ABSTRACT

Maternal BloodThe deportation of syncytiotrophoblast material into the maternal circulation has been recognized for many years.1 The syncytiotrophoblast is a unicellular sheet (syncytium) formed from fusion of monocellular cytotrophoblasts. This presents particular challenges in terms of its ex vivo isolation and culture, as mentioned in subsequent sections. Multinucleate syncytial trophoblast sprouts (approx. 80-200 µm in size) have been identified in uterine vein and

inferior vena cava blood and in lung capillaries. By virtue of its size, the majority of this material is trapped in the capillary bed of the mother’s lungs, with very little entering the peripheral circulation.2This led us to explore the possibility that smaller syncytiotrophoblast vesicles are also released, which are small enough to pass through the lung capillaries and enter the peripheral circulation. This was achieved using an in house enzyme-linked immunosorbent assay (ELISA) in which maternal plasma was ultracentrifuged to pellet the vesicles, which were then detected using a monoclonal antibody to placental alkaline phosphatase (PLAP) (NDOG2), specific for the syncytiotrophoblast.3 Increased levels of placental vesicles were found in the uterine vein blood compared to the peripheral blood, consistent with their placental origin. Levels increased with advancing pregnancy4 and labor,5 returning to zero in most cases by 48 h postdelivery. Flow cytometry has been used to confirm the particulate nature of the syncytiotrophoblast material pelleted from maternal plasma using two syncytiotrophoblast-specific monoclonal antibodies, anti-PLAP, as used in our ELISA, and ED822, which recognizes an as yet unknown antigen on the apical surface of the syncytiotrophoblast.3,6Human leukocyte antigen-G (HLA-G)-positive vesicles have also been reported to be present in the maternal circulation.7,8HLA-G is a specialized form of a class I major histocompatibility complex (MHC) antigen, which is almost uniquely expressed on invasive extravillous cytotrophoblasts of the placenta but not the syncytiotrophoblast. Orozco et al. in 2009 investigated the presence of PLAP-and HLA-G-positive vesicles in parallel.7 They found comparable levels of HLA-G-and PLAP-positive vesicles, with more of the former in the first and second trimesters but more of the latter in the third trimester, consistent with the increased shedding of syncytiotrophoblast vesicles later in pregnancy reported by others.4Pap et al. also reported the presence of HLA-G-positive vesicles in the third trimester.8 However, it is difficult to understand how so many HLA-G-positive vesicles could be present in the maternal circulation, given that the only extravillous cytotrophoblast exposed to the maternal blood is that which invades the maternal spiral arteries at the end of the first trimester. This is present in very low amounts, with a surface area in the order of 10 cm2 compared to 12 m2 for the syncytiotrophoblast at term (Professor Graham Burton, University of Cambridge, personal communication). PLAP-positive

vesicles have also been isolated directly from pregnancy plasma, taken between 26 and 28 weeks’ gestation, using agarose beads coated with an anti-PLAP antibody.9 15.2.1 Circulating Placental Vesicles in Pre-EclampsiaPre-eclampsia is a disorder of pregnancy that affects 2.5-3.0% of women. No other complication of pregnancy is both so common and so dangerous for mother and baby. For both, it is potentially lethal or detrimental to long-term health and may require preterm delivery of the baby with all its associated consequences. Its first (preclinical) stage comprises deficient remodeling of the uteroplacental circulation (8-18 weeks), dysfunctional perfusion and placental oxidative stress.10 This stimulates the release of placental factors into maternal blood that cause the second, clinical, stage (after 20 weeks). The latter results from maternal systemic vascular inflammation, which leads to the maternal syndrome of hypertension, proteinuria, edema, activation of the coagulation system, and, in worse cases, eclampsia, characterized by fits. We have shown that a maternal systemic inflammatory response is intrinsic to normal pregnancies but more severe in pre-eclampsia, including metabolic, clotting, and complement disturbances.11 Proinflammatory factors released by the oxidatively stressed syncytiotrophoblast into the maternal circulation link the two stages. While a number of soluble factors have been implicated in this process (e.g., sFlt-1 and sEndoglin, the soluble receptors for vascular endothelial growth factor [VEGF] and transforming growth factor β [TGFβ]), we believe that syncytiotrophoblast vesicles play an important role.12 Evidence for increased levels of placental vesicles in the maternal circulation in pre-eclampsia comes from a number of studies of pregnancy plasma. Using our anti-PLAP ELISA, we showed that the levels of placental vesicles were significantly increased in both uterine vein and peripheral vein blood of pre-eclamptic women at cesarean section compared to normal pregnant controls.3 We and others have shown that the increase in placental vesicle shedding was more pronounced in the more severe early-onset (before 34 weeks) compared to late-onset pre-eclampsia.13,14 Interestingly, no increase in placental vesicle levels, compared to normal pregnant controls, was seen in women with fetal growth restriction (FGR) in the absence of any other features of the maternal syndrome (i.e.,

normotensive FGR). Some types of FGR have the same placental pathology as pre-eclampsia but without the maternal features. This finding suggests that vesicle shedding is key to the development of the maternal syndrome of pre-eclampsia. Using the same ELISA, Reddy et al. showed a significant increase in placental vesicle shedding in pre-eclamptic women at the end of labor, which could account for the postpartum worsening of the disease, which is sometimes seen.5 Increased levels of placental vesicles in the plasma of preeclamptic women were also found in flow cytometry studies by Lok et al. in 20086 using the anti-syncytiotrophoblast monoclonal antibody ED822. In contrast, others who used an anti-PLAP antibody to identify syncytiotrophoblast vesicles and an anti-HLA-G antibody to identify an extravillous cytotrophoblast found no difference in the levels between women with pre-eclampsia and normal pregnant controls.7 However, these samples were taken from women with late-onset pre-eclampsia, and therefore their findings reflect those of Goswami et al., who found significant differences in placental vesicles only in early-but not late-onset disease.13 More recently a study of women with severe pre-eclampsia, using an anti-PLAP antibody, showed no significant difference in placental vesicles compared to the normal pregnant controls.15 15.2.2 Functional Characterization of Placental Vesicles

in Normal Pregnancy and Pre-EclampsiaChanges in the numbers of circulating placental vesicles throughout normal pregnancy and pre-eclampsia suggest that they may play important functional roles in both. To study this requires the isolation of large amounts of pure vesicles. Ideally these would be obtained directly from maternal blood. However, maternal plasma contains not only placental vesicles but others derived from many cell types within the vascular compartment, including erythrocytes, leukocytes, and endothelial cells, with the most prominent population (>90%) derived from platelets.6 It is therefore necessary to isolate placental vesicles from this large background contamination. While this has been reported using anti-PLAP antibody-conjugated magnetic beads,9 the yields are low, and detailed analysis would require a substantial volume of blood to be taken from the pregnant woman. However, as mentioned above, pregnancy has a distinct advantage in

terms of studying EVs in that their source, the placenta, can be easily obtained at delivery and vesicles prepared from it ex vivo. There are three main methods for preparing EVs from the placenta: mechanical separation, explant culture, and placental perfusion. 15.2.2.1 Mechanical separationIn this technique the maternal decidua overlying the chorionic villi of the placenta is scraped off and the chorionic villi dissected away from the chorionic plate. The villi are then stirred in a buffer for 1 h, during which time the vesicles are shed.16 While this method gives high yields of vesicles, the need to cut through the tissues may lead to the shedding of vesicles from placental endothelial and stromal cells, which may contaminate the preparations. 15.2.2.2 Explant culturesChorionic villi are dissected from the chorionic plate of the placenta and cultured for varying periods of time. The shed placental vesicles are harvested from the culture medium.17 While this preparation appears to be more biologically relevant than the mechanical method, there are concerns about the health of the tissue in these cultures, as the syncytiotrophoblast is initially shed and then regenerates over a period of time.18 Furthermore there are a number of nontrophoblast cell types present in the explants, which may also shed contaminating vesicles into the culture supernatant. 15.2.2.3 Placental perfusionPlacental vesicles may also be produced by dual placental perfusion.19 In this system the fetal artery and vein of a placental lobule are cannulated and perfused with a buffer to maintain the fetal circulation. Catheters are then introduced into the intervillous space on the maternal side, which is perfused with a buffer to mimic the maternal blood flow. Placental vesicles are shed into the perfusate and pelleted by ultracentrifugation for subsequent analysis.4,20-22 15.2.2.4 Trophoblast cell lines and primary trophoblast

preparationsIn addition, vesicles can be prepared from trophoblast cell lines.23,24 Of particular relevance is the choriocarcinoma cell line BeWo, as

it can be induced to fuse by treatment with forskolin or dbCAMP, thereby mimicking the formation of the syncytiotrophoblast by fusion of villous cytotrophoblast in vivo.25,26 Primary villous cytotrophoblasts isolated from placentas also syncytialize in vitro, to a far greater extent than BeWo cells, and have been shown to release syncytiotrophoblast-derived vesicles into the culture supernatant.27 It is important to note that vesicles prepared by these different methods are not comparable. They have been shown to have different functional properties.20 The reader should be aware of the type of preparation used in a study when interpreting the results. 15.3 Functional Effects of Placental Vesicles In

15.3.1 Immunosuppresive Effects of Placental VesiclesIt has long been known that maternal cell-mediated immune responses are suppressed during pregnancy as part of a wider mechanism to prevent fetal rejection. A range of immunosuppresive factors are released by the placenta into the maternal circulation, and there is growing evidence that placental vesicles are involved in this process. T-cell responses are significantly inhibited by placental vesicles in terms of phytohemagglutinin (PHA)- and mixed lymphocyte response-induced proliferation,21,28,29 Fas ligand-mediated lymphocyte apoptosis, and CD3-zeta loss9,29-31and interleukin (IL)-2-induced signal transducer and activator of transcription 3 (STAT3) phosphorylation.8 Placental vesicles have also been shown to suppress natural killer (NK) cell killing of K562 tumor target cells by blocking their activating NKG2D receptors via MHC class I chain-related (MIC A/B) proteins and UL-16 binding proteins (ULBPs) expressed on their surface.32,33 15.3.2  Proinflammatory Effects of Placental Vesicles

While the immunosuppressive effects of placental vesicles on T-cell and NK cell responses go some way to explaining how the placenta avoids immune rejection, the situation is more complicated than this. Although T-and NK cell responses are suppressed in pregnancy, maternal innate immune responses are activated to

bring about an inflammatory state.34 While at first this may appear counterintuitive, inflammatory cytokines in the right amounts are known to be beneficial to the processes of implantation and placentation and may also help the mother to fight infection as her T-and NK-cell-mediated immune responses are suppressed. However, in pre-eclampsia the mild inflammatory state of normal pregnancy decompensates, leading to a systemic inflammatory response in the mother, with associated endothelial dysfunction and activation of the clotting system. This is believed to be the cause of the maternal syndrome.11 There is growing evidence that placental vesicles play a role in the maternal inflammatory response. We and others have shown that they are rapidly taken up and internalized by monocytes in vitro4,35-37 with the subsequent release of a range of proinflammatory cytokines (including tumor necrosis factor-α [TNFα], MIP-1α, IL-1α, IL-1β, IL-6, IL-8, IL-12, IL-18, and interferon-γ [IFNγ]).22 Placental vesicles have also been shown to directly stimulate neutrophils, resulting in increased superoxide production38 and the formation of extracellular lattices (neutrophil extracellular traps [NETs]), which, ex vivo, are increased in the intervillous space of pre-eclampsia placentas.30 15.3.3 Effects of Placental Vesicles on Endothelial CellsAs well as having effects on innate immune cells, placental vesicles have also been shown to affect the function of endothelial cells, inhibiting their proliferation and growth as a monolayer in vitro20,39,40and inhibiting the relaxation of preconstricted blood vessels ex vivo.39-41 Furthermore, when human umbilical vein endothelial cells (HUVECs) are cultured with placental vesicles the culture supernatants can secondarily activate neutrophils, demonstrating the potential for a vicious cycle of inflammatory activation.42These effects may contribute to the endothelial dysfunction of pre-eclampsia in vivo, which is a characteristic of the maternal syndrome. 15.3.4 Procoagulant Effects of Placental VesiclesPregnancy is associated with a physiological increase in many procoagulant factors and inhibitors of fibrinolysis such that the mother is in a hypercoagulable state. This is an important mechanism

in normal pregnancy, which prevents postpartum hemorrhage. Placental vesicles may play a role in this process, being known to express tissue factor and phosphatidylserine, both of which initiate the coagulation cascade43,44 (see Chapter 1). 15.3.5 Differential Effects of Placental Vesicles from

Normal and Pre-Eclampsia Placentas on Immune and Endothelial Cells and the Coagulation SystemA key question is whether, in addition to the quantitative differences already described, there are qualitative differences between vesicles shed from normal and pre-eclampsia placentas that could alter their function. There is currently little information on this as the majority of studies have used vesicles from normal placentas due

to the practical difficulties of preparing vesicles from pre-eclampsia placentas. However, a recent report has shown that vesicles from pre-eclampsia placentas caused significantly higher activation of peripheral blood mononuclear cells (PBMCs) to produce a range of cytokines and chemokines, including IL-1β, compared to normal placental vesicles.45 It has also been shown that higher levels of superoxide production are induced in neutrophils by vesicles prepared from pre-eclampsia placentas compared to controls.38 This may be due to the higher content of peroxidized lipids and greater susceptibility to oxidation of vesicles prepared from pre-eclampsia placentas.46 In pre-eclampsia there is excessive activation of the coagulation system with increased platelet activation.47 Our recent work has shown that vesicles prepared from pre-eclampsia placentas express higher levels of functional tissue factor than those from normal placentas, which may account for the excessive activation of the clotting system seen in this disorder.43,44 However, no differential effects of vesicles prepared from normal and pre-eclampsia placentas on endothelial cells have been found.39 15.4 Characterization of Placental VesiclesAs discussed above, placental vesicles have been shown to have a wide range of functional activities, suggesting that they carry a variety of bioactive molecules into the maternal circulation. A

crucial step is to define which molecules are present and whether there are differences in these between vesicles from normal and pre-eclampsia placentas that could explain their role in this disorder. A summary of some candidate molecules and their potential activities is shown in Fig. 15.1. Proinflammatory

Disrupon of endothelium

Procoagulant

Immunoregulaon

Figure 15.1 The functional activities of placental vesicles and the relevant molecular cargos they carry. Proinflammatory effects of placental vesicles may be due to their expression of danger molecules, including Hsp70, HMGB1, and Syncytin 1, and their procoagulant activity may result from their expression of tissue factor and phosphatidylserine. Flt-1 (and/or sFlt-1), endoglin, integrins, and CD26 carried by placental vesicles may contribute to endothelial dysfunction. Placental vesicles have also been shown to be immunoregulatory, suppressing NK and T-cell responses in vitro, possibly due to their expression of MICA/B and UL-16, and Fas ligand, HLA-G, and the minor histocompatibility antigen DDX3Y, respectively. Adapted from Ref. 12, Copyright (2012), with permission from Elsevier.