HAL Id: hal-02197445
https://hal.archives-ouvertes.fr/hal-02197445
Submitted on 27 May 2021
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
Parathyroid hormone potentiates nucleotide-induced
[Ca2+](i) release in rat osteoblasts independently of
G(q) activation or cyclic monophosphate accumulation
-A mechanism for localizing systemic responses in bone
K. A. Buckley, S. C. Wagstaff, G. Mckay, A. Gaw, R. A. Hipskind, G. Bilbe,
J. A. Gallagher, W. B. Bowler
To cite this version:
Parathyroid Hormone Potentiates Nucleotide-induced [Ca
2ⴙ]
iRelease in Rat Osteoblasts Independently of G
qActivation or
Cyclic Monophosphate Accumulation
A MECHANISM FOR LOCALIZING SYSTEMIC RESPONSES IN BONE*
Received for publication, June 28, 2000, and in revised form, December 20, 2000 Published, JBC Papers in Press, December 21, 2000, DOI 10.1074/jbc.M005672200
Katherine A. Buckley‡§, Simon C. Wagstaff‡, Gwen McKay¶, Alasdair Gaw¶, Robert A. Hipskind储, Graeme Bilbe**, James A. Gallagher‡, and Wayne B. Bowler‡
From the ‡Human Bone Cell Research Group, Department of Human Anatomy & Cell Biology, University of Liverpool, L69 3GE, United Kingdom,¶AstraZeneca Charnwood Bioscience, Bakewell Road, Loughborough, LE115RH, United Kingdom,
the储Institut de Genetique Moleculaire de Montpellier, CNRS UMR5535, IFR 24, Montpellier, France, and **Novartis Pharma AG, CH-4002, Basel, Switzerland
The regulation of tissue turnover requires the coordi-nated activity of both local and systemic factors. Nucle-otides exist transiently in the extracellular environ-ment, where they serve as ligands to P2 receptors. Here we report that the localized release of these nucleotides can sensitize osteoblasts to the activity of systemic fac-tors. We have investigated the ability of parathyroid hormone (PTH), a principal regulator of bone resorption and formation, to potentiate signals arising from nucle-otide stimulation of UMR-106 clonal rat osteoblasts. PTH receptor activation alone did not lead to [Ca2ⴙ]
i elevation in these cells, indicating no Gqcoupling, how-ever, activation of Gq-coupled P2Y1receptors resulted in characteristic [Ca2ⴙ]
irelease. PTH potentiated this nu-cleotide-induced Ca2ⴙrelease, independently of Ca2ⴙ
in-flux. PTH-(1–31), which activates only Gs, mimicked the actions of PTH-(1–34), whereas PTH-(3–34), which only activates Gq, was unable to potentiate nucleotide-in-duced [Ca2ⴙ]
irelease. Despite this coupling of the PTHR to Gs, cAMP accumulation or protein kinase A activation did not contribute to the potentiation. 3-Isobutyl-1-methylxanthine, but not forskolin effectively potenti-ated nucleotide-induced [Ca2ⴙ]
i release, however, fur-ther experiments proved that cyclic monophosphates were not involved in the potentiation mechanism. Co-stimulation of UMR-106 cells with P2Y1 agonists and PTH led to increased levels of cAMP response element-binding protein phosphorylation and a synergistic ef-fect was observed on endogenous c-fos gene expression following costimulation. In fact the calcium responsive Ca/cAMP response element of the c-fos promoter alone was effective at driving this synergistic gene expression. These findings demonstrate that nucleotides can pro-vide a targeted response to systemic factors, such as PTH, and have important implications for PTH-induced signaling in bone.
Nucleotides exist in the extracellular microenvironment of bone due to release from damaged cells at sites of injury, or via more controlled nonlytic release (1), where they differentially activate cell surface P2 receptors. Activation of Gq-coupled P2Y receptors leads to [Ca2⫹]i1 mobilization which has ultimately been found to induce processes including proliferation (2) and differentiation (3) and to modulate osteoblast-induced bone formation (4). The bone matrix synthesizing osteoblasts are thought to express multiple P2Y receptor subtypes (5), with expression profile changing as osteoblasts differentiate (6). The mechanisms that determine focal responsiveness to systemic factors in bone to allow a local tissue turnover, characteristic of the remodeling process, remain unclear. However, our previous observations that coapplication of nucleotides and parathyroid hormone (PTH) can result in synergistic responses in osteo-blasts (7) led us to hypothesize that locally released extracel-lular nucleotides could sensitize surrounding cells to the action of circulating hormones.
Parathyroid hormone is one of the most important systemic regulators of bone and mineral homeostasis. The action of PTH on skeletal cells is complex and can result in the stimulation of both resorption and new bone formation (8, 9). This ability to stimulate coupled, but opposing processes, has been attributed to the nature of the receptor that transduces signals arising from PTH stimulation. The PTH1 receptor belongs to a sub-group of seven transmembrane receptors that include those responsive to calcitonin, secretin, and VIP (10). These receptors are distinct in their ability to couple to both Gsand Gq and hence activate dual signal transduction pathways leading to both cyclic AMP formation and release of Ca2⫹from intracel-lular stores ([Ca2⫹]i).
The coupling of the PTH receptor to Gsin osteoblasts is well characterized. Activation of the cAMP signaling pathway is responsible for a number of PTH-induced downstream re-sponses, including expression of the c-fos proto-oncogene (11). In addition, the use of truncated PTH fragments in vivo has confirmed that complete N-terminal sequence (vital for Gs
ac-* This work was supported by an AstraZeneca studentship (to K. A. B.), the North West Cancer Research Fund (to S. C. W.) the As-sociation pour la Recherche pour le Cancer (to R. A. H.), and the Ar-thritis Research Campaign (to W. B. B.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ To whom correspondence should be addressed: Tel.: 44-151-794-5505; Fax: 44-151-794-5478.
1The abbreviations used are: [Ca2⫹]
i, intracellular [Ca
2⫹]; PTH, par-athyroid hormone, where not specified PTH ⫽ PTH-(1–34); FLIPR, fluorescent imaging plate reader; BSS, basal salt solution; 2-MeSATP, 2-methylthioadenosine 5⬘-triphosphate; 2-MeSADP, 2-methylthioa-denosine 5⬘-diphosphate; IBMX, 3-isobutyl-1-methylxanthine; dibu-tyryl-cGMP, N2,2⬘-O-dibutyrylguanosine 3⬘,5⬘-cyclic monophosphate; CREB, cAMP response element-binding protein; Ca/CRE, calcium and cAMP response element; DMEM, Dulbecco’s modified Eagle’s medium.
© 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
This paper is available on line at http://www.jbc.org
9565
tivation) is essential to drive the anabolic skeletal effects of PTH (12) and cAMP accumulation is often used as a measure of PTH receptor activation (13). However, the nature of PTH receptor/Gq-coupling and subsequent downstream responses in osteoblasts following activation remains controversial. Both inositol 1,4,5-trisphosphate-dependent and independent mech-anisms for PTH-induced [Ca2⫹]
i release have been reported (14, 15). To further complicate the issue of PTH-induced [Ca2⫹]i release, recent studies in cell types other than osteoblasts have demonstrated that while PTH alone is unable to induce [Ca2⫹]i release, it can strongly potentiate the [Ca2⫹]
i elevations in-duced by agonists acting at other Gq-coupled G-protein-coupled receptor. The mechanism behind this potentiation remains un-clear but has been attributed to G protein subunit interaction or shuttling of calcium between intracellular stores (16).
In a study by Kaplan et al. (17) using rat UMR-106 cells, PTH alone was shown to produce small [Ca2⫹]ielevations and also potentiated nucleotide-induced [Ca2⫹]i elevations. How-ever, in contrast to these findings, we previously demonstrated that PTH alone did not elevate [Ca2⫹]iand was ineffective at potentiating nucleotide-induced responses in the human osteo-sarcoma cell line SaOS-2 (7). In the same study, however, PTH and nucleotide costimulation of SaOS-2 cells did result in syn-ergistic induction of gene expression.
Considering the observations described above, the current study has addressed: (i) the controversy concerning the ability of PTH to effect [Ca2⫹]
irelease in the rat osteosarcoma cell line UMR-106, (ii) the mechanisms by which PTH can potentiate [Ca2⫹]irelease induced by locally released nucleotides, and (iii) whether this potentiated [Ca2⫹]iresponse can ultimately drive downstream cellular responses in osteoblasts.
EXPERIMENTAL PROCEDURES
Reagents—Dulbecco’s modified Eagle’s medium (DMEM),␣-modified Eagle’s medium (␣-MEM), Ham’s F-12, and fetal calf/bovine serum were purchased from Life Technologies (United Kingdom or France). Human parathyroid hormones (PTH) 1–34 and 1–31, and bovine PTH-(3–34) were purchased from Peninsula Laboratories (UK). Nucleotides, bovine serum albumin, fluo-3 AM, potato apyrase, H-89, IBMX, dibutyryl-cGMP, SKF96365, nitrocellulose membranes, and peroxidase-coupled goat anti-rabbit antibodies were obtained from Sigma (UK). Phospho-CREB and Phospho-CREB specific antisera were obtained from New England Biolabs (UK). Nitrocellulose membranes and enhanced chemilumines-cence reagents were acquired from Amersham Pharmacia Biotech (UK) or PerkinElmer Life Sciences (Belgium). Zetabind hybridization mem-brane was purchased from Cuno (Meriden, CT). Luciferase lysis reagent and luciferase reagent were purchased from Promega (UK). UMR-106 cells were kindly provided by T. J. Martin, Melbourne, Australia.
Cell Culture—UMR-106 rat osteoblast-like cells were cultured in
␣-MEM supplemented with 10% fetal calf serum, 100 g/ml streptomy-cin, 100 units/ml penicillin, and 2 mM L-glutamine. The cells were maintained at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Confluent cells were either induced for subsequent preparation of whole cell extracts or RNA isolation, or harvested upon confluence for Ca2⫹ release measurements in the FLIPR system (18). The UMR-106 multi CRE c-fos reporter cell line was cultured in Ham’s F-12 and DMEM (1:1) under similar conditions.
[Ca2⫹]
i Measurements—Cells were grown in 225-cm
2 vented cap flasks, harvested upon confluence, and resuspended in basal salt solu-tion (BSS) composisolu-tion (mM): NaCl (125), KCl (5), MgCl2(1), CaCl2(1.5), HEPES (25), glucose (5), and 1 mg/ml bovine serum albumin, pH 7.3. Cells were loaded with 17Mfluo-3-AM for 1 h at 37 °C with agitation
and in the presence of apyrase (2 units/ml), after which cells were washed 3 times in BSS and aliquoted into 96-well black wall, clear base plates at a density of 2.5⫻ 105cells/well. These were then centrifuged to obtain a confluent layer of cells on the base of the plate and cells were subsequently washed 4 times in BSS. Ca2⫹flux was measured in all 96 wells simultaneously and in real time using a Fluorescent Imaging Plate Reader (FLIPR). When effects of PTH and nucleotides together were studied the cells were incubated for 60 s with PTH prior to exposure to nucleotide. The effects of IBMX and forskolin were assessed by their introduction to the cells 6 min prior to nucleotide addition, H-89 was introduced 10 min before the addition of nucleotides and
dibutyryl-cGMP 30 min before nucleotide addition. In all cases [Ca2⫹]irelease was
measured for 1 min after nucleotide addition, a measurement of fluo-rescence being taken every second.
Whole Cell Extract Preparation—Whole cell extracts were prepared
as previously described (19). At the appropriate time point, plates were placed on ice and the cell layer quickly washed twice with ice-cold phosphate-buffered saline (140 mMNaCl, 10 mMNaPO4, pH 7.3) con-taining 10 mMNaF and 100MNa3VO4. Cells were solubilized in 10 mM Tris-HCl, pH 7.05, 50 mMNaCl, 30 mMsodium pyrophosphate, 50 mM
NaF, 5 mMZnCl2, 1% (v/v) Triton X-100, 100 MNa3VO4, 20 mM -glycerophosphate, 10 mM4-nitrophenyl phosphate, 1 mM dithiothre-itol, 0.5 mMbenzamidine, 2.5g/ml aprotinin, leupeptin, pepstatin, 0.2
mMphenylmethylsulfonyl fluoride, and 200 nMokadaic acid. The cell layer was collected by scraping and lysis completed by vortexing for 60 s. The lysate was clarified by centrifugation at 14,000⫻ g at 4 °C for 15 min and the protein concentration determined with Bradford’s rea-gent using bovine serum albumin as standard. An aliquot for Western analysis was denatured immediately in 2% SDS, 2% glycerol, 50 mM
Tris-HCl, pH 6.8, 1% 2-mercaptoethanol and the remaining superna-tant stored at⫺70 °C.
Western Analysis—15g of whole cell extracts were separated on an
8.5% SDS-polyacrylamide electrophoresis minigel and transferred elec-trophoretically to a nitrocellulose membrane. The filter was blocked for 1 h at room temperature in 5% (w/v) low fat milk powder-TBST (10 mM
Tris-HCl, pH 7.5, 100 mMNaCl, 0.1% Tween 20), followed by incubation overnight at 4 °C with the specific antiserum indicated below in block-ing buffer. After washblock-ing in TBST, the blots were incubated for 1 h at room temperature with peroxidase-coupled goat anti-rabbit antibody, diluted 1/1500 in 5% (w/v) low fat milk powder-TBST. After washing, the immune complexes were visualized using enhanced chemilumines-cence. The antisera used were: anti-CREB and anti-phospho Ser133 CREB, diluted 1:1000.
RNA Isolation—Total RNA was extracted from control and
stimu-lated cells according to the protocol of Chomczynski and Sacchi (20). Briefly, cells were lysed in 4Mguanidine thiocyanate, 0.5% Sarkosyl, 0.1Mmercaptoethanol, 25 mMsodium citrate, pH 7.0, followed by acid
phenol/chloroform extraction. RNA was stored as a precipitate at ⫺70 °C.
Northern Analysis—10g of total RNA was denatured and
electro-phoresed through a 0.8% (w/v) agarose gel containing 3.7% formalde-hyde (v:v), followed by transfer to Zetabind hybridization membrane. Blots were prehybridized at 65 °C in 50% formamide, 5⫻ SSC, 5 ⫻ Denhardt’s reagent, 1% SDS, 50 mMsodium phosphate buffer, pH 6.8, 5 mg/ml total calf liver RNA, 200g of tRNA, and probed with a c-fos riboprobe spanning exons 3 and 4 of the human c-fos gene mixed with a riboprobe derived from the rat gapdh cDNA as previously described (19). Membranes were washed for 30 min at 65 °C with 0.2⫻ SSC, 1% SDS solution and mRNAs visualized using phosphorstorage technology and autoradiography with Kodak X-AR film and intensifying screens at ⫺70 °C. RNA loading and integrity were measured by UV shadowing of the filter prior to hybridization.
FIG. 1. UMR-106 cells predominantly express P2Y1receptors.
Fluo-3-loaded UMR-106 cells were treated with various nucleotide ago-nists over a concentration range of 0.1–100M(as indicated on the x
axis scale). Ca2⫹release was measured according to fluorescence every second for 1 min after nucleotide addition by a FLIPR. Fluorescence is expressed as a percentage of the response to 100M2-MeSADP (n⫽ 4). Errors represent S.D. from the mean.
Effects of PTH on Nucleotide-induced [Ca
2⫹]
i
Release in Osteoblasts
Luciferase Reporter Gene Assays—Reporter cells were seeded into
white 96-well plates (Dynex) at a density of 96,000 cells/well in 50% DMEM, 50% Ham’s F-12 containing 10% fetal calf serum. Cells were incubated for 24 h, after which time medium was replaced by 50% DMEM, 50% Ham’s F-12 medium containing 0.1% bovine serum albu-min and incubated for a further 24 h. Cells were induced for 4 h with ⫻10 concentration of agonist stock solution prepared in serum-free medium to the final concentrations indicated in Fig. 9. After induction, cells were washed twice in cold phosphate-buffered saline and lysed in luciferase cell culture lysis reagent for 15 min at room temperature (10 l/well). A microtiter plate luminometer (ML 3000, Dynex Technolo-gies) was set up in the enhanced flash mode (maximum sensitivity, delay time⫽ 2 s, integration time ⫽ 10 s). Luciferase reagent (100l) was automatically added and light emission was measured every 10 ms during a period of 10 s. Data was recorded as the peak value of relative light units. Treatments were performed in triplicate and the mean fold increase in luciferase activity was calculated relative to mock-induced cells, which was taken as 1-fold.
RESULTS
UMR-106 Cells Predominantly Express P2Y1 Receptor—
Since osteoblasts have been described to express numerous subtypes of the P2Y family, and previous studies to receptor profile UMR-106 cells have been incomplete, we used the FLIPR system to fully investigate P2Y receptor expression by these cells. The effects of a series of known P2 agonists on [Ca2⫹]i elevation were measured in fluo-3-loaded cells. The agonist potency order (p[A]50) was as follows: 2-MeSADP (5.27⫾ 0.09) ⬎ 2-MeSATP (4.89 ⫾ 0.15) ⬎ ADP (4.60 ⫾ 0.15) ⬎ ATP (4.57⫾ 0.11), suggesting that P2Y1is the predominant P2 receptor expressed by UMR-106 cells (Fig. 1). While AMP, ␣MeATP, and UDP were inactive at concentrations between 0.1 and 100M(data not shown), UTP evoked a small [Ca2⫹]i
elevation (Fig. 1). This profile differs from that suggested by Kaplan et al. (17) in an earlier study, where predominant functional effects were elicited by P2Y2/Y4 receptor agonists. We have previously reported heterogeneity of P2 receptor ex-pression in populations of primary and clonal osteoblasts (7), a process that appears to be differentiation-dependent (6). Di-verse experimental culture methods or passage number may therefore account for the apparent differences in receptor pro-file between cells in this study and those used in the study of Kaplan et al. (17).
PTH Potentiates P2Y1Agonist-induced [Ca2⫹]iRelease—In a previous study using UMR-106 cells PTH potentiated [Ca2⫹]
i elevations induced by certain nucleotide agonists (17). In one of our earlier studies, however, using the human osteosarcoma
cell line SaOS-2, we observed no potentiation of nucleotide-induced [Ca2⫹]
irelease following costimulation with PTH (7). To clarify these differences, the effects of human PTH-(1–34) (100 ng/ml), in combination with known nucleotide agonists, on [Ca2⫹]iwere investigated in UMR-106 cells using the FLIPR. In contrast to the earlier study of Kaplan et al. (17), we repeat-edly saw no elevation of [Ca2⫹]iin response to PTH (0.1–500 ng/ml, data not shown). We did, however, observe a striking potentiation to the [Ca2⫹]iresponse with all nucleotides tested (Fig. 2 shows just ADP for clarity). A potentiated calcium response was observed when cells were stimulated with UTP and PTH, therefore confirming that P2Y2 receptors are ex-pressed by these cells, albeit at low levels.
PTH-induced Potentiated Ca2⫹ Elevations Are Dependent
upon Intracellular Store Release—To begin to investigate the
mechanism behind this observed potentiation we wondered whether calcium influx contributed to this response. Again using the FLIPR we found that in the absence of extracellular Ca2⫹, ADP and PTH costimulation induced a [Ca2⫹]iresponse approximately six times greater than that induced by ADP alone (Fig. 3A). However, this response was depressed in com-parison to that resulting from the same experiment performed in the presence of extracellular Ca2⫹(Fig. 3A). Interestingly, introduction of the receptor-activated calcium channel blocker
FIG. 2. Costimulation of UMR-106 cells with nucleotides and PTH results in potentiated Ca2ⴙelevation. PTH-(1–34) (100 ng/ml) was introduced to fluo-3-loaded UMR-106 cells 1 min before nucleotide addition (nucleotide concentration range, 0.3Mto 1 mM). Ca2⫹release was measured according to fluorescence every second for 1 min after PTH addition, then for a further 1 min after nucleotide addition. Flu-orescence is expressed as a percentage of the response to 1 mMADP⫹
PTH 100 ng/ml (n⫽ 3). Errors represent S.D. from the mean.
FIG. 3. Potentiated Ca2ⴙelevations are dependent upon intra-cellular store release. A, Fluo-3 loaded UMR-106 cells were divided into 2 groups, half were washed and resuspended in Ca2⫹-free BSS, the remaining cells were resuspended in normal BSS. PTH-(1–34) (100 ng/ml) was introduced 1 min before ADP addition (0.3Mto 1 mM, as shown on the x axis scale) to all cells. [Ca2⫹]
irelease was recorded as a
fluorescence reading by the FLIPR every second for 1 min after PTH addition, and every second for 1 min after ADP addition (n⫽ 3). B, again fluo-3-loaded UMR-106 cells were divided, half were resuspended in BSS containing SFK96365 (30 M), remaining cells were
resus-pended in normal BSS. Measurements were taken as for A (n⫽ 3).
SKF96365 resulted in an increase in the observed potentiation between ADP and PTH (Fig. 3B). Thus calcium influx is not the major component in the potentiation we observe.
Truncated PTH Peptides Act Differentially to Potentiate ADP-induced [Ca2⫹]
iRelease—To further elucidate the mech-anisms behind the observed potentiated [Ca2⫹]i release we used truncated forms of the PTH peptide, which have previ-ously been reported to have selective stimulatory effects upon the G-proteins that couple second messenger signaling path-ways to the PTH receptor. It has been reported that stimula-tion of the PTH receptor by PTH-(1–34) results in activastimula-tion of both Gqand Gsand therefore utilizes both PKC and adenylyl cyclase to initiate signaling cascades (10). We report here that PTH-(1–31) (100 ng/ml), which stimulates adenylyl cyclase but not PKC (21), is able to potentiate ADP-induced Ca2⫹ re-sponses as effectively as the full-length peptide (PTH-(1–34)) (Fig. 4A). In contrast, PTH-(3–34) (100 ng/ml) which does not activate adenylyl cyclase (22) was unable to potentiate the [Ca2⫹]irelease induced by ADP (Fig. 4A). To confirm that this peptide was functionally interacting with the PTH receptor it was introduced in combination with ADP and the active pep-tides, PTH-(1–34) or -(1–31). PTH-(3–34) was shown to act as a functional antagonist at the PTH receptor as it depressed the
[Ca2⫹]i release induced by ADP in combination with either PTH-(1–34) or PTH-(1–31) (Fig. 4B).
H-89 Does Not Inhibit PTH/Nucleotide-potentiated [Ca2⫹]
i
Release—As a main downstream target for activated adenylyl
cyclase pathways, we next investigated the potential role of PKA in driving potentiated [Ca2⫹]irelease in UMR-106 cells. Preincubation of UMR-106 cells with the PKA inhibitor H-89 (1–100M) did not affect potentiated [Ca2⫹]irelease induced by
ADP and PTH, indicating that PKA activity is not necessary to initiate the potentiation (Table I). H-89 was inhibitory in these cells, since it blocked PTH-induced phosphorylation of CREB on Ser133, the residue phosphorylated by active PKA (Fig. 5).
Forskolin and IBMX Have Differential Effects on ADP-in-duced [Ca2⫹]
iRelease—The results above showed that poten-tiation occurs with a PTH peptide that activates a signaling pathway involving adenylyl cyclase but not PKA. To assess whether elevated cAMP could drive potentiated [Ca2⫹]irelease, cells were costimulated with ADP and the cell permeable ad-enylyl cyclase activator, forskolin (50 M), in the absence of
extracellular calcium. Forskolin was unable to potentiate ADP-induced calcium responses (Table II). Forskolin alone was shown to elevate intracellular cAMP levels in these cells and to induce CREB phoshorylation via PKA phosphorylation (data not shown). We also tested the ability of the nonspecific phos-phodiesterase inhibitor, IBMX (1 mM) to potentiate
ADP-in-duced [Ca2⫹]irelease. In contrast to forskolin, IBMX effectively potentiated ADP-induced [Ca2⫹]
irelease (Table II).
cGMP Does Not Potentiate ADP-induced [Ca2⫹]
i Release— Since the results obtained with forskolin and IBMX suggested that accumulation of cGMP, downstream of the activated PTH receptor, may be causing the potentiation of ADP-induced [Ca2⫹]i levels, we costimulated cells with ADP and the cell permeable cGMP analogue, dibutyryl-cGMP (300 M), in the
absence of extracellular calcium. Dibutyryl-cGMP was unable to potentiate ADP-induced [Ca2⫹]
ielevations (Fig. 6), eliminat-ing accumulation of this cyclic monophosphate as a possible mechanism of potentiation. Dibutyryl-cGMP alone produced no [Ca2⫹]ielevation (data not shown).
PTH and P2Y1Receptor Agonists Induce Phosphorylation of the Transcription Factor CREB—Phosphorylation of CREB on
Ser133 is strongly linked to signaling-induced transcriptional activation. Ser133is a substrate for PKA but also for kinases activated by elevated intracellular calcium. We investigated whether ADP and PTH alone or in combination induced CREB phosphorylation in UMR-106 cells. We performed Western analysis of whole cell extracts using antisera directed against CREB phosphorylated on Ser133. The P2Y
1 agonist ADP (100 M) induced low CREB activation following 15 min stimulation,
while PTH-(1–34) (100 ng/ml) gave rise to a robust transcrip-tion factor phosphorylatranscrip-tion. In combinatranscrip-tion, however, agonists induced levels of CREB activation greater than those seen with either agonist alone. Quantitative values for CREB and phos-pho-CREB were obtained by densitometric analysis (Fig. 7).
FIG. 4. Truncated PTH peptides act differentially to potentiate ADP-induced [Ca2ⴙ]
irelease. A, PTH-(1–31) (100 ng/ml), PTH-(3– 34) (100 ng/ml), and PTH-(1–34) (100 ng/ml) were assessed for their ability to potentiate ADP-induced (0.3M TO1 mM) [Ca2⫹]irelease in
UMR-106 cells. All PTH peptides were introduced to fluo-3-loaded cells in the FLIPR 1 min before nucleotide addition. Ca2⫹release was meas-ured according to fluorescence every second for 1 min after PTH addi-tion, then for a further 1 min after nucleotide addition. Fluorescence is expressed as a percentage of the response to 1 mMADP⫹ PTH-(1–34) 100 ng/ml (n⫽ 4). B, PTH-(3–34) (100 ng/ml) was introduced in com-bination with both PTH-(1–31) (100 ng/ml) and PTH-(1–34) (100 ng/ml) to UMR-106 cells. ADP was added 1 min after PTH addition and fluorescence was measured as for A (n⫽ 3).
TABLE I
H-89 does not depress ADP/PTH-induced potentiated Ca2⫹elevations H-89 (1–100M) was introduced to fluo-3-loaded UMR-106 cells 10 min before ADP (100M) and PTH (100 ng/ml) addition. Ca2⫹release was measured according to fluorescence every 30 s for 10 min after H-89 addition, then every second for a further 1 min after ADP/PTH addition. Fluorescence is expressed as a percentage of the response to 100M
ADP⫹ 100 ng/ml PTH. H-89 had no effect on the potentiation at any concentration tested, figures show effects of 10MH-89 (n⫽ 3).
Agonist(s) Response of ADP (10PTH (100 ng/ml)⫺4M)⫹
ADP 22%
ADP⫹ PTH 100%
ADP⫹ PTH ⫹ H89 104%
Effects of PTH on Nucleotide-induced [Ca
2⫹]
i
Release in Osteoblasts
PTH and P2Y1Receptor Agonists Synergistically Induce the Endogenous c-fos Gene—To determine the impact of the
ob-served potentiated calcium response on gene expression we measured the effects of P2Y1agonists and PTH on the levels of endogenous c-fos transcription in UMR-106 cells. This proto-oncogene is of particular relevance when studying osteoblasts as it has been implicated in many of the processes that govern skeletal tissue remodeling (23). Stimulation of quiescent UMR-106 cells with the P2Y1agonists ADP and 2-MeSATP (10M)
for 45 min had a very weak effect on c-fos mRNA expression, as assessed by Northern analysis (Fig. 8). PTH alone led to a more robust c-fos induction that was significantly enhanced by co-stimulation with PTH and P2Y1agonists (Fig. 8).
The Ca/CRE Located at Position⫺60 in the c-fos Promoter Is Sufficient to Drive Synergistic Gene Expression—The Ca/CRE
element in the c-fos promoter responds to both cAMP and certain types of calcium signals. Following the observation that nucleotides and PTH could potentiate [Ca2⫹]irelease in UMR-106 cells, we wondered whether activation of the Ca/CRE alone may be sufficient to generate the observed synergy on c-fos gene expression. To assess this we utilized UMR-106 cells stably transfected with luciferase reporters driven by a trun-cated Ca/CRE promoter element. Costimulation of stable pools
of transfected UMR-106 cells with PTH and P2Y1 nucleotide agonists resulted in synergistic luciferase activity (Fig. 9). These data show that this calcium responsive region of the fos promoter is sufficient to drive synergistic gene expression, and that the strong calcium signal induced by ADP and PTH co-stimulation converges on this promoter element.
DISCUSSION
The main findings of this study are: 1) the rat osteosarcoma cell line UMR-106 primarily expresses P2Y1receptors coupled to [Ca2⫹]imobilization, 2) PTH does not couple to Gqin UMR-106 cells, but strongly potentiates P2Y1 agonist-induced [Ca2⫹]
irelease in these cells, 3) this potentiation is primarily dependent upon calcium release from intracellular stores and relies upon the activated PTH receptor coupling to Gs, 4) po-tentiated [Ca2⫹]i release occurs independently of cAMP/PKA pathways or cGMP accumulation, 5) PTH and nucleotide co-stimulation potentiates phosphorylation of the transcription factor CREB and leads to a synergistic induction of the
endog-FIG. 5. H-89 inhibits PTH-induced CREB phosphorylation in
UMR-106 cells. To confirm the bioactivity of H-89 quiescent cells were treated with PTH (100 ng/ml) alone or in combination with this com-pound (10M). H-89 was introduced either 15 or 30 min before lysing cells, as indicated above each lane, PTH was introduced 10 min before lysing in all lanes. Whole cell extracts were prepared, and 15g of protein per sample ran on an 8.5% SDS-polyacrylamide electrophoresis minigel. After transfer to a nitrocellulose membrane the blot was im-munodetected with anti-phospho-Ser133CREB antiserum, followed by peroxidase-coupled anti-rabbit antiserum and detection using en-hanced chemiluminescence. The blot was stripped and re-probed with anti-CREB antiserum to confirm the presence of similar levels of pro-tein in each lane. The absence of a signal for phospho-CREB in lanes treated with H-89 confirms that this compound does inhibit PKA in this cell type.
TABLE II
IBMX, but not forskolin can potentiate nucleotide-induced [Ca2⫹]
i release
Forskolin (50M), IBMX (1 mM), and PTH (100 ng/ml) were intro-duced to fluo-3-loaded UMR-106 cells 6 min before ADP addition (ADP concentration range, 0.3 M to 1 mM). Ca2⫹release was measured according to fluorescence every second for 1 min after ADP addition. Fluorescence is expressed as a percentage of the response to 100M
ADP⫹ 100 ng/ml PTH (n ⫽ 3).
Agonist(s) Response of ADP (10⫺4M)
⫹ PTH (100 ng/ml)
ADP 65%
ADP⫹ PTH 100%
ADP⫹ forskolin 64%
ADP⫹ IBMX 96%
FIG. 6. cGMP accumulation is not involved in the mechanism
resulting in potentiated [Ca2ⴙ]
i elevation. The cGMP analogue dibutyryl-cGMP (300M) was introduced to fluo-3-loaded cells 30 min before nucleotide addition (nucleotide concentration range, 0.3Mto 1 mM). Ca2⫹release was measured according to fluorescence every second
for 1 min after nucleotide addition. Fluorescence is expressed as a percentage of the response to 1 mMADP ⫹ PTH 100 ng/ml (n ⫽ 3). Errors represent S.D. from the mean. FCS, fetal calf serum.
FIG. 7. ADP and PTH stimulate CREB phosphorylation in UMR-106 cells. Quiescent UMR-106 cells were stimulated with 100M
ADP and 100 ng/ml PTH alone or in combination as indicated above each lane. After 15 min cells were lysed and Western analysis for CREB phosphorylation was performed as described in the legend to Fig. 5. Following densitometry, phospho-CREB levels were expressed as a ratio of CREB and then expressed as a percentage increase over vehicle as indicated below each lane. FCS, fetal calf serum.
enous c-fos gene, and 6) the fos Ca/CRE alone is sufficient to obtain a similar synergistic induction in the context of reporter genes stably transfected into UMR-106 cells.
An important observation from these studies was that PTH-(1–34), over a wide dose range, was unable to elevate [Ca2⫹]
iin UMR-106 cells, demonstrating that Gqis not activated in re-sponse to PTH in this cell type (data not shown). This is in contradiction to Kaplan et al. (17) who reported that PTH did induce a small [Ca2⫹]
ielevation. Earlier studies also confirm that the effect of PTH on [Ca2⫹]imobilization in osteoblasts is controversial, one group finding no effect (24), but others re-porting various increases (25, 26). A possible explanation for these very different reports may be the presence of ATP in the culture medium due to cell lysis. We have also demonstrated that nucleotides can be constitutively released nonlytically
from osteoblasts, and that this may be positively regulated by fluid flow (1). Therefore one consequence of agitating cell sus-pensions will be to increase release of ATP, with the ability to modulate PTH-induced responses. Systems fitting this criteria have been used in studies that report Gqcoupling in UMR-106 cells in response to PTH. To avoid this problem, we routinely treated the cells with apyrase to eliminate extracellular nucle-otides released in this manner. It is also worth noting that other studies have observed a Ca2⫹response to PTH in a subset of cells within a population, suggesting that this response in osteoblasts is limited by the state of differentiation of the cells (27, 28).
Our studies have shown huge potentiations of [Ca2⫹]i in response to nucleotide and PTH receptor costimulation which is consistent with previous findings (17). This observed poten-tiation was primarily dependent on intracellular store release, although the potentiation was slightly depressed in the ab-sence of extracellular Ca2⫹ (Fig. 3A). Interestingly, an in-creased potentiation was observed in the presence of the recep-tor-gated Ca2⫹ channel blocker SFK96365, (Fig. 3B), again supporting the theory that intracellular Ca2⫹ store release, rather than Ca2⫹influx was primarily responsible for potenti-ation. These observations also discount the possibility that potentiated cytosolic calcium levels reflect the ability of PTH/ cAMP-mediated stimulation of membrane channels, since this process is dependent upon extracellular Ca2⫹influx (29).
Although we found no evidence for Gqcoupling to the PTH receptor in UMR-106 cells, we utilized truncated PTH peptides to confirm that Gsactivation was vital in driving the potenti-ation mechanism. As expected, PTH-(1–31), which activates G␣sindependently of Gq, was able to potentiate ADP-induced [Ca2⫹]irelease (Fig. 4A). Conversely PTH-(3–34), which report-edly activates Gq(22), was ineffective in this respect, despite its capacity to functionally interact with the PTH receptor, shown by its ability to act as an antagonist to both PTH-(1–31) and PTH-(1–34) (Fig. 4B). These data were suggestive of the in-volvement of cAMP/PKA driven pathways in potentiating [Ca2⫹]i release. Intracellular cross-talk may occur between cAMP and [Ca2⫹]i driven pathways, cAMP can elicit [Ca2⫹]i release via PKA phosphorylation of inositol 1,4,5-trisphosphate receptor on serines 1755 and 1589. Kinetically, it seemed un-likely that this mechanism drove PTH potentiation of ADP-induced [Ca2⫹]i release in our system, however, we investi-gated this possibility using the metabolic PKA inhibitor H-89. While H-89 effectively inhibited CREB phosphorylation follow-ing PTH stimulation (Fig. 5), similar concentrations were un-able to depress [Ca2⫹]i release in response to PTH and ADP (Table I).
Although activated PKA appeared not to be essential for a potentiated [Ca2⫹]irelease the possibility still remained that elevated cAMP, through a PKA-independent mechanism, could play a role. This was investigated by costimulating UMR-106 cells with ADP and the adenylyl cyclase activator forskolin. In the absence of extracellular calcium the inability of forskolin to mimic PTH-induced responses strongly suggested that poten-tiated [Ca2⫹]
irelease occurs independently of cAMP accumu-lation. Unexpectedly, the nonspecific phosphodiesterase inhib-itor IBMX potentiated ADP-induced [Ca2⫹]
i release in a manner similar to PTH (Table II), suggesting that other cyclic monophosphates may be involved in the potentiation mecha-nism in osteoblasts. In contrast, Short and Taylor (16), using PTH1 receptor-transfected human kidney cells, were unable to potentiate carbachol-induced calcium responses with IBMX, and in fact, inhibited PTH-potentiated carbachol responses with this phosphodiesterase inhibitor (16). To clarify this ap-parent contradiction, and to further elucidate the potentiation
FIG. 8. P2Y1 receptor agonists and PTH induce synergistic
endogenous c-fos gene expression in UMR-106 cells. UMR-106 cells were serum-starved and then stimulated for 45 min with 10M
ADP, 2-MeSATP, and/or PTH 100 ng/ml as indicated above each lane. 10g of total RNA for each sample was analyzed by Northern blotting and hybridization with c-fos, and glyceraldehyde-3-phosphate dehydro-genase (GAPDH) riboprobes mixed 15:1. The c-fos signal in each lane was standardized to the GAPDH internal control, and the level of induction relative to control expressed numerically below each lane.
FIG. 9. PTH and nucleotides induce synergistic c-fos gene
ex-pression via the Ca/CRE promoter alone. UMR-106 cells stably transfected with the fos Ca/CRE promoter-luciferase reporter were serum starved and then induced for 4 h with 100MADP and/or PTH 10⫺8M. Luciferase activity was measured by a microtiter plate lumi-nometer in the enhanced flash mode (n⫽ 3). Error bars indicate the S.D.
Effects of PTH on Nucleotide-induced [Ca
2⫹]
i
Release in Osteoblasts
mechanism, UMR-106 cells were costimulated with ADP and the cGMP analogue, dibutyryl-cGMP. cGMP was unable to potentiate ADP-induced Ca2⫹elevations in these cells (Fig. 6). The discrepancies between the results obtained by Short and Taylor (16) and our own, may perhaps be explained by the fact that in some cell types IBMX has been found to elevate intra-cellular calcium (30, 31), whereas in other cell types, cGMP can dampen inositol 1,4,5-trisphosphate receptor-driven responses (32).
Our findings showing that neither elevated cAMP nor cGMP are able to potentiate nucleotide-induced [Ca2⫹]
i elevations and the inability of IBMX to potentiate carbachol responses in kidney cells demonstrate that an as yet unidentified mecha-nism is responsible for [Ca2⫹]ipotentiation observed upon dual activation of two separate G-protein-coupled receptors. Other investigators have suggested that PTH may regulate Ca2⫹ mobilization by facilitating translocation of Ca2⫹between dis-crete intracellular stores, thereby regulating the Ca2⫹ pool available to receptors linked to inositol 1,4,5-trisphosphate for-mation (16). Alternatively, it has been suggested that a direct G-protein interaction may account for potentiated [Ca2⫹]i release (33).
The significance of this calcium potentiation can be seen in its effects on transcription factor activation and gene expres-sion. The transcription factor CREB binds the cyclic AMP re-sponse element (CRE) and initiates transcription in rere-sponse to a number of extracellular signals, including elevated [Ca2⫹]
i. Phospho-CREB plays a key role in signaling-driven activation of the proto-oncogene c-fos which in turn has been strongly implicated in driving many osteoblast functions, including pro-liferation and differentiation (34, 35). PTH and ADP costimu-lation of UMR-106 cells resulted in increased levels of phospho-CREB (Fig. 7), and potentiation of endogenous c-fos mRNA expression (Fig. 8). Furthermore, using UMR-106 reporter cells, stably transfected with a truncated c-fos promoter con-sisting only of the Ca/CRE, we have demonstrated that this promoter element alone is sufficient to drive synergistic c-fos expression (Fig. 9). In contrast, in a previous study using the human osteosarcoma cell line SaOS-2, we found that both the Ca/CRE and the serum response element of the c-fos promoter were required to drive the same synergy (7). This was consist-ent with an inability of PTH to potconsist-entiate nucleotide-induced [Ca2⫹]
irelease in SaOS-2 cells. It would therefore appear that multiple pathways, both calcium dependent and independent, exist to couple P2Y1and PTH receptor coactivation to syner-gistic gene expression in osteoblasts.
It has never been clear how systemic PTH can initiate re-modeling that is confined to a small area of bone. These studies suggest a mechanism whereby the localized release of nucleo-tides, via lytic and nonlytic mechanisms, can sensitize cells to surrounding systemic factors such as PTH. This costimulation appears to have profound effects on osteoblastic cells, initiating calcium-activated pathways that ultimately result in
transcrip-tion factor activatranscrip-tion and synergistic inductranscrip-tion of gene expression.
REFERENCES
1. Bowler, W. B., Tattersall, J. A., Hussein, R., Dixon, C. J., Cobbold, P. H., and Gallagher, J. A. (1998) Bone 22, 3S
2. Dixon, J. C., Bowler, W. B., Littlewood-Evans, A., Dillon, J. P., Bilbe, G., Sharpe, G. R., and Gallagher, J. A. (1999) Br. J. Pharmacol. 127, 1680 –1686
3. Koshiba, M., Apasov, S., Sverdlov, V., Chen, P., Erb, L., Turner, J. T., Weisman, G. A., and Sitkovsky, M. V. (1997) Proc. Natl. Acad. Sci. U. S. A.
94, 831– 836
4. Jones, S. J., Gray, C., Boyde, A., and Burnstock, G. (1997) Bone 21, 393–399 5. Reimer, W. J., and Dixon, S. J. (1992) Am. J. Physiol. 263, C1040 –C1048 6. Dixon, J. C., Bowler, W. B., Walsh, C. A., and Gallagher, J. A. (1997) Br. J.
Pharmacol 120, 777–780
7. Bowler, W. B., Dixon, C. J., Halleux, C., Maier, R., Bilbe, G., Fraser, W. D., Gallagher, J. A., and Hipskind, R. A. (1999) J. Biol. Chem. 274, 14315–14324
8. Raisz, L. G. (1965) Nature 197, 1015–1017
9. Dempster, D. W., Cosman, F., Parisien, M., and Shen, V. (1993) Endocrine Rev.
14, 690 –709
10. Abou-Samra, A. B., Juppner, H., Force, T., Freeman, M. W., Kong, X-F., Schipani, E., Urena, P., Richards, J., Bonventre, J. V., Potts, J. T., Jr., Kronenberg, H. M., and Segre, G. V. (1992) Proc. Natl. Acad. Sci. U. S. A.
89, 2732–2736
11. Tyson, D. R., Swarthout, J. T., and Partridge, N. C. (1999) Endocrinology 140, 1255–1261
12. ArmamentoVillareal, R., Ziambaras, K., AbbasiJarhomi, S. H., Dimarogonas, A., Halstead, L., Fausto, A., Avioli, L. V., and Civitelli, R. (1997) J. Bone Min. Res. 12, 384 –392
13. Du, P. F., Seitz, P. K., and Cooper, C. W. (2000) Endocrine 12, 25–33 14. Dunlay, R., and Hruska, K. (1990) Am. J. Physiol. 258, F223–F231 15. Tong, Y., Zull, J., and Yu, L. (1996) J. Biol. Chem. 271, 8183– 8191 16. Short, A. D., and Taylor, C. T. (2000) J. Biol. Chem. 275, 1807–1813 17. Kaplan, A. D., Reimer, W. J., Feldman, R. D., and Dixon, S. J. (1995)
Endo-crinology 136, 1674 –1684
18. Kuntzweiler, T. A., Arneric, S. P., and Donnelly-Roberts, D. L. (1998) Drug Dev. Res. 44, 14 –20
19. Hipskind, R. A., Baccarini, M., and Nordheim, A. (1994) Mol. Cell. Biol. 14, 6219 – 6231
20. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156 –161 21. Jouishomme, H., Whitfield, J. F., Gagnon, L., Maclean, S., Isaacs, R.,
Chakravarthy, B., Durkin, J., Neugebauer, W., Willick, G., and Rixon, R. H. (1994) J. Bone Min. Res. 9, 943–949
22. Goltzman, D., Peytremann, A., Callahan, E., Tregear, G. W., and Potts, J. T., Jr. (1975) J. Biol. Chem. 250, 3199 –3203
23. Grigoriadis, A. E., Wang, Z-Q., Cecchini, M. G., Hofstetter, W., Felix, R., Fleish, H. A., and Wagner, E. F. (1994) Science 266, 443– 448
24. Boland, C. J., Fried, R. M., and Tashjian, A. H., Jr. (1986) Endocrinology 118, 980 –989
25. Donahue, H. J., Fryer, M. J., Erikson, E. F., and Heath, H. (1988) J. Biol. Chem. 263, 13522–13527
26. Yamaguchi, D. T., Hahn, T. J., Iida-Klein, A., Kleeman, C. R., and Muallem, S. (1987) J. Biol. Chem. 262, 7711–7718
27. Scho˜fl, C., Cuthbertson, K. S. R., Gallagher, J. A., Pennington, S. R., Cobbold, P. H., Brabant, G., Hesch, R-D., and Von Zur Muhlen, A. (1991) Biochem. J.
274, 15–20
28. Civitelli, R., Fujimori, A., Bernier, S. M., Warlow, P. M., Goltzman, D., Hruska, K. A., and Avioli, L. V. (1992) Endocrinology 130, 2392–2400
29. Luini, A., Lewis, D., Guild, S., Corda, D., and Axelrod, J. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 8034 – 8038
30. Pannabecker, T., and Orchard, I. (1987) Brain Res. 423, 13–22 31. Dasarathy, Y., and Fanburg, B. L. (1988) J. Cell. Physiol. 137, 179 –184 32. Komalavilas, P., and Lincoln, T. M. (1994) J. Biol. Chem. 269, 8701– 8707 33. Jimenez, A. I., Castro, E., Mirabet, M., Franco, R., Delicado, E. G., and
Miras-Portugal, M. T. (1999) Glia 26, 119 –128
34. Lee, K., Deeds, J. D., Chiba, S., Un-No, M., Bond, A. T., and Segre, G. V. (1994) Endocrinology 134, 441– 450
35. Stein, G. S., Lian, J. B., Stein, J. L., Van, W.ijnen, A. J., and Montecino, M. (1996) Physiol. Rev. 76, 593– 629
