It is estimated that annually approximately 2.2 billion people are exposed to the risk for Plasmodium falciparum malaria infection and between 300–600 million clinical attacks are attributable to this parasite [1]. Ninety percent of deaths occur in sub-Saharan Africa, the majority involving children less than five years of age. In addition to children, pregnant women (particularly those in their first pregnancy) are at highest risk of severe disease [2]. A hallmark of malaria during pregnancy is the sequestration of malaria-infected red blood cells (iRBCs) containing late developmental stages in the intervillous spaces (IVS) of the placenta [3-5]. This is usually accompanied by the infiltration of maternal leukocytes, especially monocytes, in the IVS [6,7] and haemozoin deposition [4,8], resulting in what is referred to as placental malaria (PM). PM poses substantial risk to the mother, the foetus, and the neonate in the form of maternal anaemia and poor foetal outcomes such as low birth weight (LBW) and prematurity ([9,10]; reviewed in [11,12]).

The sequestration of iRBCs in the placenta is thought to be mediated in large part by the cytoadherence of iRBCs to placental receptors expressed in the IVS and on the syncytiotrophoblast (ST; foetal epithelial cells that are in direct contact with maternal blood within the IVS). Currently, it is believed that the glycosaminoglycan chondroitin sulfate A (CSA) is the principal placental iRBC receptor [13-15]. Other minor receptors are proposed to exist [16-19], although the role of hyaluronic acid [20] has recently been questioned [15]. Parasite-encoded surface ligands expressed on the membrane of iRBCs are thought to facilitate this adherence. To date, the only well-studied cytoadherence parasite protein is the P. falciparum erythrocyte membrane protein-1 (PfEMP1) encoded by the highly polymorphic members of the var gene family [21,22]. The most well characterized PfEMP1 variant identified to mediate iRBC binding to the placenta is VAR2CSA [23-25]. Despite intense effort to elucidate the placental host/parasite interaction on the molecular level, the consequences of this placental iRBC sequestration on ST cell function have largely been ignored.

The immunological consequences of malaria in pregnancy have been widely investigated. A protective IgG antibody response that blocks the binding of iRBC to CSA in the placenta has been shown to develop in a sex- and gravidity-dependent manner [26,27]. In addition, several studies have demonstrated the presence of both proinflammatory and anti-inflammatory immune factors in malaria-infected placentas [28-31]. For example, increased amounts of Th1 cytokines such as TNF-α [29,31], IFN-γ [29,30] and IL-1β [31] have been demonstrated in PM-positive placental blood. Production of IL-10 by intervillous blood mononuclear cells (IVBMC) was also shown to be increased in PM [28,30] and was hypothesized to be important in the control of the negative effects of Th1 cytokines on pregnancy [28-31]. In addition, several proinflammatory chemokines have been observed in association with PM including interleukin-8 (CXCL8/IL-8) [31,32] and beta chemokines such as macrophage inflammatory protein-1 alpha (MIP-1α/CCL3), macrophage chemoattractant protein-1 (MCP-1/CCL2), I-309/CCL1 [32] and MIP-1β/CCL4. [33]. Massively elevated levels of macrophage migration inhibitory factor (MIF) were observed in the placental plasma from women with PM [34], and immunohistochemical analysis suggests that ST contributes to this increase, particularly in cases of PM (36). However, in most cases, the cell types responsible for the production of these immune factors have not been definitively identified, although cells of both foetal and maternal origin have been implicated[31,32,35]. While proinflammatory responses are important in the clearance of iRBCs and protection against PM, they have also been shown to play a major role in malaria pathophysiology and contribute to morbidity [36]. These proinflammatory immune responses have been associated with the ensuing monocytic infiltration [6,32] which has been associated with poor foetal outcomes [6,7,9]

Foetally-derived tissues have been demonstrated to be active participants in maintaining the immunological milieu at the maternal-foetal interface during pregnancy [37]. However, their role during PM remains largely unknown. Furthermore, the consequences of iRBC sequestration in the IVS on the function of the immunologically active ST are not known. This has partly been due to the lack of a good in vitro model system with which to investigate this. Previous studies utilizing such a system [18] showed that the ST is capable of responding to iRBC binding via pan-tyrosine phosphorylation of ST proteins [18]. In the present study, the impact of iRBC interaction/binding on ST cells was further investigated. Activation of immunologic signaling in ST was observed upon stimulation with P. falciparum infected-RBC selected for binding to ST. This activation was found to be associated with changes in cytokine and chemokine expression and secretion. Furthermore, this iRBC/ST interaction stimulated the chemotactic migration of peripheral blood mononuclear cells (PBMC).

Results from this study demonstrate for the first time that the ST is capable of responding immunologically to interaction with P. falciparum-infected RBCs selected for binding to ST. This was evidenced by the induction of MAPK pathways, secretion of immune factors such as MIF and MIP-1α/CCL3, and induction of migration of PBMC towards iRBC^ST-stimulated ST.

The activation of MAPK pathways via phosphorylation is initiated by a large variety of external signals and leads to a wide range of cellular responses, including growth, differentiation, inflammation, apoptosis (reviewed in [39]), and secretion of cytokines (reviewed in [40]). This study demonstrates that the interaction of iRBC^ST with ST cells leads to the activation of the JNK1 pathway. In a previous study, the binding of iRBC^ST to ST cells was shown to result in the tyrosine phosphorylation of at least two ST proteins [18], altogether suggesting that a number of components of cellular activation pathways are activated in ST in the context of PM.

Indeed, the ST is known to be immunoactive, secreting cytokines and chemokines important for the maintenance of pregnancy [37] and in response to bacterial infections [41,42] and lipopolysaccharide [43]. The interaction with iRBC^ST led to the secretion of MIF and MIP-1α/CCL3 by ST. MIP-1α/CCL3 levels were shown to be increased in placental plasma of PM-positive women [32], although not so in another study [33]. The current study suggests that the ST may contribute to elevated MIP-1α/CCL3 levels in the placenta during PM. However, dramatic variation in the magnitude of the secretion of MIP-1α/CCL3 (~300 pg/ml of MIP-1α/CCL3 from one placenta preparation to below detectable levels in others) was observed. This is not unprecedented as other studies using isolated trophoblast cells from term placentas have reported variability in receptor expression on the trophoblast cells [44] and cytokine/chemokine secretion [45] among the different placenta preparations utilized.

Previous studies demonstrated that PM resulted in increased levels of MIF in the intervillous blood [34] and that BeWo cells, a trophoblastic cell line, secreted substantial amounts of MIF upon iRBC^ST interaction [35]. Together, these studies implicate the ST as an important source of MIF. This cytokine is known to play an important biologic role during pregnancy [46]. Importantly, it is also known to be involved in macrophage activation [47] and retention, as it inhibits macrophage migration. Therefore, MIF may play a role in the retention of maternal immune cells that may be recruited to the IVS by other chemokines such as MIP-1α/CCL3 and RANTES/CCL5 (unpublished data), leading to the increased levels of monocytes in the IVS during PM. Previous studies have shown that PM is commonly associated with monocyte recruitment to the IVS [6,7] which, in turn, is associated with high levels of intervillous chemokines [32]. Results from this study suggest that the ST contributes to the migration of immune cells into the IVS. It remains to be determined, however, what cell types are specifically migrating toward iRBC-stimulated ST and whether this activity contributes to protective or pathogenic effects in vivo.

Both IL-8/CXCL8 [31,32] and TNF-α have been associated with PM in previous studies. In this study, however, ST constitutively secreted IL-8/CXCL8 in a stimulation-independent manner and did not secrete any detectable TNF-α. This implies that the increased levels of IL-8/CXCL8 and TNF-α observed in PM-positive placentas are not derived from ST.

Variability in the secretion and gene expression of the investigated chemokines and cytokines between ST preparations from individual donors was observed. This reflects the inherent variability of human-based immunologic studies. In this context, it is important to keep in mind that differences in host genetic backgrounds contribute to the heterogeneities of malaria morbidity and disease manifestations [48-52]. Indeed, some cases of PM are accompanied by a massive infiltration of maternal mononuclear cells into the IVS but not others. It is currently unknown if these different disease manifestations are the result of variations in chemokine levels in the IVS which are driven by host genetic polymorphisms.

The experimental design utilized in the current study does not allow one to categorically claim that the effects observed were all directly due to the binding of iRBC^ST. Both late stage trophozoites and schizonts (early and mature) were used to stimulate the ST and this co-culture was incubated for up to 24 hours in some experiments. Especially at late time points, some of the schizonts would have ruptured, releasing both haemozoin and glycosylphosphatidylinositol (GPI) among other host- and parasite-derived materials. Thus, it is possible that these malarial products also stimulated the ST cells. However, this system is not physiologically irrelevant. On the contrary, this complex mixture of malarial products and iRBCs is precisely what the ST cells experience in vivo, and, in fact, ongoing unpublished work suggests that haemozoin is a potent stimulator of ST immunologic responses (N. Lucchi et al., unpublished data). Nonetheless, further experiments including controls such as unselected iRBC or var2csa-null P. falciparum would reveal the extent to which the hypothesized points of iRBC^ST/ST interaction (i.e., var2csa-encoded PfEMP1 and CSA) are required for ST activation. Furthermore, determining whether or not iRBC^ST binding affects non-immunologic ST functions such as nutrient and gaseous exchange could shed some light on the mechanisms leading to LBW associated with PM.

This study provides the first evidence that the interaction of malarial parasites with ST cells in the placenta induces immunologic changes in the ST cells as evidenced by the activation of the MAPK pathways, upregulated secretion of MIF and MIP-1α/CCL3, and stimulation of PBMC chemotaxis. Thus, ST cells play an active immunological role in response to malarial parasites in the placenta and are capable of influencing and/or contributing to the local maternal immune environment.

JMM, NWL and DSP conceived of the study and the study design and participated in data interpretation. The study was conducted by NWL. JMM and NWL drafted and critically reviewed the paper. DSP provided the parasites and maintained the parasite cultures together with NWL. All the authors reviewed and approved of the final version of the paper.