Abstract
The parathyroid hormone (PTH)-1 receptor mediates the pathophysiological effects of PTH in hyperparathyroidism and PTH-related protein (PTHrP) in humoral hypercalcemia of malignancy. A PTH1 receptor antagonist may be of therapeutic utility in these disorders. We recently identified a novel antagonist, tuberoinfundibular peptide (7-39) [TIP(7-39)], derived from the likely endogenous ligand for the PTH2 receptor TIP39. In this study its in vitro profile is evaluated and compared with that of [d-Trp12,Tyr34]bPTH(7-34) and PTHrP(7-34), representing the two previously known structural classes of PTH1 receptor antagonists. TIP(7-39) binds with higher affinity (6.2 nM) to the PTH1 receptor than [d-Trp12,Tyr34]bPTH(7-34) (45 nM) and PTHrP(7-34) (65 nM) and displays a 5.5-fold greater PTH1/PTH2 receptor selectivity. TIP(7-39) does not stimulate cAMP accumulation via the PTH1 receptor [in a sensitive assay that detects the activity of the weak partial agonist [Nle8,18,Tyr34]bPTH(3-34)] and does not increase intracellular calcium. Schild analysis for TIP(7-39) was consistent with purely competitive antagonism of PTH(1-34)'s stimulation of cAMP accumulation (slope = 0.99 ± 0.24). The pKB for TIP(7-39) (7.1 ± 0.3) was higher than that for [d-Trp12,Tyr34]bPTH(7-34) (6.5 ± 0.0) and PTHrP(7-34) (6.0 ± 0.1). Binding of125I-TIP(7-39) to the PTH1 receptor could be measured (KD = 1.3 ± 0.1 nM,Bmax = 1.3 ± 0.1 pmol/mg), whereas binding of125I-[Nle8,18,d-Trp12,Tyr34]bPTH(7-34) could not be detected. Kinetic analysis indicated that125I-TIP(7-39) dissociates much more slowly (t1/2 = 14 min) than [d-Trp12,Tyr34]bPTH(7-34) (13 s) and PTHrP(7-34) (9 s). The novel antagonist TIP(7-39) therefore displays a more favorable in vitro pharmacological profile than antagonists derived from PTH and PTHrP and may be useful for demonstrating the utility of PTH1 receptor antagonism in the treatment of hypercalcemia.
The parathyroid hormone type 1 (PTH1) receptor mediates the principal physiological responses to PTH and to PTH-related protein (PTHrP) (Martin and Moseley, 1995; Potts et al., 1995). PTH acts on the PTH1 receptor in bone and kidney to elevate blood calcium levels (Potts et al., 1995). PTHrP is a locally acting autocrine or paracrine factor and developmental regulator (Martin and Moseley, 1995; Wysolmerski and Stewart, 1998). Both of these peptides are involved in disorders of calcium metabolism. Humoral hypercalcemia of malignancy (HHM) is caused by very high levels of tumor-produced PTHrP activating the PTH1 receptor (Grill et al., 1998). In primary hyperparathyroidism (HPT), increased blood calcium results from excessive secretion of PTH from the parathyroid gland (Nemeth and Fox, 1999).
Because activation of the PTH1 receptor is involved in the pathology of these disorders, a PTH1 receptor antagonist may be of therapeutic utility. The N-terminal region of these peptides is a critical determinant for cAMP accumulation via the PTH1 receptor (Potts et al., 1995). PTH1 receptor antagonists have therefore been developed by deletion of residues from the N terminus of PTH and PTHrP. The most potent of these ligands is [Nle8,18,Tyr34]bPTH(3-34) (Rosenblatt et al., 1977, 1980; McGowan et al., 1983; Hoare et al., 1999a), which has a KD of about 1 nM. This ligand was initially described as an antagonist based on inhibition of PTH-stimulated cAMP accumulation (Rosenblatt et al., 1977; McGowan et al., 1983) but was subsequently demonstrated to act as a partial agonist in vivo (Gray et al., 1982; Segre et al., 1985). Using more sensitive in vitro techniques, [Nle8,18,Tyr34]bPTH(3-34) has been shown to be a weak partial agonist for cAMP accumulation (Martin et al., 1981; McKee et al., 1990; Pines et al., 1996). This analog also stimulates cAMP-independent signaling pathways (Azarani et al., 1996).
Removing the first six N-terminal residues from PTH and PTHrP, together with substitution of certain amino acids, yielded PTH(7-34) and PTHrP(7-34) analogs with reduced signaling efficacy but at the expense of lower binding affinity (Rosenblatt et al., 1980; McGowan et al., 1983; McKee et al., 1988, 1990; Chorev et al., 1990; Goldman et al., 1998; Hoare et al., 1999a). The residue 7-34 fragments act as antagonists, or in some cases weak partial agonists, in vitro (Goldman et al., 1988; McKee et al., 1988, 1990; Chorev et al., 1990). These peptides can antagonize the effects of exogenous PTH or PTHrP in thyroparathyroidectomized rats (Horiuchi et al., 1983; Doppelt et al., 1986; Horiuchi and Rosenblatt, 1987; Dresner-Pollak et al., 1996). Administration of the antagonist before PTH or PTHrP exposure may be required to observe significant inhibition (Dresner-Pollak et al., 1996).
We identified a novel PTH1 receptor antagonist in our investigation of the receptor selectivity of tuberoinfundibular peptide of 39 residues (TIP39). This recently discovered hypothalamic peptide activates the PTH2 receptor and may be its natural ligand (Hoare et al., 1999b; Usdin et al., 1999). The human PTH2 receptor has 52% amino-acid sequence identity to the human PTH1 receptor (Usdin et al., 1995). TIP39 shares some sequence homology with PTH and PTHrP; five residues are identical when the sequences of TIP39, PTH, and PTHrP are aligned (Usdin, 2000). TIP39 strongly activates the PTH2 receptor and binds to it with subnanomolar affinity (0.59 nM) (Hoare et al., 2000). TIP39 binds to the PTH1 receptor with moderate affinity (59 nM) but produces little or no stimulation of cAMP accumulation. Deletion of six residues from the N terminus of TIP39 reduces binding affinity for the PTH2 receptor by 72-fold but increases PTH1 receptor affinity by a factor of 10 (Hoare et al., 2000). TIP(7-39) does not detectably stimulate cAMP accumulation at PTH1 or PTH2 receptors. Therefore, TIP(7-39) is a selective, high-affinity antagonist for the PTH1 receptor.
TIP(7-39) may possess different properties from N-terminally truncated PTH and PTHrP analogs, owing to its different primary structure and because the parent peptide, unlike PTH and PTHrP, does not appreciably activate the receptor. We have now evaluated the in vitro functional and binding properties of bovine TIP(7-39) at the PTH1 receptor and compared its pharmacological profile with that of [d-Trp12,Tyr34]bovine PTH(7-34) and PTHrP(7-34), which represent the two previously known structural classes of PTH1 receptor antagonists.
Materials and Methods
Reagents and Peptides.
The following peptides were obtained from Bachem (Torrance, CA) or Peninsula Laboratories (Belmont, CA): [d-Trp12,Tyr34]bPTH(7-34) amide, [Nle8,18,d-Trp12,Tyr34]bPTH(7-34) amide, PTHrP(7-34) amide, rPTH(1-34), [Nle8,21,Tyr34]rPTH(1-34) amide, and [Nle8,18,Tyr34]bPTH(3-34) amide. The letters “b” and “r” designate the peptide sequence as bovine and rat, respectively. These peptides were dissolved in 10 mM acetic acid at a concentration of 1 mM, calculated using the peptide content and weight provided by the supplier. bTIP39 and bTIP(7-39) were purchased from Biomolecules Midwest (Waterloo, IL). bTIP(7-39) was quantified using the copper bicinchoninic acid method (Pierce, Rockford, IL) with TIP39 as the standard.125I-cAMP was obtained from NEN (Boston, MA) and Na125I (2000 Ci/mmol) was from ICN Biomedicals (Costa Mesa, CA). Lactose peroxidase was obtained from Sigma (St. Louis, MO). Cell culture supplies were obtained from Life Technologies (Frederick, MD), except for Dulbecco's modified Eagle's medium (DMEM), which was from Mediatech (Herndon, VA). Fluo-4 acetoxymethyl ester and Pluronic F-127 were from Molecular Probes (Eugene, OR). Probenecid was from Sigma.
Preparation of Radioligands.
125I-[-Nle8,21,Tyr34]rPTH(1-34) was prepared using chloramine T as catalyst followed by purification by HPLC, as previously described (Clark et al., 1998). The di-iodinated form of the radioligand (4000 Ci/mmol) was used in binding experiments.125I-TIP39 and125I-TIP(7-39) (2000 Ci/mmol) were prepared using the lactose-peroxidase method. TIP39 [5 μg in 5 μl of reaction buffer (0.1 M sodium acetate buffer, pH 6.5)] was dispensed into a siliconized microfuge tube, followed by sequential addition of 0.5 mCi Na125I, 5 μl of 20 μg/ml lactose peroxidase in reaction buffer, and 45 μl of reaction buffer. After mixing, 10 μl of 0.001% H2O2 was added. After 20 min at room temperature the reaction was terminated by addition of 0.5 ml of reaction buffer supplemented with 0.1% sodium azide. After a further 5 min, 0.5 ml of reaction buffer supplemented with 1 M NaCl, 0.1% BSA, and 1% potassium iodide was added. The radioligand was then desalted using a C18 cartridge and purified by HPLC. The radioactive peak fractions corresponded with a single peak of UV absorbance.
Cell Culture of HEK293 Cells and Isolation of Cell Membranes.
HEK293 cells stably expressing the human PTH1 or PTH2 receptor were grown as previously described (Hoare et al., 1999a). P2 membrane preparations from HEK293 cells expressing the human PTH2 and PTH1 receptors were isolated as previously described (Hoare et al., 1999a). Membrane protein was quantified using the copper bicinchoninic acid method with BSA as the standard. Cell membranes were stored at −80°C before use.
Cell Culture and Transient Expression in COS-7 Cells.
COS-7 cells were grown as previously described (Clark et al., 1998). For cAMP accumulation assays COS-7 cells were transfected as previously described (Clark et al., 1998) except that transfections were performed in 10-cm tissue culture dishes using 10 μg of plasmid DNA. The cells were transferred after trypsinization to 96-well plates at a density of 50,000 cells/well the following day. Cells were used for assays of cAMP accumulation 3 days after transfection.
Radioligand Binding Assays.
The centrifugation assay used for radioligand displacement assays (Fig.1) has been described previously (Hoare et al., 1999a). A similar assay design was used for the PTH1 and PTH2 receptors, in which radiolabeled agonist binding was displaced by the unlabeled ligands in the presence of 10 μM GTPγS. Briefly, cell membranes (45 μg), radioligand (50,000 cpm), and unlabeled ligand were incubated in a final volume of 1 ml of assay buffer [20 mM HEPES, 100 mM NaCl, 1 mM EDTA, 3 mM MgSO4 pH 7.5, supplemented with 0.3% nonfat dried milk powder, 100 μM 4-(2-aminoethyl)-benzenesulfonylflouride, 1 μg/ml bacitracin, and 10 μM GTPγS] for 2 h at 21°C. Membranes were collected at 18,000g, the surface of the pellet gently washed, and the radioactivity counted as described (Hoare et al., 1999a). For the PTH1 receptor,125I-[Nle8,21,Tyr34]rPTH(1-34) was used as the radioligand at a final concentration of approximately 5 pM. 125I-TIP39 was used to label the PTH2 receptor at a concentration of 10 pM, assuming mono-iodination of TIP39 using the lactose peroxidase method. To prevent greater than 20% of the total radioligand added from binding to the membranes, 15 to 20 μg of membranes from transfected cells was used, made up to 45 μg with membranes from nontransfected cells.
Binding of 125I-TIP(7-39) to HEK293 membranes expressing the PTH1 receptor (Figs. 5 and 6) was performed using rapid filtration to separate bound and free radioligand as previously described (Hoare and Usdin, 1999), using the assay buffer described above. Incubations were carried out in 96-well polypropylene plates. The incubation mixture was transferred to a polyvinylidene fluoride filtration plate (MAHVN45; Millipore, Bedford, MA) and the membranes collected by rapid filtration using a Millipore Multiscreen vacuum manifold. In saturation experiments, varying concentrations of125I-TIP(7-39) were incubated in triplicate with 10 μg of membranes in the absence or presence of 1 μM unlabeled TIP(7-39) (for measurement of total binding and nonspecific binding, respectively) for 1 h at 21°C. In radioligand association experiments, radioligand and buffer were brought to 21°C by incubation in a water bath for 15 min. Prewarmed membranes were then added to the wells at various time points and the assay wells harvested simultaneously. Nonspecific binding in these experiments was defined using 300 nM unlabeled TIP(7-39), incubated with membranes and radioligand for 1 and 60 min. In the experiment in Fig. 7, a second, unlabeled ligand was included in the assay incubation to estimate the association and dissociation rate constants of the unlabeled ligand (see below). In dissociation experiments radioligand and membranes were equilibrated for 60 min before addition of unlabeled TIP(7-39) (300 nM final concentration) at various time points. All time points were harvested simultaneously. (As a result the shorter time points of the time course were equilibrated with radioligand for between 1 and 2 h.) Nonspecific binding was defined using 300 nM unlabeled TIP(7-39), which was included in the equilibration phase of the assay.
Measurement of Cellular Levels of cAMP.
Slightly different procedures were used depending on the experimental paradigm. For the experiment in Fig. 2, transfected COS-7 cells were treated for 40 min at 37°C with 50 μl/well cAMP assay buffer [DMEM containing 25 mM HEPES supplemented with 0.1% bovine serum albumin, 30 μM Ro 20-1724 (Research Biochemicals International, Natick, MA), 100 μM 4-(2-aminoethyl)-benzenesulfonylflouride, and 1 μg/ml bacitracin]. This buffer was then removed and replaced with 40 μl of fresh buffer. Test agents were added in a volume of 10 μl and the cells incubated for an additional 40 min at 37°C. The assay was then terminated by the addition of 50 μl of 0.1 N HCl, 0.1 mM CaCl2. For measurement of PTH1 receptor antagonism by TIP(7-39) (Fig. 4) cells were washed with 100 μl of DMEM and then treated with 40 μl of cAMP assay buffer containing varying concentrations of antagonist (or no antagonist for the control) for 30 min at 37°C followed by addition of a range of concentrations of rPTH(1-34) in a volume of 10 μl. After a further 40 min at 37°C the assay was terminated as described above. For measurement of the effect of human plasma on antagonist potency (Fig. 5) cells were treated for 40 min with 50 μl of cAMP assay buffer. The buffer was removed and the following solutions added sequentially: 30 μl of buffer containing plasma, 10 μl of antagonist in buffer, and 10 μl of rPTH(1-34) in buffer. The cells were incubated at 37°C for 40 min before assay termination. Human plasma was prepared by addition of EDTA to whole blood at a final concentration of 10 mM followed by centrifugation at 1000g for 10 min. The plasma supernatant was collected and stored in aliquots at −80°C before use. cAMP was quantified using a radioimmunoassay as previously described (Clark et al., 1998).
Measurement of Intracellular Calcium Concentration.
HEK293 cells stably expressing the PTH1 receptor were seeded in wells of a 96-well plate at 100,000 cells/well. The following day, medium was removed and the cells washed once with 0.1 ml of Dulbecco's phosphate-buffered saline (DPBS) containing 1 mM Ca2+ and 1 mM Mg2+. Cells were then loaded with 5 μM fluo-4 acetoxymethyl ester, with 0.1% (w/v) Pluronic F-127 and 2.5 mM probenecid in DPBS for 1 h at 37°C. After two washes with DPBS supplemented with 0.1% BSA, cells were incubated in 0.1 ml of the same buffer for 30 min at 37°C. This buffer was then removed and 50 μl of prewarmed DPBS with BSA added. Baseline fluorescence was then measured for 80 s at 37°C in a Cytofluor 4000 multiwell plate fluorimeter (PerSpective Biosystems, Framington, MA) (excitation wavelength 485 ± 20 nm, emission wavelength 530 ± 25 nm). Test agents were then added and fluorescence monitored as before. Fluorescence was measured in duplicate wells of cells for each experimental condition. Cytosolic free calcium concentration ([Ca2+]i) was calculated using the following equation: [Ca2+]i =KD (F −Fmin)/(Fmax −F) where KD is the ion dissociation constant (345 nM) for the indicator and F the fluorescence signal in arbitrary units.Fmax (maximum fluorescence at Ca2+ saturation of the indicator) was determined by addition of 130 μM ionomycin and Fmin(background fluorescence) measured after addition of 20 mM EGTA.
Data Analysis.
Concentration-dependence data for ligand-stimulated cAMP accumulation and inhibition of radioligand binding (Figs. 1, 2, 4, and 5) were analyzed using the following four-parameter logistic equation using Prism 2.01 (GraphPad Software Inc., San Diego, CA):
The effect of TIP(7-39) on rPTH(1-34)-stimulated cAMP accumulation at the human PTH1 receptor was analyzed using Schild analysis (Fig. 4), using the following equation:
125I-TIP(7-39) saturation of the PTH1 receptor was analyzed as follows. First, nonspecific binding [measured in the presence of 1 μM TIP(7-39)] was estimated as a fraction of the free radioligand concentration by linear regression. The values ofKD and Bmaxwere obtained by fitting total binding data (measured in the absence of unlabeled ligand) to the following equation using Prism 2.01:
125I-TIP(7-39) association data (total binding) were fitted to a biexponential association equation to account for association to specific and nonspecific sites (Fig. 7). This procedure was used because the value of nonspecific binding measured after 60 min was slightly greater than the value measured after 1 min. In the analysis the equilibrium level of nonspecific binding was fixed at that measured at 60 min. The observed association rate constant for nonspecific binding was high (>2 min−1). The observed association rate of specific radioligand (L) binding [kon(obs)] was fitted by linear regression to the equation kon(obs) =koff +kon[L] wherekon and koffare the association and dissociation rate constants, respectively.125I-TIP(7-39) dissociation data were fitted to a monoexponential dissociation equation. A biexponential equation did not significantly improve the fit in all cases (P > .7).
The association and dissociation rate constants of unlabeled ligands were determined using the method devised by Motulsky and Mahan (1984)in which association of a radiolabeled antagonist [125I-TIP(7-39)] is measured in the presence of a fixed concentration of the unlabeled ligand. The model assumes that the ligands bind in a competitive manner according to simple bimolecular reactions. The total amount of radioligand bound to the receptor ([RL]) as a function of time was fitted to the following equation using SigmaPlot 3.0 (Jandel Scientific, SPSS Inc., Chicago, IL):
Statistical comparison of multiple means was performed initially by single-factor analysis of variance followed by post hoc analysis with the Newman-Keuls test. Statistical comparison of two means was performed using a two-tailed Student's t test.
Results
Binding of Antagonists to the Human PTH1 and PTH2 Receptors.
Radioligand binding assays were used to compare the receptor binding affinity of TIP(7-39) with that of [d-Trp12,Tyr34]PTH(7-34) and PTHrP(7-34). Membranes prepared from HEK293 cells expressing the human PTH1 receptor were labeled with125I-[Nle8,21,Tyr34]rPTH(1-34) and from HEK293 cells expressing the human PTH2 receptor with125I-TIP39. Binding was measured in the presence of 10 μM GTPγS to minimize complications arising from receptor-G-protein coupling, such as pseudoirreversible binding of the agonist radioligand (Hoare et al., 1999a).
Binding of all ligands to both receptors was described by a pseudo Hill slope of approximately unity (Table 1), consistent with a simple bimolecular reaction for the receptor-ligand interaction. TIP(7-39) bound with a significantly higher affinity to the PTH1 receptor than [d-Trp12,Tyr34]PTH(7-34) or PTHrP(7-34) (Fig. 1A; Table 1). The difference of IC50 was 7.3-fold for [d-Trp12,Tyr34]PTH(7-34) and 10-fold for PTHrP(7-34). All of the antagonist ligands bound with lower affinity to the PTH2 receptor than the PTH1 receptor (Fig. 1B; Table 1). However, TIP(7-39) displayed a 5.5-fold greater selectivity for the PTH1 receptor than [d-Trp12,Tyr34]PTH(7-34) or PTHrP(7-34) (Table 1).
Effect of TIP(7-39) on cAMP Accumulation in COS-7 Cells Expressing a C-Terminal Hemagglutinin-Tagged Human PTH1 Receptor.
Some PTH1 receptor ligands that were initially identified as antagonists based on inhibition of PTH-stimulated cAMP accumulation have since been demonstrated to possess significant efficacy in more sensitive assay systems. The best characterized example is [Nle8,18,Tyr34]bPTH(3-34). TIP(7-39) did not detectably stimulate cAMP accumulation in HEK293 expressing the human PTH1 receptor (Hoare et al., 2000) but in these cells a response to [Nle8,18,Tyr34]bPTH(3-34) was also not detected (Hoare et al., 1999a). We attempted to develop a more sensitive measure of PTH1 receptor activation to evaluate the potential agonism of TIP(7-39), and used the ability to detect the partial agonism of [Nle8,18,Tyr34]bPTH(3-34) as the criteria for this assay. In COS-7 cells expressing the wild-type PTH1 receptor a measurable cAMP response to [Nle8,18,Tyr34]bPTH(3-34) was observed in two of five assays (data not shown). However, a hemagglutinin-tagged PTH1 receptor was detectably activated by this ligand in COS-7 cells in each of five experiments, with anEmax of 26 ± 4% of the maximal response to rPTH(1-34) (Fig. 2). [This receptor contains a 12-amino-acid residue hemagglutinin epitope inserted at the C terminus (Clark et al., 1998).] TIP(7-39) did not detectably stimulate cAMP accumulation in this assay (Fig. 2): Linear regression analysis indicated that the slope defining the concentration dependence of cAMP accumulation was not significantly different from zero in three independent experiments. In addition, the level of cAMP accumulation produced by 3.2 μM TIP(7-39) (0.91 ± 0.04 pmol/well) was not significantly different (P = .51) from the accumulation measured in the absence of ligand (0.95 ± 0.04 pmol/well).
Effect of TIP(7-39) on Intracellular Calcium Concentration.
The PTH1 receptor has been demonstrated to couple to other second messenger pathways in addition to stimulation of cAMP accumulation (Abou-Samra et al., 1992; Azarani et al., 1996; Friedman et al., 1999). One of the best studied of these additional pathways is the elevation of [Ca2+]i. We therefore tested whether TIP(7-39) affects [Ca2+]i, using fluo-4-loaded HEK293 cells expressing the human PTH1 receptor. No change in [Ca2+]i was observed when these cells were incubated with a high concentration of TIP(7-39) (1 μM), whereas 3 nM rPTH(1-34) produced a robust, rapid, and transient increase in [Ca2+]i (Fig.3). TIP(7-39) (1 μM) antagonized the effect of rPTH(1-34) (3 nM); the peak [Ca2+]i increase was reduced by 79 ± 1% and the rate of increase was reduced (Fig.3).
Measurement of Antagonist Potency of TIP(7-39) at Human and Rat PTH1 Receptors Expressed in COS-7 Cells.
Schild analysis of TIP(7-39) inhibition of rPTH(1-34)-stimulated cAMP accumulation was performed to examine the mechanism of action of the antagonist at the PTH1 receptor and to measure antagonist potency in a functional assay. TIP(7-39) produced a parallel rightward shift of the rPTH(1-34) concentration dependence curve for stimulation of cAMP production at the human PTH1 receptor (Fig. 4A). The antagonist did not significantly affect theEmax for rPTH(1-34) and did not detectably activate cAMP accumulation in the absence of agonist (see legend to Fig. 4). The Schild slope was 0.99 ± 0.24 (Fig. 4B). These observations strongly suggest that TIP(7-39) acts as a competitive antagonist of rPTH(1-34)-stimulated cAMP accumulation at the human PTH1 receptor, at least over the range of antagonist concentrations tested. The pKB of TIP(7-39) at the human PTH1 receptor was 6.83 (150 nM). This value is 24-fold greater than the IC50 of TIP(7-39) for inhibition of125I-rPTH(1-34) binding to the human PTH1 receptor (Table 1). TIP(7-39) also antagonized PTHrP(1-34)-stimulated cAMP accumulation at the human PTH1 receptor, with a pKB of 6.94 ± 0.09 (110 nM) (graphical data not shown). The pKB of the antagonist was also measured for the rat PTH1 receptor expressed in COS-7 cells, using 3.2 μM TIP(7-39) and rPTH(1-34) as the agonist (graphical data not shown). The pKB value of 6.51 ± 0.23 (310 nM) was not greatly different from that for the human PTH1 receptor.
Antagonist Potency in the Presence of Human Plasma.
One explanation that has been proposed for the lack of effect of PTH1 receptor antagonists in vivo is inactivation of the ligand as a result of ligand binding to plasma proteins (Kukreja et al., 1994). We investigated this possibility by measuring the shift of rPTH(1-34) EC50 produced by the antagonist in the absence and presence of 20% human plasma. It is important to note that this experiment does not address the effects of serum proteases on the antagonist effect because protease inhibitors were included in the assay. Human plasma did not reduce the antagonist potency of TIP(7-39) (Fig. 5), [d-Trp12,Tyr34]PTH(7-34), or PTHrP(7-34) (Table 2). Indeed, plasma increased antagonist potency between 2.3- and 3.5-fold (Table 2). These experiments also demonstrate that TIP(7-39) displays a greater antagonist potency than either [d-Trp12,Tyr34]PTH(7-34) or PTHrP(7-34), in both the absence and presence of plasma (Table 2).
Binding of 125I-TIP(7-39) to the Human PTH1 Receptor in HEK293 Cell Membranes.
To enable a more detailed characterization of its ligand binding mechanism we prepared radiolabeled TIP(7-39). bTIP39 contains a tyrosine residue at position 29 and a methionine residue at position 30 (Usdin et al., 1999), so125I-TIP39(7-39) was prepared using the mildly oxidizing lactose peroxidase method. Specific binding of this radioligand was detected in membranes prepared from HEK293 cells expressing the human PTH1 receptor (using 300 nM TIP(7-39) or 300 nM TIP39 to define nonspecific binding), whereas no specific binding was detected in HEK293 membranes prepared from nontransfected cells (data not shown). The total binding/nonspecific binding ratio for125I-TIP(7-39) was approximately 5:1, which is comparable with the signal-to-noise ratio of 6:1 obtained with125I-[Nle8,18,Tyr34]bPTH(3-34) (a commonly used radiolabeled antagonist/partial agonist for the PTH1 receptor). The affinity of 125I-TIP(7-39) for the PTH1 receptor was measured in saturation experiments, using varying concentrations of the radioligand. The saturation data were fitted well by a single-site saturation isotherm (Fig.6A), a two-site model not providing a significant improvement to the fit (P values ranged from 0.75 to 0.95). The KD for125I-TIP39(7-39) was 1.3 ± 0.1 nM and theBmax 1.3 ± 0.1 pmol/mg (n = 3). This KD is comparable with that for [Nle8,18,Tyr34]bPTH(3-34) (2.0 nM, Hoare and Usdin, 1999). The Bmaxis slightly higher than that for [Nle8,18,Tyr34]bPTH(3-34) (0.7 pmol/mg, Hoare and Usdin, 1999). However this value was obtained from homologous displacement experiments, which may be less accurate than saturation experiments for measurement ofBmax if there is a difference between the binding affinities of the iodinated and noniodinated ligands.
Measurement of Antagonist Binding Kinetics at the Human PTH1 Receptor in HEK293 Cell Membranes.
The association and dissociation rate constants for 125I-TIP(7-39) binding to the PTH1 receptor were measured directly using data from the time courses of radioligand association and dissociation. The affinities of [d-Trp12,Tyr34]PTH(7-34) and PTHrP(7-34) are probably too low to permit their use as radioligands in binding assays. (We prepared125I-[Nle8,18,d-Trp12,Tyr34]PTH(7-34), the higher affinity of the two analogs, and were unable to detect specific binding to the PTH1 receptor (data not shown).) Rate constants for these peptides were measured indirectly using a method in which association of a single concentration of a radioligand [125I-TIP(7-39)] is measured in the presence of a single concentration of the unlabeled test ligand (Motulsky and Mahan, 1984). The time course data (Fig.7) were fitted to eq. 4 as described under Materials and Methods to obtain estimates of the association and dissociation rate constants of the unlabeled ligand.
Association and dissociation of 125I-TIP(7-39) binding to the PTH1 receptor was monophasic (Figs. 6C and 7) and the observed association rate constant appeared to be linearly dependent upon the concentration of radioligand (Fig. 6B). These observations are consistent with a simple bimolecular interaction between the receptor and radioligand. The kinetically derived KD(0.57 nM, Table 3) was in reasonable agreement with the KD measured directly in saturation experiments (1.3 nM, Fig. 6A). The estimate of the dissociation rate constant from the plot ofkon(obs) versus concentration of radioligand (0.077 min−1, Fig. 6B) was in good agreement with the directly measured value (0.051 min−1, Fig. 6C).
Association of 125I-TIP(7-39) in the presence of the unlabeled antagonists (Fig. 7) was fitted well by a model that assumes competitive inhibition between the radioligand and unlabeled ligand (eq. 4). The model can account for the slight “overshoot” observed for the association of125I-TIP(7-39) in the presence of [Nle8,18,Tyr34]bPTH(3-34) (Fig. 7). Equation 4 was used to estimate the association and dissociation rate constants for the unlabeled ligands. The dissociation rate constant for both [d-Trp12,Tyr34]PTH(7-34) and PTHrP(7-34) was much greater than the constant for125I-TIP(7-39) (Table 3). There was little difference between the values of the association rate constant for the three ligands (Table 3). Thus, the higher PTH1 receptor binding affinity of TIP(7-39) results from a considerably reduced rate of dissociation of the ligand from the receptor. The reliability of this indirect method for measuring the kinetic parameters was checked by comparing the kinetically derived equilibrium dissociation constant with that measured in equilibrium binding assays. For all three unlabeled ligands tested the values obtained using the two methods were in good agreement. (For [d-Trp12,Tyr34]PTH(7-34) and PTHrP(7-34) compare the values in Tables 1 and 3. TheKD of [Nle8,18,Tyr34]bPTH(3-34) for the PTH1 receptor (2.0 nM) has been reported previously (Hoare and Usdin, 1999).) Further support for the reliability of the method is provided by a reasonable agreement between the dissociation rate constant for [Nle8,18,Tyr34]bPTH(3-34) estimated by eq. 4 (0.030 ± 0.011 min−1) and the value obtained by direct measurement of125I-[Nle8,18,Tyr34]bPTH(3-34) dissociation (0.061 ± 0.002 min−1,n = 2, graphical data not shown).
Discussion
The PTH1 receptor is involved in disorders of calcium metabolism because it is the site of action of PTH and PTHrP. HHM resulting from bone resorption can be effectively treated in the long term using bisphosphonates, which inhibit resorptive processes (Singer et al., 1991; Brown and Robbins, 1999). However, the effect of these compounds is not evident until several days after treatment is initiated (Singer et al., 1991). An alternative strategy in development is neutralization of the osteoclast differentiation factor osteoprotegerin ligand by osteoprotegerin (Capparelli et al., 2000), but neither of these antiresorptive approaches target the renal effects of the PTH1 receptor. HPT can be treated surgically by parathyroidectomy but medical therapy may be required to stabilize blood calcium levels before surgery or for patients who cannot be treated surgically. Calcimemetic compounds have been proposed as potential therapies for primary HPT (Nemeth and Fox, 1999). Despite these advances, effective medical treatments for acute hypercalcemic crisis and primary HPT are lacking. PTH1 receptor antagonism may provide an alternative or complementary therapeutic strategy. However, PTH1 receptor antagonists based on the structure of PTH or PTHrP have so far not been effective (Kukreja et al., 1994; Rosen et al., 1997).
In this study we investigated the functional properties of a novel PTH1 receptor antagonist, TIP(7-39) (Hoare et al., 2000). The effects were compared with those of previously described antagonists produced by N-terminal truncation of PTH {[d-Trp12,Tyr34]PTH(7-34)} and PTHrP [PTHrP(7-34)]. The principal findings of this study are as follows: 1) TIP(7-39) acts as a purely competitive antagonist of the PTH1 receptor at the concentrations tested. 2) TIP(7-39) binds with higher affinity to the PTH1 receptor than [d-Trp12,Tyr34]PTH(7-34) or PTHrP(7-34) and displays a greater PTH1/PTH2 receptor selectivity. 3) Human plasma did not reduce the potency of any of the antagonists in the presence of protease inhibitors. 4) Specific binding of125I-TIP(7-39) to the PTH1 receptor can be measured and is well described by a simple bimolecular reaction. 5) The dissociation rate constant of 125I-TIP(7-39) is considerably lower than that of the previously described antagonist ligands. The higher PTH1 receptor binding affinity of TIP(7-39) indicates that the ligand may hold more promise for the development of highly potent, selective PTH1 receptor antagonists than PTH- or PTHrP-based peptides. The benefit of enhanced PTH1/PTH2 receptor-binding selectivity is not clear at present but such selectivity should minimize side effects resulting from blockade of the PTH2 receptor.
TIP(7-39) acts as a competitive antagonist of the PTH1 receptor at the concentrations used in this study: in assays of cAMP accumulation the peptide produces a rightward-shift of the agonist concentration-dependence curve, defined by a Schild plot slope of unity, and it does not affect the maximal stimulation produced by the agonist (Fig. 4). In [Ca2+]i assays the ligand strongly inhibits the response to rPTH(1-34) (Fig. 3). TIP(7-39) also appears to be a pure antagonist at the PTH1 receptor within the detection limits of the assays used. The ligand does not significantly activate the hemagglutinin-tagged human PTH1 receptor expressed in COS-7 cells, a highly sensitive assay in which the partial agonism of [Nle8,18,Tyr34]bPTH(3-34) can be detected (Fig. 2). [The greater sensitivity of this assay compared with that for the PTH1 receptor in HEK293 cells could be a result of a higher level of receptor expression in COS-7 cells (approximately 105 and 106receptors/cell, respectively.)] The variable results obtained for the wild-type PTH1 receptor in COS-7 cells may be a result of variable transfection efficiency. The more consistent response observed with the C-terminally modified tagged receptor versus the wild-type receptor may be result from altered receptor-G-protein coupling (Iida-Klein et al., 1995).
We found that the functional potency of the antagonist ligands was markedly less than the affinity of the ligands measured in radioligand binding assays. This observation is common in studies of PTH1 receptor antagonism (Goldman et al., 1988; McKee et al., 1988). To an extent this effect may be due to the different assay conditions used. In the cAMP accumulation assay a 37°C preincubation of TIP(7-39) with receptor before addition of the agonist reduced the antagonist potency compared with simultaneous addition of the ligands (KB values of 310 and 74 nM, respectively). This finding could be explained by degradation of the peptide in the longer incubation with the cells. The functional potency of TIP(7-39) was further increased by the addition of plasma (KB of 21 nM), which could block nonspecific binding more effectively. In contrast the radioligand binding assay used to measure antagonist binding affinity was designed to minimize ligand degradation and nonspecific binding (Hoare and Usdin, 1999). The remaining discrepancy could be explained by a lack of equilibration in the adenylyl cyclase assay, owing to the slow dissociation of rPTH(1-34) from the PTH1 receptor (Hoare et al., 1999a). Alternatively, the discrepancy may be a result of the different environments of the receptor in the two assays (cell membranes versus whole cells).
Inactivation by binding to plasma proteins has been proposed to explain the lack of in vivo efficacy of PTH1 receptor antagonists. In a previous study rat and human plasma were observed to inhibit the antagonist effect of [Leu11,d-Trp12]PTHrP(7-34) at the rat PTH1 receptor in osteosarcoma cells (Kukreja et al., 1994). In this study we examined the effect of human plasma on antagonism of the human PTH1 receptor expressed in COS-7 cells. At a concentration of 20%, plasma did not reduce the potency of TIP(7-39), [d-Trp12,Tyr34]PTH(7-34), or PTHrP(7-34) (Table 2). Plasma appeared to slightly increase antagonist potency (Table 2). One possible reason for the differences found between the two studies is the use of human versus rat PTH1 receptors. The rat PTH1 receptor binds PTH- and PTHrP-based antagonist ligands with 20-fold lower affinity than the human receptor (Jüppner et al., 1994). Depletion of the antagonist by binding to plasma proteins may have a greater effect on antagonism of the rat receptor than the human receptor because higher concentrations of the antagonist are required to effectively block the rat receptor.
The lack of PTH1 receptor antagonism in some in vivo studies and in patients with HPT may be due to complications associated with the kinetics of agonist and antagonist action. [Nle8,18,d-Trp12,Tyr34]PTH(7-34) was ineffective in a rat model of HHM and in patients with HPT, suggesting that the antagonist is ineffective when the levels of PTH or PTHrP are high at the time of the antagonist infusion (Kukreja et al., 1994; Rosen et al., 1997). However, PTH1 receptor antagonists block the effects of administered PTH or PTHrP if the antagonist is infused before the agonist (Horiuchi et al., 1983; Doppelt et al., 1986;Horiuchi and Rosenblatt, 1987; Dresner-Pollak et al., 1996). We evaluated one component of the kinetics of antagonist action, the rate of ligand association to and dissociation from the PTH1 receptor. The indirectly determined dissociation rate constant of [d-Trp12,Tyr34]PTH(7-34) and PTHrP(7-34) was very high, implying rapid dissociation of these ligands from the PTH1 receptor (t1/2 values of 13 and 9 s, respectively). Rapid dissociation may contribute to the lack of efficacy of the ligand in the studies described above, in combination with the low plasma half-life {22 min for absorption of [Nle8,18,d-Trp12,Tyr34]PTH(7-34) (Schetz et al., 1995)}. The level of receptor occupancy predicted by an equilibrium model (used to calculate the doses used in the studies above) may not have been reached if the antagonist degrades rapidly, a problem that may be exacerbated if the antagonist dissociates rapidly from the receptor. The presence of high agonist levels before antagonist administration would enhance this effect, by slowing antagonist association with the receptor and exposing the ligand in the circulation for longer. The dissociation rate constant for TIP(7-39) was much lower (t1/2 value of 14 min). The slower dissociation of this antagonist may improve the level of receptor occupancy in vivo, increasing the antagonist effect. However the effectiveness of TIP(7-39) as an antagonist in vivo will probably be most dependent on the plasma half-life, which remains to be determined.
In conclusion, we have identified a novel PTH1 receptor antagonist, TIP(7-39), that displays a more favorable in vitro pharmacological profile than antagonists derived from the structures of PTH or PTHrP. Radiolabeled TIP(7-39) provides for the first time a labeled antagonist devoid of detectable agonism for use in radioligand binding studies. TIP(7-39) should prove useful for evaluating the effectiveness of PTH1 receptor antagonism in the reduction of elevated serum calcium levels. If this utility can be demonstrated, TIP(7-39), structurally modified analogs, or more stable low-molecular-weight PTH1 receptor antagonists may provide a new therapeutic strategy for the treatment of hypercalcemia.
Acknowledgment
We gratefully acknowledge Jon Marsh for patient assistance with the measurement of intracellular calcium.
Footnotes
-
Send reprint requests to: Ted B. Usdin, Rm. 3D06, Bldg. 36, 36 Convent Dr., MSC4094, National Institutes of Health, Bethesda, MD 20892-4090. E-mail: usdin{at}codon.nih.gov
-
↵1 This study was supported by the National Institute of Mental Health, Intramural Research Program.
- Abbreviations:
- PTH1
- type 1 parathyroid hormone receptor
- PTH
- parathyroid hormone
- PTHrP
- parathyroid hormone-related protein
- HHM
- humoral hypercalcemia of malignancy
- HPT
- hyperparathyroidism
- TIP39
- tuberoinfundibular peptide of 39 residues
- DMEM
- Dulbecco's modified Eagle's medium
- GTPγS
- guanosine-5′-O-(3-thio)triphosphate
- DPBS
- Dulbecco's phosphate-buffered saline
- Received May 19, 2000.
- Accepted July 7, 2000.
- U.S. Government