lable at ScienceDirect

Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99

Contents lists avai

Placenta

journal homepage: www.elsevier .com/locate/placenta

IFPA Meeting 2010 Workshops Report II: Placental pathology; Trophoblastinvasion; Fetal sex; Parasites and the placenta; Decidua and embryonic or fetalloss; Trophoblast differentiation and syncytialisation

A. Al-Khan a, I.L. Aye b, I. Barsoum c, A. Borbely d, E. Cebral e, G. Cerchi f, V.L. Clifton g, S. Collins h,T. Cotechini c, A. Davey i, J. Flores-Martin j, T. Fournier k, A.M. Franchi l, R.E. Fretesm, C.H. Grahamc,G. Godbole n, S.R. Hansson o, P.L. Headley p, C. Ibarra q, A. Jawerbaum r, U. Kemmerling s, Y. Kudo t,P.K. Lala u, L. Lassance v, R.M. Lewisw, E. Menkhorst x, C. Morris y, T. Nobuzane t, G. Ramos z, N. Rote aa,R. Saffery bb, C. Salafia cc, D. Sarr dd, H. Schneider ee, C. Sibley ff, A.T. Singh b, T.S. Sivasubramaniyamgg,M.J. Soares hh, O. Vaughan ii, S. Zamudio jj, G.E. Lash kk,*,1

aDepartment of Obstetrics and Gynaecology, University of California San Diego, San Diego, CA, USAb School of Women’s and Infants’ Health, University of Western Australia, AustraliacDepartment of Anatomy and Cell Biology, Queen’s university, Kingston, CanadadDepartment of Developmental and Cellular Biology, Institute of Biomedical Sciences, University of Sao Paulo, Brazile Laboratory of Reproduction and Maternal-Embryonic Physiopatology, IFIBYNE-CONICET, Biodiversity and Experimental Biology Department, FCEN, University of Buenos Aires,Buenos Aires, ArgentinafCEBBAD, Universidad Maimonides, ArgentinagRobinson Institute, Department of Paediatrics and Reproductive Health, University of Adelaide, Adelaide, AustraliahNuffield Dept of Obstetrics and Gynaecology, University of Oxford, UKiDepartment of Physiology, University of Alberta, Edmonton, Canadaj Facultad de Ciencias Químicas, UNC, CIBICI-CONICET, Argentinak INSERM, University of Paris Descartes, Paris, Francel Laboratory of Physiopathology of Pregnancy and Labor, CEFYBO-CONICET, School of Medicine, University of Buenos Aires, Buenos Aires, ArgentinamDepartment of Cell Biology, Histology and Embryology, Universidad Nacional Córdoba and La Rioja, ArgentinanMolecular and Cellular Biology Lab, National Institute for Research in Reproductive Health, Parel, Mumbai, IndiaoDepartment of Immunotechnology, Lund University, Lund, SwedenpDepartment of Clinical Biochemistry and Immunology, Statens Serum Institut, Denmarkq Physiology Department, School of Medicine, University of Buenos Aires, Buenos Aires, Argentinar Laboratory of Reproduction and Metabolism, CEFYBO-CONICET, School of Medicine, University of Buenos Aires, Argentinas Program of Anatomy and Developmental Biology, University of Chile, Santiago, ChiletDepartment of Obstetrics and Gynaecology, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, JapanuDepartment of Anatomy & Cell Biology, University of Western Ontario, London, CanadavMedical University of Graz, Austriaw The Institute of Developmental Sciences, University of Southampton, Southampton, UKx Prince Henry’s Institute, Clayton, Victoria, Australiay Tulane University, New Orleans, USAzUC San Diego, USAaaDepartment of Obstetrics and Gynaecology, MacDonald Women’s Hospital, Case School of Medicine Cleveland, OH, USAbbDevelopmental Epigenetics Group, Murdoch Childrens Research Institute, Royal Children’s Hospital Melbourne, Parkville, Victoria, Australiacc Placental Analytics, LLC, NY, USAddDepartment of Infectious Diseases, Center for Tropical and Emerging Global Diseases, University of Georgia, Athens, USAeeDepartment of Obstetrics and Gynecology, Insel Spital, University of Berne, Berne, SwitzerlandffMaternal and Fetal Health Research Centre, The University of Manchester, Manchester, UKgg Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto, Canadahh Institute for Reproductive Health and Regenerative Medicine, Department of Pathology, Laboratory Medicine, University of Kansas Medical Center, Kansas City, USAiiCentre for Trophoblast Research, Department of Physiology, Development and Neuroscience, Cambridge, UKjjUMDNJ-New Jersey Medical School and Hackensack University Medical Center, NJ, USAkkReproductive and Vascular Biology Group, Institute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK

* Corresponding author. Institute of Cellular Medicine, 3rd Floor, William Leech Building, Newcastle University, Newcastle upon Tyne NE2 4HH, UK. Tel.: þ44 191 222 8578;fax: þ44 191 222 5066.

E-mail address: g.e.lash@ncl.ac.uk (G.E. Lash).1 GEL edited this manuscript based on contributions from the other authors.

0143-4004/$ e see front matter � 2011 Published by IFPA and Elsevier Ltd.doi:10.1016/j.placenta.2010.12.025

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99 S91

a r t i c l e i n f o

Article history:Accepted 21 December 2010

Keywords:PlacentaTrophoblastWorkshops

a b s t r a c t

Workshops are an important part of the IFPA annual meeting. At IFPA Meeting 2010 diverse topics werediscussed in twelve themed workshops, six of which are summarized in this report. 1. The placentalpathology workshop focused on clinical correlates of placenta accreta/percreta. 2. Mechanisms ofregulation of trophoblast invasion and spiral artery remodeling were discussed in the trophoblastinvasion workshop. 3. The fetal sex and intrauterine stress workshop explored recent work on placentalsex differences and discussed them in the context of whether boys live dangerously in the womb.4. Theworkshop on parasites addressed inflammatory responses as a sign of interaction between placentaltissue and parasites. 5. The decidua and embryonic/fetal loss workshop focused on key regulatorymediators in the decidua, embryo and fetus and how alterations in expression may contribute todifferent diseases and adverse conditions of pregnancy. 6. The trophoblast differentiation and syncyti-alisation workshop addressed the regulation of villous cytotrophoblast differentiation and how varia-tions may lead to placental dysfunction and pregnancy complications.

� 2011 Published by IFPA and Elsevier Ltd.

1. Placental pathology and clinical correlates(accrete/percreta)

Chairs: Abdulluh Al-Khan, Carolyn SalafiaSpeakers: Alexandre Borbely, Sally Collins, Gladys Ramos

1.1. Outline

Thisworkshop focused on newclinical and basic science insightsinto the mechanisms underlying abnormal trophoblast invasionand/or uterine pathology that result in placenta accreta/percreta.Discussion centered on the epidemiology, clinical presentation, andcomplications of these disorders, aswell asmedical costs. One of thegoals was to identify areas of research in need of more intensiveinvestigation.

1.2. Summary

Gladys Ramos discussed sonographic findings associated withthe diagnosis of placenta accreta. A number of sonographic signshave been described as indicative of placenta accreta although theaccuracy of these signs in predicting invasive placentation is notknown. A diagnostic accuracy study in which cases of pathology-proven placenta accreta were compared to cases of placenta previamatched by gestational age was performed whereby sonographicimages were reviewed and ultrasound signs of placenta accreta(previa, loss of myometrial interface, chaotic intraplacental bloodflow, ‘swiss cheese appearance’, placenta bulging into bladder,distance of bulge into the bladder, invasion into the bladder, andcolor Doppler crossing vessels) were coded as present, absent orindeterminate. The most common and sensitive sonographicfinding associated with placenta accreta was loss of myometrialinterface. If present all signs were highly specific (90e100%) andhad a high positive predictive value (90e100%) for the diagnosis ofplacenta accreta. No combination of sonographic signs couldaccurately predict all placenta percretas. However, placenta per-cretas often had more than one sonographic sign. It was concludedthat the presence of a previawith loss of myometrial interface is themost common finding in patients with placenta accreta.

Alexandre Borbely discussed expression of biglycan in normalhuman termplacenta and inhighly invasivepathologies.Decorin andbiglycan are members of the small leucine-rich proteoglycan familyand are important factors in the control of cellular proliferation,migration and invasion. Several placental pathologies includeincreased invasive activity of the trophoblast. These pathologies canbe lethal for both mother and fetus and they can develop during

gestation asplacenta accreta and invasivemoleor after gestation, likeclassic choriocarcinoma. The expression and localization patterns ofdecorin and biglycan in normal term placenta, placenta accreta,invasive mole and choriocarcinoma were determined. In normalterm placenta, decidual cells were immunopositive for decorinwhereas the extravillous trophoblast cells (EVT) were negative. Inplacenta accreta and invasive mole, EVT were positive for decorinwhereas the surrounding matrix was negative. In choriocarcinoma,only cytotrophoblast and metastatic cells were immunoreactive fordecorin. Biglycan displayed similar results to decorin, with exceptionof the matrix-type fibrinoid in normal term placenta and the EVTsurrounding matrix in placenta accreta, which presented strongstaining for biglycan. These results suggest that the expressionpatterns of biglycan in highly invasive cells from placenta accreta,invasive mole and choriocarcinoma might play a role in modulatingtrophoblast migration and invasion. The expression pattern pre-sented by decorin suggests that this proteoglycan might exertanother role than migration/invasion modulation.

In a comment from the floor, Peeyush K. Lala mentioned howprevious studies have shown that decidua-derived decorin isa paracrine regulator of extravillous trophoblast (EVT) invasivenessduring normal pregnancy. It serves as a storage device for inactiveTGF-b in the decidual extracellular matrix, until the decorin-TGF-b complex is cleaved by an EVT-derived proteinase cascade that alsoactivates TGF-b to exert its anti-invasive action. In addition, decorinexerts anti-proliferative, anti-migratory and anti-invasive actionson EVT independent of TGF-b, because of its binding as an antag-onistic ligand to several tyrosine kinase receptors such as EGFR,IGF-1R, MET and VEGF-R2 on the EVT. Since normal EVT nevershows immuno-detectable decorin, whereas EVT in placentaaccreta does, it was suggested that careful ultrastructural localiza-tion and in situ hybridization are needed to ascertain the source andlocation of decorin in placenta accreta.

Sally Collins discussed “spirals” and “spurts” and small forgestational age (SGA) babies.Inadequate transformation of thespiral arteries has long been implicated in adverse pregnancyoutcomes.A novel technique for measuring the spiral arterywaveform has recently been developed and it appears to bealtered in SGA babies.Women were recruited prospectively to haveseven extra ultrasound scans throughout pregnancy. Customizedbirth weight centiles were calculated with SGA defined as <10thcentile. A mixed model statistical analysis was used which takesinto account the longitudinal nature of the data. In the study 66women were recruited, 9 of which had SGA babies. The pulsatilityindex of the spiral artery ‘jets’ was increased in SGA pregnanciescompared to those with normal birth weights. A similar trend was

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99S92

seen in uterine arteries. Increased pulsatility in blood flowingforward into the intervillous space probably reflects the persis-tence of muscularity in the spiral artery wall. Despite this, nosignificant difference was seen in the uterine artery, adding weightto the theory that intra-myometrial shunts play an important partin the placental circulation.

1.3. Conclusions

Placenta accreta and percreta are increasingly importantand common obstetric catastrophes about which relatively little isunderstood. While the rise in cesarean sections has been blamedfor the increased incidence, the exact mechanism (basalis defectresulting from surgery or an abnormal vascularization resulting fromrepair of surgery) remains elusive. Multidisciplinary approaches tothis life-threatening disorder are clearly needed.

2. Trophoblast invasion

Chairs: Charles Graham, Peeyush LalaSpeakers: Ivraym Barsoum, Georgina Cerchi, Tiziana Cotechini,

Thierry Fournier, Paula Headley, Charles Graham, Geeta Godbole,Peeyush Lala, Ellen Menkhorst

2.1. Outline

This workshop aimed to discuss (a) mechanisms involved introphoblast invasion, and (b) mechanisms involved in uterinevascular remodeling and the association between the two events.

2.2. Summary

Peeyush Lala discussed whether mechanistic studies oftrophoblast invasion identify early markers for pre-eclampsia.Highly regulated trophoblast invasion of the decidua and the utero-placental arterioles are key events in the development of the feto-placental unit during early pregnancy. Compromised placentalperfusion resulting from inadequate trophoblast invasivenessduring early gestation has been linked with the origin of fetalgrowth restriction and pre-eclampsia. There are many factorsinvolved in regulating trophoblast invasion to maintain utero-placental homeostasis, including growth factors, growth factorbinding proteins, proteoglycans, enzymes, lipids and extracellularmatrix (ECM) components. Trophoblast invasiveness can bepromoted by autocrine or paracrine mediators: e.g. EGF and VEGFfamily (paracrine), IGF2 (autocrine), HGF (paracrine), uPA (auto-crine), IGF-BP-1 (paracrine), IET-1 (autocrine/paracrine), IL-8, IP-10,and PGE2 (paracrine). On the other hand, trophoblast invasivenessis controlled primarily by paracrine (decidua-derived) mediators,including TGF-b, TNFa, and decorin. Decorin can be an antagonisticligand for multiple tyrosine kinase receptors EGFR, IGFR-1, MET,and VEGF-R2 expressed by EVT cells. A derangement in any ora combination of these regulators may underlie pre-eclampsia,which is a multifactorial disease. Some of them may result inmaternal serum markers in early-mid pregnancy, which may be ofpredictive value in longitudinal studies; whereas other markersreported during clinically identified disease may represent effectsof the disease. Identification of early markers that can be harnessedfor the prevention or management of early-onset pre-eclampsia isurgently needed.

Ellen Menkhorst discussed the relative roles of decidualizedand non-decidualized stromal products in regulating trophoblastadhesion. EVT must adhere, migrate and invade through thedecidua during placental formation. Abnormal decidualization ofendometrial stromal cells leads to unregulated trophoblast invasion

and pregnancy failure inmice. Recent evidence in humans indicatespre-eclampsia may be associated with impaired decidualization.The mechanisms by which decidual cells interact with EVT and theimportance of adequate decidualization in this interaction remainlargely unknown.The regulation of EVT protein expression andadhesion by factors secreted by non-decidualized and decidualizedendometrial stromal cells was assessed. EVT showed increasedadhesion to decidualized compared to non-decidualized endome-trial stromal cells. Non-decidualized and decidualized endometrialstromal cell conditioned media stimulated the expression ofdistinct proteins in isolated EVT, some of which are associated withplacental pathologies including pre-eclampsia. These data suggestthat cell surface and secreted decidual cell factors regulate EVTadhesion and proteins. Adequate decidualization may be criticalfor appropriate EVT function and therefore placental developmentin humans.

Georgina Cerchi discussed the activity and expression of matrixmetalloproteinases in a human uterine fibroblast-trophoblast co-culture system. Successful implantation is the result of complexmolecular interactions between the hormonally primed uterus andmature blastocyst. Ethical restrictions and limited availability ofhuman placental tissue limits human implantation studies. An invitro co-culture model using human uterine fibroblasts andtrophoblast explants, both from term placenta, was established tostudy maternalefetal interaction. Matrix metalloproteinase (MMP)protein expression and activity was assessed by Western blot andzymography respectively. Pro-MMP-2 expressionwas up-regulatedin uterine fibroblast cells after co-culture with trophoblastcompared with controls; in contrast, MMP-2 expressionwas down-regulated. In trophoblast explants from co-culture, MMP-2expressionwas up-regulated. Pro-MMP-2 andMMP-2 activity wereup-regulated in co-culture conditions while MMP-2 activity wasdown-regulated. The model established may prove a good tool forfurther study of the role of MMPs during implantation.

Geeta Godbole discussed the role of decidual homeobox (HOX)A10 in controlling trophoblast invasion by regulating paracrinefactors. The decidua is a major regulator of trophoblast invasion,which dictates the degree of placentation. However, the genetic/molecular players within the decidua regulating this invasion havenot been identified. HOXA10 is a major regulator of decidualizationand was hypothesized to be involved in regulation of trophoblastinvasion. To investigate the role of decidual factors and decidualHOXA10 on trophoblast invasion, supernatants from decidualizedand HOXA10 silenced decidual-endometrial stromal cells wereused to investigate their effects on invasion and expression ofinvasion-related markers on trophoblast cell lines JEG-3 and ACH-3P.Decidual supernatants activated invasion of trophoblast celllines, and elevated expression of MMPs; the loss of HOXA10 in thedecidual cells further augmented this response. Supernatantsderived from HOXA10 knockdown cells increased the invasiveability of the trophoblast cells and activated expression of MMPsand integrins. These observations indicate that HOXA10 in decidualcells may balance the expression of pro- and anti-invasive factorssecreted by the decidua for controlled invasion and that loss of thisfactor in the decidual cells disrupts this balance leading to overinvasion. Additional data suggests that HOXA10might be critical fortransformation of endometrial stromal cells into functionaldecidual cells.

Thierry Fournierdiscussedhowinvasionof CMV-infectedhumancytotrophoblasts is decreased in a PPARg-dependent manner.Human implantation involves a major invasion of the uterine walland complete remodeling of uterine arteries by EVT. Abnormality inthese early steps of placental development leads to poor placenta-tion and fetal growth defects and is often associated with pre-eclampsia. It has previously been shown that activation of the

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99 S93

ligand-activated nuclear receptor PPARg with synthetic (rosiglita-zone) or natural (15deoxyPGJ2) agonists inhibits trophoblast inva-sion. Interestingly, it was demonstrated that for its own replicationhuman cytomegalovirus (HCMV) activates PPARg transcriptionalactivity in infected cells. It has also been shown that PPARg antag-onists dramatically impair virus production and that the majorimmediate-early promoter contains PPRE able to bind PPARg asassessed by EMSA and ChIP. By activating PPARg, HCMV dramati-cally impaired early human trophoblastmigration and invasiveness,as assessed by using EVT primary cultures and the EVT-derived cellline HIPEC. These data provide new clues to explain how earlyinfection during pregnancy could impair implantation and placen-tation and therefore embryonic development.

Paula Hedley discussed leptin and soluble leptin receptors infirst trimester pregnancy sera as markers for pre-eclampsia. Thematernal serum concentration of leptin is increased in the secondand third trimesters of pregnancies afflicted by pre-eclampsia.Leptin, its soluble receptor, and the free leptin index in firsttrimester sera from 123 women who developed pre-eclampsia and285 controls were assessed. Maternal serum leptin levels increasedin pre-eclampsia compared to controls, while soluble receptorconcentrations were reduced. There was no correlation betweenleptin or soluble receptor levels and the severity of pre-eclampsia.In severe pre-eclampsia cases there was a negative correlationbetween TNFa and free leptin index, whereas inmild pre-eclampsiacases, this correlation was positive. There was no correlationbetween TNFa and free leptin index in control pregnancies. Thesedata indicate a role for leptin and free leptin index as earlyscreening markers for pre-eclampsia and a prognostic role of TNFaand free leptin index in differentiating severe from mild pre-eclampsia

Charles Graham discussed the link between aberrant maternalinflammation and deficient utero-placental perfusion leading topregnancy complications. Trophoblast invasion and remodeling ofthe uterine spiral arteries is required for the establishment ofadequate utero-placental perfusion. Trophoblast invasion is finelyregulated and requires the participation of locally producedmolecules, some of which stimulate invasion, whereas others, suchas TGF-b, inhibit invasion. Various complications of pregnancy (e.g.pre-eclampsia and fetal growth restriction) are characterized bydeficient remodeling of the spiral arteries. These complications arealso often associated with aberrant maternal inflammation linkedto macrophage infiltration around the spiral arteries. Recentresearch has revealed that activated macrophages inhibit tropho-blast invasiveness in vitro via a mechanism dependent on TNFa.Studies with rats indicate that aberrant maternal inflammationleads to utero-placental vascular abnormalities resulting inimpaired utero-placental blood flow and fetal morbidities. Defi-cient utero-placental perfusion resulting from maternal inflam-mation may be linked to poor remodeling of the spiral arteriesand/or maternal coagulopathies. These findings provide insight onthe role of inflammation in pregnancy complications.

Tiziana Cotechini discussed how inflammation-mediated fetalgrowth restriction in a rat model is associated with shallowtrophoblast invasion. There is evidence that fetal growth restric-tion is causally linked to deficient remodeling of the uterine spiralarteries leading to inadequate placental perfusion. This pregnancycomplication is also characterized by aberrant maternal inflam-mation. Using a model in which pregnant rats are injected withrepeated low doses of lipopolysaccharide (LPS), it was shown thatabnormal maternal inflammation disrupts trophoblast invasionand placental hemodynamics leading to fetal growth restriction.Pups from rats receiving daily LPS injections beginning at10 mg/kg on gestational day (GD) 13.5 followed by 40 mg/kg on GD14.5, 15.5 and 16.5 were growth-restricted on GD 17.5. Compared

to control rats, immunohistochemistry revealed fewer interstitialtrophoblast cells invading the mesometrial triangle in LPS-treatedanimals. Ultrasound analysis revealed a decrease in spiral arteryflow velocity following LPS exposure. These data indicatethat inflammation-associated fetal growth restriction may bea consequence of deficient trophoblast invasion and remodelingof the spiral arteries.

Ivraym Barsoum discussed whether nitric oxide donors canprevent pre-eclampsia. Pre-eclampsia is associated with increasedcirculating levels of pro-inflammatory molecules such as solublefms-like tyrosine kinase 1 (sFlt-1) and soluble endoglin (sEng).In hypoxic placentas sFlt-1/sEng are released into the circulationand are associated with the hypertension and proteinuria thatcharacterize pre-eclampsia. Low concentrations of the nitricoxide (NO) mimetic nitroglycerin inhibited the accumulation ofhypoxia-inducible factor 1a (HIF-1a), an important mediator ofadaptations to hypoxia, and thereby interferedwith hypoxic effects.Therefore, it was hypothesized that NO mimetics can inhibitthe hypoxia-induced release of sFlt-1/sEng from hypoxic pla-centas.Data indicated that nitroglycerin inhibits the expression andrelease of sFlt-1/sEng from placental explants exposed to hypoxia.The nitroglycerin-mediated inhibition of expression and releaseof sFlt-1 and sEng was due to diminished HIF-1a accumulation.This study provides evidence that hypoxia, through HIF-1a, inducesplacental release of both sFlt-1 and sEng, and this effect is inhibitedby nitroglycerin. These findings indicate that NO mimetics mayhave potential applications in the treatment and/or prevention ofpre-eclampsia.

2.3. Conclusions

Studies of cellular and molecular mechanisms regulatingtrophoblast invasion continue to provide new insights into thephysiology of utero-placental homeostasis and its breakdownundera wide range of pathological conditions that compromise fetal-maternal health. While diseases associated with poor trophoblastinvasion (pre-eclampsia and fetal growth restriction) were high-lighted in this workshop, trophoblast hyper-invasive disorders suchas placenta accreta, gestational trophoblastic neoplasias, andchoriocarcinomas deserve equal attention. For example, choriocar-cinomas exhibit refractoriness to the anti-invasive action ofdecidua-derived TGF-b as well as decorin. While their TGF-b resis-tance was shown to be, at least in part, due to loss of smad-3expression, the reasons for the resistance to decorin remain to beexplored. Further studies intowhy trophoblast hypo-invasiveness isshared bymany, but not all, cases of fetal growth restriction andpre-eclampsia should be valuable. Consensus is needed regarding thechoice of early serummarkers for longitudinal screeningof pregnantmothers that can predict disease and also help in devising strategiesfor prevention and intervention. In this regard, therapeutic use ofcertain animal models of pre-eclampsia may be helpful for possibleapplication in the human.

It is also important to emphasize that the fine balance betweenpro-invasive and anti-invasive factors at the fetal-maternalinterface can be disrupted by alterations in the local microenvi-ronment. Among such alterations is scarring of the uterine wall(e.g. due to repeated caesarean deliveries) leading to a hyper-invasive trophoblast and placenta accreta/percreta, as discussed inthe pathology workshop. Furthermore, the presence of high levelsof pro-inflammatory molecules resulting from aberrant maternalimmune activation may result in shallow invasion and theimpaired utero-placental perfusion associated with variouscomplications of pregnancy.

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99S94

3. Fetal sex and intrauterine stress: boys livedangerously in the womb

Chairs: Vicki Clifton, Rohan LewisSpeakers: Luciana Lassance, Rohan Lewis, Richard Saffery, Colin

Sibley, Owen Vaughan

3.1. Outline

Sex differences in the placenta have important implications forthe developmental origins of adult disease (DOHaD). Determiningthe relationship between sex and placental function is crucial forunderstanding how poor intrauterine environment contributes tothe risk of chronic disease in later life. This workshop exploredrecent work on placental sex differences and discussed them in thecontext of the boys living dangerously in the womb hypothesis. Theaim was to better our understanding of how males and femalesrespond to suboptimal intrauterine conditions and how this mayaffect their health throughout life.

3.2. Summary

Luciana Lassance discussed how fetal sex determines insulinand IGF2 effects on the feto-placental endothelium. Insulin andinsulin-like growth factor (IGF) 2 are central growth factors for thedevelopment of the placenta and the fetus. Their receptors areexpressed on the placental endothelium but their cellular effectsremained elusive. They can activate different signaling pathwaysand regulate different cellular effects in arterial and venousplacental endothelial cells. These effects may also depend on fetalsex and on oxygen levels. Therefore, insulin and IGF2 effects onarterial and venous endothelial cells isolated from placentasderived from male and female fetuses were investigated bymicroarrays. The role of insulin and IGF2 in cell signaling wasanalyzed in arterial endothelial cells from both sexes under 12%and 21% oxygen. As 21% oxygen represents a hyperoxic situation,these experiments were performed in the presence or absence ofthe anti-oxidant NAC. Microarray analyses of the effects of insulinand IGF2 revealed differential gene expression in a sex dependentway even in the absence of the growth factors. Insulin and IGF2-stimulated signaling showed more Akt and GSK3a/b phosphory-lation in arterial endothelial cells isolated from male placentasthan in those isolated from female placentas. This activation wasoxygen dependent occurring only at 21% oxygen and in the pres-ence of NAC but not at 12% oxygen. These data suggest that fetalsex can modulate insulin and IGF2 effects on feto-placentalendothelial cells, with the male endothelium being more suscep-tible to these effects.

Rohan Lewis discussed sex differences in placental geneexpression. Sex differences in placental gene expression mayunderlie differences in the responses of male and female placentasto environmental stimuli. Analysis of genes involved in amino acidtransport andmetabolism identified sex differences both in averageplacental gene expression between male and female and in therelationships between placental gene expression and maternalfactors. Sex differences were observed in the average level ofexpression of 8 genes including those commonly used as house-keeping genes. Sex differences were primarily observed in themoststably expressed genes, suggesting that this may be a more generalphenomenon but that inmany genes it is obscured by other sourcesof variation. Analysis of target genes suggests that there are sexdependent relationships between maternal factors, includingmaternal smoking, diet and parity, andmRNA levels in the placenta.These genes are therefore good candidates for further investigationof sex differences in the placenta.

Richard Saffery discussed sex-specific differences in theplacental epigenome of newborn twins. Epigenetic modifications,such as DNA methylation, play a major role in controlling geneexpression. Mounting evidence suggests a unique epigenetic profilein human placenta and purified cell subtypes, with evidence forconsiderable inter-individual variation. However, mechanismsunderlying this variation remain unclear. It has been speculatedthat this results from a combination of genetic and environmentalfactors, which combine during pregnancy to modulate the levelof methylation present at specific gene promoters. As part of thePeri/Postnatal Epigenetics Twins (PETS) study, placental tissue fromover 150 twin pregnancies has been collected. In the current study,DNA methylation profiling of placental tissue from sixteen twinpairs was analyzed at over 27,000 genomic CpG sites, associatedwith 14,500 genes. Cluster analysis revealed that fetal sex wasthe biggest determinant of overall placental methylation profile.The majority of this effect was due to X-chromosome-associatedchanges resulting from X-inactivation in females. However, signif-icant differences were also found on autosomal chromosomes(w900 probes), highlighting the role of sex-specific epigeneticchange in placental genome regulation.

Owen Vaughan discussed sex differences in efficiency andamino acid clearance in the late gestation mouse placenta. It is welldocumented that the extent to which the intrauterine environmentprograms adult phenotype is dependent upon fetal sex. Forinstance, male offspring show increased incidence and severity ofhypertension in human and animal models of fetal growthrestriction. The supply of nutrients across the placenta is a majorfetal growth signal and programming stimulus but differences inplacental transport characteristics between the sexes have not beeninvestigated. Conceptus growth and materno-fetal amino acidclearance in the mouse in late gestation was assessed. In normalpregnancies, the female placenta is more efficient, as determinedby the number of grams of fetus produced per gram of placenta, andexhibits higher amino acid clearance per gram than that of themale; these functional differences may be related to differences inplacental architecture and expression of nutrient transporters. Sexspecificities in total placental nutrient supply and the combinationof substrates provided to the fetus may render the male placentaless able to adapt to altered maternal environment and thus thefetus is more susceptible to adverse outcomes.

Colin Sibley discussed how corticosteroids stimulate the activityof the Naþ/Hþ exchanger (NHE) in the syncytiotrophoblast inplacentas from female babies. NHE regulates intracellular pH (pHi)in the syncytiotrophoblast; its activity in this epithelium is reducedin fetal growth restriction. During a study testing the hypothesisthat corticosteroids acutely stimulate NHE in the syncytiotropho-blast, it was shown that the observed effects were dependent onthe sex of the fetus. Villous fragments from term placentas wereloaded with a pH sensitive fluorescent dye and the Naþ-dependentrecovery of pHi from an acid load was taken as a measure of NHEactivity. In placental villi from female babies aldosterone signifi-cantly increased NHE activity but there was no effect in villi fromplacentas of male babies. Simultaneous application of cortisol withcarbenoxolone (11-b-HSD-2 blocker) onto the villi again signifi-cantly increased NHE activity only in placentas from female babies.It is speculated that these sex dependant effects of corticosteroidson NHE activity may be related to differences in placental 11-b-HSD-2 activity.

3.3. Conclusions

This workshop was an interesting exploration of a new body ofevidence identifying that the placenta does function in a sex-specific manner in both animal models and humans. Data

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99 S95

presented at this workshop demonstrated that sex differences areobserved in the placenta at multiple levels, from epigeneticmodification of DNA, through gene expression to functionalchanges in the way in which the placenta responds to hormonesand transports nutrients. It is clear that these changes becomeestablished in the placenta via mechanisms such as DNA methyl-ation, and are not simply due to direct effects of fetal sexhormones. Future work is required to establish the basis of sexdifferences and where there are common underlying mechanisms.The discussion raised a number of interesting issues including thebasis of sex differences from an evolutionary perspective and howdifferent evolutionary pressures may shape the sex differences inspecific species.

4. Parasites, infections and the placenta

Chairs: Henning Schneider, Ulrike KemmerlingSpeakers: Ashley Davey, Ricardo Fretes, Thierry Fournier, Stefan

R. Hansson, Ulrike Kemmerling, CindyMorris, Demba Sarr, HenningSchneider

4.1. Outline

This workshop addressed various inflammatory responses asa sign of interaction between placental tissue and parasites likePlasmodium falciparum, Trypanosoma cruzi as well as cytomegalo-virus. Changes in morphology and function of the placenta asa result of infection were discussed. The significance of differenttissue reactions were also discussed in the context of defensive ortoxic effects. Congenital transmission of infectious agents can leadto miscarriages and low birth weight newborns. Data on molecularmechanisms involved in the transmission of the organisms fromthe maternal into the fetal compartment were presented.

4.2. Summary

Ulrike Kemmerling discussed the possible mechanisms ofT. cruzi infection in congenital Chagas’ disease. Chagas’ disease, oneof the major public health concerns in Latin America, is caused bythe hemoflagellate protozoanT. cruzi. In vector-related diseases, it issecond to malaria in prevalence andmortality. In the past few yearscongenital transmission of T. cruzi has become more important,and partly responsible for the “globalization of Chagas’ disease”,constituting a public health problem of increasing relevance.Diverse pathogens, including T. cruzi, are able to cross the placentalbarrier and infect both the placenta and fetus. However, knowledgeabout the mechanism of tissue invasion and infection of humanplacenta by this parasite is scarce. T. cruzi induces syncytio-trophoblast destruction and detachment, selective disorganizationof the basal lamina and of collagen I in the connective tissue ofvillous stroma and apoptosis in the chorionic villi. These resultssuggest that the penetration of this parasite in the placenta isa consequence of its proteolytic activity on the basal lamina and onthe connective tissue. Together with the induction of apoptosisthese may be part of the mechanisms of infection and tissueinvasion by this parasite.

Ricardo Fretes discussed analysis of the human placental cho-rionic villi and trophoblast infection by T. cruzi in vitro. Chorionicvilli of human placenta may actively participate in the control ofcongenital Chagas transmission. In accordance with the lowtransmission rate of T. cruzi to the fetus, there is a low productiveinfection of T. cruzi in placental tissue with no viable parasites inculture media. It was questioned whether syncytiotrophoblast andcytotrophoblast have the same susceptibility to infection by T. cruzi.Isolated cytotrophoblast and trypsin-denuded chorionic villi were

very susceptible to infection, contrary to isolated syncytiotropho-blast and the complete chorionic villus barrier.In addition, T. cruziinduced detachment of the syncytiotrophoblast associated withnitrative-oxidative stress and increased expression of eNOS mRNAand protein. These data support a resistance of the completeplacental barrier to infection by T. cruzi and a possible mechanismthat this parasite might employ to infect the chorionic villi.

Ashley Davey discussed how differences in binding of humancytomegalovirus to cultured syncytiotrophoblast and Caco2epithelial cells are due to differential expression of maternalhuman cytomegalovirus (HCMV) receptors. Active HCMV infec-tions increase the risk of fetal transmission; however, this is notcorrelated with viral load in maternal blood or urine.It has beenproposed that infectious HCMV persists and is protected inplacental syncytiotrophoblast through reversible binding. This willmaintain the maternal infection leading to negative vasculareffects as has previously been shown and increase the risk of fetaltransmission. Inoculum and not progeny HCMV reversibly binds toheparan sulfate proteoglycans on cultured syncytiotrophoblastand is protected from degradation for >48 h. Interestingly, Caco2,microvilliated epithelial cells, show equivalent reversible bindingbut a threefold higher infection efficiency compared to syncytio-trophoblast. These differences could be attributed to differentialexpression of HCMV receptors. Thus, even though the placentaacts like a viral reservoir, with infectious HCMV persisting in syn-cytiotrophoblast, a strong protective barrier to syncytial infectionis maintained.

Thierry Fournier discussed how activation of PPARg by humancytomegalovirus for de novo replication impairs trophoblast inva-sion. The HCMV, a virus of the beta herpes family, which generallyhas no effect on an individual’s health, may cause serious disordersin immunodeficient individuals and lead to miscarriage and fetalinjury in pregnant women. Indeed, an initial infection, or reinfec-tion of a mother during pregnancy may be responsible formiscarriage, low birth weight, mental retardation or serioussensory problems in newborn babies. It is known that infection ofthe fetus is always preceded by infection of the placenta. It has beenpreviously shown that activation of the ligand-activated nuclearreceptor PPARg inhibits trophoblast invasion. Interestingly, wheninfecting trophoblast cells, HCMV activates and uses PPARg for itsown replication, thus upsetting some of the cellular mechanismssuch as invasion and migration involved in the physiologicalprocess of anchoring the placenta to the uterine wall. This can leadto abnormal development of the placenta resulting, in turn, ininsufficient nutrition of the fetus and, as a consequence, theappearance of growth and neurological developmental disorders,independent of infection of the fetus itself by HCMV.

Cindy Morris discussed differential permissiveness of extra-villous trophoblast-like cell lines to human cytomegalovirusinfection. HCMV infection at the fetal-maternal interface may resultin poor pregnancy outcomes for both mother and child. The exactmechanism(s) required for HCMV binding and entry into EVT isunclear. To determine how HCMV binds and enters EVT, studieswere performed to compare infection kinetics between 2 EVT-likecell lines, SGHPL-4 and SGHPL-5. Unlike SGHPL-4 cells that arehighly permissive to HCMV infection, the SGHPL-5 cells hadminimal viral protein expression at high multiplicity of infection.Flow cytometric and Western blot analyses demonstrated similarsurface receptor expression patterns on both cell lines at differentlevels; therefore the published fibroblast receptors for HCMV arelikely not the reason for the lack of permissiveness of the SGHPL-5cells. Experiments are ongoing to determine whether the HCMVgH/gL/UL128-131 pentameric complex plays a role in HCMVbinding and internalization in cytotrophoblasts, as seen in endo-thelial and epithelial cells.

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99S96

Demba Sarr discussed how the malaria parasite toxin, hemo-zoin, activates MAP kinases and promotes a chemotactic andimmunostimulatory secretory response in primary human syncy-tiotrophoblast. Malaria during pregnancy is characterized bysequestration of parasitized erythrocytes and accumulation of theparasite hemoglobin metabolite, hemozoin, in the intervillousspace. Primary syncytiotrophoblast cells respond immunologicallyto cytoadherent P. falciparum but their responsiveness to hemozoinhas not been reported. In vitro exposure of syncytiotrophoblast tohemozoin induced ERK1/2 and JNK mitogen-activated proteinkinase phosphorylation, secretion of chemokines, and release ofcell surface receptors. The stimulated cells also elicited the specificmigration of peripheral blood mononuclear cells. Finally, condi-tioned medium from hemozoin-stimulated syncytiotrophoblastinduced upregulation of intercellular adhesion molecule-1 onprimarymonocytes. The dependence of the hemozoin responses onmitogen-activated protein kinase stimulation was confirmed byinhibition of chemokine release in syncytiotrophoblast pre-treatedwith MEK1/2 inhibitor. These findings expand the cell types knownto be responsive to native hemozoin to include fetal syncytio-trophoblast and provide further evidence that syncytiotrophoblastcells can influence the local maternal immune response toplacental malaria.

Henning Schneider discussed adherence of P. falciparum infec-ted red cells to the trophoblast and placental inflammatoryresponse studied by dual ex vivo perfusion of an isolated cotyledon.The pathogenesis of placental malaria is associated with accumu-lation of malaria-infected erythrocytes in the intervillous space.The P. falciparum erythrocyte membrane protein-1 (PfEMP-1)mediates the adherence of red cells infected with the Var2CSAstrain of the parasite via binding to chondroitin A sulfate in themicrovillous membrane of the syncytiotrophoblast. After closedloop dual ex vivo perfusion of an isolated cotyledon with malaria-infected erythrocytes in the maternal circuit, only the schizontstage of the Var2CSA showed a marked decline indicating seques-tration in the placenta. The ring and early trophozoite stages, whichdo not express PfEMP-1, and all stages of the non-adherent 3D7 didnot drop. As an inflammatory response, the trophoblast in perfu-sions with malaria-infected erythrocytes compared to perfusionswith plain medium or with uninfected erythrocytes releases themacrophage chemokines MIF and MIP-1a into the maternalperfusate at a significantly increased rate. Microarray-based geneexpression shows substantial upregulation of several pro-inflam-matory cytokines (IL-1b, TNFa, IL-6) and the chemokines IL-8, MIP-1a, CCL3. Only the release of MIF and MIP-1a and the upregulationof the transcription factor c-fos seem to be malaria induced,whereas the changes in expression of other chemokines and pro-inflammatory cytokines is an unspecific response to stress relatedto ex vivo perfusion.

Stefan Hansson discussed links between pregnancy-associatedmalaria and pre-eclampsia. Pregnancy-associated malaria resultsin sequestration of parasites in the placenta, causing placentalinflammation and impaired placental function. Placental malariahas been associated with an increased risk of pre-eclampsia, butthe mechanisms linking these conditions are not known.Pre-eclampsia is associated with placental over-expression ofhemoglobin and increased plasma concentrations of free fetalhemoglobin and the heme scavenger, alpha-1-microglobulin(A1M). In a longitudinal prospective study in Korogwe, northeastTanzania, 1000 pregnant women were followed throughout theirpregnancy with clinical and parasitological examination andcollection of blood samples at 4 antenatal visits, emergency visits,and at delivery. Plasma fromwomen with positive rapid diagnostictest and women developing pre-eclampsia, or eclampsia wereanalyzed for free HbF/A1M.

4.3. Conclusions

The present understanding of infection of the placenta by T. cruzi(Chagas’ disease), human cytomegalovirus (HMCV) and P. falcipa-rum (Malaria) was reviewed. The impressive diversity of mecha-nisms of infection and penetration of the placental barrier on thepart of the infectious organism and of the defensive responses onthe part of the placenta became readily apparent.

5. Decidua, embryonic and fetal development and loss

Chairs:Elisa Cebral, Alicia JawerbaumSpeakers: Elisa Cebral, Vicki Clifton, Ana Franchi, Cristina Ibarra,

Michael Soares, Stacy Zamudio

5.1. Outline

Embryos and fetuses are highly susceptible to adverse changesin the maternal environment, and the differences between theircapacity to adapt to those changes and their damage and loss maybe related to the adverse effects exerted in the decidua and theplacenta. Throughout intrauterine development both the cells andthe extracellular matrix components interact with complexmechanisms to determine the proper development that will lead tofetal survival at term. Underlying the pathologies of pregnancy,dysregulation of these mechanisms may lead to embryo or fetalloss. In this workshop, the aimwas to discuss the latest informationon key regulatory mediators in the decidua, the embryo and thefetuses and to explore their alterations in different diseases andadverse conditions. A main focus was to distinguish between thoseadverse situations that lead to embryonic/fetal survival andthose that threaten embryonic/fetal life. An important questionaddressed was if there is a limit in the quantity or quality of damagethat induces adaptive responses compatible with embryonic/fetaldevelopment when compared to those that cause severe develop-mental damage or lead to embryonic/fetal loss.

5.2. Summary

Michael Soares discussed decidual development in the sub-fertile brown Norway rat. The brown Norway inbred rat strainexhibits a subfertility phenotype and possesses a growth-restrictedplacentation site. Subfertility in the brown Norway rat is associatedwith ovarian and uterine dysfunction. Brown Norway rats ovulatefewer eggs and their ovaries show disruptions in organization andin ability to produce steroid hormones. These ovarian anomaliescontribute to uterine dysfunction. The brown Norway rat uterus isresistant to progesterone, which leads to an attenuated decidualreaction and a less than optimal maternal milieu for placental/fetaldevelopment. Collectively, these maternal factors and potentialintrinsic differences in brown Norway rat trophoblast developmentcontribute to growth-restricted placentas. There are a wealth oftools developed for the brown Norway rat that can facilitate geneticdissection of pathophysiology, which makes the brown Norway ratan ideal model organism for investigating the genetics of preg-nancy failure.

Elisa Cebral discussed early embryo loss, growth restrictionand abnormal regulatory mechanisms after periconceptionalcohol ingestion. It was addressed whether moderate alcoholconsumption prior to gestation and up to mid-gestation in miceis able to cause early retarded growth-differentiation, altermorphogenesis and cause embryo loss. Trophoblast cells, innercell mass in blastocysts and cylinder eggs, and the mesenchymaland neuroepithelial cells in organogenic embryos are potentialtargets for alcohol-induced early embryo loss.It was discussed

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99 S97

how the embryonic cells, molecules and developmental processessuch as proliferation and growth, migration, invasion, timing ofdifferentiation andmorphogenesis can be altered by direct alcoholaction and indirectly through defective placentation. Deregulateddifferentiation, defective trophectoderm expansion/migration,delayed trophoblast growth, implanting embryo loss, embryonuclear alterations, defective neural tube, disorganized neu-repithelium, microcephaly, disruption of metabolic, cellular andmolecular pathways in organogenic embryos, were some of themain effects found after alcohol exposure. Periconception ethanolintake exerts a strongly negative influence on the early embryoleading to fetal growth restriction and skeletal malformation inmouse fetuses at term.

Stacy Zamudio discussed links between hypoxia and fetal loss.Among humans 78/100 of all conceptions fail to result in a livebirth.The rates at which embryonic/fetal loss occurs at various timepoints in pregnancy and their known causes were examined.Inaddition, the time course for oxygen changes in placenta and fetusand whether abnormalities of oxygen tension or transport are orare not likely to be causally involved in miscarriage/fetal loss/stillbirth was also assessed. This analysis allows identification ofkey time points in pregnancy when oxygen-related issues shouldbe discounted or considered as a consequence of other pathologicalprocesses rather than a proximate cause of pregnancy failure. Thesekey time points include the first trimester of pregnancy whenprofound fluctuations in placental oxygen tension occur, the latesecond trimester inwhich other pathological processes may impairplacental oxygen diffusion, and late in gestationwhen the fetus, fora variety of reasons, can outgrow its supply line.

Ana Franchi discussed the role of the endocannabinoid systemand early pregnancy loss. Endocannabinoids are unsaturated fattyacid derivatives which act as endogenous ligands for cannabinoidreceptors and mimic the effects of natural existing cannabinoids. Ithas been clearly demonstrated that endocannabinoid signalinginfluences female reproduction during early pregnancy. Ananda-mide, the major endocannabinoid studied so far, has a role inimplantation and embryo development. However, high levels ofthis molecule correlate with fetal weight loss and abortion.Theparticipation of anandamide in a murine model of early pregnancyloss in order to develop useful tools to prevent miscarriage wasassessed.It was observed that anandamide is involved in thesynthesis of the pro-inflammatory molecules such as nitric oxideand prostaglandins that contribute to embryonic resorption. It wasalso observed that the metabolism of anandamide and nitric oxidesynthesis are regulated by progesterone.

Vicki Clifton discussed maternal asthma and fetal development.It has been questioned how a fetus can respond to an adversematernal environment in order to survive and the impact ofmaternal asthma during pregnancy has been investigated toaddress this question. Asthma is a highly common complication ofhuman pregnancy and exerts significant effects on fetal develop-ment. Uncontrolled asthma, regardless of its severity, exerts sex-specific effects on fetal growth. The female fetus reduces growth atthe end of gestation to cope with the presence of the asthma whilethe male fetus continues to grow normally. If asthmatic mothers’had an acute, asthma exacerbation during pregnancy, the malefetus was at risk of a low birth weight outcome, preterm delivery orstillbirth. The female fetus was unaffected by an asthma exacer-bation and more likely to deliver at term. These findings raise thequestion of whether fetal sex should be considered when assessingfetal growth and survival in an adverse environment.

Cristina Ibarra discussed the role of nitric oxide in shiga toxin 2-induced premature delivery of dead fetuses in rats. Shiga toxin-producing Escherichia coli infections are a possible cause of fetalmorbidity and mortality. It has been previously reported that

intraperitoneal injection of rats in the late stage of pregnancy witha combination of shiga toxin 2 and LPS induces premature deliveryof dead fetuses. Now it has been demonstrated that there is anincrease in nitric oxide production and high nitric oxide synthase(iNOS) expression in placental tissues from rats injected with shigatoxin 2/LPS and killed 12 h after treatment. Aminoguanidine, aninhibitor of iNOS, caused a significant reduction of shiga toxin 2/LPSeffects on the feto-maternal unit but did not prevent the prematuredelivery. Placental tissues from rats treated with aminoguanidine/shiga toxin 2/LPS presented normal histology that was indistin-guishable from controls. These results reveal that placental damageand fetus mortality is mediated, at least in part, by an increase innitric oxide production by iNOS.

5.3. Conclusions

A complex interaction of molecular signals guides embryonic/fetal development. In an adverse environment, embryonic/fetalsusceptibility is highly dependent on the developmental stage aswell as on other factors such as fetal sex. Furthermore, it is clearlyrelevant that embryonic/fetal developmental damage is in partinduced by the adverse effects exerted by different pathologies onthe decidua and the placenta. Damaging mechanisms havecommon features in different pathologies and can result inembryonic and fetal adaptations or loss. The similar insults that canlead to different effects on embryonic/fetal survival or loss illustratethe narrow gap delimiting intrauterine life and death, a key pointthat deserves future research.

6. Trophoblast differentiation and syncytialization

Chairs: Yoshiki Kudo, Neal RoteSpeakers: Irving Aye, Jesica Flores-Martin, Takahiro Nobuzane,

Neal Rote, Ambika Singh, Tharini Sivasubramaniyam

6.1. Outline

Successful humanplacentation is dependent upon the transitionfrom mononuclear villous cytotrophoblast into multinuclear syn-cytiotrophoblast. Major differentiation events include, but are notlimited to, cellular fusion, production and secretion of hCG, devel-opment of relative resistance to apoptosis, and loss of proliferativecapacity. This workshop addressed advances in the study of allcomponents of the villous cytotrophoblast differentiation process,their interrelationships, and how variations may lead to placentaldysfunction and pregnancy complications.

6.2. Summary

Nobuzane Takahiro discussed whether syncytin and its receptorASCT2 have an influence on the fusion activity in human placentalBeWo cells. Syncytin, the human endogenous retrovirus envelopeprotein, is highly expressed in the placental syncytiotrophoblastlayer and contributes to placental trophoblast fusion. ASCT2 isa syncytin receptor and known to be identified as the amino acidtransporter B0. To investigate the role of syncytin and ASCT2 onfusion activity RNA interference techniques were used in humanplacental BeWo cells. BeWo cells were treated with siRNA forsyncytin-1 and ASCT2, and then incubated with forskolin to inducefusion. The effect of those specific siRNA treatments on theexpression of mRNA and protein was evaluated by means of real-time RT-PCR and Western blotting, respectively. Cell fusion activitywas evaluated by flow-cytometry and hCG secretion. The expres-sion of both mRNA and protein of syncytin-1 and ASCT2 wasreduced after incubation of specific siRNA treatment. The secretion

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99S98

of hCG was also reduced after treatment with siRNA for syncytinand ASCT2, but fusion activity was only reduced after treatmentwith siRNA for ASCT2. This result suggests that syncytin hasa different effect on fusion activity and hCG secretion comparedwith ASCT2 in BeWo cells.

Tharini Sivasubramaniyam discussed a novel role for Parti-tioning defective protein 6 (Par6) in trophoblast cell fusion. Humanplacental development is dependent upon the establishment ofproper trophoblast cell differentiation events including fusion andinvasion, shaping proper organogenesis. This study examined thecontribution of polarity to trophoblast cell fusion by examiningPar6, a key regulator of cell polarity. Human placental tissuesacross gestation, pre-eclamptic placentae and age-matchedcontrols were collected. To establish a role for Par6 in trophoblastfusion, primary isolated trophoblast cells were cultured at 3% and20% oxygen and Par6 expression examined spatially and tempo-rally in conjunction with a polarity marker, Zona Occludin-1(ZO-1). Par6 expressionwas assessed following forskolin treatmentin BeWo cells. Par6 siRNA strategy was employed in BeWo cellsand ZO-1 expression examined. Additionally, the role of Par6 ontrophoblast fusion was assessed by examining the fusogenicmarker GCM-1 following Par6 silencing. Early in gestation, Par6localized mainly to the nuclei of cytotrophoblast cells while, withadvancing gestation, Par6 expression increased and its localizationshifted to the cytoplasm where it was found at the interfacebetween cytotrophoblast cells and the syncytium. In primarytrophoblast cells, Par6 levels were increased after 48 h of exposureto 3% oxygen and localized to tight junctions while at 20% oxygen,Par6 expression remained cytoplasmic. Silencing Par6 in BeWocells resulted in a disruption of ZO-1 localization and this wasassociated with an increase in GCM-1 expression. Following for-skolin treatment, Par6 expression decreased in association with anincrease in syncytin expression. Interestingly, Par6 expression wasdecreased in pre-eclampsia relative to age-matched controls. Thesefindings provide novel insights into a role for Par6 in regulatingtrophoblast cell fusion via its effect on polarity during humanplacental development, which may be disrupted and contribute tothe pathogenesis of pre-eclampsia.

Ambika Singh discussed how ceramide, acid sphingomyelinaseand ceramidase exert differential effects on trophoblast differen-tiation and fusion. Although trophoblast syncytialisation has beenactively investigated for decades, various contradicting theoriesdescribing the fundamental nature of trophoblast differentiationand fusion exist.A novel role of ceramide, and sphingolipidsynthesis/metabolic enzymes in trophoblast differentiation hasrecently been described. The distinct roles of ceramide in differ-entiation and fusion, and the signaling pathways involved in thisprocess were investigated. Cytotrophoblast cells, isolated fromterm human placentas were allowed to differentiate over 7 days inculture, as assessed by measuring hCG secretion, placental alkalinephosphatase (PLAP) activity and E-cadherin expression andimmunostaining. Exposure of trophoblast cells for 72 h to shortchain ceramide, acid sphingomyelinase and a ceramidase inhibitor(B13) enhanced hCG secretion, indicating a role for ceramide indifferentiation but not fusion. Syncytialised trophoblast expressedhigh levels of ceramidase, suggesting that this enzyme may beinvolved in maintaining the syncytial phenotype. Consistent withthis view, B13 treatment reduced fusion as shown by enhancedE-cadherin expression. Inhibition of the ceramide-responsivepathways JNK-II and PP2A did not abolish the effects of ceramide,and in fact increased hCG productionwhile having divergent effectson fusion (PLAP and E-cadherin). This study highlights thecomplexity of the pathways that regulate trophoblast differentia-tion and fusion (syncytialisation). Different agents (e.g. ceramide)can regulate differentiation independently of effects on fusion,

while regulators of fusion can have no or even opposite effects ondifferentiation. Pregnancy complications such as pre-eclampsia,where a moderate increase in syncytialisation is observed with noeffect on differentiation, may be the end result of perturbations inone or more of these pathways.

Irving Aye discussed how oxysterols inhibit trophoblast syncy-tialisation by activating the liver X receptor. Oxygenated cholesterolmetabolites known as oxysterols display potent biological activitiesranging from regulation of lipid homeostasis to cytotoxic and pro-apoptotic effects, and have been linked to lipid disorders such asatherosclerosis. Oxysterols have previously been shown to inhibitinvasion of first trimester trophoblast cells, an effect which involvesactivation of the nuclear liver X receptor, suggesting a link betweenoxysterols and pre-eclampsia. In this study, the effect of severaloxysterols on trophoblast syncytialisation (differentiation andfusion) in term placental trophoblast cells was investigated.Primary trophoblast cells were isolated from term placentas andallowed to syncytialise in vitro. Trophoblast cells were treated withvarious oxysterols [25-hydroxycholesterol (25-OHC), 7-ketocho-lesterol (7-ketoC), 22(R)-hydroxycholesterol (22R-OHC)] and thesynthetic liver X receptor agonist (T0901317) at non-toxic doses,with or without pre-treatment with liver X receptor antagonistgeranylgeranyl diphosphate (GGPP). Trophoblast differentiationwas monitored by measuring GCM-1 mRNA expression, hCGsecretion and placental alkaline phosphatase activity, while cellfusion was determined by E-cadherin immunostaining and quan-tification of syncytialised nuclei. Trophoblast differentiation andfusion were both significantly inhibited by all oxysterols used.GCM-1 expressionwas reduced between 30 and 50%, hCG secretionbetween 20 and 35%, alkaline phosphatase activity between 15 and25% and cell fusion between 30 and 50%. The synthetic liver Xreceptor agonist T0901317 also potently inhibited trophoblastdifferentiation and fusion. Moreover, treatment with the liver Xreceptor antagonist GGPP abrogated the inhibitory effects of oxy-sterols and T0901317 on trophoblast syncytialisation. These find-ings suggest that oxysterols impair differentiation and fusion ofterm trophoblast cells via a liver X receptor dependent mechanism.Excessive oxysterol exposure may impair placental formation andregeneration via inhibition of syncytialisation, thereby contributingto placental pathologies.

Jésica Flores-Martin discussed how down-regulation of StarD7by RNA interference increases b-hCG production and secretion inJEG-3 cells. StarD7 is a member of the START-domain protein familywhose function remains poorly defined. It has recently beenreported that StarD7 is partially relocated from the cytoplasm tothe plasma membrane during in vitro cytotrophoblast differentia-tion into syncytiotrophoblast. Furthermore, it has been shown thatb-catenin activates human StarD7 expression through Wnt/b-cat-enin signaling.To explore its function in JEG-3 cells StarD7expression was down-regulated by siRNA. Silencing of StarD7expression led to a marked decrease of b-catenin, Cnx43, iNOS,MBD2, ABCG2 and TGFbRII mRNA levels, all of them associatedwithWnt signaling. In contrast, expression of syncytial formationmarkers, such as b-hCG protein production and secretion, as well asb-hCG mRNA levels, was increased by StarD7 siRNA. In addition,immunostaining for desmoplakin suggested that there wasa reduction of intercellular desmosomes between adjacent JEG-3cells after knockdown of StarD7 expression. These findings indicatethat the inhibition of StarD7 transcript level alters the expression ofseveral critical genes suggesting that it may play a role in placentaldevelopment.

Neal Rote discussed how silencing of caspase 8 in BeWosuppresses b-hCG induction. Development of the human syncy-tiotrophoblast from mononuclear villous cytotrophoblast results inincreased resistance to apoptosis. It has previously been reported

A. Al-Khan et al. / Placenta 32, Supplement B, Trophoblast Research, Vol. 25 (2011) S90eS99 S99

that forskolin-induced differentiation and syncytialization of theBeWo cell line resulted in increased expression of both protein andmRNA for the anti-apoptotic protein Bcl-2 and increased resistanceto apoptosis. However, in cells treated with solvent alone approx-imately 10% underwent spontaneous intercellular fusion, and inforskolin-treated cells approximately 20% did not fuse. It wasinvestigated whether increased expression of Bcl-2 was a result offorskolin-induced differentiation or the result of syncytialization.BeWo treated with DMSO or with forskolin for 72 h were evaluatedby immunofluorescence for E-cadherin (to identify fused cells) andBcl-2.The immunostaining intensity for Bcl-2 was similar inmononuclear cells from cultures treated with DMSO or forskolin.Compared with mononuclear cells, staining for Bcl-2 was increasedby 17% in spontaneously fused cells in DMSO-treated cultures andby 30% in fused cells in forskolin-treated cultures. Forskolin treat-ment enhanced the production of Bcl-2 by 8% in fused cells. Thus,

increased production of Bcl-2 in BeWo appears to be related moreclosely to syncytialization, whether spontaneous or forskolin-induced, than to differentiation.

6.3. Conclusions

Trophoblast cell differentiation and syncytial formation are twokey processes in the development of the placenta. Several keyregulatory molecules of these processes were discussed in thisworkshop. What became clear is that these two processes aredifferentially regulated and this perhaps highlights a need toconsider these processes as being linked but quite distinct.

Conflict of interest

The authors have no conflict of interest.