Both v-Ha-Ras and v-Raf stimulate expression of the vascular endothelial growth factor in NIH 3T3 cells.

Stimulation of NIH 3T3 cells with platelet-derived growth factor (PDGF)-BB and 12-O-tetradecanoylphorbol-13-acetate (TPA) enhances vascular endothelial growth factor (VEGF) gene expression. To address the question of whether Ras and Raf are involved in the induction of VEGF gene expression by PDGF and TPA, we examined the effects of both factors on NIH 3T3 cells stably transfected with v-Ha-ras or v-raf. In serum-starved NIH 3T3 cells, only low levels of mRNA expression can be detected, whereas both ras and raf transformed cell lines express enhanced levels of a 4.3-kilobase VEGF transcript. Stimulation with PDGF or TPA resulted in increased VEGF mRNA in all cell lines, with highest levels found in the transformed cells. Immunofluorescence studies confirmed that the elevated VEGF mRNA expression correlated with enhanced protein levels. Positive immunofluorescence signals could be detected in v-Ha-ras or v-raf transformed cell lines but not in unstimulated NIH 3T3 cells. VEGF from conditioned medium of v-raf transformed NIH 3T3 cells was partially purified by chromatography on heparin-Sepharose. Biological activity of this VEGF protein was demonstrated by competition with binding of recombinant I-VEGF to human umbilical vein endothelial cells and by its ability to stimulate proliferation of these cells.


Stimulation of NIH 3T3 cells with platelet-derived growth factor (PDGF)-BB and 12-O-tetradecanoylphorbol-13-acetate (TPA) enhances vascular endothelial growth factor (VEGF) gene expression. To address the question of whether Ras and Raf are involved in the induction of VEGF gene expression by PDGF and TPA, we examined the effects of both factors on NIH 3T3 cells stably transfected with v-Ha-ras or v-raf. In serumstarved NIH 3T3 cells, only low levels of mRNA expression can be detected, whereas both ras and raf transformed cell lines express enhanced levels of a 4.3kilobase VEGF transcript. Stimulation with PDGF or TPA resulted in increased VEGF mRNA in all cell lines, with highest levels found in the transformed cells. Immunofluorescence studies confirmed that the elevated VEGF mRNA expression correlated with enhanced protein levels. Positive immunofluorescence signals could be detected in v-Ha-ras or v-raf transformed cell lines but not in unstimulated NIH 3T3 cells. VEGF from conditioned medium of v-raf transformed NIH 3T3 cells was partially purified by chromatography on heparin-Sepharose. Biological activity of this VEGF protein was demonstrated by competition with binding of recombinant 125 I-VEGF 165 to human umbilical vein endothelial cells and by its ability to stimulate proliferation of these cells.
Vascular endothelial growth factor (VEGF) 1 has several features as a mediator of both normal and pathological angiogenesis. It specifically stimulates growth of endothelial cells and induces angiogenesis in vivo (1)(2)(3). VEGF is produced by several differentiated cell types (1, 4 -6), and VEGF mRNA is also expressed in various transformed cell lines (7,8). The VEGF gene is transcribed into at least four mRNA types by alternative splicing, giving rise to four distinct VEGF proteins of 121, 165, 189, and 206 amino acids after cleavage of the 26-amino acid signal peptide (9,10). The two smaller proteins are efficiently secreted, whereas the larger forms apparently remain associated with heparin-containing proteoglycans on the cell surface (10). Four AP-1 binding sites have been identified within the promoter sequence of the VEGF gene (9). Cells either treated with phorbol ester, leading to activation of protein kinase C (PKC), or stimulated with PDGF-BB induce VEGF gene expression, suggesting the participation of AP-1 binding sites (9,11). Therefore, PKC is one essential component in signal transduction leading to VEGF expression (11).
Several lines of evidence strongly implicate Ras as an essential component of growth-regulating signal transduction by growth factor receptor protein-tyrosine kinases (12). The c-ras proto-oncogene encodes a 21,000-Da guanine nucleotide-binding protein (13,14), which is thought to be involved in the regulation of cell growth, differentiation, and carcinogenesis (15)(16)(17). When microinjected into quiescent fibroblasts, the Ras protein has been shown to be a potent mitogen (15). In addition, microinjection into pheochromocytoma PC 12 cells has been shown to induce neurite outgrowth (18). Ras also plays an important role in the pathogenesis of several malignancies including tumor growth. Mutations in the ras gene that increase the proportion of Ras in the GTP-bound state relative to Ras-GDP activate the biological function of Ras and have been found in a large number of human tumors (19,20).
The raf proto-oncogene products are cytoplasmic serine/ threonine-specific protein kinases, which are activated by growth factors or phorbol esters (21)(22)(23). 2 They transmit the incoming signals to the transcriptional machinery in the nucleus. Expression of constitutively activated Raf-1 and A-Raf trans-activate transcription from the oncogene-responsive element in the polyomavirus enhancer, which contains an AP-1/ Ets binding site (24), in a Ras-independent manner (25,26). In contrast, mitogen-or Ras-induced expression from the oncogene-responsive element was completely dependent on functional Raf (25). Therefore, Raf has been positioned downstream of Ras in the signal transduction pathway leading to proliferation and transformation of NIH 3T3 cells. Recently, direct interaction of normal and oncogenic Ras with Raf has been reported (27,28). Ras-mediated activation of Raf thus initiates a protein kinase cascade leading to mitogen-activated protein kinase activation and finally to increased AP-1 activity (29,30).
A pronounced synergism between PKC and mutant p53 in the induction of VEGF expression was previously reported (31). We therefore addressed the question of whether a functional link exists between oncogenic transformation of cells and the expression of the tumor angiogenesis factor, VEGF. This would generate a connection between genetic alterations, which are thought to play a role in the process of tumor initiation, and the process of tumor progression, mediated by enhanced expression of VEGF. For this purpose, we have used NIH 3T3 cell lines transformed by v-Ha-ras or v-raf, the viral forms of the respective cellular oncogenes, which have been found to be frequently activated by mutation or overexpressed in certain human tumors (19,20,(32)(33)(34). 3 We show that v-Ha-Ras or v-Raf are sufficient to enhance VEGF expression in NIH 3T3 cells without further stimulation. Furthermore, NIH 3T3 cells transformed by v-raf are able to secrete VEGF, which shows biolog-* 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.
Northern Blot and Slot-Blot Analysis-Total RNA was prepared as described (37), subjected to electrophoresis on a 1.25% agarose gel containing 2.2 M formaldehyde according to the surface tension method (38), and transferred to a nitrocellulose membrane (Hybond C-extra, Amersham) by capillary blotting. Blots were hybridized as described (39,40).
Western Blot Analysis-Cells were lysed in TBST buffer (50 mM Tris hydrochloride, pH 7.2, 150 mM NaCl, 0.2% Triton X-100). Insoluble material was removed by centrifugation at 4°C for 20 min at 12,000 ϫ g. Protein concentrations were determined by the method of Bradford (41). Samples were resolved by SDS-polyacrylamide gel electrophoresis. The gels were electroblotted onto nitrocellulose, and blots were blocked with 5% (w/v) gelatin in TBST buffer and incubated with polyclonal rabbit antiserum against VEGF. After intensive washing with TBST buffer, the blots were incubated with an alkaline phosphatase-conjugated anti-rabbit antibody.
Immunofluorescence-Cells were seeded in chamber slides (Nunc) and cultivated for 24 h. Afterward, the medium was changed to Dulbecco's modified Eagle's medium supplemented with 0.05% bovine calf serum, and cells were incubated for 20 h at 37°C. After incubation for 1 h at 4°C to prevent protein export, cells were fixed with 3.7% formaldehyde in phosphate-buffered saline for 10 min. After washing with phosphate-buffered saline, cells were incubated for 30 min with PTB buffer (0.2% Triton X-100 in phosphate-buffered saline containing 0.1% bovine serum albumin). Then, cells were incubated for 2 h at room temperature with anti-VEGF antibody (K7.16) in the same buffer. Finally, cells were washed three times for 5 min with PTB buffer. Bound VEGF antibody was detected with fluoroscein isothiocyanateconjugated goat anti-rabbit antibody (Sigma).
Binding Competition Assay-Confluent HUVE cells (passage 3) in 24-well plates were incubated with 1.5 ng/ml 125 I-VEGF 165 and increasing amounts of unlabeled recombinant VEGF 165 from Sf9 cells or VEGF from conditioned medium. The concentration of VEGF protein in conditioned medium was estimated by comparison with increasing amounts of recombinant VEGF 165 in a Western blot analysis. After 3 h at 4°C, cells were lysed and radioactivity was measured.
Proliferation Assay-A suspension of HUVE cells (passage 3) was prepared with endothelial cell basal medium supplemented with 10% fetal calf serum. Cells were seeded into 24-well plates (5000 cells per well). After 24 h, either recombinant VEGF 165 from Sf9 cells or VEGF protein from conditioned medium (concentration was estimated as described above) were added. On day 4, medium and VEGF were replaced by fresh material. On day 7, cells were washed, trypsinized, and counted with a coulter counter. The experiment was done three times with essentially identical results.

Effects of v-Ha-Ras on VEGF Gene
Expression-To examine the effects of v-Ha-ras transformation of NIH 3T3 cells on expression of the VEGF gene, VEGF mRNA was assayed in v-Ha-ras transformed and control NIH 3T3 cells by Northern blot analysis. VEGF mRNA steady state levels in serumstarved cells were significantly increased in v-Ha-ras transformed cells as compared to control cells (Fig. 1). Furthermore, VEGF expression could be induced in both cell lines by TPA and PDGF-BB, as previously reported (11). But after stimulation with the PKC activator TPA, VEGF expression was augmented in v-Ha-ras transformed cells as compared to control cells (Fig. 2), suggesting that v-Ha-Ras and PKC act in an additive manner on the VEGF mRNA level. This effect was not detectable when v-Ha-ras transformed cells were stimulated with PDGF-BB (Fig. 2).
It was previously reported that at least in some cell types

FIG. 1. Basal VEGF mRNA expression in v-Ha-ras or v-raf transformed NIH 3T3 cells. NIH 3T3 cells transformed by v-Ha-ras
or v-raf and control NIH 3T3 cells were serum starved for 24 h. Total RNA was isolated, and 10 g were analyzed by Northern blotting using a VEGF 165 cDNA probe (39). The filter was exposed for 48 h at Ϫ80°C with intensifying screen. The rRNA 28 S band is indicated on the left. Equal loading of the gel was confirmed by ethidium bromide staining.

FIG. 2. Time course of VEGF expression of v-Ha-ras transformed and control NIH 3T3 cells. The v-Ha-ras transformed cells (NIH 3T3/v-Ha-ras) and control NIH 3T3 cells (NIH 3T3
) were serum starved for 24 h to achieve quiescence and then stimulated by the addition of TPA (100 ng/ml) and PDGF-BB (50 ng/ml). RNA was extracted at the indicated time (h) after stimulation. 10 g of total RNA were loaded in each lane. The filters were probed with the 600-base pair VEGF 165 complementary DNA (cDNA) probe (39) and exposed for 48 h at Ϫ80°C with intensifying screen. The rRNA 28 S band is indicated on the left. Equal loading of the gel was confirmed by ethidium bromide staining. tumorigenic progression induced by oncogenic ras is associated with increased PKC gene transcription and PKC expression (42)(43)(44). Therefore, we have addressed the question of whether the enhanced TPA inducibility of VEGF expression in NIH 3T3 cells transformed by v-Ha-ras may be due to increased PKC levels in the transfected cell line. Western blot analysis using a polyclonal antibody that recognizes PKC ␣, ␤I/␤II, ␥, and ⑀ did not show any difference in PKC expression between the two cell lines (data not shown), indicating that the increased TPA inducibility of VEGF expression in v-Ha-ras transformed cells is not due to increased PKC levels.
VEGF Gene Expression in raf-transfected NIH 3T3 Cells-To address the question whether constitutive activation of Raf leads to alteration in VEGF expression, v-raf transformed NIH 3T3 cells were examined in Northern blot analysis for VEGF mRNA levels. In serum-starved cells, VEGF expression was significantly increased compared to NIH 3T3 control cells (Fig.  1). The time courses for VEGF expression after stimulation with PDGF-BB and TPA revealed an additional increase in VEGF mRNA to the already elevated basal expression in v-raf transformed cells (Fig. 3). In addition, after stimulation with TPA, VEGF expression was intensified, suggesting a combined action of v-Raf and the PKC activator TPA on VEGF expression.
These results suggest that the Raf kinase is involved in the signal transduction leading to VEGF gene expression. Furthermore, it is strongly indicated that constitutive activation of Raf-1 releases the cell for the necessity of external stimuli to induce VEGF expression.
VEGF Protein Expression in ras-or raf-transformed Cell Lines-To confirm the high basal expression of VEGF mRNA, VEGF protein expression in v-Ha-ras or v-raf transformed cells was analyzed by indirect immunofluorescence using affinitypurified VEGF antibodies (Fig. 4). The specific VEGF fluorescence signal in serum-starved unstimulated NIH 3T3 cells was very weak referring to low VEGF protein expression, whereas the signal in the transformed cell lines was significantly increased. To evaluate the specificity of the VEGF antiserum, immunofluorescence studies were also performed with the corresponding pre-immuneserum, and no fluorescence signal could be detected (data not shown).
By expressing VEGF transiently in mammalian cells, it could be shown that only VEGF 165 and VEGF 121 are efficiently secreted into the medium, whereas the longer forms VEGF 189 and VEGF 206 are rather cell associated (10). Based on this observation, together with the strong VEGF immunofluorescence in v-Ha-ras and v-raf transformed cells, we partially purified VEGF from medium conditioned by the v-raf transformed cell line. VEGF could be identified in fractions eluted from heparin-Sepharose columns by Western blot analysis (Fig.  5A).
The fraction containing the highest amount of VEGF protein was tested for competition with 125 I-labeled rVEGF 165 in a binding assay (Fig. 5B). The competition was similar to the positive control with rVEGF 165 , whereas PDGF-BB and bFGF even at a concentration of 100 ng/ml showed no competition at all. Furthermore, VEGF protein isolated from v-raf transformed cells promoted proliferation of HUVE cells similar to rVEGF 165 (Fig. 5C), suggesting that VEGF expressed in NIH 3T3 cells is biologically active.

DISCUSSION
The identification of mitogen-activated protein kinase kinase as a substrate of Raf (45) together with the demonstration of direct phosphorylation of c-Jun (46) by mitogen-activated protein kinase extended the signal transduction pathway (receptor kinase, Ras, Raf, mitogen-activated protein kinase kinase, mitogen-activated protein kinase (Ras/Raf pathway)) into the nu-

FIG. 3. Time course of VEGF expression in v-raf transformed NIH 3T3 cells. NIH 3T3 cells transformed by v-raf (NIH 3T3
/v-raf) as well as control cells (NIH 3T3) were serum starved for 24 h. Cells were stimulated with PDGF-BB (40 ng/ml) and TPA (100 ng/ml) for the indicated periods (h). Total RNA was isolated, and 10 g were analyzed by Northern blotting using a VEGF 165 cDNA probe (39). The filters were exposed for 48 h at Ϫ80°C with intensifying screens. The rRNA 28 S band is indicated on the left. Equal loading was assured by staining the RNA gels with ethidium bromide.

FIG. 4. Basal VEGF protein expression in v-Ha-ras or v-raf transformed NIH 3T3 cells. NIH 3T3 cells transformed by v-Ha-ras
(NIH 3T3/v-Ha-ras) or v-raf (NIH 3T3/v-raf) and NIH 3T3 control cells (NIH 3T3) were maintained for 24 h in medium supplemented with 0.05% serum. Immunofluorescence with a VEGF-specific antiserum was performed as described under "Experimental Procedures." cleus. A transforming mutant of raf leads to trans-activation of promoters containing AP1 binding sites (25). In addition, it could be shown that various natural promoters containing AP1/ Ets binding sites respond to Raf activation (47)(48)(49)(50)(51). It was also shown that expression of oncogenic Ras leads to an increase in the expression of c-Jun and c-Fos and therefore to an increased AP1 activity (29). The four AP1 binding sites in the promoter region of the VEGF gene (9) render this gene a potential downstream effector in Ras and Raf activation.
Finkenzeller et al. (11) could show that VEGF mRNA expression in NIH 3T3 cells is intensified after PDGF-BB and TPA stimulation. Furthermore, they postulate an essential role of PKC in the signal transduction leading to VEGF induction. Together, with the demonstration of direct phosphorylation of Raf-1 by PKC␣ (52), the Raf kinases are placed in a central position in the signal transduction pathway of VEGF expression.
The enhanced basal expression of VEGF mRNA in serumstarved NIH 3T3 cells transformed by v-Ha-ras or v-raf (Fig. 1) also indicates that the Ras/Raf signal transduction pathway plays an essential role in the regulation of VEGF gene expression. Despite the increased expression in ras-or raf-transformed cell lines, VEGF expression was still further enhanced by treatment of cells with PDGF and TPA (Figs. 2 and 3). This increase of VEGF indicates that a second, Ras-and Raf-independent, signal transduction pathway may exist or, with regard to Raf, that the already constitutively activated Raf kinase becomes additionally positively regulated by phosphorylation of specific sites in the kinase domain after PDGF or TPA stimulation.
The enhanced basal VEGF mRNA expression also resulted in higher VEGF protein expression, as was shown by indirect immunofluorescence (Fig. 4). Furthermore, VEGF protein could be isolated from conditioned medium of v-raf transformed cells (Fig. 5A). Characterization of the secreted VEGF unveiled a specific competition to 125 I-labeled VEGF 165 (Fig. 5B), and it was shown to be biologically active for HUVE cells (Fig. 5C).
Taken together, these results clearly demonstrate the essential role of the Ras/Raf signal transduction pathway in the regulation of VEGF expression in NIH 3T3 cells. This indicates that Ras as well as Raf are not only involved in cell proliferation and differentiation but also in tumor progression, mediated by increased expression of the tumor angiogenesis factor VEGF.
FIG. 5. Partial purification and characterization of VEGF protein from conditioned medium of v-raf transformed NIH 3T3 cells. A, cells were maintained for 72 h in medium supplemented with 0.05% serum. 100 ml of conditioned medium were run through a heparin-Sepharose column, and proteins were eluted with 1 M NaCl and fractionated in 1-ml aliquots. 15 l of fractions (1-6) were mixed with non-reducing sample buffer and run on a 10% SDS polyacrylamide gel. Western blot analysis was performed with a VEGF-specific polyclonal rabbit antiserum. L, load; FT, flow through; C, control recombinant VEGF 165 expressed in Sf9 cells; M, molecular weight marker. B, competition of VEGF binding. Confluent HUVE cells were incubated with 1.5 ng/ml 125 I-VEGF 165 and increasing amounts of either unlabeled recombinant VEGF 165 (open circles) or VEGF from conditioned medium from v-raf transformed NIH 3T3 cells (closed circles). As negative control, a 100-fold excess of unlabeled PDGF-BB (open triangle) or bFGF (closed triangle) was used. The experiment was done three times with essentially identical results. C, proliferation of HUVE cells. HUVE cells were seeded into 24-well plates and treated with either recombinant VEGF 165 (open circles) or VEGF protein from conditioned medium (closed circles). After 7 days, cells were counted with a coulter counter. The experiment was done three times with essentially identical results.
In B and C, the amount of purified VEGF was estimated by Western blot analysis.