Platelet-Derived Growth Factor Stimulates Sodium-Dependent Pi Transport in Osteoblastic Cells via Phospholipase Cγ and Phosphatidylinositol 3′-Kinase

Authors

  • Xuechu Zhen,

    1. Division of Clinical Pathophysiology, Department of Medicine, University of Geneva, Geneva, Switzerland
    Search for more papers by this author
  • Jean-Philippe Bonjour,

    1. Division of Clinical Pathophysiology, Department of Medicine, University of Geneva, Geneva, Switzerland
    Search for more papers by this author
  • Joseph Caverzasio Ph.D.

    Corresponding author
    1. Division of Clinical Pathophysiology, Department of Medicine, University of Geneva, Geneva, Switzerland
    • Division of Clinical Pathophysiology, Department of Medicine, University of Geneva, CH-1211 Geneva 14, Switzerland
    Search for more papers by this author

Abstract

Inorganic phosphate (Pi) is a major regulator of cell metabolism. The Pi transport activity in the plasma membrane is a main determinant of the intracellular level of this ion. In bone-forming cells, Pi transport is important for the calcification of the bone matrix. In this study, the effect of platelet-derived growth factor (PDGF) on Pi transport activity and the signaling mechanism involved in this cellular response were analyzed. The results indicate that PDGF is a potent and selective stimulator of sodium-dependent Pi transport in the mouse calvaria-derived MC3T3-E1 osteoblast-like cells. The change in Pi transport induced by PDGF-BB was dependent on translational processes and affected the Vmax of the Pi transport system. These observations suggested that enhanced Pi transport activity in response to PDGF resulted from insertion of newly synthesized Pi transporters in the plasma membrane. The role of activation of mitogen activated protein (MAP) kinase, phospholipase C (PLC)γ or phosphatidylinositol 3-kinase (PI-3–kinase), in mediating this effect of PDGF, was investigated. A selective inhibitor of the PDGF receptor tyrosine kinase activity (CGP 53716) completely blocked PDGF-induced protein tyrosine phosphorylation of several proteins including the PDGF receptor, PLCγ, MAP kinase, and association of the p85 subunit of PI-3′-kinase. Associated with this effect, the increase in Pi transport induced by PDGF was completely blunted by 5 μM CGP 53716. Inhibition of MAP kinase activity by cAMP agonists did not influence Pi transport stimulation induced by PDGF. However, inhibitors of protein kinase C completely blocked this response. A selective inhibitor of PI-3-kinase, LY294002, also significantly reduced this effect of PDGF. In summary, these results indicate that PDGF is a potent and selective stimulator of Pi transport in osteoblastic cells. The mechanism responsible for this effect is not mediated by MAP kinase but involves tyrosine phosphorylation-dependent activation of PLCγ and PI-3-kinase.

INTRODUCTION

Platelet-derived growth factor (PDGF) is a potent mitogen and chemotactic factor for a variety of cells of mesenchymal origin such as fibroblasts, vascular smooth muscle cells, and skeletal cells.1 In addition to its potent mitogenic effect on mesenchymal cells, PDGF has been shown to have many other biological effects including activation of Na+/H+ exchange activity in epithelial cells2 and enhancement of inorganic phosphate (Pi) uptake in C3H fibroblasts.3 Pi is a major regulator of intracellular metabolism and plays a central role in substrate-level and oxydative phosphorylation. Although intracellular Pi concentration is known to vary as the result of the breakdown and synthesis of molecules containing high Pi-bond energy, intracellular Pi level also depends on the rates of uptake and loss of Pi across the cell membrane. In addition to its role in cell metabolism, Pi transport is also an important function of bone-forming cells for the calcification of the bone matrix. Matrix vesicles, which are structures derived from the plasma membrane of osteogenic cells, are endowed with a Pi transport system.4 The activity of this Pi transport system, which can be stimulated by several calciotropic factors,5 plays a critical role in initial events responsible for the accumulation of Ca and Pi and the calcification of these structures.5,6 The cellular mechanism by which growth factors, including PDGF, modulate Pi transport in differentiated cells remains largely unknown. Recent studies from this laboratory suggested that tyrosine phosphorylation represents an essential step in the regulation of Pi transport by IGF-1,7 but the molecular mechanism responsible for this response has not yet been elucidated. Three isoforms of PDGF have been described,8,9 which bind with different affinities to two related tyrosine kinase receptors.10,11 Binding of PDGF to its receptor leads to dimerization of receptor molecules and initiates autophosphorylation on multiple tyrosine residues,12 including tyrosines in the kinase insert region,13,14 and the C-terminal domain.15 After phosphorylation, the receptor physically associates with signaling molecules, including phospholipase C (PLC)γ-1,16–19 phosphatidylinositol 3-kinase (PI-3-kinase),20,21 Syp (SH-PTP2),22 and GTPase-activating protein (GAP).23–25 In this work, we present results indicating that PDGF is a potent and selective stimulator of sodium-coupled Pi transport activity in osteoblast-like cells. The mechanism responsible for this effect of PDGF on Pi transport involves tyrosine phosphorylation-dependent activation of PLCγ and PI-3-kinase.

MATERIALS AND METHODS

Materials

Cell culture reagents were purchased from Flow laboratories (ICN Biochemicals Inc, Costa Mesa, CA, U.S.A.) and fetal calf serum (FCS) was from Gibco (Life Technologies Ltd., Paisley, U.K.). Human PDGF B chain homodimer was obtained from A.F. Schützdeller (Tübingen, Germany) and was dissolved as a concentrated solution in 0.1 mM acetic acid and 1 mg/ml bovine serum albumin (BSA). 3-isobutylmethylxanthine (IBMX) and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma (St. Louis, MO, U.S.A.). The protein kinase C (PKC) inhibitors calphostin C and staurosporine and the cAMP agonist forskolin were obtained from Calbiochem-Novabiochem (La Jolla, CA, U.S.A.). The PDGF receptor inhibitor CGP53716 was generously supplied by Dr. N.B. Lydon (Ciba-Geigy, Basel, Switzerland). LY 294002 was obtained from Biomol Research Labs (Plymouth Meeting, PA, U.S.A.). All these agents were dissolved as concentrated solution in dimethylsulfoxide (DMSO). [γ-32P]ATP (3000 Ci/mmol) and [3H]thymidine were from Amersham International plc (Little Chalfont, U.K.). H3[32P]O4 and [3H]-alanine were from DuPont de Nemours (Brussel, Belgium).

Electrophoresis reagents were obtained from Bio-Rad (Richmond, CA, U.S.A.). Immobilon P PVDF-membranes and milliblot semidry graphite electroblotter were from Millipore Corp. (Bedford, MA, U.S.A.). Agarose conjugated antiphosphotyrosine Ab, clone 4G10, anti-PLCγ polyclonal Ab and the antirat mitogen activated protein (MAP) kinase R2 were obtained from Upstate Biotechnology (Lake Placid, NY, U.S.A.). Anti-Pan ERK mAb and anti-p85 polyclonal Ab were from Transduction Laboratories (Lexington, KY, U.S.A.). Horseradish peroxidase linked antimouse and antirabbit secondary antibodies as well as enhanced chemiluminescence (ECL) reagents and hyperfilm were from Amersham International plc.

Other chemicals were obtained from standard laboratory suppliers and were of the highest purity available.

Cell culture

The murine calvaria-derived MC3T3-E1 osteoblast-like cells were grown in alpha modified essential medium (α-MEM) containing 10% FCS, 1% (v/v) nonessential amino acids, 100 IU/ml penicillin, and 100 μg/ml streptomycin. All cultures were maintained at 37°C in a humidified atmosphere of 5% CO2−95% air. With these culture conditions, the cells reached confluency after 4 days. Subcultures were obtained once a week by removing the cells from the dish using 0.1% collagenase and 0.25% trypsin in a Ca- and Mg-free Earle's salt solution containing 2 mmol/l of EDTA. For the studies of Pi transport and DNA measurements, the cells were plated (20 × 103) in 16-mm 24-well tissue culture clusters whereas 25 cm2 cell culture flasks were used for immunoprecipitation experiments.

Influence of PDGF on Pi transport activity

The effect of PDGF on Pi transport was analyzed in confluent MC3T3-E1 cells cultured in α-MEM containing 1% FCS for 24 h. Following PDGF exposure for various incubation times, Pi transport activity was determined in Earle's buffered salt solution (EBSS) with 15 mM HEPES, pH 7.4, containing 0.05–1.5 mmol/l labeled H3[32P]O4 or 0.1 mM [3H] alanine as previously described.26 In one type of experiment, confluent MC3T3 cells were preincubated with 2 μM cycloheximide for 1 h and during PDGF exposure before the determination of Pi transport activity. Transport experiments were performed in EBSS containing no PDGF. Before the transport assay, the cell layer was rinsed three times with EBSS without radioactive or cold substrate. The transport measurement started after adding 0.3 ml of EBSS containing the labeled substrate (1 μCi/ml). After various incubation times, the uptake solution was aspirated, and the cell layer was rinsed three times with 0.3 ml of ice-cold (4°C) substrate free Earle's salt solution. The cells were then solubilized with 0.25 ml of 0.2 N sodium hydroxyde, and the radioactivity contained in a 200 μl aliquot was counted by a standard liquid scintillation technique. The Pi transport activity was normalized by DNA content. DNA was measured as described by Burton27 with calf thymus DNA as standard.

Immunoprecipitation of cellular proteins

As for Pi transport analysis, confluent MC3T3-E1 cells were cultured in α-MEM containing 1% FCS for 24 h before addition of the various agents. For experiments related to the effect of PDGF on MAP kinase, the cells were preincubated with either 0.5 mM IBMX and 10 μM forskolin or the vehicle (0.2% DMSO) for 20 minutes before and during PDGF (10 ng/ml) exposure for 15 minutes. The influence of the specific PDGF receptor tyrosine kinase inhibitor CGP 5371628 was analyzed in cells preincubated for 90 minutes with the inhibitor or its vehicle (0.2% DMSO) prior to and during PDGF exposure for 15 minutes. Immunoprecipitation of phosphotyrosine containing proteins was performed in cells exposed to PDGF (10 ng/ml) for 15 minutes. Then, the culture medium was aspirated and the cells were immediately frozen in liquid nitrogen. Cell lysates were prepared by incubating the cells at 4°C with gentle agitation for 30 minutes in 1.1 ml Tris buffer A (20 mM, pH 7.6) containing 40 mM β-glycerophosphate, 1.5 mM EGTA, 0.5 mM EDTA, 25 mM NaF, 1 mM Na-pyrophosphate, 0.5 mM Na3VO4, 1 mM dithiothreitol (DTT), 1 mM benzamidine, 2 μg/ml leupeptin, 2 μg/ml aprotinin, 0.1 mg/ml phenylmethylsulfonylfluoride (PMSF) and 0.5% Triton X-100. Extracts were cleared by centrifugation (15 minutes at 16,000g and 4°C), and the proteins contained in the supernatant were immunoprecipitated overnight at 4°C with either 2.5 μg of antirat MAP kinase R2 and 50 μl of protein A/G beads for 2 h (Oncogene Science, Inc., Lake Placid, NY, U.S.A.) or with agarose conjugated antiphosphotyrosine monoclonal antibodies, clone 4G10.

MAP kinase activity

MAP kinase immunoprecipitates were washed four times with Tris buffer A and once with kinase buffer containing 25 mM HEPES (pH 7.2), 10 mM MgCl2, 1 mM DTT, 0.2 mM Na3VO4, 2 μM PKA inhibitor peptide (PKI; Life Technologies Ltd.). MAP kinase activity was determined in 0.5 ml of kinase buffer with 50 μM [γ-32P]ATP (50 μCi/ml, 3000 Ci/mmol), and 0.3 mg/ml myelin basic protein (MBP) as previously described.29 Enzymatic reaction was performed during 20 minutes at 30°C. It was stopped by adding 100 μl of ice-cold 0.1 M EDTA followed by rapid centrifugation at 4800 g for 5 min. Four aliquots of the supernatants (quadruplicate determinations) were filtered using multiscreen GV 96-well plate fitted with a low binding 0.22 μM pore size membrane, type MAGVN2250 (Millipore Corporation). After three washes with 5% TCA, the filters were added to scintillation fluid for radioactivity counting.

Analysis of tyrosine phosphorylated proteins

Antiphosphotyrosine immunoprecipitates were washed with Tris buffer A and resuspended in 100 μl sodium dodecyl sulphate (SDS) sample buffer (10% glycerol, 2% SDS, Tris 0.0625 M, pH 6.8, 100 mM DTT, and 0.0125‰ bromophenol blue). After heating of the samples for 20 minutes at 60°C, the agarose beads were pelleted by centrifugation and the supernatant loaded on 6–10% SDS-PAGE gels, using Laemmli buffer system.30 The resolved proteins were transferred to immobilon-P PVDF membranes using the milliblot graphite electroblotter system following the manufacturer's instructions. Membranes were blocked for 4 h at room temperature in Tris-buffered saline, 0.1% Tween 20 (TBST) containing 1% BSA (w/v) and incubated overnight at 4°C in TBST, 1% BSA containing 0.2 μg/ml antiphosphotyrosine monoclonal antibody, clone 4G10. The membranes were then washed at room temperature with TBST. Immunoreactive bands and biotinylated MW standards were visualized by the ECL detection technique using a sheep antimouse secondary antibody conjuguated to horseradish peroxidase or streptavidin conjuguates.

Detection of specific proteins recovered in anti-phosphotyrosine immunoprecipitates

To detect the presence of PLCγ, MAPK and the p85 subunit of PI-3-kinase within the antiphosphotyrosine immunoprecipitates, the PVDF membranes used for the antiphosphotyrosine immunoblotting experiments described above were stripped for 30 minutes at 60°C in 62.5 mM Tris-HCl, pH 6.8, containing 2% SDS and 100 mM β-mercaptoethanol and washed at room temperature with TBST. The membranes were then blocked for 4 h at room temperature in TBST, 1% BSA and reprobed by overnight incubation at 4°C, in TBST, 1% BSA containing the specific antibody against the protein of interest. Immunoreactive bands and biotinylated MW standards were visualized by the ECL detection technique using the appropriate horseradish peroxidase secondary antibody or streptavidin conjuguates.

Statistical analysis

All results are expressed as mean ± SEM. A two-sided unpaired Student's t-test was used for statistical analysis. A difference between experimental groups was considered significant when the p value was <5.0%.

RESULTS

Characteristics of Pi transport stimulation induced by PDGF in MC3T3-E1 osteoblast-like cells

The initial rate of Pi transport and Pi accumulation in MC3T3-E1 cells, determined in presence of sodium in the uptake medium, were markedly enhanced by 30 ng/ml PDGF for 24 h (Fig. 1). The initial rate of Pi transport, assessed by the measurement of Pi uptake during the first 6 minutes of incubation was increased by 2.5- to 3.0-fold by PDGF. Pi accumulation, estimated by the amount of Pi taken up after 20 minutes of incubation, was already 5.5- to 6.0-fold higher in PDGF, compared with vehicle-treated cells, although a steady-state level of Pi was not yet reached in PDGF-treated cells. The sodium-independent component of Pi uptake, determined in the presence of choline chloride, was less than 10% of the total uptake measured at 6 minutes incubation time and was not influenced by PDGF (Fig. 1). Unless specified, this component was therefore neglected in further experiments.

Figure FIG. 1.

Effect of PDGF on Pi transport activity in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. Then, they were incubated for 24 h in the same culture medium with either 30 ng/ml PDGF (closed symbols) or its vehicle (open symbols). Pi uptake was determined in EBSS containing 0.1 mM32PO4 and either 143 mM NaCl (circles) or 143 mM choline chloride (squares). Each symbol represents mean ± SEM of five or six determinations from a representive study out of three experiments.

Pi transport activity was already significantly enhanced by PDGF after 30 minutes incubation and further increased during the next 6 h of incubation. A maximal stimulation of Pi transport activity was reached after 24 h of treatment (Fig. 2, left panel). In cells exposed to PDGF for 24 h, a significant increased Pi transport activity was detected with 2 ng/ml PDGF. A submaximal stimulation was obtained with 10 ng/ml and maximal effect observed with 50 ng/ml PDGF (Fig. 2, right panel).

Figure FIG. 2.

Time and dose-effect of PDGF on Pi transport in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. Then, they were exposed to either PDGF (•) or vehicle (○) for various incubation times with 10 ng/ml PDGF (left panel) or with increasing concentrations of PDGF for 24 h (right panel). Pi uptake was determined in EBSS medium containing 0.1 mM32PO4. Each point represents mean ± SEM of five or six determinations of a representative study out of five experiments. *p < 0.01, **p < 0.001 compared with the respective vehicle.

The change in Pi transport activity induced by PDGF was not associated with the alteration of sodium-dependent alanine transport (Table 1) suggesting that a selective mechanism mediates this cellular response. As previously shown for most agonists of Pi transport stimulation in various eucaryotic cells, the increased Pi transport induced by PDGF was due to alteration of the maximal velocity of the Pi transport system (Vmax) with no change of the affinity constant (Km) of the carrier for Pi (Table 2). Pi transport stimulation induced by 6 h of exposure of 10 ng/ml PDGF (11.1 ± 0.9 nmol/μg DNA) compared with vehicle (3.2 ± 0.2 nmol/μg DNA) was completely blunted by 2 μM cycloheximide (PDGF + cyclo: 3.5 ± 0.1; Veh + cyclo: 3.0 ± 0.1 nmol/μg DNA).

Table Table 1. Selectivity of Pi Transport Stimulation Induced by PDGF in MC3T3-E1 Osteoblast-like Cells
original image
Table Table 2. Kinetic Analysis of Pi Transport Stimulation Induced by PDGF in MC3T3-E1 Cells
original image

Signaling mechanisms involved in PDGF stimulation of Pi transport

Exposure of MC3T3-E1 cells to 10 ng/ml PDGF for 15 minutes markedly increased the tyrosine phosphorylation of several proteins with molecular weights of approximately 184, 154, 112, 84, 72, 57, 55, 45, 42, 39, and 38 kD (Fig. 3). The 184 kD protein likely represents the PDGF receptor. Reblotting analysis with specific antibodies indicated that p154 and p45 are PLCγ and p44 MAP kinase, respectively (Fig. 3). p72 has been identified as the SHPTP2/Syp tyrosine phosphatase (data not shown) and other phosphorylated proteins were not characterized. The amount of the p85 subunit of PI-3-kinase associated with tyrosine phosphorylated proteins was also markedly increased by PDGF (Fig. 3). Preincubation of MC3T3-E1 cells with 5 μM CGP 53716 for 90 minutes completely blunted the change in protein tyrosine phosphorylation and the binding of p85 to tyrosine phosphorylated proteins induced by PDGF (Fig. 3). Associated with these effects of CGP 53716 on activation of signaling molecules, Pi transport stimulation induced by PDGF was dose dependently and maximally inhibited by 5 μM of this agent (Fig. 4). These results strongly suggest that tyrosine phosphorylation induced by the PDGF receptor, including its autophosphorylation, is the initial event responsible for Pi transport stimulation induced by PDGF.

Figure FIG. 3.

Inhibition by the PDGF receptor tyrosine kinase inhibitor CGP53716 of enhanced protein tyrosine phosphorylation induced by PDGF in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. They were then preincubated with either CGP53716 (5 μM) or 0.2% DMSO in the same medium for 90 minutes. Then, the cells were exposed to either PDGF (10 ng/ml) or vehicle in presence or absence of CGP53716 for 15 minutes. Following treatment, the cells were rapidly frozen in liquid nitrogen and lysed at 4°C. Lysate proteins were used to immunoprecipitate phosphotyrosyl proteins with agarose-conjugated anti-TyrP monoclonal antibody 4G10. Proteins were separated by SDS-PAGE on a 6–10% acrylamide gradient gel, transferred on an immobilon membrane, immunoblotted with mAb 4G10, and detected by the enhance chemiluminescent (ECL) method using horse radish peroxidase-conjugated anti-mouse antibody: (1) Veh; (2) PDGF (10 ng/ml); (3) CGP53716 (5 μM); (4) CGP53716 + PDGF. The blot used for identification of phosphotyrosine-containing proteins was stripped and reprobed with either anti-PLCγ polyclonal Ab, anti-p85 polyconal Ab or anti-Pan ERK mAb using the ECL system.

Figure FIG. 4.

Effect of the PDGF receptor tyrosine kinase inhibitor CGP53716 on Pi transport stimulation induced by PDGF in osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. They were then preincubated with either CGP53716 or 0.2% DMSO for 90 minutes prior to or during incubation with either 10 ng/ml PDGF (inline image) or its vehicle (□) for 6 h. Pi transport was determined in EBSS medium containing 0.1 mM32PO4. Each column represents the mean ± SEM of five or six determinations of a representive study out of two experiments. **p < 0.001 compared with the vehicle.

MAP kinase is a key enzyme in controlling mitogenic and differentiation processes induced by growth factors.31 Its functional role in mediating the change in Pi transport induced by growth factors is unknown and was investigated for PDGF. Selective inhibition by PKA of the Ras-dependent regulatory pathway leading to activation of MAP kinase was recently reported29,32,33 and was used to assess whether MAP kinase may be involved in the change of Pi transport induced by PDGF. As shown in the left part of Fig. 5, preincubation of MC3T3-E1 cells with 0.5 mM IBMX and 10 μM forskolin for 20 minutes completely prevented MAP kinase activation induced by 10 ng/ml PDGF. In the same experimental conditions, the mitogenic expression of PDGF (3.5 ± 0.2-fold [treated/control], n = 4, p < 0.001) was completely blunted by stimulators of cAMP level (0.8 ± 0.1-fold [treated/control], n = 4). In contrast, the change in Pi transport activity induced by PDGF was not altered (Fig. 5, right panel). Also, and as previously described in UMR-106 cells,34 cAMP stimulating agents significantly enhanced Pi transport in untreated MC3T3-E1 cells (Fig. 5, right panel).

Figure FIG. 5.

Relationship between MAP kinase activity and Pi transport induced by PDGF in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. They were preincubated with either 0.5 mM IBMX plus 10 μM forskolin or vehicle (0.2% DMSO) for 20 minutes. PDGF (10 ng/ml) was then added for 15 minutes (MAP kinase analysis) or 6 h (Pi transport) with or without cAMP agonists. MAP kinase immunoprecipitation and assay of activity using MBP as substrate (left panel) were performed as described in Methods. Pi transport (right panel) was determined in EBSS medium containing 0.1 mM32PO4. The results of MAP kinase activity are mean ± SEM of triplicates of one determination, representative of two similar observations. For Pi transport, the results are expressed as the mean ± SEM of five or six determinations from a representative study out of five experiments. **p < 0.001 compared with the vehicle.++p < 0.001 compared with foskolin + IBMX.#p < 0.05 compared with the vehicle.

The role of the phosphoinositide-specific phospholipase C signaling pathway in mediating the stimulatory effect of PDGF on Pi transport was also investigated since previous observations suggested that PKC can influence Pi transport activity in osteoblast-like cells.35 This relationship was analyzed either by using PKC inhibitors or by testing the PDGF response on Pi transport in MC3T3-E1 cells down-regulated for PKC activity. As shown in the left part of Fig. 6, 1 μM calphostin C, a specific PKC inhibitor, did not affect the basal Pi transport activity but completely blunted the stimulation of Pi transport induced by PDGF. A mixed inhibitor of PKC and tyrosine kinase, staurosporine (0.5 μM), significantly reduced the basal Pi transport activity and also blocked this hormonal response (Fig. 6, right panel). The stimulatory effect of PDGF on Pi transport was also blunted in MC3T3-E cells down-regulated for PKC activity by pretreatment with 5 μM PMA for 24 h (Fig. 7, right panel). In contrast to the absence of Pi transport stimulation induced by either PMA or PDGF in these cells, the stimulatory effect of IGF-1 remained fully expressed (Fig. 7).

Figure FIG. 6.

Effect of PKC inhibitors on Pi transport stimulation induced by PDGF in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with αMEM containing 1% FCS for 24 h. Then, they were preincubated with 1 μM calphostin C (left panel) or 0.5 μM staurosporine (right panel) for 1 h and 20 minutes, respectively, before and during exposure to PDGF (10 ng/ml) or vehicle for 6 h. Pi transport was determined in EBSS medium containing 0.1 mM32PO4. Results are expressed as the mean ± SEM of five or six determinations from a representative study out of two experiments. **p < 0.001 compared with the vehicle.++p < 0.001 compared with the vehicle.

Figure FIG. 7.

Effect of PDGF on Pi transport stimulation in MC3T3-E1 cells down-regulated for PKC. MC3T3-E1 cells cultured for 4 days were changed with α-MEM containing 1% FCS for 24 h. Then, they were exposed to 5 μM PMA or vehicle (0.2% DMSO) for 24 h. After two washes with culture medium, the cells were exposed to either vehicle, PDGF (10 ng/ml), IGF-1 (10−7 M) or PMA (1 μM) for 6 h. Pi transport was determined in EBSS medium containing 143 mM of either NaCl or choline chloride for determination of the Na-dependent Pi transport and 0.1 mM32PO4. The results are expressed as mean ± SEM from a representive study out of two experiments. *p < 0.01, **p < 0.001 compared with the vehicle.

The PI-3-kinase pathway is another important cellular mechanism previously shown to influence the stimulation of solute transport activities in response to activated growth factor receptors.2,36 As shown in Fig. 8, LY294002, a specific inhibitor of PI-3-kinase,37 significantly and dose-dependently reduced the change in Pi transport activity induced by PDGF.

Figure FIG. 8.

Effect of the PI-3′-kinase inhibitor LY294002 on Pi transport stimulation induced by PDGF in MC3T3-E1 osteoblast-like cells. MC3T3-E1 cells cultured for 4 days were changed with α-MEM medium containing 1% FCS for 24 h. They were then preincubated with various concentrations of LY 294002 for 10 minutes before and during exposure to either vehicle (□) or 10 ng/ml PDGF (inline image) for 6 h. Pi uptake was determined in EBSS medium containing 0.1 mM32PO4. Data are expressed as mean ± SEM of five or six determinations from a representative study out of two experiments. **p < 0.001 compared with the vehicle.++p < 0.001 compared with PDGF in absence of the LY294002 inhibitor.

DISCUSSION

The results of the present study indicate that PDGF is an important activator of Pi transport in osteoblast-like cells. Pi uptake in MC3T3E-1 cells was linear during the first 6 minutes of incubation in both vehicle- and PDGF-treated cells. A linear Pi uptake process corresponds to Pi influx into the cells and can be considered as a good estimate of the activity of Pi transport system(s) in the plasma membrane. As shown in Fig. 1, PDGF enhanced a sodium-dependent and not a sodium-independent component of Pi transport. This effect was not associated with alteration of the Na-coupled alanine uptake, which uses the same sodium gradient driving force. These observations indicate that PDGF selectively increased the activity of a Na-dependent Pi transport system(s) located in the plasma membrane of osteoblast like cells. Compared with other growth factors or experimental conditions such as IGF-138 or Pi limitation39 known to stimulate Pi transport in epithelial cells, PDGF appears to be a potent stimulator of Pi transport in bone forming and fibroblastic cells.3 In response to exposure to 50 ng/ml PDGF-BB for 24 h, a 6- to 7-fold increased Pi transport activity was recorded in MC3T3-E1 cells (Fig. 2). The change in Pi transport induced by PDGF was maximally expressed after several hours of PDGF exposure (Fig. 2, left panel), and this response was impaired by inhibiting translational processes. These observations, and the information that PDGF increased the Vmax of the Pi transport system without alteration of the Km (Table 2), suggest that the synthesis of Pi carriers and their insertion in the plasma membrane were enhanced. A change in Pi transport activity was already significantly recorded after 15 minutes of exposure to PDGF, suggesting that post-translational mechanisms might also be involved in activation of Pi transport in response to this growth factor. These regulatory processes are similar to those recently described for the regulation of renal Pi transport in kidney epithelial cells of the proximal tubule in response to alteration in dietary Pi.40,41

The stimulation of Pi transport activity induced by PDGF in MC3T3-E1 cells was associated with a pronounced change in protein tyrosine phosphorylation (Fig. 3). The selective inhibitor of the PDGF receptor tyrosine kinase, CGP 53716, completely blocked autophosphorylation of the PDGF receptor and tyrosine phosphorlation of most of the proteins (Fig. 3). Associated with this effect, CGP 53716 completely blunted Pi transport stimulation induced by PDGF, suggesting that this hormonal response is mediated by a tyrosine phosphorylation-dependent signaling process. Among the signaling proteins activated in response to PDGF and inhibited by CGP 53716, the PLCγ and the p85 subunit of PI-3-kinase were molecules of particular interest for mediating the change in Pi transport. Indeed, it has been shown that activation of PKC by phorbol esters enhances Pi transport in osteoblast-like cells35 and that PI-3-kinase plays an essential role in insulin-induced glucose transport.36 PLCγ and PI 3-kinase have also been shown to mediate activation of Na+/H+ exchange induced by PDGF in normal murine mammary gland epithelial and chinese hamster ovary cells.2 In addition to these two signaling pathways, the role of the MAP kinase was also of interest since, as mentioned above, this enzyme activates major pathways for regulation of mitogenic and differentiation processes. Its role in regulating Pi transport activity in response to growth factors has not been investigated. The results obtained in this study using cAMP stimulating agents, which have been shown to inhibit Ras-mediated activation of MAP kinase,29,32,33 indicate that the MAP kinase pathway is not involved in the regulation of Pi transport in response to PDGF. Indeed, inhibition of MAP kinase activation by PDGF in response to stimulators of cAMP level did not influence the stimulation of Pi transport induced by this growth factor. Also, and as previously reported in UMR-106 osteoblast-like cells,34 cAMP-stimulating agents significantly increased basal Pi transport activity in MC3T3-E1 cells. As expected, however, the mitogenic response induced by PDGF was completely blocked. These observations indicate the existence of distinct cellular processes for the regulation of Pi transport and those controlling cell proliferation in response to growth factors.

Activation of PLCγ results in production of diacylglycerol which stimulates PKC and the synthesis of 1,4,5-inositol trisphosphate, which induces intracellular calcium release. As mentioned earlier, activation of PKC by phorbol esters has been shown to stimulate Pi transport in osteoblast-like cells35 suggesting that this pathway could be involved in the regulation of Pi transport in MC3T3-E1 cells. Using different PKC inhibitors and experimental conditions which inhibit PKC activity, the results of the present study strongly suggest that this enzyme is an essential mediator of Pi transport stimulation induced by PDGF. Indeed, both calphostin C and staurosporine completely blocked Pi transport stimulation induced by PDGF. Down-regulation of PKC by pretreatment of the cells with phorbol esters also prevented this hormonal response. In these cells, Pi transport stimulation induced by IGF-1 remained unchanged, suggesting that PKC does not mediate this hormonal effect. This observation also suggests that different cellular processes are likely to be involved in the regulation of Pi transport in response to various growth factors. Indirect evidence that other tyrosine phosphorylation-dependent processes than the PLCγ-PKC pathway might be involved in the regulation of Pi transport is provided by data with staurosporine. Indeed, in MC3T3-E1 cells treated with this mixed inhibitor of tyrosine kinase and PKC, the basal Pi transport activity was markedly reduced (Fig. 6, right panel), an effect not observed with calphostin C, a more specific PKC inhibitor. This observation suggests that Pi transport activity in osteoblast-like cells is also regulated by tyrosine phosphorylation-dependent processes not involving PKC activation as recently described in MC3T3-E1 cells in response to fluoroaluminate.42

Phosphoinositide metabolism has been implicated as an important process of growth factor signal transduction cascades. The 3-phosphoinositide pathway was suggested to play an important role in regulation of solute transport systems.2,36 Two selective PI-3-kinase inhibitors, LY294002, a derivative of quercetin, and wortmannin, an antifungal antibiotic were recently described.37,43 In MC3T3-E1 cells, LY294002 decreased Pi transport stimulation induced by PDGF (Fig. 8). Similar inhibitory effects were obtained with wortmannin (data not shown). However, a relatively high concentration (>200 nM) of this inhibitor, compared with inhibition of PI-3-kinase–dependent functions in other cell systems, was required for alteration of PDGF-induced Pi transport stimulation. Nevertheless, the data obtained with LY294002 suggest that the 3-phosphoinositide pathway is probably also involved in the regulation of Pi transport in response to PDGF. The cellular mechanism involved in this regulatory process remains unclear. By analogy with the role of PI-3-kinase activity in the regulation of glucose transport in CHO cells,44 it is tempting to speculate that enhanced PI-3-kinase activity might be responsible for PDGF-stimulated translocation of Pi transporters from a cellular vesicular pool to the plasma membrane of MC3T3-E1 cells.

The consequences of enhanced Pi transport activity in response to PDGF in osteoblast-like cells on either alteration in cell metabolism or/and initial events of bone matrix calcification are important functional aspects which remain to be investigated.

In conclusion, PDGF is a potent and selective stimulator of Na-coupled Pi transport activity in osteoblast-like cells. The response probably results from the synthesis and insertion in the plasma membrane of newly synthesized Na-dependent Pi transporters. The initial signaling events involve tyrosine phosphorylation of PLCγ and of the p85 regulatory subunit of PI-3-kinase. Activation of PLCγ and PKC is an essential pathway for the regulation of Pi transport in response to PDGF. PI-3-kinase also participates in this regulatory process by a yet unknown cellular mechanism which could be common to several ion transport systems.

Acknowledgements

We are indebted to Pierre Apostolides for excellent technical help. This work was supported by the Swiss National Science Foundation (Grant No. 32-324411 91).

Ancillary