Flt-1, but not Flk-1 mediates hyperpermeability through activation of the PI3-K/Akt pathway

ORIGINAL ARTICLE 236J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology
Flt-1, But Not Flk-1 Mediates
Hyperpermeability ThroughActivation of thePI3-K/Akt Pathway
CARSTEN VOGEL,2 ALEXANDER BAUER,3 MARION WIESNET,2 KLAUS T. PREISSNER,4
WOLFGANG SCHAPER,3 HUGO H. MARTI,3 AND SILVIA FISCHER1,4*1Department of Anesthesiology and Intensive Care, Kerckhoff-Clinic, Bad Nauheim, Germany2Max-Planck Institute for Heart and Lung Research, Bad Nauheim, Germany3Institute of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany4Institute of Biochemistry, University of Giessen, Giessen, Germany

Vascular endothelial growth factor (VEGF), a potentmediator of endothelial proliferation andmigration, has an important role also in brainedema formation during hypoxia and ischemia. VEGF binds to the tyrosine kinase receptors Flt-1 and Flk-1. Yet, their relative importancefor hypoxia-induced hyperpermeability is not well understood.We used an in vitro blood-brain barrier (BBB)model consisting of porcinebrain microvascular endothelial cells (BMEC) to determine the role of Flt-1 in VEGF-induced endothelial cell (EC) barrier dysfunction.Soluble Flt-1 abolished hypoxia/VEGF-induced hyperpermeability. Furthermore, selective antisense oligonucleotides to Flt-1, but not toFlk-1, inhibited hypoxia-induced permeability changes. Consistent with these data, addition of the receptor-specific homolog placenta-derived growth factor, which binds Flt-1 but not Flk-1, increased endothelial permeability to the same extent as VEGF, whereas addingVEGF-E, a viral VEGF molecule from the orf virus family activating Flk-1 and neuropilin-1, but not Flt-1, did not show any effect. Using thecarcinoma submandibular gland cell line (CSG), only expressing Flt-1, it was demonstrated that activation of Flt-1 is sufficient to inducehyperpermeability by hypoxia and VEGF. Hyperpermeability, induced by hypoxia/VEGF, depends on activation of phosphatidylinositol3-kinase/Akt (PI3-K/Akt), nitric oxide synthase (NOS) and protein kinase G (PKG). The activation of the PI3-K/Akt pathway by hypoxiawas confirmed using an in vivo mice hypoxia model. These results demonstrate that hypoxia/VEGF-induced hyperpermeability can bemediated by activation of Flt-1 independently on the presence of Flk-1 and indicate a central role for activation of the PI3-K/Akt pathway,followed by induction of NOS and PKG activity.J. Cell. Physiol. 212: 236–243, 2007. � 2007 Wiley-Liss, Inc.

H.H. Marti and S. Fischer contributed equally to this study.

*Correspondence to: Silvia Fischer, Justus-Liebig Universitat,Biochemisches Institut, Friedrichstr. 24, 35385 Giessen, Germany.E-mail: [email protected]

Received 23 November 2006; Accepted 15 December 2006

DOI: 10.1002/jcp.21022

Vascular endothelial growth factor (VEGF) stimulatesendothelial cell (EC) growth and migration in vitro (Ferrara andHenzel, 1989; Senger et al., 1996) and angiogenesis in vivo(Connolly et al., 1989; Dvorak et al., 1995). VEGFwas originallydescribed as a potent vascular permeability factor (VPF)responsible for the accumulation of plasma protein-rich fluid inthe ascites of tumor patients (Senger et al., 1983).VEGF is inducible by hypoxia/ischemia in vitro and in vivo andwas suggested to be a likely candidate for the development ofischemia- and tumor-induced vasogenic brain edema(Baethmann, 1978; Goldman et al., 1997; Schoch et al., 2002).We recently confirmed this hypothesis using an in vitro modelof the blood-brain barrier (BBB) consisting of porcine brainderived microvascular ECs (BMEC) by demonstrating thathypoxia-induced hyperpermeability is mediated by the VEGF/VEGF receptor (VEGFR) system in an autocrine manner viasimultaneous activation of the phospholipase C (PLC) andPI3-K/Akt pathways followed by the release of nitric oxide andactivation of the protein kinase G (PKG) (Fischer et al., 1999).However, it remains to be elucidated through which VEGFRpermeability changes are mediated.VEGF exerts its multiple actions by binding to two tyrosinekinase receptors, VEGFR-1 or fms-like tyrosine kinase Flt-1(Flt-1) and VEGFR-2 or fetal liver kinase receptor Flk-1 (Flk-1)(De Vries et al., 1992; Quinn et al., 1993; Shibuya, 2001) and toneuropilin-1, a class 3 semaphorin receptor, which binds VEGFin an isoform-specific manner (Becker et al., 2005). Manystudies have shown that Flt-1 and Flk-1 play a leading role inVEGF-induced angiogenesis. Stimulation of either receptor

� 2 0 0 7 W I L E Y - L I S S , I N C .

triggers its own phosphorylation, followed by an intracellularcalcium elevation (Waltenberger et al., 1994). Most studiessuggest that the single activation of Flk-1 alone is sufficient toinduce the vast majority of VEGF effects (Park et al., 1994;Waltenberger et al., 1994; Seetharam et al., 1995; Gilbert et al.,1998). However, since the murine knockout of either Flt-1 orFlk-1 is lethal (Fong et al., 1995; Shalabi et al., 1995) and sinceVEGFRs can possibly form heterodimers (Waltenberger et al.,1994), it remains unclear whether the contribution or thepresence of both receptors is required independently orsimultaneously in different VEGF-related events. Furthermore,there is still significant uncertainty concerning the role ofindividual VEGFRs for special biological functions such asvascular permeability. It has been shown that Flk-1 isresponsible for EC proliferation and migration (Millauer et al.,1993). In contrast, activation of Flt-1 induces various signalingpathways in vitro (De Falco et al., 2002) and furthermore, the

F L T - 1 M E D I A T E D H Y P E R P E R M E A B I L I T Y 237

maintenance and survival function of VEGF has been associatedwith Flt-1, as Flt-1 mRNA is highly expressed in quiescentendothelium (Peters et al., 1993). Furthermore, there isclear evidence that Flt-1 can function as a negative regulatorof Flk-1 (Hiratsuka et al., 1998) because Flt-1 activationdown-modulates Flk-1-mediated EC proliferation, which ismediated by activation of the PI3-K/Akt pathway(Zeng et al., 2002).We therefore investigated the selective role of Flt-1 and Flk-1for hypoxia-induced permeability changes via VEGF in BMECsand analyzed the contribution of Flt-1 using selective antisenseoligonucleotides. Furthermore, Flt-1-mediated permeabilitychangeswere confirmed in an epithelial cell line, only expressingFlt-1, but not Flk-1. Finally, the underlying signaling eventsfollowing activation of Flt-1 in this epithelial cell line wereexamined.

Materials and MethodsCell culture

Capillary brain-derived endothelial cells (BMEC) were isolated andcultured as described. Epithelial cells from porcine choroid plexus(CPE) were isolated and cultured as described (Crook et al., 1981).CSG 120/7 cells (Carcinoma submandibular gland cells) and MDCKcells (Madin-Darby canine kidney epithelial cells) were cultured inDMEM supplemented with 10% (v/v) FCS, 200 U/ml penicillin, and200 U/ml streptomycin. Cells were used when reaching confluency.The plates were washed once with PBS and incubated for the indicatedtime in M199 medium containing the indicated agents (VEGF, PlGF-1,polyclonal anti human VEGF antibody [all purchased from PeproTech;London, UK], soluble Flt-1 receptor (sFlt) [R & D; Wiesbaden,Germany], human VEGF R2 (KDR)/Fc chimera [Relia Tech;Braunschweig, Germany], VEGF-E [friendly supplied by M. Clauss;Indianapolis], U73122 [Biotrend Chemikalien; Koln, Germany],bisindolylmaleimide (BIM), ODQ, wortmannin (WM), LY294002,PD98059, and SB203580 [all purchased from Calbiochem; Darmstadt,Germany]).

In vitro hypoxia model

To induce hypoxia (2% oxygen), confluent monolayers of BMEC, CPE,CSG, or MDCK were incubated at 378C for the indicated time bygassing the chamber with 93% N2, 2% O2, and 5% CO2. Controlcultures were incubated under normoxic conditions for the samelength of time.

Assay of endothelial and epithelial monolayer permeability

BMEC, CPE, CSG, or MDCK cells were seeded and cultured toconfluency on polycarbonate membranes. Permeability across the cellmonolayer was measured and calculated as described previously(Fischer et al., 2005).

Antisense Oligonucleotide treatment

BMEC or CSG monolayers were treated 48 h before the start of theexperiment with antisense oligonucleotides (MWG-Biotech;Ebersberg, Germany) containing sequences complementary to bovineFlt-1 andmouse Flk-1mRNA (genBank accession numbersX94263 and70842). The sequences of the oligonucleotides were for the Flt-1antisense (AS): 50-CAAAGATGGACTCGGGAG-30 and for the Flk-1AS:GGATACCTAGCGCAAAGA-30. Two scrambled (scr) sequenceswere used as negative controls: scr Flt-1, scr-Flt-1: 50-AGCTAGGCACGAGAGTGA-30 and scr Flk-1, scr-Flk-1:50-AGACAATAAGGGCATCCG-30 . The uptake of theoligonucleotides was confirmed by using Cy3-50 end modifiedoligonucleotides and fluorescence microscopy.

Extraction of poly (AR)-RNA

Poly(Aþ)-RNA was isolated directly from cultured cell monolayersusing Dynabeads1 oligo(dT)25 (Dynal Biotech, Oslo, Norway).Extraction was performed according to manufacturer’s instructions.

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

Reverse transcriptase (RT)-PCR analysisof Flt-1- and Flk-1 receptor

Total RNA (500 ng) was reverse-transcribed at 258C for 10 min and at428C for 50 min using the Superscript II reverse transcriptase(Invitrogen, Groningen, Netherlands). The reaction was stopped byheat inactivation. For amplification the Taq PCR Core-Kit (Promega,Mannheim, Germany) was used. The following primers were used:sense 50-CTGCCACTCTCATTGTTAACG-30 and antisense 50-GACAGTTTCAGGTCCTCTCC-30 for Flt-1 (PCR product length:470 bp) and for Flk-1, sense 50-GTTTGCAAGAACTTGGATGCT-30

and antisenese 50-CCTGGCAGGTGTAGAGGCC-30 (PCR productlength: 410 bp). PCR (35 cycles) was performed using the followingamplification protocol: 948C for 1 min, 558C for 1 min, and 728C for1 min. 10 ml of the PCR products were electrophoresed on a 1.3%agarose gel containing ethidium bromide and photographed under UVlight. The PCR products of both receptors obtained were cloned intopGEM-T vector (Promega) according to manufacturer’s instructions.DNAs were analyzed by sequencing using the Pharmacia DNASequencer and both DNAs showed more than 95% homology to thecorresponding VEGFR.

Northern blot analysis

PolyAþ mRNA samples were electrophoresed on a 1%-agarose gelcontaining 2.2 M formaldehyde and transferred to a hybond-Nmembrane. Hybridizations were performed using cDNA probes ofFlt-1 and Flk-1 generated as described before according to standardprocedures. To control the loading of polyAþ RNA the blots wererehybridized with a b-actin cDNA probe.

Western blotting

Western blotting was performed according to standard protocolsusing pAkt or Akt antibodies (New England Biolab; Frankfurt,Germany; dilution 1:1000), or Flt-1 or Flk-1 antibodies (Santa Cruz,Heidelberg, Germany; dilution 1:1000) followed by incubation withHRP-conjugated goat-anti rabbit IgG (1:3500). Detection wasperformed using ECL reagents according to manufacturer’sinstructions (Amersham, Buchler, Germany).

Determination of vascular permeability in vivo

All experiments were performed with male NMRI mice (25–35 g) andwere approved by the governmental animal care authorities. WM wasdissolved in DMSO (25 mg/ml), diluted in aqua ad iniectabilia Braun(1:133) and injected (1.5 mg/kg) ip (intraperitoneal). Controls wereperformed with the appropriate amount of aqua-DMSO. Mice wereexposed to normobaric hypoxia (8% oxygen) for 24 h or were kept atroom air as described (Schoch et al., 2002). Following hypoxicexposure, the animals were killed by decapitation and the dissectedbrains were frozen in liquid nitrogen. Permeability of brain vessels wasquantified using sodium fluorescein as described (Schoch et al., 2002).Results are presented as relative fluorescence units (RFU) per mg ofbrain tissue.

Statistical analysis

Results were expressed as the mean� SEM. The unpaired Students t-test or analysis of variance and subsequent multiple comparisons usingDunn’s method were used for statistical analysis. Results wereconsidered as statistically different at P< 0.05.

ResultsFlt-1 receptor only mediates hypoxia-inducedhyperpermeability in BMECs

In earlier studies we demonstrated that hypoxia-inducedpermeability changes were mediated by VEGF since hypoxia-induced permeabililty changeswere completely abolished in thepresence of anti VEGF antibody in vitro (Fischer et al., 1999) andin vivo (Schoch et al., 2002). To investigate which VEGFR isinvolved in hypoxia- and VEGF-induced permeability changes,soluble Flt-1 (sFlt-1) was added to BMEC. As in the case of theneutralizing VEGF antibody, sFlt-1 prevented hypoxia-inducedpermeability changes completely (Fig. 1A). Furthermore, PlGF,which only binds to Flt-1, increased the permeability to the

Fig. 1. Activation of Flt-1 increased the permeability across theBMEC monolayer. A: Neutralizing antibodies to VEGF and solublesFlt-1 abolished hypoxia-induced hyperpermeability. BMECmonolayers were cultured either under normoxia (filled) or hypoxia(hatched) for 3 h in the absence and presence of a neutralizingantibody to VEGF (anti VEGF-Ab, 10 mg/ml) or sFlt-1 (100 ng/ml).B: VEGF or PlGF, but not VEGF-E increased the permeability acrossthe BMEC monolayer. BMEC monolayers were treated for 3 h withVEGF (10 ng/ml), PlGF (10 ng/ml), or VEGF-E (10 ng/ml), all incombination with the antioxidans a-lipoic acid (2 mmol/L). Theamount of [3H]inulin that passed across the BMEC monolayer after3 h at normoxia in the absence of any added compound was set to100%. Values are expressed as the meansWSEM (nU 10, MP<0.05).

Fig. 2. Antisense oligonucleotides against Flt-1 abolishedhypoxia-induced hyperpermeability in BMEC. A: BMEC monolayerswere preincubated for 2 days with the antisense oligonucleotidesfollowed by culturing either under normoxia (filled) or hypoxia(hatched) for 3 h in the absence and presence of antisenseoligonucleotides to Flt-1 (Flt-1-AS), Flk-1 (Flk-1 AS) or nonsenseoligonucleotides to both receptors (scr Flt-1, scr Flk-1). Theconcentration of all oligonucleotides was 2 mmol/L. The amount of[3H]inulin that passed across the BMEC monolayer after 3 h atnormoxia in the absence of any added compound was set to 100%.Values are expressed as the meansWSEM (nU 10, MP<0.05).B:Western blot analysis for Flt-1 or Flk-1 of cell lysates prepared fromBMECs after preincubation for 2 dayswith antisense oligonucleotidesagainst Flt-1 or Flk-1.

238 V O G E L E T A L .

same extent as VEGF whereas VEGF-E, only activating Flk-1,had no effect on permeability (Fig. 1B).These results suggest that permeability changes in BMEC aremediated by activation of the Flt-1 receptor only. To confirmthese results, selective AS oligonucleotides against bothreceptors were incubated with cells 2 days prior to the start ofthe experiment. The Flt-1-AS inhibited hypoxia-inducedpermeability changes whereas blocking Flk-1 expression wasineffective (Fig. 2A). The corresponding scrambledoligonucleotides did not show any effect.Western blot analysisconfirmed that after preincubation of the cells with each of theantisense oligonucleotides, the protein expression wasdecreased (Fig. 2B). Preincubation with the scrambledoligonucleotides did not change the protein expression of thecorresponding receptor (data not shown).

VEGFRs are differentially expressed in variousepithelial cells

We next aimed to explore permeability changes mediated byactivation and signaling of Flt-1 in the absence of Flk-1 in moredetail. As no ECs, expressing only one of the VEGFRs areavailable, primary choroid plexus epithelial cells (CPE) and thetwo epithelial cell lines, MDCK and CSG, were analyzed. UsingPCR, we demonstrate that BMEC and CPE express both, Flt-1and Flk-1, CSG Flt-1 only, and MDCK cells none of the VEGFRs(Fig. 3A). The PCR products had the appropriate size (470 bp


for Flt-1 and 410 bp for Flk-1) and their identity was verified bysequence analysis. To confirm these PCR results by Northernblot analysis the cloned cDNA fragments were used as probes.Analysis of poly Aþ mRNA isolated from BMECs, CPE, CSG,and MDCK cells confirmed the PCR results. While BMEC andCPE expressed Flt-1 and Flk-1 mRNA (7.5 kb, 5.5 kb), CSGexpressed Flt-1 only, andMDCK none of the VEGFRs (Fig. 3B).

Presence of Flt-1 is sufficient to mediate hypoxia- andVEGF-induced permeability changes

In agreementwith these results, permeability ofMDCKwas notincreased by exposure to either hypoxia, VEGF, or PlGF,whereas CPE and CSG cells showed a significant increase inpermeability by all three conditions (Fig. 4A). Hypoxia-inducedpermeability changes of CPE and CSG cells were abolished inthe presence of a neutralizing antibody to VEGF to the sameextent as in BMEC (Fig. 4A) suggesting that hypoxia-inducedpermeability changes of these epithelial cells are also mediatedby VEGF. Taken together, these results further demonstratethe presence of Flt-1 alone to be sufficient to mediateVEGF- and hypoxia-induced hyperpermeability.To further confirm these results, hypoxia-induced permeabilitychanges of the CSG cell monolayer were investigated in thepresence of the AS oligonucleotides against Flt-1 and Flk-1. TheASoligonucleotide against Flt-1 only abolished hypoxia-inducedpermeability changes, whereas the Flk-1 AS oligonucleotide

Fig. 3. CSG cells expressed only Flt-1. A: Reverse transcription-polymerase chain reaction (RT-PCR). Total RNA from CPE, CSG,MDCK, and BMEC cells was isolated and used for PCR analysis usingspecific primers for the Flt-1 and Flk-1 receptor. B: Northern blotanalysis from poly AR RNA isolated from BMEC, CSG, MDCK, andCPE. cDNAprobes of Flt-1 and Flk-1 were used as probes. Same blotswere stripped and hybridized with a-actin cDNA probe for control ofequal loading.

Fig. 4. Permeability of CPE and CSG, but not of MDCK cells wasincreased by hypoxia, VEGF, or PlGF. A: CPE (filled), CSG (hatched),or MDCK (empty) monolayers were cultured either under normoxia(N)orhypoxia (H) for 3h in theabsenceandpresenceof aneutralizingantibody to VEGF (VEGF-Ab, 10mg/ml), ormonolayers were treatedfor 3 h with VEGF (10 ng/ml) or PlGF (10 ng/ml), both in combinationwith the antioxidans a lipoic acid (2 mmol/L). B: Hypoxia-inducedhyperpermeability in CSG cells was decreased by specific antisenseoligonucleotides against Flt-1, but not against Flk-1. CSGmonolayerswere cultured either under normoxia (filled) orhypoxia (hatched)for 3 h in the absence and presence of antisense oligonucleotides toFlt-1 (Flt-1-AS), Flk-1 (Flk-1AS) or nonsense oligonucleotides to bothreceptors (scr Flt-1, scr Flk-1). The concentration of alloligonucleotides was 2mmol/L. The amount of [3H]inulin that passedacross the BMEC monolayer after 3 h at normoxia in the absence ofany added compound was set to 100%. Values are expressed as themeansWSEM (nU 10, MP<0.05).


showed no effect. Again, using scrambled controloligonucleotides no change in hypoxia-inducedhyperpermeability was noticed (Fig. 4B).

Hypoxia-induced permeability changes of CSG wereprevented by inhibition of the PI3-K/Akt pathway, theNO synthase and cyclic guanylate cyclase

To investigate signal transduction pathways induced by Flt-1activation, CSG cells were treated with various signalinginhibitors.Wepreviously demonstrated activation of the PI3-K/Akt and PLC pathways in hypoxia- and VEGF-inducedpermeability changes of BMEC, expressing both, Flt-1 and Flk-1(Fischer et al., 2004). We thus first investigated these pathwaysin hypoxia-induced permeability changes of CSG cells. Asshown in Fig. 5A, hypoxia-induced permeability changes areinhibited by LY294002 and WM, which are both inhibitors ofthe PI3-K/Akt pathway, while U73122, a blocker of PLCactivation, was without effect.Next, downstream events of the PI3-K/Akt pathway wereanalyzed by investigating the activation of nitric oxide synthase(NOS) and cyclic guanylate cyclase (cGC). Hypoxia-inducedpermeability changes in CSG cells were abolished by NMMA, ablocker of NOS, and by ODQ, an inhibitor of cGC (Fig. 5B).BIM, blocking activation of PKC, did not show any effect onpermeability, neither during normoxia nor hypoxia.Furthermore, inhibition of the p38 and p44/p42 MAP kinases,using their specific inhibitors SB203580 and PD98059, did notprevent hypoxia-induced permeability changes (Fig. 5B).


To further confirm the activation of the PI3-K/Akt pathway,phosphorylation of Akt in response to VEGF, PlGF, and hypoxiawas determined. The phosphorylation of Akt was increased in atime-dependent manner up to 3 h. Whereas VEGF and PlGFincreased the phosphorylation of Akt about twofold, hypoxiaresulted in a nearly fourfold increase (Fig. 6A,B). Longertreatments attenuated the increase of phosphorylated Akt(data not shown). In agreement to the permeabilitymeasurements, hypoxia-induced Akt phosphorylation wasdecreased significantly in CSG cells by adding neutralizing VEGFantibodies (Fig. 6C,D).

Hypoxia-induced permeability is mediated bythe PI3-K/Akt pathway in vivo

Having demonstrated that inhibition of PI3-K/Akt abolishesFlt-1-mediated hypoxia-induced hyperpermeability in vitro, we

Fig. 5. Hypoxia-induced hyperpermeability of CSG cells wasabolished after blockage of the PI3-K/Akt pathway. A: CSGmonolayers were incubated for 3 h during normoxia (filled) andhypoxia (hatched) for 3 h in the absence and presence of LY294002(50 mmol/L), WM (100 nmol/L), and U73122 (2.5 mmol/L). B: CSGmonolayers were incubated for 3 h during normoxia (filled) andhypoxia (hatched) for 3 h in the absence and presence of BIM(2.5 mmol/L), NMMA (50 mmol/L), ODQ (50 mmol/L), SB2035580(10 mmol/L), and PD90059 (20 mmol/L). The amount of [3H]inulinthat passed across the CSG monolayer after 24 h of normoxia in theabsence of any added compound was set to 100%. Values areexpressed as the meansWSEM (nU 10, MP<0.05).

240 V O G E L E T A L .

tested whether this effect could also be observed in vivo.Unfortunately, no specific blocker of Flt-1 activation for in vivouse was available to us. Nevertheless, we could analyze theinvolvement of the PI3-K/Akt pathway during systemic hypoxiain vivo. In animals submitted to systemic hypoxia, VEGFproduction is continuously increased in the brain when oxygenavailability is progressively decreased, leading subsequently tobrain edema formation (Schoch et al., 2002). Pretreatment ofanimals with the PI3-K/Akt inhibitor WM completely inhibitedthe increase in vascular permeability under hypoxic conditionswhereas it has no influence under normoxic settings (Fig. 7).

Discussion

Increasing experimental evidence supports a significant role forVEGF in the development of brain edema during hypoxia(Schoch et al., 2002) and ischemia (Zhang et al., 2000).However, the molecular mechanisms whereby VEGF inducesvascular permeability are notwell understood. In particular, thecell-surface receptors transducing the permeability signal arenot well characterized. VEGF binds to Flt-1, Flk-1, and


neuropilin-1 which are all expressed on ECs as well as othercell types (Ferrara et al., 2003). Both Flk-1 (Joukov et al.,1998; Murohara et al., 1998; Gille et al., 2001) and neuropilin-1(Becker et al., 2005) have been suggested to mediatevascular permeability.Here, we demonstrate that hypoxia-induced permeabilitychanges in BMEC are mediated by VEGF-induced activation ofFlt-1, based on the following findings: (i) soluble Flt-1, known toinhibit VEGF-induced migration and proliferation by forminginactive complexes with VEGF (Kendall and Thomas, 1993;Roeckl et al., 1998) abolished hypoxia/VEGF-inducedhyperpermeability, (ii) selective antisense oligonucleotides toFlt-1, but not to Flk-1 inhibited hypoxia/VEGF-inducedpermeability changes, and (iii) adding the receptor-specifichomolog PlGF-1, which binds Flt-1 but neither neuropilin-1 norFlk-1 (Park et al., 1994), increased the permeability to the sameextent as VEGF, whereas the addition of VEGF-E, a viral VEGFhomolog from the orf virus family activating Flk-1 andneuropilin-1, but not Flt-1 (Ogawa et al., 1998; Wise et al.,1999), did not show any effect. As from the three differentisoforms of PlGF generated by mRNA alternative splicing (Caoet al., 1997), only PlGF-2 binds to neuropilin-1, known to act as aco-receptor for VEGF165 activation of Flk-1 (Migdal et al., 1998;Mamluk et al., 2002), activation of both neuropilin-1 and Flk-1can be excluded.Although, Flt-1 was the first receptor tyrosine kinase to beidentified as a VEGFR (De Vries et al., 1992), the precisefunction of this molecule is still under debate. Initially, it wasproposed that Flt-1 might not be primarily a receptortransmitting a mitogenic signal, but rather a ‘‘decoy’’ receptor,able to regulate in a negative fashion the activity of VEGF on thevascular endothelium, by preventing VEGF binding to Flk-1(Park et al., 1994). In the meantime, however, several studiesdemonstrated activation and signaling mediated via Flt-1. Flt-1,expressed in porcine aortic ECs, was able to transduce signalsfor increased DNA synthesis and proliferation and to increasetyrosine phosphorylation of PLCg in response to VEGF (Itoet al., 2001). Accordingly, it was demonstrated that a mutantform of VEGF that lacks Flk-1 activity retains the ability toinduce vascular permeability (Stacker et al., 1999). Our findingthat hypoxia-induced hyperpermeability in BMEC is mediatedby activation of Flt-1 is in accordance with findingsdemonstrating that hypoxia upregulates the expression of Flt-1(Gerber et al., 1997; Fischer et al., 1999) and also the binding ofVEGF to Flt-1 (Fischer et al., 1999).It has been reported that an intra- and intermolecular cross talkbetween the VEGFRs Flt-1 and Flk-1 occurs. Activation of Flt-1by PlGF resulted in the intermolecular transphosphorylation ofFlk-1, thereby amplifying VEGF-driven angiogenesis throughFlk-1 (Autiero et al., 2003). To elucidate whether the presenceof Flk-1 is necessary for Flt-1 mediated permeability changes,we aimed to analyze cells that express only one of the VEGFRs.As no such untransfected ECs are available, we used epithelialcells instead, since the expression of VEGFRs has been detectedin several types of non-ECs (Kimet al., 1999;Ohno-Matsui et al.,2003). Furthermore, previous studies indicated that VEGF andFlt-1 were expressed in tumor epithelial cells and that theymight play an important role in tumor metastasis (Ghanemet al., 2003). Therefore, we analyzed several epithelial cell linesby RT-PCR and Northern blot with respect to their VEGFRexpression pattern. We found that CSG cells only expressedFlt-1, while MDCK cells expressed none of the VEGFRs.Primary epithelial choroid plexus cell (CPE), expressing Flt-1and Flk-1, as shown before (Nico et al., 2004), were used ascontrol. In agreement with the results using BMEC, thepermeability of the CPE and CSG monolayer was increased byhypoxia, VEGF, and PlGF-1, whereas the permeability of theMDCK monolayer remained unchanged during the sametreatments. Since CSG cells do not express the Flk-1 receptor,

Fig. 6. VEGF-, PlGF, andhypoxia-inducedactivationpAkt.A:CSGwere treated for60, 90, and180minwithVEGF (10ng/ml, filled), PlGF (10ng/ml, hatched), and hypoxia (empty). Total cell lysates were prepared and analyzed byWestern blot analysis for pAkt and Akt expression usingspecificphospho-AktandAktantibodies.Thefigureshowsonetypicalblotoutof four.B:QuantificationofAktphosphorylation.TheexpressionsofpAkt andAkt at each time point and treatment were divided by each other. Results are expressed corresponding to the value determined in theabsenceof any added compound.Values are expressed as themeansWSEM(nU 4).C:Hypoxia-inducedphosphorylation ofAktwasdecreasedbyneutralizing antibodies to VEGF. CSG were incubated for 180 min under normoxia and hypoxia in the absence and presence of neutralizingantibodiestoVEGF(VEGF-Ab,10mg/ml).Totalcell lysateswerepreparedandanalyzedbyWesternblotanalysis forpAktandAktexpressionusingphospho-Akt and specificAkt antibodies.D:QuantificationofAktphosphorylation.The rationof pAkt andAkt determined in the absenceof anti-VEGF-Ab under normoxia was set to 100 %. Values are expressed as the meansWSEM (nU 4, MP<0.05).


permeability changes induced by VEGF and PlGF are related toactivation of Flt-1. This suggestionwas further confirmedbyourresult that specific antisense oligonucleotides to Flt-1 but not toFlk-1 prevented hypoxia/VEGF-induced permeability changesof the CSG monolayer. Thus, our results strongly indicate that


the presence of Flk-1 is not necessary to mediate permeabilitychanges by activation of Flt-1. Flt-1 can be activated by eitherVEGF or PlGF. PlGF can thus play an important role in vascularpermeability. For example, it was demonstrated that miceoverexpressing PlGF exhibit increased vascularization and

Fig. 7. PI3-K/Akt inhibition abrogates vascular permeability in vivo.Adult mice were intraperitoneally injected with 45 mg Wortmannin(WT) or vehicle as control (C) and were exposed to normobarichypoxia at 8% oxygen for 24 h (H) or were kept at room air (N). Toquantify vascular permeability of brain vessels, 200 ml sodium-fluorescein (6mg/ml inPBS)was injected through the tail vein and thefluorescence of the brain parenchyma per mg of brain tissue (inrelative fluorescent units, RFU) was quantified. PI3-K/Akt inhibitioncompletely blocked the hypoxia-induced increase in vascularpermeability, although it had no effect under normoxic conditions.Values are meanWSD (nU 8–12) MP<0.05.

242 V O G E L E T A L .

vessel permeability (Odorisio et al., 2002). This effect might berelated to binding and activation of its own receptor, Flt-1, or,alternatively, PlGFmight also amplify the effect of VEGF throughupregulation of Flk-1 (Odorisio et al., 2002).The analysis of hypoxia/VEGF-induced downstream signalingevents in vitro and by using an in vivo mice hypoxia modelrevealed the activation of the PI3-K/Akt pathway as inhibition ofthis pathway by wortmannin reduced brain edema formation.Results were in accordance with in vitro studies demonstratingPI3-K/Akt activation (Suzuma et al., 2000) and in vivo studiesconfirming that Akt signaling mediates VEGF vascularpermeability (Six et al., 2002). Upon receptor activation, Akt isknown to be recruited to the plasma membrane and to bind toinositol lipids via its pleckstrin homology domain. Akt is thenphosphorylated by phosphoinositide-dependent kinases, andthis phosphorylation enhances its catalytic activity toward avariety of diverse substrates (Downward, 1998). Akt is adownstream target of VEGF as neutralizing antibodies to VEGFdecreased but not abolished hypoxia-induced Aktphosphorylation. Our results suggest that other hypoxia-inducible agents, such as IGF-1, might additionally induce Aktphosphorylation (Wang et al., 2004; Brywe et al., 2005). Themain function of PI3-K/Akt activation is promotion of survival inECs (Fujio and Walsh, 1999), which has been found to beassociated with activation of Flt-1 (Peters et al., 1993) but alsoFlk-1 (Gerber et al., 1998). Studies demonstrating activation ofFlk-1 were performed using human umbilical vein ECs, whereasthe PI3-K/Akt pathway was induced via Flt-1 activation inprimary cultures of microvascular bovine retinal ECs (Cai et al.,


2003). Thus, VEGF activation of Flt-1 or Flk-1 seems to dependon cell-specific differences. Results of this study indicate thatpermeability changes induced by hypoxia/VEGF can bemediated by induction of the PI3-K/Akt pathway via activationof Flt-1. As BMEC express both VEGFRs, Flt-1 and Flk-1, wecannot rule out completely that also activation of Flk-1might beinvolved in hypoxia/VEGF-induced permeability changes ofBMECs. Nevertheless, this study clearly demonstrates thatpermeability changes of the used tumor cell line (CSG) aremediated via activation of Flt-1 followed by activation of thePI3-K/Akt pathway independently on the presenceof Flk-1. Thispathway might be important in edema formation associatedwith tumor growth, since several reports demonstrate that Flt-1 is present and functional on different human cancer cells andthat activation of Flt-1 by VEGF can activate processes involvedin tumor progression andmetastasis (Hiratsuka et al., 2002; Fanet al., 2005; Wey et al., 2005). Recently, it was shown that Aktcan phosphorylate endothelial nitric-oxide synthase (eNOS)resulting in eNOS activation and NO production (Dimmeleret al., 1999; Fulton et al., 1999; Michell et al., 2002), which is inagreement with our results demonstrating the activation ofNOS and PKG but not of PKC andMAP kinases p38 and p42/44in CSG as well as in BMEC (Fischer et al., 2004). The finding thatVEGF did not activate the p38 MAP kinase is in agreement withresults, demonstrating that PI3-kinase/Akt signaling promotesEC survival by inhibiting p38 MAP kinase-dependent apoptosis(Gratton et al., 2001).Conclusively, our data indicate that permeability changes can bemediated by Flt-1 engagement leading to the activation of thePI3-K/Akt and NOS. These results will be helpful to developtherapeutical strategies to prevent hypoxia/VEGF-inducedhyperpermeability leading to vasogenic brain edema formation,without interfering with the beneficial angiogenic andneuroprotective effects of VEGF.

Acknowledgments

We thank I. Keller for technical assistance and G. Froelich forartwork.

Literature Cited

Autiero M, Waltenberger J, Communi D, Kranz A, Moons L, Lambrechts D, Kroll J,Plaisance S, De Mol M, Bono F, Kliche S, Fellbrich G, Ballmer-Hofer K, Maglione D,Mayr-Bearle U, Dewerchin M, Dombrowski S, Stanimirovic D, van Hummelen P, Dehio C,Hicklin DJ, Persico G, Herbert J-M, Communi D, Shibuya M, Collen D, Conway EM,Carmeliet P. 2003. Role of PlGF in the intra- and intermolecular cross talk between theVEGF receptors Flt1 and Flk1. Nat Med 9:936–943.

Baethmann A. 1978. Pathophysiological and pathochemical aspects of cerebral edema.Neurosurg Rev 1:85–100.

Becker PM, Waltenberger J, Yachechko R, Mirzapoiazova T, Sham JSK, Lee CG, Elias JA,Verin AD. 2005. Neuropilin-1 regulates vascular endothelial growth factor-mediatedendothelial permeability. Circ Res 96:1257–1265.

BryweKG,MallardC,GustavssonM,HedtjarnM, LeverinAL,WangX, BlomgrenK, Isgaard J,Hagberg H. 2005. IGF-1 neuroprotection in the immature brain after hypoxia-ischemia,involvement of Akt and GSK3beta? Eur J Neurosci 21:1489–1502.

Cai J, Ahmad S, Jiang WG, Huang J, Kontos CD, Boulton M, Ahmed A. 2003. Activationof Vascular Endothelial Growth Factor Receptor-1 sustains angiogenesis and Bcl-2Expression via the phosphatidylinositol-3-kinase pathway in endothelial cells. Diabetes52:2959–2968.

Cao Y, Ji WR, Qi P, Rosin A. 1997. Placenta Growth Factor: Identification andcharacterization of a novel isoform generated by RNA alternative splicing. BiochemBiophys Res Commun. 235:493–498.

Connolly DT, Heuvelman DM, Nelson R, Olander JV, Eppley BL, Delfino JJ, Siegel NR,Leimgruber RM, Feder J. 1989. Tumor vascular permeability factor stimulates endothelialcell growth and angiogenesis. J Clin Invest 84:1470–1478.

Crook RB, Kasagami H, Prusiner SB. 1981. Culture and characterization of epithelial cellsfrom bovine choroid plexus. J Neurochem 37:845–854.

De Falco S, Gigante B, Persico G. 2002. Structure and function of placental growth factor.Trends Cardiovasc Med 12:241–246.

De Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N, Williams LT. 1992. Thefms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science255:989–991.

Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM. 1999. Activation ofnitric oxide synthase in endothelial cells by Akt-dependent phosporylation. Nature399:601–605.

Downward J. 1998.Mechanisms and consequencesof activationof protein kinaseB/Akt.CurrOpin Cell Biol 10:262–267.


Dvorak HF, Brown LF, Detmar M, Dvorak AM. 1995. Vascular permeability factor/vascularendothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol146:1029–1039.

Fan F, Wey JS, McCarty MF, Belcheva A, Liu W, Bauer TW, Womico RJ, Wu Y, Hooper A,Hicklin DJ. 2005. Oncogene. 24:2647–2653.

Ferrara N, Henzel WJ. 1989. Pituitary follicular cells secrete a novel Heparin-bindinggrowth factor specific for vascular endothelial cells. Biochem Biophys Res Commun161:851–858.

Ferrara N, Gerber H-P, LeLouter J, 2003. The biology of VEGF and its receptor. Nature Med9:669–676.

Fischer S, Clauss M, Wiesnet M, Renz D, Schaper W, Karliczek GF. 1999. Hypoxia inducespermeability in brain microvessel endothelial cells via VEGF and NO. Am J Physiol276:C812–C820.

Fischer S, Wiesnet M, Marti HH, Renz D, Schaper W. 2004. Simultaneous activation ofseveral second messengers in hypoxia-induced hyperpermeability in brain derivedendothelial cells. J Cell Physiol 198:359–369.

Fischer S, Wiesnet M, Renz D, Schaper W. 2005. H2O2 induces paracellular permeabilityof porcine brain-drived microvascular endothelial cells by activation of the p44/42MAP kinase pathway. Eur J Cell Biol 84:687–697.

Fong G-H, Rossant J, Gertsenstein M, Breitman M. 1995. Role of the Flt-1 receptor tyrosinekinase in regulating the assembly of vascular endothelium. Nature 376:66–70.

Fujio Y, Walsh K. 1999. Akt mediates cytoprotection of endothelial cells by vascularendothelial growth factor in an anchorage-dependent manner. J Biol Chem 274:16349–16354.

Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF, Pappetropoulos A,Sessa WC. 1999. Regulation of endothelium-derived nitric oxide production by proteinkinase Akt. Nature 399:597–601.

Gerber HP, Condorelli F, Park J, Ferrara N. 1997. Differential transcriptional regulationof the 2 vascular endothelial growth-factor receptor genes—flt-1, but not flk-1/kdr, isup-regulated by hypoxia. J Biol Chem 272:23659–23667.

Gerber HP, McMurtrey A, Kowalski J, Yan MH, Keyt BA, Dixit V, Ferrara N. 1998. Vascularendothelial growth-factor regulates endothelial-cell survival through thephosphatidylinositol 3’-kinase akt signal-transductionpathway—requirement for flk-1/kdractivation. J Biol Chem 273:30336–30343.

GhanemMA, van SteenbruggeGJ, SudaryoMK, Mathoera RB, Nijman JM, van der Kwast TH.2003. Expression and prognostic relevance of vascular endothelial growth factor (VEGF)and its receptor (flt-1) in nephroblastoma. J Clinical Pathol 56:107–113.

Gilbert RE, Vranes D, Berka JL, Kelly DJ, Cox A, Wu LL, Stacker SA, Cooper ME. 1998.Vascular endothelial growth-factor and its receptors in control and diabetic rat eyes. LabInvest 78:1017–1027.

Gille H, Kowalski J, Li B, LeCouter J, Moffat B, Zioncheck TF, Pelletier N, Ferrara N.2001. Analysis of biological effects and signaling properties of Flt-1 (VEGFR-1) and KDR(VEGFR-2)—A reassessment using novel receptor-specific vascular endothelial growthfactor mutants. J Biol Chem 276:3222–3230.

Goldman CK, Bharara S, Palmer CA, Vitek J, Tsai JC, Weiss HL, Gillespie GY. 1997. Brainedema in meningiomas is associated with increased vascular endothelial growth-factorexpression. Neurosurg 40:1269–1277.

Gratton J-P, Morales-Ruiz M, Kureishi Y, Fulton D, Walsh K, Sessa WC. 2001. AktDown-regulation of p38 signaling provides a novel mechanism of vascular endothelialgrowth factor-mediated cytoprotection in endothelial cells. J Biol Chem 276:30359–30365.

Hiratsuka S,NakamuraK, Iwai S,MarakamiM, ItohT,KijimaH, Shipley JM, SeniorRM, ShibuyaM. 2002. MMP9 induction by vascular endothelial growth factor receptor-1 is involved inlung-specific metastasis. Cancer Cell 2:289–300.

Hiratsuka S, Minova O, Kuno J, Noda T, Shibuya M. 1998. Flt-1 lacking the tyrosinekinase domain is sufficient for normal development and angiogenesis in mice. Proc NatlAcad Sci USA 4:9349–9354.

Ito N, Huang K, Claesson-Welsh L. 2001. Signal transduction by VEGF receptor-1 wild typeand mutant proteins. Cell Signal 13:849–854.

Joukov V, Kumar V, Sorsa T, Arighi E, Weich H, Saksela O, Alitalo K. 1998. Arecombinantmutant vascular endothelial growth factor-c that has lost vascular endothelialgrowth-factor receptor-2 binding, activation, and vascular-permeability activities. J BiolChem 273:6599–6602.

Kendall R, Thomas K. 1993. Inhibition of vascular endothelial growth factor activity by anendogenously encoded soluble receptor. Proc Natl Acad Sci USA 90:10705–10709.

Kim I, ryan AM, Rohan R, Amano S, Agular S, Miller JW, Adamis AP. 1999. Constitutiveexpression of VEGF, VEGFR-1, and VEGFR-2 in normal eyes. Invest Ophthalmol Vis Sci40:2115–2121.

Mamluk R, Gechtman Z, Kutcher ME, gasiunas N, Gallagher J, Klagsbrun M. 2002.Neuropilin-1 binds vascular endothelial growth factor 165, placenta growth factor-2, andheparin via its b1b2 domain. J Biol Chem 277:24818–24825.

Michell BJ, Harris MB, Chen Z-P, Ju H, Venema VJ, Blackstone MA, Huang W, Venema RC,Kemp BE. 2002. Identification of regulatory sites of phosphorylation of the bovineendothelial nitric-oxide synthase at serine 617 and serine 635. J Biol Chem 277:42344–42351.

Migdal M, Huppertz B, Tessler S, Comforti A, Shibuya M, Reich R, Baumann H. 1998.Neuropilin-1 is a placenta growth factor-2 receptor. J Biol Chem 28:22272–22278.


Millauer B, Wizigmann-Voos S, Schnurch H, Martinez R, Moller NP, Risau W, Ullrich A.1993. High affinity VEGF binding and developmental expression suggest Flk-1 as a majorregulator of vasculogenesis and angiogenesis. Cell 72:835–846.

Murohara T, Horowitz JR, Silver M, Tsurumi Y, Chen DF, Sullivan A, Isner JM. 1998. Vascularendothelial growth-factor vascular-permeability factor enhances vascular-permeability vianitric-oxide and prostacyclin. Circulation 97:99–107.

Nico B, Mangieri D, Corsi P, De Giorgis M, Vacca A, Roncali L, Ribatti D. 2004. Vascularendothelial growth factor-A, vascular endothelial growth factor receptor-2 andangiopoietin-2 expression in the mouse choroid plexuses. Brain Res 1013:256–259.

Odorisio T, Schietroma C, Zaccaria ML, Cianfarani F, Tiveron C, Tatangelo L, Failla CM,Zambruno G. 2002. Mice overexpressing placenta growth factor exhibit increasedvascularization and vessel permeability. J Cell Sci 115:2559–2567.

Ogawa S, OkuA, SawanoA, Yamanuchi S, Yazaki Y, ShibuyaM. 1998. A novel type of vascularendothelial growth factor, VEGF-E (NZ-7 VEGF), preferentially utilizes KDR/Flk-1receptor and carries a potentmitotic activity without heparin-binding domain. J Biol Chem273:31273–31282.

Ohno-Matsui K, Yoshida T, Uetama T, Mochizuki M, Morita J. 2003. Vascular endothelialgrowth factor upregulates pigment epithelium-derived factor expression via VEGFR-1 inhuman retinal pigment epithelial cells. Biochem Biophys Res Commun 303:962–967.

Park J, Chen H, Winer J, Houck K, Ferrara N. 1994. Placenta growth factor: Potentiationof vascular enodothelial growth factor bioactivity, in vitro and in vivo, and high affinitybinding to Flt-1 but not to Flk-1/KDR. J Biol Chem 269:25646–25654.

Peters KG, De Vries C, Williams LT. 1993. Vascular endothelial growth factor receptorexpression during embryogenesis and tissue repair suggest a role in endothelialdifferentiation and blood vessel growth. Proc Nat Acad Sci 90:8915–8919.

Quinn TP, Peters KG, De Vries C, Ferrara N, Williams LT. 1993. Fetal liver kinase 1 is areceptor for vascular endothelial growth factor and is selectively expressed in vascularendothelium. Proc Natl Acad Sci USA 90:7533–7537.

Roeckl W, Hecht D, Sztajer H, Waltenberger J, Yayon A, Weich HA. 1998. Differentialbinding characteristics and cellular-inhibition by soluble vegf receptor-1 and receptor-2.Exp Cell Res 241:161–170.

Schoch HJ, Fischer S, Marti HH. 2002. Hypoxia-induced VEGF expression causes vascularleakage in the brain. Brain 125:2549–2557.

Seetharam L, Gotoh N, Maru Y, Neufeld G, Yamaguchi S, Shibuya M. 1995. A uniquesignal transduction from FLT tyrosine kinase, a receptor for vascular endothelial growthfactor VEGF. Oncogene 10:135–147.

Senger DR, Dvorak AM, Peruzzi CA, Harvey VS, Dvorak HF. 1983. Tumor cells secrete avascular permeability factor that promotes accumulation of ascites fluid. Science 219:983–985.

Senger DR, Ledbetter SR, Claffey KP, Papadopoulos-Sergiou A, Perruzzi CA, Detmar M.1996. Stimulation of endothelial cell migration by vascular permeability factor/vascularendothelial growth factor through cooperative mechanisms involving avb3 integrin,osteopontin, and thrombin. Am J Pathol 149:293–305.

Shalabi F, rossant J, Yamaguchi TP, Gertenstein M, Wu XF, Breitman ML, Schuh AC. 1995.Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature376:62–66.

Shibuya M. 2001. Structure and function of VEGF/VEGF-receptor system involved inangiogenesis. Cell Struct Funct 26:25–35.

Six I, Kureishi Y, Zhengyu L, Walsh K. 2002. Akt signaling mediates VEGF/VPF vascularpermeability in vivo. FEBS Letters 532:67–69.

Stacker SA, Vitali A, CaesarC, Domagala T, Groenen LC,Nice E, AchenMG,Wilks AF. 1999.A mutant form of vascular endothelial growth factor (VEGF) that lacks VEGF receptor-2activation retains the ability to induce vascular permeability. J Biol Chem 274:34884–34892.

Suzuma K, Naruse K, Suzuma I, Takahara N, Ueki K, Aiello LP, King GL. 2000. Vascularendothelial growth factor induces expression of connective tissue growth factor via KDR,Flt1, and phosphatidylinositot 3-kinase-Akt-dependent pathways in retinal vascular cells.J Biol Chem 275:40725–40731.

Waltenberger J, Claesson-Welsh L, SiegbahnA, ShibuyaM,HeldinC-H. 1994.Different signaltransduction properties of KDR and Flt1, two receptors for vascular endothelial growthfactor. J Biol Chem 269:26988–26995.

Wang X, Deng J, Boyle DW, Zhong J, Lee WH. 2004. Potential role of IGF-1 in hypoxiatolerance using a rat hypoxic-ischemic model: Activation of hypoxia-inducible factor 1alpha. Pediatr Res 55:385–394.

Wey JS, Fan F, Gray MJ, Bauer TW, McCarty MF, Somicio R, LiuW, Evans DB,Wu Y, HicklinDJ. 2005. Vascular endothelial growth factor receptor-1 promotes migration and invasionin pancreatic carcinoma cell lines. Cancer 104:427–438.

Wise LM, Veikkola T, Mercer AA, Savory LJ, Fleming SB, Caesar C, Vitali A, Makinen T,Alitalo K, Stacker SA. 1999. Vascular endothelial growth factor (VEGF)-like proteinfrom orf virus NZ2 binds to VEGFR2 and neuropilin-1. Proc Natl Acad Sci USA96:3071–3076.

Zeng HY, Zhao DZ, Mukhopadhyay D. 2002. Flt-1-mediated down-regulation of endothelialcell proliferation through pertussis toxin-sensitive G proteins, beta gamma subunits, smallGTPase CDC42, and partly by Rac-1. J Biol Chem 277:4003–4009.

Zhang ZG, Zhang L, Jiang Q, Zhang RL, Davies K, Powers C, van Bruggen N, ChoppM. 2000.VEGF enhances angiogenesis and promotes blood-brain barrier leakage in the ischemicbrain. J Clin Invest 106:829–838.

Documents

Flt-1, but not Flk-1 mediates hyperpermeability through activation of the PI3-K/Akt pathway