Flt-1-mediated Down-regulation of Endothelial Cell Proliferation through Pertussis Toxin-sensitive G Proteins, βγ Subunits, Small GTPase CDC42, and Partly by Rac-1

Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) promotes its function primarily by activating two receptor tyrosine kinases, Flt-1 (VEGFR-1) and KDR (VEGFR-2). Recently, it has been shown that KDR is responsible for VPF/VEGF-stimulated endothelial cell (EC) proliferation and migration, whereas Flt-1 activation down-modulates KDR-mediated EC proliferation. Although KDR-mediated EC proliferation and migration have been extensively studied, much less is known about Flt-1-mediated antiproliferation. Here, we demonstrate that Flt-1-mediated antiproliferative activity can be blocked completely by the dominant negative mutant of CDC42 (CDC42-17N) and partially by a Rac1 dominant negative mutant (Rac1-17N) but is not affected by a RhoA dominant negative mutant (RhoA-19N). Both CDC42-17N and Rac1-17N increase the intracellular Ca2+ mobilization in response to VPF/VEGF but have no effect on KDR and MAPK phosphorylation. Using the chimeric-receptor EGLT in which the extracellular domain of epidermal growth factor receptor was fused to the transmembrane and intracellular domains of Flt-1, we also demonstrate that CDC42 and Rac1 are activated by EGLT. Previously, we showed that phosphatidylinositol 3-kinase is required for Flt-1-mediated antiproliferative activity, but phospholipase C is not required. As expected, CDC42 and Rac1 activation mediated by EGLT can be completely inhibited by PI3K inhibitors, wortmannin and LY294002, and the p85 dominant negative mutant but not by either the phospholipase C inhibitor, U73122, or an intracellular Ca2+chilator BAPTA/AM. Surprisingly, pertussis toxin and overexpression of the free Gβγ-specific sequestering minigene hβARK1(495) also inhibit EGLT-mediated CDC42 and Rac1 activation completely. Moreover, pertussis toxin treatment also increases the intracellular Ca2+ mobilization and inhibits the antiproliferation activity, thus suggesting that pertussis toxin-sensitive G proteins and the Gβγ subunits are involved in the signaling pathway of Flt-1 that down-regulates EC proliferation. Taken together, these results further expand our understanding of Flt-1-mediated antiproliferative activity in VPF/VEGF-stimulated endothelium.


Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) promotes its function primarily by activating two receptor tyrosine kinases, Flt-1 (VEGFR-1) and KDR (VEGFR-2). Recently, it has been shown that KDR is responsible for VPF/VEGF-stimulated endothelial cell (EC) proliferation and migration, whereas Flt-1 activation down-modulates KDR-mediated EC proliferation. Although KDR-mediated EC proliferation and migration have been extensively studied, much less is known about Flt-1-mediated antiproliferation. Here, we demonstrate that Flt-1-mediated antiproliferative activity can be blocked completely by the dominant negative mutant of CDC42 (CDC42-17N) and partially by a Rac1 dominant negative mutant (Rac1-17N) but is not affected by a RhoA dominant negative mutant (RhoA
. Both CDC42-17N and Rac1-17N increase the intracellular Ca 2؉ mobilization in response to VPF/VEGF but have no effect on KDR and MAPK phosphorylation. Using the chimeric-receptor EGLT in which the extracellular domain of epidermal growth factor receptor was fused to the transmembrane and intracellular domains of Flt-1, we also demonstrate that CDC42 and Rac1 are activated by EGLT. Previously, we showed that phosphatidylinositol 3-kinase is required for Flt-1-mediated antiproliferative activity, but phospholipase C is not required. As expected, CDC42 and Rac1 activation mediated by EGLT can be completely inhibited by PI3K inhibitors, wortmannin and LY294002, and the p85 dominant negative mutant but not by either the phospholipase C inhibitor, U73122, or an intracellular Ca 2؉ chilator BAPTA/AM. Surprisingly, pertussis toxin and overexpression of the free G␤␥-specific sequestering minigene h␤ARK1(495) also inhibit EGLT-mediated CDC42 and Rac1 activation completely. Moreover, pertussis toxin treatment also increases the intracellular Ca 2؉ mobilization and inhibits the antiproliferation activity, thus suggesting that pertussis toxinsensitive G proteins and the G␤␥ subunits are involved in the signaling pathway of Flt-1 that down-regulates EC proliferation. Taken together, these results further expand our understanding of Flt-1-mediated antiproliferative activity in VPF/VEGF-stimulated endothelium.
The Rho family of the small GTPase superfamily has been shown to play an important role in cell growth, migration, transformation, and gene expression (12). The Rho family includes Rho (RhoA, RhoB, RhoC, RhoE, and RhoG), Rac (Rac1, Rac2, Rac3, and RhoG), CDC42 (CDC42Hs, G25K, and TC10), Rnd (RhoE/Rnd3, Rnd1/Rho6, and Rnd2/Rh07), RhoD, and TTF (12). Among them, RhoA, Rac1, and CDC42 are the most extensively studied members of this family. RhoA primarily induces the formation of stress fibers, whereas Rac1 and CDC42 promote the formation of lamellipodia and filopodia, respectively, when they are expressed in cells (13,14). It was reported that VPF/VEGF induces actin-based mobility (15), suggesting that Rho family proteins might be involved in this response. However, there is no evidence of Rho family proteins playing a role in VPF/VEGF-induced cellular responses and whether VPF/VEGF activates these GTPases.
In this study, we examined whether Rho GTPases are involved in the antiproliferative function of Flt-1. We found that a blockade of endogenous functions of CDC42 and Rac1 but not RhoA by their respective dominant negative mutants (CDC42-17N, Rac1-17N, and RhoA-19N) increases VPF/VEGF-induced HUVEC proliferation, indicating that CDC42 and Rac1 are involved in Flt-1-antiproliferative effect. In addition, overexpression of CDC42-17N and Rac1-17N increases VPF/VEGFstimulated intracellular calcium release but not KDR and MAPK phosphorylation. Our results further demonstrate that Flt-1 stimulation activates CDC42 and Rac1, and that the activation of CDC42 and Rac1 is mediated through PI3K, free G␤␥ subunits, and pertussis toxin-sensitive G proteins. Furthermore, the pretreatment with pertussis toxin increases VPF/VEGF-induced calcium mobilization and proliferation. Moreover, our data also indicate that there are two pathways for CDC42 activation, Rac1-dependent and Rac1-independent. The inhibition of the Rac1-dependent pathway by Rac1-17N partially inhibits CDC42 activation, resulting in partial inhibition of antiproliferation, but the inhibition of CDC42 completely prevented antiproliferation in VPF/VEGF-stimulated HUVEC.

EXPERIMENTAL PROCEDURES
Materials-Recombinant VPF/VEGF was obtained from R&D Systems (Minneapolis, MN). EGM-MV BulletKit, trypsin-EDTA, and trypsin neutralization solution were obtained from Clonetics (San Diego, CA). Vitrogen 100 was purchased from Collagen Biomaterials (Palo Alto, CA). Mouse monoclonal antibodies against the KDR C-terminal domain and rabbit polyclonal antibodies against CDC42 and Rac1 were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). [ 3 H]Thymidine was obtained from PerkinElmer Life Sciences. Fura-2 AM and Pluronic F-127 were obtained from Molecular Probes (Eugene, Oregon). Pertussis toxin was obtained from Calbiochem.
Cell Culture-Primary HUVEC were obtained from Clonetics. Cells were grown on plates coated with 30 g/ml vitrogen in EGM-MV Bul-letKit (5% fetal bovine serum in endothelial basal medium with 12 g/ml bovine brain extract, 1 g/ml hydrocortisone, 1 l/ml GA-1000, and human EGF). HUVEC transduced with EGLT were grown in the same medium without human EGF. HUVEC (passage 3 or 4) that were ϳ80% confluent were used for most experiments. Cells were serumstarved in 0.1% fetal bovine serum in endothelial basal medium for 24 h prior to treatment.
Overexpression of Proteins in HUVEC-CDC42-17N, Rac1-17N, and RhoA-19N were kindly provided by Margaret M. Chou (University of Pennsylvania, Philadelphia, PA). The fragments encoding the genes were subcloned to a retroviral vector. The h␤ARK1(495) minigene was cloned by reverse transciption-PCR from HUVEC RNA and subcloned to a retrovirus vector (16). Retrovirus preparation and HUVEC infection with retrovirus were carried out as described previously (11,17).
Proliferation Assays-Assays were carried out as described previously (11,17). 2 ϫ 10 3 HUVEC/well (not infected or infected with the indicated retroviruses) were seeded in 24-well plates. After 2 days, cells were serum-starved (0.1% serum) for 24 h and then stimulated with 10 ng/ml VEGF for 20 h. For the experiment with pertussis toxin, pertussis toxin was added to a final concentration of 100 ng/ml when cells were changed to serum-starved conditions. 1 Ci/ml of [ 3 H]thymidine was added to each well, and 4 h later, cells were washed with ice-cold PBS three times, fixed with 100% cold methanol for 15 min at 4°C, precipitated with 10% cold trichloroacetic acid for 15 min at 4°C, washed with water three times, and lysed with 200 l of 0.1 N NaOH for 30 min at room temperature. [ 3 H]Thymidine incorporation was measured in scintillation solution. Data were expressed as the values for stimulated cell relative to the mean for its control group. Data are expressed as the mean Ϯ S.D. of triplicate values.
Immunoprecipitation and Western Blotting-48 h after infection, HUVEC were serum-starved for 24 h and stimulated with 10 ng/ml of VPF/VEGF for various times. Stimulation was halted by the addition of ice-cold PBS, and cells were washed three times with ice-cold PBS and lysed with cold radioimmune precipitation buffer (20 mM Tris-HCl, pH 7.5, 0.15 M NaCl, 1% Triton X-100, 1 mM PMSF, 1 mM Na 3 VO 4 , 1 mM EGTA, 1 g/ml leupeptin, 0.5% aprotinin, and 2 g/ml pepstatin A). Cell lysates were collected after centrifugation at 14,000 ϫ g for 15 min at 4°C. 1 mg of lysate protein was incubated with 1 g of antibody against KDR for 1 h, and 50 l of protein A-conjugated agarose-beads were incubated at 4°C for an additional hour. The beads were washed with radioimmune precipitation buffer three times, and immunoprecipitates were resuspended in 2ϫ SDS sample buffer for Western blot analysis. All experiments were repeated at least three times.
Intracellular Ca 2ϩ Release-Serum-starved HUVEC transfected with LacZ or CDC42-17N or pretreated with pertussis toxin for 16 h were loaded with Fura-2 AM and stimulated with 10 ng/ml of VPF/ VEGF. Assay was carried out as described previously (11,17). All experiments were repeated at least three times.
CDC42 and Rac1 Activation Assays-The CDC42 and Rac1 activity assay was modified from Ren et al. (18). The glutathione S-transferase (GST)-Pak-Rac binding domain (CRIB) fusion protein (19) was kindly provided by Dr. Rick Cerione (Cornell University, Ithaca, NY). Bacteria were grown to A 600 of 0.8 and induced with 1 mM isopropylthiogalactoside for 3 h. Bacteria were collected 50 ml/aliquot by centrifugation at 5000 rpm for 20 min and frozen at Ϫ80°C. To prepare the GST-Pak-CRIB beads, each aliquot of frozen bacteria was resuspended in 2 ml of cold PBS, and 20 l of dithiothreitol (1 M), 20 l of PMSF (0.2 M), and 40 l of lysozyme (50 mg/ml) were added and incubated on ice for 30 min. 225 l of 10% Triton X-100, 22.5 l of 1 M MgCl 2 , and 22.5 l of DNase I (2000 kilounits/ml) were added. After additional incubation for 30 min on ice, the bacterial lysates were centrifuged at 10,000 rpm. The supernatant was incubated with 200 l of glutathione-coupled Sepharose-4B beads (Amersham Biosciences, Inc.) that were washed three times with bead-washing buffer (PBS with 10 mM dithiothreitol and 1% Triton X-100). After incubation at 4°C for 45 min, beads were washed with bead-washing buffer for three times and resuspended in bead-washing buffer to give out 50% bead slur. GST-Pak-CRIB bound to Sepharose beads is approximately 1 g/l as detected by a protein assay kit (Bio-Rad Laboratories, Hercules, CA).
24-h serum-starved HUVEC with or without retroviral infection were stimulated with 10 ng/ml of VPF/VEGF or EGF at different time intervals. Stimulation was stopped by the addition of ice-cold PBS. Cells were washed with PBS three times and lysed with lysis buffer (150 mM NaCl, 0.8 mM MgCl 2 , 5 mM EGTA, 1% IGEPAL, 50 mM HEPES, pH 7.5, 1 mM PMSF, 10 g/ml leupeptin, and 10 g/ml aprotinin). Cell lysates were centrifuged at 14,000 rpm for 3 min. The supernatant was incubated with 50 l of GST-Pak-CRIB beads at 4°C for 45 min. Protein bound to beads were washed three times with AP wash buffer (50 mM Tris-HCl, pH 7.2, 1% Triton X-100, 150 mM NaCl, 10 mM MgCl 2 , 1 mM PMSF, 10 g/ml leupeptin, and 10 g/ml aprotinin) and analyzed by SDS-PAGE with antibodies against CDC42 or Rac1 as indicated. For inhibitor experiments, different concentrations of inhibitors as indicated with the exception of pertussis toxin were added 5 min before EGF treatment. Pertussis toxin was added 16 h before EGF treatment. All experiments were repeated at least three times.

Effect of the Dominant Negative Mutants of CDC42, Rac1
, and RhoA on VPF/VEGF-stimulated HUVEC Proliferation-To examine whether the Rho family proteins CDC42, Rac1, and RhoA are involved in VPF/VEGF-stimulated HU-VEC proliferation, we overexpressed the dominant negative mutants of CDC42 (CDC42-17N), Rac1 (Rac1-17N), or RhoA (RhoA-19N) using a recently developed retroviral gene-delivery system, which demonstrated an almost 100% infection yield in HUVEC to block the function of endogenous CDC42, Rac1, or RhoA (11,17). HUVEC transduced with LacZ as a control, CDC42-17N, Rac1-17N, or RhoA-19N were serum-starved and stimulated with 10 ng/ml of VPF/VEGF. Proliferation assay was measured by the incorporation of [ 3 H]thymidine. Data are expressed as -fold increase as compared with the non-stimulated control. Surprisingly, as shown in Fig. 1 (7)(8)(9)(10)(11). Therefore, we tested whether CDC42 and Rac1 play any role in the activation of these signaling molecules. Serumstarved HUVEC transduced with LacZ, CDC42-17N, and Rac1-17N were stimulated with VPF/VEGF for 0, 1, 5, and 10 min. Cellular extracts were immunoprecipitated with an antibody against KDR and immunoblotted with an antibody against phosphotyrosine. Fig. 2a clearly indicates that neither CDC42-17N nor Rac1 has any effect on KDR phosphorylation. Surprisingly, in a VPF/VEGF-stimulated intracellular Ca 2ϩ mobilization assay, the kinetic slope and the maximum response of intracellular Ca 2ϩ mobilization was increased in HUVEC transduced with CDC42-17N compared with that of those transduced with LacZ (Fig. 2b). The overexpression of Rac1-17N also increases the kinetic slope and maximum response of intracellular Ca 2ϩ mobilization, however, the effects of Rac1-17N are less than that of CDC42-17N (Fig. 2b). However, MAPK was phosphorylated at the same level in HUVEC transduced with LacZ, CDC42-17N, or Rac1 after treatment with VPF/VEGF for different time intervals (Fig. 2c).
Flt-1 Mediates Activation of both CDC42 and Rac1-Because the dominant negative mutants of CDC42 and Rac1 inhibit VPF/VEGF-stimulated HUVEC proliferation and intracellular Ca 2ϩ mobilization, we tested whether VPF/VEGF directly activates CDC42 and Rac1. The activity of CDC42 and Rac1 was measured using a pull-down assay of GST-Pak-CRIB fusion proteins that bound only to the GTP-bound forms of CDC42 and Rac1 (19). Serum-starved HUVEC were stimulated with VPF/VEGF for various times. Cellular extracts were incubated with freshly prepared GST-Pak-CRIB beads. Proteins bound to the beads were subjected to Western blot analysis using the antibodies against CDC42 or Rac1. Fig. 3, panel A, shows that CDC42 and Rac1 activity increases as early as 0.5 min and remains high at 1 min after VPF/VEGF treatment. To examine whether Flt-1 mediates VPF/VEGF-induced CDC42 and Rac1 activation, we used the recently developed receptor chimera, EGLT, in which the N-terminal domain of Flt-1 is replaced with that of epidermal growth factor receptor (11). It was previously shown that HUVEC were not responsive to EGF treatment in the experimental conditions used (11). As expected, in the HUVEC transduced with LacZ-expressing viruses, EGF did not activate CDC42 and Rac1 (Fig. 4, a and b,  top panel). When serum-starved HUVEC transduced with EGLT-expressing viruses were stimulated with EGF for different time intervals, CDC42 and Rac1 were activated in a similar time frame as that in HUVEC stimulated with VPF/VEGF (Fig.  4, a and b, bottom panel).
PI3K Mediates CDC42 and Rac1 Activation-To identify the  signaling molecules that mediate CDC42 and Rac1 activation, EGLT-expressing HUVEC were pretreated with an inhibitor of the phospholipase C family (U73122), PI3K inhibitors (wortmannin and LY294002), or BAPTA/AM, an intracellular Ca 2ϩ chelator, for 5 min and treated with EGF for 1 min. Cellular extracts were used to measure the activation of CDC42 and Rac1. The data show that wortmannin and LY294002 completely inhibited both CDC42 and Rac1 activation, but neither U73122 nor BAPTA/AM inhibited CDC42 and Rac1 activation in EGF-stimulated EGLT/HUVEC (Fig. 5a). To further confirm that PI3K is required for EGF-stimulated CDC42 and Rac1 activation in EGLT-transduced HUVEC, HUVEC were transduced with EGLT together with LacZ as control or a dominant negative mutant of PI3K, p85DN. Serum-starved cells were stimulated with EGF, and cellular extracts were subjected to CDC42 and Rac1 activation assay. As shown in Fig. 5b, p85DN completely inhibited EGF-stimulated EGLT-mediated CDC42 and Rac1 activation. These results indicate that PI3K is a key mediator of EGLT-mediated CDC42 and Rac1 activation.
Pertussis Toxin-sensitive G proteins and G␤␥ Subunits Mediate CDC42 and Rac1 Activation-It has been shown that pertussis toxin inhibited Flt-1-mediated microphage migration stimulated by VPF/VEGF (1) and suggested that pertussis toxin-sensitive G proteins may participate in Flt-1 signaling. Therefore, we examined whether pertussis toxin has any effect on CDC42 and Rac1 activation mediated by EGLT. Serumstarved HUVEC transduced with EGLT were pretreated with pertussis toxin overnight followed by stimulation with EGF for 1 min. Cellular extracts were subjected to CDC42 and Rac1 activation assay. As shown in Fig. 6, pertussis toxin completely inhibited the CDC42 and Rac1 activation mediated by EGLT. These data suggest that pertussis toxin-sensitive G proteins are involved in EGLT-mediated CDC42 and Rac1 activation. It is known that after activation heterotrimeric G proteins disso-ciate into ␣ and ␤␥ subunits, and the released ␤␥ subunits from pertussis toxin-sensitive G proteins can trigger several downstream signaling pathways (20,21). Therefore, to test whether  G␤␥ subunits are required for CDC42 and Rac1 activation, we overexpressed the h␤ARK1(495) peptide to sequester free G␤␥. h␤ARK1(495) corresponds to the C-terminal domain of human ␤ARK1 that physically interacts with free G␤␥ and therefore acts as a specific intracellular G␤␥ antagonist inhibiting G␤␥mediated downstream events (22,23). HUVEC were transduced with EGLT-expressing viruses together with LacZ-or h␤ARK1(495)-expressing viruses. The transduced cells were then stimulated with EGF for 1 min. Cellular extracts were subjected to a CDC42 or Rac1 activation assay. The data show that overexpression of h␤ARK1(495) completely inhibits the activation of CDC42 and Rac1 (Fig. 7), indicating that G␤␥ subunits are required for Flt-1-mediated CDC42 and Rac1 activation.
Rac1-17N Partially Inhibited CDC42 Activation-Currently, our data indicate that both CDC42 and Rac1 activation are regulated by pertussis toxin-sensitive G proteins, free G␤␥ subunits, and PI3K but not by phospholipase C or intracellular Ca 2ϩ . However, the effect of CDC42-17N on VPF/VEGF-stimulated HUVEC proliferation and intracellular Ca 2ϩ mobilization is almost twice that of Rac1-17N (Figs. 1 and 2b). Therefore, we examined whether there was any cross-talk between CDC42 and Rac1. EGLT/HUVEC transduced with LacZ or Rac1-17N was stimulated with EGF for 1 min. Cellular extracts were subjected to a CDC42 activation assay. As shown in Fig. 8a, Rac1-17N partially inhibits CDC42 activation. However, when EGLT/HUVEC transduced with CDC42-17N was stimulated with EGF and cellular extracts were subjected to Rac1 activation assay, CDC42-17N had no effect on Rac1 activation (Fig. 8b). These data indicate that Flt-1-stimulated CDC42 activation is mediated by Rac1-dependent as well as independent pathways.
Effect of Pertussis Toxin on VPF/VEGF-induced Intracellular Ca 2ϩ Mobilization and EC Proliferation-Next, we tested whether pertussis toxin has any effect on Flt-1-stimulated HU-VEC proliferation and intracellular calcium mobilization in response to VPF/VEGF. Serum-starved HUVEC were pretreated with 100 ng/ml of pertussis toxin for 16 h and stimulated with 10 ng/ml of VEGF/VEGF. The data show that pretreatment with pertussis toxin increases the kinetic slope and the maximum response of VEGF/VEGF-induced intracellular Ca 2ϩ mobilization (Fig. 9a). We then examined the effect of pertussis toxin on Flt-1-mediated down-regulation of HUVEC proliferation. Serum-starved HUVEC were pretreated with 100 ng/ml of pertussis toxin for 16 h and stimulated with 10 ng/ml of VPF/VEGF for 24 h. As expected, the pretreatment with pertussis toxin increases the rate of VPF/VEGF-induced HU-VEC proliferation (Fig. 9b) at the same rate as that in HUVECtransduced CDC42-17N (Fig. 1a) and p85DN (11). DISCUSSION It is known that KDR mediates cell proliferation and migration of EC in response to VPF/VEGF (7)(8)(9)(10)(11), whereas Flt-1 down-modulates KDR-mediated cell proliferation (10,11). Recently, with the chimeric receptors EGDR and EGLT, in which the extracellular domains were replaced with that of epidermal growth factor receptor, respectively, we found that EGLT cotransduction inhibited EGDR-stimulated HUVEC proliferation in response to EGF of up to 50% (11), and that both the inhibition of PI3K activity by wortmannin and p85 dominant mutant (11) and the inhibition of Flt-1 function with an antibody against Flt-1 2 increased the proliferation rate of approximately 50%. In this study, we show that VPF/VEGF-induced HUVEC proliferation is increased approximately 50 and 25% in CDC42-17N-and Rac1-17N-transduced cells, respectively. These data clearly indicate that CDC42-17N completely inhibits Flt-1-mediated HUVEC antiproliferative activity, but Rac1-17N partially inhibits this activity.
Whereas the downstream pathway of CDC42 and Rac1 regulation of actin rearrangement has been extensively studied (24 -26), much less is known about how CDC42 and Rac1 are activated (27). fMLP has been reported to activate CDC42 through a pertussis toxin-sensitive pathway (28) 8. Rac1-17N partially inhibited CDC42 activation, but CDC42-17N has no effect on Rac-1 activation. a, EGLT-transduced HUVEC were transduced with LacZ and Rac1-17N. Serum-starved cells were stimulated with EGF for 1 min. Cellular extracts were incubated with GST-Pak-CRIB beads. Proteins bound on GST-Pak-CRIB beads (panel A) and cellular extracts (panel B) were analyzed with an antibody against CDC42. b, EGLT-transduced HUVEC were transduced with LacZ and CDC42-17N. Serum-starved cells were stimulated with EGF for 1 min. Cellular extracts were incubated with GST-Pak-CRIB beads. Proteins bound on GST-Pak-CRIB beads (panel A) and cellular extracts (panel B) were analyzed with an antibody against Rac1. but not in others (33,34). Recently, Jimenez et al. (35) reported that platelet-derived growth factor-stimulated CDC42 activation is independent of PI3K enzymatic activity but is dependent on the p85 regulatory subunit of PI3K. In this study, with the results of a pull-down assay we show that VPF/VEGF activates CDC42 and Rac1 in HUVEC, and that this activation is mediated by EGLT (i.e. Flt-1). Furthermore, CDC42 and Rac1 activation through EGLT (Flt-1) can be inhibited by PI3K inhibitors (wortmannin and LY294002) and p85DN, but not by U73122 nor by intracellular calcium chilator BAPTA/AM. Taken together, our findings demonstrate that Flt-1-stimulated PI3K is upstream of CDC42 and Rac1 activation that modulates KDR-stimulated intracellular calcium mobilization and cell proliferation.
It was reported that pertussis toxin inhibited VPF/VEGFstimulated microphage migration (1). Because microphages express Flt-1 and not KDR, it was suggested that the Gi family proteins might play a role in Flt-1-mediated signaling (1). Our result that pretreatment with pertussis toxin inhibits EGLTmediated CDC42 and Rac1 activation further confirms the involvement of Gi family proteins in Flt-1 signaling. Pretreatment with pertussis toxin also increased the kinetic slope and maximum response of VPF/VEGF-stimulated intracellular Ca 2ϩ mobilization in a similar way to the effect of overexpression of dominant negative mutants of CDC42 or PI3K (11) in HUVEC. The results indicate that Flt-1-induced PI3K, CDC42, and Rac1 activation is via pertussis toxin-sensitive G proteins. It is known that many responses mediated by pertussis toxinsensitive G proteins are through the released G␤␥ (20,21). Our results show that G␤␥-sequestering peptide h␤ARK (495) completely inhibits Flt-1-induced CDC42 and Rac1 activation, indicating that G␤␥ is the upstream mediator of Flt-1 signaling to CDC42 and Rac1 activation. It is known that G␤␥ can activate PI3K␥ (36 -39), but it is not known whether G␤␥ can also activate PI3K␣ and/or PI3K␤, which use p85 as their regulatory subunits. Another possibility is that the p85 dominant negative mutant (p85DN) has some effect on PI3K␥.
Our results also indicate that overexpression of Rac1-17N inhibits CDC42 activation by ϳ50%, but CDC42-17N has no effect on Rac1 activation. These findings indicate that the antiproliferative effect of Rac1 is mediated through its effect on CDC42 activation. This also correlates with the effect of Rac1-17N and CDC42-17N in VPF/VEGF-stimulated HUVEC proliferation and intracellular Ca 2ϩ mobilization. Meanwhile, these data suggest that VPF/VEGF-induced CDC42 activation can be mediated through Rac1-dependent as well as independent pathways. Our results have demonstrated that both CDC42 and Rac1 activation in response to VPF/VEGF requires pertussis toxin-sensitive G proteins, free G␤␥ subunits, and PI3K activation, indicating that VPF/VEGF-induced Rac1 and CDC42 activation is mediated by a common upstream signaling event. Although it has been shown that PI3K binds to Rac1 and CDC42 (27), it is not clear how PI3K regulates Rac1 and CDC42 activity and how Rac1 triggers CDC42 activation.
In summary, the current study demonstrates that CDC42 and Rac1 rather than RhoA mediate the antiproliferative effect of Flt-1. The effect of CDC42 and Rac1 is through reducing VPF/VEGF-induced intracellular calcium mobilization but not KDR and MAPK phosphorylation. Our results further demonstrate that Flt-1 stimulation activates CDC42 and Rac1, and that the activation of CDC42 and Rac1 is mediated through PI3K, free G␤␥ subunits, and pertussis toxin-sensitive G proteins. Furthermore, pretreatment with pertussis toxin increases VPF/VEGF-induced calcium mobilization and proliferation. Together, our results extend the pathway that mediates Flt-1 down-modulated VPF/VEGF-induced HUVEC proliferation. The findings in this study significantly contribute to our understanding of how Flt-1 functions as an antiproliferative modulator during VPF/VEGF-induced vasculogenesis and angiogenesis.