Endocrinology Vol. 139, No. 10 4293-4299
Copyright © 1998 by The Endocrine Society
Type-1 Parathyroid Hormone (PTH)/PTH-Related Peptide (PTHrP) Receptors Activate Phospholipase C in Response to Carboxyl-Truncated Analogs of PTH(1–34)1
Hisashi Takasu and
F. Richard Bringhurst
Endocrine Unit, Massachusetts General Hospital, Harvard Medical
School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: F. Richard Bringhurst, M.D., Endocrine Unit, Wellman 5, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail:
bringhurst.richard{at}mgh.harvard.edu
 |
Abstract
|
|---|
The carboxyl(C)-truncated human (h) PTH (hPTH) analog hPTH(1–31),
which activates adenylyl cyclase (AC), but not protein kinase C,
in rat osteosarcoma cells, exerts an anabolic effect on rat bone
in vivo similar to that of hPTH(1–34). It has been
proposed, therefore, that this action of PTH(1–34) is mediated
exclusively by stimulation of AC via the rat type-1 PTH/PTH-related
peptide (PTHrP) receptor (PTH1R).
To determine whether this selective signaling pattern also might be a
property of the hPTH1R, we studied signal transduction via
heterologously expressed hPTH1Rs in response to activation by
hPTH(1–34), hPTH(1–31), and a C-truncated analog that does not
increase rat bone mass in vivo, hPTH(1–30). In porcine
LLC-PK1 cells that stably expressed recombinant hPTH1Rs, these three
peptides activated AC identically (EC50 = 1–2
nM). In cells with comparable expression of rat PTH1Rs, AC
activation by hPTH(1–34) and hPTH(1–31) again was identical, whereas
full activation by hPTH(1–30) required higher concentrations
(EC50 = 10 nM vs. 1
nM). Surprisingly, hPTH(1–31) fully stimulated
phospholipase C (PLC), via both species of PTH1Rs, with potency that
was similar (hPTH1Rs) or slightly reduced (rat PTH1Rs), relative to
that of hPTH(1–34). hPTH(1–30), however, was 5-fold less potent than
hPTH(1–34) in activating PLC via hPTH1Rs and showed weak and only
partial activity via the rat PTH1R. Comparable results were obtained
when human and rat PTH1Rs were transiently expressed heterologously in
COS-7 cells or homologously in HEK 293 and UMR 106–01 cells,
respectively. Binding affinities of these C-truncated peptides to human
and rat PTH1Rs were concordant with their relative potencies in
activating PLC.
We conclude that hPTH(1–31) and, to a lesser extent, hPTH(1–30) can
activate PLC, as well as AC, via both rat and human PTH1Rs.
Accordingly, a role for PLC activation in the anabolic action of PTH
in vivo cannot be excluded.
|
Introduction
|
|---|
PTH CONTROLS the differentiation and
function of target cells via specific G protein-linked cell-surface
receptors that can generate multiple intracellular second messengers.
The molecular sequences of two related PTH receptors, type 1
[PTH/PTH-related peptide (PTHrP) receptor] and type 2, have been
determined and shown to be products of different genes (1). Of these,
the type 1 receptor, referred to hereafter as the PTH1R, is activated
equivalently by both PTH and PTHrP to stimulate multiple effectors,
including adenylyl cyclase (AC), phospholipase C (PLC), and transient
elevations of cytosolic free calcium (2, 3, 4, 5, 6). The PTH1R is known to be
expressed in classical PTH target tissues, such as bone and kidney (7),
whereas the type 2 receptor (PTH2R) seems to be expressed mainly in
hypothalamus, lung, endothelium, exocrine pancreas, and other
nonclassical PTH target tissues (8).
The involvement of these various intracellular second messengers in
eliciting key cellular responses to PTH (or PTHrP) via the PTH1R is not
well understood. The complex effects of PTH on bone, most notably its
ability to increase bone mass when administered intermittently but not
continuously, have served to focus much recent interest on the roles of
PTH1R second messengers in osteoblasts, the principal PTH target cells
of bone. Accordingly, efforts have been undertaken in a number of
laboratories to identify signal-selective analogs of PTH that might be
used to activate only a subset of the normal signaling responses to PTH
in osteoblasts and thereby reveal the importance of specific PTH1R
second messengers for cellular responses of interest (9, 10). Although
PTH normally circulates mainly as the intact molecule [PTH(1–84)],
the amino-terminal fragment 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) has been shown to fully activate
the PTH1R.
In a series of recent reports, the carboxyl(C)-terminal region of human
(h) 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) has been implicated as a critical region
necessary for activation of protein kinase C (PKC) via the PTH1R (9, 11, 12, 13, 14, 15). For example, the C-truncated analog 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), which
activated AC as well as 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) in ROS 17/2 rat osteosarcoma cells,
failed to stimulate membrane-associated PKC activity in these cells,
whereas the fragment hPTH(29, 30, 31, 32) retained this PKC-stimulating
activity (11, 15). The subsequent demonstration that intermittently
administered 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) was almost as active as 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) in
stimulating increased bone mass in ovariectomized rats then was
interpreted as evidence that this anabolic effect of PTH in
vivo did not require PKC activation via the PTH1R (10, 11, 16).
Further, amino (N)-truncated analogs such as hPTH(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), which were
active in the PKC assays, neither activated AC nor exerted anabolic
actions in vivo. These findings suggested that AC activation
by PTH1Rs in osteoblasts is both necessary and sufficient for the
anabolic effect of the hormone in vivo (17).
Although plausible, these concepts are based mainly upon analyses of
PTH1R signaling conducted in a single rodent malignant cell line that
may not be representative of normal osteoblasts. Moreover, significant
species differences in ligand selectivity previously observed among
cloned PTH1Rs (18, 19) require that caution be used in extrapolating
results from rodent to human systems.
In the studies presented here, we have analyzed the impact of
C-truncation of the 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) ligand on binding to, and signaling
via, both rat and human PTH1Rs expressed in host cells that lack
endogenous functional PTH receptors. The results show that 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)
and 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) can activate PLC, as well as AC, via rat and human
PTH1Rs. The findings also demonstrate significant differences between
these two species of PTH1R and highlight the role of ligand binding
affinity, as well as that previously shown for receptor density (20, 21), in determining the balance of PTH1R signaling along the AC
vs. PLC pathways.
|
Materials and Methods
|
|---|
Cell culture
HKRK B28 cells and EW5 cells, which are clonal LLC-PK1 porcine
renal epithelial cells that stably express human and rat receptors,
respectively, were cultured under 5% CO2 in air in DMEM
containing 7% FBS and 1% penicillin/streptomycin (all from GIBCO-BRL,
Grand Island, NY). Both cell lines express similar numbers of PTH1Rs,
i.e. 280,000 sites/cell in HKRK B28 cells and 320,000
sites/cell in EW5 cells, as determined by Scatchard analysis (21).
Subcultivation and plating of cells for experiments were performed as
previously described (20, 21). COS-7 cells, HEK 293 cells, and UMR
106–01 cells were maintained in DMEM with 10% FBS and 1%
penicillin/streptomycin. These three cell lines were plated into 6-well
multiwell plates at densities required to produce confluent monolayers
1 day later.
Plasmid transfections
Confluent monolayers of COS-7 cells, HEK 293 cells, and
UMR-106–01 cells were transfected with full-length human or rat PTH1R
complementary DNAs (cDNAs) [HKRK or R15B, respectively (21)]. For
COS-7 cells, the diethylaminoethyl-dextran method was used, as
previously described (22). Otherwise, the lipofectAMINE method
(GIBCO-BRL) was employed, using a 5-h transfection at 37 C. All acutely
transfected cells were studied 3 days after transfection.
Intracellular cAMP accumulation
The procedure and buffer components used were described
previously (21). In brief, cells were washed once and incubated with
agonists in the presence of isobutylmethylxanthine (1 mM)
at 37 C for 15 min. The reactions were terminated by rapidly aspirating
the buffer and freezing the cell layers on liquid nitrogen. Accumulated
cAMP then was extracted with 50 mM HCl and measured using a
commercial RIA kit (Dupont-New England Nuclear, Boston, MA).
PLC activation
PLC activation was examined by measuring the production of
inositol 1,4,5-triphosphate (for HKRK B28 cells and EW5 cells) or total
inositol phosphates (for COS-7, HEK, and UMR 106–01 cells). Cells were
labeled with [3H]myo-inositol (3 µCi/ml) in
inositol-free DMEM containing 0.1% heat-inactivated BSA at 37 C for
16 h. After washing the cells with the same medium containing 30
mM LiCl (for HKRK B28, EW5, and COS-7 cells) or 10
mM LiCl (for HEK 293 and UMR 106–01 cells), peptides were
added and incubations carried out at 37 C for 4 min (HKRK B28 and EW5
cells) or for 30 min (COS-7, HEK 293, and UMR 106–01 cells). PLC
activation was terminated by rapid aspiration and addition of cold 5%
trichloroacetic acid (TCA). Acid extraction was conducted at 4 C
for at least 1 h before separation of the inositol
poly-phosphate fractions by ion-exchange chromatography, as
previously described (21).
Radioligand binding
Radioligand binding to PTH1Rs was measured as previously
described (5). In brief, confluent cell monolayers in 24-well plates
were rinsed and then incubated with
125I-[Tyr34]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,
with or without increasing concentrations of each nonradioactive hPTH
analog, for 6 h in the cold room (2–8 C). After aspiration of the
ligand mixture, the cells were washed three times and then solubilized
with 0.5 M NaOH-0.1% Triton X-100 for determination
of cell-associated radio-activity.
Peptides and other reagents
All reagents, unless otherwise specified, were obtained from
Sigma Chemical Co. (St. Louis, MO). All isotopes were purchased from
Dupont-New England Nuclear. All hPTH peptides were synthesized
in the Core Laboratory of the Endocrine Unit, and the C-termini of all
hPTH peptides were amidated. Radioiodination and purification of
[Tyr34]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 were performed as
previously described (5).
|
Results
|
|---|
Cyclic AMP accumulation in LLC-PK1 cells that stably express
PTH1Rs
When tested in EW5 cells, which are clonal LLC-PK1 porcine renal
epithelial cells that express 320,000 rat PTH1Rs per cell, 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),
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), and 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) each demonstrated comparable maximal
activation of AC. The EC50s for 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) and
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) were identical (1 nM), whereas that for
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) was 10-fold higher (i.e. 10 nM)
(Fig. 1B
). These results are similar to
those of previous studies using the rat osteosarcoma cell line ROS 17/2
(11, 23, 24). In LLC-PK1-derived HKRK B28 cells, however, which
expressed nearly the same number of hPTH1Rs (i.e. 280,000
per cell), the activation curves for the three peptides were identical
(EC50s = 2 nM) (Fig. 1A).
PLC activation in LLC-PK1 cells that stably expressed PTH1Rs
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) had been reported not to activate PKC in ROS 17/2
cells or murine proximal tubular cells (11, 15, 25). As shown in Fig. 2B, however, 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) fully activated
PLC in EW5 cells, although the EC50 of this peptide was 3-
to 4-fold higher than that of 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). 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) also weakly
activated PLC via the rat PTH1R, although a maximal response was not
observed with this peptide at the highest concentration investigated
(10,000 nM). The EC50 for 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),
therefore, was at least 100-fold higher than that of 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). A
different pattern was observed, however, in LLC-PK1 cells (HKRK B28)
that express hPTH1Rs. In these cells, 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) and 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)
exhibited similar activity, with indistinguishable EC50s
(approximately 200 nM) (Fig. 2A). Furthermore, 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)
achieved full agonism at 10,000 nM, with an
EC50 (1000 nM) only 5-fold higher than that of
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).
PLC activation in COS-7 cells that acutely expressed PTH1Rs
The finding that 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) activated PLC as effectively as
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) via hPTH1Rs, and nearly as well via rat PTH1Rs receptors
expressed in LLC-PK1 cells, was quite unexpected, in light of previous
reports that this peptide did not activate PKC in membranes of rodent
ROS 17/2 cells (11, 15, 16). To determine whether this was a feature
peculiar to the LLC-PK1 host cells used or, perhaps, a more general
phenomenon, key aspects of these studies were repeated using COS-7
cells that were acutely transfected with recombinant human or rat
PTH1Rs. Preliminary studies were undertaken to determine the amounts of
plasmid DNA of each type required to assure comparable levels of human
and rat PTH1R expression (1,400,000 receptors per cell) in the COS-7
cells. As shown in Fig. 3, the same
general pattern of PLC-activating potency was observed for the three
peptides with COS-7 cells that expressed human or rat PTH1Rs as with
their LLC-PK1-cell counterparts. Specifically, the activation curves
for 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) and 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) were identical in COS-7 cells that
expressed hPTH1Rs but again diverged (i.e. EC50s
were 3- to 4-fold different) in cells that expressed rat PTH1Rs. Also,
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) again was much weaker, relative to 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), via rat
than human PTH1Rs, although it did achieve full agonism via the human
receptors (as in LLC-PK1 cells). Interestingly, PLC activation via the
hPTH1R was more efficient (3- to 4-fold lower EC50) than
via the rat PTH1R in COS-7 cells (a difference not observed in LLC-PK1
cells that expressed roughly 5-fold fewer receptors of each type).
PLC activation via PTH1Rs expressed in homologous cell
systems
To further clarify whether the differences in peptide selectivity
observed between human and rat PTH1Rs in LLC-PK1 porcine or COS-7
monkey cells might relate in some way to species-specific differences
among host-cell G proteins or other such receptor-independent factors,
further experiments were carried out using completely homologous
systems. Thus, human and rat PTH1Rs were transiently expressed in HEK
293 human kidney cells and UMR 106–01 rat osteosarcoma cells,
respectively.
As shown in Fig. 4, a pattern of relative
potency for PLC activation via the hPTH1R comparable with that seen in
LLC-PK1 and COS-7 cells was observed for the three peptides during
testing in acutely transfected HEK 293 cells, as had been seen in
LLC-PK1 and COS-7 cells. Again, the activities of 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) and
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) were indistinguishable (EC50s =
approximately 10 nM), whereas 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) showed full
agonism but with reduced efficacy (EC50 = 20–30
nM).
Although UMR 106–01 cells do express endogenous rat PTH1Rs, inositol
phosphate production was not detectable in untransfected cells within
30 min at 37 C, presumably because of the relatively low number of
endogenous receptors expressed (estimated at 20,000/cell) (21, 26).
Accordingly, it was necessary to transiently transfect the UMR 106–01
cells with rat PTH1R cDNA to increase receptor expression to a level
that could support detectable PLC activation. In such rat
PTH1R-transfected UMR 106–01 cells, 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) was again less
effective than 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), and 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) showed only partial agonism
at a concentration (1000 nM) that was almost 100-fold
higher than the EC50 for 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) (Table 1).
Binding affinity of 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) and 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) to human and rat
PTH1Rs
Because we previously had observed that the efficiency of PLC
signaling by both rat and human PTH1Rs was strongly influenced by the
density of receptors expressed on the cell surface (20, 21), it seemed
likely that the differences in PLC-activating potency observed among
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), 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), and 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), in cells expressing
comparable numbers of rat and human PTH1Rs, might be caused by
differences in relative binding affinity among these peptides. We
therefore performed competitive displacement binding experiments using
the EW5 (rat PTH1R) and HKRK B28 (hPTH1R) cells. As shown in Fig. 5, 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) and 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) exhibited
similar IC50s (approximately 2 nM) in HKRK B28
cells, whereas 10-fold higher concentrations of 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) were
required for comparable displacement of the radioligand. In EW5 cells,
in contrast, the IC50s for 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), 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), and
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) all were different (5, 30, and 500 nM,
respectively). Thus, the apparent binding affinities of these three
peptides to the two species of PTH1Rs correlated quite well with their
relative PLC-stimulating activities (compare Figs. 2
and 5). In
contrast, AC activation was less strongly influenced by changes in
binding affinity and apparently only significantly impaired when
binding affinity was reduced by 100-fold or more (compare Figs. 1 and 5). These findings are concordant with previous results obtained
by varying the number of human or rat PTH1Rs expressed in these cells
(20, 21).
|
Discussion
|
|---|
Our experiments show that human and rat PTH1Rs differ, with
respect to reductions in signaling efficiency that accompany
C-truncation of 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 to the shortened peptides,
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)NH2 and 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)NH2. These
changes in signal transduction are greater with rat than with human
PTH1Rs, are more pronounced in assays of PLC than of AC activation, and
correlate with alterations in ligand-receptor binding affinity. The
greater vulnerability of PLC signaling to such structural alterations
in the hPTH ligand, evident with both species of PTH1Rs, suggests that
coupling of the PTH1R to the G protein(s) responsible for activating
PLC, such as Gq/G11/G14, in these cells must be
less robust than that to Gs, a conclusion we previously had reached
independently by analyzing a series of LLC-PK1 cell lines that stably
expressed a broad range of rat or human PTH1Rs (20, 21).
Because of this differential coupling to signaling effectors, it seems
likely that the differences in binding of the 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) and
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) peptides to the human and rat PTH1Rs could cause them to
exhibit disparate signaling selectivity via the two receptors. Thus, in
the case of the hPTH1R, the apparent binding affinity and signaling
EC50s (for both AC and PLC) of 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) are identical to
those of 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), and this peptide therefore exhibits no signaling
selectivity. The binding of 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) to the human receptor, on the
other hand, is reduced sufficiently to impair PLC activation by 5- to
6-fold, yet not enough to affect AC activation. As a result,
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) shows modest selectivity for AC over PLC via the hPTH1R.
Similar considerations would account for the relative selectivity for
AC seen with both truncated peptides via the rat PTH1R, to which each
binds with lower affinity than to the hPTH1R.
These differences in analog selectivity between human and rat PTH1Rs
seem to be intrinsic to the respective receptor molecules themselves,
because the disparities in PLC-activating efficiency were observed when
each receptor was analyzed in three different cell lines, including one
homologous system in each case. Although the precise structural
features of the two receptors that are responsible for these
differences are not yet known, our observations indicate that effects
observed with PTH analogs in rodent systems should not be assumed to be
directly applicable to humans.
Because it failed to activate PKC and yet fully activated AC in ROS
17/2 osteosarcoma cells, 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) has been considered to be a highly
signal-selective analog for the cAMP pathway in the rat (11, 15).
Accordingly, the finding that 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) and 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) induce
equivalent anabolic effects on rat bone in vivo has been
widely interpreted as evidence that this action of PTH is linked
exclusively to activation of AC, and not PLC/PKC, in bone cells. Our
results suggest that this conclusion should be viewed more cautiously.
In the three rat PTH1R-expressing cell lines studied here, 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)
activated PLC as fully as did 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), with EC50s that
were only modestly (3- to 5-fold) elevated. If the same were true in
normal bone cells in vivo, it would be difficult to exclude
a requirement for PLC signaling in this action of the hormone, given
the limited AC selectivity observed with this analog via rat PTH1Rs.
Moreover, if 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) were to activate PLC, it is unlikely that it
would fail to activate PKC in the same cell, even if PLC-independent
pathways of PKC activation also exist.
It is important to consider why our results with 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) seem to
differ so strikingly from those previously reported by others (11, 15, 25). One obvious issue is that different responses have been measured.
We measured PLC, whereas Jouishomme et al. (11, 15) measured
PKC in unextracted membranes. Because PKC may be activated
independently of PLC, it is possible that the stimulation of PKC
previously observed in ROS 17/2 cells was unrelated to activation of
PLC via the PTH1R. Second, previous work was conducted using ROS 17/2
osteosarcoma cells that expressed endogenous, rather than transfected,
rat PTH1R genes (11, 15). The ROS 17/2 cells almost certainly express
the same rat PTH1R that we have studied here, because they were the
source of the ROS 17/2.8 subclone from which the rat PTH1R cDNA used in
our studies originally was cloned (2, 5, 21). On the other hand, it is
quite possible [perhaps even likely (27, 28, 29, 30)] that ROS 17/2 cells and
spleen cells express alternate species of receptors for PTH that could
have inhibited the response(s) to 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) in these cells or that
might have supported a PLC-independent activation of PKC by hPTH(1, 2, 3, 4, 5, 6, 7,
Top
Abstract
Introduction
), and hPTH(
) was measured, as
described in Materials and Methods, in LLC-PK1 cells
that stably expressed human (A, HKRK B28 cells) or rat (B, EW5 cells)
PTH1Rs. Results were expressed as percentages of the maximal response
to hPTH(
