REGULAR ARTICLE

Distribution of Notch protein members in normaland preeclampsia-complicated placentas

Luigi Cobellis & Annunziata Mastrogiacomo &

Elisabetta Federico & Maria Teresa Schettino &

Maria De Falco & Lucrezia Manente &

Gabriele Coppola & Marco Torella & Nicola Colacurci &Antonio De Luca

Received: 6 March 2007 /Accepted: 12 September 2007 / Published online: 23 October 2007# Springer-Verlag 2007

Abstract Notch proteins are a transmembrane receptorfamily that is structurally and functionally conserved fromworms to humans. The mammalian family of Notch proteinsconsists of several genes encoding Notch receptors and relatedNotch ligands. Notch signaling is involved in different aspectsof the cell-fate decision tree: differentiation, proliferation, andapoptosis. These three processes are finely regulated in humanplacenta in order to allow a successful pregnancy and correctfetal growth. Notch and its ligands also participate in vascularremodeling and stabilization. Vasculogenesis and bloodregulation are of importance in the human placenta for normalfetal development and growth; any disorder of these systemsleads to preeclampsia. Drawing on this background, wehave investigated the expression of Notch-1, Notch-4, andJagged-1, together with two members related to the Notchpathway in angiogenesis: VEGF and p21. Normal andpreeclamptic human placentas have been evaluated byimmunohistochemistry. In preeclamptic samples, a down-regulation of Notch pathway members occurs with a weak/

moderate expression of the Notch protein members in allcomponents of placenta compared with physiological pla-centas that, at term, exhibit the strong expression of Jagged-1and a moderate expression of both Notch-1 and Notch-4 in allcompartments of the placental villi. Moreover, preeclampticsamples also reveal a down-regulation of VEGF expression,together with a moderate nuclear expression of p21Cip1 in thesyncytiotrophoblast, cytotrophoblast, and endothelial cells.This down-regulation of VEGF in preeclamptic placentas, inturn, probably decreases Notch protein expression inplacental compartments and in endothelial cells and couldoffer an ethiopathogenetic explanation for the onset of thispathology.

Keywords Placenta . Preeclampsia . Notch pathway .

Angiogenesis . Immunohistochemistry . Human

Introduction

The Notch signaling pathway was initially identified inDrosophila when a mutation in Notch was found to causean increase in the number of neuroblasts at the expense ofepidermal cells (Hoppe and Greenspan 1986). However, wenow know that this pathway is present in all metazoans,where it functions as one of the major pathways thatdetermine cell identity during development (Artavanis-Tsakonas et al. 1999; Egan et al. 1998). Notch proteins(Notch-1 to Notch-4 in vertebrates) are single-pass trans-membrane proteins that are activated by Delta (Dll1, 3, 4)and Jagged/Serrate (Jagged-1, -2) ligands (Egan et al. 1998;Fleming 1998). Each of these proteins shows cell-type- andtissue-specific expression during development. This ligand/receptor interaction leads to proteolytic cleavage and the

Cell Tissue Res (2007) 330:527–534DOI 10.1007/s00441-007-0511-6

DO00511; No of Pages

L. Cobellis :A. Mastrogiacomo : E. Federico :M. T. Schettino :M. Torella :N. ColacurciDepartment of Gynaecology, Obstetric, and ReproductiveScience, Second University of Naples,Naples, Italy

M. De FalcoDepartment of Biological Sciences, Section of Evolutionaryand Comparative Biology, University of Naples “Federico II”,Naples, Italy

L. Manente :G. Coppola :A. De Luca (*)Department of Medicine and Public Health,Section of Human Anatomy, Second University of Naples,Via L. Armanni 5,80138 Naples, Italye-mail: antonio.deluca@unina2.it

release of the intracellular domain of Notch (NICD), whichtranslocates into the nucleus and interacts with the DNA-binding protein RBP-Jκ (also known as CBF1 or suppres-sor of Hairless) encoded by the RBPSUH gene, and whichthen mediates the program of cell identity. This transcrip-tional activator complex induces the transcription of targetgenes, most notably the Hey and Hes genes (Diez et al.2007; Kadesch 2004). The Hey bHLH transcription factorfamily (Hey1, Hey2, and HeyL) is essential for varioussteps of embryonic development (Fischer and Gessler2003). The Notch pathway has been demonstrated to beinvolved in multiple aspects of vascular development andangiogenesis, but the mechanisms by which Notch exertsits effects on the vasculature remain to be elucidated.Mutations inhibiting the Notch pathway, and sustainedactivation of Notch, similarly affect vascular development,suggesting a requirement for fine temporal and spatialregulation of Notch signaling (Diez et al. 2007). Targeteddeletion of Notch-1 causes embryonic lethality because ofdefects in blood vessel development, and this phenotype isenhanced in Notch-1/Notch-4 double knock-out mice(Krebs et al. 2000; Huppert et al. 2000; MacKenzie et al.2004). Paradoxically, similar vascular abnormalities occurin transgenic mice expressing a constitutively activeNotch-4 mutant in the endothelium, demonstrating the needfor the critical regulation of Notch activity during vasculardevelopment (MacKenzie et al. 2004; Uyttendaele et al.2001). Notch activation might be absent in vessels at theearly stages of angiogenesis when endothelial cells areproliferating, but Notch might be reactivated when endo-thelial cells stop proliferating, and vessels begin to stabilize(Henderson et al. 2001; Taylor et al. 2002) in response tovarious factors such as vascular endothelial growth factor(VEGF, also VEGF-A) and fibroblast growth factor-2(Compagni et al. 2000; Colville-Nash and Willoughby1997; MacKenzie et al. 2004; Senger et al. 1983). Notchactivation can stimulate or inhibit proliferation by modu-lating the regulation of the cell cycle in a cell-type-specificand context-dependent manner (Carlesso et al. 1999; Jundtet al. 2002; Noseda et al. 2004; Rangarajan et al. 2001;Ronchini and Capobianco 2001; Sriuranpong et al. 2001;Weng et al. 2003). In human arterial endothelial cells,VEGF induces the up-regulation of Notch-1 and its ligand,Dll4, in vitro (Limbourg et al. 2005; Liu et al. 2003).Moreover, Noseda et al. (2004) have demonstrated that, inendothelial cells, Notch activation down-regulates p21Cip1.

Angiogenesis, the sprouting of new capillaries fromexisting vessels, is an essential component of placentadevelopment. In humans, vascular transformation occurs inthe placental bed in which extravillous trophoblast cellswithin the maternal decidual tissue transform the spiralarteries into high-flow low-resistance vessels that are moreefficient at supplying the placenta with maternal blood

(Meekins et al. 1994). The placenta is a relatively richsource of angiogenic growth factors, and the regulation ofvascular development in the placenta is thought to beintimately associated with such factors. In recent years,great emphasis has been focused on placenta tissue inwhich vascular remodeling and blood pressure and flow aredramatically important for normal fetal development andgrowth, and in which alterations of these elements causetypical pregnancy pathology: preeclampsia. Preeclampsia isone of the most important complications of pregnancy andis the leading cause of maternal and perinatal morbidity andmortality in the world. It is a complex clinical syndromepotentially involving all the organ systems. Despiteintensive research, the etiology of preeclampsia remainsunknown. Current hypotheses include vascular-mediatedfactors, placental ischemia, genetic predisposition, andimmune maladaptation, all of which have been stated tocontribute to the development of preeclampsia (Su et al.2001). The purpose of this study has been to investigate thepattern of expression of Notch protein members in normaland preeclampsia-complicated placentas. In order to clarifythe role of Notch signaling in preeclampsia-complicatedplacentas, we have also investigated the expression ofVEGF and p21, which are respectively supposed to bemembers upstream and downstream of the Notch signalingpathway in angiogenesis.

Materials and methods

Samples

Placentas from the third trimester of gestation wereobtained immediately after cesarean section (n=15). Sam-ples of pregnant patients with preeclampsia were obtainedfrom cesarean deliveries (n=15). All the samples wereobtained with informed consent of patients. The gestationalage of the third trimester specimens ranged from 28 to40 weeks. The gestational age of preeclamptic specimensranged from 35 to 40 weeks. All the collected specimenswere immediately fixed in formalin (n=45; 15 from eachtrimester of gestation and 15 from preeclampsia complicat-ed pregnancies) for immunohistochemistry. Representativesections of each specimen were stained with hematoxylin-eosin and examined by a pathologist to confirm histologicalpreservation of the microanatomical structure for normalsamples. Protocols involving human tissues in this researchwere approved by our institutional human research studiescommittee.

Preeclamptic placentas were selected as describedpreviously (Cobellis et al. 2007). In brief, sample categoryconsisted of normotensive pregnant patients and pregnantpatients with preeclampsia (an increase in systolic blood

528 Cell Tissue Res (2007) 330:527–534

pressure of 30 mm Hg, an increase in diastolic bloodpressure of 15 mm Hg, or blood pressure greater than140 mm Hg/90 mmHg as an absolute reading, and asignificant increase of proteinuria and of weight).

Immunoblotting

Human placental samples both of the third trimester ofgestation and preeclamptic placentas were homogenized at4°C in 300 μl lysis buffer (50 mM TRIS-HCl pH 7.4,5 mM EDTA, 250 mM NaCl, 50 mM NaF, 0.1% TritonX-100, 0.1 mM Na3VO4, 1 mM phenylmethane sulfonyl-fluoride, 10 mg/ml leupeptin) for 30 min in ice. Lysateswere centrifuged at 14,000g for 10 min at 4°C. Proteinlevels were tested by the Bradford assay and normalized byCoomassie blue staining. Protein (30 μg) was resolved by10% SDS-polyacrylamide gel electrophoresis, transferred topolyvinylidene fluoride (PVDF) membranes (Millipore) inCAPS buffer (10 mM CAPS, 20% methanol, pH 11.0). Theloading and transfer of equal amounts of proteins wereconfirmed after transfer to PVDF membranes by stainingthe membranes with Red Ponceau (Sigma, St. Louis, Mo.,USA). After extensive washes with TBS-T buffer (2 mMTRIS, 13.7 mM NaCl, 0.1% Tween-20, pH 7.6) to removethe Red Ponceau, the membranes were blocked with 5%milk in TBS-T buffer and then washed in TBS-T. Forimmmunoblotting, goat polyclonal antibodies (Santa CruzBiotechnology) raised against human Notch-1 (sc-6014),Notch-4 (sc-8643), and Jagged-1 (sc-6011), rabbit poly-clonal antibody raised against VEGF (sc-152), and mousemonoclonal raised against p21 (sc-817) were incubatedwith the membranes in 3% milk for 1 h (all antibodiesdiluted 1:100) and then washed in TBS-T. The membraneswere subsequently incubated with a suitable secondaryantibody coupled with horseradish peroxidase (diluted1:3,000; Amersham, Piscatanay, N.J., USA) and washedin TBS-T. The presence of secondary antibody bound to themembrane was detected by using the ECL system (DuPontNEN, Wilmington DE, USA). Immunoblotting with β-actin(Santa Cruz Biotechnology) was performed as a loadingcontrol.

Immunohistochemistry

Immunohistochemistry was carried out essentially asdescribed previously (Cobellis et al. 2007; De Falco et al.2007). Briefly, paraffin sections from each specimen werecut at a thickness of 5 μm, mounted on glass, dried overnightat 37°C, deparaffinized in xylene, rehydrated through agraded series of alcohol, and washed in phosphate-bufferedsaline (PBS). PBS was used for all subsequent washes andfor antiserum dilution. Tissue sections were quenchedsequentially in 3% hydrogen peroxide and blocked with

PBS plus 6% non-fat dry milk (Biorad) for 1 h at roomtemperature. Slides were then incubated at 4°C overnightwith goat polyclonal antibodies (Santa Cruz Biotechnology)raised against human Notch-1 (sc-6014), Notch-4 (sc-8643),and Jagged-1 (sc-6011) at a 1:100 dilution, with rabbitpolyclonal antibody raised against VEGF (sc-152) at a 1:100dilution, and with mouse monoclonal raised against p21(sc-817) and mouse monoclonal antibody raised againstCD-31 (DAKO, clone JC70A) at a 1:100 dilution. Afterseveral washes (3×5 min) to remove excess antibody, theslides were incubated with diluted anti-goat biotinylatedantibody, anti-rabbit biotinylated antibody, or anti-mousebiotinylated antibody (Vector Laboratories) for 1 h. All theslides were then processed by the ABC method (VectorLaboratories) for 30 min at room temperature. DAB (3,3-diaminobenzidine; Vector Laboratories) was used as thechromogen, and hematoxylin was used as the nuclearcounterstain. Negative controls for each tissue section wereprepared by substituting the primary antiserum with non-immune IgG. For each experiment, all slides were stained ina single batch and thus received equal staining. Immuno-histochemical staining intensity was evaluated and ranked as0 (absent), 1 (weak), 2 (moderate), or 3 (intense) as describedby Seelam et al. (2001).

For each specimen, an HSCORE value was derived bysumming the percentages of cells/areas that stained at eachintensity and multiplying this value by the weighted intensityof the staining. An average of 22 fields was observed foreach tissue by three observers at different times, and theaverage score was used. All values were expressed as themean±SEM, and differences were compared by usingStudent’s t-test.

Results

Pattern of expression of proteins of Notch pathwayin normal and preeclamptic placenta

To compare the expression levels of Notch pathway proteinsbetween normal and preeclamptic placentas, we performedimmunoblotting for: Notch-1, Notch-4, Jagged-1, VEGF,and p21. Figure 1 shows that the proteins are differentlymodulated between normal placenta and preeclampsia-complicated placenta. We detected a decreased expressionof all the members of Notch pathway in preeclampsia-complicated placentas compared with normal third trimesterplacentas. Specifically, we observed a marked decrease ofNotch-1 and Notch-4 expression on comparing normalplacentas with preeclamptic placentas during the thirdtrimester (Fig. 1a). Moreover, Jagged-1 and VEGF exhibiteda decrease of expression in preeclampsia-complicatedplacentas compared with normal third trimester placentas

Cell Tissue Res (2007) 330:527–534 529

(Fig. 1a,b). On the contrary, the expression levels of p21were similar in normal and preeclampsia-complicatedplacentas (Fig. 1b).

Distribution of Notch-1, Notch-4, and Jagged-1 in normaland preeclamptic placenta

We evaluated the localization and distribution of Notch-1,Notch-4, and Jagged-1 in normal and preeclamptic humanplacentas during the third trimester by immunohistochem-istry (for a description of the pattern of expression of Notchprotein members in human placenta throughout pregnancy,see De Falco et al. 2007). Samples from the physiologicaland preeclampsia-complicated placentas were processedunder the same conditions and compared. In preeclampticplacentas, we observed a down-regulation of all themembers of the Notch signaling pathway compared withnormal physiological placentas.

In agreement with our previous results, we observed that, innormal human placentas during the third trimester, Notch-1was localized at intense levels in the syncytiotrophoblastmembrane but was weak in the cytoplasm of cytotrophoblastand syncytiotrophoblast (Fig. 2a). In preeclampsia-complicated placentas, Notch-1 showed a low membraneimmunopositivity in the syncytiotrophoblast, in contrast to

Fig. 2 Localization of Notch signaling proteins in normal andpreeclamptic samples. a Notch-1 membrane localization in thesyncytiotrophoblast in normal placenta from the third trimester ofgestation. b Low Notch-1 membrane immunopositivity in thesyncytiotrophoblast of preeclamptic placenta. c Moderate Notch-4expression in the cytotrophoblast and endothelial cells (arrows) innormal placenta. d Lower Notch-4 localization in the cytotrophoblastand endothelial cells (arrow) of preeclamptic placental villi. e IntenseJagged-1 immunostaining in all the compartments of normal placenta(arrows endothelial cells). f Intermediate Jagged-1 immunopositivityin the preeclamptic placental villi (arrows endothelial cells). g IntenseCD31 localization in the cytoplasm of endothelial cells in the normalplacenta. h Representative negative control of a placenta from thethird trimester of gestation. Bar 30 μm

Fig. 1 Pattern of expression of proteins of Notch pathway in humanplacenta during pregnancy. Western blot analysis of Notch-1, Notch-4,Jagged-1, VEGF, and p21 in normal placentas (NP) and inpreeclampsia-complicated placentas (PP), at the third trimester ofgestation

530 Cell Tissue Res (2007) 330:527–534

undetectable staining in the cytotrophoblast, stroma, andendothelial cells (Fig. 2b). In normal placentas, Notch-4immunopositivity was weak in the syncytiotrophoblastcytoplasm and membrane and intermediate in the cytotro-phoblast. Its immunopositivity was strongly observed in theplacental stroma and in the cytoplasm of endothelial cells(Fig. 2c). In preeclamptic placentas, Notch-4 immuno-positivity was localized at a low level of expression in thecytoplasm of cytotrophoblast and in syncytiotrophoblast,together with moderate immunopositivity in the stroma andin the endothelial cells (Fig. 2d). In normal physiologicalplacentas, Jagged-1 was localized at intense levels in thecytoplasm of both cytotrophoblast and syncytiotrophoblast,together with strong immunopositivity in the placentalstroma and in the cytoplasm of endothelial cells (Fig. 2e).In preeclampsia, Jagged-1 showed a lower expression in thesyncytiotrophoblast, together with an intermediate expres-sion level in the cytotrophoblast, stroma, and endothelialcells (Fig. 2f). CD31 was used as a marker of endothelialcells (Fig. 2g).

The expression of Notch protein members, as detectedby immunohistochemical staining intensity analysis, innormal and preeclamptic human placentas during the thirdtrimester is depicted in Fig. 3. In the normal humanplacenta during third trimester, we observed that Notch-1was well expressed in the syncytiotrophoblast cells but waslower in the cytotrophoblast and stroma of placental villi(Fig. 3a), whereas Notch-4 had the opposite expression(Fig. 3b). Moreover, strong expression of Jagged-1 waspresent in all the placental components (Fig. 3c). On thecontrary, in preeclampsia-complicated human placenta, weobserved a down-regulation of Notch proteins in all theplacental compartments (Figs. 3a–c).

Distribution of VEGF and p21Cip1 in normaland preeclamptic placenta

In agreement with our previous results (De Falco et al.2007), we observed that, in normal human placentas duringthe third trimester, VEGF was localized almost exclusivelyin the cytoplasm of the cytotrophoblast at intense levels,together with an intense immunopositivity in the cytoplasmof endothelial cells of small capillaries inside the villousstroma (Fig. 4a). In preeclamptic placentas, we found thatVEGF was localized at moderate levels of expression in thecytotrophoblast (Fig. 4b). Moderate VEGF expressionlevels were detected in the cytoplasm of endothelial cellsinside placental villi (Fig. 4b).

p21Cip1 was expressed in normal placenta during thethird trimester at moderate/intense levels but was preva-lently localized in some nuclei of the syncytiotrophoblastand of endothelial cells (Fig. 4c). Moreover, in preeclampticplacentas, we observed a moderate immunopositivity for

p21Cip1 localized in the nuclei of the syncytiotrophoblast,cytotrophoblast, and endothelial cells. The p21Cip1 immuno-positivity in the stroma was almost absent (Fig. 4d). CD31was used as a marker of endothelial cells (Fig. 4e).

The expression of VEGF and p21Cip1, as detected innormal and preeclamptic human placentas during the thirdtrimester, was examined by immunohistochemical stainingintensity analysis (Fig. 5). In the normal human placenta,strong expression of VEGF was observed in the cytotropho-blast and in the endothelial cells inside the villous stroma incontrast to a lower expression in the syncytiotrophoblast. Onthe contrary, p21Cip1 showed weak expression in thecytotrophoblast and higher expression level in the syncytio-

Fig. 3 Intensity staining of cytotrophoblast, syncytiotrophoblast, andstromal cells for proteins of Notch signaling pathway. Expression levelof Notch-1 (a), Notch-4 (b), and Jagged-1 (c) in placental villi ofnormal and preeclamptic placentas (vertical lines SEM)

Cell Tissue Res (2007) 330:527–534 531

trophoblast and stroma in normal human placenta from thethird trimester of gestation. In preeclampsia-complicatedplacenta, VEGF had the same pattern of expression as innormal human placenta during the third trimester but with alower level of expression, whereas p21Cip1 showed a slightlydecreased expression in the syncytiotrophoblast and in thecytotrophoblast with localization almost exclusively in thenucleus.

Discussion

Preeclampsia is a widespread and potentially dangerousdisorder that may occur from 20 weeks of gestation (Bateset al. 2006; Borzychowski et al. 2006). It occurs in 3%–5%of first pregnancies and is characterized by polymorphicmaternal systemic disturbances and widespread endothelialdysfunction (Bates et al. 2006; Borzychowski et al. 2006;Roberts et al. 1989). Failure of adequate trophoblastinvasion and placental development results in an under-

perfused placenta, which is proposed to release unidentifiedproducts into the maternal circulation, resulting in aspectrum of clinical manifestations, including hypertension,proteinuria, cerebral edema and infarction, pulmonaryedema, liver hemorrhage, renal failure, and coagulopaty(Bates et al. 2006). However, preeclampsia is currentlyimpossible to predict or prevent as its precise causes areunknown, although many candidates have been proposed(Bates et al. 2006; Borzychowski et al. 2006). Increasingevidence suggests that an imbalance between factorspromoting angiogenesis, such as VEGF, and factors antag-onizing angiogenesis, such as soluble fms-like tyrosinekinase 1 (soluble FLT1), has a fundamental role in thepathogenesis of preeclampsia (Stepan et al. 2006). VEGF isan important survival factor for the endothelium and is thedominant angiogenic molecule in physiological and patho-logical angiogenesis; hence, VEGF systemic inhibitionwould cause generalized endothelial dysfunction (Bates etal. 2006; Borzychowski et al. 2006; Ferrara 2004).However, although the properties of VEGF have rapidlyled to its investigation as a potential pathophysiologicalmolecule in preeclampsia, there has been considerablecontroversy over the levels of circulating VEGF duringthe last 10 years (Baker et al. 1995; Bates et al. 2006;Sharkey et al. 1996). A large number of studies have shown

Fig. 5 Intensity staining of cytotrophoblast, syncytiotrophoblast, andstromal cells for VEGF (a) and p21Cip1 (b). Expression levels inplacental villi of normal and preeclamptic placentas are shown(vertical lines SEM)

Fig. 4 Localization of VEGF and p21 Cip1 proteins in normal andpreeclamptic samples. a Intense VEGF expression in the cytoplasm ofcytotrophoblast and endothelial cells (arrow) of normal placenta.b VEGF staining in cytotrophoblast and endothelial cells (arrow) insidethe villous stroma of preeclamptic placental villi. c Moderate/intenselevels of expression of p21Cip1 in the nuclei of the syncytiotrophoblastand endothelial cells (arrow) in normal placenta. d Moderate p21Cip1

expression in the nuclei of the syncytiotrophoblast (arrowhead) and inthe nuclei of endothelial cells (arrow) in preeclampsia-complicatedplacenta. e Intense CD31 localization in the cytoplasm of endothelialcells in the preeclamptic placenta. f Representative negative control ofa preeclamptic placenta. Bar 30 μm (a,b,e,f), 16 μm (c,d)

532 Cell Tissue Res (2007) 330:527–534

that free available VEGF levels are significantly reduced inpreeclampsia (Bates et al. 2006; Livingston et al. 2000;Reuvekamp et al. 1999). Therefore, although the totallevels of VEGF may be increased in this disorder, thephysiologically active concentration may actually be re-duced (Bates et al. 2006). In addition, in preeclampsia,placental soluble FLT1 is up-regulated, and these increasedFLT1 concentrations are accompanied by decreased levelsof free VEGF, suggesting that soluble FLT1 binds VEGF inthe maternal circulation and thereby blocks its angiogeniceffects (Levine et al. 2004; Stepan et al. 2006). In thepresent study, we have observed, by immunohistochemis-try, that VEGF is slightly down-regulated in preeclampticplacentas, viz., it exhibits moderate expression in thecytoplasm of the cytotrophoblast and endothelial cells ofplacental villi (see above), compared with physiologicalplacentas at term in which intense VEGF expression hasbeen found in the cytotrophoblast and in the endothelialcells (De Falco et al. 2007). In zebrafish embryos, VEGFhas been proposed to function as an inducer of Notchsignaling in immature endothelial progenitor cells, leadingto the adoption of arterial cell fate (Lawson et al. 2002;Diez et al. 2007) and pointing to the crucial role of Notchsignaling in arteriogenesis and angiogenesis. Thus, Notch/Delta is a critical downstream effector of the arterigenic andangiogenic response to VEGF (Liu et al. 2003). Notchsignals in combination with other cellular factors are able toinfluence apoptotic events, proliferation, and differentiationduring all stages of development. In particular, Notchsignaling seems to function as a general developmentaltool to direct cell fate and to build an organism (Weinmaster1997). Notch-4 functions in endothelial cells to regulateangiogenic remodeling of the vasculature and is predomi-nantly expressed in the vascular endothelium. Loss ofNotch-4 function in mice enhances the vascular remodelingdefects caused by loss of Notch-1 function (Krebs et al.2000; MacKenzie et al. 2004). However, Notch activationprevents mitogen-induced p21Cip1 up-regulation in endo-thelial cells providing a novel mechanism for Notch-mediated inhibition of proliferation (Noseda et al. 2004).The regulation of cell-fate choices is extremely important inorgans such as human placenta since this tissue arisesthrough the proliferation and differentiation of the blasto-cyst trophoblast cells. Our recent study of human placentahas shown that Notch proteins are differently modulatedthroughout the pregnancy with a slight decreased expres-sion of Notch-1, in contrast with a stable intense expressionof Notch-4 and a strong increase in Jagged-1 expressionlevels. Although little variation is seen in the localizationand distribution of Notch pathway members throughoutpregnancy, a strong difference is evident in the expressionlevel of these proteins. In particular, during the thirdtrimester of pregnancy, we have observed the strong

expression of Jagged-1 in all the placental compartmentsand a decrease of both Notch-1 and Notch-4. These datasupport the hypothesis that high levels of ligand accompaniedby a lower steady-state level of receptors enhances theactivation of the Notch pathway. Thus, we have proposedthat VEGF is able to induce an increase of Jagged-1expression, and that the activation of Notch/Jagged probablyaffects p21Cip1 expression, which not only decreasesthroughout the pregnancy, but also shifts from the nucleusto the cytoplasm of the syncytiotrophoblast and endothelialcells during the third trimester of gestation (De Falco et al.2007). These observations are consistent with data publishedby Liu et al. (2003) demonstrating that VEGF is able toinduce gene expression of Notch-1 in human arterialendothelial cells and also of those by Noseda et al. (2004)showing that, in endothelial cells, Notch activation down-regulates p21Cip1.

Drawing on this background, we decided to compare thepattern of expression of Notch-1, Notch-4, Jagged-1,VEGF, and p21Cip1 in normal and preeclamptic humanplacentas during the third trimester. Moreover, we observeda down-regulation of Notch pathway members in all theplacental compartments compared with their expression inphysiological placentas. The Notch down-regulation prob-ably was induced by the decreased expression of VEGFobserved in preeclamptic placentas. Furthermore, in pre-eclamptic samples, we observed a slight decrease of p21Cip1

expression, although we also observed that, contrary tonormal human placentas during the third trimester ofgestation, p21Cip1 localized in the nuclei of the syncytio-trophoblast, endothelial cells, and also cytotrophoblastcells, thus suggesting its permanent activation in these cells.

Taken together, our data direct attention to a VEGF/Notch pathway imbalance in preeclamptic placentas. Thisimbalance could be the cause of poor placentationassociated with the onset of preeclampsia.

References

Artavanis-Tsakonas S, Rand MD, Lake RJ (1999) Notch signaling:cell fate control and signal integration in development. Science284:770–776

Baker PN, Krasnow J, Roberts JM, Yeo KT (1995) Elevated serumlevels of vascular endothelial growth factor in patients withpreeclampsia. Obstet Gynecol 86:815–821

Bates DO, Macmillan PP, Manjaly JG, Qiu Y, Hudson SJ, Bevan HS,Hunter AJ, Soothill PW, Read M, Donaldson LF, Harper SJ(2006) The endogenous anti-angiogenic family of splice variantsof VEGF, VEGFxxxb, are down-regulated in pre-eclampticplacentae at term. Clin Sci 110:575–585

Borzychowski AM, Sargent IL, Redman CWG (2006) Inflammationand pre-eclampsia. Semin Fetal Neonatal Med 11:309–316

Carlesso N, Aster JC, Sklar J, Scadden DT (1999) Notch1-induceddelay of human hematopoietic progenitor cell differentiation isassociated with altered cell cycle kinetics. Blood 93:838–848

Cell Tissue Res (2007) 330:527–534 533

Cobellis L, De Falco M, Mastrogiacomo A, Giraldi D, Dattilo D,Scaffa C, Colacurci N, De Luca A (2007) Modulation of apelinand APJ receptor in normal and preeclampsia-complicatedplacentas. Histol Histopathol 22:1–8

Colville-Nash PR, Willoughby DA (1997) Growth factors in angio-genesis: current interest and therapeutic potential. Mol MedToday 3:14–23

Compagni A, Wilgenbus P, Impagnatiello MA, Cotten M, ChristoforiG (2000) Fibroblast growth factors are required for efficienttumor angiogenesis. Cancer Res 60:7163–7169

De Falco M, Cobellis L, Giraldi D, Mastrogiacomo A, Perna A,Colacurci N, Miele L, De Luca A (2007) Expression anddistribution of Notch protein members in human placentathroughout pregnancy. Placenta 28:118–126

Diez H, Fischer A, Winkler A, Hu C-J, Hatzopoulos AK, Breier G,Gessler M (2007) Hypoxia-mediated activation of Dll4-Notch-Hey2 signaling in endothelial progenitor cells and adoption ofarterial cell fate. Exp Cell Res 313:1–9

Egan SE, St-Pierre B, Leow CC (1998) Notch receptors, partners andregulators: from conserved domains to powerful functions. CurrTop Microbiol Immunol 228:273–324

Ferrara N (2004) Vascular endothelial growth factor: basic science andclinical progress. Endocr Rev 25:581–611

Fischer A, Gessler M (2003) Hey genes in cardiovascular develop-ment. Trends Cardiovasc Med 13:221–226

Fleming RJ (1998) Structural conservation of Notch receptors andligands. Semin Cell Dev Biol 9:599–607

Henderson AM, Wang SJ, Taylor AC, Aitkenhead M, Hughes CC(2001) The basic helix-loop-helix transcription factor HESR1regulates endothelial cell tube formation. J Biol Chem 276:6169–6176

Hoppe PE, Greenspan RJ (1986) Local function of the Notch gene forembryonic ectodermal pathway choice in Drosophila. Cell46:773–783

Huppert SS, Le A, Schroeter EH, Mumm JS, Saxena MT, Milner LA,Kopan R (2000) Embryonic lethality in mice homozygous for aprocessing deficient allele of Notch1. Nature 405:966–970

Jundt F, Anagnostopoulos I, Forster R, Mathas S, Stein H, Dorken B(2002) Activated Notch1 signaling promotes tumor cell prolifer-ation and survival in Hodgkin and anaplastic large celllymphoma. Blood 99:3398–3403

Kadesch T (2004) Notch signaling: the demise of elegant simplicity.Curr Opin Genet Dev 14:506–512

Krebs LT, Xue Y, Norton CR, Shutter JR, Maguire M, Sundberg JP,Gallahan D, Closson V, Kitajewski J, Callahan R, Smith GH,Stark KL, Gridley T (2000) Notch signaling is essential forvascular morphogenesis in mice. Genes Dev 14:1343–1352

Lawson ND, Vogel AM, Weinstein BM (2002) Sonic hedgehog andvascular endothelial growth factor act upstream of the Notchpathway during arterial endothelial differentiation. Dev Cell3:127–136

Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF,Schisterman EF, Thadhani R, Sachs BP, Epstein FH, Sibai BM,Sukhatme VP, Karumanchi SA (2004) Circulating angiogenicfactors and the risk of preeclampsia. N Engl J Med 350:672–683

Limbourg FP, Takeshita K, Radtke F, Bronson RT, Chin MT, Liao JK(2005) Essential role of endothelial Notch1 in angiogenesis.Circulation 111:1826–1832

Liu ZJ, Shirakawa T, Li Y, Soma A, Oka M, Dotto GP, Fairman RM,Velazquez OC, Herlyn M (2003) Regulation of Notch1 and Dll4by vascular endothelial growth factor in arterial endothelial cells:implications for modulating arteriogenesis and angiogenesis. MolCell Biol 23:14–25

Livingston JC, Chin R, Haddad B, McKinney ET, Ahokas R, SibaiBM (2000) Reductions of vascular endothelial growth factor and

placental growth factor concentrations in severe preeclampsia.Am J Obstet Gynecol 183:1554–1557

MacKenzie F, Duriez P, Larrivée B, Chang L, Pollet I, Wong F, Yip C,Karsan A (2004) Notch4-induced inhibition of endothelialsprouting requires the ankyrin repeats and involves signalingthrough RBP-Jκ. Blood 104:1760–1768

Meekins JW, Pijnenborg R, Hanssens M, McFadyen IR, Asshe A van(1994) A study of placental bed spiral arteries and trophoblastinvasion in normal and severe preeclamptic pregnancies. Br JObstet Gynaecol 101:669–674

Noseda M, Chang L, McLean G, Grim JE, Clurman BE, Smith LL,Karsan A (2004) Notch activation induces endothelial cell cyclearrest and participates in contact inhibition: role of p21Cip1

repression. Mol Cell Biol 24:8813–8822Rangarajan A, Talora C, Okuyama R, Nicolas M, Mammucari C,

Oh H, Aster JC, Krishna S, Metzger D, Chambon P, Miele L,Aguet M, Radtke F, Dotto GP (2001) Notch signaling is a directdeterminant of keratinocyte growth arrest and entry intodifferentiation. EMBO J 20:3427–3436

Reuvekamp A, Velsing-Aarts FV, Poulina IE, Capello JJ, Duits AJ(1999) Selective deficit of angiogenic growth factors characterisespregnancies complicated by pre-eclampsia. Br J Obstet Gynaecol106:1019–1022

Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA,McLaughlin MK (1989) Preeclampsia: an endothelial celldisorder. Am J Obstet Gynecol 161:1200–1204

Ronchini C, Capobianco AJ (2001) Induction of cyclin D1 transcrip-tion and CDK2 activity by Notch(ic): implication for cell cycledisruption in transformation by Notch(ic). Mol Cell Biol21:5925–5934

Seelam B, Kayisli UA, Mulayim N, Arici A (2001) Regulation of Fasligand expression by estradiol and progesterone in humanendometrium. Biol Reprod 65:979–985

Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, DvorakHF (1983) Tumor cells secrete a vascular permeability factor thatpromotes accumulation of ascites fluid. Science 219:983–985

Sharkey AM, Cooper JC, Balmforth JR, McLaren J, Clark DE,Charnock-Jones DS, Morris NH, Smith SK (1996) Maternal plasmalevels of vascular endothelial growth factor in normotensivepregnancies and in pregnancies complicated by pre-eclampsia. EurJ Clin Invest 26:1182–1185

Sriuranpong V, Borges MW, Ravi RK, Arnold DR, Nelkin BD,Baylin SB, Ball DW (2001) Notch signaling induces cellcycle arrest in small cell lung cancer cells. Cancer Res 61:3200–3205

Stepan H, Faber R, Dornhofer N, Huppertz B, Robitzki A, Walther T(2006) New insights into the biology of preeclampsia. BiolReprod 74:772–776

Su YN, Lee CN, Cheng WF, Shau WY, Chow SN, Hsieh FJ (2001)Decreased maternal serum placenta growth factor in early secondtrimester and preeclampsia. Obstet Gynecol 97:898–904

Taylor KL, Henderson AM, Hughes CC (2002) Notch activationduring endothelial cell network formation in vitro targets thebasic HLH transcription factor HESR-1 and downregulatesVEGFR-2/KDR expression. Microvasc Res 64:372–383

Uyttendaele H, Ho J, Rossant J, Kitajewski J (2001) Vascularpatterning defects associated with expression of activated Notch4in embryonic endothelium. Proc Natl Acad Sci USA 98:5643–5648

Weinmaster G (1997) The ins and outs of Notch signaling. Mol CellNeurosci 9:91–102

Weng AP, Nam Y, Wolfe MS, Pear WS, Griffin JD, Blacklow SC,Aster JC (2003) Growth suppression of pre-T acute lymphoblasticleukemia cells by inhibition of Notch signaling. Mol Cell Biol23:655–664

534 Cell Tissue Res (2007) 330:527–534