1 Department of Obstetrics and Gynecology, 2 Department of Clinical Chemistry, Helsinki University Central Hospital, Haartmaninkatu 2, FIN-00290 Helsinki and 3 Molecular/Cancer Biology Laboratory, Haartman Institute, University of Helsinki, Finland
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: amniotic fluid/PlGF/pregnancy-associated binding/VEGF
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Two specific endothelial cell receptors, VEGF receptor-1 (VEGFR-1) or fms-like tyrosine kinase-1 (flt-1) (de Vries et al., 1992) and VEGFR-2 or kinase insert domain-containing receptor (KDR) (Terman et al., 1994
), are known to bind VEGF, and a soluble form of VEGFR-1 has been identified (Kendall and Thomas, 1993
; Kendall et al., 1996
). VEGFR-1 also binds PlGF with high affinity (Park et al., 1994
). Serum VEGF concentrations have earlier been reported to rise during the first trimester of pregnancy (Evans et al., 1998
), but it is not known to what extent the VEGF receptors or other possible factors regulate the bio-availability of VEGF and PlGF during early or later pregnancy.
The presence of some, so far unknown, gestational factors regulating angiogenesis seems obvious considering the strong local neovascularization occurring in the placenta and the placental bed. The original aim of this study was to measure VEGF and PlGF in amniotic fluid and serum during pregnancy, and to identify possible factors regulating their activity.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Amniotic fluid
A pool of 50 random amniotic fluid samples from gestational weeks 1517 was prepared from samples submitted for assessment of fetal chromosomal abnormalities and assay of -fetoprotein (AFP). Eleven additional individual samples of amniotic fluid were collected during Caesarean section of healthy mothers at term. Samples were stored at 20°C until analysis.
Non-pregnant subjects
Single blood samples were collected from 22 healthy non-pregnant, non-medicated and non-smoking women (median age 29 years, range 1739 years). As serum VEGF concentrations have been shown to correlate with serum progesterone concentrations (Evans et al., 1998), all the samples were collected at the same phase of the menstrual cycle, i.e. the follicular phase.
Although no diurnal variation of urinary VEGF concentrations has been observed in gonadotrophin-treated women (Robertson et al., 1995), all blood samples were collected between 8 and 9 a.m. Following separation, serum was frozen and stored at 20°C until analysis.
Immunoassays of VEGF and PlGF
VEGF and PlGF were quantified by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instructions [Quantikine recombinant human (rh)VEGF and Quantikine rhPlGF, R&D Systems Europe Ltd, Abingdon, UK]. The detection limit of the assays was 16 ng/l. Recovery of rhVEGF and rhPlGF was tested by the ability of the ELISA to detect added rhVEGF (200 ng/l) or rhPlGF (100 ng/l) (proteins were provided as standard proteins in the ELISA) in various samples. The inter- and intra-assay coefficients of variation were 6.5 and 8.5% (Quantikine rhVEGF) and 6.3 and 7.9% (Quantikine rhPlGF). Recovery of added VEGF was also studied in serial dilutions of amniotic fluid and maternal serum samples from early and term pregnancy. According to the manufacturer, rhVEGF and rhP1GF ELISA show ~ 20% and 5% cross-reactivity respectively, with the naturally occurring PlGF/VEGF heterodimer (DiSalvo et al., 1995). No cross-reactivity of either ELISA was observed with the novel growth factors VEGF-B (Olofsson et al., 1996
) and VEGF-C (Joukov et al., 1996
) when tested at concentrations 0.5750 ng/l.
The possible influence of heparin on the ability of amniotic fluid to inhibit VEGF immunoreactivity in ELISA was studied by first incubating amniotic fluid with VEGF (see below) for 1 h at room temperature, after which heparin (0.055 mg/ml; Løvens Kemiske Fabrik, Ballerup, Denmark) was added to the samples and incubation was continued for 1 h at room temperature. VEGF concentrations were then measured by ELISA.
Immunodiffusion
The reactivity of the VEGF-binding factor with antibodies against 2-macroglobulin (anti-
2M, 7.2 g/l; DAKO, Denmark), pregnancy zone protein (anti-PZ, 5.8 g/l; DAKO) and pregnancy-associated plasma protein-A (anti-PAPP-A, 4.2 g/l; DAKO) was studied by immunodiffusion in 1% agarose gels. Samples of amniotic fluid and serum from pregnant women at term and from non-pregnant women were studied, either directly or following 1 h incubation with [125I]VEGF at room temperature. A total of 10 µl of antibody and sample were pipetted into the wells and incubated for 48 h at +4°C. Gels were viewed for immunoprecipitates at 24 and 48 h, dried, and, after drying, subjected to autoradiography (X-Omat®; Eastman Kodak Company, Rochester, NY, USA).
Time-resolved immunofluorometric assay (IFMA)
The concentration of 2M in amniotic fluid was determined by an immunofluorometric method (IFMA) as described previously (Leinonen et al., 1996
). The detection limit of the assay was 5 µg/l.
Radio-iodination of VEGF
Five µg of recombinant human VEGF165 (Genzyme Diagnostics, Cambridge, MA, USA) was radiolabelled using Iodo Gen® (Pierce, Rockford, IL, USA). Sixty-one µl phosphate buffered saline (PBS; 0.14 µl NaCl, 2.7 µl KCl, 0.01 µl Na2HPO4, 1.76 µl KH2PO4, pH 7.4), 4 µl [125I]Na (2 mCi) and 5 µl of rhVEGF in 25 µl of PBS were pipetted into a test tube coated with 200 µl of dichloromethane containing 40 mg/l Iodo Gen®. After 15 min 10 µl of 0.01 µl KI was added. The labelled protein was separated from free iodine by gel filtration on a 1.5x5 cm Sephadex® G-25 column (PD-10; Pharmacia Fine Chemicals, Uppsala, Sweden) in PBS containing 1% bovine serum albumin (BSA; Sigma Chemical Co., St Louis, MO, USA). Specific activity was 36 000 c.p.m./ng protein.
Gel filtration
A Sephacryl® S-300 HR (Pharmacia Biotech, Uppsala, Sweden) column (84x1.5 cm; Pharmacia LKB Biotechnology, Uppsala, Sweden) was equilibrated with tris-buffered saline (TBS; 0.05 mol/l TrisHCl, 0.15 mol/l NaCl, pH 7.7). The exclusion volume of the column was 40 ml, flow rate was 15 ml/h (Microperpex® peristaltic pump, Pharmacia LKB Biotechnology) and 1 ml fractions were collected (Redifrac®, Pharmacia LKB Biotechnology). Thyroglobulin 669 kDa (Pharmacia Fine Chemicals), ferritin 440 kDa (Pharmacia Fine Chemicals), human immunoglobulin G (IgG) 168 kDa (Sigma Chemical Co.), bovine serum albumin 67 kDa (Sigma Chemical Co.), soybean trypsin inhibitor 20.1 kDa (Sigma Chemical Co.) were used as molecular weight markers.
Two ml of amniotic fluid pool was fractionated on the column. Five consecutive fractions were pooled, and recovery of VEGF in the pooled fractions was analysed by ELISA. Pooling of fractions was done to reduce expense. Despite this, it was found that the accuracy of the analysis still allowed verification of the VEGF binding activity eluting in fractions similar to those of the high molecular weight complexes observed in gel filtration analysis with [125I]VEGF. [125I]VEGF in 200 µl TBS was loaded onto the column alone or after pre-incubation for 1 h at room temperature, with 200 µl of amniotic fluid, serum from maternal, umbilical and non-pregnant samples. The fractions were monitored for absorbance at 280 nm (Lambda 3B UV/VIS Spectrophotometer®; Perkin-Elmer, Überlingen, Germany) and radioactivity (1260 Multigamma gamma counter®; LKB, Wallac, Sweden).
In another experiment 200 µl of amniotic fluid and sera were separated on the column. Recombinant VEGF was added to the fractions (200 ng/l) and the recovery was measured by ELISA.
The effect of low pH and high salt concentration on the VEGF-binding complex was studied with a 50x1 cm column packed with Sephacryl® S-300 HR and equilibrated with TBS. The flow rate was 15 ml/h and 0.5 ml fractions were collected. [125I]VEGF was incubated with amniotic fluid at room temperature for 1 h and loaded onto the column (200 µl) as such or following acidification to pH 2 with HCl for 15 min and neutralization with NaOH. Then [125I]VEGF incubated with amniotic fluid at room temperature for 1 h was separated at pH 2 without neutralization. In other experiments 200 µl of [125I]VEGF pre-incubated with amniotic fluid for 1 h at room temperature was loaded onto the column which had been equilibrated with either 0.5, 1, 2, 3, 4 mol/l NaCl or with 4 mol/l KSCN.
Isoelectric focusing
NOVEX Precast Gel®, pH 310 (Novel Experimental Technology, San Diego, CA, US), was used for estimation of the isoelectric point of the complex studied according to the manufacturer's instructions. A total of 1 ng of [125I]VEGF in 4 µl TBS-1%BSA was pipetted into the wells following 1 h incubation at room temperature with 9 µl of either PBS, amniotic fluid or amniotic fluid with added 1 ng, 4 ng or 16 ng of unlabelled rhVEGF (Genzyme®), or the VEGF-binding fractions of amniotic fluid or pregnancy serum from early or term pregnancy. These VEGF-binding fractions were obtained from separation of samples by gel filtration in Sephacryl® S-300 HR columns, as described above.
Statistics
Between-group comparisons were performed using Student's unpaired t-test.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
When [125I]VEGF was incubated with amniotic fluid and then separated by gel filtration on a column (S-300 HR, 50x1 cm) equilibrated with 4 mol/l KSCN the elution profile was similar to that of [125I]VEGF alone. When [125I]VEGF incubated with amniotic fluid was separated in the same column using NaCl at concentrations of 0.54 mol/l in PBS, [125I]VEGF was not dissociated from the complex. Furthermore, the ability of amniotic fluid to shift the radioactive peak to high molecular weight fractions was retained after acidification of the complex to pH 2 and neutralization. However, when the [125I]VEGF complex was pre-incubated and separated at pH 2 without neutralization, no shift of the radioactive peak to the high molecular weight fractions was observed. Incubation with various concentrations of heparin before analysis in ELISA did not release VEGF from the complex in amniotic fluid.
In immunodiffusion with antibodies against 2M, PZ and PAPP-A no precipitation lines with amniotic fluid were observed, whether studied alone or after pre-incubation with [125I]VEGF. Serum of pregnant women at term showed precipitation lines with anti-
2M and anti-PZ, but not with anti-PAPP-A. Serum from non-pregnant women showed precipitation lines with anti-
2M, but not with anti-PAPP-A or anti-PZ. Autoradiography revealed bands corresponding to the precipitation lines for anti-
2M, whereas no bands were seen with PZ or PAPP-A (data not shown). By IFMA no
2M was observed in amniotic fluid.
In isoelectric focusing [125I]VEGF added to amniotic fluid, or to the VEGF-binding fractions of amniotic fluid or serum, displayed an isoelectric point of approximately 8. Addition of increasing amounts of non-radioactive rhVEGF caused dissociation of [125I]VEGF, which then focused in the pH range 45 (Figure 3).
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The absence of detectable VEGF immunoreactivity in maternal serum in early and term pregnancy, followed by a post-partum rise in serum VEGF concentrations, suggested the presence of a pregnancy associated factor suppressing VEGF immunoreactivity in ELISA. Measurement of the recovery of VEGF added to serum showed the presence of a factor which probably bound VEGF. This activity was also detected in amniotic fluid at much higher concentrations. Gel filtration of [125I]VEGF added to pregnancy serum and amniotic fluid showed that this binding activity occurred in the high molecular weight fractions.
In early pregnancy, the VEGF-binding capacity was about 50 times higher in amniotic fluid than in maternal serum. The binding capacity in term amniotic fluid was two-fold that of early amniotic fluid. The origin and source, as well as the exact nature of the binding compound need further studies, but from the above data it can be concluded that the binding activity is strongly associated with gestation and dependent on gestational age. The high concentrations of the binding factor in amniotic fluid suggest that it originates from the amniotic fluid compartment. Interestingly, in gel filtration the complex formed in early amniotic fluid showed two distinct components, whereas term amniotic fluid contained only a single high molecular weight complex corresponding to the later eluting compound in early amniotic fluid.
The identity of the VEGF-binding compound is not yet known, but well known proteins of the corresponding molecular weight can be excluded. 2M, which is a major protease inhibitor in serum, has been shown to bind VEGF irreversibly (Soker et al., 1993
), and in the immunodiffusion studies some radioactivity was found to be associated with
2M, but this could not be detected in amniotic fluid. Assays of
2M by IFMA confirmed that it is not detected in amniotic fluid (Bhat and Pattabiranam, 1980
). Heparin did not affect VEGF binding in amniotic fluid or serum of pregnant women, whereas it has been reported to inhibit its binding by
2M (Soker et al., 1993
). These results indicate that
2M is not responsible for the VEGF binding activity in amniotic fluid. However, on the basis of our immunodiffusion results,
2M binds some VEGF in serum of non-pregnant subjects, and it probably contributes to a minor part of the binding activity in maternal serum.
The serum concentrations of two high molecular weight proteins, PZ and PAPP-A have been shown to increase with advancing pregnancy, reaching peak at term, and declining during the post-partum days (von Schoultz, 1974; Lin et al., 1976
). PAPP-A is also present in amniotic fluid (Bischof et al., 1982
). We therefore studied the possible role of these proteins in VEGF-binding activity. In immunodiffusion neither PZ nor PAPP-A reacted with complexed VEGF. Furthermore, the molecular weight and isoelectric point (pI) of the complex, approximately 440 kDa and pI approximately 8, differ from those of PAPP-A, Mr 820 kDa (Sinosich et al., 1980
) and pI 4.4 (Lin et al., 1976
) and from the molecular weight of PZ, Mr 326 kDa (von Schoultz, 1974
).
We studied whether the soluble form of VEGFR-1 (Kendall and Thomas, 1993; Kendall et al., 1996
; Banks et al., 1998
) would be responsible for the binding of VEGF by measuring the recovery of PlGF. VEGFR-1 binds PlGF with high-affinity (Park et al., 1994
) but no inhibition of PlGF immunoreactivity in ELISA was observed. Therefore it seems unlikely that soluble VEGFR-1 is responsible for the VEGF-binding activity observed in amniotic fluid. Furthermore, the molecular size of the soluble VEGFR-1 is only about 90 kDa (Kendall and Thomas, 1993
).
The nature of binding of VEGF to the complex in amniotic fluid was characterized by studying the ability of high salt, low pH, heparin and the chaotropic salt KSCN to release [125I]VEGF from the complex. [125I]VEGF was not released from the complex either by high concentrations of NaCl or by heparin, but 4 mol/l KSCN and low pH caused dissociation of the complex. Taken together, these results indicate that VEGF is bound strongly by an acid stable protein. In amniotic fluid, the binding protein is heterogeneous in molecular size. Such a size heterogeneity can be due to dimerization, but the existence of two separate binding proteins cannot be excluded. Interestingly, the dimeric 2M molecule forms a tetramer when it reacts with proteases. However, VEGF is not known to exert proteolytic activity and protease-
2M complexes are covalent.
It has recently been suggested that VEGF-binding protein or proteins occur in maternal serum and that such binding proteins were also present in blood from non-pregnant subjects and the umbilical vein, but were saturated to leave unbound VEGF available for detection by ELISA (Anthony et al., 1994). The results of this study suggest that the VEGF-binding activity of umbilical cord and post-partum maternal samples is low, and that most of the VEGF is unbound.
In conclusion, a putative heterodimeric high molecular weight VEGF-binding protein is present in maternal serum and amniotic fluid. Its concentrations are dependent on gestational age, it disappears after delivery, and it is not clearly detectable in umbilical blood. Because it does not bind PlGF, it does not appear to be the presently known soluble VEGF receptor. Further studies are needed to identify this VEGF-binding compound and to elucidate its potential role in the regulation of vasculogenesis of human pregnancy.
![]() |
Acknowledgments |
---|
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Banks, R., Forbes, M., Searles, J. et al. (1998) Evidence for the existence of a novel pregnancy associated soluble variant of the vascular endothelial grown factor receptor, Flt-1. Mol. Hum. Reprod., 4, 377386.[Abstract]
Bhat, A. and Pattabiranam, T. (1980) Comparative study of proteinase inhibitors in human amniotic fluid and maternal serum. Brit. J. Obstet. Gynecol., 87, 11091114.[ISI][Medline]
Bischof, P., Duberg, S., Herrman, W. et al. (1982) Amniotic fluid and plasma concentrations of pregnancy associated plasma protein-A (PAPP-A). Brit. J. Obstet. Gynecol., 89, 358363.[ISI][Medline]
de Vries, C., Escobedo, J., Ueno, H. et al. (1992) The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science, 255, 989991.[ISI][Medline]
DiSalvo, J., Bayne, M.L., Conn, G. et al. (1995) Purification and characterization of a naturally occurring vascular endothelial growth factor. placenta growth factor heterodimer. J. Biol. Chem., 270, 77177723.
Evans, P., Wheeler, T., Anthony, F. et al. (1998) A longitudinal study of maternal serum vascular endothelial growth factor in early pregnancy. Hum. Reprod., 13, 10571062.[Abstract]
Gospodarovicz, D., Abraham, J. and Schilling, J. (1989) Isolation and characterization of a vascular endothelial cell mitogen produced by pituitary-derived folliculo stellate cells. Proc. Natl Acad. Sci. USA, 86, 73117315.[Abstract]
Joukov, V., Pajusola, K., Kaipainen, A. et al. (1996) A novel vascular endothelial growth factor, VEGF-C, is the ligand for the Flt-4 receptor tyrosine kinase (VEGFR-3). EMBO J., 15, 290298.[Abstract]
Kaipainen, A., Korhonen, J., Pajusola, K. et al. (1993) The related FLT4, FLT1, and KDR receptor tyrosine kinases show distinct expression patterns in human fetal endothelial cells. J. Exp. Med., 178, 20772088.[Abstract]
Keck, P., Scott, D., Krivi, G. et al. (1989) Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science, 246, 13091312.[ISI][Medline]
Kendall, R. and Thomas, K. (1993) Inhibition of vascular endothelial growth factor activity by an endogenously encoded soluble receptor. Proc. Natl Acad. Sci. USA, 90, 1070510709.[Abstract]
Kendall, R., Wang, G. and Thomas, K. (1996) Identification of a natural soluble form of the vascular endothelial growth factor receptor, FLT-1, and its heterodimerization with KDR. Biochem. Biophys. Res. Commun., 226, 324328.[ISI][Medline]
Ladoux, A. and Frelin, C. (1993) Hypoxia is a strong inducer of vascular endothelial growth factor mRNA expression in the heart. Biochem. Biophys. Res. Commun., 195, 10051010.[ISI][Medline]
Leinonen, J., Zhang, W-M. and Stenman, U-H. (1996) Complex formation between PSA isoenzymes and protease inhibitors. J. Urol., 155, 10991014.[ISI][Medline]
Leung, D.W., Cachianes, G., Kuang, W.J. et al. (1989) Vascular endothelial growth factor is a secreted angiogenic mitogen. Science, 246, 13061309.[ISI][Medline]
Lin, T., Halbert, S., Spellacy, W. et al. (1976) Human pregnancy-associated plasma proteins during the post partum period. Am. J. Obstet. Gynecol., 124, 382387.[ISI][Medline]
Maglione, D., Guerriero, V., Viglietto, G. et al. (1991) Isolation of a human placenta cDNA coding for a protein related to the vascular permeability factor. Proc. Natl Acad. Sci. USA, 88, 92679271.[Abstract]
Olofsson, B., Pajusola, K., Kaipainen, A. et al. (1996) Vascular endothelial growth factor B (VEGF-B), a novel growth factor for endothelial cells. Proc. Natl Acad. Sci. USA, 93, 25762581.
Orlandini, M., Marconcini, L., Ferruzzi, R. et al. (1996) Identification of a c-fos-induced gene that is related to the platelet-derived growth factor/vascular endothelial growth factor. Proc. Natl Acad. Sci. USA, 93, 1167511680.
Park, J.E., Chen, H.H., Winer, J. et al. (1994) Placenta growth factor. Potentiation of vascular endothelial growth factor bioactivity, in vitro and in vivo, and high affinity binding to Flt-1 but not to Flk-1/KDR. J. Biol. Chem., 269, 2564625654.
Robertson, D., Selleck, K., Suikkari, A-M. et al. (1995) Urinary vascular endothelial growth factor concentrations in women undergoing gonadotrophin treatment. Hum. Reprod., 10, 24782482.[Abstract]
Sharkey, A.M., Charnock, J.D., Boocock, C.A. et al. (1993) Expression of mRNA for vascular endothelial growth factor in human placenta. J. Reprod. Fertil., 99, 609615.[Abstract]
Sinosich, M., Sim, R. and Teisner, B. (1980) Characterization of pregnancy-associated plasma protein-A. Biochem. Int., 20, 579589.
Soker, S., Svahn, C. and Neufeld, G. (1993) Vascular endothelial growth factor is inactivated by binding to alpha2-macroglobulin and the binding is inhibited by heparin. J. Biol. Chem., 268, 76857691.
Terman, B., Khandke, L., Dougher, V.M. et al. (1994) VEGF receptor subtypes KDR and FLT1 show different sensitivities to heparin and placenta growth factor. Growth Factors, 11, 187195.[ISI][Medline]
von Schoultz, B. (1974) A quantitative study of the pregnancy zone protein in the sera of pregnant and puerperal women. Am. J. Obstet. Gynecol., 119, 792797.[ISI][Medline]
Vuorela, P., Hatva, E., Lymboussakis, A. et al. (1997) Expression of vascular endothelial growth factor and placenta growth factor in human placenta. Biol. Reprod., 56, 489494.[Abstract]
Yamada, Y., Nezu, J., Shimane, M. et al. (1997) Molecular cloning of a novel vascular endothelial growth factor, VEGF-D. Genomics, 42, 483488.[ISI][Medline]
Submitted on October 14, 1998; accepted on February 2, 1999.