Endocrinology Vol. 142, No. 4 1534-1545
Copyright © 2001 by The Endocrine Society
Selective and Nonselective Inverse Agonists for Constitutively Active Type-1 Parathyroid Hormone Receptors: Evidence for Altered Receptor Conformations1
Percy H. Carter,
Brian D. Petroni,
Robert C. Gensure,
Ernestina Schipani,
John T. Potts Jr. and
Thomas J. Gardella
Endocrine Unit, Massachusetts General Hospital and Harvard Medical
School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Thomas J. Gardella, Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114. E-mail:
Gardella{at}helix.MGH.Harvard.edu
 |
Abstract
|
|---|
The spontaneous signaling activity of some G protein-coupled receptors
and the capacity of certain ligands (inverse agonists) to inhibit such
constitutive activity are poorly understood phenomena. We investigated
these processes for several analogs of PTH-related peptide (PTHrP) and
the constitutively active human PTH/PTHrP receptors (hP1Rcs)
hP1Rc-H223R and hP1Rc-T410P. The N-terminally truncated antagonist
PTHrP(5-36) functioned as a weak partial/neutral agonist with both
mutant receptors but was converted to an inverse agonist for both
receptors by the combined substitution of Leu11 and
D-Trp12. The N-terminally intact analog
[Bpa2]PTHrP(1–36)—a partial agonist with the wild-type
hP1Rc—was a selective inverse agonist, in that it depressed basal cAMP
signaling by hP1Rc-H223R but enhanced signaling by hP1Rc-T410P. The
ability of [Bpa2]PTHrP(1–36) to discriminate between
the two receptor mutants suggested that H223R and T410P confer
constitutive receptor activity by inducing distinct conformational
changes. This hypothesis was confirmed by the observations that: 1) the
double mutant receptor hP1Rc-H223R/T410P exhibited basal cAMP levels
that were 2-fold higher than those of either single mutant; and 2)
hP1Rc-H223R and hP1Rc-T410P internalized 125I-PTHrP(5–36)
to markedly different extents. The overall results thus reveal that two
different types of inverse agonists are possible for PTHrP ligands
(nonselective and selective) and that constitutively active PTH-1
receptors can access different conformational states.
|
Introduction
|
|---|
THE TYPE-1 PTH receptor (PTH/PTHrP receptor
or P1Rc) mediates the homeostatic calcium-regulating actions of PTH and
the paracrine developmental actions of PTH-related peptide
(1). Three different constitutively active mutant PTH-1
receptors have been identified in patients with Jansen’s metaphyseal
chondrodysplasia, a rare disease characterized by hypercalcemia and
short stature (2, 3, 4, 5). Each of these activating
mutations occurs near the cytoplasmic terminus of one of the
transmembrane domains: the His223
Arg mutation in TM 2
(2), the Thr410Pro mutation in TM6 (4, 5),
and the Ile458Arg mutation in TM7 (3). In COS-7 cell
expression systems, these mutations resulted in 4- to 10-fold increases
in the basal level of cAMP, relative to the basal level of cAMP in
cells expressing the wild-type PTH-1 receptor (2, 3, 4, 5).
Confirmation of the predicted in vivo hyperactivity of these
mutant receptors was provided by a recent study of transgenic mice in
which it was shown that the targeted expression of hP1Rc-H223R in
chondrocytes rescues the lethal phenotype associated with the deletion
of the gene for PTH-related peptide (6).
We have previously reported that certain competitive antagonists for
the PTH receptor act as inverse agonists with these constitutively
active PTH-1 receptors, in that they reduce the elevated basal cAMP
signaling (7). The capacity of a competitive antagonist to
behave as an inverse agonist when studied with a constitutively active
G protein-coupled receptor has been described for a number of other
ligand/receptor systems (for review see Ref. 8). The
molecular mechanisms that underlie inverse agonism and constitutive
receptor activity have been discussed, largely in theoretical terms
(9, 10), but these are still only poorly understood
processes. The continued interest in understanding both the structural
basis for inverse agonism, as well as the mechanisms by which these
ligands modulate receptor signaling activity, is likely to provide
important insights into the fundamental mechanism(s) of ligand
recognition and receptor activation for the G protein-coupled receptors
(11). In our field, potent peptidic inverse agonists for
the constitutively active variants of the PTH-1 receptor could serve as
useful reagents in studying the transgenic mouse models that are now
being developed to dissect the role of the receptor in development
(6). In addition, small molecule mimmetics of these
peptides could provide a therapeutic benefit to patients with Jansen’s
metaphyseal chondrodysplasia.
Our earlier work revealed that the N-terminally truncated
peptides
[Leu11,D-Trp12]PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2
and [D-Trp12]PTH
(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2 were inverse agonists with both the
hP1Rc-H223R and hP1Rc-T410P constitutively active receptors
(7). These two antagonist/inverse agonist peptides
exhibited 3- to 30-fold higher apparent binding affinities than did the
other (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) antagonist peptides in the study, and each contained Leu
and D-Trp at positions 11 and 12, respectively. These
observations suggested two possibilities regarding the molecular basis
for inverse agonism in PTH/PTHrP ligands: 1) all high affinity
N-terminally truncated antagonists act as inverse agonists for the
constitutively active PTH-1 receptors; and 2)
Leu11, D-Trp12,
or their combination serves as a structural determinant of inverse
agonism. We addressed these possibilities in the present study by
using two newly described high-affinity antagonist
peptides—[Ile5,Trp23,Tyr36]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)NH2
and
[Bpa2,Ile5,Trp23,Tyr36]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)NH2
(12)—both of which were previously shown to be
approximately 17-fold more potent as competitive antagonists with the
wild-type PTH-1 receptor than was
[Leu11,D-Trp12,Trp23,Tyr36]PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)NH2,
and neither of which contained the Leu11 or
D-Trp12) substitutions. As described
herein, our analysis of these peptide analogs and their derivatives
reveal that neither of our initial hypotheses was entirely correct and
that different mechanistic modes of inverse agonism in the PTH/PTHrP
system are possible. Furthermore, the discovery that
[Ile5,Bpa2,Trp23,Tyr36]-
PTHrP1–36NH2 acted as a selective inverse
agonist for hP1Rc-H223R led us to reinvestigate the mechanisms of
constitutive activation used by hP1Rc-H223R and hP1Rc-hT410P, and, in
so doing, to determine that the H223R and T410P mutations induce
different activated states of the receptor.
|
Materials and Methods
|
|---|
Peptides
Amino acid sequences of the peptides used are summarized in
Table 1
. All peptides used in this study
contained a free amino terminus (with the exception of
[desNH2-Ala1,Tyr34]hPTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2)
and a carboxamide at the C-terminus; all PTHrP(x-36) analogs contained
the modifications of Ile5 (unless truncated at
residue 7), Trp23 and Tyr36
(13). In most cases, these shared peptide modifications
are not indicated in the subsequent textual references to the peptide
analogs. The peptide
[Nle8,18,Tyr34]bPTH(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2
{PTH(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)} was purchased from Bachem (Torrance, CA).
All other peptides were prepared on a PE Applied Biosystems (Norwalk, CT) model 430A peptide synthesizer
using Fmoc main-chain protecting group chemistry, HBTU/HOBt/DIEA (1:1:2
molar ratio) for coupling reactions, and TFA-mediated
cleavage/sidechain-deprotection (MGH Biopolymer Synthesis Facility,
Boston, MA). Each peptide was purified by HPLC, lyophilized,
reconstituted in 10 mM acetic acid, and stored at -80
C. The purity, identity, and stock concentration of each compound was
secured by analytical HPLC, mass spectrometry, and amino acid analysis,
respectively. Radiolabeling was performed using
125I-Na (2,200 Ci/mmol, NEN Life Science Products) and chloramine-T; the resultant
[125I-Tyr]-ligand was purified by HPLC.
Cell culture and DNA transfection
Stock solutions of EGTA/trypsin and antibiotics were obtained
from Life Technologies, Inc. (Gaithersburg, MD); FBS was
obtained from HyClone Laboratories, Inc. (Logan, UT).
COS-7 cells were cultured at 37 C in DMEM supplemented with FBS (10%),
penicillin G (20 U/ml), streptomycin sulfate (20 µg/ml), and
Amphotericin B (0.05 µg/ml) in a humidified atmosphere
containing 5% CO2. The wild-type and mutant
PTH-1 receptor complementary DNAs (cDNAs) were contained in the pcDNA1
vector (Invitrogen, Carlsbad, CA) (5, 14).
The plasmid encoding hP1Rc-H223R/T410P was constructed by replacing a
SacI-SacI restriction DNA fragment of
hP1Rc-T410P, which extended from the 5' polylinker region to a site
corresponding to the middle of extracellular loop 2, with the
corresponding fragment of hP1Rc-H223R. The COS-7 cells were transfected
in 24-well plates when the cells were of 85–95% confluency using
DEAE-dextran (15). For each well, 200 ng of plasmid DNA
was used [except for the expression normalization studies (see Fig. 6), where varied amounts of DNA were used]. All DNA was
purified by cesium chloride/ethidium bromide gradient centrifugation.
Assays were conducted 72–96 h after transfection, at which point
approximately 20% of the cells expressed surface wild-type PTH
receptors at a density of about 5 x 106 per
cell (15).
View larger version (30K):
[in this window]
[in a new window]
|
Figure 6. Effects of combining the H223R and T410P mutations
on PTH-1 receptor function. A, COS-7 cells were transiently transfected
using plasmid DNA; 200 ng/well) encoding either hP1Rc-WT, hP1Rc-H223R,
hP1Rc-T410P, or hP1Rc-H223R/T410P, and then subsequently treated with
either buffer alone (solid columns) or buffer containing
1 µM [Nle8,21,
Tyr34]-rPTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )NH2 (hatched
columns). The data (mean ± SEM) are from
three independent experiments, each performed in duplicate or
triplicate. B and C, COS-7 cells were transfected with hP1Rc-WT DNA at
1.6 ng or 400 ng of DNA per well or with hP1Rc-H223R/T410P DNA at 400
ng/well. In panel B, the cells were treated as in A. In panel C, the
level of cell-surface expression of the receptor was measured using a
primary antibody directed against the receptor’s extracellular domain
and an iodinated secondary antibody. The data of panels B and C were
extracted from a larger experiment (performed three times in duplicate)
in which the amount of DNA used in the transfection for each receptor
varied from 0.2–400 ng per well in 2-fold increments. When the
resulting surface expression levels (and cAMP responses) were measured,
the expression levels of hP1Rc-WT and hP1Rc-H223R/T410P were most
closely matched (P = 0.5) using 1.6 and 400 ng of
DNA encoding the respective receptors. The data (mean ±
SEM) are from the three experiments performed in duplicate.
D, COS-7 were cells transfected with DNA encoding hP1Rc-H223R/T410P
(200 ng/well) and subsequently treated with varying doses of
[Ile5, Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 ( ),
[Leu11,D-Trp12, Trp23,
Tyr36]-PTHrP(7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 ( ),
[Nle8,21, Tyr34]-rPTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )NH2
(•), [Bpa2, Ile5, Trp23,
Tyr36]hPTHrP(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 ( ),
[Trp2, Ile5, Trp23,
Tyr36]hPTHrP(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 ( ),
[Arg2, Ile5, Trp23,
Tyr36]hPTHrP(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 ( ), or
[desNH2-Ala1,
Tyr34]-hPTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )NH2 (X), and then
assayed for cAMP accumulation. In panels A, B, and D, the differences
between the basal and PTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )-modulated cAMP levels for
hP1Rc-H223R/T410P were statistically different (P =
0.04, 0.00001 and 0.0002, respectively). The data (mean ±
SEM) were derived from three to six independent
experiments, each performed in duplicate; in each experiment, the cAMP
values were normalized to the cAMP level detected in the absence of
added ligand (100%).
|
|
cAMP accumulation assays
Assays of transfected COS-7 cells were performed in 24-well
plates. Cells were rinsed with 0.5 ml of binding buffer (50
mM Tris-HCl, 100 mM NaCl, 5 mM KCl,
2 mM CaCl2, 5% heat-inactivated
horse serum, 0.5% FBS, adjusted to pH 7.7 with HCl) and treated with
100 µl of binding buffer containing varying amounts of peptide analog
and 200 µl of cAMP assay buffer (DMEM containing 2 mM
3-isobutyl-1-methylxanthine, 1 mg/ml BSA, 35 mM HEPES-NaOH,
pH 7.4). The medium (total volume = 300 µl) was removed after a
30 min incubation at room temperature. The cells were then frozen (-80
C), lysed with 0.5 ml 50 mM HCl, and refrozen (-80 C). The
cAMP content of the diluted lysate was determined by RIA. Nonlinear
regression was used to calculate the EC50 values
for cAMP accumulation and to curve-fit the data (see below).
Competitive antagonism studies
The cAMP accumulation protocol described above was used for
studies of antagonist peptides with some minor modifications. Cells
were rinsed with 0.5 ml binding buffer and treated successively with
100 µl of binding buffer containing an antagonist peptide, 100 µl
of cAMP assay buffer, and 100 µl of cAMP assay buffer containing the
agonist PTH (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
([Nle8,21,Tyr34]rPTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2)
or inverse agonist
{[Leu11,D-Trp12]-PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
or
[Leu11,D-Trp12]-PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)}
for a final volume of 300 µl. The cells were then incubated for 30
min at room temperature and processed as above for quantification of
intracellular cAMP levels.
Competition binding assays
Binding reactions were performed in 24-well plates. Cells were
rinsed with 0.5 ml of binding buffer and then treated successively with
100 µl binding buffer, 100 µl of binding buffer containing various
amounts of unlabeled competitor ligand, and 100 µl of binding buffer
containing approximately 100,000 cpm of
125I-tracer (ca. 26 fmol; final volume = 300
µl). Incubations were 4 h at 15 C, except for experiments
designed for Scatchard analysis, which were incubated 6 h at 4 C.
Cells were then placed on ice, the binding medium was removed, and the
monolayer was rinsed three times with 0.5 ml of cold binding buffer.
The cells were subsequently lysed with 0.5 ml 5 N NaOH and
counted for radioactivity. The nonspecific binding for each experiment
was determined by competition with a 1 µM dose of
unlabeled
[Nle8,21,Tyr34]rPTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)NH2.
The maximum specific binding (B0) was the
total radioactivity bound in the absence of unlabeled ligand, corrected
for nonspecific binding. Nonlinear regression was used to calculate
binding IC50 values (see below).
Antibody binding analyses
Indirect antibody binding studies were performed to quantify
levels of receptor surface expression in intact transfected COS-7
cells. The primary antibody was an affinity purified rabbit polyclonal
antibody (anti-H2) that recognizes an epitope in the amino-terminal
extracellular domain of the human PTH-1 receptor, and the secondary
antibody was 125I-labeled goat antirabbit Ig. The
binding and washing steps were performed as described
(5).
Photochemical cross-linking
Photochemical cross-linking was carried out with transiently
transfected COS-7 cells as described (12). All
manipulations were executed on ice using chilled reagents. COS-7 cells
in 6-well plates were rinsed with 2.0 ml binding buffer, treated with 1
ml of binding buffer containing ca. 4 x 106
cpm (ca. 1 pmol) of
125I-[Bpa2]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36),
and incubated for 4 h at 13 C. The cells were then chilled on ice,
the medium was removed, and the cells were rinsed twice with 1 ml of
binding buffer. Next, the cells were covered with 800 µl of binding
buffer and exposed to irradiation with a Black-Ray UV lamp (366
nM, 7 mW/cm2, source-to-cell =
4.5 cm) for 15 min on ice in a cold room (4 C). The medium was
withdrawn and the cells were rinsed twice with 2 ml of ice-cold
acid-saline buffer and twice with 2 ml of ice-cold binding buffer. The
cells were lysed with 250 µl of a Triton buffer (50 mM
Tris-HCl, 10% Triton X-100, 1 mM phenylmethylsulfonyl
fluoride, 0.05 mg/ml Bacitracin, 150 mM NaCl, pH 7.8) for
30 min at 0 C before being harvested (50 µl rinse with the Triton
buffer) and centrifuged at 2,000 x g for 20 min at 4
C. The supernatant was collected and equal-volume aliquots were
analyzed using SDS-PAGE (5–20% acrylamide gradient) (16)
followed by autoradiography of the dried gel at -80 C with an
intensifying screen. For the experiment shown (see Fig. 5), the
relative cross-linking efficiency for each receptor was determined by
excising the major band (
80 kDa) from the dried gel, counting the
radioactivity, dividing the resultant CPMs by the total radioactivity
that specifically bound to that receptor (determined in a parallel
experiment in which the cells were washed with neutral buffer instead
of acid buffer), and normalizing the resulting value to the
corresponding value obtained for hP1Rc-WT.
View larger version (53K):
[in this window]
[in a new window]
|
Figure 5. Cross-linking of [Bpa2]PTHrP(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )
to wild-type and constitutively active PTH-1 receptors in COS-7 cells.
The analog 125I-[Bpa2, Ile5,
Trp23, Tyr36]PTHrP(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 was
bound to COS-7 cells transiently transfected with either hP1Rc-WT,
hP1Rc-H223R, or hP1Rc-T410P and, after washing the cells with neutral
binding buffer to remove unbound ligand, was exposed to UV irradiation.
The cells were then washed again with acid-saline buffer, lysed with
SDS loading buffer, and equal-volume aliquots of the resulting lysates
were analyzed by SDS-PAGE/autoradiography. The relative cross-linking
efficiencies calculated in this experiment for each receptor mutant,
compared with hP1Rc-WT (1.0), were 3.4 (H223R) and 1.6 (T410P). The
results were replicated in two other independent experiments.
|
|
Internalization assays
Cells were treated as above for competition binding assays, but
incubations were at room temperature. At each time point, the binding
process was terminated by the removal of the binding medium and the
cell monolayer was washed three times with either 0.5 ml of binding
buffer (pH 7.7), to determine total cell-associated cpm, or 0.5 ml of
acid-saline buffer (50 mM glycine, 150 mM NaCl,
adjusted to pH 2.5 with HCl), to determine the internalized cpm. The
cells were then lysed with 0.5 ml 5 N NaOH and counted for
radioactivity. Nonspecific binding (subtracted) was determined at each
time point and for each washing condition using COS-7 cells transfected
with pcDNA1.
Other data calculation
Calculations were performed using Microsoft Corp.
Excel. Nonlinear regression analyses of binding and cAMP dose-response
data were performed using the four-parameter equation:
yP = Min + [(Max - Min)/(1 +
(IC50/x)slope)]. The Excel
Solver function was used for parameter optimization, as described
previously (12, 17). Scatchard transformations of
homologous competition binding data performed with iodinated PTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
(26 fmol/well) and varying amounts of the same noniodinated ligand
(1.2–300 pmol/well) were used to calculate cell surface expression
levels and apparent dissociation constants
(kDapps). The calculations assumed a transfection
efficiency of 20% a cell density of 500,000 cells per well (verified
in several representative transfections), and a single class of
ligand-binding sites. The statistical significance between two data
sets was determined using a one-tailed Student’s t test,
assuming unequal variances for the two sets.
|
Results
|
|---|
Based on the hypotheses described above, we first compared the
abilities of two newly described N-terminally truncated PTHrP
antagonist analogs to function as inverse agonists in COS-7 cells
transiently transfected with the constitutively active mutant PTH
receptors hP1Rc-H223R and hP1Rc-T410P. As previously found for
[Leu11,D-Trp12]-PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34),
[Leu11,D-Trp12]-PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)NH2
depressed intracellular cAMP levels in a dose-dependent fashion in
cells expressing either hP1Rc-H223R or hP1Rc-T410P (Fig. 1, B and C). At the highest peptide
concentration tested, the maximum reductions achieved for the H223R and
T410P receptors were 42% and 51% of the corresponding basal cAMP
levels, respectively, and these reductions from basal were significant
(P < 0.0001). In contrast,
PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)NH2 did not act as an inverse agonist
with either hP1Rc-H223R or hP1Rc-T410P, but instead, induced a weak
partial agonist response with hP1Rc-H223R (27% of the maximum
PTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)-induced response for this receptor) and was nearly inert
with hP1Rc-T410P (Fig. 1, B and C). The inability of PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) to
behave as an inverse agonist with the two constitutively active
receptors was not due to weak binding affinity, as the
IC50 values observed for this analog in
competition binding analyses performed with either mutant receptor were
3- to 4-fold lower than the corresponding IC50
values observed for
[Leu11,D-Trp12]PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
(hP1Rc-H223R: 7.2 ± 1.8 nM vs.
28 ± 9 nM, P = 0.02; hP1Rc-T410P: 6.0
±1.5 nM vs. 22 ± 9
nM, P = 0.05, Table 2).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. cAMP-signaling responses of PTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) and
N-terminally truncated PTHrP(x-36) analogs in COS-7 cells expressing
wild-type and constitutively active PTH-1 receptors. Shown are the
effects of the analog ligands, [Nle8,21,
Tyr34]rPTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )NH2, (•),
[Ile5, Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 (), or
[Leu11,D-Trp12, Trp23,
Tyr36]hPTHrP(7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 () on the
cAMP-signaling properties of hP1Rc-WT (A), hP1Rc-H223R (B), or
hP1Rc-T410P (C) in transiently transfected COS-7 cells. The cells were
treated with varying doses of each peptide for 30 min at room
temperature, and the resulting intracellular cAMP levels were
quantified by RIA. The dotted lines in B and C indicate
the basal cAMP levels. The data (mean ± SEM) are from
three independent experiments, each performed in duplicate.
|
|
The experiments described above eliminated our first possibility
(vide supra), in that not all high-affinity N-terminally
truncated PTHrP fragments exhibited inverse agonism. To address the
second possibility, namely that inverse agonism arises from the
specific substitutions at positions 11 and/or 12, we synthesized three
peptides—[Leu11]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36),
[D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
and
[Leu11,D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
(Table 1
)—and tested their effects on cAMP-signaling by the
wild-type and constitutively active mutant receptors. In cells
expressing hP1Rc-WT, none of the three PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) analogs produced a
detectable change in basal cAMP levels (data not shown). In cells
expressing the mutant receptors, the singly substituted analogs,
[Leu11]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) and
[D-Trp12]-PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36),
behaved similarly to the unmodified PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) parent peptide and
induced weak increases or no change in cAMP production levels (Fig. 2, A and B). However, the doubly
substituted analog
[Leu11,D-Trp12]-PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
depressed cAMP levels with both hP1Rc-H223R and hP1Rc-T410P (Fig. 2, A
and B). Dose response analyses (Fig. 2, C and D) indicated that with
both hP1Rc-H223R and hP1Rc-T410P the potency of
[Leu11,D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
as an inverse agonist was not significantly different from that seen
with
[Leu11,D-Trp12]PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36),
as the EC50s obtained for the respective peptides
with hP1Rc-H223R were 59 ± 20 nM vs. (34 ± 12
nM, P = 0.2) and with hP1Rc-T410P they were
63 ± 12 nM vs. 65 ± 21
nM (P = 0.5).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 2. Effects of position 11 and 12 modifications on the
inverse agonism of PTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 ). A and B, COS-7 cells were transiently
transfected with either hP1Rc-H223R (A) or hP1Rc-T410P (B) and treated
with buffer alone (basal), buffer containing [Ile5,
Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2, or buffer
containing an analog of that same peptide with the substitutions
of Lys11Leu, Gly12D-Trp, or
both. The peptide concentrations were 3 µM and the
treatment was for 30 min at room temperature. C and D, COS-7 cells were
transiently transfected with hP1Rc-H223R (C) or hP1Rc-T410P (D), and
treated with varying doses of [Ile5,
Leu11,D-Trp12, Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 () or
[Leu11,D-Trp12, Trp23,
Tyr36]hPTHrP(7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 (•) for 30 min at
room temperature. The dashed lines indicate the basal
cAMP levels. Shown are data (mean ± SEM) from three
(A, B) or four (C, D) independent experiments, each performed in
duplicate.
|
|
Although the introduction of both the Leu11 and
D-Trp12 substitutions was sufficient
to confer inverse agonism to PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36), it had a surprisingly
deleterious effect on receptor-binding affinity. Thus, with each
receptor, the IC50 observed for the disubstituted
peptide was approximately 15-fold higher than that observed for
PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) (P < 0.002). This reduction in apparent
binding affinity could be largely attributed to the
Gly12D-Trp
substitution, which by itself reduced affinity approximately 45-fold,
relative to the binding affinity that PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) exhibited for each
receptor (P < 0.05) (Table 2). This result contrasts
with the 5- to 13-fold improvement in binding affinity that the
Gly12D-Trp substitution produced in
PTH(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) and PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) (18, 19). Consistent with the
high apparent binding affinity of unmodified PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36), relative to
that of
[Leu11,D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
and
[Leu11,D-Trp12]PTHrP(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36),
PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) could inhibit the inverse agonist actions of the latter
two peptides (Fig. 3, A and B, and data
not shown). We tested whether or not the inverse agonism inherent to
[Leu11,D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
would compensate for its weaker binding affinity when assayed for
competitive antagonism of PTH (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) at the wild-type PTH-1 receptor.
As shown in Fig. 3C, this was not the case, as
[Leu11,D-Trp12]PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
was approximately 4-fold less effective in inhibiting
PTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)-induced cAMP signaling at the wild-type receptor than was
the unmodified PTHrP(5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) analog (IC50s =
590 ± 170 nM vs. 145 ± 30
nM, respectively, P = 0.02), and
a lower maximum inhibition was observed for the modified peptide than
for the parent peptide (47 ± 7% vs. 79 ± 3%,
P = 0.001).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Antagonism of inverse agonists and classical
agonists by PTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 ) analogs. A and B, Shown is the capacity of the
near-neutral antagonist analog, [Ile5, Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2, to block the inverse
agonist action of [Leu11,D-Trp12
Trp23, Tyr36]-PTHrP(7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 in
COS-7 cells expressing hP1Rc-H223R (A) or hP1Rc-T410P (B). The cells
were treated with varying doses of the PTHrP(7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 ) inverse agonist
analog alone (•) or with a constant dose (1 µM) of the
PTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 ) analog (). Note the small partial agonist effect that
the PTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 ) analog has on these two mutant receptors. C, Shown are
the capacities of [Ile5, Trp23,
Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2 () or
[Ile5, Leu11,D-Trp12,
Trp23, Tyr36]hPTHrP(5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )NH2
() to inhibit the agonist response induced by [Nle8,21,
Tyr34]-rPTH(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )NH2 (3 nM) in
COS-7 cells transiently transfected with the wild-type hP1Rc. All data
(mean ± SEM) were derived from three independent
experiments, each performed in duplicate.
|
|
In parallel with the studies described above on truncated peptides, we
also examined the antagonist [Bpa2]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
(12) (Table 1), as well as several new structurally
related positition-2 modified PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) analogs, for inverse agonism
on the constitutively active mutant PTH-1 receptors (Fig. 4). These experiments allowed us to test
whether or not the Leu11 and
D-Trp12 modifications, along with the
N-terminal truncation, were strictly required for inverse agonism.
The effect of these position-2-substituted PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) analogs
on cAMP accumulation in COS-7 cells expressing the wild-type human
PTH-1 receptor is shown in Fig. 4A, and their receptor-binding
properties are presented in Table 3. Each
of the peptides displayed a diminished maximum signaling response with
hP1Rc-WT, in comparison to PTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), and the rank order of
potency/efficacy was: PTH(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) >
[Arg2]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) >
[Trp2]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) >
[Bpa2]PTHrP(2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36) >
[D-Bpa2]PTHrP(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26
Top
Abstract
Introduction
