BIOLOGY OF REPRODUCTION 56, 489-494 (1997)
Expression of Vascular Endothelial Growth Factor and Placenta Growth Factor in Human Placenta' 5 5 Piia Vuorela,2,3 Erika Hatva,4 5 Athina Lymboussaki, 5 Arja Kaipainen, Vladimir Joukov,
Maria Graziella Persico,6 Kari Alitalo,5 and Erja Halmesmaki 3
Department of Obstetrics and Gynecology, 3 'Helsinki University Central Hospital, 00290 Helsinki, Finland Neurobiology Laboratory 4 and Molecular/Cancer Biology Laboratory, 5 University of Helsinki, SF-00014 Helsinki, Finland International Institute of Genetics and Biophysics,6 CNR, 12-80125 Naples, Italy creted, N-glycosylated homodimeric protein, and it potentiates the angiogenic activity of VEGF [9]. Abundant P1GF mRNA has been shown to be present in placenta by Northern blot analysis [10]. The function and precise location of P1GF mRNA and protein in placenta have not been determined. Different isoforms of VEGF and P1GF are expressed as a result of alternative RNA splicing. There are currently five known forms of VEGF, with 121, 145, 165, 189, 206 amino acid residues [11-13], and two isoforms of P1GF, with 131 and 152 amino acid residues (PlGF-1 and -2, respectively) [14]. VEGF and P1GF are known to form a VEGF/P1GF heterodimer, which is a highly potent endothelial cell mitogen [3]. In this study, we analyzed the site of VEGF and P1GF mRNA expression as well as the distribution of VEGF and P1GF protein and putative sites of action in normal human term placenta by in situ hybridization and immunohistochemistry.
ABSTRACT Normal development and function of the placenta requires invasion of the maternal decidua by trophoblasts, followed by abundant and organized vascular growth. Little is known of the significance and function of the vascular endothelial growth factor (VEGF) family, which includes VEGF, VEGF-B, and VEGF-C, and of placenta growth factor (PIGF) in these processes. In this study we have analyzed the expression of VEGF and PIGF mRNAs and their protein products in placental tissue obtained from noncomplicated pregnancies. Expression of VEGF and PIGF mRNA was observed by in situ hybridization in the chorionic mesenchyme and villous trophoblasts, respectively. Immunostaining localized the VEGF and PIGF proteins in the vascular endothelium, which was defined by staining for von Willebrand factor and for the Tie receptor tyrosine kinase, an early endothelial cell marker. VEGF-B and VEGF-C mRNAs were strongly expressed in human placenta as evidenced by Northern blot analysis. These data imply that VEGF and PIGF are produced by different cells but that both target the endothelial cells of normal human term placenta.
MATERIALS AND METHODS INTRODUCTION
Tissue Specimens
Intensive blood vessel growth comparable to that occurring in the placenta can be observed in only a few normal physiological processes such as ovulation [1], embryogenesis [2], and wound healing. Therefore, factors that induce and maintain placental vascular growth and function are of considerable developmental and clinical significance. Vascular endothelial growth factor (VEGF) is a major regulator of blood vessel growth and has also been shown to induce vascular permeability [3, 4]. VEGF knock-out experiments on mouse embryos resulted in malformed vasculature followed by early embryonic death [5, 6]. Recently two growth factors sharing structural similarity with VEGF-VEGF-B, and VEGF-C-have been characterized [7, 8]. VEGF-B is 42% homologous to VEGF at the amino acid level, and both VEGF-B and VEGF-C contain a conserved pattern of eight cysteine residues involved in the folding and dimerization of the growth factors of this family [7, 8]. Placenta growth factor (PlGF) is a member of the VEGF family of growth factors and is known to share 53% identity with the platelet-derived growth factor-homologous region of VEGE Like VEGF, P1GF is a se-
Placental tissue was obtained, after informed consent, from normal delivery by women with uncomplicated pregnancies. Immediately after delivery of the placenta, samples were collected for paraffin treatment for further in situ hybridization or were frozen in liquid nitrogen for further immunohistochemistry. For in situ analysis, tissues were fixed with 4% paraformaldehyde for about 20 h before dehydration and paraffin embedding. A total of five placentas were included in the study. In Situ Hybridization The VEGF probe was generated from a 581-base pair BamHI fragment (nucleotides 57-638) in the pGEM3Zf plasmid and the P1GF probe from a 640-base pair EcoRI fragment (nucleotides 304-944) in the same plasmid. To obtain the antisense probe template, the VEGF plasmid was linearized by HindIII restriction endonuclease (Boehringer Mannheim, Mannheim, Germany) digestion and purified. Antisense and sense P1GF templates were generated by linearizing and purifying the plasmid DNA with Pst I and Sty I restriction endonucleases (Boehringer Mannheim), respectively. Antisense VEGF radiolabeled RNA probe was synthesized via incorporation of [3 5S]UTP (Amersham International, Buckinghamshire, UK) by addition of SP6 polymerase (Promega Riboprobe Gemini Core II System; Promega Corporation, Madison, WI), and antisense and sense P1GF probes were synthesized by addition of SP6 and T7 polymerases (Promega
Accepted September 30, 1996. Received May 14, 1996. 'This study was supported by the Helsinki University 350th Anniversary Fund. 2Correspondence: Piia Vuorela, Department of Obstetrics and Gynecology, Helsinki University Central Hospital, Haartmaninkatu 2, 00290 Helsinki, Finland. FAX: 358-0-471 4801; e-mail:
[email protected]
489
490
VUORELA ET AL.
Riboprobe Gemini Core II System), respectively. The probes were then treated with ribonuclease-free DNase I (Boehringer Mannheim). The hybridization technique used was based on that described by Wilkinson et al. [15, 16] with the following modifications: 1) instead of toluene, xylene was used before embedding in paraffin wax; 2) 6-jLm sections were cut and then placed on a layer of diethyl pyrocarbonate-treated water on the surface of glass slides pretreated with 2% 3-aminopropyltriethoxysilane; 3) alkaline hydrolysis was omitted; 4) the hybridization mixture contained 50% deionized formamide; 5) two high-stringency washes, for 30 min and 90 min, were performed at 65°C in a solution containing 30 mM dithiotreitol and single-strength standard saline citrate (SSC; 0.15 M sodium chloride and 0.015 M sodium citrate). Autoradiography was carried out using NTB-2 emulsion (Eastman Kodak, Rochester, NY). The emulsioncoated slides were stored at 4°C with desiccant for an exposure time of 6 wk. The slides were developed in Kodak D 19 developer (Eastman Kodak), counterstained with Mayer's hematoxylin (Merck, Darmstadt, Germany), and mounted with Permount mounting medium (Fisher Chemicals, Springfield, MO). Hybridization with sense probe was used as a control to demonstrate signal specificity. Immunohistochemistry Cryostat sections, 6 jim, were cut onto poly-L-lysinecoated slides and stored without fixation at -20 0 C until analysis. The sections were stained immunohistochemically with the peroxidase-antiperoxidase method using rabbit polyclonal antibodies against human P1GF [17], mouse monoclonal antibodies against recombinant human VEGF 16 5 (R&D Systems Europe, Dorset, UK), mouse monoclonal antibodies against the extracellular domain of human Tie [18], and rabbit polyclonal antibodies against human von Willebrand factor (vWF; Dakopatts, Klostrup, Denmark). Sections were fixed in ice-cold acetone for 10 min and washed in PBS. Before incubation with the primary antibody, the sections were incubated with diluted normal swine serum for 30 min, normal horse serum for 30 min, Vectastain blocking serum (Vector Laboratories, Peterborough, UK) for 20 min, and normal swine serum for 30 min at room temperature (RT). The P1GF antibody (1: 500) was then added for 60 min at RT; VEGF antibody (1:500) overnight at 4°C; Tie antibody (1:500) overnight at 4C; and F VIII antibody (1:200) (Dakopatts) for 60 min at RT. After PBS washes, secondary antibodies were added onto the test slides: diluted (1:100) nonconjugated swine antibody against rabbit IgG (Dakopatts) onto the P1GF slides, diluted (1:300) biotinylated goat antibody against rabbit/mouse IgG (DAKO StreptABComplex/HRP Duet; Dakopatts) onto the VEGF slides, diluted (1:100) biotinylated goat antibody against rabbit/mouse IgG (Vectastain ABC Elite biotin-avidin system; Vector Laboratories) onto the Tie slides, and diluted (1:200) biotinylated swine antibody against rabbit IgG (Dakopatts) onto the vWF slides. The P1GF slides were then incubated at RT for 60 min; the VEGF, Tie, and vWF slides were incubated for 30 min. Peroxidase-antiperoxidase complex produced in rabbit (Dakopatts) was then added to the PIGF test slides for 45 min, Dako ABC (Dakopatts) reagent to the VEGF test slides for 30 min, Vectastain ABC reagent (Vector Laboratories) to the Tie test slides for 30
min, and Vectastain ABC reagent to the F VIII test slides for 45 min at RT. Bound antibody conjugates were visualized with 3-amino-9-ethylcarbazole, 0.03% H2 0 2, 14 mM acetic acid, and 33 mM sodium acetate. The test slides were counterstained with Mayer's hematoxylin (Merck) and mounted with Aquamount improved mounting medium (BDH Laboratory Supplies, Poole, Dorset, England). Normal serum, IgG, and PBS were used as negative controls. Northern Blotting and Hybridization Expression of VEGF, VEGF-B, and VEGF-C transcripts was analyzed by hybridization of a human multiple-tissue Northern blot (Clontech Laboratories, Palo Alto, CA). The probe contained a mixture of the 32 P-labeled fragment of human VEGF 16 5 cDNA (a kind gift from Dr. Daniel Connolly), and a labeled polymerase chain reaction fragment of nucleotides 1-382 of the human VEGF-B 1 65 [8] and of nucleotides 495-1661 of the human VEGF-C cDNA (Acc X94216) [7]. The probes do not cross-react, and they were labeled to approximately the same specific activity. The blot was hybridized overnight at 42°C using 50% formamide, 5-strength SSPE buffer (3 mM NaCl, 200 mM NaH 2PO 4, 20 mM EDTA, pH 7.0), 2% SDS, 10-strength Denhardt's solution, 100 mg/ml salmon sperm DNA, and single-strength 106 cpm/ml of each labeled probe. The blot was washed at RT for 30 min in double-strength SSC containing 0.05% SDS and then two times, for 20 min each, at 52°C in 0.1-strength SSC containing 0.1% SDS. The blot was then exposed at -70°C for 5 days using intensifying screens and Fuji RX film (Fuji, Tokyo, Japan). RESULTS The expression of VEGF, VEGF-B, and VEGF-C mRNAs was analyzed in human placenta RNA by Northern blotting and hybridization with radiolabeled cDNA probes. As can be seen from the results shown in Figure 1, placenta contained low levels of the 4.5- and 3.7-kilobase (kb) VEGF mRNAs and higher levels of the 2.4-kb VEGF-C and 1.4-kb VEGF-B mRNAs. For comparison, analysis was conducted on adult human heart and brain RNAs, which have been reported to express, respectively, abundant amounts of all three mRNAs, particularly of VEGF-B mRNA [7, 8]. Placental tissue, analyzed by in situ hybridization using the sense and antisense VEGF probes, showed a specific mRNA signal in cells of mesenchymal origin within the chorionic plate. No VEGF mRNA signal was observed in the villous structures (Fig. 2, A-B). In contrast, P1GF mRNA expression was observed outlining the villi, i.e., in villous trophoblasts, but not in the chorionic plate (C and D). Analysis of the placental tissue by immunohistochemistry showed strong positive staining of the blood vessel endothelium for VEGF protein. Little or no staining was observed in the villous stroma (Fig. 3, A and B). P1GF immunostaining showed a similar pattern but with more intense staining (C and D). The endothelial cell-specific receptor tyrosine kinase Tie showed clear positive staining lining the blood vessel endothelium. Tie staining was also observed in villous trophoblasts (Fig. 4). The endothelium was strongly positive for the endothelial cell marker vWF (data not shown). The chorionic plate and maternal decidual cells were negative for all of the proteins studied (for Tie staining, see Fig. 4A; for P1GF staining, see Fig. 3C).
P1GF AND VEGF IN HUMAN PLACENTA
491
FIG. 1. Expression of VEGF mRNAs in placenta. A Northern blot containing polyadenylated RNA from heart, brain, and placental tissue was hybridized with a pool of the VEGF, VEGF-B, and VEGF-C radioactively labeled probes and analyzed on x-ray film.
DISCUSSION In this study we demonstrated, by Northern blot analysis, the presence of the novel growth factors VEGF-B and VEGF-C in term human placenta. Our results show that these are expressed in excess in comparison to VEGE The precise location, role, and function of VEGF-B and VEGF-C within the placenta remain to be studied. We localized VEGF mRNA expression, by in situ hybridization analysis, to fetal-derived mesenchymal cells within the chorionic plate. An earlier study of VEGF mRNA expression in term placenta demonstrated VEGF mRNA expression in extravillous trophoblasts and in maternal and fetal-derived macrophages, as well as in the amnion [19]. In the present study, VEGF protein was immunohistochemically observed in the vascular endothelium of blood vessels within the villi. Some staining was also observed in the villous stroma. This finding is consistent with earlier results showing that VEGF was specific for vascular endothelium [20-22]. However, other studies on VEGF expression in term placenta have demonstrated positive VEGF immunoreactivity in villous cytotrophoblast [23] and syncytiotrophoblast [24, 25] as well as extravillous trophoblasts [25, 26]. These earlier findings are logical in view of mRNA for one of the two known VEGF receptors, VEGFR-1, in trophoblasts [27]. The discrepancy between the immunohistochemical findings of our investigation and earlier studies may be attributable to the use of different antibodies. In the present study we used a commercial monoclonal antibody against recombinant human VEGF, whereas three of the previous studies used a polyclonal antibody raised against a 20 amino acid peptide corresponding
FIG. 2. Detection of human VEGF and PIGF mRNAs in normal human term placenta by in situ hybridization. Dark- and brightfield views of sections hybridized with VEGF antisense probe (A, B) and PIGF antisense (C,D) and sense (E,F) probes are shown. Note that the VEGF mRNA signal is observed in mesenchymal cells of the chorionic plate (Ch), a layer of mesenchymal connective tissue giving rise to the placental villi (V), whereas the villi are negative for VEGF mRNA. PIGF mRNA signal is observed in trophoblasts aligning the villi, whereas the chorionic plate was negative for PIGF mRNA (not shown). Scale bar = 100 i.m.
to residues 1-20 at the amino terminus of VEGF [23-25]. However, one of the studies [24] used a commercial monoclonal antibody against recombinant human VEGF but from a different source than the one employed in the present study. Also, we cannot exclude the possibility of the existence of VEGF/P1GF heterodimers, which may crossreact with the antibody used. Our results show that VEGF mRNA is expressed in cells of mesenchymal origin within the chorionic plate and VEGF protein in vascular endothelium of villous blood vessels and to some extent in the villous stroma. This pattern of expression may imply a paracrine mode of action whereby VEGF produced by cells in the chorionic plate diffuses throughout the villi, accumulating at its presumed site of action, vascular endothelium. P1GF mRNA expression was observed in villous trophoblasts by in situ hybridization analysis. This result
492
VUORELA ET AL.
FIG. 3. Detection of VEGF and PIGF by immunoperoxidase staining in normal human term placenta. A section stained with the VEGF antibody is shown (A)with a corresponding negative control (B). Similar stainings for PIGF are shown in C and D, respectively. VEGF and PIGF protein can be observed lining the endothelial cells of blood vessels within villi (V). Little staining for VEGF was observed in the villous stroma. The chorion and the maternal decidua were negative for VEGF and PIGF; a low magnification of section including maternal decidua (d) and villi stained for PIGF is shown in the lower left corner of C. x330.
FIG. 4. Tie protein in normal human term placenta. Sections stained with Tie antibody (A) and its negative control (B)are shown. Tie protein expression occurs in the endothelial cells of blood vessels within the villi (V), shown here within a large villus. Tie immunostaining can also be observed in the trophoblasts aligning villi, shown here in smaller villi. The chorion and the maternal decidua are negative for Tie protein; a low magnification of a section including chorion (Ch) and villi, stained for Tie protein, is shown in the upper right corner of A. x190.
P1GF AND VEGF IN HUMAN PLACENTA
implies that P1GF is produced in trophoblasts, cells characterized by a variety of secretory action, as they secrete a wide range of proteins and hormones and therefore play an important endocrinologic and nutritional role [28, 29]. P1GF protein was immunohistochemically detected in vascular endothelium, implying that P1GF would diffuse across from the trophoblasts to its site of action in the vascular endothelium, and that P1GF could also act in a paracrine manner. It has been shown that the PIGF antibody cross-reacts with VEGF/PIGF heterodimers (unpublished results), and therefore, it is possible that our positive P1GF immunostainings are not entirely specific for the PIGF homodimer. Apart from this possible cross-reaction, our results suggest that both P1GF and VEGF target vascular endothelium of villous blood vessels but possibly via separate mechanisms. Due to the site of mRNA expression, PIGF could respond to stimuli from the maternal circulation and VEGF to stimuli within fetal tissues. In this study we used immunostaining of the endothelial cell-specific receptor Tie and F VIII to identify endothelial cells. Tie is a receptor tyrosine kinase crucial for vascular development, as Tie knock-out experiments with mouse embryos resulted in early embryonic death from hemorrhage [30, 31]. In this study, however, Tie protein did not appear to be restricted to vascular endothelium, as positive staining was also observed in villous trophoblasts. This new finding leads to speculation on the possible role of Tie in trophoblastic function, perhaps influencing the invasion of trophoblasts into the maternal decidua at implantation. Therefore, it would be interesting to define the cells expressing Tie mRNA in placenta. In this study the presence of the VEGF family of growth factors-VEGF, VEGF-B, VEGF-C, and P1GF-in term human placenta was demonstrated. These growth factors may form a complex regulatory system underlying placental vascular growth and maintenance. Inadequate blood vessel formation or maintenance in the placenta results in impaired placental growth, malfunction, hypoxia, and fetal malnutrition. Hypoxia may lead to impaired fetal growth and subsequent central nervous system damage, or even intrauterine death. Identification of the factors controlling the development, growth, and maintenance of placental vasculature may help to clarify the pathogenesis of poor placentation. In future, such data may be of potential use in the diagnosis and treatment of high-risk pregnancies with placental degeneration. ACKNOWLEDGMENTS We would like to thank Dr. Arto Orpana for expert advice on biochemistry. Professor Olavi Ylikorkala is thanked for providing excellent working facilities.
REFERENCES 1. Gaede SD, Sholley MM, Quattropani SL. Endothelial mitosis during the initial stages of corpus luteum neovascularization in the cycling adult rat. Am J Anat 1985; 172:173-180. 2. Sadler T. Special embryology, cardiovascular system. In: Langman's Medical Embryology, 5th ed. Baltimore: Williams & Wilkins; 1985: 168-214. 3. DiSalvo J, Bayne ML, Conn G, Kwok PW, Trivedi PG, Soderman DD, Palisi TM, Sullivan KA, Thomas KA. Purification and characterization of a naturally occurring vascular endothelial growth factor.placenta growth factor heterodimer. J Biol Chem 1995; 270:77177723. 4. Ferrara N, Houck K, Jakeman L, Leung D. Molecular and biological properties of the vascular endothelial growth factor family of proteins. Endocr Rev 1992; 13:18-22.
493
5. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoek J, Kendraprasad H, Eberhardt C, Declercq C, Pawling J, Moons L, Collen D, Risau W, Nagy A. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 1996; 380:435-439. 6. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O'Shea KS, Powell-Braxton L, Hillan KJ, Moore MW, Quattropanio SL. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996; 380:439-442. 7. Joukov V, Pajusola K, Kaipainen A, Chilov D, Lahtinen I, Kukk E, Saksela O, Kalkkinen N, Alitalo K. A novel vascular endothelial growth factor, VEGF-C, is the ligand for the Flt-4 (VEGFR-3) and KDR (VEGFR-2) tyrosine kinases. EMBO J 1996; 15:290-298. 8. Olofsson B, Pajusola K, Kaipainen A, von Euler G, Joukov V, Saksela O, Orpana A, Petterson R, Alitalo K, Eriksson U. Vascular endothelial growth factor B, a novel growth factor for endothelial cells. Proc Natl Acad Sci USA 1996; 93:2576-2581. 9. Park JE, Chen HH, Winer J, Houck KA, Ferrara N. Placenta growth factor. Potentiation of vascular endothelial growth factor bioactivity, in vitro and in vivo, and high affinity binding to Flt-i but not to FlkI/KDR. J Biol Chem 1994; 269:25646-25654. 10. Hauser S, Weich HA. A heparin-binding form of placenta growth factor (PIGF-2) is expressed in human umbilical vein endothelial cells and in placenta. Growth Factors 1993; 9:259-268. 11. Charnock-Jones DS, Sharkey AM, Rajput-Williams J, Burch D, Schofield JP, Fountain SA, Boocock CA, Smith SK. Identification and localization of alternately spliced mRNAs for vascular endothelial growth factor in human uterus and estrogen regulation in endometrial carcinoma cell lines. Biol Reprod 1993; 48:1120-1128. 12. Charnock-Jones DS, Sharkey AM, Boocock CA, Ahmed A, Plevin R, Ferrara N, Smith SK. Vascular endothelial growth factor receptor localization and activation in human trophoblast and choriocarcinoma cells. Biol Reprod 1994; 51:524-530. 13. Tischer E, Mitchell R, Hartman T, Silva M, Gospodarovicz D, Fiddes J, Abraham J. The human gene for vascular endothelial growth factor. Multiple protein forms are encoded through alternative exon splicing. J Biol Chem 1991; 266:11947-11954. 14. Maglione D, Guerriero V, Viglietto G, Ferraro MG, Aprelikova O, Alitalo K, Del Vecchios S, Lei K-J, Chou JY, Persico MG. Two alternative mRNAs coding for the angiogenic factor, placenta growth factor (PIGF), are transcribed from a single gene of chromosome 14. Oncogene 1993; 8:925-931. 15. Wilkinson DG, Bailes JA, Champion JE, McMahon AP A molecular analysis of mouse development from 8 to 10 days post coitum detects changes only in embryonic globin expression. Development 1987; 99: 493-500. 16. Wilkinson DG, Bailes JA, McMahon AP Expression of the protooncogene int-i is restricted to specific neural cells in the developing mouse embryo. Cell 1987; 50:79-88. 17. Viglietto G, Maglione D, Rambaldi M, Cerutti J, Romano A, Trapasso E Fedele M, Ippolito P, Chiappetta G, Botti G, Fusco A, Persico MG. Upregulation of vascular endothelial growth factor (VEGF) and downregulation of placenta growth factor (PIGF) associated with malignancy in human thyroid tumors and cell lines. Oncogene 1995; 11:15691579. 18. Batard P, Sansilvestri P, Scheinecker C, Knapp W, Debili N, Vainchenker W Buhring H-J, Monier M-N, Kukk E, Partanen J, Matikainen M-T, Alitalo R, Hatzfeld J, Alitalo K. The Tie tyrosine kinase is expressed by human hematopoietic progenitor cells and by a subset of megakaryocytic cells. Blood 1996; 87:2212-2220. 19. Sharkey AM, Charnock-Jones DS, Boocock CA, Brown KD, Smith SK. Expression of mRNA for vascular endothelial growth factor in human placenta. J Reprod Fertil 1993; 99:609-615. 20. Leung DW, Cachianes G, Kuang W-J, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246:1306-1309. 21. Keck P, Hauser S, Krivi G, Sanzo K, Warren T, Feder J, Connolly D. Vascular permeability factor, an endothelial cell mitogen related to PDGE Science 1989; 246:1309-1312. 22. Plouet J, Schilling J, Gospodarovicz D. Isolation and characterization of a newly identified endothelial cell mitogen produced by AtT-20 cells. EMBO J 1989; 8:3801-3806. 23. Ahmed A, Xiao F Li, Dunk C, Whittle JM, Rushton DI, Rollason T. Colocalisation of vascular endothelial growth factor and its Flt-i receptor in human placenta. Growth Factors 1995; 12:235-243. 24. Jackson MR, Carney EW, Lye SJ, Ritchie JW. Localization of two
494
VUORELA ET AL.
angiogenic growth factors (PDECGF and VEGF) in human placentae throughout gestation. Placenta 1994; 15:341-353. 25. Shiraishi S, Nakagava K, Kinukawa N, Nakano H, Sueshi K. Immunohistochemical localization of vascular endothelial growth factor in the human placenta. Placenta 1996; 17:111-121. 26. Cooper J, Sharkey A, McLaren J, Charnock-Jones D, Smith S. Localization of vascular endothelial growth factor and its receptor gflt, in human placenta and decidua by immunohistochemistry. J Reprod Fertil 1995; 105:205-213. 27. de Vries C, Escobedo J, Ueno H, Ferrara N, Williams L. The fmslike tyrosine kinase, a receptor for vascular endothelial growth factor. Science 1992; 255:989-991.
28. Ross M, Reith E. Female reproductive system. In: Histology: A Text and Atlas. New York: Harper & Row Publ. Inc.; 1985: 681-689, 722725. 29. Talamentes A, Ogren L. The placenta as an endocrine organ: polypeptides. In: The Physiology of Reproduction, vol. 2. New York: Raven Press Ltd.; 1988: 2093-2144. 30. Puri M, Rossant J, Alitalo K, Bernstein A, Partanen J. The receptor tyrosine kinase TIE is required for the integrity and survival of vascular endothelial cells. EMBO J 1995; 14:5884-5891. 31. Sato TN, Tozawa Y, Deutsch U, Wolburg-Buchholz BK, Fujiwara Y, Gendron-Maguire MM, Gridley T, Wolburg H, Risau W, Qin Y. Distinct roles of the receptor tyrosine kinases Tie-l and Tie-2 in blood vessel formation. Nature 1995; 376:70-74.
