Phenotype of the Endothelium in the Human Term Placenta

Placenta (2001), 22, 32–43 doi:10.1053/plac.2000.0579, available online at http://www.idealibrary.com on

Phenotype of the Endothelium in the Human Term Placenta J. F. Dyea,b, R. Jablenskab, J. L. Donnellyb, L. Lawrenceb, L. Leachc, P. Clarkb and J. A. Firthb b Leukocyte Biology, Division of Biomedical Sciences, Imperial College School of Medicine, London and c School of Biomedical Sciences, Faculty of Medicine and Health Sciences, Queen’s Medical Centre, University of Nottingham, UK Paper accepted 7 June 2000

The placental endothelium contributes to regulating transplacental exchange and maintaining the immunological maternofetal barrier. We characterized the endothelial phenotype in human normal term placentae with a panel of antibodies to endothelial antigens using a standardized immunofluorescence method. Placental endothelium strongly expressed vWF, PAL-E, H-antigen, thrombomodulin, PECAM-1, CD34, CD36, ICAM-1, CD44, thy-1, A10/33-1, VE-cadherin, caveolin-1 and HLA-G, whereas occludin, claudin-1, eNOS, angiotensin converting enzyme (ACE), ICAM-2, endoglin and integrin-
3 were weakly expressed. PGI2 synthase, tissue factor, E-selectin and VCAM-1 were not detected. Some antigens were heterogenously expressed along the vascular tree or within individual villi. Expression of ACE, eNOS, vWF, P-selectin, E-selectin, integrin
v3 and endoglin was stronger in the maternal decidual vessels, while PECAM-1, CD44, thy-1 and caveolin-1 expression was stronger in fetal vessels. Some endothelial markers were present in trophoblasts and stroma. Endothelial proliferation was apparent in mature intermediate and terminal villi. There was limited inflammatory response to TNF
in explants, characterized by upregulation of vWF, P-selectin, PECAM-1 and CD44, downregulation of thrombomodulin, but no increase in ICAM-1 expression, nor induction of E-selectin, VCAM-1 or tissue factor. These patterns of heterogeneity, proliferative activity and inflammatory activation may underlie the specific physiological roles of the placental endothelium.  2001 Harcourt Publishers Ltd Placenta (2001), 22, 32–43

INTRODUCTION Maternofetal exchange across the placenta is widely considered to be limited by the syncytiotrophoblast layer, while the role of the fetal capillary endothelium has perhaps been overlooked. However, it is clear that the placental endothelium does play a role in restricting exchange (Firth and Leach, 1996) and may also be involved in maintaining the immunological barrier provided by the placenta (Simister and Story, 1997). The specialized functions typical of the vascular endothelium (e.g. regulation of transendothelial permeability, blood flow, clotting and inflammation) are reflected by the expression of specific proteins. These proteins may then be used to characterize the endothelial phenotype (Table 1). Although these markers are considered definitive of endothelial cell phenotype, considerable heterogeneity in structure and function is recognized (Milner, Loesch and Burnstock, 1997). Differences occur between tissues [e.g. (Belloni, Carrey and Nicolson, 1992)], between large vessels and capillaries [e.g. (Kumar, West and Ager, 1987)] and between individual endothelial cells in a tissue [e.g. (Diglio et al., 1988; Rupnick, Carrey and Williams, 1988; Furuya et al., 1990; Spanel Borowski and Fenyves, a

To whom correspondence should be addressed at: Leukocyte Biology, Division of Biomedical Sciences, Sir Alexander Fleming Building, Imperial College School of Medicine, South Kensington, London SW7 2AZ, UK. Fax: +44 (0)171 594 3269; E-mail: [email protected]

0143–4004/01/010032+12 $35.00/0

1994)]. Whether these are determined early, at the stage of vasculogenesis, or whether they arise during terminal differentiation of cells in response to their local tissue environment (Aird et al., 1997), is unknown. These levels of heterogeneity make it necessary to evaluate the endothelial phenotype within each tissue in order to develop an understanding of its differentiated functions. This may be particularly so in the placenta, since histological evidence suggests that, unusually, the placental endothelial cells differentiate from mesenchymal precursor cells as they assemble into capillaries (Demir et al., 1989). There is also evidence of differences between placental microvascular endothelial cells and umbilical vein endothelial cells in vitro, in response to growth factors and expression of markers (Dye et al., 1996). Some characteristics of the placental endothelium are well established. Much of this work relates to specific transport properties, such as for immunoglobulin via FcRII (Sedmak et al., 1991; Lang et al., 1993; Bright et al., 1994) and transporter proteins such as glucose transporters and transferrin (Lang et al., 1993). There is also increasing interest in the expression of growth factor receptors (Barleon et al., 1994; Ahmed, 1997; Kaipainen, 1993 #282). Previous studies have shown the expression of certain ‘typical’ endothelial markers in placental endothelium (vWF:RAg, thrombomodulin, PECAM-1, CD34 and VE-cadherin) (Lang et al., 1993; Leach et al., 1993; Leach et al., 1994) although there is a paucity of  2001 Harcourt Publishers Ltd

Dye et al.: Phenotype of the Placental Endothelium

33

Table 1. Phenotypic profiling of endothelial cells Function

System

Anticoagulant

Thrombomodulin/protein C/protein S PGI2 synthesis Extrinsic pathway initiation VIII activation, platelet adhesion Platelet adhesion Renin/angiotensin system NO system Tight Junction

Pro-coagulant, thrombogenesis Vasopressor Vasodilator Homotypic adhesion paracellular barrier Leukocyte binding

Marker

Adherens Junction

Immune-related Antigen presentation/adhesion

H-antigen HLA system

Other markers

Scavenger receptor TGF co-receptor Angiogenic integrin Quiescence Proliferation

QB/END30 QB/END40 PGI2 synthase Tissue Factor vWF:RAg P-selectin (CD62-P) ACE eNOS Occludin Claudin-1 VE-cadherin E-selectin (CD62-E) P-selectin (CD62-P) ICAM-1 (CD54), ICAM-2 (CD102) VCAM-1 (CD106) PECAM-1 (CD31)a CD34? CD44a? thy-1? UEA-1 HLA B/C HLA-G? CD36? endoglin (CD105)
v3 (CD51/61) PAL-E PCNA

‘?’ denotes uncertain or multiple functions; a also show homotypic adhesion.

literature to support or augment this work. The placental endothelium is also known to express at least some vasoregulator components (Myatt et al., 1993; Eis et al., 1995; Shellhaas et al., 1997). Functional changes in the placental endothelium have been implicated in pathological pregnancies (Ghabour et al., 1995; Myatt et al., 1997). Therefore, characterization of the role of this endothelium is important and requires an appreciation of its differentiated phenotype. In addition, although the placenta expands throughout gestation (Benirschke and Kaufmann, 1995), cellular mechanisms of angiogenesis in the placenta are poorly understood and where cell division occurs is not clear. The aim of this study was to systematically characterize the phenotype of the placental endothelium in normal, term pregnancies, using a panel of antibodies to endothelial cell antigens (Table 1), with a standardized immunochemical technique. The main focus was the phenotype of the placental microvessels, and comparisons were made with larger vessels within stem villi and with vessels of maternal origin, within the decidua basalis. Staining for proliferating cell nuclear antigen (PCNA) was also performed to assess the proliferation of endothelial cells in term placentae. Since pro-inflammatory cytokines cause profound effects on the expression of response genes, changing the endothelial pheno-

type, we investigated the responses of placental endothelium to TNF
using an explant model. Some of these data have been published previously in abstract form (Dye, Leach and Firth, 1995; Dye et al., 1997).

MATERIALS AND METHODS Tissue preparation Human placentae, delivered electively by caesarian section, were collected from the Labour ward, St Mary’s Hospital NHS Trust, Paddington, London, within 5 min of delivery, and used with consent. Tissue specimens of placental villi and decidua were excised from the basal surface of the placenta (approximately 1 cm3), washed in PBS, then immersion fixed for approx 1–2 h in 4 per cent paraformaldehyde/PBS at 0–4C, washed in PBS, cut into 0.5 cm3 blocks for cryomicrotomy, immersed in OCT (Merk, Poole, UK) and frozen in isopentane (cooled to freezing point with liquid nitrogen). Some samples of tissue were immersion-fixed in parallel, with Mirsky’s fixative (National Diagnostics, Hull, UK), a commercial non-volatile aldehyde formulation, claimed to be suitable for greater antigen preservation. Frozen sections were cut at 7–8 m onto poly--lysine coated glass slides (three per slide).

Placenta (2001), Vol. 22

34

Table 2. Suppliers of primary antibodies Marker

Clone/ID

ab class

Source

vWF ACE (CD143) PGI2 synthase eNOS

F8 (21-43) 4057 160630

mIgG1 mIgG1 mcab r-poly mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG1 mIgG2a mIgG2b mIgG1 mIgG1 mIgG2a mIgG2a mIgG1 mIgG1 mIgG1 mIgG1 mIgG mIgG1 gIgG r-poly r-poly mIgG2a mIgG1 mIgM r-poly mIgG1 IgG2a

Serotec Chemicon Cayman Santa Cruz Transduction Labs R&D DSHB Serotec Sanbio Serotec DSHB Serotec Serotec Serotec DSHB DSHB Serotec Serotec Serotec Biogenesis Biogenesis Serotec Serotec Pharmingen Dr M. Pignatelli Takara Santa-Cruz Zymed Zymed Serotec Serotec Serotec Santa-Cruz Serotec Dr P. Le Bouteiller

mIgG1

Serotec

CD31 CD34 CD36 ICAM-1 (CD54) CD62E CD62P CD102 CD105 CD106 CDw90 panCD44 HUVEC-derived tissue factor (CD142) thrombospondin-related CD51/61 VE-cadherin(CD144) E-cadherin P-cadherin N-cadherin occludin claudin-1 thrombomodulin (CD141) Kidney-related caveolin-1 PCNA HLA-G -ve IgG

N30020(3) 9G11(BBA7) P2B1 QB/END 10 VM58 84H10 P2H3 1.2B6 AK-6 B-T1 P3D1 P8B1 1.4c3 F15-42-1 F10-44-2 PAL-E II A10-33-1 23C6 55-7H1 HECD NCC-CAD-299 sc-1502 71-1500 71-7800 QB/END 40 QB/END 30 QB/END 20 sc-894 PC10 BFL-1 87G B-Z1

Reagents The primary antibodies used in the study are listed in Table 2. Other reagents were as follows: goat-anti-mouse f(ab)2 biotin conjugate and extravidin-FITC (Sigma); goat-antimouse f(ab)2-Cy3 conjugate and streptavidin-Cy3 (Caltag, Burlinghame, CA, USA); Biotinylated UEA-1 (Vector labs, Peterborough, UK); hrTNF
(Innogenetics, Ghent, Belgium); M199, Earles salts, with bicarbonate and hepes (Gibco, Paisley, UK); FBS (Harlan Seralab, Loughborough, UK).

fixed immediately (0 hour control) or transferred to culture medium (M199 plus 5 per cent FBS) either with or without 100 U/ml hrTNF
. This concentration of TNF
causes reorganization of interendothelial adhesion molecules in perfused placentae (Leach et al., 1995), and also causes maximal upregulation of leukocyte adhesion molecules in endothelial culture systems (Haraldsen et al., 1996). The explants were incubated in a humidified 5 per cent CO2 atmosphere at 37C (for 4 h or 24 h) before fixation as described. Explants maintained their viability, as assessed by trypan-blue exclusion, for the duration of the incubations. Sections from control and treated tissue were immunostained in parallel.

TNF treatment The response of the placental endothelium to the proinflammatory cytokine TNF
was assessed in placental explant culture. Placental explants from four placentae (approx 0.2 cm3, 2–3 per group) were taken, washed in PBS and either

Immunohistochemistry A standardized procedure was followed using freshly made reagents at pH 7.4 and room temperature. Frozen sections were allowed to thaw and dry. They were permeabilized with 0.5 per cent Triton-X100 in PBS for 5 min, then washed twice

Frozen sections of other adult human tissues (skin, nasal polyp) were used as controls for some antibodies.

Dye et al.: Phenotype of the Placental Endothelium

in PBS and blocked in block buffer (5 per cent BSA/PBS) for 30 min. Sections were briefly washed and primary reagents were added in 5 per cent BSA/PBS at 1 : 100 dilution, for 1–2 h at room temperature or overnight at 4C. Sections were then twice rinsed briefly and washed three times for 5 min in wash buffer (0.1 per cent BSA/0.1 per cent Triton-X100/ PBS). Secondary and tertiary reagents were similarly used at 1 : 50 to 1 : 100 dilution in block buffer, and incubated with the sections for 1 h, followed by the same washing procedure. Finally, sections were washed twice in PBS then twice in water, before being mounted with Immunofluor (ICN). In the case of PCNA staining and evaluation of TNF, tissue was doubled-labelled with UEA-1, to identify the endothelium, and a specific antibody. Appropriate negative controls were run in parallel (unreactive mIgG for murine monoclonal antibodies, purified rabbit Ig for rabbit polyclonals and detection reagent alone for UEA-1, relaced primary antibodies). Positive controls for most antibodies were established with cultured human endothelial cells. Adult human tissue was used as an additional positive control (for CD31, CD44, thy-1 and PGI2 synthase). Epifluorescence microscopy and data collection Sections were examined on a Zeiss Axiophot epifluorescence microscope equipped with a mercury vapour lamp and filter sets for FITC and TRITC, as well as phase contrast interference optics. Sections were examined by switching between the epifluorescence and phase contrast modes to confirm the localization of markers. Paired fluorescence/phase images were photographed onto TMAX400 film (Kodak). Intensities of each marker in the endothelium were scored on a scale of 0 (negative) to 4 (intense), in comparison to UEA-1 staining which was scored 4. The number of maternal vessels in the decidual basalis varies widely between blocks but at least 20 profiles were examined for each antigen. Since PCNA-positive nuclei are relatively rare (even in proliferating endothelium), their frequency was determined from 10 selected microscopic fields of intermediate and terminal villi, selected randomly under illumination for UEA-1 fluorescence, in each specimen. This method enabled selection of fields with clear endothelial profiles, without bias for antigen staining. The results were expressed as positive nuclei per number of vessel profiles. Effects of TNF treatment were similarly evaluated by photographic comparison of at least four fields of terminal villi, randomly selected under illumination for UEA-1, using a fixed exposure time (and uniform print conditions) for each marker. In addition, comparisons of fluorescent intensity on sections were made independently by several observers. RESULTS Expression and localization of endothelial markers in placental villi and decidua basalis The expression and distribution patterns of the markers by the endothelium of fetal microvessels and maternal vessels in

35

the placenta are summarized in Table 3 and illustrated for some in Figure 1. Each marker was examined in at least four placentae and consistent patterns of staining were found. Immunoreactivity in the placental endothelium was found for many of the proteins in Table 1; vWF, thrombomodulin (CD141), CD62-P, eNOS, ACE (CD143), VE-cadherin (CD144), occludin, claudin-1, PECAM-1 (CD31), CD34, CD44, thy-1 (CDw90), ICAM-1 (CD54), ICAM-2 (CD102), CD36, endoglin (CD105), caveolin-1, and the histological markers PAL-E and A10/33-1. Additionally, UEA-1 binding was specific for the endothelium, and was co-localized with endothelial markers such as PECAM-1, PAL-E and A10/33-1 (not shown). PGIS reactivity was observed in adult nasal polyp endothelium and in cells within the decidua basalis, but not in placental endothelium (not shown). Immunostaining of tissue factor was found only in stromal cells. eNOS, vWF, P-selectin, occludin and claudin-1 were expressed more intensely in arteries and veins of the villous vascular tree, whereas CD44, ICAM-1 and thy-1 expression were strongest in the sinusoidal capillaries. No differences in the endothelial staining intensity along the vascular tree were found for other markers (UEA-1, thrombomodulin, PECAM, CD34, CD36, PAL-E, A10-33-1, VE-cadherin). Marked variation within or between individual villi in the staining was found for some antigens (CD36, P-selectin, occludin, claudin-1, endoglin), (shown for P-selectin in Figure 3, 0 hour control). Comparison between the endothelial phenotype in the fetal microvessels and the maternal vessels of the decidua basalis showed quantitative differences. ACE, eNOS, vWF, P-selectin, E-selectin, integrin
v3 (CD51/61) and endoglin were expressed more intensely in maternal vessels than fetal. PECAM-1, CD44 and caveolin-1 were expressed more intensely in fetal microvessels. Several endothelial markers were positive in cells covering parts of the inner decidual surface [presumptive vascular trophoblasts (Kaufmann and Castellucci, 1997)]. In addition, the expression of markers in villous pericytes (stromal cells), syncytiotrophoblasts (STB), extravillous trophoblasts (EVT) and decidual stromal cells were recorded (Table 3). No differences in immunostaining reactions for several antigens in tissue fixed with Mirsky’s fixative compared to paraformaldehyde were detected, excepting occludin. In paraformaldehyde-fixed tissue, occludin was only observed in larger vessels. Staining was enhanced in Mirsky-fixed tissue, where a relatively weak punctate stain in the capillaries could be observed.

Proliferative activity of chorionic microvascular endothelium Proliferating endothelial cells were observed in intermediate and terminal villi but not in vessels of stem villi. These cells were identified by positive staining for PCNA, labelling cells in the G1-S phase of the mitotic cycle (Iatropoulos and Williams, 1996) (Figure 2 and Table 4). From random sampling of 544

Placenta (2001), Vol. 22

36

Table 3. Endothelial phenotype in human caesarean-section term placenta Marker

Fetal vessels

Maternal vessels

Vascular trophoblast

Tissue distribution

Cellular distribution

H-Antigen (UEA-1)

++++

++++

+++

ENDO, VT

thrombomodulin (CD141) tissue factor (CD142) ACE (CD143) PGI2 synthase

+++
+


+++
+ + locally


+++




VT>ENDO villous stroma STB & VT>ENDO EVT & STB

eNOS

+

+ (locally)




EVT & STB>ENDO

QB/END 20 PAL-E


++++


++++


++


ENDO, VT

A10-33/1 vWF P-selectin (CD62P)

+++ + + (locally+ +)

+++ +++ ++++

++ +++ ++

ENDO, VT VT(intense)>ENDO ENDO, VT>STB

E-selectin (CD62E) ICAM-1 (CD54) ICAM-2 (CD102) VCAM-1 (CD106) CD34 CD44 Thy-1 (CDw90) PECAM-1 (CD31)


+++ +
++++ +++ ++ ++++

+ ++ ?+
++++ ?+ + +++

+ ++


++++
++

STB, stroma (weak) ENDO, VT ENDO villous stroma (weak) ENDO VT, PC & VSMC>ENDO VT, PC & VSMC>ENDO ENDO, VT

Lumenal cell membranes>basolateral membranes punctate cytoplasmic stain Lumenal cell membrane Cell membranes Fine punctate stain of whole cell Punctate, esp in basal aspect of CTB Punctate stain, and in large vessel EC, more diffuse
Cell membranes and punctate cytoplasmic stain, endothelial basement membrane deposits Punctate cytoplasmic stain Punctate cytoplasmic stain Heterogenous punctate stain, similar to vWF Punctate stain Lumenal cell membranes Cell membranes

CD51/61 Cadherin 5 (CD144) E-cadherin N-cadherin P-cadherin occludina claudin-1 HLA-G


++


+ (locally+ +) + (locally + + +) +++

+ ++


+
++











STB>ENDO ENDO STB
STB (weak) ENDO, STB ENDO, STB, decidual stroma EVT, ENDO>pericytes

CD36 (SR-B1)

+ (locally+ + +)

++




ENDO, DM stroma

endoglin (CD105) caveolin-1

+ (locally+ +) ++++

+ +++


+

STB, VT & DM>ENDO ENDO, VSMC (not pericytes)

Lumenal membrane and clefts Cell membranes Diffuse Lumenal membrane>basal membrane Basal aspect of CTB Clefts and ablumenal membrane Clefts Weak diffuse staining of EC Clefts Clefts, punctate in STB Clefts, punctate Cell membrane in EVT and PC, cytoplasmic in ENDO Lumenal membrane (some villi weak or negative) Cell membranes (esp STB) Dense punctate cytoplasmic stain

The relative intensity of marker staining from a uniform procedure was assessed by direct visual comparison of samples (+ =weak, + + + + =strong). ENDO: endothelium; PC: pericytes; VSMC: vascular smooth muscle cells; STB: syncytiotrophoblast layer; DM: decidual membrane; VT: vascular trophoblasts (internal surface of DM). a Reactivity in microvessels detected only in Mirsky-fixed tissue.

vessel profiles, in 10 fields per placental sample, the median value of 8.8 per cent (range 6.3–16.5 per cent) PCNA positive nuclei per vessel profile was found (in four placentae). This contrasted with the endothelial cells in stem villi and in maternal vessels of the decidua, in which no PCNA positivity was found by random sampling, and where extensive inspection of tissue revealed very few positive cells. However, PCNA positive cells were found in the trophoblast layer, the villous stroma and the fibrinous stroma of the decidua (not shown).

Regulation of markers by TNF There were consistent changes with TNF incubation in staining intensity of several markers (vWF, P-selectin, PECAM-1, CD44, ICAM-1, E-selectin, VCAM-1, thrombomodulin, tissue factor, endoglin) in the placental endothelium, in comparison to control tissue fixed at time zero or incubated without TNF (n=4) (Table 5 and Figure 3). In the case of P-selectin and vWF, the increased expression was local, reflecting its heterogenous expression in control tissue. There

Dye et al.: Phenotype of the Placental Endothelium

37

Figure 1. Immunofluorescent localization of representative endothelial markers in fetal capillary endothelium of human term placenta. Frozen sections of fixed tissue were immunostained for fluorescent microscopy as described in the methods. (A) UEA-1; (B) UEA-1/phase-contrast double exposure; (C) CD36; (D) PAL-E; (E) A10-33/1; (F) caveolin-1; (G) endoglin (CD105); (H) ACE; (I) negative immunoglobulin control. In A-F, staining is localized to the endothelium, whereas in G & H, although there is strong expression in the syncytiotrophoblast layer, definitive staining of the endothelium is present (arrows). Bar=20 m.

was no immunostaining for E-selectin, VCAM-1 or tissue factor in the placental endothelium even of explants which were incubated with TNF for 24 h (not shown).

DISCUSSION In this study we characterize the fetal endothelium in the pre-labour term placenta using an immunocytochemical approach, with markers for a wide range of endothelial antigens, as outlined in Table 1. The expression of many of these markers has been previously reported in various contexts,

but the present study compared immunohistochemical reactivity with a standardized immunofluorescence technique. Expression patterns of ACE, CD36, A10-33/1, P-selectin, intergrin
v3, CD44, Thy-1 and caveolin-1 in the placenta have not been previously examined. Some of the present findings confirm other data, and provide detail about the spatial distribution and tissue specificity of markers. For example, expression of HLA-G in placental endothelium has been described (Le Bouteiller et al., 1999), but the present study with fluorescence microscopy resolved a distinct expression in pericytes and vascular smooth muscle cells as well as in endothelium.

38

Placenta (2001), Vol. 22

Figure 2. Double-labelled placental villous endothelium, demarked with UEA-1(A), detected with Z-avidin-FITC (green), showing PCNA immunoreactivity (B), detected by goat-anti-mouse-Cy3 conjugate (red). Superimposition of red and green fluorescence (C) shows an example of a PCNA positive nucleus (arrow) within the endothelium, which occurs at a relatively high frequency in intermediate and terminal villi, but not in stem villi. Bar=20 m.

Markers expressed by placental endothelium The pattern of expression of endothelial antigens we find in the placental endothelium in some ways reflects a ‘typical’ endothelial profile. The binding of biotinylated UEA-1 lectin was specific to the endothelium, and is suitable for identifying the endothelium in double-labelling with antibody reagents. We were unable to confirm a previous report of UEA-1 binding to fibroblasts, vascular smooth muscle cells and fibrinoid in the villous stroma and decidua (Lang et al., 1993). Some of the markers examined can be associated with the status of the endothelium (e.g. quiescent, activated, angiogenic). CD34 is distributed lumenally on quiescent endothelium but during angiogenesis it is localized on basolateral sprouting projections, suggesting a role in angiogenic migration (Fina et al., 1990; Schlingemann et al., 1990). Its expression and lumenal distribution in placental microvascular endothelium, found here and reported previously (Lang et al., 1993), conflicts with other data (Fina et al., 1990). This may be explained by an alternative antigenic form of CD34 to the type III form identified by QB/END10 in the present study. The scavenger receptor, CD36, is a multifunctional receptor which can inhibit angiogenesis (Dawson et al., 1997). PAL-E is a marker generated from HUVEC, indicative of ‘differentiated endothelium’ (Schlingemann et al., 1985) and its expression in placental microvascular endothelium confirms a previous report (Lang et al., 1993). By contrast, A10-33/1 is not regarded as a constitutive marker, and has been described in proliferating, transformed and malignant endothelial cells, as well as in some normal endothelial cells (Bruggen and Sorg, 1983; Suter et al., 1983; Nahler, 1985). Its global expression in placental endothelium indicates that the antigen has a ‘normal’ placental function. A10/33-1 is claimed to block the interaction of endothelial cells with thrombospondin (Bachem catalogue, Bachem, Saffron Walden, UK). The failure to identify prostaglandin I2 synthase, a vasodilatory mediator, in placental vessels conflicts with previous immunohistochemical evidence describing its expression in

chorionic villous endothelium rather than syncytiotrophoblasts (Wetzka et al., 1996; Shellhaas et al., 1997). The absence of integrin
v3 from placental endothelium confirms previous data (Leach et al., 1993) and is notable, since it is expressed by many other endothelia, on both lateral (Lampugnani et al., 1991) and abluminal surfaces (Albelda, 1991). Its expression is required for sprouting-type angiogenesis (Brooks et al., 1994).

Heterogeneous expression of endothelial markers The differences we observed between fetal and large uteroplacental vessels in the decidua were mainly quantitative rather than qualitative. The markers expressed more strongly in the maternal vessels, such as vWF, P-selectin, ACE and eNOS, are typical of large vessels, while those stronger in fetal vessels may represent a ‘microvascular’ phenotype. This is supported by our finding that CD44 and thy-1, which are expressed in placental capillary endothelium, were present in dermal microvessels. A role of placental microvessels in macromolecular transport is suggested by their strong expression of caveolin-1, the major structural protein of caveolae (Schnitzer, 1997). The greater expression of vWF and eNOS found in the large vessel endothelium of stem villi concurs with known patterns generally described for vWF (Kumar, West and Ager, 1987) and for eNOS in the placenta (Myatt et al., 1993; Eis et al., 1995). This data supports previous descriptions of differences between placental microvessels and umbilical vessels, notably for immunoglobulin FcRII (Lang et al., 1993). Local variations in the expression of some proteins were found between placental microvessels (microheterogeneity), particularly for P-selectin, occludin and claudin-1, even within a single villus. These patterns have not been described previously and imply local variations in microenvironment or different sub-types of endothelial cell. Heterogenous expression of endoglin, a TGF- receptor subunit, could be

Dye et al.: Phenotype of the Placental Endothelium

0 h control

39

4 h control

4 h TNF

PECAM-1

P-selectin

CD44

ICAM-1

TM

E-selectin

Figure 3. Effect of TNF treatment on endothelial marker expression in placental explants. Placental tissue from caesarean-section term placentae was either fixed immediately or incubated at 37C in culture medium with or without hrTNF
(100 U/ml) for 4 h before fixation. Frozen sections were immunostained and fluorescence micrographs were recorded at uniform exposure for each marker: (A) PECAM-1 (6 sec); (B) P-selectin (20 sec); (C) CD44 (15 sec); (D) ICAM-1 (10 sec); (E) thrombomodulin (TM) (15 sec); (F) E-selectin (20 sec). Endothelial expression is increased in A–C, show little change in D, is decreased in E and is weak or undetectable in F. The distribution of P-selectin is markedly heterogenous within the microvascular endothelium (arrows), but is increased in intensity with TNF treatment. CD44 expression in the endothelium (arrows) is normally weaker than in surrounding mesenchymal cells, and leukocytes (LC), but endothelial expression is increased with TNF. Bar=20 m.

due to local activation since its expression is known to occur in activated endothelial cells (Westphal et al., 1993). The present findings of microheterogeneity may explain the conflicting reports of endoglin expression in placental endothelium (Gougos et al., 1992; St Jacques et al., 1994).

Proliferative activity of the placental endothelium One way in which the placental endothelium differs from that of other organs is its cellular proliferation rate. In the mouse, the turnover time of placental endothelium, measured by

Placenta (2001), Vol. 22

40

Table 4. Random field measurement of PCNA expression in the endothelium of intermediate and terminal chorionic villi

Placenta no 1 2 3 4

No vessel profiles from 10 fields 109 159 154 122 (total=544)

PCNA+ve 18 10 12 12

%PCNA+ve/no profiles 16.5 6.29 7.79 9.83 (median=8.81)

Table 5. The effects of 4 hrTNF exposure on endothelial markers Marker

Effect of TNF


PECAM-1 P-selectin vWF CD44 cadherin 5 Endoglin ICAM-1 Thy-1 thrombomodulin E-selectin VCAM-1 tissue factor

increase increase increase increase no change/increase no change no change/decrease no change/decrease decrease weak/absent absent absent

thymidine incorporation, of around 2–3 days, contrasts with the endothelium in mature organs (turnover times between 60 and 10 000 days) (Denekamp, 1984). Proliferative activity has been measured in human placentae by thymidine uptake into explants, and by labelling for proliferation markers, but endothelial cells were not identified and therefore not quantified in this study (Kosanke et al., 1998). Using UEA-1 double-labelling to identify the endothelium, our findings indicate high rates of endothelial cell proliferation in the human term placenta, occurring within and limited to intermediate and terminal villi. This finding suggests expansion of existing vessels is an important mechanism, rather than ‘sprouting’ angiogenesis, where proliferation occurs in migratory endothelial cells (expressing
v3) forming new vessel loops (Varner and Cheresh, 1996). This degree of proliferation may also relate to the microheterogenous distribution patterns described above.

Homotypic adhesion molecules of the endothelium The barrier function of the endothelium, arguably a key cellular property, is determined by the cell–cell adhesion molecules which form adherens junctions and tight junctions. VE-cadherin is the main adhesion molecule of endothelial

adherens junctions (Lampugnani et al., 1992), but there is evidence that several cadherin isoforms can be expressed in a single cell type (Salomon et al., 1992). In this study we have confirmed our previous findings, showing the endothelium specifically expresses VE-cadherin, which is localized at endothelial cell–cell contacts, but not N-cadherin (Leach et al., 1993). E-cadherin expression was found only in the syncytiotrophoblast layer, and no discrete staining was found for Por N-cadherin. Evidence that the tight junction adhesion molecules occludin and claudin-1 are expressed heterogenously in placental microvascular endothelium suggests that other molecules are likely to be involved in forming tight junctions in the placenta. We observed two distribution patterns of PECAM-1 were found in different vessels; either uniformly distributed over the cell surface as in activated endothelial cells (Romer et al., 1995) or concentrated in endothelial cell clefts.

Leukocyte–endothelial adhesion molecules The expression profile of leukocyte adhesion molecules in the placental endothelium was unusual. The strong expression of ICAM-1 is indicative of endothelial activation, yet neither E-selectin nor VCAM-1 were detected, in eight specimens (see below), and the punctate distribution of vWF and P-selectin is consistent with its localization in Weibel–Palade bodies. Also, the expression of ICAM-2, which is generally constitutive in resting endothelium (de Fougerolles et al., 1991) was only expressed weakly. These findings, from term placentae prelabour, support previous data showing expression of ICAM-1 and ICAM-2, but not E-selectin or VCAM-1, in placental endothelium (Lyall et al., 1995; Steinborn et al., 1999), but do not support the finding of E-selectin expression in proliferating endothelial cells in the placenta (Kraling et al., 1996). In addition, the fetal endothelium expressed two other molecules with roles in cell adhesion interactions: CD44, which is usually described as a non-endothelial marker (Flanagan et al., 1989), and thy-1, found to be expressed in activated endothelium (Fivenson et al., 1992). CD44 expression has previously been reported (using monoclonal antibody F10441) in trophoblasts but not in placental endothelium (Flanagan et al., 1989). Using the same antibody and two other monoclonal antibodies to CD44 we observed localization in mesenchymal and endothelial cells, but not in the trophoblast layer. These findings are corroborated by in vitro expression of CD44 and thy-1 in umbilical vein and placental microvascular endothelial cells (Dye et al., 1996).

Regulation of markers by TNF
In the endothelium the induction of the leukocyte adhesion receptors, ICAM-1, E-selectin and VCAM-1 by proinflammatory cytokines such as TNF is an important inflammatory response. Because of the strong expresssion of ICAM-1 in the absence of E-selectin and VCAM-1 in the

Dye et al.: Phenotype of the Placental Endothelium

placental endothelium, we investigated whether an inflammatory stimulus could upregulate these inducible adhesion molecules. The genes for these molecules are strongly suppressed unless activated by NF-B. Induction of E-selectin is typically rapid and transient, peaking at around 4 h, whereas VCAM-1 expression is more sustained, peaking at 24 h (Mackay et al., 1993). In addition, tissue factor is induced by TNF in endothelial cells after 2–4 h (Reverdiau et al., 1995; Conway and Rosenberg, 1998). However, the placental endothelium showed an unusually limited expression of ‘classical’ reponses to TNF over 4 h, with little further change after 24 h. This implies a low potential for leukocyte recruitment during inflammation. This behaviour is unlikely to be explained by absence of TNF receptors, since both TNFr-I and TNF-rII have been immunolocalized to placental endothelium (Austgulen et al., 1992; Steinborn et al., 1998). We also found evidence for some ‘typical’ responses to the TNF treatment in this explant model. TNF normally stimulates rapid secretion of vWF and release of P-selectin into the cell membrane from Weibel–Palade bodies (Giddings, 1990; Ewenstein, 1997). The TNF-induced changes in their distribution in the placental endothelium we observed are consistent with this, although increased expression has not been previously described. Similarly, increased expression of PECAM-1 seen in this model accords with reports that TNF stimulates PECAM-1 redistribution out of the endothelial cleft in cultured cells (Romer et al., 1995). Upregulation of CD44 by TNF has also been described in HUVEC (Mackay et al., 1993). Downregulation of thrombomodulin by proinflammatory stimuli is also a well documented pro-coagulant response (Conway and Rosenberg, 1988) which also occurred in this model. An alternative explanation for this behaviour may be interacting factors or secondary mediators within the placenta. Hofbauer cells are thought to play a role in ‘immunodeviation’ within the placenta, suppressing both fetal and maternal immune responses (Hunt, Chu and Miller, 1997). Th-2 type cytokines or endothelial mitogens (e.g. FGFs I and II, VEGF and PlGF), produced by trophoblasts and Hofbauer cells (Ahmed, 1997), may contribute to this pattern of endothelial behaviour. This data, showing only limited inflammatory responses to TNF, may explain previous findings (Lyall et al., 1995; Steinborn et al., 1999). This could be relevant in several types of fetal pathophysiology where high levels of TNF are produced by placental cells (Conrad and Benyo, 1997; Dammann and Leviton, 1997).

41

Expression of endothelial markers in other cell types The expression of endothelial markers by trophoblasts is likely to have general relevance to placentation (Zhou et al., 1997). We found that each placental cell type expressed a different profile of endothelial markers (Table 3). eNOS, PGIS and ACE were found in syncytiotrophoblasts and extravillous trophoblasts but not in vascular trophoblasts. Integrin
v3 and endoglin were strongly expressed in syncytiotrophoblasts. These findings concur with published data showing expression in syncytiotrophoblasts of eNOS (Myatt et al., 1993; Eis et al., 1995), ACE (Poisner, 1998) and endoglin (Gougos et al., 1992). HLA-G was strongly expressed by extravillous trophoblasts, as previously described (Kaufmann and Castellucci, 1997; Le Bouteiller et al., 1999) but was also found in pericytes and vascular smooth muscle. Expression of ICAM-1 (Burrows et al., 1997), E- and P-selectin, and integrin
v3 (Kaufmann and Castellucci, 1997) has been reported in extravillous trophoblasts in the first trimester, though was not found in the present study at term. Vascular trophoblasts expressed several endothelial antigens (vWF, PECAM-1, thrombomodulin, PAL-E, A10-33-1, ICAM-1, E- and P- selectin, CD44) but not CD34, CD36, VE-cadherin, eNOS, PGIS or ACE. The expression of tissue factor only in stromal cells fits the hypothesis that its expression in pericytes is required for angiogenesis (Carmeliet et al., 1996) and, within the decidua, to prevent bleeding (Erlich et al., 1999). In conclusion, this study found the placental endothelium to express a typical profile of markers, but with some distinctive features. Several markers were differentially expressed within the placental vascular tree while micro-heterogenous expression of some markers could indicate either microenvironments or the presence of sub-types of endothelial cell. In explants, the inflammatory response to TNF treatment was limited. It would be interesting to extend this study to investigate the responses to other inflammatory mediators produced in the placenta. The emerging picture of the fetal endothelial phenotype is one in which the anti-coagulant/pro-coagulant and some of the vasodilator/vasoconstrictor systems are present but the potential for expressing some inflammatory adhesion molecules is lacking. This may represent a physiological status important for placental function, since full-blown placental inflammation has dire consequences (Benirschke and Kaufmann, 1995; Villegas et al., 1996).

ACKNOWLEDGEMENTS We thank the Developmental Studies Hybridoma Bank for the supply of antibodies to PECAM, VCAM-1 and E-selectin, Dr Clare Isacke, Dept. Biology, Imperial College, London for antibodies to CD44, Dr Philippe Le Bouteiller, INSERM, Toulouse, France, for antibodies to HGL-G and Dr Macimo Pignatelli, University of Bristol for antibody HECD and Dr Joselyn Glazier, Institute of Child Health, University of Manchester for helpful comments. The financial support of the Wellcome Trust and the British Heart Foundation is gratefully acknowledged.

REFERENCES Ahmed A (1997) Heparin-binding angiogenic growth factors in pregnancy. Trophoblast Research, 10, 215–258.

Aird WC, Edelberg JM, Weiler Guettler H, Simmons WW, Smith TW & Rosenberg RD (1997) Vascular bed-specific expression of an endothelial cell gene is programmed by the tissue microenvironment. J Cell Biol, 138, 1117–1124.

42 Albelda SM (1991) Endothelial and epithelial cell adhesion molecules. Am J Respir Cell Mol Biol, 4, 195–203. Austgulen R, Espevik T, Mecsei R & Scott H (1992) Expression of receptors for tumor necrosis factor in human placenta at term. Acta Obstet Gynecol Scand, 71, 417–424. Barleon B, Hauser S, Schollmann C, Weindel K, Marme D, Yayon A & Weich HA (1994) Differential expression of the two VEGF receptors flt and KDR in placenta and vascular endothelial cells. J Cell Biochem, 54, 56–66. Belloni PN, Carney DH & Nicolson GL (1992) Organ-derived microvessel endothelial cells exhibit differential responsiveness to thrombin and other growth factors. Microvasc Res, 43, 20–45. Benirschke K & Kaufmann P (1995) Pathology of the Human Placenta. New York: Springer-Verlag Inc. Bright NA, Ockleford CD & Anwar M (1994) Ontogeny and distribution of Fc gamma receptors in the human placenta. Transport or immune surveillance? J Anat, 184, 297–308. Brooks PC, Clark RA & Cheresh DA (1994) Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science, 264, 569–571. Bruggen J & Sorg C (1983) Detection of phenotypic differences on human malignant melanoma lines and their variant sublines with monoclonal antibodies. Cancer Immunol Immunother, 15, 200–205. Burrows TD, King A, Smith SK & Loke YW (1997) Trophoblast–Matrix interactions in human placentation. Trophoblast Research, 10, 163–172. Carmeliet P, Mackman N, Moons L, Luther T, Gressens P, Van Vlaenderen I, Demunck H, Kasper M, Breier G, Evrard P, Muller M, Risau W, Edgington T & Collen D (1996) Role of tissue factor in embryonic blood vessel development. Nature, 383, 73–75. Conrad KP & Benyo DF (1997) Placental cytokines and the pathogenesis of preeclampsia. Am J Reprod Immunol, 37, 240–249. Conway EM & Rosenberg RD (1988) Tumor necrosis factor suppresses transcription of the thrombomodulin gene in endothelial cells. Mol Cell Biol, 8, 5588–5592. Dammann O & Leviton A (1997) Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res, 42, 1–8. Dawson DW, Pearce SF, Zhong R, Silverstein RL, Frazier WA & Bouck NP (1997) CD36 mediates the In vitro inhibitory effects of thrombospondin-1 on endothelial cells. J Cell Biol, 138(3), 707–717. Demir R, Kaufmann P, Castellucci M, Erbengi T & Kotowski A (1989) Fetal vasculogenesis and angiogenesis in human placental villi. Acta Anat Basel, 136, 190–203. Denekamp J (1984) Vascular endothelium as the vulnerable element in tumours. Acta Radiol Oncol, 23(4), 217–225. Diglio CA, Grammas P, Giacomelli F & Wiener J (1988) Rat heartderived endothelial and smooth muscle cell cultures: isolation, cloning and characterization. Tissue Cell, 20, 477–492. Dye JF, Clark P, Leach L & Firth JA (1996) The phenotype of cultured human placental microvascular endothelial cells. J Anat, 189, 215–216. Dye JF, Jablenska R, Donnelly JL, Clark P & Firth JA (1997) Is the human placental endothelium resistant to inflammatory stimuli? Placenta, 18, A21. Dye JF, Leach L & Firth JA (1995) Marker expression in placental endothelial cells in situ and in vitro. Placenta, 16, A.17. Eis AL, Brockman DE, Pollock JS & Myatt L (1995) Immunohistochemical localization of endothelial nitric oxide synthase in human villous and extravillous trophoblast populations and expression during syncytiotrophoblast formation in vitro. Placenta, 16, 113–126. Erlich J, Parry GC, Fearns C, Muller M, Carmeliet P, Luther T & Mackman N (1999) Tissue factor is required for uterine hemostasis and maintenance of the placental labyrinth during gestation. Proc Natl Acad Sci USA, 96, 8138–8143. Ewenstein BM (1997) Vascular Biology of von Willebrand factor. In Vascular Endothelium (Eds) Born VR & Schartz CJ, pp. 107–122. Stuttgart: Schattauer. Fina L, Molgaard HV, Robertson D, Bradley NJ, Monaghan P, Delia D, Sutherland DR, Baker MA & Greaves MF (1990) Expression of the CD34 gene in vascular endothelial cells. Blood, 75, 2417–2426. Firth JA & Leach L (1996) Not trophoblast alone: a review of the contribution of the fetal microvasculature to transplacental exchange. Placenta, 17, 89–96. Fivenson DP, Douglass MC & Nickoloff BJ (1992) Cutaneous expression of Thy-1 in mycosis fungoides. Am J Pathol, 141, 1373–1380.

Placenta (2001), Vol. 22 Flanagan BF, Dalchau R, Allen AK, Daar AS & Fabre JW (1989) Chemical composition and tissue distribution of the human CDw44 glycoprotein. Immunology, 67, 167–175. Furuya S, Edwards C & Ornberg R (1990) Morphological behavior of cultured bovine adrenal medulla capillary endothelial cells. Tissue Cell, 22, 615–628. Ghabour MS, Eis AL, Brockman DE, Pollock JS & Myatt L (1995) Immunohistochemical characterization of placental nitric oxide synthase expression in preeclampsia. Am J Obstet Gynecol, 173, 687–694. Giddings JC (1990) Cytokines and thrombomodulin. In The Endothelium: An Introduction to Current Research (Ed.) Warren JB. New York: Wiley-Liss inc. Gougos A, St Jacques S, Greaves A, PJ OC, d’Apice AJ, Buhring HJ, Bernabeu C, van Mourik JA & Letarte M (1992) Identification of distinct epitopes of endoglin, an RGD-containing glycoprotein of endothelial cells, leukemic cells, and syncytiotrophoblasts. Int Immunol, 4, 83–92. Hagemeier, HH, Vollmer E, Goerdt S, Schulze Osthoff K & Sorg C (1986) A monoclonal antibody reacting with endothelial cells of budding vessels in tumors and inflammatory tissues, and non-reactive with normal adult tissues. Int J Cancer, 38, 481–488. Haraldsen G, Kvale D, Lien B, Farstad IN & Brandtzaeg P (1996) Cytokine-regulated expression of E-selectin, intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1) in human microvascular endothelial cells. J Immunol, 156, 2558–2565. Hunt JS, Chu W & Miller L (1997) Non-specific immunity of the uterus and placenta. Placenta, 18, A.3. Iatropoulos MJ & Williams GM (1996) Proliferation markers. Exp Toxicol Pathol, 48, 175–81. Kaufmann P & Castellucci M (1997) Extravillous trophoblast in the human placenta. Trophoblast Research, 10, 21–65. Kosanke G, Kadyrov M, Korr H & Kaufmann P (1998) Maternal Aneamia results in increased proliferation in human placental villi. Trophoblast Research, 11, 339–357. Kraling BM, Razon MJ, Boon LM, Zurakowski D, Seachord C, Darveau RP, Mulliken JB, Corless CL & Bischoff J (1996) E-selectin is present in proliferating endothelial cells in human hemangiomas. Am J Pathol, 148, 1181–1191. Kumar S, West DC & Ager A (1987) Heterogeneity in endothelial cells from large vessels and microvessels. Differentiation, 36, 57–70. Lampugnani MG, Resnati M, Dejana E & Marchisio PC (1991) The role of integrins in the maintenance of endothelial monolayer integrity. J Cell Biol, 112, 479–490. Lampugnani MG, Resnati M, Raiteri M, Pigott R, Pisacane A, Houen G, Ruco LP & Dejana E (1992) A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol, 118, 1511–1522. Lang I, Hartmann M, Blaschitz A, Dohr G, Skofitsch G & Desoye G (1993) Immunohistochemical evidence for the heterogeneity of maternal and fetal vascular endothelial cells in human full-term placenta. Cell Tissue Res, 274, 211–218. Le Bouteiller P, Solier C, Proll J, Aguerre Girr M, Fournel S & Lenfant F (1999) Placental HLA-G protein expression in vivo: where and what for? Hum Reprod Update, 5, 223–233. Leach L, Bhasin Y, Clark P & Firth JA (1994) Isolation of endothelial cells from human term placental villi using immunomagnetic beads. Placenta, 15, 355–364. Leach L, Clark P, Lampugnani MG, Arroyo AG, Dejana E & Firth JA (1993) Immunoelectron characterisation of the inter-endothelial junctions of human term placenta. J Cell Sci, 104, 1073–1081. Leach L, Westcott EDA, Westgeest J & Firth JA (1995) Effect of histamine and tumour necrosis factor on the distribution and localisation of cadherin-5 and PECAM-1 in the perfused human term placental microvessels. J Anat, 187, 206. Lyall F, Greer IA, Boswell F, Young A, Macara LM & Jeffers MD (1995) Expression of cell adhesion molecules in placentae from pregnancies complicated by pre-eclampsia and intrauterine growth retardation. Placenta, 16, 579–587. Mackay F, Loetscher H, Stueber D, Gehr G & Lesslauer W (1993) Tumor necrosis factor alpha (TNF-alpha)-induced cell adhesion to human endothelial cells is under dominant control of one TNF receptor type, TNF-R55. J Exp Med, 177, 1277–1286. Milner P, Loesch A & Burnstock G (1997) Plasticity of expression of vasoactive agents. In Vascular Endothelium: Physiology, Pathology and Therapeutic Opportunities (Eds) Born GVR & Scwartz CJ, pp. 243–260. Stuttgart: Schattauer.

Dye et al.: Phenotype of the Placental Endothelium Myatt L, Brockman DE, Eis AL & Pollock JS (1993) Immunohistochemical localization of nitric oxide synthase in the human placenta. Placenta, 14, 487–495. Myatt L, Eis AL, Brockman DE, Greer IA & Lyall F (1997) Endothelial nitric oxide synthase in placental villous tissue from normal, pre-eclamptic and intrauterine growth restricted pregnancies. Hum Reprod, 12, 167– 172. Nahler N (1985). Gewebespezifische heterogenitat kapiller endothelzellen: dargestellt durch reaktionen monoklonaler antikorper auf kultivierten zellen. Zell- und gewebskulturmodel in der pathobiochemie der bindegewebserkrankungen, Bad Waldsee, PCS-Verlag Basel. Poisner AM (1998) The human placental renin-angiotensin system. Front Neuroendocrinol, 19, 232–252. Reverdiau P, Jarousseau AC, Thibault G, Khalfoun B, Watier H, Lebranchu Y, Bardos P & Gruel Y (1995) Tissue factor activity of syncytiotrophoblast plasma membranes and tumoral trophoblast cells in culture. Thromb Haemost, 73, 49–54. Romer LH, McLean NV, Yan HC, Daise M, Sun J & DeLisser HM (1995) IFN-gamma and TNF-alpha induce redistribution of PECAM-1 (CD31) on human endothelial cells. J Immunol, 154, 6582–6892. Rupnick MA, Carey A & Williams SK (1988) Phenotypic diversity in cultured cerebral microvascular endothelial cells. In Vitro Cell Dev Biol, 24, 435–444. Salomon D, Ayalon O, Patel King R, Hynes RO & Geiger B (1992) Extrajunctional distribution of N-cadherin in cultured human endothelial cells. J Cell Sci, 102, 7–17. Schlingemann RO, Dingjan GM, Emeis JJ, Blok J, Warnaar SO & Ruiter DJ (1985) Monoclonal antibody PAL-E specific for endothelium. Lab Invest, 52, 71–76. Schlingemann RO, Rietveld FJ, de Waal RM, Bradley NJ, Skene AI, Davies AJ, Greaves MF, Denekamp J & Ruiter DJ (1990) Leukocyte antigen CD34 is expressed by a subset of cultured endothelial cells and on endothelial abluminal microprocesses in the tumor stroma. Lab Invest, 62, 690–696. Schnitzer JE (1997) The endothelial cell surface and caveolae in health and disease. In Vascular Endothelium: Physiology, Pathology and Therapeutic Opportunities (Eds) Born GVR & Schwartz CJ, pp. 77–96. Stuttgart: Schattauer. Sedmak DD, Davis DH, Singh U, van de Winkel JG & Anderson CL (1991) Expression of IgG Fc receptor antigens in placenta and on endothelial cells in humans. An immunohistochemical study. Am J Pathol, 138, 175–181.

43 Shellhaas CS, Coffman T, Dargie PJ, Killam AP & Kay HH (1997) Intravillous eicosanoid compartmentalization and regulation of placental blood flow. J Soc Gynecol Investig, 4, 58–63. Simister NE & Story CM (1997) Human placental Fc receptors and the transmission of antibodies from mother to fetus. J Reprod Immunol, 37, 1–23. Spanel Borowski K & Fenyves A (1994) The heteromorphology of cultured microvascular endothelial cells. Arzneimittelforschung, 44, 385–391. St Jacques S, Forte M, Lye SJ & Letarte M (1994) Localization of endoglin, a transforming growth factor-beta binding protein, and of CD44 and integrins in placenta during the first trimester of pregnancy. Biol Reprod, 51, 405–413. Steinborn A, Geisse M & Kaufmann M (1998) Expression of cytokine receptors in the placenta in term and preterm labour. Placenta, 19, 165–170. Steinborn A, Sohn C, Heger S, Niederhut A, Hildenbrand R & Kaufmann M (1999) Labour-associated expression of intercellular adhesion molecule-1 (ICAM-1) in placental endothelial cells indicates participation of immunological processes in parturition. Placenta, 20, 567–573. Suter L, Brocker EB, Bruggen J, Ruiter DJ & Sorg C (1983) Heterogeneity of primary and metastatic human malignant melanoma as detected with monoclonal antibodies in cryostat sections of biopsies. Cancer Immunol Immunother, 16, 53–58. Varner JA & Cheresh DA (1996) Tumor angiogenesis and the role of vascular cell integrin alphavbeta3. Important Adv Oncol, 69–87. Villegas H, Gonzalez Jimenez MA, Paredes Vivas Y, Aguilera Suarez G & Karchmer S (1996) [Ultrastructural analysis of placentas with villitis. Retrospective study]. Ginecol Obstet Mex, 64, 368–376. Wang JM, Kumar S, Pye D, van Agthoven AJ, Krupinski J & Hunter RD (1993) A monoclonal antibody detects heterogeneity in vascular endothelium of tumours and normal tissues. Int J Cancer, 54, 363–370. Westphal JR, Willems HW, Schalkwijk CJ, Ruiter DJ & de Waal RM (1993) A new 180-kDa dermal endothelial cell activation antigen: in vitro and in situ characteristics. J Invest Dermatol, 100, 27–34. Wetzka B, Charnock Jones DS, Viville B, Cooper JC, Nusing R, Zahradnik HP & Smith SK (1996) Expression of prostacyclin and thromboxane synthases in placenta and placental bed after pre-eclamptic pregnancies. Placenta, 17, 573–581. Zhou Y, Fisher SJ, Janatpour M, Genbacev O, Dejana E, Wheelock M & Damsky CH (1997) Human cytotrophoblasts adopt a vascular phenotype as they differentiate. A strategy for successful endovascular invasion? J Clin Invest, 99, 2139–2151.