Histochem Cell Biol (2007) 128:55–63 DOI 10.1007/s00418-007-0293-7
ORIGINAL PAPER
Expression of the blood-group-related antigens Sialyl Lewis a, Sialyl Lewis x and Lewis y in term placentas of normal, preeclampsia, IUGR- and HELLP-complicated pregnancies Vassilis Minas · Ioannis Mylonas · Barbara Schiessl · Doris Mayr · Sandra Schulze · Klaus Friese · Udo Jeschke · Antonis Makrigiannakis
Accepted: 2 May 2007 / Published online: 6 June 2007 © Springer-Verlag 2007
Abstract Lewis antigens belong to the blood group of antigens and mediate cellular adhesion through interaction with selectins. Invasive trophoblasts use an array of adhesion molecules to facilitate cell–cell and cell–extracellular matrix interactions. Here, we examined immunohistochemically the expression of Sialyl Lewis a (sLea), Sialyl Lewis x (sLex) and Lewis y (Ley) in term placentas obtained from cases of normal, intrauterine growth retardation (IUGR), preeclamptic (PE) and hemolysis, elevated liver enzymes and low platelets syndrome (HELLP) pregnancies. We report the expression of sLex in third trimester extravillous trophoblasts (EVT). sLex was signiWcantly decreased in IUGR and moderately decreased in PE compared to normal placentas. sLex was additionally found in syncytiotrophoblast, without however any signiWcant diVerences in staining intensity between normal and pathological cases. sLea was restricted to amnion epithelium. Finally, Ley was expressed in cytotrophoblasts and villous endothelial cells. Ley expression was signiWcantly upregulated in IUGR and
U. Jeschke and A. Makrigiannakis have contributed equally. V. Minas · A. Makrigiannakis Laboratory of Human Reproduction, Department of Obstetrics and Gynaecology, Faculty of Medicine, University of Crete, 71003 Heraklion, Greece I. Mylonas (&) · B. Schiessl · S. Schulze · K. Friese · U. Jeschke First Department of Obstetrics and Gynaecology, Ludwig-Maximilians-University of Munich, Maistrasse 11, 80337 Munich, Germany e-mail:
[email protected] D. Mayr Institute of Pathology, Ludwig Maximilians University of Munich, Thalkirchner Str. 36, 80337 Munich, Germany
HELLP, whereas there was a trend toward increase in PE compared to normal placentas. The present study suggests that downregulation of sLex in EVT might be associated with IUGR and PE. Furthermore, Ley, which was recently described as a potent angiogenic factor, is upregulated in placental villi in conditions associated with placental malperfusion. Keywords Lewis antigens · Trophoblast · IUGR · Preeclampsia · HELLP
Introduction During initial stages of placentation the cytotrophoblast of the anchoring villi gives rise to the cytotrophoblast columns, which penetrate the syncytiotrophoblast “shell” to invade maternal tissues. Cytotrophoblasts that penetrate beyond the syncytiotrophoblast shell are termed extravillous trophoblasts (EVT). A subpopulation of EVT, named interstitial trophoblast, invades the maternal decidua and myometrium and provides the anchorage of the placenta and the conceptus to the uterus (Chaddha et al. 2004). Another EVT subpopulation, the endovascular trophoblast, initially occludes the spiral arteries at the implantation site and migrates in a retrograde direction along the vessels. These cells adopt endothelial-like characteristics and gradually replace the maternal endothelium (Zhou et al. 1997a). By the middle of second trimester both the decidual and myometrial segments of spiral arteries have been remodeled to produce vessels of increased diameter and low resistance, able to meet the growing fetal demands in blood supply (Pijnenborg et al. 1983). Failure of trophoblasts to properly invade maternal tissues may lead to a spectrum of pregnancy complications.
123
56
Histochem Cell Biol (2007) 128:55–63
Impaired trophoblast invasion and altered patterns of EVT apoptosis have been identiWed in conditions associated with uteroplacental malperfusion such as intrauterine growth retardation (IUGR), preeclampsia (PE) and haemolysis, elevated liver enzymes and low platelets syndrome (HELLP) (Brosens et al. 1972; Zhou et al. 1997b; Reister et al. 2001; Zhou et al. 2002). PE is the leading cause of maternal mortality in the Western world and also increases perinatal mortality Wvefold (Walker 2000). The endovascular invasion in this condition is often so limited, that the identiWcation of cytotrophoblasts in maternal vessels is very diYcult (Brosens et al. 1972). The Sialyl Lewis (sLe) antigens are blood group-related carbohydrates. sLex (NeuNAc2!3Gal1!4[Fuc1!3] GlcNAc!R) and sLea (NeuNAc2!3Gal1!3 [Fuc1!4]GlcNAc!R) are monofucosylated tetrasaccharides which represent cognate ligands of E- and P-selectins. Less is known about the physiological role of Ley (Fuc1!2Gal1!4[Fuc1!3]GlcNAc1!R), a difucosylated tetrasaccharide, which has been implicated in the processes of cellular adhesion and cellular motility (Hakomori 1991). In an early study, King et al. reported the presence of Lewis antigens in extravillous trophoblasts. Given the strong resemblance of EVT and cancer cells, the authors suggested that these factors could act as targets for uterine natural killer cell (uNK)-mediated control of EVT apoptosis and invasion (King and Loke 1988). Interestingly, a recent study showed that excessive expression of sLex in tumor cells resulted in rejection by NK cells (Ohyama et al. 1999). The ability of Lewis antigens to mediate adhesion of tumor cells to endothelial cells expressing E-selectin, has implicated these factors in the pathophysiology of bloodborn tumor metastasis (Takada et al. 1993; Alpaugh et al. 2002). We have previously reported increased levels of blood group related carbohydrate antigens sLea and sLex in invasive carcinoma without lymph node metastasis compared to carcinoma in situ. The diVerences were even stronger in cases of invasive carcinoma with lymph node metastasis, suggesting that these antigens might contribute to the invasive and metastatic potential of this malignancy (Jeschke et al. 2005). Furthermore, endovascular trophoblasts have
been suggested to utilize components of the leukocyte emigration system such as selectins and their carbohydratebased ligands (Zhou et al. 2003). In the present study we tested the hypothesis that altered levels of Lewis antigens’ expression in EVT might be associated with conditions characterized by impaired trophoblast invasion and/or apoptosis. In particular, we examined the expression of sLea, sLex and Ley in term placenta and compared the levels of the antigens’ expression between normal and IUGR, PE and HELLP placentas.
Materials and methods Tissue samples Decidual tissues were obtained from 27 placentas of women giving birth in the Wrst Department of Obstetrics and Gynecology of the LMU Munich. Six placentas with IUGR (mean date of delivery § SD 33 § 3 weeks of gestation), eight placentas with PE (mean date of delivery 33 § 3.2 weeks of gestation), Wve placentas with HELLP (mean date of delivery 33.8 § 2.9 weeks of gestation) and eight normal placentas (mean date of delivery 38.2 § 3 weeks of gestation) were used for the study. The study was approved by the ethics committee of the LMU Munich, Germany and informed consent was obtained from the patients. The clinical details of the cases enrolled in the study are shown in Table 1. It would be better for comparison if the gestational age of the controls were closer to that of the pathological pregnancies. Physiologic ongoing pregnancies last up to 42 weeks and reach maturity after the completion of 37 weeks. Normal placental material was taken from that period of time. Earlier deliveries just for research purposes were impossible due to ethical reasons. Fetal intrauterine growth restriction (IUGR) was deWned as an estimated intrauterine weight below the Wfth centile after gestational age had been conWrmed by an early pregnancy ultrasound. IUGR needed to be conWrmed after delivery; otherwise the patient was excluded from the study. PE was deWned as maternal hypertension >140/ 90 mmHg without previous history of hypertension and
Table 1 Clinical details on the patients and newborns in the HELLP, IUGR, PE and normal control group IUGR (n = 6)
Preeclampsia (n = 8)
HELLP (n = 5)
Normal control (n = 8) 38.2 (SD 3.9)
Week of gestation at delivery
33.0 (SD 3.0)
33.18 (SD 0.83)
33.8 (SD 2.8)
Neonatal birthweight (g)
1213.6 (SD 545.3)
1,409 (SD 547)
1796.2 (SD 689.1)
3,435 (SD 412.2)
pH umbilical artery
7.26 (SD 0.09)
7.29 (SD 0.06)
7.332 (SD 0.09)
7.27 (SD 0.09)
Apgar score at 5 min
9.0 (SD 0.7)
9.36 (SD 0.67)
9.2 (SD 0.84)
9.4 (SD 0.9)
Apgar score at 10 min
9.6 (SD 0.5)
9.73 (SD 0.47)
9.8 (SD 0.5)
10 (SD 0)
SD standard deviation
123
Histochem Cell Biol (2007) 128:55–63
57
300 mg/l proteinuria without any history of renal disease. The deWnition for HELLP-syndrome is elevated liver enzymes, low platelets and haemolysis. Patients included into the study had a minimum of twofold increased liver enzymes, platelets below 100,000/l and detectable haptoglobin below the normal range. Immunohistochemistry Immunohistochemistry on paraYn sections (7 m) of the diVerent decidual tissue specimens was performed by incubating the slides in methanol/H2O2 (30 min) to inhibit endogenous peroxidase activity, followed by washing in phosphate-buVered saline (PBS, pH 7.4) (5 min) and treatment with goat serum (20 min, 22°C) to reduce non-speciWc background staining. Tissues were incubated with the primary antibodies (Table 2) for 1 h at room temperature. Sections were incubated with the biotinylated secondary antibody (1 h, 22°C) and avidin-biotinylated peroxidase (45 min, room temperature). Between each step, sections were washed with PBS. Peroxidase staining reaction was done with diaminobenzidine/H2O2 (1 mg/ml; 5 min) and stopped in tap water (10 min). Sections were counterstained in hemalaun (1 min) and then cover-slipped. In controls, the primary antibody was replaced with pre-immune mouse serum with positive and negative controls included. All specimens were evaluated by a pathologist (D.M.). The intensity and distribution patterns of the staining reaction were evaluated by two blinded, independent observers, including a gynecological pathologist, using the semi-quantitative immunoreactive score (IRS), as previously described (Remmele et al. 1986), and used to assess steroid receptors (Mylonas et al. 2000) and cathepsin D (Mylonas et al. 2003) expression in human endometrial tissue. The IRS was calculated by multiplication of optical-staining intensity (graded as 0 = no, 1 = weak, 2 = moderate and 3 = strong staining) and the percentage of positive-stained cells (0 = no staining, 1 = 81% of the cells) and without having knowledge of the pathological evaluation, the diagnosis or the standard performed hematoxylin reaction of each specimen.
Table 2 Antibodies used in the study
Antigen
Antibody
ImmunoXuorescence double staining For the characterization of Lewis antigens-expressing cells in placental tissue, cryosections were examined. All samples were Wxed in 5% buVered formalin. Slides were incubated with primary antibodies (Table 2) overnight at 4°C. After washing, Cy2-labelled goat anti-mouse IgM and Cy3labeled goat anti-mouse IgG (both Dianova, Hamburg, Germany), diluted 1:200 and served as secondary antibodies. For double-staining procedures AYniPure F(ab’)2 fragment goat anti-mouse IgM, chain-speciWc antibodies were used. These antibodies react with the heavy chain of mouse IgM and showed no reaction with mouse IgG (data obtained from Dianova). The slides were Wnally embedded in mounting buVer containing 4⬘,6-diamino-2-phenylindole, resulting in staining of the nucleus. Slides were examined with a Zeiss (Jena, Germany) Axiophot photomicroscope. Digital images were obtained with a digital-camera system (Axiocam, Zeiss) and saved on computer. Dot blot analysis The relevant sLex/ Ley carbohydrate structures were identiWed by dot blot analysis as described previously (Stahn et al. 2005). Aliquots of diVerent villous placental tissue (5 l each, containing 1 g protein) were spotted on nitrocellulose strips and dried for 20 min at room temperature. UnspeciWc binding sites were blocked (4% milk powder in PBS) for 1 h. Individual strips were then incubated with diVerent carbohydrate-recognizing antibodies (1 g/ml) for 1 h at room temperature, washed twice in PBS containing 0.05% Tween 20 and detected with polyclonal rabbit anti mouse IgG conjugated to AP (1:1000, Dako, Hamburg, Germany) by incubating the strips for 1 h at room temperature and developed with the BCIP reagent (Sigma, Germany). Statistical analysis The SPSS/PC software package, version 10.01 (SPSS, Munich, Germany), was used for collection, processing and statistical analysis of all data. Statistical analysis was
Isotype
Concentration/dilution
Source Calbiochem
Sialyl Lewis a
KM231
Mouse IgM
1 g/ml
Sialyl Lewis a
Ca19-9
Mouse IgG
1:15
Calbiochem
Sialyl Lewis x
CSLEX
Mouse IgM
1:200
Pharmingen
Lewis y
MCA 2160
Mouse IgM
1 g/ml
Serotec
Cytokeratin-7
OV-TL12/30
Mouse IgG1
1:30
Novocastra
123
58 Fig. 1 Immunohistochemical staining of sLea in term placenta. sLea is expressed in amnion epithelium (a), while other placental cell types are negative (b) (magniWcation £250)
Fig. 2 Immunohistochemical staining of sLex in term placenta. sLex is expressed in syncytiotrophoblast and extravillous trophoblast in normal (a, b), IUGR (c, d), PE (e, f) and HELLP (g, h) cases (magniWcation £250). Double immunoXuorescence was applied to show identity of sLex expressing cells by double staining with Cytokeratin-7. sLex appears in green and CK-7 in red (b, d, f, h)
123
Histochem Cell Biol (2007) 128:55–63
Histochem Cell Biol (2007) 128:55–63
59
performed using the non-parametrical Mann–Whitney U test for comparison of the mean values. P < 0.05 values were considered statistically signiWcant.
12 10
IRS score
Results
x
SLe in Syncytiotrophoblast
8 6 4
Expression of Sialyl Lewis a in term placenta in normal and pathological cases
2
A
Expression of Sialyl Lewis x in term placenta in normal and pathological cases
0 normal
IUGR
PE
HELLP
x
SLe in Extravillous Trophoblast 10 9 8 7 IRS sc or e
To assess the expression of Lewis antigens in placenta, immunohistochemistry was performed. sLea was strongly expressed in amnion epithelial tissue in all cases investigated. However, there was no staining in decidua, EVT, villous syncytiotrophoblast, cytotrophoblast, stroma or endothelium, in any sample (Fig. 1).
*
6 5
*
4 3 2 1
Sialyl Lewis x was detected in syncytiotrophoblast (Fig. 2a, c, e, g) and EVT (Fig. 2b, d, f, h). Comparison of syncytiotrophoblast staining intensity between normal and pathological cases revealed no statistically signiWcant diVerences (Fig. 3a). For EVT, sLex expression was signiWcantly decreased in IUGR and moderately decreased in PE compared to normal cases (Fig. 3b). The identity of SLexpositive EVT was analyzed with double immunoXuorescence experiments and with sLex and cytokeratin-7 antibodies (Fig. 2b, d, f, h). Expression of Lewis y in term placenta in normal and pathological cases Lewis y was immunolocalized in cytotrophoblast and villous endothelial cells in almost all of the specimens examined, whereas syncytiotrophoblast, decidua and EVT were negative (Fig. 4a–d). IRS scoring revealed a statistically signiWcant increase in Ley expression in cytotrophoblast (Fig. 5a) and villous endothelium (Fig. 5b) of IUGR and HELLP compared to normal cases. A trend toward increase in PE was also noted; however, it was not statistically signiWcant. Characterization of Lewis y-expressing cells in placental villi To verify that both cytotrophoblast and villous endothelial cells express Ley, a Xuorescence double labeling with trophoblast marker (cytokeratin-7) and Ley antibodies was performed. Cytotrophoblast in placental villi was stained with both antibodies, whereas endothelium showed only Ley expression (Fig. 6).
B
C
0 normal
Normal
IUGR
IUGR
PE
HELLP
PE
HELLP
SLeX in Dot-Blot in villous Trophoblast
Fig. 3 Staining intensity of sLex expression in syncytiotrophoblast (a) and EVT (b) of normal, IUGR, PE and HELLP cases determined by the semi-quantitative immunohistochemical immunoreactive score on the diVerent tissue slides. Semi-quantitative data were obtained by using just the immunohistochemical staining reaction and not immunoXuorescence results. Data represent mean § SEM (*P < 0.05). Dot blot analysis of sLex expression in villous trophoblast tissue of normal, IUGR, PE and HELLP placentas (c)
Dot blot analysis Dot blot analysis was performed to visualize the expression of the tetrasaccharides sLex and Ley. Figure 3c summarizes the results for the individual SLex expression from diVerent sources of dot blot analysis. SLex was identiWed in villous trophoblast in normal placental villous trophoblast tissue and in IUGR, PE and HELLP villous trophoblasts. Expression of sLex was rather low in normal and HELLP villous trophoblasts. On the contrary, sLex showed a higher expression in IUGR and PE villous trophoblasts. Figure 5c shows the expression of Ley in villous cytotrophoblasts obtained by dot blot analysis. Expression of Ley was rather low in normal villous cytotrophoblasts. On the contrary, Ley showed a higher expression in IUGR and
123
60
Histochem Cell Biol (2007) 128:55–63
Fig. 4 Immunohistochemical staining of Ley in term placenta. Ley is expressed in villous cytotrophoblast and endothelial cells in normal (a), IUGR (b), PE (c) and HELLP (d) cases. Primary antibody was omitted for negative control (e) (magniWcation £250)
PE villous cytotrophoblasts and highest expression in HELLP villous cytotrophoblasts.
Discussion In the present study, we investigated the expression of the blood group related carbohydrates Sialyl Lewis a, Sialyl Lewis x and Lewis y in term placentas obtained from normal, IUGR, PE and HELLP pregnancies. sLea was expressed in amnion epithelium in all cases. sLex was expressed in syncytiotrophoblast and EVT. There was no signiWcant diVerence in syncytiotrophoblast sLex staining intensity between normal and pathological cases although we identiWed an upregulation in IUGR and PE, which was conWrmed by dot blot analysis. Contrarily, sLex expression in EVT was signiWcantly downregulated in IUGR and moderately downregulated in PE cases. Finally, Ley was detected in cytotrophoblasts and endothelial cells of chorionic villi. The expression of Ley in these cell types was signiWcantly increased in IUGR and HELLP compared to normal placentas. In PE, Ley was also upregulated; however this diVerence was not statistically signiWcant.
123
Findings of Ley expression in villous trophoblast tissue was conWrmed by dot blot analysis. A functional role for Lewis antigens in placentation has been supported by two early studies (King and Loke 1988; Burrows et al. 1994). The authors reported the expression of P- and E-selectin and their ligand sLex at the implantation site. Selectins were detected in vascular endothelial cells in decidua basalis (Burrows et al. 1994) and sLex was detected in Wrst and second trimester EVT (King and Loke 1988). It was therefore suggested that these molecules could be utilized by EVT to facilitate their migration within decidual spiral arterioles (Burrows et al. 1994). In agreement with the aforementioned studies, we report the strong expression of sLex in third trimester EVT, whereas sLea and Ley were absent from this cell type. The reduced expression of sLex in EVT of IUGR and PE placentas might indicate a functional role for this antigen in trophoblast invasion. We propose that reduced sLex expression might contribute to shallow endovascular trophoblast invasion, which occurs in IUGR and PE. Elegant studies by Zhou et al. have indicated that cytotrophoblasts adopt a vascular phenotype as they diVerentiate along the invasive pathway (Zhou et al. 1997a). This process includes the
Histochem Cell Biol (2007) 128:55–63
61
y
Le in Cytotrophoblast 14 12 10 IRS score
*
*
8 6 4 2
A
0 normal
IUGR
PE
HELLP
y
Le in Villous Endothelium 12
*
10
*
IRS score
8 6 4 2
B
0
Normal
C
normal
IUGR
IUGR
PE
PE
HELLP
HELLP
LeY in Dot-Blot in Cytotrophoblast
Fig. 5 Staining intensity of Ley expression in cytotrophoblast (a) and endothelial cells (b) of normal, IUGR, PE and HELLP cases determined by the semi-quantitative immunohistochemical immunoreactive score on the diVerent tissue slides. Data represent mean § SEM (*P < 0.05). Dot blot analysis of Ley expression in villous trophoblast tissue of normal, IUGR, PE and HELLP placentas (c)
downregulation of a number of adhesion molecules and the induction of others, which promote invasion. Failure of cytotrophoblasts to appropriately modulate the expression of these molecules was associated with PE (Zhou et al. 1997b). Our data show that sLex could be an additional adhesion mediator, which is absent in cytotrophoblasts, but is strongly upregulated in invasive trophoblasts. Likewise, we show here that impaired expression of the sLex antigen by invasive trophoblasts is present in PE and IUGR. Recently, Ohyama et al. proposed that, although sLex facilitates attachment of tumor cells to endothelial cells and thus blood-born metastasis, when excessively expressed, the antigen might rather lead to tumor cell rejection by NK cells (Ohyama et al. 1999). In our study sLex was downregulated in EVT in PE, i.e. in a condition where EVT apoptosis
has been reported to be markedly increased (Zhou et al. 1997b). Therefore, it appears that there is rather no clear connection between expression of sLex and apoptosis of EVT in placenta. Furthermore, in contrast with King et al. we report the expression of sLex in syncytiotrophoblast, as well as expression of Ley in cytotrophoblast of third trimester villi. The two studies agree on the expression of Ley in villous endothelium. Prakobphol et al. showed that villous trophoblast cells express both, L-selectin and its carbohydrate ligand sLex (Prakobphol et al. 2006). Cytotrophoblast aggregates that bridge the gap between the placenta and the uterus are suspended as cell columns in the intervillous space, where they experience signiWcant amounts of shear stress generated by maternal blood Xow. The proper formation of these structures is crucial to pregnancy outcome as they play a vital role in anchoring the embryo/fetus to the decidua. Prakobphol et al. present evidence that the interactions of L-selectin with its carbohydrate ligands, a specialized adhesion system that is activated by shear stress, play an important role. In our study, Ley was immunolocalized in both villous cytotrophoblast and endothelial cells, since double Xuorescence experiments revealed expression of Ley in both cytokeratin-7 positive and negative villous structures. It is pertinent to mention that King et al. examined a very small number of cases, which could explain the fact that they failed to detect expression of Lewis antigens in villous trophoblasts. Interestingly, a recent study reported angiogenic eVects for Ley-6/H-5-2 (Ley/H). The glycoconjugate was found to be chemotactic for human dermal microvascular endothelial cells in vitro and to induce neovascularization in vivo. The angiogenic response to Ley/H was similar to that of potent angiogenic factors such as vascular endothelial growth factor and basic Wbroblast growth factor (Halloran et al. 2000). In our study Ley expression was increased in cytotrophoblasts and villous endothelial cells in IUGR, PE and HELLP cases. Taken together, these data prompt us to suggest that upregulation of Ley in villi could be a response of fetal villous vessels in compensation for the placental hypoxia, which is associated with these conditions. Adjacent villous trophoblasts might also upregulate soluble Ley to enhance angiogenesis inside the villous stroma. One of the diYculties of this study was to obtain placental tissue from normal and IUGR, PE and HELLP pregnancies adjusted to gestational age. Pregnancies complicated with IUGR, PE and HELLP are delivered preterm due to susceptibility of fetal cardiotocography or abnormal umbilical artery Doppler waveforms. Since it is still quite unclear if placental gestation can inXuence expression of blood group related carbohydrate antigens, careful acquirement of normal placental tissue samples was envisaged. Therefore,
123
62
Histochem Cell Biol (2007) 128:55–63
Fig. 6 Double Xuorescence staining for Ley (a, green) and cytokeratin-7 (b, red) in term placenta counterstained with 4⬘6⬘-diamidino-2-phenylindole for nuclei visualization. Triple-Wlter excitation (c), (magniWcation £400). Endothelial cells express Ley, but not cytokeratin-7 (white arrows)
tissue specimens of normal pregnancies were obtained at the course of an elective cesarean section for breech presentation during the 38th week of gestation to avoid any inXuencing factors due to the physiological “stress” during normal delivery. However, this is a major issue, since the observed diVerences can be either due to the disease or due to the gestation of the tissue and further investigations are warranted to elucidate these questions. In summary, we show here that EVT from PE and IUGR pregnancies display in situ reduced expression of the sLex antigen, compared to uncomplicated pregnancies. Whether this disparity contributes to the limited endovascular EVT invasion observed in these conditions, is a tempting hypothesis that warrants further investigation. Placenta malperfusion might account for the increased expression of Ley in the conditions examined. We speculate that this could be a mechanism for the embryo to induce placental angiogenesis and thus compensate for the reduced oxygen and nutrient supply. Acknowledgments This work was supported by a project and grant from the German (DAAD) & Greek State Scholarships Foundation (IKYDA 2003)
References Alpaugh ML, Tomlinson JS, Kasraeian S, Barsky SH (2002) Cooperative role of E-cadherin and sialyl-Lewis X/A-deWcient MUC1 in the passive dissemination of tumor emboli in inXammatory breast carcinoma. Oncogene 21:3631–3643
123
Brosens IA, Robertson WB, Dixon HG (1972) The role of the spiral arteries in the pathogenesis of preeclampsia. Obstet Gynecol Annu 1:177–191 Burrows TD, King A, Loke YW (1994) Expression of adhesion molecules by endovascular trophoblast and decidual endothelial cells: implications for vascular invasion during implantation. Placenta 15:21–33 Chaddha V, Viero S, Huppertz B, Kingdom J (2004) Developmental biology of the placenta and the origins of placental insuYciency. Semin Fetal Neonatal Med 9:357–369 Hakomori S (1991) Possible functions of tumor-associated carbohydrate antigens. Curr Opin Immunol 3:646–653 Halloran MM, Carley WW, Polverini PJ, Haskell CJ, Phan S, Anderson BJ, Woods JM, Campbell PL, Volin MV, Backer AE, Koch AE (2000) Ley/H: an endothelial-selective, cytokine-inducible, angiogenic mediator. J Immunol 164:4868–4877 Jeschke U, Mylonas I, Shabani N, Kunert-Keil C, Schindlbeck C, Gerber B, Friese K (2005) Expression of sialyl Lewis X, sialyl Lewis A, E-cadherin and cathepsin-D in human breast cancer: immunohistochemical analysis in mammary carcinoma in situ, invasive carcinomas and their lymph node metastasis. Anticancer Res 25:1615–1622 King A, Loke YW (1988) DiVerential expression of blood-group-related carbohydrate antigens by trophoblast subpopulations. Placenta 9:513–521 Mylonas I, Speer R, Makovitzky J, Richter DU, Briese V, Jeschke U, Friese K (2000) Immunohistochemical analysis of steroid receptors and glycodelin A (PP14) in isolated glandular epithelial cells of normal human endometrium. Histochem Cell Biol 114:405– 411 Mylonas I, Makovitzky J, Richter DU, Jeschke U, Briese V, Friese K (2003) Cathepsin D expression in normal, hyperplastic and malignant endometrial tissue: an immunohistochemical analysis. Acta Histochem 105:245–252 Ohyama C, Tsuboi S, Fukuda M (1999) Dual roles of sialyl Lewis X oligosaccharides in tumor metastasis and rejection by natural killer cells. Embo J 18:1516–1525
Histochem Cell Biol (2007) 128:55–63 Pijnenborg R, Bland JM, Robertson WB, Brosens I (1983) Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta 4:397–413 Prakobphol A, Genbacev O, Gormley M, Kapidzic M, Fisher SJ (2006) A role for the L-selectin adhesion system in mediating cytotrophoblast emigration from the placenta. Dev Biol 298:107–117 Reister F, Frank HG, Kingdom JC, Heyl W, Kaufmann P, Rath W, Huppertz B (2001) Macrophage-induced apoptosis limits endovascular trophoblast invasion in the uterine wall of preeclamptic women. Lab Invest 81:1143–1152 Remmele W, Hildebrand U, Hienz HA, Klein PJ, Vierbuchen M, Behnken LJ, Heicke B, Scheidt E (1986) Comparative histological, histochemical, immunohistochemical and biochemical studies on oestrogen receptors, lectin receptors, and Barr bodies in human breast cancer. Virchows Arch A Pathol Anat Histopathol 409:127–147 Stahn R, Goletz S, Wilmanowski R, Wang X, Briese V, Friese K, Jeschke U (2005) Human chorionic gonadotropin (hCG) as inhibitior of E-selectin-mediated cell adhesion. Anticancer Res 25:1811–1816 Takada A, Ohmori K, Yoneda T, Tsuyuoka K, Hasegawa A, Kiso M, Kannagi R (1993) Contribution of carbohydrate antigens sialyl
63 Lewis A and sialyl Lewis X to adhesion of human cancer cells to vascular endothelium. Cancer Res 53:354–361 Walker JJ (2000) Pre-eclampsia. Lancet 356:1260–1265 Zhou Y, Fisher SJ, Janatpour M, Genbacev O, Dejana E, Wheelock M, Damsky CH (1997a) Human cytotrophoblasts adopt a vascular phenotype as they diVerentiate. A strategy for successful endovascular invasion? J Clin Invest 99:2139–2151 Zhou Y, Damsky CH, Fisher SJ (1997b) Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. One cause of defective endovascular invasion in this syndrome? J Clin Invest 99:2152–2164 Zhou Y, McMaster M, Woo K, Janatpour M, Perry J, Karpanen T, Alitalo K, Damsky C, Fisher SJ (2002) Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol 160:1405–1423 Zhou Y, Genbacev O, Fisher SJ (2003) The human placenta remodels the uterus by using a combination of molecules that govern vasculogenesis or leukocyte extravasation. Ann N Y Acad Sci 995:73–83
123
