Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Neuromedin B (NMB), a decapeptide first isolated from porcine spinal cord (Minamino et al., 1983), shares similarity with gastrin-releasing peptide (GRP) and is a member of the bombesin (Bn) family of peptides. Previous studies revealed the presence of this peptide in the central nervous system, including the pituitary (Namba et al., 1985; Hearn et al., 1992), olfactory bulb (Moody et al., 1986), and spinal cord. NMB-containing fibers innervate the gastrointestinal tract, particularly the esophagus and rectum (Namba et al., 1985), and NMB is found in some small cell and non-small cell lung carcinoma cell lines (Ladenheim and Guss, 1995). The rodent (Wada et al., 1991), murine (Ohki-Hamazaki et al., 1997), and human (Corjay et al., 1991) NMB receptors (NMB-Rs) have been cloned, and their patterns of expression have been determined. Native NMB-R has been detected in rodent esophagus (von Schrenck et al., 1989) and bladder (Rouissi et al., 1991), in normal (DeMichele et al., 1994) and neoplastic (Corjay et al., 1991; Moody et al., 1992) human lung, and in the C6 rodent glioblastoma line (Wang et al., 1992). Like other Bn receptor subtypes, the NMB-R has a heptahelical topology and couples to G proteins to elicit its effects. NMB-R activation results in stimulation of phospholipases C (Wang et al., 1992), D (Hou et al., 1998), and A₂ (Ryan et al., 1996); increases in intracellular calcium concentration ([Ca²⁺]_i; (Moody et al., 1992); and tyrosine phosphorylation of p125 focal adhesion kinase (Tsuda et al., 1997). Although the potential physiological role of GRP has been a major focus of research, the role of NMB in physiological or pathophysiological processes has received much less attention. Some studies suggest that NMB may also play an important part in a numerous biological processes. In some tumor cells expressing NMB-Rs, receptor activation exerts trophic effects (Moody et al., 1992). Additional studies have demonstrated a modulatory role of NMB in suppression of feeding behavior and gastric emptying (Varga et al., 1995), control of the hypothalamic-pituitary-adrenocortical axis and thyrotropin release (Rettori et al., 1989), sensory transmission in the spinal cord (Cridland and Henry, 1992), excitation of serotonin neurons in the dorsal raphe nucleus (Pinnock et al., 1994), control of potassium secretion by the blood-brain barrier (Vigne et al., 1997), and bladder smooth muscle contractility (Rouissi et al., 1991), and NMB has been implicated in the stimulation of bronchial mucosal secretion (Ali et al., 1996). The development of selective, high-affinity NMB-R agonists and antagonists would enable a more precise definition of the role of NMB in various physiological or pharmacological processes.

In contrast to GRP receptor antagonists (Jensen and Coy, 1991), the discovery process for NMB-R antagonists has been slower. When strategies used to develop GRP receptor antagonists were applied to the synthesis of potential NMB antagonists, none yielded high-affinity NMB-R antagonists (Lin et al., 1995). It was subsequently discovered that an unrelated peptide, the cyclic somatostatin analog D-Nal,Cys,Tyr,D-Trp,Lys,Val,Cys,Nal-NH₂ (SS-octa) had a 100-fold selectivity for NMB-Rs over GRP receptors (Orbuch et al., 1993), and this class of agents has been used in a number of pharmacological studies (Ryan et al., 1996). However, nonpeptide NMB-R antagonists resistant to degradation in vivo would be ideal for use in biological systems. Two nonpeptide GRP antagonists, kuwanons G and H, have been described (Mihara et al., 1995). However, their affinities for GRP receptors are in the submicromolar range, with even lower affinities for NMB-Rs, making these reagents impractical in biological systems.

Recently, a peptoid based on a single amino acid template, PD 165929, was characterized as a high-affinity NMB-R antagonist (Eden et al., 1996). In the present study, we evaluated the pharmacology of a second-generation peptoid from this series, PD 168368, for all three mammalian Bn receptor subtypes and for the amphibian frog [Phe¹³]Bn (fBB₄) receptor. We also compared the pharmacological profile of PD 168368 with those of multiple Bn receptor subtypes in a number of different animal species because it was shown previously that some synthetic Bn/GRP analogs can act as antagonists or agonists on the same receptor subtype from different animal species (Wang et al., 1990) or on different receptor subtypes from the same animal (Ryan et al., 1996). Our results confirm that PD 168368 is a potent and selective antagonist for NMB-Rs regardless of origin, which could prove useful in understanding the biology and pharmacology of the NMB-R.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. NCI-N417 human small cell lung carcinoma cells, NCI-H1299 human non-small cell lung carcinoma cells, and AR42J cells were gifts from Herb Oie (National Cancer Institute, Navy Medical Oncology Branch, Naval Medical Center, Bethesda, MD). HuTu-80 duodenal adenocarcinoma cells, C6 rat glioblastoma cells, and CHO-K1 cells were purchased from American Type Culture Collection (Rockville, MD). RPMI-1640 medium, Dulbecco's modified Eagle's medium, Dulbecco's PBS, G418 sulfate, and FBS were obtained from Life Technologies, Inc. (Grand Island, NY). Soybean trypsin inhibitor was obtained from Sigma Chemical Co. (St. Louis, MO). Hydroxypropyl--cyclodextrin was obtained from Aldrich Chemical Co. (Milwaukee, WI). BSA fraction V was purchased from ICN Biomedicals Inc. (Aurora, OH). AG 1-X8 resin was obtained from Bio-Rad (Richmond, CA). myo-[2-³H]Inositol (20 Ci/mmol) was purchased from DuPont-New England Nuclear (Boston, MA). Bn, [Tyr⁴]Bn, NMB, and GRP were obtained from Bachem (Torrence, CA). [Phe¹³]Bn and SS-octa were gifts from John Taylor (Biomeasure, Inc., Milford, MA). PD 168368 was a gift from Robert Pinnock (Parke-Davis Neuroscience Research Center, Cambridge, UK). All other chemicals were reagent grade.

Cell Culture. NCI-H1299 cells transfected with human (h)NMB-Rs or hBRS-3 receptors (Mantey et al., 1997) and NCI-N417 cells were grown in RPMI-1640. AR42J, HuTu-80, C6, or BALB 3T3 cells transfected with human (Mantey et al., 1997), rat (Benya et al., 1992), or mouse (Benya et al., 1994) GRP or NMB-Rs were grown in Dulbecco's modified Eagle's medium. CHO-K1 cells transfected with fBB₄ (Pradhan et al., 1998) receptors were grown in Ham's F-12 medium. All cell media were supplemented with 10% (v/v) FBS, and 300 µg/ml G418 sulfate was included for the NCI-H1299, BALB 3T3, and CHO-K1 transfectants. All cell lines were incubated at 37°C in a 5% CO₂ atmosphere.

Creation of NCI-H1299 hNMB-R Transfectants. NCI-H1299 cells stably expressing hNMB-Rs were prepared by transfection methods previously described by workers in our laboratory (Benya et al., 1992; Mantey et al., 1997). Briefly, 15 µg of plasmid DNA containing an hNMB-R cDNA in the pCD2 expression vector (InVitrogen, San Diego, CA) (Benya et al., 1992) were transfected into the NCI-H1299 cells using LIPOFECTamine (Life Technologies, Gaithersburg, MD). Seventy-two hours after transfection, the cells were split 1:3 and 800 µg/ml G418 (Life Technologies, Gaithersburg, MD) was added to select for hNMB-R positive cells. Single colonies were isolated 2 weeks later and maintained in standard growth medium containing 300 µg/ml G418. Expression of hNMB-Rs in these cells was confirmed by reverse transcriptase-polymerase chain reaction and by binding of ¹²⁵I-[D-Tyr⁰]NMB. Four clones (3, 6, 11, and 12) with the best binding were tested for their selectivity for Bn-related peptides and gave similar results. One of these clones, hNMB-R-transfected NCI-H1299 clone 3, was used in the present study.

Preparation of Peptides and Radioligands. [D-Phe⁶]Bn(6-13) methyl ester, [D-Phe⁶,Ala¹¹,Phe¹³,Nle¹⁴]Bn(6-14), [D-Tyr⁶,Ala¹¹, Phe¹³,Nle¹⁴]Bn(6-14), and [D-Tyr⁰]NMB were synthesized by solid phase methods as described previously (Wang et al., 1990) and characterized by amino acid analysis and matrix-assisted laser desorption mass spectroscopy (Finnegan, Hemel Hemstead, UK). ¹²⁵I-[D-Tyr⁰]NMB, ¹²⁵I-[Tyr⁴]Bn, and ¹²⁵I-[D-Tyr⁶,Ala¹¹,Phe¹³, Nle¹⁴]Bn(6-14) were prepared using IODO-GEN as described previously (von Schrenck et al., 1989; Wang et al., 1992; Mantey et al., 1997). Before use in binding experiments or bioassays, PD 168368 was dissolved in either 0.9% saline (w/v) containing 50% (v/v) hydroxypropyl--cyclodextrin or 33% (v/v) dimethyl sulfoxide (DMSO)/deionized water or incubation buffer.

Binding of Radioligands. Cells were incubated with 75 pM of ¹²⁵I-[D-Tyr⁶,Ala¹¹,Phe¹³,Nle¹⁴]Bn(6-14) for NCI-N417 cells and hBRS-3- and fBB₄-transfected cells; ¹²⁵I-[D-Tyr⁴]Bn for the HuTu-80, AR42J, and murine GRP or human GRP (hGRP) receptor-transfected cells; or ¹²⁵I-[D-Tyr⁰]NMB for the C6 cells and the rat or human NMB-R-transfected cells for the indicated duration and temperature in binding buffer solution [24.5 mM HEPES (pH 7.4), 98 mM sodium chloride, 6 mM potassium chloride, 2.5 mM monobasic sodium phosphate, 5 mM sodium pyruvate, 5 mM sodium fumarate, 5 mM sodium glutamate, 2 mM glutamine, 11.5 mM glucose, 0.5 mM calcium chloride, 1.15 mM magnesium chloride, 0.01% soybean trypsin inhibitor, 0.2% (v/v) amino acid mixture, 0.2% (w/v) BSA, and 0.1% (w/v) bacitracin]. Nonsaturable binding was the amount of radioactivity seen with 75 pM radioligand in the presence of 1 µM [D-Tyr⁶,Ala¹¹,Phe¹³,Nle¹⁴]Bn(6-14), Bn, or NMB and was <10% of total binding in all experiments. Receptor affinities of ligands were determined using a least-squares, curve-fitting program (LIGAND) and the Cheng-Prusoff equation.

Measurement of Inositol Phosphates. All cell lines tested were subcultured onto 24-well plates (5 × 10⁴ cells/well) in their respective propagation media and incubated at 37°C for 24 h before loading with 3 µCi/ml myo-[2-³H]inositol in growth medium containing 2% FBS for an additional 24 h. Ten minutes before assay, the cells were incubated at 37°C with 1 ml/well PBS (pH 7.0) containing 20 mM lithium chloride. The buffer was then aspirated, and 500 µl of an assay buffer [135 mM sodium chloride, 20 mM HEPES (pH 7.4), 2 mM calcium chloride, 1.2 mM magnesium sulfate, 1 mM EGTA, 20 mM lithium chloride, 11.1 mM glucose, and 0.05% BSA (w/v)] was added with or without any of the indicated agonists or antagonists. After a 30-min incubation period (37°C), 1 ml of ice-cold 0.1% hydrochloric acid/methanol (v/v) was added to terminate the experiment. Extraction and collection of total [³H]inositol phosphates (IP) using Dowex AG1-X8 anion exchange resin were performed as reported previously (Ryan et al., 1998a).

[Ca²⁺]_i. Cells were harvested by scraping or centrifugation and resuspended to a concentration of 1.5 × 10⁶ cells/ml in an assay buffer containing 24.5 mM HEPES (pH 7.4), 98 mM sodium chloride, 6 mM potassium chloride, 2.5 mM monobasic sodium phosphate, 5 mM sodium pyruvate, 5 mM sodium fumarate, 5 mM sodium glutamate, 2 mM glutamine, 11.5 mM glucose, 1.45 mM calcium chloride, 1.15 mM magnesium chloride, 0.01% soybean trypsin inhibitor, 0.2% (v/v) amino acid mixture, and 0.2% BSA (w/v). Cell suspensions were incubated with the calcium fluorophore fura-2 acetoxymethyl ester (Molecular Probes, Eugene, OR) at a concentration of 2 µM for 45 min at 37°C and washed twice with assay buffer before beginning the experiments. The change in fluorescence of 2-ml aliquots with or without agonists or antagonists was measured in a Delta PTI Scan 1 spectrofluorimeter (Photon Technology International, South Brunswick, NJ) equipped with a stir bar and water bath (37°C). The fluorescence was measured at two excitation wavelengths (340 and 380 nm), using a single emission wavelength of 510 nm, and the calcium concentration was calculated as described previously (Ryan et al., 1998b). To determine autofluorescence, an aliquot of unlabeled cells was examined in identical experimental conditions.

Statistical Analysis. Statistical analysis of the data were performed using StatView version 1.01 (BrainPower, Inc., Calabasas, CA). KELL for Windows v. 5 (Biosoft, Ferguson, MO) was used for Scatchard analysis of binding data. KaleidaGraph graphing software (Synergy Software, Reading, PA) was used for data plotting and iterative curve-fitting. The Student's t test was used to determine the statistical significance between group means. Values of p >.05 were considered significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

PD 168368 was poorly soluble in incubation buffer and gave almost no inhibition of ¹²⁵I-[D-Tyr⁰]NMB to hNMB-R containing cells with concentrations up to 30 µM (Table 1). However, PD 168368 was soluble in 33% DMSO or 50% hydroxypropyl--cyclodextrin at a stock concentration of 10 mM. We therefore examined the effect of these vehicles on the binding of PD 168368 or putative Bn ligands to human GRP, NMB, and BRS-3 receptors. As shown in Table 1, GRP and NMB had high affinity for their respective receptors in binding buffer containing hydroxypropyl--cyclodextrin or DMSO, and their affinities in the binding buffers with these vehicles were not significantly different from the binding buffer without either solvent. In addition, [D-Phe⁶,Ala¹¹, Phe¹³,Nle¹⁴]Bn(6-14), which has high affinity for BRS-3 (Mantey et al., 1997), bound with similar affinities in all three buffers (Table 1). The affinities of PD 168368 in hydroxypropyl--cyclodextrin were determined for all three human Bn receptors and, in contrast to GRP or NMB, were affected by the solvent used in the buffer (Table 1). PD 168368, in the hydroxypropyl--cyclodextrin buffer, bound with relatively high affinity to hNMB-Rs (K_i = 30 nM), followed by a 40-fold lower affinity for hGRP receptors and almost no affinity for hBRS-3 (K_i > 10 µM; Table 1). However, when DMSO was used as a vehicle, the affinities were reduced 3-fold for the hGRP receptor and 5-fold for hNMB-Rs compared with that seen with hydroxypropyl--cyclodextrin (Table 1). No significant change in the very low affinity of PD 168368 for native hBRS-3 (>10 µM) was observed in the presence of DMSO. Because it appeared that the use of DMSO contributed to a reduction in PD 168368 binding affinities possibly by affecting solubility or some other mechanism, whereas the use of hydroxypropyl--cyclodextrin had no effect on the affinity of other highly soluble peptides, hydroxypropyl--cyclodextrin was used as the vehicle for the remaining binding experiments.

View this table: [in this window] [in a new window]   TABLE 1 Effect of various solvents on binding of Bn-like peptides or PD 168368 to Bn receptor subtypes BALB 3T3 cells transfected with human GRP, NMB, or BRS-3 receptors were incubated with 75 pM of their respective radioligand as described in Experimental Procedures for 45 min with either Bn, NMB, [D-Phe6,-Ala11,Phe13,Nle14]Bn(6-14), or PD 168368 in binding buffer alone; binding buffer containing 0.5% cyclodextrin (v/v); or 0.3% DMSO (v/v) vehicles. Increasing concentrations of unlabeled peptide or PD 168368 were added with the same final concentration of vehicle in all cases, and the dose-inhibition curves were analyzed by least-squares curve fitting program. Ki values were calculated according to the method of Cheng-Prusoff and are the mean ± S.E.M. from at least three experiments performed in duplicate.

To determine the selectivity of PD 168368 for the mammalian Bn receptor subtypes across species, competitive binding studies were performed in cells containing rodent or human NMB-Rs and GRP receptors, as well as cells containing human BRS-3 (Fig. 1, left). In the C6 cells (containing rNMB-R) and the rNMB-R-transfected BALB 3T3 cells, PD 168368 displaced ¹²⁵I-[D-Tyr⁰]NMB binding with a greater potency than that seen with ¹²⁵I-[Tyr⁴]Bn in AR42J cells, which contain rGRP receptors. Specifically, in the C6 cells, detectable inhibition was observed at 1 nM, half-maximal inhibition was observed at 30 nM and 80% inhibition was achieved at 320 nM, whereas in the rNMB-R-transfected cells, detectable inhibition was observed at 1 nM, half-maximal inhibition was observed at 60 nM and 80% inhibition of binding was observed at 600 nM (Fig. 1, left). For AR42J cells (rGRP-R), inhibition was not detected until 300 nM, half-maximal inhibition was seen at 1000 nM, and 80% inhibition of binding was observed at 5 µM. Among the cell lines containing human Bn receptors, PD 168368 had the highest affinity for the hNMB-R-transfected BALB cells with half-maximal inhibition of binding of 30 nM (Fig. 1, right). The concentration of PD 168368 causing half-maximal inhibition of ¹²⁵I-[Tyr⁴]Bn binding to hGRP receptors in HuTu-80 cells (1000 nM) was 33-fold higher than the concentration of PD 168368 needed to displace ¹²⁵I-[D-Tyr⁰]NMB in the hNMB-R-transfected BALB cells. Among the two hGRP receptor-containing cells tested, PD 168368 had a slightly greater affinity for native hGRP receptors in HuTu-80 cells than transfected hGRP receptors in BALB 3T3 cells (Fig. 1, right). In the NCI-N417 cells, which contain native hBRS-3 (Ryan et al., 1998b), no significant displacement of binding was seen in the presence of PD 168368 at concentrations up to 10 µM, whereas 40% maximal inhibition of binding was seen with 10 µM PD 168368 in the hBRS-3-transfected BALB cells (Fig. 1, right).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Ability of PD 168368 to inhibit binding of 125I-[D-Tyr4]Bn, 125I-[D- Tyr0]NMB, or 125I-[D-Tyr6,Ala11,Phe13, Nle14]Bn(6-14) to rat (left) or human (right) GRP, NMB, or BRS-3 receptors, respectively. Cells were incubated for 45 min at 25°C with 75 pM 125I-[Tyr4]Bn, 125I-[D-Tyr0]NMB, or 125I-[D-Tyr6,Ala11,Phe13,Nle14]Bn(6-14), with or without PD 168368 at the concentrations indicated. Results are expressed as the percentage of saturable binding in the absence of unlabeled peptide and are the mean ± S.E.M. from at least three experiments using duplicate determinations.

To compare the affinity of PD 168368 to the various putative ligands for various Bn receptors, we first determined the affinity of these naturally occurring peptides (Bn, GRP, NMB, [Phe¹³]Bn) for the four classes of Bn receptors (Table 2) and summarized the results in Table 3. Bn had the highest affinity for mouse, rat, and human GRP receptors (K_i = 0.5-2.2 nM); moderate affinity for fBB₄ receptors (K_i = 14 nM); and slightly lower affinity for NMB-Rs (K_i = 21-44 nM; Tables 2 and 3). Among the cell lines studied, the HuTu-80 cells, which express native hGRP receptors, and the mGRP-R-transfected BALB 3T3 cells had 4-fold greater affinity for Bn than that observed on AR42J cells or hGRP receptor-transfected BALB 3T3 cells. No detectable affinity for Bn was observed in the cells containing hBRS-3 (Tables 2 and 3). Similar to Bn, GRP also had high affinity for cells containing GRP receptors (K_i = 0.3-6.2 nM), moderate affinity for fBB₄ receptors (K_i = 79 nM) as well as human and rodent NMB-Rs (K_i = 403-5080 nM), but no affinity for hBRS-3 (Table 2). NMB had the highest affinity for NMB-Rs (rat, human), slightly lower affinity for fBB₄-transfected CHO-K1 cells (K_i = 11 nM), relatively low affinities for GRP receptor-containing cells (K_i = 174-437 nM), and low affinities for cells containing hBRS-3 (K_i = 2800-10,000 nM; Table 2). [Phe¹³]Bn, a naturally occurring ligand in amphibians, had the greatest affinity for fBB₄-transfected CHO-K1 cells and GRP receptors with a K_i value of 1 to 4.7 nM but also had high affinity for rat and human NMB-Rs (Tables 2 and 3). The cells containing hBRS-3 receptors had low affinity for [Phe¹³]Bn (K_i = 6600-10,000 nM; Table 2).

We next examined the affinities of PD 168368 for the presently known Bn receptor subtypes. PD 168368 had the highest affinity (Tables 2 and 3) for the NMB-R subtype. Compared with the affinities of the other natural ligands for NMB-Rs, PD 168368 had a 2- to 10-fold lower affinity than NMB or [Phe¹³]Bn, similar affinity to Bn, and 10- and 20-fold higher affinity than GRP (Table 2). For GRP receptors, PD 168368 had 600- to 2000-fold lower affinity than Bn, [Phe¹³]Bn, or GRP and 3- to 5-fold lower affinity than NMB (Table 2). For BRS-3 receptors, PD 168368, as well as each of the natural peptides, had very low affinity (>7000 nM). For fBB₄ receptors, PD 168368 had a 10,000-fold lower affinity than [Phe¹³]Bn and greater than 100-fold lower affinity than Bn, GRP, or NMB.

We also compared the selectivity of the peptoid with the profile of the NMB-R antagonist SS-octa. PD 168368 had a 2- to 7-fold higher affinity for rat or human NMB-Rs than SS-octa (Table 2). However, PD 168368 had a similar selectivity for NMB-Rs to SS-octa in regard to GRP receptors (Tables 2 and 3). Specifically, PD 168368 had a 20- to 40-fold higher affinity for each species of NMB-R over GRP receptors, whereas SS-octa had a 12- to 46-fold selectivity (Table 2). When NMB-Rs are compared with BRS-3 receptors, PD 168368 exhibited greater than 200-fold selectivity, whereas SS-octa had a 10-fold selectivity (Table 2). Last, for fBB₄ receptors, PD 168368 had greater than 250-fold selectivity for NMB-Rs, whereas SS-octa had no ability to distinguish these two receptors, having equal affinity for both (Tables 2 and 3).

To determine the mechanisms by which PD 168368 interacted with NMB-Rs, the ability of 30 nM PD 168368 to inhibit the dose-inhibition curve of NMB for ¹²⁵I-[D-Tyr⁰]NMB binding was examined in the hNMB-R-transfected NCI-H1299 cells (Fig. 2). Scatchard analysis of the inhibition binding curve using nonlinear, best-squares curve-fit analysis (Fig. 2, inset) revealed that PD 168368 decreased NMB binding in a competitive manner because 30 nM PD 168368 caused a significant decrease in the K_d of NMB for hNMB-R from 5.4 ± 1.0 to 9.4 ± 1.4 nM (p < .05) without a significant change in the total number of NMB binding sites (without 30 nM PD 168368, 240 ± 33 fmol/10⁶ cells; with 30 nM PD 168368, 248 ± 27 fmol/10⁶ cells).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Ability of PD 168368 to inhibit binding of 125I-[D-Tyr0]NMB to hNMB-R-transfected NCI-H1299 cells. hNMB-R-transfected NCI-H1299 cells were incubated for 45 min with 75 pM 125I-[D-Tyr0]NMB in the presence of NMB at the concentrations indicated (Control) or in combination with 30 nM PD 168368. Results are expressed as the percentage of saturable binding seen without addition of NMB. Inset, dose-inhibition data plotted in the form of Scatchard. Results are mean ± S.E.M. from at least three experiments using duplicate determinations.

Because the binding affinity of PD 168368 for hNMB-Rs differed according to the vehicle used possibly by altering solubility or by some other mechanism, we examined the ability of PD 168368 to inhibit the total [³H]IP release observed with a submaximal concentration of NMB in the hNMB-R-transfected BALB 3T3 cells in either hydroxypropyl--cyclodextrin (final concentration = 0.5% v/v) or DMSO vehicle (final concentration = 0.3% v/v; Fig. 3). At the concentrations studied, neither PD 168368/hydroxypropyl--cyclodextrin nor PD 168368/DMSO had significant agonist activity alone (data not shown). In both vehicles, PD 168368 inhibited the NMB-stimulated increase of total [³H]IP in a concentration-dependent manner (Fig. 3). In the cells exposed to PD 168368 in hydroxypropyl--cyclodextrin, detectable inhibition was observed at 3 nM PD 168368, half-maximal inhibition was observed at 30 nM, and complete inhibition was observed at 300 nM PD 168368 (Fig. 3). In the cells exposed to PD 168368 in DMSO, detectable, half-maximal, and complete antagonism was observed at concentrations of 10, 50, and 300 nM, respectively (Fig. 3). Iterative curve-fit analysis of the PD 168368 in hydroxypropyl--cyclodextrin dose-inhibition curve revealed an IC₅₀ value of 27 ± 3 nM, whereas the PD 168368 in DMSO dose-inhibition curve revealed an IC₅₀ value of 47 ± 7 nM. Because these results demonstrated that DMSO diminished the ability of PD 168368 to inhibit NMB-R activation, similar to its effect on NMB-R binding (Table 1), the remaining experiments examining the biological activity of PD 168368 on Bn receptor subtypes were performed using the hydroxypropyl--cyclodextrin vehicle (final assay concentration = 0.5% v/v).

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Ability of various solvents to affect the antagonist profile of PD 168368 against NMB-induced [3H]IP formation in hNMB-R-transfected BALB 3T3 cells. hNMB-R-transfected BALB 3T3 cells were incubated with 3 nM NMB alone or with the indicated concentrations of PD 168368 in either 0.5% (v/v) hydroxypropyl--cyclodextrin or 0.3% (v/v) DMSO vehicles at the indicated concentrations. Values represent the percentage of increase in [3H]IP seen with 3 nM NMB alone and are the mean ± S.E.M. from at least three experiments performed in duplicate. For the hydroxypropyl--cyclodextrin and DMSO vehicles, respectively, the control values were 1884 ± 60 and 2098 ± 115 dpm, and the values for 3 nM NMB alone were 34,289 ± 364 and 47,508 ± 1275 dpm.

The biological activity of PD 168368 against all four Bn receptor subtypes was examined using the [³H]IP assay (Table 4). At a concentration of 4 µM, PD 168368 alone had no agonist activity on total [³H]IP release, whereas activation of NMB, GRP, BRS-3, and fBB₄ receptors by 30 nM NMB, Bn, [D-Phe⁶,Ala¹¹,Phe¹³,Nle¹⁴]Bn(6-14), or [Phe¹³]Bn, respectively, stimulated a 1.3- to 14-fold increase in total [³H]IP (Table 4). No significant inhibition of total [³H]IP release by PD 168368 was observed in the hBRS-3/BALB- or fBB₄/CHO-K1-transfected cells; 15% inhibition of total [³H]IP was observed in the human, mouse, and rodent GRP-R-transfected cells; and 85 to 90% inhibition was observed in human and rodent NMB-R-transfected cells (Table 4).

View this table: [in this window] [in a new window]   TABLE 4 Ability of PD 168368 to inhibit increases in [3H]IP stimulated by activation of NMB-Rs, GRP-RS, BRS-3, and BB4 receptors Cells expressing NMB-Rs, GRP-Rs, BRS-3, or BB4 receptors were incubated with 30 nM NMB, Bn, [D-Phe6,-Ala11,Phe13,Nle14]Bn(6-14), or [Phe13]Bn, respectively, for 45 min with or without 4 µM PD 168368 in 0.5% cyclodextrin (v/v). Percentage of agonist alone represents the percentage of [3H]IP formed in the presence of agonist and 4 µM PD 168368 compared with the agonist alone. Results are expressed as mean ± S.E.M. from at least three experiments performed in duplicate.

Because mobilization of [Ca²⁺]_i on receptor activation occurs with each of the Bn receptor subtypes (Wang et al., 1992; Pradhan et al., 1998; Ryan et al., 1998a; Ryan et al., 1998b), we examined the ability of PD 168368 to inhibit the release of [Ca²⁺]_i caused by activation of NMB, GRP, BRS-3, and fBB₄ receptors (Fig. 4). Bn, NMB, [D-Phe⁶,Ala¹¹,Phe¹³,Nle¹⁴] Bn(6-14), or [Phe¹³]Bn, at a 100 nM concentration, caused an elevation of [Ca²⁺]_i in the cells containing receptors for their respective ligands (Fig. 4). The rise in calcium was rapid in all cells, reaching a peak level within 15 s and returning to basal levels by 1 to 2 min. In this assay, PD 168368, at a concentration of 4 µM, failed to stimulate a significant release of [Ca²⁺]_i alone in any of the cell lines studied (Fig. 4, Table 5). In cell lines containing NMB-Rs, the NMB-stimulated increase in [Ca²⁺]_i was either markedly attenuated (60-75%) or completely abolished in the presence of 4 µM PD 168368 (Fig. 4, left; and Table 5). Furthermore, the latency to the peak of the transient was increased (Fig. 4). When the NMB concentration was increased to 100 nM in hNMB-R-transfected BALB 3T3 cells, PD 168368 did not attenuate the calcium transient (data not shown). PD 168368 also inhibited the rise in calcium observed in the murine and human GRP receptor-transfected 3T3 cells but to a lesser extent (50%) than that seen with the hNMB-R-containing cells (Fig. 4, right; and Table 5). No significant inhibition of [Ca²⁺]_i by PD 168368 was observed in NCI-N417 cells, which natively express hBRS-3, or in the fBB₄-transfected CHO-K1 cells (Fig. 4, right; and Table 5), although the transient was attenuated by 10% in the hBRS-3-transfected BALB 3T3 cells (Fig. 4, right; Table 5).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Ability of PD 168368 to inhibit increases in [Ca2+]i resulting from GRP-R, NMB-R, BRS-3, or fBB4 receptor activation. Cells were loaded with fura-2 acetoxymethyl ester and assayed under conditions described in Experimental Procedures. NMB was used as the agonist for the cells containing NMB-Rs (left) at a concentration of 100 nM for the C6 cells and rNMBR-transfected BALB 3T3 cells and 10 nM for the hNMB-R-transfected cells. Right, 100 nM Bn, 100 nM [D-Phe6,Ala11,Phe13,Nle14]Bn(6-14) (BRS-3 peptide), and 100 nM [Phe13]Bn were used as agonists for the mGRP-R-transfected BALB 3T3 cells, the hBRS-3-transfected BALB 3T3 cells, and the fBB4-R-transfected CHO-K1 cells, respectively. Results shown are the effect of the peptide agonists on [Ca2+]i, in the absence or presence of 4 µM PD 168368 as indicated, and are representative of a typical experiment performed at least three times.

View this table: [in this window] [in a new window]   TABLE 5 Ability of PD 168368 to inhibit increases in [Ca2+]i stimulated by activation of NMB-Rs, GRP-Rs, BRS-3, and BB4 receptors Cells expressing NMB-Rs, GRP-Rs, hBRS-3, or BB4 receptors were incubated with 100 nM NMB, Bn, [D-Phe6,-Ala11,Phe13,Nle14]Bn(6-14), or [Phe13]Bn, respectively, with or without 4 µM PD 168368 in 0.5% cyclodextrin (v/v). hNMB-R-transfected BALB 3T3 cells were incubated with 10 nM NMB, which gave a comparable rise in [Ca2+]i to that seen with the other cells containing NMB-Rs. Percentage of agonist alone represents the percent of [Ca2+]i mobilized in the presence of agonist and 4 µM PD 168368 compared with the agonist alone. Results are expressed as mean ± S.E.M. from at least three experiments performed in duplicate.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

There have been numerous attempts to develop small molecule, nonpeptide antagonists for peptide hormone receptors based on the rationale that such compounds, due to their lower susceptibility to degradation in vivo, could prove superior to peptide analogs as potential therapeutic agents. In this study, we examined the comparative pharmacology of the peptoid PD 168368 against NMB-Rs from a number of species, as well as comparing different Bn receptor subtypes of the same species.

The examination of PD 168368's affinity or ability to function as an antagonist at a given receptor in different species was performed because the assumption that an agent tested in one animal species will share the same pharmacology for the same receptor in comparable tissue from a different animal species is not always valid. A study of the effects of des-methionine alkylamide Bn analogs on Bn-induced amylase release on pancreatic acini (Wang et al., 1990) revealed that some of these analogs behaved as GRP receptor agonists or partial agonists in the rat. However, when this class of peptides was assayed for their effects in guinea pig acini, they were found to behave as GRP receptor antagonists (Jensen and Coy, 1991). Similarly, various synthetic cholecystokinin (CCK) analogs behave as CCK_A receptor antagonists in some species and as a partial agonist in others (Howard et al., 1984; Schmassmann et al., 1994). Significant differences have been demonstrated for various peptide antagonists among various species with NMB and GRP receptors as well as other peptide hormone receptors. For example, it was found that the NMB-R antagonist SS-octa had a 4-fold greater affinity for rodent versus human NMB-Rs (Orbuch et al., 1993). Likewise, in a study of GRP receptor antagonists, some of the Bn analog peptides had greater antagonist potency for guinea pig than for rat GRP receptors (Wang et al., 1990). For other peptide hormone receptors, such as the tachykinin receptors, species variation in antagonist potency and/or affinity has also been described for nonpeptide antagonists of tachykinin receptors (Maggi et al., 1993; Beattie et al., 1995), which displayed varying degrees of potency against a variety of mammalian tissues. In the present study, several lines of evidence demonstrated that PD 168368 was a potent NMB-R antagonist and exhibited neither a species-specific preference nor variance in pharmacological activity (e.g., agonist versus antagonist activity) for NMB-Rs. First, it was determined that PD 168368 lacked intrinsic agonist activity in the assays tested because in cell lines containing rodent or human NMB-Rs, peptoid concentrations as high as 4 µM failed to elicit two known cellular responses normally associated with NMB-R activation (Benya et al., 1992; Moody et al., 1992; Wang et al., 1992), elevation of [Ca²⁺]_i, or IPs. Second, PD 168368 displacement of NMB binding was concentration dependent, and Scatchard analysis of PD 168368 binding revealed that this peptoid appeared to attenuate NMB binding in a competitive manner because it was capable of decreasing the affinity of NMB for its receptor without significantly altering the number of binding sites, although allosteric antagonism cannot be excluded by the results of our studies. Third, PD 168368 antagonized IP release over a concentration range similar to that observed for inhibition of NMB binding and attenuated NMB-stimulated calcium mobilization. Fourth, PD 168368 shared similar nanomolar affinities for both rodent and human NMB-Rs. These results support the conclusion that PD 168368 is likely to be useful in both rodents and higher-order species as an NMB-R antagonist.

A second aim of this study was to examine the potential selectivity of PD 168368 for NMB-Rs over other Bn receptors and to investigate its ability to function as an antagonist or agonist at other subtypes of Bn receptors. This comparison was performed because previous studies have shown that antagonists for one peptide hormone receptor can have agonist activity in another receptor subtype of the same family. For example, some members of one class of potent GRP receptor antagonists, the Des-methionine Bn alkylamide analogs, behaved as agonists on NMB-Rs (Gibril et al., 1994; Ryan et al., 1996). A similar phenomenon has been observed with CCK receptors, where PD 135158 behaved as a CCK_B antagonist and also as a CCK_A receptor agonist. In the present study, we found that PD 168368 did not have agonist activity at any of the other Bn receptor subtypes in the assays tested, and therefore, at least in human or rat studies, it should not be subject to nonselective agonist activity. In terms of its antagonist selectivity for Bn receptor subtypes, our results demonstrate that PD 168368 had high selectivity (>200-fold) for NMB-Rs over BRS-3 or fBB₄ receptors. For differentiation of NMB from GRP receptors in rat or human, PD 168368 had a lower selectivity, having 30- to 60-fold higher affinity for NMB-Rs than for GRP receptors. This result raises the possibility that if high concentrations of PD 168368 were used in various experimental systems in the two species, some inhibition would also be seen at the GRP receptor.

PD 168368 and the other selective NMB-R antagonist SS-octa (Orbuch et al., 1993) have both similarities and important differences in their pharmacological characteristics. Both antagonists have nanomolar affinities for NMB-Rs, have greater selectivity for NMB-Rs compared with GRP receptors, and inhibit NMB binding in a competitive manner (Orbuch et al., 1993). In addition, both compounds attenuated the IP release and calcium mobilization seen with NMB-R activation (Orbuch et al., 1993). However, there were some distinct differences between these two compounds. Unlike the peptoid PD 168368, SS-octa had relatively high affinity for some of the other Bn receptor subtypes. SS-octa interacted with fBB₄ receptors with the same affinity that it interacted with NMB-Rs and had a 3- to 10-fold higher affinity for BRS-3 receptors than PD 168368. Furthermore, the general use of SS-octa is limited because it can also function as a somatostatin receptor agonist (Orbuch et al., 1993) and interact with the µ opioid receptor (Orbuch et al., 1993). Recently, an ornithine analog of SS-octa (Ladenheim et al., 1994) was synthesized that decreases the affinity for the µ opioid receptor and probably overcomes this limitation. Another important difference found between these two NMB-R antagonists is the effect of the solvents used on the potency of the peptoid PD 168368. SS-octa was soluble in physiological buffers (Orbuch et al., 1993), whereas PD 168368 had poor solubility in physiological buffers. The data from the present study demonstrated that the choice of vehicle used to dissolve PD 168368 can have an effect on the ability of the peptoid to inhibit binding or biologic activity. We found that both the receptor affinity of PD 168368 and its ability to function as an antagonist were significantly lower if DMSO was used as the solvent rather than hydroxypropyl--cyclodextrin. The basis for this result was not investigated in detail; however, it was not due to a nonspecific effect because agonists at each of the three mammalian Bn receptor subtypes had similar affinity in DMSO, hydroxypropyl--cyclodextrin, or physiological buffers. Furthermore, it was not a specific effect on the NMB-Rs because the affinity of PD 168368 for GRP receptors was affected in a similar manner. The most likely explanation for these results is that the cycloglucan structure of hydroxypropyl--cyclodextrin, which is distinctly different from the structure of DMSO, provides better solubility for PD 168368, especially at the final vehicle concentrations (0.3-0.5%) used in the assays. It remains possible the cyclodextrin could be having some direct effect on the affinity of PD 168368 for receptors; however, this is unlikely because cyclodextrin had no effect on Bn or NMB affinity for the same receptors.

In conclusion, we find that PD 168368 is a potent and selective NMB-R antagonist with similar characteristics against human and rodent NMB-Rs. PD 168368 should prove useful for further studies of NMB-R biology and represents a significant step toward the development of nonpeptide antagonists for receptors of the Bn-like peptides. Our results demonstrate that when interpreting results with PD 168368, it will be important to remember it retains some affinity for the GRP receptor (30-60-fold lower than the NMB-R) and that the vehicle selection could have a significant effect on its affinity and potency.