhe7]NKB binding to CHO hNK-3 cell membranes by [MePhe7]NKB, NKB, senktide, NKA and substance P is presented in figure 2 and table1. The inhibition constants (Ki values), determined from IC50 values of concentration response curves, were as expected for the rank order of potency for tachykinin agonists at the hNK-3 receptor (Sadowski et al., 1993).
The active (S)-enantiomer SB 223412 inhibited the binding of [125I][MePhe7]NKB to CHO hNK-3 cell membranes with a Ki value of 1.0 ± 0.1 nM (n = 10), whereas for the racemate and the inactive (R)-isomer, SB 223411, theKi values were 2.5 ± 0.1 nM (n = 4) and 161 ± 29 nM (n = 3), respectively (fig. 3). The other reported nonpeptide NK-3 receptor antagonist, SR 142801 (Emonds-Alt et al., 1995), has similar affinity as SB 223412, with aKi value of 1.2 ± 0.2 nM (n = 4) in this assay (fig. 3). Thus, the affinity of both antagonists is similar to that of the natural ligand, NKB (Ki = 0.81 ± 0.21 nM,n = 3; table 1).
Saturation binding experiments in the presence and absence of 0.2 nM SB 223412 were performed to determine the competitive nature of the binding inhibition. As shown in figure 4, SB 223412 produced a decrease in affinity (Kd ) with no change in maximal binding, which is consistent with competitive antagonism. In the presence of 2 nM SB 223412, the meanKd value was 1.60 nM, and in the absence, it was 0.71 nM (n = 2).
SB 223412 was evaluated for affinity at the NK-3 receptor in rats and guinea pigs using [125I][MePhe7]NKB binding to brain cortical membrane preparations. SB 223412 demonstrated high affinity for the guinea pig receptor (Ki = 0.79 ± 0.19 nM,n = 3) but lower affinity for the rat receptor (Ki = 30.1 ± 6.9 nM,n = 3).
Selectivity profile.
Selectivity relative to other tachykinin receptors was assessed by competitive binding experiments using membranes prepared from CHO cells stably expressing the human NK-2 (CHO hNK-2) and human NK-1 (CHO hNK-1) cells and [125I]NKA and [3H]substance P, respectively. The results of competition binding experiments with CHO hNK-2 membranes are summarized in table 1. Analysis of the competition of [125I]NKA binding by NKA, SR 48968 (a potent NK-2 receptor antagonist; Emonds-Alt et al., 1992), SB 223412 and SR 142801 revealed subnanamolar affinity for NKA and SR 48968, whereas the NK-3 antagonists SB 223412 and SR 142801 had ∼140- and ∼33-fold selectivity for NK-3 vs.NK-2 receptors, respectively.
Investigation of the competition of [3H]substance P binding to CHO hNK-1 membranes by substance P, CP 99994 (a potent NK-1 receptor antagonist, McLean et al., 1993), SB 223412 and SR 142801 demonstrated high affinity for substance P and CP 99994, submicromolar affinity for SR 142801 and very low affinity for SB 223412 (Ki > 10,000 nM; table 1).
Selectivity of SB223412 was further assessed comprehensively by testing it in >60 receptor binding, enzyme and ion channel assays. SB 223412, at concentrations up to 1 or 10 μM, was without effect in this battery of assays, except for the peripheral benzodiazepine and N-formyl-Met Leu Phe (fMLP) receptor binding assays, in which it had IC50 values of ∼1 μM (n = 2) and −3.1 ± 0.1 μM (n = 3), respectively. It should be noted that these concentrations are ∼1000-fold higher than its affinity for the hNK-3 receptor.
Ca++ mobilization studies.
Cellular functional NK-3 receptor antagonist activity of compounds in Ca++ mobilization studies was assessed using the hNK-3 receptor stably transfected into HEK 293 cells (HEK 293 hNK-3), because the suspended CHO hNK-3 cells demonstrated unexpectedly high basal Ca++ levels. HEK 293 hNK-3 cells responded in a concentration-dependent manner to tachykinin agonists with Ca++ transients. Activity of the standard tachykinin agonists gave the expected rank order potency for hNK-3 receptor : [MePhe7]NKB = NKB > senktide > NKA = substance P, with EC50 values of 0.47 ± 0.16 nM (n = 4), 0.45 ± 0.10 (n = 4), 3.1 ± 1.0 nM (n = 3), 77 ± 30 nM (n = 3) and 91 ± 25 nM (n = 3), respectively (fig. 5A).
SB 223412 inhibited Ca++ mobilization induced by 1 nM NKB with an IC50 value of 16.6 ± 1.6 nM (n = 7), whereas the inactive (R)-isomer, SB 223411, had an IC50 value of 780 nM (n = 2) and the racemate had an IC50 value of 38.6 ± 7.9 nM (n = 3; fig. 5B). SR 142801, the other nonpeptide NK-3 receptor antagonist, required pretreatment with the cells to obtain maximal inhibitory activity: with a 5-min pretreatment at 37°C, the IC50 value was 6.1 ± 1.4 nM (n = 4; fig. 5B).
The cellular functional NK-3 receptor antagonist activity of SB 223412 was not time dependent [i.e., the inhibition was identical with 10-sec (IC50 = 13 nM) or 5-min (IC50 = 12 nM) pretreatment with antagonist]. Furthermore, inhibition of the Ca++ response was rapidly reversible, because treatment with varying concentrations of SB 223412 for 5 min followed by rapid centrifugation and resuspension in original buffer (IC50 = 11 nM) or fresh buffer without antagonist (IC50 >1000 nM) results in significant loss of inhibitory activity.
The Ca++ cellular functional assay was also used to investigate the competitive nature of the inhibitory activity of SB 223412. SB 223412 produced a concentration-dependent inhibition of NKB-induced Ca++ mobilization, resulting in rightward parallel shifts in the NKB concentration response curves (fig.6). Schild analysis of the data revealed a meanKb value of 3 nM (n = 2) and a slope of the regression line that was not significantly different from 1 (0.9), which is consistent with competitive antagonism.
Senktide-induced contraction in isolated rabbit iris sphincter muscle.
Senktide was a potent contractile agonist in the isolated rabbit iris sphincter muscle with a pD 2 value of 9.1 ± 0.1 (n = 4). SB 223412 (10 nM) surmountably antagonized the contractile responses to senktide (fig. 7) with a Kb value of 1.6 ± 0.53 nM (n = 4).
In Vivo Activity
NK-3 receptor antagonist activity.
The NK-3-selective ligand senktide induces a characteristic set of behaviors in rodents that appear to be mediated by serotonin release in brain and spinal cord (Stoessl et. al., 1990). A mouse model of this phenomenon was developed to investigate the in vivo activity of SB 223412. Oral administration of SB 223412 (5–20 mg/kg in PEG-400/1% methylcellulose, 30-min pretreatment, 6 mice/group) produced a dose-dependent inhibition of senktide (1 mg/kg s.c.)-induced behavioral effects (rapid head shakes and tail whips, counted for 10 min after senktide) with an ED50 value of 12.2 mg/kg. For comparison, SR 142801 (5–15 mg/kg) administered orally in the same vehicle demonstrated dose-dependent inhibition with an ED50 value of 14.7 mg/kg.
Pharmacokinetic profile.
The pharmacokinetic profile of SB 223412, in PEG-400 solution, was assessed in rats and dogs after oral and intravenous (infusion) administration; the results are summarized in figure 8 and tables 2 and3. When administered by oral gavage to rats at a dose of 8 mg/kg, high and sustained plasma concentrations (in the μg/ml range) of SB 223412 were detected. Bioavailability was determined to be 62 ± 9% (n = 3) with a long half-life (t 1/2 = 218 ± 36 min) and a high Cmax value of 2.80 ± 0.54 μg/ml (fig. 8, table3).
Similarly, in the dog, after oral administration of 5 mg/kg SB 223412, high and sustained plasma levels were obtained. Oral bioavailability was determined to be high at 71 ± 7% with a Cmaxvalue of 8.0 ± 0.1 μg/ml and a half-life of 386 ± 92 min (n = 3; fig. 8, table 3). Secondary peaks in the plasma concentration-time profiles indicated the presence of enterohepatic recirculation in both species. Preliminary analytical assessment revealed no evidence of major circulating metabolites in either species.
After 6 hr of continuous intravenous infusion of SB 223412 to rats at 1 mg/kg/hr, brain tissue concentrations were 356 ± 99 ng/g (n = 3). Monitoring of plasma concentrations of SB 223412 over the last 2 hr of the infusion established that they were not changing appreciably, with a mean value 1750 ng/ml. Thus, the brain tissue/plasma concentration ratio was 0.20 ± 0.02.
Discussion
The findings of the present series of experiments indicate that SB 223412, a novel, nonpeptide NK-3 receptor antagonist, has the following features: (1) high affinity (Ki = 1 nM vs. the hNK-3 receptor), (2) selectivity (∼140-fold selectivity for hNK-3 vs. hNK-2 and >10,000-fold selectivity for hNK-3 vs. hNK-1 and no effect in concentrations up to 1 or 10 μM in >60 receptor, ion channel and enzyme assays), (3) oral activity vs. NK-3 receptor-induced behavioral effects in mouse and (4) good pharmacokinetic profile after oral administration in rats and dogs, with high, sustained plasma concentrations, low clearance, bioavailability of ∼60% to 70% and no evidence of a major circulating metabolite in both species.
Although the tachykinins and their receptors have been extensively studied for many years, the focus of research has been on the NK-1 receptor and its preferred ligand, substance P, and to a lesser extent the NK-2 receptor and NKA, with significantly less work conducted on the NK-3 receptor and NKB. The lower emphasis on NK-3 receptor research has been partly due to the lack of potent and selective nonpeptide NK-3 receptor antagonists with which to define more clearly the physiological and pathophysiological roles of the NK-3 receptor. NK-3 receptors have been demonstrated by biochemical, pharmacological and molecular biological techniques to be present in both the central nervous system and peripheral nervous system, where they may have a neuromodulatory role, influencing the release of various transmitters (Arenas et al., 1991; Ramirez et al.; 1994;Schemann and Kayser, 1991; Stoessl et al., 1990). Recent reports have presented information on the first nonpeptide NK-3 receptor antagonist, SR 142801 (Emonds-Alt et al., 1995;Nguyen-Le, et al., 1996; Oury-Donat et al., 1995). SR 142801 is a piperidine derivative that is structurally very similar to the NK-2 receptor antagonist SR 48968, which was shown to have moderate affinity for NK-3 receptors in guinea pig cerebral cortex membranes (IC50 = 320 nM; Petitet et al., 1993a) and hNK-3 receptors (table 1). SB 223412 is a member of a novel class of nonpeptide NK-3 receptor antagonists, structurally unrelated to SR 142801, that are based on the 2-phenylquinoline backbone (Giardinaet al., 1996). SB 223412 is a high-affinity antagonist on the basis of binding studies in CHO hNK-3 cell membranes, with aKi value identical to that obtained with SR 142801. The affinity of SR 142801 in the present receptor assay is lower than that previously reported (i.e., 0.2 nM; Emonds-Alt et el., 1995). Functional studies confirmed the high potency of SB 223412, with a Kb value of 3 nM demonstrated for inhibition of NKB-induced calcium mobilization in HEK-293 cells expressing the hNK-3 receptor.
In addition to high potency, an important component of the profile of a receptor antagonist is selectivity. Human tachykinin receptor binding studies indicated that SB 223412 has ∼140-fold selectivity for hNK-3 receptors vs. hNK-2 receptors and >100,000-fold selectivityvs. the hNK-1 receptor. Thus, SB 223412 has moderate affinity for the hNK-2 receptor and little or no affinity for the hNK-1 receptor. By way of comparison, SR 142801 had ∼35-fold selectivity for hNK-3 vs. hNK-2 and ∼625-fold selectivityvs. hNK-1. The selectivity of SB 223412 for NK-3 receptors was confirmed and highlighted by its lack of effect, in concentrations of 1 or 10 μM, in >60 separate receptor, enzyme and ion channel assays, including opioid receptors (mu, kappa anddelta) and sodium channel (site 2) for which SR 142801 was reported to have affinities in the 0.1 to 1 μM concentration range (Emonds-Alt et al., 1995). SB 223412 demonstrated some affinity for the benzodiazepine (peripheral) receptor and fMLP receptor, but the affinities were 1000-fold lower than that for the hNK-3 receptor. Therefore, overall SB 223412 is a very selective NK-3 receptor antagonist.
Species differences in the affinities of the nonpeptide NK-1 and NK-2 receptor antagonists have been demonstrated (Fong et al., 1992; Patacchini et al., 1991; Petitet et al., 1993b). Similarly, preliminary data in support of species effects for NK-3 receptors have been provided, with differences between the affinities of SR 48968 for rat and guinea pig receptors noted (Petitetet al., 1993a), and SR 142801 reported to possess much higher affinity (36–136-fold) for human, gerbil and guinea pig NK-3 receptors vs. rat NK-3 receptor (Chung et al., 1995; Emonds-Alt et al., 1995). In the present study, SB 223412 had similar affinity for human, rabbit and guinea pig NK-3 receptors and lower affinity (∼40-fold) for the rat NK-3 receptor. Several attempts were made to determine the affinity of the NK-3 receptor antagonists for the mouse NK-3 receptor because this was the species chosen for the in vivo pharmacological activity, but these were unsuccessful due to insufficient specific binding. Thus, for the NK-3 receptor it appears that the receptor antagonists, on the basis of the structural classes identified thus far, have a lower affinity for the rat receptor vs. the human and other animal receptors, including guinea pig and rabbit.
It has been reported previously that administration of NK-3 receptor agonists, such as senktide, to rodents produces a characteristic set of behavioral responses, including wet dog shakes, head shaking and tail flicks (Stoessl et al., 1988, 1990). The mechanism responsible for this phenomenon appears to be due, at least in part, to the release of 5-hydroxytryptamine from the central nervous system. We established this mouse central nervous system model and demonstrated the ability of oral administration of SB 223412 to inhibit senktide-induced head shakes and tail flicks. Thus, SB 223412 has oral activity against NK-3 receptor-induced central nervous system effects in mouse. Direct evidence for central nervous system penetration of SB 223412 was provided by disposition studies, involving infusion of the compound in the rat, which revealed a brain tissue concentration of 356 ng/g and a mean brain/plasma ratio of 0.20.
The pharmacokinetic characteristics of SB 223412 were assessed in two species, rats and dogs, after intravenous and oral administration. The profile of the compound in both species was similar, with the notable features being low systemic clearance, long t 1/2values and high oral bioavailability resulting in high, sustained plasma levels (μg/ml range) after oral administration. Secondary peaks in the plasma concentration-time profiles, apparently due to enterhepatic recirculation, contributed to long maintenance of high plasma concentrations. In addition, preliminary assessment indicated no evidence of a major circulating metabolite in either species.
It has been reported by Patacchini et al. (1995) that the functional effects of SR 142801 against NK-3-receptor induced contractions in isolated guinea pig ileum longitudinal muscle preparations are essentially irreversible, reflected by the lack of reversal of the effects by washing out for up to 2 hr, and time dependence (i.e., increasing blockade with longer incubation times); in addition, the antagonism of the responses by SR 142801 is insurmountable. In contrast, the present results, from binding and calcium mobilization studies, indicate that the inhibitory effects of SB 223412 are reversed by washout and are not time dependent. Furthermore, the antagonism produced by the compound is surmountable as demonstrated in the calcium mobilization and rabbit iris contraction studies.
In summary, the data indicate that SB 223412 is a high affinity, selective, reversible and competitive antagonist of hNK-3 receptors. It is orally active in an NK-3 receptor-induced central nervous system behavioral model in mouse. In addition, SB 223412 has a good pharmacokinetic profile. The preclinical pharmacological and pharmacokinetic profile of SB 223412 suggests that it will be a useful tool compound to assist in the elucidation of the physiological and pathophysiological roles of NK-3 receptor activation.
Acknowledgments
We thank John Adamou for assistance in cloning the human tachykinin receptors; Punam Sandhu, Michael Spengler and Frank Dixon for help in conducting of the pharmacokinetic studies; Mario Grugni, Roberto Rigolio and Karl F. Erhard for the synthesis of SR 142801 and Luca F. Raveglia for the preparation of CP 99994.
Footnotes
-
Send reprint requests to: Henry M. Sarau, Ph.D., SmithKline Beecham Pharmaceuticals, Department of Pulmonary Pharmacology (UW 2531), 709 Swedeland Road, King of Prussia, PA 19406. E-mail:Skip_Sarau-1{at}sbphrd.com
- Abbreviations:
- NK-1
- neurokinin 1
- NK-2
- neurokinin 2
- NK-3
- neurokinin 3
- CHO
- Chinese hamster ovary
- CHO hNK-3
- CHO cells stably expressing the human neurokinin -3 receptor
- CHO hNK-2
- CHO cells stably expressing the human neurokinin-2 receptor
- CHO hNK-1
- CHO cells expressing the human neurokinin -1 receptor
- HEK
- human embryonic kidney
- HEK 293 hNK-3
- HEK 293 cells stably expressing the human NK-3 receptor
- NKA
- neurokinin A
- NKB
- neurokinin B
- BSA
- bovine serum albumin
- KRH
- Krebs-Ringer-Henseleit
- EC50
- concentration of agonist producing 50% of maximal response
- IC50
- concentration of antagonist causing 50% inhibition of agonist response
- Ki
- apparent inhibition constant
- Kb
- dissociation constant
- PCR
- polymerase chain reaction
- LC/MS/MS
- liquid chromatography with triple quadrupole mass spectrometric detection
- Received September 18, 1996.
- Accepted February 14, 1997.
- The American Society for Pharmacology and Experimental Therapeutics