March 14, 2005 17:24 801 TFJF041-12-56085

Endothelium, 12:1–7, 2005Copyright c© Taylor & Francis Inc.ISSN: 1062-3329 print / 1029-2373 onlineDOI: 10.1080/10623320590933888

Antiangiogenic Effect of Soluble Vascular EndothelialGrowth Factor Receptor-1 in Placental Angiogenesis

Shakil Ahmad and Asif AhmedDepartment of Reproductive and Vascular Biology, The Medical School, University of Birmingham,Edgbaston, Birmingham, West Midlands, United Kingdom5

Differential splicing of the flt-1 mRNA generates soluble variantof vascular endothelial growth factor (VEGF) receptor-1 (sVEGFR-1/sFlt-1). The action of VEGF is antagonized by sVEGFR-1. SolubleVEGFR-1 binds to VEGF with a high affinity and therefore worksto modulate VEGF and VEGF signaling pathway. In this study,the authors tested the hypothesis that VEGF-mediated endothelialcell angiogenesis is tightly modulated by the release of sVEGFR-1and placental expression of sVEGFR-1 is up-regulated by hypoxia.Immunolocalization studies showed progressively intense stainingfor sVEGFR-1 and VEGF in the trophoblast of placental villousexplants throughout gestation. Endothelial cell migration studiesusing a modified Boyden’s chamber showed a significant increasein cell migration in response to VEGF that was significantly at-tenuated in the presence of exogenous sVEGFR-1. Furthermore,stimulation of endothelial cells with VEGF led to a dose-dependentincrease in the release of sVEGFR-1 as determined by enzyme-linked immunosorbent assay (ELISA). Exposure of normal pla-cental villous explants to hypoxia (1% pO2) increased trophoblastexpression of sVEGFR-1 when compared with tissue normoxia (5%pO2). In addition, conditioned media from hypoxia treated placen-tal villous explants induced a significant increase in endothelial cellmigration that was significantly reduced in presence of sVEGFR-1.In this study the authors tested the hypothesis that VEGF-mediatedangiogenesis is tightly modulated by the release of sVEGFR-1 andits protein is strongly expressed in the trophoblast, which is up-regulated by hypoxia.

10

15

20

25

30

Keywords Angiogenesis, Hypoxia, Preeclampsia, Soluble VascularEndothelial Growth Factor Receptor, Vascular EndothelialGrowth Factor

35

Successful placentation involves the development of a low-impedance uteroplacental circulation following trophoblast in-

Q1 Received ; accepted .This work was supported by grants from the Wellcome Trust

053347/B/98/Z and British Heart Foundation Programme GrantRG/98003. The authors would like to express their appeciation to theclinical and laboratory staff of Birmingham Women’s Health Care NHSTrust for help with tissue collection.

Address correspondence to Asif Ahmed, Department of Reproduc-tive and Vascular Biology, The Medical School, University of Birm-ingham, Edgbaston, Birmingham, West Midlands, B15 2TG, UnitedKingdom. E-mail: a.s.ahmed@bham.ac.uk

vasion and transformation of the maternal intramyometrial por- 40tion of the spiral arterioles (Benirschke and Kaufmann. 1995). Inpreeclampsia, there is insufficient adaptation of the decidual andintramyometrial portions of the spiral arterioles (Frusca et al.1989), resulting in reduced uteroplacental blood flow (Lunellet al. 1982) and leading to local hypoxia in the placenta. 45Vascular endothelial growth factor (VEGF) is up-regulated byhypoxia and is a potent vascular protective and angiogenic factorin the placenta (Ahmed et al. 1995). VEGF mediates its signalvia two tyrosine kinase receptors, VEGF receptor-1 (VEGFR-1/Flt-1) and VEGFR-2 (KDR/Flk-1). VEGFR-1 can also be ex- 50pressed as a soluble protein and is generated by alternativesplicing of the fms-like tyrosine kinase (flt-1) gene (Kendalland Thomas 1993). Soluble VEGFR-1 (sVEGFR-1) has strongantagonistic activity and neutralises the effects mediated byVEGF and placenta growth factor (PlGF) (Kendall et al. 551996).

Recent studies suggest that sera and amniotic fluids frompregnant women contain a high-affinity VEGF-binding protein.In preeclampsia, levels of the VEGF-binding protein are higherin the amniotic fluid (Vuorela et al. 2000). In addition, maternal 60serum levels of VEGF are reported to be increased in preeclamp-sia and this correlates with the severity of the disease (Hunteret al. 2000). Soluble VEGFR-1 mRNA is expressed at very highlevels in the placenta and is produced by both villous and ex-travillous trophoblast (Clark et al. 1998). Recently, an in vivo 65animal model demonstrated that overexpression of sVEGFR-1leads to hypertension, proteinuria, and glomerular endothelio-sis, conditions that are similar to preeclampsia, which is char-acterized by an elevated blood pressure, proteinuria, and edema(Friedman et al. 1991; Maynard et al. 2003). In preeclampsia 70the microvilli are unduly thin and abnormally shaped (Jonesand Fox 1980), the defective development of placental bloodvessels clearly suggests serious impairment of angiogenesis.Indeed, a recent study from our laboratory demonstrated thatelevated sVEGFR-1 in preeclampsia may be responsible for 75the poor placental angiogenesis in this disorder and removalof sVEGFR-1 restores the ability of endothelial to form cap-illary networks (Ahmad and Ahmed 2004). In this study wehave tested the hypothesis that VEGF-mediated angiogenesis is

1

March 14, 2005 17:24 801 TFJF041-12-56085

2 S. AHMAD AND A. AHMED

tightly modulated by the release of sVEGFR-1 and its protein is80strongly expressed in the trophoblast, which is up-regulated byhypoxia.

MATERIALS AND METHODS

ReagentsRecombinant VEGF165 and sVEGFR-1 ectodomain85

(sVEGFR-1) were purchased from (RELIATech, Braunschweig,Germany). Polycarbonate filters (8-µm pore size, polyvinylpy-rrolidone-free) were obtained from Receptor Technologies(Adderbury, Oxford, UK). All other cell culture reagents andchemicals were obtained from Sigma Chemical (Poole, Dorset,90UK) unless otherwise stated.

Cell CultureHuman umbilical vein endothelial cells (HUVECs) were iso-

lated, characterized, and cultured as previously described(Bussolati et al. 2001). Experiments were performed on second-95or third-passage HUVECs. All cell types were grown at 37◦C,5% CO2 in a humidified incubator and routinely passaged when90% confluent.

Tissue CollectionHuman placental tissue were collected following surgical ter-100

mination of pregnancy during the first and second trimesters,and immediately following nonlabouring caesarean section atterm. Samples were analyzed from uncomplicated, ‘normal’first-trimester, second-trimester, and term pregnancies.Full-thickness biopsies were taken from a central, macroscopi-105cally normal portion of placenta, stripped of overlying mem-branes, and thoroughly washed in phosphate-buffered saline(PBS) as described previously (Ahmed et al. 1995). Informedconsent was obtained from the patients and the study had the ap-proval of the South Birmingham Ethical Committee110(Birmingham, UK).

Villous and Cell Culture in Increasing Oxygen TensionAfter dissection, three pieces of normal villous fragments

were equilibrated overnight in phenol-free Dulbecco’s modi-fied Eagle’s medium (DMEM) containing 0.1% bovine serum115albumin (BSA) in 24-well plates before transfer to Modular In-cubator chambers (ICN, Thame, Oxfordshire, UK) as previouslydescribed (Khaliq et al. 1999; Perkins et al. 2002). Under normalphysiological conditions the oxygen tension within the intervil-lous space at term is approximately 50 to 60 mm Hg (Soothill120et al. 1986; Fujikura and Yoshida 1996). In brief, the cham-bers were purged with N2 gas mixtures comprising of either 1%(hypoxia: pO2 in medium = 16 mm Hg) or 5% (normoxia: pO2

in medium = 60 mm Hg) supplemented with 5% CO2 for 5 min,then sealed for the duration of the experiment. After 24 h, the125

placental explants were fixed overnight in 10% buffered forma-lin and embedded in paraffin wax.

Enzyme-Linked Immunosorbent Assay (ELISA)Sandwich ELISA for the detection of total (free and VEGF

complexed) soluble VEGFR-1 was performed as according to 130manufacturer’s (RELIA Tech, Braunschweig, Germany) spec-ifications. Aliquots of supernatants from HUVECs and nor-mal placental villous explants treated with growth factors orcontrolmedia were assayed in triplicate.

Immunocytochemistry 135

Serial 3-µm sections of formalin-fixed, paraffin-embeddedtissue from oxygen-treated normal villous explant culture andnormal term, preeclamptic, and fetal growth–restricted placentawere used for immunohistochemistry as previously described(Li et al. 1998). Anti-VEGF (Santa Cruz, CA) (1:250) and anti- 140sVEGFR-1 (R&D Systems, Oxon, UK) (1:200) antibodies wereused. Nonimmune goat serum (10% in PBS) was used as adiluent of the primary antibody (1:100) to reduce nonspecificstaining. Amplification of the primary antibody reaction wasachieved using a goat anti-rabbit secondary antibody (diluted 1451:100) in PBS for 30 min followed by a complex of streptavidin(DAKO, Bucks, UK) and biotinylated peroxidase (DAKO) foran additional 30 min. Finally, the binding was visualised by theaddition of diaminobenzidine and 0.01% hydrogen peroxide inPBS to form the insoluble antigen-antibody complex. Between 150each step the sections were washed in PBS for 5 min. All in-cubations of antiserum were performed in at room temperaturein a wet chamber. Sections were counterstained with Mayer’shaematoxylin, dehydrated, and mounted.

Cell Migration Assay 155

Chemotaxis of HUVECs was performed in a modifiedBoyden’s chamber. Polycarbonate filters (8-µm pore size) werecoated with 0.1% gelatine for 8 h at room temperature. Medium199 containing 0.5% fetal calf serum (FCS) and 10 to 50 ng/mLof VEGF165 or the vehicle alone was placed in the lower cham- 160ber. One hundred microliter of HUVEC-treated supernatant wasadded to the lower chamber. Migration was carried out at 37◦Cfor 6 h. For inhibition studies VEGF165 or conditioned mediawere preincubated with sVEGFR-1 (100 ng/mL) for 30 min. Atthe end of the experiment the upper surface of the filter was 165scraped with a rubber policeman. The filters were then fixedand stained with Diff-Quik (Harleco, Gibbstown, USA), and 10fields at ×200 magnification were counted.

Statistical AnalysisAll data are expressed as mean ±SEM. Statistical comparison 170

was performed using one-way ANOVA followed by the Student-Newman-Keuls test as appropriate. Statistical significance wasset at a value of p < .05.

March 14, 2005 17:24 801 TFJF041-12-56085

ANTIANGIOGENIC EFFECT OF sVEGFR-1 3

FIG. 1. Immunolocalization of VEGF and sVEGFR-1 throughout gestation. Serial sections were incubated with antibodies raised against VEGF and sVEGFR-1.(B, E , and H ) Immunoreactivity for VEGF protein was detected in the syncytiotrophoblast of first trimester (1T), second trimester (2T), and term placenta.(C , F , and I ) Weak to progressively intense staining for sVEGFR-1 was detected in the syncytiotrophoblast and stroma of first and second trimester and termplacenta. (A, D, and G) Negative control with omission of primary antibody. Results are representative of five separate experiments performed in triplicate. Originalmagnification: ×200.

RESULTS

Immunolocalization of VEGF and sVEGFR-1throughout Gestation175

In the first and second trimester placental villi, VEGF dis-played weak immunoreactivity in association with the syncy-tiotrophoblast (Figure 1B and E). Isolated cells within the fi-Q2

brous stroma also showed positive VEGF staining. WeaksVEGFR-1 immunoreactivity was observed in the trophoblast180bilayer of the first trimester placental villi (Figure 1C). Placentafrom second-trimester and term pregnancies showed strong lo-calized staining for sVEGFR-1 in the syncytiotrophoblast(Figure 1F and I ). Strong VEGF immunostaining was observedin the syncytiotrophoblast and in endothelial cells lining the fetal185capillaries of term placental villi (Figure 1H ). No immunostain-ing was detected in negative control, with a nonimmune antibody(Figure 1A, D, and G).

Soluble VEGFR-1 Inhibits VEGF-Induced HUVECMigration190

As endothelial cell migration is an essential component of an-giogenesis, we investigated whether the increase in sVEGFR-1 levels may be responsible for the compromised angiogene-sis seen in preeclampsia. A significant increase in endothelialcell migration was seen using a modified Boyden’s chamber195in response to VEGF. Preincubation of exogenous sVEGFR-1

(100 ng/mL) with VEGF significantly attenuated HUVEC mi-gration in a dose-dependent manner (Figure 2).

FIG. 2. Soluble VEGFR-1 inhibits VEGF-induced HUVEC migration.HUVECs were stimulated with VEGF (10 and 50 ng/mL) (black column) andmigration was assessed in a modified Boyden’s chamber. Exogenous VEGF wasincubated with sVEGFR-1 (100 ng/mL) (white column) for 30 min before addi-tion to the lower chamber. VEGF induced concentration-dependent migration ofHUVECs was attenuated with the addition of sVEGFR-1. Results are expressedas mean (±SEM) of five separate experiments performed in triplicate. Statisticalcomparison was performed using one-way ANOVA followed by the Student-Newman-Keuls test. ∗ p < .01 versus control; ∗∗ p < .001 versus control.

March 14, 2005 17:24 801 TFJF041-12-56085

4 S. AHMAD AND A. AHMED

VEGF Stimulates the Release of sVEGFR-1We sought to investigate the effect of exogenous VEGF on200

sVEGFR-1 release in endothelial cells. VEGF (10 and 50 ng/mL)induced a concentration-dependant release of sVEGFR-1 intothe conditioned media from HUVECs after 24 h of incubation,demonstrating that there is an autocrine regulation of free andactive VEGF (Figure 3).205

Immunolocalisation of sVEGFR-1 and VEGF in NormalPlacenta Exposed to Hypoxia

To determine the cell type responsible for the increased ex-pression, we exposed placental villous explants to hypoxia andexamined placental sections for sVEGFR-1 and VEGF immunore-210activity. Immunostaining for sVEGFR-1 and VEGF displayeda strong immunoreactivity in the syncytium and the blood ves-sels with diffused staining in the stroma in hypoxia-treated tis-sue in comparison to tissue sections treated under normoxia(Figure 4C–F). No immunostaining was detected in negative215control, with nonimmune antibody (Figure 4A and B).

Hypoxia Induces Release of Angiogenic Factorsfrom Placental Villi

To investigate whether hypoxia stimulated the release of proan-giogenic factors, endothelial cell migration in response to con-220ditioned media from normal term placental villi exposed to hy-poxia and normoxia was assessed. Conditioned media from hy-poxic normal term placental villi significantly increased HUVECmigration when compared to conditioned media from normalterm placental villi cultured under normoxia. More importantly,225preincubation of conditioned media from hypoxia treated nor-mal placental villi with exogenous sVEGFR-1 (100 ng/mL),significantly attenuated the levels of migration (Figure 5A). Pho-tomicrographs of representative filters of migrated cells provide

Q2

FIG. 3. VEGF stimulates the release of sVEGFR-1. HUVECs were stimulatedwith VEGF (10 and 50 ng/mL) for 24 h and the conditioned medium was assayedfor sVEGFR-1 by ELISA. VEGF induced a concentration-dependent increasein release of sVEGFR-1. Results are expressed as mean (±SEM) of five separateexperiments performed in triplicate. Statistical comparison was performed usingone-way ANOVA followed by the Student-Newman-Keuls test. ∗ p < .05 versuscontrol; ∗∗ p < .01 versus control; ∗∗∗ p < .001 versus control.

FIG. 4. Immunolocalization of VEGF and sVEGFR-1 in normal-term pla-centa exposed to hypoxia. Immunohistochemical localisation of VEGF andsVEGFR-1 in normal term placental villous explants cultured under normoxiaand hypoxia. (C and E) Weak VEGF and VEGFR-1 staining was demon-strated in the syncytiotrophoblast (syn) and the smaller capillaries (cap) oftissues xposed to normoxia. (D and F) Strong immunoreactivity for VEGFand sVEGFR-1 was observed in the syncytiotrophoblast (syn) and of the en-dothelial cells of the blood vessels in stem villi (v) of hypoxic treated placentaltissue. (A and B) Control sections incubated with nonspecific preimmune serumshow no immunostaining. Results are representative of five separate experimentsperformed in triplicate. Original magnification: ×200.

a visual representation of migrated endothelial cells per field 230(Figure 5B).

DISCUSSIONIn normal pregnancy, the rapid growth of placenta and the

associated vascularisation occurs from second trimester of preg-nancy onwards. In preeclampsia, circulating levels of sVEGFR- 2351 are elevated during this period (Levine et al. 2004). Recently,an in vivo animal model demonstrated that overexpression ofsVEGFR-1 leads to hypertension, proteinuria, and glomeru-lar endotheliosis, conditions that are similar to preeclampsia(Maynard et al. 2003). In addition to the effect of sVEGFR-1 on 240the maternal circulation as reported by Maynard et al. (2003), weaddressed whether the elevated levels of sVEGFR-1 may have anantiangiogenic effect during placental development. The causeof preeclampsia remains unknown. However, the placenta isclearly involved as preeclampsia can occur in hydatidiform mole 245when placental tissue alone is present; the delivery of the pla-centa is the only known cure for preeclampsia (Redman 1990).The data presented here shows that normal placenta expressedsVEGFR-1 as early as first trimester and its expression increasedthrough gestation. Exposure of normal placental villous explants 250

March 14, 2005 17:24 801 TFJF041-12-56085

ANTIANGIOGENIC EFFECT OF sVEGFR-1 5

FIG. 5. Hypoxia induces release of angiogenic factors from placental Villi.(A). Migration of HUVECs was observed in a modified Boyden’s chamber. Con-ditioned media from normal-term placental villous explants exposed to normoxiaand hypoxia were placed in the lower chamber (black column). Conditioned me-dia were preincubated with sVEGFR-1 (100 ng/mL) (white column) for 30 min.(B) Photomicrograph of migrated endothelial cells on a polycarbonate filters.Results are expressed as mean (±SEM) of five separate experiments performedin triplicate. Statistical comparison was performed using one-way ANOVA fol-lowed by the Student-Newman-Keuls test. ∗ p < .01 versus control; ∗∗ p < .001versus control.

to hypoxia increased expression of sVEGFR-1 in trophoblastand the endothelium. Although conditioned media from normalplacental villous promoted angiogenesis, preincubation of theconditioned media with exogenous sVEGFR-1 significantly at-tenuated endothelial cell migration. Furthermore, stimulation of255endothelial cells with VEGF induced a concentration-dependentrelease of sVEGFR-1, suggesting that sVEGFR-1 is a key factorthat regulates the bioavailability of VEGF during placental de-velopment and that elevated levels of sVEGFR-1 in preeclamp-sia may impair angiogenesis (Ahmad and Ahmed 2004).260

The elevated level of sVEGFR-1 detected from women withpreeclampsia is probably due to placental hypoxia resulting fromuteroplacental insufficiency (Ahmad and Ahmed 2001, 2004;

Zhou 2002; Maynard et al. 2003). Indeed our data shows thatexposure of normal placental villous explants to hypoxia sig- 265nificantly increases expression of sVEGFR-1. Defective remod-elling of the endometrial spiral arteries is the most widely recog-nised predisposing factor for preeclampsia (Hubel 1999). Asa result, perfusion of the intervillous space is impaired lead-ing to placental hypoxia. The mechanism responsible for the 270elevation of sVEGFR-1 in preeclampsia was recently defined(Ahmad and Ahmed 2004). The finding that VEGF-mediated en-dothelial cell migration was blocked by sVEGFR-1 suggests thatsVEGFR-1 may modulate VEGF activity in uteroplacental re-modeling. Consistent with this finding, production of sVEGFR-1 275was significantly increased upon stimulation of endothelial cellswith VEGF. In contrast, a recent study demonstrated that in pla-cental cytotrophoblast hypoxia induced an excess production ofsVEGFR-1 that led to VEGF deficiency and consequently andantiangiogenic state (Nagamatsu et al. 2004). Futhermore, other 280groups have showed that there is indeed an increased expressionand release of VEGF in placenta from women with preeclamp-sia (Baker et al. 1995; Sharkey et al. 1996; Kupferminc et al.1997; Hunter et al. 2000). Likewise, in preeclampsia, but not innormal pregnancies, there is activation of neutrophils and mono- 285cytes during the uteroplacental passage (Mellembakken et al.2002). Upon activation, leukocytes release their granular con-tents, which are capable of mediating vascular damage. SolubleVEGFR-1 may be one such culprit in this process. The additionof exogenous VEGF can induce the release of sVEGFR-1 from 290cultured endothelial cells, in a concentration-dependent manner,and human hematopoietic cell lines also produce sVEGFR-1(Inoue et al. 2000). Thus, the raised level of VEGF in the mater-nal circulation may contribute to the increased level of sVEGFR-1 by stimulating release of sVEGFR-1 from the maternal en- 295dothelium and leukocytes.

It has been hypothesized that the excess sVEGFR-1 produc-tion is a consequence of the placental hypoxia that occurs dur-ing abnormal placental. However, early placental developmentdoes indeed takes place in an environment of reduced oxygen 300tensions relative to maternal tissues, but, during the course ofpregnancy, intervillous oxygen tensions increase. Observationsthat different forms of angiogenesis occur under differing oxy-gen tensions in vivo, and in culture systems in vitro, suggest thatthe interactions of VEGF and its receptors appear to be regu- 305lated by oxygen partial pressure (pO2) (Castellucci et al. 1990;Shore et al. 1997). Earlier studies showed that sVEGFR-1 isup-regulated by hypoxia using Western blot analysis of placen-tal lysates and ELISA (Hornig et al. 2000; Ahmad et al. 2001).However, it remains unclear how hypoxia would affect the net 310angiogenic balance in the placenta because VEGF, a proangio-genic molecule is also upregulated by hypoxia. In this study wehave demonstrated that exposure of placental villous explantsto hypoxia promote a strong expression of sVEGFR-1 in thesyncytiotrophoblast layer. 315

PlGF expression is increased under elevated pO2 and down-regulated by a low pO2 (Khaliq et al. 1999; Ahmed et al. 2000),

March 14, 2005 17:24 801 TFJF041-12-56085

6 S. AHMAD AND A. AHMED

whereas VEGF and its receptors are upregulated by a low pO2

(Tuder et al. 1995; Brogi et al. 1996; Taylor et al. 1997). Thusraised sVEGFR-1 levels will result in a net loss of VEGF and320PlGF activity. Indeed PlGF levels have been reported to be sig-nificantly attenuated in pregnancies complicated by preeclamp-sia (Su et al. 2001; Tidwell et al. 2001; Chappell et al. 2002).Moreover, administration of anti-VEGF compounds can inducehypertension and proteinuria in nonpregnant animals (Sugimoto325et al. 2003). The evidence provided here suggests that the bal-ance between VEGF, PlGF, and their receptors is important foreffective placental development during pregnancy and hypoxiamediated increase in levels of sVEGFR-1 may lead to dysregu-lated angiogenesis associated with preeclampsia.330

REFERENCESAhmad S., and Ahmed A. (2001) Regulation of soluble vegfr-1 by vegf and oxy-

gen and its elevation in pre-eclampsia and fetal growth restriction. Placenta,22, A.7.

Ahmad S., and Ahmed A. (2004) Elevated placental soluble vascular endothelial335growth factor receptor-1 inhibits angiogenesis in preeclampsia. CirculationResearch, 95, 884–891.

Ahmed A., Dunk C., Ahmad S., and Khaliq A. (2000) Regulation of placentalvascular endothelial growth factor (vegf) and placenta growth factor (pigf)and soluble flt-1 by oxygen—a review. Placenta, 21 (Suppl. A), S16–S24.340

Ahmed A., Li X.F., Dunk C., Whittle M.J., Rushton D.I., and Rollason T. (1995)Colocalisation of vascular endothelial growth factor and its flt-1 receptor inhuman placenta. Growth Factors, 12, 235–243.

Baker P.N., Krasnow J., Roberts J.M., and Yeo K.T. (1995) Elevated serum levelsof vascular endothelial growth factor in patients with preeclampsia. Obstetrics345and Gynecology, 86, 815–821.

Benirschke and Kaufmann (1995) Pathology of the Human Placenta, 3rd ed.Springer-Verlag, Berlin.

Brogi E., Schatteman G., Wu T., Kim E.A., Varticovski L., Keyt B., and IsnerJ.M. (1996) Hypoxia-induced paracrine regulation of vascular endothelial350growth factor receptor expression. Journal of Clinical Investigation, 97, 469–476.

Bussolati B., Dunk C., Grohman M., Kontos C.D., Mason J., and Ahmed A.(2001) Vascular endothelial growth factor receptor-1 modulates vascular en-dothelial growth factor-mediated angiogenesis via nitric oxide. American355Journal of Pathology, 159, 993–1008.

Castellucci M., Scheper M., Scheffen I., Celona A., and Kaufmann P. (1990) Thedevelopment of the human placental villous tree. Anatomy and Embryology(Berlin), 181, 117–128.

Chappell L.C., Seed P.T., Briley A., Kelly F.J., Hunt B.J., Charnock-Jones D.S.,360Mallet A.I., and Poston L. (2002) A longitudinal study of biochemical vari-ables in women at risk of preeclampsia. American Journal of Obstetrics andGynecology, 187, 127–136.

Clark D.E., Smith S.K., He Y., Day K.A., Licence D.R., Corps A.N., LammogliaR., and Charnock-Jones D.S. (1998) A vascular endothelial growth factor365antagonist is produced by the human placenta and released into the maternalcirculation. Biology of Reproduction, 59, 1540–1548.

Friedman S.A., Taylor R.N., and Roberts J.M. (1991) Pathophysiology ofpreeclampsia. Clinical Perinatology, 18, 661–682.

Frusca T., Morassi L., Pecorelli S., Grigolato P., and Gastaldi A. (1989) Histo-370logical features of uteroplacental vessels in normal and hypertensive patientsin relation to birthweight. British Journal of Obstetrics and Gynaecology, 96,835–839.

Fujikura T., and Yoshida J. (1996) Blood gas analysis of placental and uter-ine blood during cesarean delivery. Obstetrics and Gynecology, 87, 133–375136.

Hornig C., Barleon B., Ahmad S., Vuorela P., Ahmed A., and Weich H.A. (2000)Release and complex formation of soluble vegfr-1 from endothelial cells andbiological fluids. Laboratory Investigation, 80, 443–454.

Hubel C.A. (1999) Oxidative stress in the pathogenesis of preeclampsia. Pro- 380ceedings of the Society for Experimental Biology and Medicine, 222, 222–235.

Hunter A., Aitkenhead M., Caldwell C., McCracken G., Wilson D., and McClureN. (2000) Serum levels of vascular endothelial growth factor in preeclampticand normotensive pregnancy. Hypertension, 36, 965–969.

Inoue T., Kibata K., Suzuki M., Nakamura S., Motoda R., and Orita K. (2000) 385Identification of a vascular endothelial growth factor (vegf) antagonist, sflt-1,from a human hematopoietic cell line nalm-16. FEBS Letters, 469, 14–18.

Jones C.J., and Fox H. (1980) An ultrastructural and ultrahistochemical studyof the human placenta in maternal pre-eclampsia. Placenta, 1, 61–76.

Kendall R.L., and Thomas K.A. (1993) Inhibition of vascular endothelial cell 390growth factor activity by an endogenously encoded soluble receptor. Proceed-ings of the National Academy of Sciences of the United States of America, 90,10705–10709.

Kendall R.L., Wang G., and Thomas K.A. (1996) Identification of a naturalsoluble form of the vascular endothelial growth factor receptor, flt-1, and its 395heterodimerization with kdr. Biochemical Biophysical Research Communi-cations, 226, 324–328.

Khaliq A., Dunk C., Jiang J., Shams M., Li X.F., Acevedo C., Weich H., WhittleM., and Ahmed A. (1999) Hypoxia down-regulates placenta growth factor,whereas fetal growth restriction up-regulates placenta growth factor expres- 400sion: molecular evidence for ”placental hyperoxia” in intrauterine growthrestriction. Laboratory Investigation, 79, 151–170.

Kupferminc M.J., Daniel Y., Englender T., Baram A., Many A., Jaffa A.J., GullI., and Lessing J.B. (1997) Vascular endothelial growth factor is increased inpatients with preeclampsia. American Journal of Reproductive Immunology, 40538, 302–306.

Levine R.J., Maynard S.E., Qian C., Lim K.H., England L.J., Yu K.F., Schister-man E.F., Thadhani R., Sachs B.P., Epstein F.H., Sibai B.M., Sukhatme V.P.,and Karumanchi S.A. (2004) Circulating angiogenic factors and the risk ofpreeclampsia. New England Journal of Medicine, 350, 672–683. 410

Li X., Shams M., Zhu J., Khalig A., Wilkes M., Whittle M., Barnes N., andAhmed A. (1998) Cellular localization of AT1 receptor mRNA and protein innormal placenta and its reduced expression in intrauterine growth restriction.Angiotensin II stimulates the release of vasorelaxants. Journal of ClinicalInvestigation, 101, 442–454. 415

Lunell N.O., Nylund L.E., Lewander R., and Sarby B. (1982) Uteroplacen-tal blood flow in pre-eclampsia measurements with indium-113m and acomputer-linked gamma camera. Clinical and Experimental Hypertension,B, 1, 105–117.

Maynard S.E., Min J.Y., Merchan J., Lim K.H., Li J., Mondal S., Libermann 420T.A., Morgan J.P., Sellke F.W., Stillman I.E., Epstein F.H., Sukhatme V.P.,and Karumanchi S.A. (2003) Excess placental soluble fms-like tyrosine ki-nase 1 (sflt1) may contribute to endothelial dysfunction, hypertension, andproteinuria in preeclampsia. Journal of Clinical Investigation, 111, 649–658. 425

Mellembakken J.R., Aukrust P., Olafsen M.K., Ueland T., Hestdal K., andVidem V. (2002) Activation of leukocytes during the uteroplacental passagein preeclampsia. Hypertension, 39, 155–160.

Nagamatsu T., Fujii T., Kusumi M., Zou L., Yamashita T., Osuga Y., MomoedaM., Kozuma S., and Taketani Y. (2004) Cytotrophoblasts up-regulate soluble 430fms-like tyrosine kinase-1 expression under reduced oxygen: An implicationfor the placental vascular development and the pathophysiology of preeclamp-sia. Endocrinology, 145, 4838–4845.

Perkins J., St John J., and Ahmed A. (2002) Modulation of trophoblast cell deathby oxygen and egf. Molecular Medicine, 8, 847–856. 435

Redman C.W. (1990) Platelets and the beginnings of preeclampsia. N Engl JMed, 323, 478–480.

Sharkey A.M., Cooper J.C., Balmforth J.R., McLaren J., Clark D.E., Charnock-Jones D.S., Morris N.H., and Smith S.K. (1996) Maternal plasma levelsof vascular endothelial growth factor in normotensive pregnancies and in 440

March 14, 2005 17:24 801 TFJF041-12-56085

ANTIANGIOGENIC EFFECT OF sVEGFR-1 7

pregnancies complicated by pre-eclampsia. European Journal of Clinical In-vestigation, 26, 1182–1185.

Shore V.H., Wang T.H., Wang C.L., Torry R.J., Caudle M.R., and Torry D.S.(1997) Vascular endothelial growth factor, placenta growth factor and theirreceptors in isolated human trophoblast. Placenta, 18, 657–665.445

Soothill P.W., Nicolaides K.H., Rodeck C.H., and Campbell S. (1986) Effectof gestational age on fetal and intervillous blood gas and acid-base values inhuman pregnancy. Fetal Theropy, 1, 168–175.

Su Y.N., Lee C.N., Cheng W.F., Shau W.Y., Chow S.N., and Hsieh F.J. (2001)Decreased maternal serum placenta growth factor in early second trimester450and preeclampsia. Obstetrics and Gynecology, 97, 898–904.

Sugimoto H., Hamano Y., Charytan D., Cosgrove D., Kieran M., SudhakarA., and Kalluri R. (2003) Neutralization of circulating vascular endothelialgrowth factor (vegf) by anti-vegf antibodies and soluble vegf receptor 1 (sflt-1)induces proteinuria. Journal of Biological Chemistry, 278, 12605–12608.

Taylor C.M., Stevens H., Anthony FW., and Wheeler T. (1997) Influence of 455hypoxia on vascular endothelial growth factor and chorionic gonadotrophinproduction in the trophoblast-derived cell lines: JEG, JAr and BeWo.Placenta, 18, 451–458.

Tidwell S.C., Ho H.N., Chiu W.H., Torry R.J., and Torry D.S. (2001) Lowmaternal serum levels of placenta growth factor as an antecedent of clin- 460ical preeclampsia. American Journal of Obstetrics and Gynecology, 184,1267–1272.

Tuder R.M., Flook B.E., and Voelkel N.F. (1995) Increased gene expressionfor vegf and the vegf receptors kdr/flk and flt in lungs exposed to acute or tochronic hypoxia. Modulation of gene expression by nitric oxide. Journal of 465Clinical Investigation, 95, 1798–1807.

Vuorela P., Helske S., Hornig C., Alitalo K., Weich H., and Halmesmaki E.(2000) Amniotic fluid—soluble vascular endothelial growth factor receptor-1in preeclampsia. Obstetrics and Gynecology, 95, 353–357.