ARTICLE IN PRESS

Ann Anat 190 (2008) 477—489

0940-9602/$ - sdoi:10.1016/j.

�CorrespondE-mail addr

www.elsevier.de/aanat

RESEARCH ARTICLE

Soluble and membranous vascular endothelialgrowth factor receptor-1 in pregnanciescomplicated by pre-eclampsia

Richa Tripathia,�, Gayatri Ratha, Anju Jainb, Sudha Salhanc

aDepartment of Anatomy, Vardhman Mahavir Medical College and Safdarjang Hospital, New Delhi 110029, IndiabDepartment of Biochemistry, Lady Hardinge Medical College and S.K. Hospital, New Delhi, IndiacDepartment of Obstetrics and Gynaecology, VMMC & Safdarjang Hospital, New Delhi, India

Received 4 March 2008; accepted 4 August 2008

KEYWORDSPre-eclampsia;sVEGFR1;MembranousVEGFR-1;ELISA;Immunohisto-chemistry;Placenta

ee front matter & 2008aanat.2008.08.002

ing author. Tel.: +011 26esses: richashubhi@yah

SummaryVascular endothelial growth factor receptor-1 (VEGFR-1) is essential for the normaldevelopment and function of the placenta. Defective placental vasculogenesis andtrophoblast function may lead to pre-eclampsia, a pregnancy-specific syndrome ofhypertension and proteinuria. In order to study the association of VEGFR-1 with thedevelopment of pre-eclampsia, a cross-sectional study was carried out to evaluatethe concentration of soluble VEGFR-1 (sVEGFR-1) in 360 serum samples and toanalyze the expression of membranous VEGFR-1 in 40 placental samples of normaland pre-eclamptic pregnant women. Serum and placental samples at differentgestational ages were collected from the Department of Obstetrics and Gynaecology,VMMC and Safdarjang Hospital, New Delhi. The serum levels of sVEGFR-1 and theexpression of membranous VEGFR-1 were estimated by enzyme-linked immunosor-bent assay and immunohistochemistry, respectively. The serum levels of sVEGFR-1were seen to be positively increased (p ¼ 0.0001) in patients with pre-eclampsiaat different gestational intervals as compared to the healthy pregnant womenthey were matched with. However, receiver operating characteristic (ROC) curveanalysis showed a higher sensitivity (89.17%) and specificity (90.0%) in early onset(p34 weeks) in contrast with the late-onset (434 weeks) pre-eclamptic group. Also,significant up-regulation of membranous VEGFR-1 immunoreactivity was observed inall placental cells (syncytiotrophoblast, cytotrophoblast, endothelial cells andHofbauer cells) of pre-eclamptic groups in both p34 weeks (p ¼ 0.0001) and 434weeks (p ¼ 0.0001) as compared to the normal group. Elevated sVEGFR-1 serumlevels and up-regulated membranous VEGFR-1 expression in placenta denoteabnormality in VEGF-mediated function in all placental cells, and thus maycontribute to etiopathogenesis of pre-eclampsia. Nevertheless, this study also

Elsevier GmbH. All rights reserved.

716440 (O)/9891082708 (mobile).oo.co.in (R. Tripathi), gayatrirath@rediffmail.com (G. Rath).

ARTICLE IN PRESS

R. Tripathi et al.478

shows the possible diagnostic utility of sVEGFR-1 as a sensitive and specificbiomarker for the early onset (p34 weeks) of pre-eclampsia.& 2008 Elsevier GmbH. All rights reserved.

Introduction

Pre-eclampsia is a syndrome diagnosed by thepresence of hypertension, proteinuria and/or sei-zures and is thought to be due to endothelialdysfunction (Roberts and Lain, 2002). It affectsabout 5% of pregnancies, resulting in substantialmaternal and neonatal morbidity and mortality(Roberts and Cooper, 2001). The mechanismsresponsible for the development of endothelialdysfunction in this disorder have yet to beelucidated.

Vascular endothelial growth factor (VEGF) is adimeric glycoprotein with potent angiogenic prop-erties that promote endothelial cell proliferation,migration and survival. VEGF has a family ofreceptors, the most important of which areVEGFR-1 (Flt-1, fms-like tyrosine kinase-1) andVEGFR-2 (KDR, kinase insert domain receptor)(Dvorak, 2002). VEGFR-1 has two isoforms: asoluble isoform –soluble vascular endothelialgrowth factor receptor-1 (sVEGFR-1) – and atrans-membranous isoform. sVEGFR-1 is generatedas a splice variant of VEGFR-1 gene and contains anextra-cellular binding domain, lacking tyrosinekinase signalling. sVEGFR-1 binds to VEGF andplacental growth factor (PLGF) and thus preventsthe interaction of these angiogenic growth factorswith VEGFR-1 receptors on the surface of endothe-lial cells, leading to endothelial cell dysfunction(Autiero et al., 2003; Sugimoto et al., 2003). It hasalso been suggested that VEGF affects multipletrophoblast functions, i.e. invasion, differentia-tion, proliferation and release of vasorelaxants viaan autocrine loop. Abnormal trophoblast invasionand differentiation are associated with pre-eclampsia (Roberts and Lain, 2002; Zhou et al.,2002). VEGFR-1 is the major mediator in mitogenic,angiogenic and permeability enhancing effects ofVEGF (Ferrara et al., 2003). It has recently beendemonstrated that exogenous sVEGFR-1 adminis-tered to pregnant rats induces hypertension,proteinuria and glomerular endotheliosis, suggest-ing that an excess of sVEGFR-1 may have apathogenic role in pre-eclampsia (Maynard et al.,2003).

In a series of elegant experiments, variousauthors have reported an increased concentration

of sVEGFR-1 as well as membranous VEGFR-1expression in serum, amniotic fluid and placentaof pregnancies complicated by pre-eclampsia,intra-uterine foetal growth retardation (IUGR) andtrisomy-13 (Hunter et al., 2000; Vuorela et al.,2000; Koga et al., 2003; Tsatsaris et al., 2003;Levine et al., 2004, 2006; Bdolah et al., 2006;Marini et al., 2007; Masuyama et al., 2006).sVEGFR-1 antagonizes VEGF during circulation bybinding and reducing the availability of free VEGF,thus preventing its interaction with VEGFR-1receptor in pre-eclampsia (Maynard et al., 2003;Nagamatsu et al., 2004). Therefore, it is concludedthat elevated levels of sVEGFR-1 result in excessiveVEGF inhibition and interferes with endothelial cellfunction, thereby reducing placental angiogenesis(Robinson and Stringer, 2001). Lower concentra-tions of free circulating PLGF and VEGF have beennoticed before and after the inception of pre-eclampsia (Maynard et al., 2003; Taylor et al.,2003). The severity of pre-eclampsia correlateswith the circulating levels of sVEGFR-1 and proxi-mity to the onset of hypertension or proteinuria(Chaiworapongsa et al., 2005; Powers et al., 2005).In-vitro studies have shown that sVEGFR-1 in-creases (Ahmed et al., 2000) in trophoblast cellsunder conditions of placental hypoxia associatedwith incomplete remodelling of maternal spiralarteries that persistently offer high resistance touterine artery blood flow. This can be predisposedto vascular rupture in the placental bed, especiallyafter an onrush of hypertension (Eskes, 1997).However, there is still not enough evidence tosupport the concept that altered levels of thisangiogenic factor in the maternal serum of pre-eclamptic women is the consequence or the causeof placental insufficiency.

Accumulating evidence suggests that a balanceamong VEGF, PLGF and their receptors are impor-tant for effective vasculogenesis, angiogenesis andplacental development during pregnancy. Severalrecent studies in both animals and humans havesuggested that blockage of VEGF action plays a rolein the pathophysiology of pre-eclampsia. Adminis-tration of anti-VEGF compounds can inducehypertension and proteinuria in non-pregnant ani-mals (Aiello et al., 1995) and humans enrolled inanti-angiogenic trials (Rodesch et al., 1992).

ARTICLE IN PRESS

Soluble and membranous vascular endothelial growth factor receptor-1 479

Furthermore, anti-VEGF therapy in patients withcancer has been associated with hypertension,proteinuria and reversible posterior leukoencepha-lopathy syndrome, all of which are the hallmarks ofpre-eclampsia and eclampsia (Hinchey et al., 1996;Yang et al., 2003).

Several studies have been undertaken to inves-tigate the expression of VEGF and its receptors innormal human placenta in trophoblast, endothelialand Hofbauer cells in placental villi (Cooper et al.,1995; Clark et al., 1996; Vuckovic et al., 1996;Demir et al., 2004). Membranous VEGFR-1 expres-sion is found to be associated with hypertensive aswell as IUGR patients (Regnault et al., 2002; Mariniet al., 2007). Therefore, an imbalance in angio-genic factors may be associated with vascularendothelial dysfunction and a maternal syndromeof pre-eclampsia (Bdolah et al., 2004).

Till now, only controversial data are availablewith regard to gestational sVEGFR-1 concentrationand expression of membranous VEGFR-1 in early-and late-onset pre-eclamptic patients. Therefore,this cross-sectional study has been performed to(1) chronologically determine and correlate thesVEGFR-1 in serum at different intervals inpre-eclamptic women and gestationally matchednormal pregnant women, and to (2) studythe membranous VEGFR-1 expression pattern inearly- (p34 weeks) and late-onset (434 weeks)pre-eclamptic placentae.

Materials and methods

A total of 360 subjects were recruited for thiscross-sectional study from the OPD clinic/wardsbetween the years 2004 and 2006 from theDepartment of Obstetrics and Gynaecology, Safdar-jang Hospital, including obtaining a written consentfrom the patients. The research and ethics com-mittee of Vardhman Mahavir Medical College andSafdarjang Hospital approved the study. Pre-eclampsia is defined as diastolic blood pressureabove 90mmHg and systolic blood pressure above140mmHg on at least two consecutive measure-ments, 6 h apart; oedema and/or proteinuriaX300mg in a 24 h urine collection or X1+ dipstickon a random urine sample. Patients with a historyof hypertension or any renal disease were excludedfrom the study. Healthy pregnant women withuncomplicated pregnancies served as the controlgroup. All the subjects were classified andnumbered prior to any laboratory analysis so as toavoid bias.

A total of 360 serum samples were collected, 180were from diagnosed pre-eclampsia patients and anequal number were from gestationally matcheduncomplicated pregnant women (control group).Serum was separated by centrifugation and storedat �80 1C till the batch was analyzed.

Forty placentae were collected at the time ofdelivery, 20 were from patients with pre-eclamp-sia, which were sub-classified as either early-(p34 weeks, n ¼ 10) or late-onset (434 weeks,n ¼ 10) disease according to the gestational age ofthe pre-eclamptic development. Twenty normalplacentae were collected and sub-classified intopre-term (p34 weeks, n ¼ 10) and term placentae(434 weeks, n ¼ 10) to be compared with thestudy group. Some tissues were randomly sampledfrom each placenta, then immersion-fixed in 10%formalin and thereafter embedded in paraffin wax.

ELISA

Serum samples were randomly sorted and quan-titatively analyzed for sVEGFR-1 by enzyme-linkedimmunosorbent assay (Quantikine human solubleVEGFR-1, R&D Systems Europe, Ltd.). This assayemploys a quantitative sandwich immunoassaytechnique. Briefly, recombinant human VEGFR-1standards and maternal plasma specimens wereincubated in duplicate wells of the microtiterplates pre-coated with monoclonal antibodiesspecific for VEGFR-1. After an incubation period,the assay plates were subjected to a wash step toremove unbound antibody–enzyme reagent. Then,an enzyme-linked polyclonal antibody specific toVEGFR-1 was added to the wells. Following a washto remove any unbound antibody–enzyme reagent,a substrate solution (tetramethylbenzidine) wasadded to the wells. Colour developed in the assayplates proportionally to the amount of VEGFR-1bound in the initial step. All assays were performedin their respective duplicate samples. The opticaldensity was determined by subtracting readings at540 nm from readings at 450 nm. The protein levelswere calculated using a standard curve derivedfrom a known concentration of respective recom-binant proteins. The minimum detection dose ofassays was 3.5 pg/ml. The inter- and intra-assaycoefficient of variation were 3.2% and 7.4%,respectively.

Immunohistochemistry

Five micron serial sections of fresh placentalvillous tissue were collected on poly-L-lysine-coated slides (Sigma, St. Louis, MO, USA),

ARTICLE IN PRESS

R. Tripathi et al.480

de-waxed and then dehydrated. Antigen retrievaltreatment was done in citrate buffer in a waterbath at 95–100 1C for 20min and then cooled for20–30min. After washing in 0.01M Tris-bufferedsaline (pH 7.4), sections were incubated with 1%BSA followed by overnight incubation at 4 1C withprimary monoclonal anti-human antibody ofVEGFR-1 (C-17, sc-316, Santa Cruz Biotechnology)at a dilution of 1:100. Sections were then washedwith 0.05% Tween-20 in 0.01M TBS before theapplication of secondary biotinylated antibodies.Biotinylated secondary antibodies and LSAB stain-ing systems from Dako Cytomation were used inaccordance with the manufacturer’s instructions.Staining was completed after incubation withsubstrate–chromogen solution (liquid DAB, Dako,code: K0679) and applied until the maximum levelof reaction product was obtained. Sections werethen counter stained with Mayer’s haematoxylinsolution and mounted with DPX. Negative controlstaining was performed using IgG non-immuneserum (Dako), and for the positive controlrabbit anti-human chorionic gonadotropin (hCG,A-023102, Dako, Denmark) was used. For eachtarget staining, all slides were processed at thesame time. The stained tissue slides were examinedunder an Olympus microscope (model BX-51,Olympus America, Inc., Melville, NY). Images werecaptured by a CCD camera and ‘‘Olysia Bio-report’’software linked with a personal computer. Immu-nohistochemical staining was used mainly to loca-lize and study the intensity of target molecules inthe placental villous tissue. The intensity ofimmunoreactivity was semi-quantitatively evalu-ated by grouping positively stained cells accordingto the following categories: 0 (no staining), +1(weak, but detectable), +2 (moderate) and +3(intense). In each slide, five areas were evaluatedunder a microscope using 400� original magnifica-tion. For each tissue, an H-score value wascalculated by adding the percentage of cellsgrouped in one intensity category and multiplyingthis number with the weighed intensity of the

Table 1. Clinical details of the study groups

Parameters Control (n ¼ 180

Maternal age (year) (mean7SD) 25.4373.0Baby wt. (g) (mean7SD) 2042.87704.6Primigravida no. (%) 50 (90/180)Systolic BP (mean7SD) 119.1878.13Diastolic BP (mean7SD) 78.9477.77

BP, blood pressure.�po0.05 is considered to be significant.

staining, using the formula (H-score ¼ Pc (s+1),where ‘‘s’’ represents the intensity scores and Pc isthe corresponding percentage of the cells (Demiret al., 2004). All tissues were examined by twoobservers who were blinded to the slides,i.e. patient’s identity, clinical details, gestationalages and their average scores were used.

Statistical analysis

Statistical analysis was performed using statisti-cal software STATA 9.0 (College station, TX, USA).Data were presented as number (%), mean7SD ormedian (range) as per the requirement. Thedifference in mean/median amongst the groupswas compared using the one-way ANOVA/Kruskal–Wallis test with Bonferroni correction.

The receiver operating characteristic (ROC)curve analysis was also performed to identify thecut-off serum levels for pre-eclamptic patients incomparison to normal subjects. A probability levelof pp0.05 was considered statistically significant.

Results

There was no significant difference in maternalage between the control and the pre-eclampticgroups. The weight of the baby was found to bereduced in pre-eclamptic patients as compared tothe control ones, but the difference was notsignificant. The primigravidae women constituted50% of the control group and 57.78% of the pre-eclamptic group. The difference in the meansystolic blood pressure and diastolic blood pressurewas significant between the two groups (Table 1).

ELISA

The total serum sVEGFR-1 concentration inthe pre-eclamptic group was significantly higher(median: 39,559.84 pg/ml; p ¼ 0.0001) than in the

) Pre-eclampsia (n ¼ 180) p-Value

25.4473.8 0.471948.9764.9 0.2957.78 (76/180) 0.13150.21714.75 0.0001�

99.0177.15 0.0001�

ARTICLE IN PRESS

Soluble and membranous vascular endothelial growth factor receptor-1 481

control group (median: 3360 pg/ml; p ¼ 0.0001),which is statistically significant. sVEGFR-1 concen-trations were found to decrease with an increase ingestational age from 22 to 39 weeks in pre-eclampsia (p ¼ 0.0001). Therefore, there was anegative correlation between the serum concen-trations of sVEGFR-1 and the advancing gestationalage from 22 to 39 weeks. In contrast, a positiveassociation was seen in the control group

Table 2. sVEGFR-1 levels (pg/ml) in serum at different ges

Gestational age (weeks) sVEGFR-1 (pg/ml)

Control (median, range)

22–23 (n ¼ 20) 456 (342–3360)24–25 (n ¼ 20) 456 (342–3360)26–27 (n ¼ 20) 1514.9 (622–4150)28–29 (n ¼ 20) 2059.9 (842–5500)30–31 (n ¼ 20) 13,063.03 (3250–42,000)32–33 (n ¼ 20) 15,840.41 (3800–42,000)34–35 (n ¼ 20) 16,245.26 (1020–42,000)36–3 (n ¼ 20) 16,245.26 (1160–42,000)38–39 (n ¼ 20) 6291.04 (522–42,000)

Total (n ¼ 180) 3360 (342–42,000)

�po0.05 is considered to be significant.

Table 3. Median value of sVEGFR-1 levels (pg/ml) in cont434 gestational weeks

Gestational age(weeks)

Control (n ¼ 180) (median pg/ml,range)

p34 3125 (342–42,000)434 10,557.63 (522–42,000)

�po0.05 is considered to be significant.

Figure 1. Serum sVEGFR-1 concentration (pg/ml) in norconcentration of sVEGFR-1 was significantly higher in prevalues. (a) At p34 gestational weeks (p ¼ 0.0001) and (b) a

(p ¼ 0.0001, Table 2). The serum sVEGFR-1 levelsin the pre-eclamptic group were highest from 22 to29 weeks (median: 42,000 pg/ml) and from 36 to 39weeks (median: 25,312.98 pg/ml) in the controlgroup.

In patients characterized as having eitherearly- or late-onset pre-eclampsia, the resultswere similar (Table 3). The median serum concen-tration of sVEGFR-1 in pre-eclamptic women was

tational weeks of control and pre-eclamptic groups

Pre-eclampsia (median, range) p-Value

42,000 (23,485–42,000) 0.0001�

42,000 (23,485–42,000) 0.0001�

42,000 (2642–42,000) 0.0001�

42,000 (2130–42,000) 0.0001�

31,766.15 (1148–42,000) 0.0002�

23,885.87 (1248–42,000) 0.410324,390.5 (3649.64–42,000) 0.2508

25,312.98 (3996.36–42,000) 0.434525,312.98 (3996.36–42,000) 0.0126

39,559.84 (1148–42,000) 0.0001�

rol and pre-eclamptic groups categorized as either p or

Pre-eclampsia (n ¼ 180) (median pg/ml,range)

p-Value

42,000 (1248–42,000) 0.0001�

24,390 (3649.64–42,000) 0.01�

mal and pre-eclamptic patients. The median serum-eclampsia patients. Horizontal lines represent mediant 434 gestational weeks (p ¼ 0.01).

ARTICLE IN PRESS

0.00

0.25

0.50

0.75

1.00S

ensi

tivity

0.00

0.25

0.50

0.75

1.00

Sen

sitiv

ity

0.00 0.25 0.50 0.75 1.001 - Specificity

Area under ROC curve = 0.9247

0.00 0.25 0.50 0.75 1.001 - Specificity

Area under ROC curve = 0.6342

Sensitivity:89.17%Specificity:90%AUC:0.92

Sensitivity:55%%Specificity:58.33%AUC:0.63

≤ 34 weeks > 34 weeks

Figure 2. Receiver operating characteristic (ROC) curve of sVEGFR-1 serum levels in pre-eclampsia, showing ability ofmaternal serum sVEGFR-1 to differentiate pre-eclampsia from normal pregnancies. (a) At p34 gestational weeks(AUC, area under curve ¼ 0.92) and (b) at 434 gestational weeks (AUC, area under curve ¼ 0.63).

Table 4. Test performance of serum sVEGFR-1 in diagnosing pre-eclampsia as compared to controls

Study group (weeks) Conc. cut off TP TN FP FN Sensitivity (%) Specificity (%) LR Area (95% CI)

p34 X15,855.86 107 108 12 13 89.17 90.0 8.9 0.92 (0.88, 0.95)434 X14,784.18 33 35 25 27 55.0 58.33 1.3 0.634 (0.53, 0.73)

TP, total positive; TN, total negative; FP, false positive; FN, false negative; LR, likelihood ratio; CI ¼ confidence interval.

R. Tripathi et al.482

significantly higher in p34 weeks (median:42,000 pg/ml, Figure 1a) as well as those of 434weeks (median: 24,390.5 pg/ml, p ¼ 0.0002,Figure 1b) as compared to normal pregnant women(median: 3125 pg/ml, p ¼ 0.0001). However, themedian serum concentration of sVEGFR-1 in controlgroup was lower in pre-term (p34 weeks, median:3125 pg/ml) than in the term patients (434 weeks,median: 10,557.63 pg/ml, p ¼ 0.0001).

According to the ROC curve analysis, the serumsVEGFR-1 level in early-onset pre-eclampsia(p34 weeks) had a sensitivity and specificity of89.17% and 90.0%, respectively. However, in thelate onset of disease (434 weeks), the sensitivityand specificity was found to be 55% and 58.33%,respectively (Figure 2a and b). The highest positiveLR of serum sVEGFR-1 in women within p34 weekspre-eclampsia was 8.9 and in 434 weeks was 1.3(Table 4).

Immunohistochemistry

In the control group, at p34 weeks, mild VEGFR-1 immunoreactivity was observed in syncytiotro-phoblast, cytotrophoblast, endothelial andHofbauer cells. However, in the pre-eclampsia

group, moderate immunoreactivity was noted(Figure 3a–d). This up-regulation of VEGFR-1immunoreactivity in the pre-eclampsia group wassignificant in all the placental cells, i.e. insyncytiotrophoblast (p ¼ 0.0001), cytotrophoblast(p ¼ 0.0001), endothelial (p ¼ 0.0001) and inHofbauer cells (p ¼ 0.0001).

At 434 weeks, moderate VEGFR-1 immunoreac-tivity was observed in tertiary villi cells of controlplacenta. A significant increase in intensity ofVEGFR-1 staining was observed in the pre-eclamp-sia group (p ¼ 0.0001, Figure 3e–h). Also, ascompared to pre-eclamptic patients of p34 weeks,it is over expressed (p ¼ 0.0001) with advancinggestational age. Immunoreactivity was also de-tected in syncytial knots of pre-eclamptic patientsof both sub-groups. The intensity pattern of allimmunoreactive cells in control and pre-eclampticgroups at p34 and 434 weeks are shown in Table 5(Figure 4).

Discussion

An imbalance in angiogenic factors is thought toplay an intimate role in the pathogenesis of clinical

ARTICLE IN PRESS

cba d

e f hg

ji

V S

E

C

S

E

S

EHC

S EEC

S

E

SE S S

C

Figure 3. Immunohistochemical staining for VEGFR-1 (a)–(j) in normal and pre-eclamptic placenta. (a) 22 weeks (p34weeks) control placentae showing mild VEGFR-1 expression syncytiotrophoblast (S) and endothelial cells (E) (scale bar:20 mm). (b) 22 weeks (p34 weeks) control placentae showing mild VEGFR-1 expression in syncytiotrophoblast (S),cytotrophoblast (C), endothelial cells (E) and Hofbauer cells (H) (scale bar: 5mm). (c) 22 weeks (p34 weeks) pre-eclamptic placentae showing moderate VEGFR-1 expression in syncytiotrophoblast (S), endothelial cells (E) andHofbauer cells (H) (scale bar: 20 mm). (d) 22 weeks (p34 weeks) pre-eclamptic placentae showing moderate VEGFR-1expression in syncytiotrophoblast (S), endothelial cells (E) and Hofbauer cells (H) (scale bar 5mm). (e) 37 weeks (434weeks) control placentae showing moderate VEGFR-1 expression in syncytiotrophoblast (S) and endothelial cells (E)(scale bar: 20 mm). (f) 37 weeks (434 weeks) control placentae showing moderate VEGFR-1 expression insyncytiotrophoblast (S) and endothelial cells (E) (scale bar: 5 mm). (g) 37 weeks (434 weeks) pre-eclamptic placentaeshowing intense VEGFR-1 expression in syncytiotrophoblast (S) and endothelial cells (E) (scale bar: 20 mm). (h) 37 weeks(434 weeks) pre-eclamptic placentae showing intense VEGFR-1 expression in syncytiotrophoblast (S) andcytotrophoblast cells (C) (scale bar: 5 mm). (i) Negative control – VEGFR-1 control sections incubated with IgG non-immune serum showing placental villi (V) with no staining (scale bar: 20 mm). (j) Positive control – hCG showingplacental villi with intense staining in syncytiotrophoblast (S), cytotrophoblast (C) and endothelial cells (E) (scale bar:5 mm).

Soluble and membranous vascular endothelial growth factor receptor-1 483

pre-eclampsia. Maynard et al. (2003) reported anincreased placental sVEGFR-1 mRNA expression inpre-eclampsia using microarray technology. Kendalland Thomas (1993) suggested that the serum ofnormal pregnant women is capable of inducingendothelial cells to form tube-like structures, abiological effect inhibited by the addition ofsVEGFR-1. Also, serum from pre-eclamptic patientswith higher sVEGFR-1 levels inhibited endothelialtube formation and the addition of VEGF canrestore its effect. The inhibiting effect of serumfrom pre-eclamptic women disappeared after

delivery, suggesting that this factor may have beenreleased by the placenta (Kendal et al., 1993). Thisis also supported by Barleon et al. (2001) whostated that plasma sVEGFR-1 concentration ishigher in normal pregnancy than in non-pregnantwomen and decreases dramatically after delivery(Koga et al., 2003). All such studies support theconcept that the placenta is a likely source ofserum or plasma sVEGFR-1 during pregnancy.

In the last few years, although attempts havebeen made to measure sVEGFR-1 concentrations inpre-eclampsia before and during the clinical onset

ARTICLE IN PRESS

0

50

100

150

200

250

300

350

400

Hofbauer

Control Preeclampsia Control Preeclampsia

H-s

core

(VE

GFR

-1 E

xpre

ssio

n)

**

**

** * *

Syncytio Cyto Endo0

100

200

300

400

500

600

Hofbauer

H-s

core

(VE

GFR

-1 E

xpre

ssio

n)

Syncytio Cyto Endo

≤ 34 weeks > 34 weeks

Figure 4. Comparison of immunohistochemical H-scores for membranous VEGFR-1 in cells of control and pre-eclampticpatients. Data are expressed as mean values (*p ¼ 0.0001 vs. control). (a) At p34 gestational weeks and (b) at 434gestational weeks.

Table 5. Assessment of membranous VEGFR-1 immunostaining in placenta

Cells p34 weeks (n ¼ 20) 434 weeks (n ¼ 20)

Control group(n ¼ 10)

Pre-eclamptic group(n ¼ 10)

Control group(n ¼ 10)

Pre-eclamptic group(n ¼ 10)

Syncytiotrophoblast + ++ ++ +++Cytotrophoblast + ++ ++ +++Endothelial cells + ++ ++ +++Hofbauer cells + ++ ++ +++Syncytial knots � ++ � ++

Control vs. pre-eclampsia in p34 weeks and 434 weeks, p ¼ 0.0001.Control vs. control in p34 weeks and 434 weeks, p ¼ 0.0001.Pre-eclampsia vs. pre-eclampsia in p34 weeks and 434 weeks, p ¼ 0.0001.+ Mild reactivity, ++ moderate reactivity, +++ intense reactivity.

R. Tripathi et al.484

of disease, few studies describe sVEGFR-1 concen-trations at various intervals of second and thirdtrimesters of pre-eclamptic pregnancy. Levineet al. (2006) reported alterations in circulatingsVEGFR-1 levels several weeks preceding anyclinical sign of the disease. According to Chaiwor-apongsa et al. (2005), sVEGFR-1 elevation begins at6–10 weeks and is more pronounced at 2–5 weeksbefore the clinical manifestation of symptoms.However, Wathen et al. (2006) noticed that thelevels did not dramatically alter during earlypregnancy (14 weeks of gestation) when theabnormal placentation which is associated withsevere premature pre-eclampsia occurs but were

found to be elevated at 16–20 weeks of gestation.Rana et al. (2007) reported that patients withelevated levels of sVEGFR-1 at 17–20 weeks weredestined to develop pre-eclampsia. In contrast,Unal et al. (2007) reported that sVEGFR-1 does notalter in the second trimester of patients whosubsequently develop pre-eclampsia. Others havealso reported a high risk of pre-eclampsia withelevated sVEGFR-1 concentrations between theirfirst and second trimesters (Kim et al., 2007; Vattenet al., 2007; Erez et al., 2008). Chaiworapongsaet al. (2008) noticed that this angiogenic factoralteration may be strongly associated with thedevelopment of severe clinical pre-eclampsia.

ARTICLE IN PRESS

Soluble and membranous vascular endothelial growth factor receptor-1 485

Findings by Chaiworapongsa et al. (2004) andMoore Simas et al. (2007) depicted a higher meanserum concentration of sVEGFR-1 in patientswho had already developed clinical symptoms ofpre-eclampsia (less than 34 weeks). However, thecausative factors accounting for the increase incirculating sVEGFR-1 in pre-eclampsia remains tobe determined. An in-vitro study by Ahmed et al.(2000) reported that sVEGFR-1 becomes increasedin trophoblast cells under conditions of reducedoxygen tension. Due to varying results of sVEGFR-1concentration, an attempt was made to measure itsconcentration from 22 to 39 weeks of gestation at2-week intervals, i.e. (a) 22–23, (b) 24–25, (c)26–27, (d) 28–29, (e) 30–31, (f) 32–33, (g) 34–35,(h) 36–37 and (i) 38–39 weeks of gestation in pre-eclamptic and healthy pregnant women. There is aconsistency between our study and other previousreports that found elevated plasma/serum concen-trations of sVEGFR-1 in patients with pre-eclampsiaat the time of diagnosis. Our study showed apositive association of serum concentrations ofsVEGFR-1 with advancing gestational weeks from22 to 39 weeks (p ¼ 0.0001) in pre-eclampsia ascompared to gestationally matched normal controlones. This increased concentration of sVEGFR-1 wasstable till 29 weeks of gestation after which itshowed a late-gestational decline (p ¼ 0.0001,Table 2).

The gradual increase in sVEGFR-1 concentrationcoincides with the onset of disease during preg-nancy, showing its vital role in pathogenesis.Salahuddin et al. (2007) reported 90% sensitivityand specificity of sVEGFR-1 in differentiating thirdtrimester (late onset) pre-eclampsia patients fromnormal ones. However, in our study, we found asensitivity and specificity of 89.17% and 90.0%,respectively, in early onset of disease, whichsuggests that this anti-angiogenic factor may beused as a sensitive and specific marker in diagnos-ing early-onset pre-eclampsia (p34 weeks). Thepositive LR of this marker can increase theprobability of early identification of pre-eclampsia,required to alter clinical decision making (Jaeschkeet al., 1994).

This study showed an increased expression ofmembranous VEGFR-1 during maximal placentaldevelopment in control placentae, while othersreported lower levels of VEGFR-1 during mid- tolate gestation (Cheung and Brace, 1999; Bogicet al., 2001; Regnault et al., 2002). In pre-eclampsia patients, Vuorela and Halmesmaki(2006) observed lower levels of immunoreactivityin myometrium, deciduas and in vascular smoothmuscle cells but not in decidual vascular endothe-lium. However, Marini et al. (2007) detected lower

levels of VEGFR-1 immunoreactivity only in someportions of trophoblast and some clusters ofstromal cells in pre-eclamptic patients.

We observed complete immunoreactivity intertiary villi of control as well as pre-eclampticpatients. Significant up-regulated immunoreactiv-ity has been noticed in villous endothelial cells andother non-endothelial cells – syncytiotrophoblast(p ¼ 0.0001), cytotrophoblast (p ¼ 0.0001), en-dothelial (p ¼ 0.0001) and in Hofbauer cells(p ¼ 0.0001) in both the sub-groups of pre-eclamp-tic patients as compared to control group. Thissuggests that membranous VEGFR-1 may be in-volved not only in the regulation of placentalangiogenesis in a paracrine manner but also playan autocrine role in trophoblast invasion, thussupporting the hypothesis that the VEGF ligandreceptor system, apart from its endothelial mito-genic effect, must have additional functions whenexpressed in non-endothelial cells. This observationis partially supported by Helske et al. (2001) andYang et al. (2008). Helske et al. (2001) reportedincreased immunoreactivity only in syncytiotropho-blast cells of 50% pre-eclamptic patients and Yanget al. (2008) observed increased VEGFR-1 immu-noreactivity in these cells of pre-eclamptic placen-tae. We also found increased VEGFR-1 expression inplacental syncytial knots in pre-eclamptic placen-tae, suggesting that shed syncytial fragments mayserve as a vehicle to carry excess VEGFR-1 into thematernal circulation, thereby enhancing the detri-mental aberrant anti-angiogenic function ofVEGFR-1 on the maternal peripheral vasculature.Thus, the increased VEGFR-1 levels observed mayalter normal trophoblast function in pre-eclampsia.Endothelial dysfunction, a feature of pre-eclamp-sia, is characterized by increased endothelial cell-mediated vasoconstriction, vascular permeabilityand platelet aggregation leading to maternalhypertension and proteinuria (Roberts and Lain,2002). The expression of VEGFR-1 in endothelialcells and in Hofbauer cells may affect normal fusionof angiogenic cell cords as well as organization ofvascular capillaries. Hence, it suggests that VEGFR-1 is a key factor associated with pre-eclampticpregnancy and its increased expression may alterVEGF-mediated function on trophoblast, endothe-lial and Hofbauer cells.

Experimental observations suggest that the pla-centa produces sVEGFR-1 and is a rich source of aVEGF receptor in amniotic fluid during pregnancy(Vuorela et al., 2000). Its concentration in amnioticfluid is elevated in pre-eclampsia due to reducedoxygen tension or hypoxia (Kendall and Thomas,1993; Ahmed et al., 2000; Li et al., 2005; Pandavalaet al., 2006). Hypoxia also up-regulates VEGF,

ARTICLE IN PRESS

R. Tripathi et al.486

membranous VEGFR-1 along with soluble VEGFR-1in the pre-eclamptic placenta (Pandavala et al.,2006; Munaut et al., 2008). Various studies havereported that secretion of sVEGFR-1 in maternalcirculation might be due to injury in villoustrophoblast (Huppertz et al., 2003), by cytokines(Matsubara et al., 2005), remodelling of the villoustree (Geva et al., 2002; Kearney et al., 2004),reduced uterine perfusion (Rajakumar et al.,2004a, b), complement activation (Girardi et al.,2006) or immune mechanism (Stepan et al., 2006).Our results are in concordance with Munaut et al.(2008) and showed that both the isoforms ofreceptor (soluble and membranous) might be up-regulated in the pre-eclamptic condition andsupport the hypothesis that in pre-eclampsia,overproduction of VEGF family factors by pre-eclamptic placenta is a consequence of inducedhypoxia. It is also possible that an ischemic placentacould induce expression of membranous VEGFR-1and release of sVEGFR-1 from villous trophoblastcells, which are in direct contact with maternalblood in intervillous space, in an autocrine manner.

The summing of our study is that the significantrise of serum level of sVEGFR-1 and expression ofmembranous VEGFR-1 in the placenta during earlyand late onset of pre-eclamptic patients may behighly valuable in developing rational managementplans for these patients. High sensitivity (89.17%)and specificity (90.0%) in serum samples of early-onset (p34 weeks) study group strongly supportsthe diagnostic utility of sVEGFR-1 as a sensitive andspecific biomarker in pregnancies complicated bypre-eclampsia. This may lead to better insights intothe treatment and prevention of severity ofdisease. However, a larger prospective study isrequired to further substantiate our findings.

Acknowledgement

This work was supported by a grant of Council ofScientific and Industrial Research (CSIR Project no.27-0133/04/EMR II), New Delhi, India. Ms. RichaTripathi is a recipient of the Senior ResearchFellowship of Indian Council of Medical Research(ICMR), New Delhi, India.

References

Ahmed, A., Dunk, C., Ahmad, S., Khaliq, A., 2000.Regulation of placental vascular endothelial growthfactor (VEGF) and placenta growth factor (PIGF) andsoluble Flt-1 by oxygen – a review. Placenta 21,S16–S24.

Aiello, L.P., Pierce, E.A., Foley, E.D., Takagi, H., Chen,H., Riddle, L., Ferrara, N., King, G.L., Smith, L.E.,1995. Suppression of retinal neovascularization in vivoby inhibition of VEGF using soluble VEGF-receptorchimeric proteins. Proc. Natl. Acad. Sci. USA 92,457–461.

Autiero, M., Lutton, A., Tjwa, M., Carmeliet, P., 2003.Placental growth factor and its receptor, vascularendothelial growth factor receptor-1: novel targetsfor stimulation of ischemic tissue revascularizationand inhibition of angiogenic and inflammatory dis-orders. J. Thromb. Haemostasis 1, 1356–1370.

Barleon, B., Reusch, P., Totzke, F., Herzog, C., Keck, C.,Martiny-Barron, G., Marme, D., 2001. Soluble VEGFR-1secreted by endothelial cells and monocytes is presentin human serum and plasma from healthy donors.Angiogenesis 4, 143–154.

Bdolah, Y., Sukhatme, V.P., Karumanchi, S.A., 2004.Angiogenic imbalance in the pathophysiology of pre-eclampsia: newer insights. Semin. Nephrol. 24 (6),548–556.

Bdolah, Y., Palomaki, G.E., Yaron, Y., Abram, T.B.,Goldman, M., Levine, R.J., Sachs, B.P., Haddow,J.E., Karumanchi, S.A., 2006. Circulating angiogenicproteins in trisomy 13. Am. J. Obstet. Gynecol. 194,39–45.

Bogic, L.V., Brace, R.A., Cheung, C.Y., 2001. Develop-mental expression of vascular endothelial growthfactors (VEGF) receptors and VEGF binding in ovineplacenta and fetal membranes. Placenta 22, 265–275.

Chaiworapongsa, T., Romero, R., Espinoza, J., Bujold, E.,Mee, K.Y., Goncalves, L.F., Gomez, R., Edwin, S.,2004. Evidence supporting a role for blockade of thevascular endothelial growth factor system in thepathophysiology of pre-eclampsia. Young InvestigatorAward. Am. J. Obstet. Gynecol. 190, 1541–1547.

Chaiworapongsa, T., Romero, R., Kim, Y.M., Kim, G.J.,Kim, M.R., Espinoza, J., Bujold, E., Goncalves, L.,Gomez, R., Edwin, S., Mazor, M., 2005. Plasma solubleVEGFR-1 concentration is elevated prior to the clinicaldiagnosis of pre-eclampsia. J. Maternal Fetal NeonatalMed. 17, 3–18.

Chaiworapongsa, T., Romero, R., Gotsch, F., Espinoza, J.,Nien, J.K., Goncalves, L., Edwin, S., Kim, Y.M., Erez,O., Kusanovic, J.P., Pineles, B.L., Papp, Z., Hassan, S.,2008. Low maternal concentrations of soluble vascularendothelial growth factor receptor-2 in pre-eclampsiaand small for gestation age. J. Maternal FetalNeonatal Med. 21 (1), 41–52.

Cheung, C.Y., Brace, R.A., 1999. Developmental recep-tors of vascular endothelial growth factors and itsreceptor in ovine placenta and fetal membranes.J. Soc. Gynecol. Invest. 6, 179–185.

Clark, D.E., Smith, S.K., Sharkey, A.M., Harnock Jones,D.S., 1996. Localisation of VEGF and expression of itsreceptors Flt and KDR in human placenta throughoutpregnancy. Hum. Reprod. (5), 1090–1098.

Cooper, J.C., Sharkey, A.M., McLaren, J., Charnock-Jones, D.S., Smith, S.K., 1995. Localization ofvascular endothelial growth factor and its receptor,

ARTICLE IN PRESS

Soluble and membranous vascular endothelial growth factor receptor-1 487

Flt in human placenta and deciduas by immunohis-tochemistry. J. Reprod. Fertil. 105 (2), 205–213.

Demir, R., Kayisli, U.A., Seval, Y., Celik Ozenci, C.,Korgun, E.T., Demir-Weusten, A.Y., Huppertz, B.,2004. Sequential expression of VEGF and its receptorsin human placental villi during very early pregnancy:differences between placental vasculogenesis andangiogenesis. Placenta 25, 560–572.

Dvorak, H.F., 2002. Vascular permeability factor/vascularendothelial growth factor: a critical cytokine intumour angiogenesis and a potential target fordiagnosis and therapy. J. Clin. Oncol. 20, 4368–4380.

Erez, O., Romero, R., Espinoza, J., Fu, W., Todem, D.,Kusanovic, J.P., Francesca Gotsch, Edwin, S., Nien,J.K., Chaiworapongsa, T., Mittal, P., Mazaki-Tovi1, S.,Than, N.G., Gomez, R., Hassan, S.S., 2008. Thechange in concentrations of angiogenic and anti-angiogenic factors in maternal plasma between thefirst and second trimesters in risk assessment for thesubsequent development of pre-eclampsia and small-for-gestational age. J. Maternal Fetal Neonatal Med.21 (5), 279–287.

Eskes, T.K., 1997. Abruptio placentae. A ‘‘classic’’dedicated to Elizabeth Ramsey. Eur. J. Obstet.Gynecol. Reprod. Biol. 75, 63–70.

Ferrara, N., Gerber, H.P., LeCouter, J., 2003. The biologyof VEGF and its receptors. Nat. Med. 9, 669–676.

Geva, E., Ginzinger, D.G., Zaloudek, C.J., Moore, D.H.,Byrne, A., Jaffe, R.B., 2002. Human placental vasculardevelopment: vasculogenic and angiogenic (branchingand nonbranching) transformation is regulated byvascular endothelial growth factor-A, angiopoietin-1and angiopoietin-2. J. Clin. Endocrinol. Metab. 87,4213–4224.

Girardi, G., Yarilin, D., Thurman, J.M., Holers, V.M.,Salmon, J.E., 2006. Complement activation inducesdysregulation of angiogenic factors and causes foetalrejection and growth restriction. J. Exp. Med. 203,2165–2175.

Helske, S., Vuorela, P., Carpen, O., Hornig, C., Weich, H.,Halmesmaki, E., 2001. Expression of vascular en-dothelial growth factor receptors 1, 2 and 3 inplacentas from normal and complicated pregnancies.Mol. Hum. Reprod. 7, 205–210.

Hinchey, J., Chaves, C., Appignani, B., Breen, J., Pao, L.,Wang, A., Pessin, M.S., Lamy, C., Mas, J.L., Caplan,L.R., 1996. A reversible posterior leukoencephalopa-thy syndrome. N. Engl. J. Med. 334, 494–500.

Hunter, A., Aitkenhead, M., Caldwell, C., McCracken, G.,Wilson, D., McClure, N., 2000. Serum levels ofvascular endothelial growth factor in pre-eclampticand normotensive pregnancy. Hypertension 36, 965.

Huppertz, B., Kingdom, J., Caniggia, I., Desoye, G.,Black, S., Korr, H., Kaufmann, P., 2003. Hypoxiafavours necrotic versus apoptotic shedding of placen-tal syncytiotrophoblast into the maternal circulation.Placenta 24, 181–190.

Jaeschke, R., Guyatt, G., Sackett, D., 1994. Users guideto the medical literature III: how to use an articleabout a diagnostic test. JAMA 271, 703–707.

Kearney, J.B., Kappas, N.C., Ellerstorm, C.,Dipaola, F.W., Bautch, V.L., 2004. The VEGF receptorflt-1 is a positive modulator of vascular sproutformation and branching morphogenesis. Blood 103,4527–4535.

Kendall, R.L., Thomas, K.A., 1993. Inhibition of VEGFactivity by an endogenously encoded soluble receptor.Proc. Natl. Acad. Sci. USA 90, 10705–10709.

Kim, S.Y., Ryu, H.M., Yang, J.H., Kim, M.Y., Han, J.Y.,Kim, J.O., 2007. Increased sFlt-1 to PlGF ratio inwomen who subsequently develop pre-eclampsia.J. Korean Med. Sci. 22 (5), 873–877.

Koga, K., Osuga, Y., Yoshino, O., Hirota, Y., Ruimeng, X.,Hirata, T., Takeda, S., Yano, T., Tsutsumi, O.,Taketani, Y., et al., 2003. Elevated serum solublevascular endothelial growth factor receptor-1(sVEGFR-1) levels in women with pre-eclampsia.J. Clin. Endocrinol. Metab. 88 (5), 2348–2351.

Levine, R.J., Maynard, S.E., Qian, C., Lim, K.H., Lucinda,J., Yu, K.F., Schisterman, E.F., Thadhani, R., Sachs,B.P., Epstein, F.H., Sibai, B.M., Sukhatme, V.P.,Karumanchi, S.A., 2004. Circulating angiogenic factorsand the risk of pre-eclampsia. N. Engl. J. Med. 350,672–683.

Levine, R.J., Lam, C., Yu, K.F., Maynard, S.E., Sachs,B.P., Sibai, B.M., Epstein, F.H., Romero, R., Thadhani,R., Karumanchi, S.A., 2006. Soluble endoglin andother circulating antiangiogenic factors in pre-eclampsia. N. Engl. J. Med. 355, 992–1005.

Li, H., Gu, B., Zhang, Y., Lewis, D.F., Wang, Y., 2005.Hypoxia-induced increase in soluble Flt-1 productioncorrelates with enhanced oxidative stress in tropho-blast cells from the human placenta. Placenta 26,210–217.

Marini, M., Vichi, D., Toscano, A., Zappoli Thyrion, G.D.,Parretti, E., Mello, G., Gher, G., Pacini, A., Sgambati,E., 2007. Expression of vascular endothelial growthfactor receptor types 1, 2 and 3 in placenta frompregnancies complicated by hypertensive disorders.Reprod. Fertil. Dev. 9, 641–651.

Masuyama, H., Suwaki, N., Nakatsukasa, H., Masumoto,A., Tateishi, Y., Hiramatrsu, Y., 2006. Circulatingangiogenic factors in pre-eclampsia, gestational pro-teinuria and pre-eclampsia superimposed on chronicglomerulonephritis. Am. J. Obstet. Gynecol. 194,551–556.

Matsubara, K., Nagamatsu, T., Fujii, T., Kozuma, S.,Taketani, Y., 2005. Lymphokine activated killer cellsinduced from decidual lymphocytes reduce the angio-genic activity of trophoblasts by enhancing the releaseof sFlt-1 from trophoblasts: an implication for thepathophysiology of pre-eclampsia. J. Reprod. Immu-nol. 68, 27–37.

Maynard, S.E., Min, J.Y., Merchan, J., Lim, K.H., Li, J.,Mondal, S., Libermann, T.A., Morgan, J.P., Sellke,F.W., Stillman, I.E., Epstein, F., Sukhatme, V.P.,Karumanchi, S.A., 2003. Excess placental solubleFms-like tyrosine kinase 1(sFlt-1) may contribute toendothelial dysfunction, hypertension and proteinuriain pre-eclampsia. J. Clin. Invest. 111, 649–658.

ARTICLE IN PRESS

R. Tripathi et al.488

Moore Simas, T., Crawford, S., Solitro, M., Frost, S.,Meyer, B., Maynard, S., 2007. Angiogenic factors forthe prediction of pre-eclampsia in high-risk women.Am. J. Obst. Gynecol. 197, 244.e1–244.e8T.

Munaut, C., Lorquet, S., Pequeux, C., Blacher, S.,Berndt, S., Frankenne, F., Foidart, J.M., 2008.Hypoxia is responsible for soluble vascular endothelialgrowth factor receptor-1 (VEGFR-1) but not for solubleendoglin induction in villous trophoblast. Hum.Reprod. 23, 1407–1415.

Nagamatsu, T., Fujii, T., Kusumi, M., Zou, L., Yamashita,T., Osuga, Y., Momoeda, M., Kozuma, S., Taketani, Y.,2004. Cytotrophoblasts up-regulate soluble fms-liketyrosine kinase-1 expression under reduced oxygen: animplication for the placental vascular developmentand the pathophysiology of preeclampsia. Endocrinol.145, 4838–4845.

Pandavala, S., Pope, N., Baker, P., Crocker, I., 2006. Animbalance between VEGF and its soluble receptor inplacental villous explants of intra uterine growthrestricted pregnancies. J. Soc. Gynecol. Invest. 13,40–47.

Powers, R.W., Roberts, J.M., Cooper, K.M., Galaher, M.,Frank, M.P., Harger, G.F., 2005. Maternal serumsoluble Fms like tyrosine kinase-1 concentration arenot increased in early pregnancy and decrease moreslowly postpartum in women who develop pre-eclampsia. Am. J. Obstet. Gynaecol. 193, 185–191.

Rajakumar, A., Brandon, H.M., Daftary, A., Ness, R.,Conrad, K.P., 2004a. Evidence for the functionalactivity of hypoxia-inducible transcription factorsoverexpressed in pre-eclamptic placentae. Placenta25, 763–769.

Rajakumar, A., Michael, H.M., Rajakumar, P.A., Shibata,E., Hubel, C.A., Karumanchi, S.A., Thadhani, R., Wolf,M., Harger, G., Markovic, N., 2004b. Extra-placentalexpression of vascular endothelial growth factorreceptor-1, (Flt-1) and soluble Flt-1 (sFlt-1), byperipheral blood mononuclear cells (PBMCs) in normo-tensive and pre-eclamptic pregnant women. Placenta26, 563–573.

Rana, S., Karumanchi, S.A., Levine, R.J., Venkatesha, S.,Rauh-Hain, J.A., Tamez, H., Thadhani, R., 2007.Sequential changes in antiangiogenic factors in earlypregnancy and risk of developing pre-eclampsia.Hypertension 50, 137–142.

Regnault, T.R., Orbus, R.J., de Vrijer, B., Davidsen, M.L.,Galan, H.L., Wilkening, R.B., Anthony, R.V., 2002.Placental expression of VEGF, PlGF and their receptorsin a model of placental insufficiency-intrauterinegrowth restriction (PI-IUGR). Placenta 23, 132–144.

Roberts, J.M., Cooper, D.W., 2001. Pathogenesis andgenetics of pre-eclampsia. Lancet 357, 53–56.

Roberts, J.M., Lain, K.Y., 2002. Recent insight into thepathogenesis of pre-eclampsia. Placenta 23, 359–372.

Robinson, C.J., Stringer, S.E., 2001. The splice variants ofvascular endothelial growth factor and their recep-tors. J. Cell Sci. 114, 853–865.

Rodesch, F., Simon, P., Donner, C., Jauniaix, E., 1992.Oxygen measurements in endometrial and trophoblas-

tic tissues during early pregnancy. Obset. Gynaecol.80, 283–285.

Salahuddin, S., Lee, Y., Vadnais, M., Sachs, B.P., Karuman-chi, S.A., 2007. Diagnostic utility of soluble Fms-liketyrosine kinase 1 and soluble endoglin in hypertensivediseases of pregnancy. AJOG 197, 28.e1–28.e6.

Stepan, H., Faber, R., Wessel, N., Wallukat, G.,Schultheiss, H.P., Walther, T., 2006. Relation betweencirculating angiotensin II type 1 receptor agonisticautoantibodies and soluble Fms-like tyrosine kinase 1in the pathogenesis of pre-eclampsia. J. Clin. En-docrinol. Metab. 91, 2424–2427.

Sugimoto, H., Hamano, Y., Charytan, D., Cosgrove, D.,Kieran, M., Sudhakar, A., Kalluri, R., 2003. Neutraliza-tion of circulating vascular endothelial growth factor(VEGF) by anti-VEGF antibodies and soluble VEGFreceptor-1 induces proteinuria. J. Biol. Chem. 278,12605–12608.

Taylor, R.N., Grimwood, J., Taylor, R.S., McMaster, M.T.,Fisher, S.J., North, R.A., 2003. Longitudinal serumconcentrations of PLGF: evidence for normal placentalangiogenesis in pathogenic pregnancies. Am. J. Obset.Gynaecol. 188, 177–182.

Tsatsaris, V., Goffin, F., Munaut, C., Brichant, J.F.,Pignon, M.R., Noel, A., Schaaps, J.P., Cabrol, D.,Frankenne, F., Foidart, J.M., 2003. Over expression ofsoluble endothelial growth factor receptor in pre-eclamptic patients: patho-physiological conse-quences. J. Clin. Endocrinol. Metab. 88, 5555–5563.

Unal, E., Robinson, C., Johnson, D., Chang, E., 2007.Second-trimester angiogenic factors as biomarkers forfuture-onset pre-eclampsia. Am. J. Obstet. Gynecol.197, 0s211.e1–211.e4 E.

Vatten, L.J., Eskild, A., Nilsen, T.I., Jeansson, S., Jenum,P.A., Staff, A.C., 2007. Changes in circulating level ofangiogenic factors from the first to second trimesteras predictors of pre-eclampsia. Am. J. Obstet.Gynecol. 196, 239–246.

Vuckovic, M., Ponting, J., Terman, B.I., Kikelie, V., Self,M.W., Kumar, S., 1996. Expression of VEGF receptorKDR in human placenta. J. Anat. 188, 361–366.

Vuorela, Piia, Halmesmaki, Erja, 2006. Vascular endothe-lial growth factor, its receptors, and the tie receptorsin the placental bed of women with pre-eclampsia,diabetes, and intrauterine growth retardation. Am.J. Perinatol. 23, 255–263.

Vuorela, P., Helske, S., Hornig, C., Alitalo, K., Herbert,W., Halmesmaki, E., 2000. Amniotic fluid-solublevascular endothelial growth factor receptor-1 in pre-eclampsia. Obstet. Gynaecol. 95, 353–357.

Wathen, K.A., Tuutti, E., Stenman, U.H., Alfthan, H.,Halmesma, E., Finne, P., Ylikorkala, O., Vuorela, P.,2006. Maternal serum-soluble vascular endothelialgrowth factor receptor-1 in early pregnancy endingin pre-eclampsia or intrauterine growth retardation.J. Clin. Endocrinol. Metab. 91, 180–184.

Yang, J.C., Haworth, L., Sherry, R.M., Hwu, P., Schwart-zentruber, D.J., Topalian, S.L., Steinberg, S.M., Chen,H.X., Rosenberg, S.A., 2003. A randomized trial of ananti-vascular endothelial growth factor antibody for

ARTICLE IN PRESS

Soluble and membranous vascular endothelial growth factor receptor-1 489

metastatic renal cancer. N. Engl. J. Med. 349,427–434.

Yang, Gu., Lewis, D.F., Wang, Y., 2008. Placentalproductions and expressions of soluble endoglin,soluble Fms-like tyrosine kinase receptor-1, andplacental growth factor in normal and pre-eclampticpregnancies. J. Clin. Endocrinol. Metab. 93, 260–266.

Zhou, Y., McMaster, M., Woo, K., Janatpour, M., Perry, J.,Karpanen, T., Alitalo, K., Damsky, C., Fisher, S.J.,2002. Vascular endothelial growth factor ligands andreceptors that regulate human cytotrophoblast survi-val are dysregulated in severe pre-eclampsia andhemolysis, elevated liver enzymes and low plateletsyndrome. Am. J. Pathol. 160, 1405–1423.