Full text of DOI:10.1359/jbmr.2001.16.3.441

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 3, 2001
©
2001 American Society for Bone and Mineral Research
Stimulation of Protein Kinase C Activity in Cells
Expressing Human Parathyroid Hormone Receptors
by C- and N-Terminally Truncated Fragments
of Parathyroid Hormone 1-34
1
1
1
1
1
J.F. WHITFIELD, R.J. ISAACS, B. CHAKRAVARTHY, S. MACLEAN, P. MORLEY,
1
2
2
G. WILLICK, P. DIVIETI, and F.R. BRINGHURST
ABSTRACT
The parathyroid hormone (PTH) fragment PTH(1-34) stimulates adenylyl cyclase, phospholipase C (PLC),
and protein kinase C’s (PKCs) in cells that express human, opossum, or rodent type 1 PTH/PTH-related
protein (PTHrP) receptors (PTHR1s). Certain carboxyl (C)-terminally truncated fragments of PTH(1-34),
such as human PTH(1-31) [hPTH-(1-31)NH ], stimulate adenylyl cyclase but not PKCs in rat osteoblasts or
2
PLC and PKCs in mouse kidney cells. The hPTH(1-31)NH peptide does fully stimulate PLC in HKRK B7
2
porcine renal epithelial cells that express 950,000 transfected hPTHR1s per cell. Amino (N)-terminally
truncated fragments, such as bovine PTH(3-34) [bPTH(3-34)], hPTH(3-34)NH , and hPTH(13-34), stimulate
2
PKCs in Chinese hamster ovary (CHO) cells expressing transfected rat receptors, opossum kidney cells, and
rat osteoblasts, but an intact N terminus is needed to stimulate PLC via human PTHR1s in HKRK B7 cells.
We now report that the N-terminally truncated analogs bPTH(3-34)NH and hPTH(13-34)OH do activate
2
PKC via human PTHR1s in HKRK B7 cells, although less effectively than hPTH(1-34)NH and hPTH(1-
2
3
1)NH . Moreover, in a homologous human cell system (normal foreskin fibroblasts), these N-terminally
2
truncated fragments stimulate PKC activity as strongly as hPTH(1-34)NH and hPTH(1-31)NH . Thus, it
2
2
appears that unlike their opossum and rodent equivalents, hPTHR1s can stimulate both PLC and PKCs when
activated by C-terminally truncated fragments of PTH(1-34). Furthermore, hPTHR1s, like the PTHR1s in rat
osteoblasts, opossum kidney cells, and rat PTHR1-transfected CHO cells also can stimulate PKC activity by
a mechanism that is independent of PLC. The efficiency with which the N-terminally truncated PTH peptides
stimulate PKC activity depends on the cellular context in which the PTHR1s are expressed. (J Bone Miner Res
2
001;16:441–447)
Key words: parathyroid hormone, receptors, adenylyl cyclase, protein kinase C
INTRODUCTION
late bone growth in rodents, monkeys, and osteoporotic
humans, but fragments that lack an intact N terminus do not
NTERMITTENT SUBCUTANEOUS injections of parathyroid hor- stimulate bone growth, at least in rats. Because differ-
mone (PTH) or of N-terminal fragments such as human ences in PTH receptor signal transduction induced by these
(1)
I
PTH(1-34) [hPTH-(1–34)NH ] or hPTH(1-31)NH stimu- various peptides are presumed to account for these obser-
2
2
1
2
Institute for Biological Sciences, National Research Council of Canada, Ottawa, Ontario, Canada.
Endocrine Unit, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA.
4
41

4
42
WHITFIELD ET AL.
vations, it is important to find out what signals are generated When LLC-PK1 cells express heterologous rat receptors,
when these fragments activate human receptors.
hPTH(1-31)NH and hPTH(1-34)NH equally stimulate ad-
2
(14)
2
(2,3)
Jouishomme et al.
found that hPTH fragments needed enylyl cyclase.
However, in contrast to the inability of
amino acids 29–32 to stimulate membrane-associated pro- hPTH(1-31)NH to stimulate PKC activity in ROS 17/2 rat
2
tein kinase-C (PKC) activity in ROS 17/2 rat osteosarcoma cells, hPTH(1-31)NH can stimulate PLC (IP production)
2
3
cells. Thus, although N-terminally truncated fragments via rat PTHR1s in the porcine cells, although only 25–30%
(14)
could not stimulate adenylyl cyclase, they could stimulate as effectively as hPTH(1-34)NH2.
In LLC-PK1 cells
PKC activity provided residues 29–32 were present. In expressing human PTHR1s the two fragments are equipo-
(15)
contrast, the C-terminally truncated hPTH(1-31)NH pep- tent stimulators of adenylyl cyclase and IP production.
2
3
tide fully stimulated adenylyl cyclase but not PKC activity. Importantly, PTH also needs an unmodified N terminus to
Increasing the bioactivity of this peptide (as indicated by a stimulate IP production in PTHR1-expressing porcine
3
(16)
1
6
-fold lower concentration for half-maximally stimulating cells.
Thus, removing the ␣-amino group from Ser of
adenylyl cyclase, the EC ) by replacing Lys with Leu and hPTH(1-34)NH or removing the N-terminal Ser eliminates
linking Glu to Lys to produce the [Leu ]cyclo(Glu - PLC stimulation without affecting binding to PTHR1 or ad-
Lys )hPTH(1-31)NH lactam still did not enable it to stim- enylyl cyclase activation.
2
7
1
5
0
2
2
2
26
27
22
2
6
(15,16)
Thus, N-terminally truncated
2
(4,5)
ulate PKC activity in ROS 17/2 cells.
By contrast, certain N-terminally modified or truncated cells expressing large numbers of hPTHR1s or, surprisingly,
PTH fragments do not stimulate IP production in LLC-PK1
3
(
15)
PTH fragments that are not osteogenic (at least in rats) such rat PTHR1s.
as 1-desamino-hPTH(1-34), bovine PTH(3-34) [bPTH(3- Because PLC activation is a common mechanism for
4)], hPTH(28-34), hPTH(13-34), hPTH(28-48), and activating PKC, and because both PLC and PKC can be
3
hPTH(3-84) had little or no ability to stimulate adenylyl activated by recombinant PTHR1s, these disparities in PLC
cyclase; yet all of them, especially the more potent versus PKC activation by N- and C-truncated PTH frag-
hPTH(13-34), strongly stimulated PKC activity in ROS ments seem surprising. They could arise from intrinsic
(
1–3)
1
7/2 cells.
In rat osteosarcoma cells, PTH(3-34) and differences between hPTHR1s and opossum and rat
2
ϩ
PTH(7-34) cause a rapid cytosolic Ca surge although they PTHR1s, from differences in cellular PTHR1 expression or
do so less effectively than rat PTH(1-34). PTH(3-34) can effector coupling or in the case of N-truncated fragments
stimulate phospholipase C (PLC; as indicated by increased from PLC-independent PKC-activating mechanisms. To ad-
(6)
inositol-1,4,5-tris phosphate [IP ]) as much as PTH(1-34), dress these questions, we have measured the activation of
3
and both PTH(3-34) and PTH(7-34) can stimulate PKC membrane PKC activity by hPTH fragments in LLC-PK1 cells
(7)
activity in UMR 106–01 rat osteosarcoma cells. PTH(3- expressing hPTHR1s, for which adenylyl cyclase and PLC
4), PTH(28-42), and PTH(28-48) also strongly stimulate PKC responses are known in detail and in a homologous human
activity but not adenylyl cyclase activity, in Chinese hamster foreskin fibroblast system (fibroblasts have PTH/PTHrP
3
(
17–22)
(14–16)
ovary (CHO) AP-1 cells expressing transfected rat type 1 receptors
PTH/PTH-related protein (PTHrP) receptors (PTHR1s).
). We will show that PLC-stimulating
(8)
C-terminally truncated fragments such as hPTH(1-31)NH2
These signaling patterns are not limited to rat osteoblasts and hPTH(1–28)NH stimulate PKC activity as efficiently as
2
or hamster cells with transfected rat PTHR1s. hPTH(1- hPTH(1-34)NH in LLC-PK1 cells, whereas N-terminally
2
3
1)NH stimulates adenylyl cyclase fully, but cannot acti- truncated fragments that do not detectably stimulate IP
2
3
(14–16)
vate PLC in mouse kidney cells although neither bPTH(3- production in LLC-PK1 cells
stimulate PKC activity
3
4) nor hPTH(7-34) can stimulate adenylyl cyclase or PKC in human fibroblasts.
(9)
activity in these cells. hPTH(28-48) stimulates the expres-
sion of insulin-like growth factor I (IGF-I) and cartilage
growth in neonatal mice.
PTH(28-48), and PTH(28-34) can stimulate PKC activity as
effectively as PTH(1-34)
(
10)
hPTH(3-34), PTH(28–42),
MATERIALS AND METHODS
PTH fragments
hPTH(1-34)NH , hPTH(1-31)NH , and hPTH(1–28)NH
2
(
11,12)
whereas hPTH(1-30) can
stimulate adenylyl cyclase but not PKC activity in OK
opossum kidney cells.
2
2
(
11–13)
were synthesized in the Institute for Biological Science
It is clear from these observations that PTHR1s on mouse using the Fmoc protocol and a continuous-flow peptide
and opossum kidney cells and those on neoplastic rat os- synthesizer (model 9050; PerSeptive Biosystems, Framing-
(23)
teoblasts respond similarly to C-terminally truncated PTH ham, MA, USA) as described previously.
fragments. Moreover, the receptors on mouse bone and bPTH(3-34)NH2 and hPTH(13-34)OH were purchased
Bovine
3
4
cartilage cells, opossum kidney cells, rat osteoblasts, and from Bachem (Torrance, CA, USA). [Tyr ]hPTH(3-
CHO cells expressing rat receptors, though not the receptors 34)NH was synthesized by the Peptide and Oligonucleo-
2
on mouse kidney cells, also respond similarly to tide Core Laboratory of the Massachusetts General Hospi-
N-terminally truncated PTH fragments. The question re- tal’s Endocrine Unit.
mains as to what signals hPTHR1s use to stimulate bone
formation in PTH-treated osteoporotic humans.
One approach to answering this question has been to use
Cell culture: HKRK B7 cells
porcine LLC-PK1 kidney cells, which have no endogenous
The derivation of HKRK B7 cells, stably expressing
PTHR1s, to compare the signaling properties of exogenous about 950,000 transfected PTHR1s/cell, from LLC-PK1
(14,15)
stably transfected hPTHR1 and rat PTHR1 receptors.
porcine renal epithelial cells, has been described previous-

PTH RECEPTORS AND PKCs
443
(14,15)
ly.
The cells were grown in an antibiotic-free complete already in the membranes at the time of exposure to PTH or
medium consisting of 93% (vol/vol) high-glucose (4.5 TPA and the PKCs that had subsequently moved from the
g/liter) Dulbecco’s modified Eagle’s medium (DMEM; cytosol to the membrane.
Sigma Chemical Co., St. Louis, MO, USA) and 7% (vol/
The peaks, troughs, and relative increases of PKC activity
(15)
vol) heat-inactivated fetal bovine serum.
The cultures in the treated cultures varied from experiment to experiment
were incubated at 37°C in 5% CO and 95% air.
depending on unavoidable fluctuations in the states of the
2
6
For each experiment, 0.5 or 0.6 ϫ 10 cells in 6 ml of cells.
medium were plated in 60-mm plastic dishes and the cul-
tures were allowed to grow 3 days or 4 days until they were
nearly confluent. Two hours before adding a PTH fragment,
the complete medium was discarded, the cultures were
Cell culture: human dermal fibroblasts
Human dermal fibroblasts from newborns’ foreskins were
washed twice with phosphate-buffered saline (PBS), the provided by Dr. Eric DesRosiers of Apotex Research, Inc.
complete medium was replaced with 3 ml of serum-free (Weston, Ontario, Canada) as frozen second-passage cells.
DMEM, and cultures were returned to 37°C in the 5% They were incubated at 37°C in a complete medium con-
CO /air atmosphere. The cells were exposed to a PTH sisting of 90% low-glucose DMEM and 10% heat-
2
fragment for 3 minutes (the time of peak activity) at 37°C. inactivated fetal bovine serum in an atmosphere of 5% CO2.
The cultures were then washed with ice-cold PBS. The PBS For an experiment, cells from confluent cultures in T25
was discarded and the culture dishes put on ice and 1.2 ml flasks were plated in 60-mm dishes. The cultures were then
(2)
of lysis buffer was added. Two minutes later, the cells left until they were completely confluent before starting an
were scraped with a rubber policeman. The resulting sus- experiment.
pensions were put into a 15-ml centrifuge tube and vortexed
The assay for membrane-associated PKC activity in the
for 2 minutes. The lysates were centrifuged at 600g for 10 human fibroblasts was basically the same as that for the
minutes at 4°C and the postnuclear supernatant fractions activity in HKRK B7 cells. However, the fibroblasts were
were pelleted by centrifugation at 100,000g for 10 minutes much more resistant to the lysis buffer and needed two, 5-s
at 4°C. The PKC activity associated with these membranes sonications to be lysed in the buffer. We used 4 ␮g instead
was measured without prior extraction, reconstitution, or of 10 ␮g of protein per assay and the time of peak PKC
(
24,25)
artificial activation according to Chakravarthy et al.
The level of PKC activity in equal amounts of membrane B7 cells.
protein was indicated by the extent of phosphorylation by
The adenylyl cyclase activity in human fibroblasts was
activity was 10 minutes instead of the 3 minutes in HKRK
3
the pelleted membranes of the PKC-specific peptide Ac- expressed as the rate of formation of [ H]cyclic adenosine
FKKSFKL (acetyl-Phe-Lys-Lys-Ser-Phe-Lys-Leu-NH2), monophosphate (cAMP) from the cellular adenosine
which corresponds to residues 160–166 of the membrane- triphosphate (ATP) pool that had been labeled with
associated myristoylated alanine-rich C-kinase substrate [ H]adenine before the cells were exposed to a PTH frag-
MARCKS) protein, the standard marker of PKCs activa- ment. The cells were incubated for 10 minutes after
3
(2)
(
(22)
tion in intact cells.
The extent of phosphorylation of this adding a peptide. The reaction was stopped with 10% tri-
peptide substrate was determined as described by Chakra- chloroacetic acid, and the [ H]cAMP was then separated
3
(24,25)
(26)
(2)
varthy et al.
and Whitfield et al.
and expressed as and measured.
3
2
counts per minute of P radioactivity per microgram of
Ac-FKKSFKL substrate. As expected, the phosphorylation
of the Ac-FKKSFKL substrate by the membranes from control
and PTH-treated cells was prevented by including in the
Statistical analysis
All data were expressed as means Ϯ SEM. Comparisons
assay mixture the RFARKGALRQKVNHEVKN peptide were made by one-way analysis of variance (ANOVA).
corresponding to the autoinhibitory 19–36 pseudosubstrate When significant effects were found, Scheffe’s or Fisher’s
domain of PKCs at concentrations that are known not to affect projected least significant difference (PLSD) tests were used
other protein kinases in intact cell membranes such as those for multiple comparisons and p Ͻ 0.05 was considered to be
used in this study. Cultures used as positive controls significant.
were exposed to the standard PKC activator 12–0-
tetradecanolyphorbol-13-acetate (TPA), at a concentra-
tion of 1 ␮M.
RESULTS
This assay measures activity of membrane-associated
PKCs in situ and not a shift of activated PKCs from the
cytosol to membranes. Substantial fractions of cellular
HKRK B7 cells
We first determined the optimal time at which to measure
PKCs may reside on membranes in an inactive but stimu- membrane-associated PKC activity after exposing HKRK
latable state, and the extent of translocation of enzyme B7 cells to TPA or PTH fragments. The stimulation of PKC
activity from cytosol to membranes may not be an accurate activity by the positive control agent, 1 ␮M TPA, was
(25)
or sufficiently sensitive measure of PKC stimulation.
1.7-fold by 5 minutes, after which it increased again to
Therefore, we did not measure the cytosolic PKCs in the 2.7-fold higher than controls at 7 minutes (Fig. 1). The PKC
present experiments because the actual peak membrane– activity peaked around 3 minutes after adding 50 nM
associated PKC activities in the PTH- or TPA-stimulated hPTH(1–28)NH or hPTH(1-34)NH , and it returned to the
2
2
cells were the sums of the activities of the PKCs that were control level by 5 minutes. Therefore, measurements of the

4
44
WHITFIELD ET AL.
FIG. 1. Changes in PKC activity in HKRK B7 cells with time after
addition of 1 ␮M TPA, 50 nM hPTH(1-34)NH , or 50 nM hPTH(1–
2
2
8)NH . The points are the combined means Ϯ SEMs of the results
2
from three separate experiments.
effects of various PTH fragments on PKC activity were
carried out 3 minutes after their addition to the cultures, a
time when IP production also peaks in PTH-treated HKRK
3
(16)
B7 cells.
The response to increasing concentrations of hPTH(1–
(2)
2
8)NH tended to be biphasic, as in ROS 17/2 rat cells.
FIG. 2. Stimulation of PKC activity in HKRK B7 cells by hPTH(1-
2
The activity reached peak values at 0.25 nM (1.78 Ϯ 0.21- 34)NH
and two of its C-terminally truncated fragments hPTH(1-
and hPTH(1–28)NH . Each point is the combined mean Ϯ
2
2
2
fold higher than controls) and again at 25 nM (1.79 Ϯ 31)NH
SEM of measurements of PKC activity from three separate experi-
ments. The actual membrane-associated PKC activities in the untreated
control cultures were 2322 Ϯ 266 cpm, 1961 Ϯ 79 cpm, and 2552 Ϯ
0
.26-fold higher than controls; Fig. 2). The response to
hPTH(1-31)NH was also biphasic. The activity peaked at
2
0
.25 nM at 1.54 Ϯ 0.10 times the control value, dropped
between 0.25 nM and 5 nM, and then rose again to 1.83 Ϯ
.24 times the control activity at 500 nM (Fig. 2). There was
3
2
1
38 cpm of P radioactivity per microgram of FKKSFKL peptide
substrate, respectively. The (*) superscript indicates that the value is
significantly different (p Ͻ 0.05) from control.
0
a similar biphasic response after exposure to various con-
centrations of hPTH(1-34)NH (Fig. 2). The three peptides
2
equally stimulated PKC activity when tested in the same
experiment at a concentration of 500 nM (data not shown). PKC activity was hPTH(13-34)OH (Fig. 5). For example,
Smaller peptides hPTH(1–27)NH and hPTH(1–26)NH did the PKC activity in cultures exposed to 10 nM hPTH(13-
2
2
not increase PKC activity at any concentration between 0.1 34)OH was 4.83 Ϯ 0.16-fold greater than in control cul-
and 500 nM (data not shown).
tures.
We considered the possibility that the enhanced PKC
significantly (p Ͻ 0.05) in- responses to the N-terminally truncated PTH fragments
Bovine PTH(3-34)NH , which stimulates PKC activity in
2
(2,8)
ROS 17/2 and other cells,
creased PKC activity in HKRK B7 cells 1.29 Ϯ 0.02-fold at could have been triggered by a PTH receptor different from
0 nM (Fig. 3). Similar results (not shown) were observed PTHR1 such as the type 2 PTH receptor (the receptor for the
5
3
4
with [Tyr ]hPTH(3-34)NH , a peptide that does not acti- distantly related neuropeptide tuberoinfundibular protein
2
(15,16)
vate PLC in HKRK B7 cells.
hPTH(13-34)OH, which [TIP] 39), which is not activated by PTHrP, or maybe the
was the most potent stimulator of PKC activity in ROS 17/2 PTH receptor in keratinocytes and lymphocytes, which
(2)
cells, increased PKCs activity 1.16 Ϯ 0.03-fold at 25 nM stimulates PKC activity but cannot stimulate adenylyl cy-
(26,27)
(Fig. 3).
clase.
However, both possibilities were ruled out by
the nearly equal abilities of hPTH(1-34)NH2 and
hPTHrP(1-34)NH to stimulate adenylyl cyclase in these
cells (Fig. 6).
Human dermal fibroblasts
2
hPTH(1-34)NH significantly (p Ͻ 0.05) increased PKC
2
activity 2.69 Ϯ 0.36-fold at 0.5 nM in human foreskin
fibroblasts (Fig. 4); however, the response was not biphasic
like it was in HKRK B7 cells. As in the HKRK B7 cells,
DISCUSSION
there was a similar response to hPTH(1-31)NH (Fig. 4). In
The ability of N-terminally truncated PTH fragments
2
contrast to the PKC activity in HKRK B7 cells, PKC such as bPTH(3-34)NH and hPTH(13-34)OH to stimulate
2
activity in human fibroblasts was stimulated as strongly by PKC activity, even if only weakly, in HKRK B7 porcine
hPTH(3-34)NH as by hPTH(1-34)NH or hPTH(1-31)NH
(
cells with their high density of hPTHR1s contrasts with the
Fig. 5). In fact, the strongest stimulator of dermal fibroblast inability of such N-terminally truncated hPTH fragments to
2
2
2

PTH RECEPTORS AND PKCs
445
3
4
FIG. 5. Ability of [Tyr ]hPTH(3-34)NH2 and hPTH(13-34)OH to
stimulate PKC activity in human dermal fibroblasts. The results are the
combined means Ϯ SEMs from three separate experiments. The actual
PKC activities in the untreated control cultures were 3661 Ϯ 303 cpm
and 2232 Ϯ 72 cpm of P radioactivity per microgram of FKKSFKL
peptide substrate, respectively. The (*) superscript indicates that the
value is significantly different (p Ͻ 0.05) from control.
FIG. 3. Stimulation of PKC in HKRK B7 cells by the N-terminally
truncated fragments bPTH(3-34)NH and hPTH(13-34)OH. The points
2
3
2
are the combined means Ϯ SEMs of the results of three or four separate
experiments. The actual PKC activities in the untreated control cultures
3
2
were 1957 Ϯ 156 cpm and 2577 Ϯ 53 cpm of P radioactivity per
microgram of FKKSFKL peptide substrate, respectively. The (*) su-
perscript indicates that the value is significantly different (p Ͻ 0.05)
from control.
FIG. 6. Ability of hPTH(1-34)NH and hPTHrP(1-34)NH to stim-
2
2
ulate adenylyl cyclase in normal human dermal fibroblasts. Points are
the combined means Ϯ SEMs from four separate experiments.
FIG. 4. Ability of hPTH(1-31)NH and hPTH(1-34)NH to stimulate
2
2
PKC activity in normal early passage human dermal fibroblasts. The via the PTHR1. The existence of such a mechanism also is
curves are combined means Ϯ SEMs from three and five separate
suggested by the fact that despite the stimulation of PKC
activity by both N- and C-terminally truncated fragments in
experiments, respectively. The actual PKC activities in the untreated
3
2
control cultures were 1852 Ϯ 99 cpm and 4104 Ϯ 111 cpm of
P
human fibroblasts, we have been unable to detect either a
radioactivity per microgram of FKKSFKL peptide substrate, respec-
tively. The (*) superscript indicates that the value is significantly
different (p Ͻ 0.05) from control.
2
ϩ
rise in IP or the Ca surge that should have been triggered
3
if there had been an increase in IP production (data not
3
shown). The absence of a detectable rise in IP in these
3
fibroblasts could have been caused by them having a low
PTHR1 density, which is known to reduce the efficiency of
(
15–16)
stimulate IP production in these cells.
The ability of
3
(14)
PLC activation.
says were carried out as in Whitfield et al.
Indeed, when competitive binding as-
hPTH(3-34)NH to activate the PTHR1 receptors also is
indicated by the ability of the fragment to stimulate
mitogen-activated protein (MAP) kinase activity and mod-
2
(29)
we found that
PTH binding, hence the receptor density, was too low for an
accurate Scatchard analysis (data not shown). The efficiency
(28)
erately stimulate the proliferation of HKRK B7 cells.
The failure of the N-terminally truncated fragments to stim- of such an alternative mechanism appears to depend on the
ulate IP production but at the same time stimulate PKC and type of cell expressing the PTHR1 because the same
3
MAP kinase activities and the proliferation of these cells N-terminally truncated PTH fragments stimulated PKC ac-
points to the existence of a PLC-independent mechanism of tivity at least as strongly as the N-terminally intact hPTH(1-
PKC activation by N-terminally truncated PTH fragments 34)NH in the dermal fibroblasts.
2

4
46
WHITFIELD ET AL.
As noted previously, the ability of a PTH peptide to 34), as indicated by an equally increased urinary cAMP
(14)
stimulate PLC or PKC is a function of receptor density,
concentration. However, the truncated peptide is far less
the receptor species, and the availability within the target effective than hPTH(1-34) in stimulating bone resorption,
cell of the components needed to couple the receptor to this as indicated by its failure to increase either the plasma
mechanism. The dependence of PTHR1’s signaling proper- calcium level or the urinary cross-linked N-terminal colla-
ties on the nature of the target cell is shown by the ability of gen telopeptides. The same result has been observed in
an N-terminally truncated PTH fragment to stimulate PKC mice: hPTH(1-31)NH stimulates osteoblast activity as ef-
2
strongly (relative to hPTH(1-34)NH ) in rat osteoblasts with fectively as hPTH(1-34)NH , but it is only half as effective
2
2
(2,7)
(38)
rat receptors,
CHO hamster cells expressing rat recep- a stimulator of osteoclast activity as hPTH(1-34)NH2.
(8)
tors, and dermal fibroblasts expressing normal levels of The results of experiments on cells of an established line of
endogenous human receptors, but only weakly in LLC-PK1 human fetal osteoblasts suggest that these cells respond to
porcine cells expressing large numbers of human receptors. C-terminally truncated PTHs like ROS osteoblasts rather
An example of the influence of the target cell and receptor than the primary human fibroblasts. In these human osteo-
origin is the fact that the C-terminally truncated hPTH(1- blasts, hPTH(1-31) can stimulate the cAMP/PKA-dependent
3
1)NH activates PKC as strongly in HKRK B7 cells and expression of transforming growth factor ␤2 (TGF-␤2) but
2
human dermal fibroblasts as hPTH(1-34)NH , but it fails to not the PKC-dependent expression of TGF-␤1 whereas
2
activate PKCs in ROS 17/2 rat osteoblastic cells, murine hPTH(1-34) can stimulate the expression of both TGF-
(
30)
proximal tubular cells, or probably human fetal osteo- ␤s.
blasts.
Clearly, the next step is to find out what signals are
In this respect, the hPTHR1s in the HKRK B7 in fact given to normal primary or very early passage human
cells and dermal fibroblasts behave like the unconventional osteoblasts by osteogenic fragments such as hPTH(1-
(3,9,30)
(5)
PTH receptors on rat spleen lymphocytes.
31)NH and how these signals are related to the processes of
2
These various observations suggest that there may be at bone nodule formation and osteoclast induction by osteo-
least two mechanisms by which an activated hPTHR1 can blastic cells.
stimulate PKC. One of these, like the stimulation of adeny-
lyl cyclase, requires an interaction of the receptor with a
(16)
PTH fragment’s unmodified N terminus
and may pro-
ACKNOWLEDGMENTS
ceed via activation of PLC when conditions favor this
response. It seems that cells of an established line of human
Funding for this work was provided by the Canadian
fetal osteoblasts lack this mechanism because they cannot Arthritis Network and the National Research Council of
stimulate PKC-dependent processes when exposed to Canada.
(30)
hPTH(1-31)NH2.
The other mechanism can be activated
by a portion of the PTH ligand that does not include the N
terminus and can proceed without measurable PLC activa-
tion. The PLC-independent mechanism might involve the
REFERENCES
stimulation of phospholipase A and/or phospholipase
1. Whitfield JF, Morley P, Willick GE 1999 The parathyroid
hormone, its fragments and analogs—potent bone builders for
treating osteoporosis. Exp Opin Invest Drugs 9:1293–1315.
. Jouishomme H, Whitfield JF, Chakravarthy BB, Durkin JP,
Gagnon L, Isaacs RJ, Maclean S, Neugebauer W, Willick G,
Rixon RH 1992 The protein kinase-C activation domain of the
parathyroid hormone. Endocrinology 130:53–60.
3. Jouishomme H, Whitfield JF, Gagnon L, Maclean S, Isaacs R,
Chakravarthy B, Durkin J, Neugebauer W, Willick G, Rixon
RH 1994 Further definition of the protein kinase C activation
domain of the parathyroid hormone. J Bone Miner Res 9:943–
2
(9,31–33)
(34)
D.
However, Bringhurst and Singh
have reported
that hPTH(3-34)NH does not stimulate PKC activity via
2
2
phospholipase (PLD) in LLC-PK1 cells and we have not
been able to show a stimulation of PLD in the human
fibroblasts (data not shown). LLC-PK1 cells are relatively
deficient in what human and rat PTH/PTHrP receptors need
to stimulate the second mechanism when they are activated
by N-terminally truncated PTH fragments whereas CHO
cells, human fibroblasts, opossum kidney tubule cells, and
rat osteoblasts can provide the necessary environment to
enable PTHR1s to activate fully PKC on interaction with
N-terminally truncated PTHs.
9
49.
4
. Whitfield JF, Morley P, Willick GE, Langille R, Ross V,
Maclean S, Barbier J-R 1997 Cyclizzaqtion by a specific
lactam linkage increases the ability of human parathyroid
An important practical question arising from the present
experiments is what signals would be given by hPTH(1-
hormone (hPTH)-(1-31)NH to stimulate bone growth in
2
ovariectomized rats. J Bone Miner Res 12:1246–1252.
. Whitfield JF, Isaacs RJ, Maclean S, Morley P, Barbier J-R,
Willick G 1999 Stimulation of membrane-associated protein
5
3
1)NH to induce osteoblasts to start making bone in an
2
osteoporotic patient? (From the clinical standpoint whether
or not N-terminally truncated PTH fragments can stimulate
PKCs is unimportant—they cannot stimulate adenylyl cy-
clase, which [at least in rats] is a sine qua non for inducing
kinase-C activity in spleen lymphocytes by hPTH(1-31)NH ,
2
27
22
26
its lactam derivative, [Leu ]-cyclo(Glu -Lys )-hPTH-(1-
31)NH , and hPTH-(1-30)NH . Cell Signal 11:159–164.
2
2
6. Donahue HJ, Fryer MJ, Eriksen EF, Heath H 1988 Differential
effects of parathyroid hormone and its analogues on cytosolic
calcium ions and cAMP levels in cultured rat osteoblast-like
cells. J Biol Chem 263:13522–13527.
. Fujimori A, Cheng SL, Avioli LV, Civitelli R 1992 Stucture-
function relationship of parathyroid hormone: Activation of
phospholipase-C, protein kinase-A and -C in osteosarcoma
cells. Endocrinology 130:29–36.
(
35,36)
new bone formation.
results and those of Takasu and Bringhurst
hPTH(1-31)NH and hPTH(1-34)NH in humans should be
) According to both the present
(14)
the effects of
2
2
7
(37)
the same, but they are not. According to Fraher et al.
hPTH(1-31)NH subcutaneously infused into human volun-
2
teers stimulates adenylyl cyclase as effectively as hPTH(1-

PTH RECEPTORS AND PKCs
447
8
. Azzaqrani A, Goltzman D, Orlowski J 1996 Structurally di-
verse N-terminal peptides of parathyroid hormone (PTH) and
PTH-related peptide (PTHrP) inhibit the Na /H exchanger 24. Chakravarthy BR, Bussey A, Whitfield JF, Sikorska M, Wil-
mone: Cyclic lactams of the receptor-binding region. J Med
Chem 40:1373–1380.
ϩ
ϩ
NH3 isoform by binding to the PTH/PTHrP receptor type I and
activating distinct signaling pathways. J Biol Chem 271:
liams RE, Durkin JP 1991 The direct measurement of protein
kinase C (PKC) activity in isolated membranes using a selec-
tive peptide substrate. Anal Biochem 196:144–150.
1
4931–14936.
9
. Friedman PA, Gesek FA, Morley P, Whitfield JF, Willick GE 25. Chakravarthy BR, Whitfield JF, Durkin JP 1994 Inactive
1
999 Cell-specific signaling and structure-activity relations of
membrane protein kinase Cs: A possible target for receptor
signalling. Biochem J 304:809–816.
parathyroid hormone analogs in mouse kidney cells. Endocri-
nology 140:301–309.
2
6. Whitfield JF, Chakarvarthy BR, Durkin JP, Isaacs RJ, Jou-
ishomme H, Sikorska M, Williams RE, Rixon RH 1992 Para-
thyroid hormone stimulates protein kinase C but not adenylate
cyclase in mouse epidermal keratinocytes. J Cell Physiol 150:
1
1
0. Rihani-Bisharat S, Maor G, Lewinson D 1998 In vivo anabolic
effects of parathyroid hormone (PTH) 28-48 and N-terminal
fragments of PTH and PTH-related protein on neonatal mouse
bones. Endocrinology 139:974–981.
2
99–303.
2
2
2
7. Usdin TB, Hoare SRJ, Wang T, Mezey ⌱, Kowalak JA 1999
TIP: A new neuropeptide and PTH2-receptor agonist from
hypothalamus. Nat Neurosci 2:941–943.
8. Cole JA, Bringhurst F 2000 PTH and P42/P44 activation in
OK cells and LLC-PK1 cells expressing differential densities
of human PTH1 receptor (PTHR1) J Bone Miner Res 15:S318.
9. Whitfield JF, Morley P, Willick GE, Isaacs RJ, MacLean S,
Ross V, Barbier J-R, Divieti P, Bringhurst FR 2000 Lactam
formation increases receptor binding, adenylyl cyclase stimu-
lation and bone growth stimulation by human parathyroid
1. Azzaqrani A, Goltzman D, Orlowski J 1995 Parathyroid hor-
mone and parathyroid hormone-related peptide inhibit the api-
ϩ
ϩ
cal Na /H exchanger NHE-3 isoform in renal cells (OK) via
a dual signaling cascade involving protein kinase A and C.
J Biol Chem 270:20004–20010.
1
2. Siegfried G, Vrtovsnik F, Prie D, Amiel C, Friedlander G 1995
Parathyroid hormone stimulates ecto-5 78-nucleotidase activ-
ity in renal epithelial cells: Role of protein kinase-C. Endocri-
nology 136:1267–1275.
1
3. LiChong KP, Li DY, Orlowski D, Tenenhouse HS, Goltzman
D 1998 Selective domains of PTH and PTHrP activate the
protein kinase C pathways and differentially influence the
hormone (hPTH)(1-28)NH . J Bone Miner Res 15:964–970.
2
3
3
3
3
0. Wu Y, Kumar R 2000 Parathyroid hormone regulates trans-
forming growth factor ␤1 and ␤2 synthesis in osteoblasts via
divergent signaling pathways. J Bone Miner Res 15:879–884.
1. Derrickson BH, Mandel LJ 1997 Parathyroid hormone inhibits
ϩ
ϩ
ϩ
Na /H exchanger (NHE3) and the type II Na /PI co-
transporter. Bone 23:S356.
1
4. Takasu H, Bringhurst FR 1998 Type-1 parathyroid hormone
ϩ
ϩ
Na -K -ATPase through G /G and the calcium-independent
q
11
(PTH)/PTH-related peptide (PTHrP) receptors activate phos-
phospholipase A . Am J Physiol 272:F781–F788.
2
pholipase C in response to carboxyl-truncated analogs of PTH.
Endocrinology 139:4293–4299.
2. Ribeiro CMP, Dubay GR, Falck JR, Mandel LJ 1994 Parathy-
ϩ
ϩ
roid hormone inhibits Na -K -ATPase through a cytochrome
P-450 pathway. Am J Physiol 266:F497–F505.
1
5. Takasu H, Guo J, Bringhurst FR 1999 Dual signaling and
ligand selectivity of the human PTH/PTHrP receptor. J Bone
Miner Res 14:11–20.
3. Singh AT, Kunnel JG, Strieleman PJ, Stern PH 1999 Parathyroid
hormone (PTH)-(1-34), [Nle(8,18), Tyr34]PTH-(3-34)amide, and
PTH-related peptide-(1-34) stimulate phosphatidylcholine hydro-
lysis in UMR-106 osteoblastic cells: Comparison with effects of
phorbol 12,13-dibutyrate. Endocrinology 140:131–137.
4. Bringhurst FR, Singh R 2000 Activation of phospholipase D
by transfected PTH/PTHrP receptors in LLCPK1 cells. J Bone
Miner Res 15:S319.
1
6. Takasu H, Gardella TJ, Luck MD, Potts JT Jr, Bringhurst FR
1
999 Amino-terminal modifications of human parathyroid hor-
mone (PTH) selectively alter phospholipase C signaling via
the type 1 receptor: Implications for design of signal-specific
PTH ligands. Biochemistry 38:13453–13460.
3
3
1
1
7. el Hessni A, Authier F, Silve C 1993 Parathyroid hormone and
glucagons compete for binding to low affinity sites on human
skin fibroblasts. Mol Cell Endocrinol 92:183–188.
5. Whitfield JF, Morley P, Willick GE 1999 The Parathyroid
Hormone: An Unexpected Bone Builder for Treating Osteo-
porosis. Landes Bioscience, Austin, TX.
8. Hanafin NM, Chen TC, Heinrich G, Segre GV 1995 Cultured
human fibroblasts and not cultured human keratinocytes ex- 36. Rixon RH, Whitfield JF, Gagnon L, Isaacs RJ, Maclean S,
press a PTH/PTHrP receptor mRNA. J Invest Dermatol 105:
33–137.
9. Pun KK, Ho PW 1989 Identification and characterization of
parathyroid hormone receptors on dog kidney, human kidney,
Chakravarthy B, Durkin JP, Neugebauer W, Ross V, Sung W,
Willick GE 1994 Parathyroid hormone fragments may stimu-
late bone growth in ovariectomized rats by activating adenylyl
cyclase. J Bone Miner Res 9:1179–1189.
1
1
chick bone and human dermal fibroblast. A comparative study 37. Fraher LJ, Avram R, Watson PH, Hendy GN, Henderson JE,
of functional and structural properties. Biochem J 259:785–
89.
Chong KL, Goltzman D, Morley P, Willick GE, Whitfield JF,
Hodsman AB 1999 Comparison of the biochemical responses
7
2
0. Pun KK, Tam SM 1995 Parathyroid hormone-activated phos-
phoinositide degradation and calcium channels in human der-
mal fibroblasts. Biol Signals 4:19–23.
1. Silve C, Santora A, Spiegel A 1985 A factor produced by
cultured rat leydig tumor (Rice-500) cells associated with
humoral hypercalcemia stimulates adenosine 3Ј,5Ј-
monophosphate production via the parathyroid hormone re-
ceptor in human skin fibroblasts. J Clin Endocrinol Metab
to human parathyroid hormone-(1-31)NH and hPTH-(1-34)
2
in healthy humans. J Clin Endocrinol Metab 84:2739–2743.
38. Mohan S, Kutilek S, Zhang C, Shen HG, Kodama Y, Srivas-
tava AK, Baylink DJ 2000 Comparison of bone formation
responses to parathyroid hormone (1-34), (1-31) and (2-34) in
mice. Bone 27:471–478.
2
Address reprint requests to:
Dr. James F. Whitfield
Institute for Biological Sciences
National Research Council of Canada
Ottawa, Ontario K1A 0R6, Canada
6
0:1144–1147.
2. Wu TL, Insogna KL, Hough LM, Milstone L, Stewart AF
987 Skin-derived fibroblasts respond to human parathyroid
2
1
hormone-like adenylate cyclase-stimulating proteins. J Clin
Endocrinol Metab 65:105–109.
2
3. Barbier J-R, Neugebauer W, Morley P, Ross V, Soska M,
Whitfield JF, Willick G 1997 Bioactivities and secondary Received in original form April, 5, 2000; in revised form Septem-
structures of constrained analogues of human parathyroid hor- ber 6, 2000; accepted October 2, 2000.