Vascular endothelial growth factor stimulates chemotactic migration of primary human osteoblasts

Bone Vol. 30, No. 3 March 2002:472– 477

Vascular Endothelial Growth Factor Stimulates Chemotactic Migration of Primary Human Osteoblasts ¨ NTHER,1 U. MAYR-WOHLFART,1 J. WALTENBERGER,2 H. HAUSSER,1 S. KESSLER,1 K.-P. GU 3 1 1 C. DEHIO, W. PUHL, AND R. E. BRENNER 1

Orthopaedic Department (RKU), University of Ulm, Ulm, Germany Department of Internal Medicine II, Ulm University Medical Center, Ulm, Germany 3 Department Molecular Microbiology, Biozentrum of the University of Basel, Basel, Switzerland 2

Introduction Recent studies have indicated a critical role for vascular endothelial growth factor (VEGF) during the process of endochondral ossification, in particular in coupling cartilage resorption with bone formation. Therefore, we studied the chemoattractive and proliferative properties of human VEGF-A on primary human osteoblasts (PHO) and compared these data with the effects of human basic fibroblast growth factor (bFGF) and human bone morphogenetic protein-2 (BMP-2). Furthermore, initial experiments were carried out to characterize VEGF-binding proteins on osteoblastic cells possibly involved in the response. For the first time, to our knowledge, we could demonstrate a chemoattractive effect of VEGF-A, but not VEGF-E, on primary human osteoblasts. The effect of VEGF-A was dose-dependent and did not reach a maximum within the concentration range tested (up to 10 ng/mL). The maximal effect observed was a chemotactic index (CI) of 2 at a concentration of 10 ng/mL. bFGF and BMP-2 exhibited maxima at 1.0 ng/mL with CI values of 2.5 and 2, respectively. In addition to its effect on cell migration, VEGF-A stimulated cell proliferation by up to 70%. Reverse transcription-polymerase chain reaction (RTPCR) analysis revealed the expression of VEGF receptors VEGFR-1 (Flt-1), VEGFR-2 (Kdr), and VEGFR-3 (Flt-4), as well as neuropilin-1 and -2. An in vitro kinase assay failed to demonstrate activation of VEGFR-2 upon stimulation with either VEGF-E or VEGF-A, consistent with the idea that the effect of VEGF-A on primary human osteoblasts is mediated via VEGFR-1. Taken together, our data establish that human osteoblasts respond to VEGF-A, suggesting a functional role for this growth factor in bone formation and remodeling. (Bone 30:472– 477; 2002) © 2002 by Elsevier Science Inc. All rights reserved.

The critical role of angiogenesis for successful osteogenesis during endochondral ossification and fracture repair is well documented.2,30 Vascular endothelial growth factor (VEGF) is a potent mitogen for endothelial cells, and plays a key role in normal and pathological angiogenesis.8,12,17,19 VEGF is secreted by many cell types, including osteoblasts and osteoblast-like cells, and its expression is regulated by several growth factors, hormones, and cytokines (insulin-like growth factor-1 [IGF-1], prostaglandin E1 [PGE1], PGE2, 1,25-dihydroxyvitamin D3, parathyroid hormone [PTH], and transforming growth factor-␤1 [TGF-␤1]).11,13,34,41 In addition to its reported effects on endothelial cells, VEGF indirectly induces proliferation and differentiation of osteoblasts by stimulating endothelial cells to produce osteoanabolic growth factors.40 A direct effect of VEGF on fetal bovine osteoblast differentiation has been described.28 VEGF mRNA is present in hypertrophic chondrocytes in the mouse epiphyseal growth plate, wherein VEGF-dependent blood vessel invasion appears to be essential for coupling cartilage resorption with bone formation.10 Horner et al.14 described the expression of VEGF by chondrocytes in the lower hypertrophic and mineralized region of human neonatal growth plate cartilage. Garcia-Ramirez et al.9 observed a more widespread expression in human fetal growth plate cartilage that was maintained in primary culture of human fetal epiphyseal chondrocytes. Chemotactic migration of bone forming cells is an important physiological event during bone formation, bone remodeling, and fracture healing. Accordingly, the chemotactic response of osteoblasts and osteosarcoma cells has been investigated using a number of different growth factors, such as TGF-␤, platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF), and bone morphogenetic proteins (BMPs),15,23,24,31,37 and some of these factors have been shown to potently stimulate chemotaxis of osteoblastic cells. However, a direct effect of VEGF on primary human osteoblasts has not been described so far. Therefore, the aim of this study was to assess and quantify the chemotactic effect of VEGF on normal primary human osteoblasts in comparison to BMP-2 and bFGF, and to establish the expression of VEGF receptors by osteoblastic cells.

Key Words: Chemotaxis; Vascular endothelial growth factor (VEGF); Bone morphogenetic protein (BMP); Cell migration; Human osteoblast; VEGF receptors.

Materials and Methods Address for correspondence and reprints: Dr. R. E. Brenner, Orthopaedic Department (RKU), Division for Biochemistry of Joint and Connective Tissue Diseases, University of Ulm, Oberer Eselsberg 45, D-89081 Ulm, Germany. E-mail: [email protected] © 2002 by Elsevier Science Inc. All rights reserved.

Cell Culture Osteoblast cultures were established from cancellous human bone fragments derived from routine hip and knee replacements 472

8756-3282/02/$22.00 PII S8756-3282(01)00690-1

Bone Vol. 30, No. 3 March 2002:472– 477

according to the method described by Robey and Termine.32 The fragments obtained from six donors (age 55–73 years) were digested with collagenase for 2 h and the cells obtained were plated and cultured in Dulbecco’s minimal essential medium (DMEM) containing 10% FCS as described previously.25 Experiments were performed only in the first three cell passages and in the stage of cell maturation. In previous studies, we showed that this technique of cell isolation maintains the osteoblastic phenotype as deduced from low expression of collagen type III compared with type I, intracellular cAMP response to PTH, activity of alkaline phosphatase, osteocalcin synthesis, and von Kossa staining.25,29 In addition, immunostaining with an anti-CD31 monoclonal antibody (MAb; JC 70, Dako, Carpinteria, CA) revealed ⬍1% positive cells, indicating that there was no major contamination with endothelial cells from microvessels. Growth Factors and Antibodies Recombinant human (rh)-bFGF was purchased from TEBU GmbH (Frankfurt, Germany), rh-VEGF-A was obtained from GBFmbH (Braunschweig, Germany), and rh-BMP-2 was kindly provided by the Theodor-Boveri-Institut, University of Wu¨ rzburg, Germany. Recombinant VEGF-E (B074) was expressed and purified as described previously for D1701.18,27 Both preparations showed similar activity. rh-PIGF-1 was purchased from R & D (Abingdon, UK). The affinity-purified antibody against a peptide of human VEGFR-1 was purchased from Alpha Diagnostics International (San Antonio, TX). Measurement of Proliferation Osteoblastic cells were plated at a density of 20,000 cells/well in 96 well flat-bottomed Nunclon microwell plates (Nunc GmbH, Wiesbaden, Germany) and grown in DMEM culture medium supplemented with 10% FCS, 2 mmol/L L-glutamine, 100 U/mL streptomycin/penicillin, and 1% amphotericin B (all substances from Biochrom KG, Berlin, Germany). Cells were allowed to adhere for 24 h. Thereafter, serum-free DMEM culture medium alone or with one of four different concentrations of growth factors (0.1, 1, 10, or 100 ng/mL) was added and the plates were incubated for an additional 72 h. Trypan blue staining after the incubation period revealed about 2% positive cells, which indicated that the cell viability was not reduced. The proliferation was determined by a colorimetric hexosaminidase assay as described elsewhere.21 p-nitrophenyl-N-acetyl-␤-glucose aminide (Sigma Aldrich, Deisenhofen, Germany) was dissolved at 7.5 mmol/L in 0.1 mol/L citrate buffer (pH 5), mixed with an equal volume of 0.5% Triton X-100 (Merck KG, Darmstadt, Germany), and added in quantities of 100 ␮L to 50 ␮L of cell suspension. After incubation for 90 min at 37°C and 95% humidity, the reaction was stopped by adding 100 ␮L of 50 mmol/L glycine buffer (pH 10.4) containing 5 mmol/L ethylenediamine tetraacetic acid (EDTA), and absorbance at 405 nm was recorded. In every assay, a standard curve using 10,000 –50,000 cells was included. Chemotaxis Assay Cells from six different donors as well as SaOS-2 cells were used for the chemotaxis experiments. Chemotactic responses were measured by a modified Boyden chamber assay using a 48 well microchemotaxis chamber (NeuroProbe, Inc., Baltimore, MD) with polycarbonate filters with 8 ␮m pores (Nucleopore, Corning Costar Corp.).6 Filters were coated with 5 mg/L gelatine (Merck GmbH, Darmstadt, Germany) according to the instructions of the

U. Meyr-Wohlfart et al. VEGF stimulates chemotactic migration of osteoblasts

473

supplier. Growth factor dilutions in DMEM medium were added to the lower well in 27 ␮L aliquots. The filter was secured in place and 50 ␮L of osteoblast suspension in serum-free DMEM medium (2 ⫻ 105 cells/mL) was added to each of the upper wells of the chamber. After a 4 h incubation period, the chamber was disassembled and the filter was carefully removed. Cells on the upper aspect of the filter were removed by rinsing in cold phosphate-buffered saline (PBS; pH 7.0) and scraping over a rubber wiper. The remaining cells on the lower aspect of the filter were then fixed in 4% formaldehyde for 4 min and stained with toluidine blue for 10 min. Control wells with medium only in the bottom well were applied for each experiment. Conditioned medium of cultured human fetal osteoblasts was used as positive control in a manner similar to that with conditioned medium from fetal cultures for fibroblast chemotaxis.26 The number of migrated cells was counted for 12 random fields per well at ⫻100 magnification. After 4 h, ⬎700 of 10,000 cells had migrated through the filter in the positive control, an average of 35 in the negative control, and between 55 and 170 after stimulation with optimal concentrations of various growth factors. Results were expressed as chemotactic index (CI), giving the average number of migrated cells upon stimulation over the average number of migrated cells without stimulation. Chemotaxis control assessment, a Zigmond–Hirsch checkerboard analysis, was performed in triplicate to distinguish between concentration-dependent cell migration (chemotaxis) and random migration (chemokinesis). The analysis was performed after eliminating the concentration gradient by adding the chemoattractant to the upper chamber with the cell suspension.42 Bottom wells were examined for cells, which had migrated through the filter but did not attach to the membrane. Reverse Transcription-Polymerase Chain Reaction (RT-PCR) Total RNA was extracted from cultured cells using the RNeasy Mini Kit (Qiagen, Hilden, Germany) and digested with RNasefree DNase (Qiagen) to remove any contaminating genomic DNA. cDNA synthesis from total RNA was performed with Omniscript reverse transcriptase (Qiagen) using (dT)15 (1 ␮mol/L) and random hexanucleotide primers (5 ␮mol/L; Roche Diagnostics, Mannheim, Germany) simultaneously. Aliquots of the cDNAs were incubated with HotStarTaq DNA-polymerase (Qiagen) and the following primers (each at a final concentration of 1 pmol/L): for VEGFR-1 (accession NM_002019), 5⬘-GGA ACA AGG CAA GAA ACC AA-3⬘ and 5⬘-CGA TGA ATG CAC TTT CTG GA-3⬘ (product size 216 bp); for VEGFR-2 (accession AF063658), 5⬘-ATC CCT GTG GAT CTG AAA CG-3⬘ and 5⬘-CCA AGA ACT CCA TGC CCT TA-3⬘ (product size 196 bp); for VEGFR-3 (accession NM_002020), 5⬘-TGA AAG CAT CTT CGA CAA GG-3⬘ and 5⬘-TTC AGC ATG ATG TGG CGT AT-3⬘ (product size 201 bp); for neuropilin-1 (accession XM_005798), 5⬘-GGT GGA TGA ATG TGA TGA CG-3⬘ and 5⬘-GCA CGT GAT TGT CAT GTT CC-3⬘ (product size 212 bp), for neuropilin-2 (accession XM_002670), 5⬘-CAT CAG GTT CAC CTC CGA CT-3⬘ and 5⬘-GCT CCA GGT CAA AGA TCA GG-3⬘ (product size 230 bp); and for GAPDH (accession NM_002046), 5⬘-GAG TCC ACT GGC GTC TTC AC-3⬘ and 5⬘-GGT GCT AAG CAG TTG GTG GT-3⬘ (product size 188 bp). All primer pairs were designed to encompass at least one intron in the genomic sequence to allow for discrimination of any sequences amplified from contaminating genomic DNA. All primers were synthesized by MWG Biotech (Munich, Germany) and were of HPSF quality. The polymerase was activated (14 min at 94°C) and then up to 40 cycles (1 min at 94°C, 45 sec at 60°C, 1 min at 72°C) were performed on a RoboCycler Gradient 96 (Stratagene, Amsterdam, The Netherlands). Amplification

474

U. Meyr-Wohlfart et al. VEGF stimulates chemotactic migration of osteoblasts

Bone Vol. 30, No. 3 March 2002:472– 477

Figure 1. Dose-dependent migration of PHOs stimulated with different growth factors. Results presented as chemotactic index (CI) from six independent experiments (A–C), measured in triplicate (means ⫾ SD). (A) rh-VEGF-A165 stimulated migration in a dose-dependent manner with a maximum at 10 ng/mL, CI ⫽ 2.4 ⫾ 0.3 (p ⫽ 0.0004). (B) rh-bFGF stimulated with a maximum at 1.0 ng/mL, CI ⫽ 2.5 ⫾ 0.2 (p ⫽ 0.001). (C) rh-BMP-2 stimulated with a maximum at 1.0 ng/mL, CI ⫽ 1.9 ⫾ 0.1 (p ⫽ 0.003). (D) rh-PIGF-1 stimulated with a maximum at 30 ng/mL, CI ⫽ 3.6 ⫾ 0.1. Statistics obtained using Student’s t-test for paired samples.

products were visualized by agarose gel electrophoresis after staining with ethidium bromide. In Vitro Kinase Analysis Primary human osteoblasts and porcine aortic endothelial cells expressing VEGFR-2 (PAEC/KDR) were cultured in DMEM and Ham’s F-12, respectively, both containing 10% fetal calf serum (FCS). Prior to the experiments, subconfluent cultures were starved overnight by serum deprivation. In vitro kinase assays were performed essentially as described recently in another study.39 Cells were preincubated for 5 min with sodium orthovanadate (100 ␮mol/L) at 37°C, then stimulated for 5 min with either VEGF-A or VEGF-E (50 ng/mL, 37°C), and subsequently solubilized in lysis buffer (1% CHAPS, 10% glycerol, 150 mmol/L NaCl, 20 mmol/L Tris-HCl [pH 7.4], 10 mmol/L EDTA, 100 ␮mol/L Na3VO4, 1% Trasylol, and 1 mmol/L phenylmethylsulfonyl fluoride). Immunoprecipitation was performed either with PY20, a monoclonal antiphosphotyrosine antibody (Transduction Laboratories, Germany), or with a polyclonal antiserum recognizing VEGFR-2 (NEF38). Immunoprecipitates bound to Protein A-Sepharose CL-4B were used for immune-complex kinase reaction in 25 ␮L of 50 mmol/L HEPES (pH 7.4), 10 mmol/L MgCl2, 0.05% Triton X-100, 1 mmol/L dithiothreitol. The kinase reaction was carried out for 7 min at

room temperature in the presence of 5 ␮Ci [␥-32P]-adenosine triphosphate (ATP). After solubilization with sample buffer the proteins were subjected to sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and the incorporated radioactivity was visualized by autoradiography using Hyperfilm MP (Amersham Life Sciences, UK). Statistics Experiments were performed either in duplicate in six independent experiments (with cells from six different donors; experiments presented in Figures 1A–C and 2) or in triplicate (experiments presented in Figures 1D and 5 and Table 1) and the mean value was used for statistical analysis. Results are presented as mean ⫾ standard deviation. The significance of differences between control and stimulation groups was determined using Student’s t-test for paired samples. p ⬍ 0.05 was considered statistically significant. Results bFGF, VEGF-A, and BMP-2 all induced a migratory response in primary human osteoblasts (Figure 1A–C), but to a different degree and at different optimal concentrations. The stimulation curves were bell-shaped for bFGF and BMP-2 with a maximum

Bone Vol. 30, No. 3 March 2002:472– 477

U. Meyr-Wohlfart et al. VEGF stimulates chemotactic migration of osteoblasts

475

Table 1. Checkerboard analysis of response of human osteoblasts to vascular endothelial growth factor (VEGF)-A165 VEGF-A165 conc. (ng/mL) lower chamber

0.0

0.1

1

10

0.0 0.1 1 10

100 196 232 335

67 109 187 212

31 69 80 169

31 38 75 120

VEGF-A165 conc. (ng/mL), upper chamber

Data expressed as percentage of migrated cells in controls in triplicate wells (SD ⬍25%).

at 1.0 ng/mL and CI values of 2.5 ⫾ 0.2 for bFGF and 1.9 ⫾ 0.1 for BMP-2. VEGF-A also led to a dose-dependent stimulation of migration, with the maximal value for CI (2.4 ⫾ 0.3) being observed at the maximal concentration tested (10 ng/mL) (Figure 1A). The chemotactic response to conditioned medium of cultured human fetal osteoblasts served as a positive control (CI value of at least 10). Most of the cells that migrated toward VEGF-A were positive for ALP, indicating that the migrating cells were indeed osteoblasts. To rule out the possibility that VEGF-A merely induced chemokinesis in osteoblastic cells, a checkerboard analysis was performed. This analysis showed that the VEGF-A-stimulated migration was present only at a positive concentration gradient, indicating that the cell migration seen in response to VEGF-A was due to true chemotaxis, not chemokinesis (Table 1). The addition of bFGF induced an increase in cell proliferation by up to 150% at a concentration of 1.0 ng/mL, the addition of VEGF led to an increase of 70% at the same concentration compared with nonstimulated cells, whereas BMP-2 showed no stimulatory effect on proliferation (Figure 2). Consistent data concerning the expression of VEGF receptors by osteoblastic cells are lacking so far. To identify the receptor(s) possibly involved in mediating the observed effects of VEGF-A on osteoblastic cells, we first studied the expression of the known VEGF receptors and binding proteins by RT-PCR, using the same cells as used in the experiments described in the last paragraph of this section. Primary human osteoblasts as well as SaOS-2 cells expressed VEGFR-1, VEGFR-2, and VEGFR-3 (Figure 3A). However, unambiguous identification was possible only after increasing the cycle number from 32 to 40, suggesting rather low steady-state levels of the respective mRNAs in these

Figure 3. Expression of VEGF receptors by osteoblastic cells. PHO and SaOS-2 cells used for the experiments (see Results) were subcultivated for 24 h before isolation of total RNA, cDNA synthesis, and PCR (see Materials and Methods). Forty amplification cycles were employed to demonstrate expression of VEGF receptors (VEGFR-1, VEGFR-2, and VEGFR-3) (A). Thirty-two cycles were used for the amplification of neuropilin (Np)-1 and -2 (B).

cells. Expression of VEGFR-1 was more prominent in primary human osteoblasts (PHOs) compared with SaOS-2 cells, whereas SaOS-2 cells exhibited stronger expression of VEGFR-3. Expression of VEGFR-1 in osteoblasts was corroborated by positive immunostaining using an affinity-purified polyclonal antipeptide antibody against VEGFR-1 (data not shown). In addition, both cell types expressed the VEGF binding proteins neuropilin-1 and -2, with the expression of neuropilin-2 being considerably stronger in SaOS-2 cells (Figure 3B). To clarify the matter of which VEGF receptors present on osteoblastic cells mediate PHO activation, we analyzed the activation of VEGFR-2 using an in vitro kinase assay following stimulation of VEGFR-2 (using VEGF-E) or stimulation of VEGFR-2 and VEGFR-1 (using VEGF-A). This assay clearly showed a lack of activation of VEGFR-2 (Figure 4). Because of the high sensitivity of this assay, we conclude that the effect of VEGF-A in PHOs is not mediated via KDR. These data are consistent with further chemotaxis experiments using VEGF-E, which mediates its effects via VEGFR-2 and did not show a stimulation of PHO migration at the concentration of 10 ng/mL (CI 0.9). On the other hand, PIGF-1, which interacts with VEGFR-1 but not VEGFR-2, did stimulate the migration of PHOs (Figure 1D). Similar results for VEGF-A, VEGF-E, and PIGF-1 were obtained with the SaOS-2 cell line (Figure 5). These data are consistent with the idea that VEGFR-1 mediates the migratory response of osteoblasts. Discussion

Figure 2. Effects of rh-VEGF-A165, rh-bFGF, and rhBMP-2 on proliferation of PHOs. Cells were cultivated for 72 h in the presence of 0.1, 1.0, 10, and 100 ng/mL of growth factors. Proliferation rates were determined by the hexosaminidase assay as described in Materials and Methods. PHOs of six donors were split and independently cultivated in three wells. Cell proliferation was measured from cells in each well in triplicate. Data expressed as mean ⫾ SD.

The chemotactic effects of growth factors and cytokines have been studied in osteogenic cells. Rat osteoblast-like cells and rat osteosarcoma cells have been investigated for a chemotactic reaction to TGF-␤, PDGF, and interleukin (IL)-1␤.15,28,31 Chemotactic properties of aFGF, bFGF, and BMPs were studied with human osteoblasts.23,24 Our study, however, focused on the chemotactic and the proliferative effect of VEGF-A, an angiogenic factor, on human osteoblasts, and on the identification of the VEGF receptor involved in mediating VEGF signals in osteoblasts. Toward this goal, we studied the chemotactic and proliferative response of osteoblastic cells, primary human osteoblasts, and SaOS-2 cells to VEGF-A, VEGF-E, and PIGF-1 and compared their effects with the effects of other osteogenic growth factors like bFGF and BMP-2. Our results demonstrate that human osteoblasts respond to VEGF-A with chemotactic cell migration and with increased proliferation. Our data further

476

U. Meyr-Wohlfart et al. VEGF stimulates chemotactic migration of osteoblasts

Bone Vol. 30, No. 3 March 2002:472– 477

Figure 4. Lack of activation of VEGFR-2/KDR in primary human osteoblasts (PHOs) in response to stimulation with VEGF-A165. PHOs from two different patients were stimulated with either VEGF-A165 (50 ng/mL) or VEGF-E (50 ng/mL) for 8 min prior to cell lysis, in vitro kinase assay, SDS-PAGE, and autoradiography. Whereas both VEGF-A and VEGF-E stimulated tyrosine phosphorylation of VEGFR-2/KDR in PAE/KDR cells (the phosphorylated receptor appeared at 210 kDa), no such signal could be detected in PHOs from two different patients. This was not only true after immunoprecipitation with the VEGFR-2/KDR-specific anitserum, NEF, but also after immunoprecipitation with the antiphosphotyrosine antibody, PY20, which indicates that no other tyrosine phosphorylation events were activated in the cells after stimualtion with either VEGF-A or VEGF-E.

suggest that these effects of VEGF-A on human osteoblasts are mediated via VEGFR-1. Midy and Plouet28 noted that fetal bovine osteoblasts bind VEGF at an apparent single site and can migrate and increase their alkaline phosphatase expression when exposed to VEGF. The cells did not show any significant change in their DNA synthesis. Our results show that proliferation of primary human osteoblasts could be stimulated by VEGF-A in a dose-dependent manner. The increase, however, was markedly lower compared with the effect of bFGF. The biological function of VEGF in endothelial cells is mediated by binding to specific tyrosine kinase receptors, VEGFR-1 and VEGFR-2.5,36,38,39 The majority of biological effects of VEGF in these cells, including migration, proliferation, and angiogenesis, are mediated via VEGFR-2.27,33,38 To achieve these effects, the isolated stimulation of VEGFR-2 using VEGF-E is sufficient.27 In contrast, the biological activities of VEGF-A on human monocytes are mediated via VEGFR-1.1,4 Recently, mRNA for VEGF and its receptors, VEGFR-1 and VEGFR-2, has been detected in cells of human fracture callus.7 Here, we show that human osteoblastic cells, primary human osteoblasts, as well as SaOS-2 cells express both of these VEGF receptors, VEGFR-1 and -2, as well as a third VEGF receptor, VEGFR-3, which binds VEGF-C and VEGF-D and has so far been demonstrated only in lymphatic tissue.22 In addition to these signal-transducing receptors, two VEGF binding proteins with proposed coreceptor function,

neuropilin-1 and -2, have been identified. These two molecules were also shown to be expressed by human osteoblasts in the present study. Although VEGF-A mediates its effects via both VEGFreceptors, VEGFR-1 and -2, in endothelial cells, several lines of evidence suggest that its effect in human osteoblasts is due to activation of VEGFR-1: (1) PIGF-1, which specifically signals via VEGFR-1 and not via VEGFR-2 has similar effects on osteoblast migration as VEGF-A; (2) VEGF-E, which specifically signals via VEGFR-2 but not via VEGFR-1, did not exhibit any effect on osteoblasts in this study; and finally (3) there was no activation of VEGFR-2 detectable upon stimulation of osteoblasts with VEGF-A. The situation in human osteoblasts therefore appears to be similar to the situation in human monocytes.4 However, the roles of VEGFR-3 and the neurophilins deserve further investigation. bFGF, an angiogenic and proliferative growth factor produced by osteoblasts, is known to increase cell proliferation of bone cells.3,25 In this study, we also observed stimulation of osteoblast migration by bFGF, with a maximal effect at a concentration of 1 ng/mL. These findings are somewhat different from those of Lind et al.,23 who found a maximal stimulation of human trabecular osteoblasts from the iliac crest by bFGF at a much higher concentration of 100 ng/mL. Using 10 ng/mL bFGF, Midy and Plouet28 observed no marked increase in the migration of fetal bovine osteoblasts from calvariae. These conflicting data may be due to the well-known differences in the biological behavior of osteoblastic cells from varying location and species of donor tissue.16,35 However, our results obtained with BMP-2 are in accordance with data described by Lind et al.24 Taken together, our in vitro data indicate that VEGF-A, in addition to other factors such as BMPs or bFGF, may contribute to the activation of PHOs in the context of bone formation and remodeling. Thus, VEGF-A may induce not only angiogenesis to facilitate bone formation, but may also stimulate this process in a direct fashion by induction of proliferation and migration of osteoblasts.

Figure 5. Chemotactic effects of different growth factors on SaOS-2 cells. Chemotaxis assays with SaOS-2 cells were performed as described in Materials and Methods using 10 ng/mL of VEGF-A, VEGF-E, and PlGF-1, respectively. The experiments were done in triplicate, and results are expressed as chemotactic index (CI).

Acknowledgments: The authors thank Ulrike Mayr and Giovanni Ravalli for expert technical assistance. The study was supported in part by grants from the BMBF (JP05SB) and the Deutsche Forschungsgemeinschaft (SFB 497/Project C1 and Wa 734/5-1).

Bone Vol. 30, No. 3 March 2002:472– 477

U. Meyr-Wohlfart et al. VEGF stimulates chemotactic migration of osteoblasts

References 1. Barleon, B., Sozzani, S., Zhou, D., Weich, H. A., Mantovani, A., and Marme, D. Migration of human monocytes in response to vascular endothelial growth factor (VEGF) is mediated via the VEGF receptor flt-1. Blood 87:3336 –3343; 1996. 2. Brighton, C. T. and Hunt, R. M. Early histological and ultrastructural changes in medullary fracture callus. J Bone Jt Surg [Am] 73:832– 847; 1991. 3. Canalis, E., Centrella, M., and McCarthy, T. Effects of basic fibroblast growth factor on bone formation in vitro. J Clin Invest 81:1572–1577; 1988. 4. Clauss, M., Weich, H., Breier, G., Knies, U., Rockl, W., Waltenberger, J., and Risau, W. The vascular endothelial growth factor receptor Flt-1 mediates biological activities. Implications for a functional role of placenta growth factor in monocyte activation and chemotaxis. J Biol Chem 271:17629 –17634; 1996. 5. de Vries, C., Escobedo, J. A., Ueno, H., Houck, K., Ferrara, N., and Williams, L. T. The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 255:989 –991; 1992. 6. Falk, W., Goodwin, R. H., Jr., and Leonard, E. J. A 48-well micro chemotaxis assembly for rapid and accurate measurement of leukocyte migration. J Immunol Meth 33:239 –247; 1980. 7. Ferguson, C., Alpern, E., Miclau, T., and Helms, J. A. Does adult fracture repair recapitulate embryonic skeletal formation? Mech Dev 87:57– 66; 1999. 8. Ferrara, N. and Bunting, S. Vascular endothelial growth factor, a specific regulator of angiogenesis. Curr Opin Nephrol Hypertens 5:35– 44; 1996. 9. Garcia-Ramirez, M., Toran, N., Andaluz, P., Carrascosa, A., and Audi, L. Vascular endothelial growth factor is expressed in human fetal growth cartilage. J Bone Miner Res 15:534 –540; 2000. 10. Gerber, H. P., Vu, T. H., Ryan, A. M., Kowalski, J., Werb, Z., and Ferrara, N. VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5:623– 628; 1999. 11. Goad, D. L., Rubin, J., Wang, H., Tashjian, A. H., Jr., and Patterson, C. Enhanced expression of vascular endothelial growth factor in human SaOS-2 osteoblast-like cells and murine osteoblasts induced by insulin-like growth factor I. Endocrinology 137:2262–2268; 1996. 12. Gospodarowicz, D., Abraham, J. A., and Schilling, J. Isolation and characterization of a vascular endothelial cell mitogen produced by pituitary-derived folliculo stellate cells. Proc Natl Acad Sci USA 86:7311–7315; 1989. 13. Harada, S., Nagy, J. A., Sullivan, K. A., Thomas, K. A., Endo, N., Rodan, G. A., and Rodan, S. B. Induction of vascular endothelial growth factor expression by prostaglandin E2 and E1 in osteoblasts. J Clin Invest 93:2490 – 2496; 1994. 14. Horner, A., Bishop, N. J., Bord, S., Beeton, C., Kelsall, A. W., Coleman, N., and Compston, J. E. Immunolocalisation of vascular endothelial growth factor (VEGF) in human neonatal growth plate cartilage. J Anat 194:519 –524; 1999. 15. Hughes, F. J., Aubin, J. E., and Heersche, J. N. Differential chemotactic responses of different populations of fetal rat calvaria cells to platelet-derived growth factor and transforming growth factor beta. Bone Miner 19:63–74; 1992. 16. Kasperk, C. H., Wergedal, J. E., Strong, D., Farley, J., Wangerin, K., Gropp, H., Ziegler, R., and Baylink, D. J. Human bone cell phenotypes differ depending on their skeletal site of origin. J Clin Endocrinol Metab 80:2511– 2517; 1995. 17. Keck, P. J., Hauser, S. D., Krivi, G., Sanzo, K., Warren, T., Feder, J., and Connolly, D. T. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science 246:1309 –1312; 1989. 18. Kroll, J. and Waltenberger, J. A novel function of VEGF receptor-2 (KDR): Rapid release of nitric oxide in response to VEGF-A stimulation in endothelial cells. Biochem Biophys Res Commun 265(3):636 – 639; 1999. 19. Kroll, J. and Waltenberger, J. Regulation of the endothelial function and angiogenesis by vascular endothelial growth factor-A (VEGF-A). Z Kardiol 89:206 –218; 2000. 20. Ku¨ bler, N. R., Reuther, J. F., Faller, G., Kirchner, T., Ruppert, R., and Sebald, W. Inductive properties of recombinant human BMP-2 produced in a bacterial expression system. Int J Oral Maxillofac Surg 27:305–309; 1998. 21. Landegren, U. Measurement of cell numbers by means of the endogenous enzyme hexosaminidase. Applications to detection of lymphokines and cell surface antigens. J Immunol Meth 67:379 –388; 1984. 22. Larrivee, B. and Karsan, A. Signaling pathways induced by vascular endothelial growth factor. Int J Mol Med 5:447– 456; 2000. 23. Lind, M., Deleuran, B., Thestrup-Pedersen, K., Soballe, K., Eriksen, E. F., and Bunger, C. Chemotaxis of human osteoblasts. Effects of osteotropic growth factors. APMIS 103:140 –146; 1995. 24. Lind, M., Eriksen, E. F., and Bunger, C. Bone morphogenetic protein-2 but not

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

477

bone morphogenetic protein-4 and - 6 stimulates chemotactic migration of human osteoblasts, human marrow osteoblasts, and U2-OS cells. Bone 18:53– 57; 1996. Mayr-Wohlfart, U., Kessler, S., Kno¨ chel, S., Puhl, W., and Gu¨ nther, K.-P. BMP-4 of Xenopus laevis stimulates differentiation of human osteoblast-like cells. J Bone Jt Surg [Br] 83:144 –147; 2001. Mensing, H., Pontz, B. F., Muller, P. K., and Gauss-Muller, V. A study on fibroblast chemotaxis using fibronectin and conditioned medium as chemoattractants. Eur J Cell Biol 29:268 –273; 1983. Meyer, M., Clauss, M., Lepple-Wienhues, A., Waltenberger, J., Augustin, H. G., Ziche, M., Lanz, C., Buttner, M., Rziha, H. J., and Dehio, C. A novel vascular endothelial growth factor encoded by Orf virus, VEGF- E, mediates angiogenesis via signalling through VEGFR-2 (KDR) but not VEGFR-1 (Flt-1) receptor tyrosine kinases. Eur Med Biol Org J 18:363–374; 1999. Midy, V. and Plouet, J. Vasculotropin/vascular endothelial growth factor induces differentiation in cultured osteoblasts. Biochem Biophys Res Commun 199:380 –386; 1994. Mo¨ rike, M., Schulz, M., Nerlich, A., Koschnik, M., Teller, W. M., Vetter, U., Brenner, R. E. Expression of osteoblastic markers in cultured human bone and fracture callus cells. J Mol Med 73(11):571–575; 1995. Parfitt, A. M. Osteonal and hemi-osteonal remodeling: The spatial and temporal framework for signal traffic in adult human bone. J Cell Biochem 55:273– 286; 1994. Pfeilschifter, J., Wolf, O., Naumann, A., Minne, H. W., Mundy, G. R., and Ziegler, R. Chemotactic response of osteoblastlike cells to transforming growth factor beta. J Bone Miner Res 5:825– 830; 1990. Robey, P. G. and Termine, J. D. Human bone cells in vitro. Calcif Tissue Int 37(5):453– 460, 1985. Rousseau, S., Houle, F., Kotanides, H., Witte, L., Waltenberger, J., Landry, J., and Huot, J. Vascular endothelial growth factor (VEGF)-driven actin-based motility is mediated by VEGFR2 and requires concerted activation of stressactivated protein kinase 2 (SAPK2/p38) and geldanamycin-sensitive phosphorylation of focal adhesion kinase. J Biol Chem 275:10661–10672; 2000. Saadeh, P. B., Mehrara, B. J., Steinbrech, D. S., Dudziak, M. E., Greenwald, J. A., Luchs, J. S., Spector, J. A., Ueno, H., Gittes, G. K., and Longaker, M. T. Transforming growth factor-beta1 modulates the expression of vascular endothelial growth factor by osteoblasts. Am J Physiol 277:C628 –C637; 1999. Siggelkow, H., Rebensdorff, K., Niedhart, C., Engel, I., Schulz, H., Atkinson, M. J., and Hufner, M. Development of the osteoblast phenotype in primary human osteoblasts in culture: Comparison with rat calvarial cells in osteoblast differentiation. J Cell Biochem 75:22–35; 1999. Terman, B. I., Dougher-Vermazen, M., Carrion, M. E., Dimitrov, D., Armellino, D. C., Gospodarowicz, D., and Bohlen, P. Identification of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem Biophys Res Commun 187:1579 –1586; 1992. Tsukamoto, T., Matsui, T., Fukase, M., and Fujita, T. Platelet-derived growth factor B chain homodimer enhances chemotaxis and DNA synthesis in normal osteoblast-like cells (MC3T3-E1). Biochem Biophys Res Commun 175:745– 751; 1991. Waltenberger, J., Claesson-Welsh, L., Siegbahn, A., Shibuya, M., and Heldin, C. H. Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem 269:26988 –26995; 1994. Waltenberger, J., Mayr, U., Pentz, S., and Hombach, V. Functional upregulation of the vascular endothelial growth factor receptor KDR by hypoxia. Circulation 94:1647–1654; 1996. Wang, D. S., Miura, M., Demura, H., and Sato, K. Anabolic effects of 1,25-dihydroxyvitamin D3 on osteoblasts are enhanced by vascular endothelial growth factor produced by osteoblasts and by growth factors produced by endothelial cells. Endocrinology 138:2953–2962; 1997. Wang, D. S., Yamazaki, K., Nohtomi, K., Shizume, K., Ohsumi, K., Shibuya, M., Demura, H., and Sato, K. Increase of vascular endothelial growth factor mRNA expression by 1,25-dihydroxyvitamin D3 in human osteoblast-like cells. J Bone Miner Res 11:472– 479; 1996. Zigmond, S. H. and Hirsch, J. G. Leukocyte locomotion and chemotaxis. New methods for evaluation, and demonstration of a cell-derived chemotactic factor. J Exp Med 137:387– 410; 1973.

Date Received: December 14, 2000 Date Revised: October 19, 2001 Date Accepted: October 19, 2001