Calcium channels activated by endothelin-1 in human trophoblast

Abstract

Ca²⁺ transfer across the syncytiotrophoblast (ST) of the human placenta is essential for normal fetal development. However, the nature of Ca²⁺ conductance in the ST and the mechanisms by which it is regulated are poorly understood. With the major signal transduction pathway of endothelin-1 (ET1) acting via phospholipase C (PLC) and Ca²⁺, we used ET1 to analyse the nature of Ca²⁺ channels on cultured trophoblastic cells by means of cytofluorimetric analysis using the ratiometric Ca²⁺ indicator Indo-1. Results indicate that ET1 (10⁻⁷ m) stimulates a biphasic (transient and sustained) increase in [Ca²⁺]_i in trophoblastic cells. This response is mediated by the endothelin receptor B (ETB) coupled to PLC, since treatment with BQ788 (10⁻⁶ m) or U73122 (2 μm) totally abolished the response. Persistence of the rapid transient rise in [Ca²⁺]_i in Ca²⁺-free extracellular medium confirms the release of Ca²⁺ from intracellular stores in response to ET1 stimulation. Furthermore, abolition of the sustained increase in [Ca²⁺]_i in Ca²⁺-free extracellular medium argues in favour of the entry of Ca²⁺ during the plateau phase. Abolition of this plateau phase by Ni²⁺ (1 mm) in the presence of extracellular Ca²⁺ confirmed the existence of an ET1-induced Ca²⁺ entry. No evidence for the presence of voltage-operated channels was demonstrated during ET1 action since nifedipine (10⁻⁶ m) did not reduce the Ca²⁺ response and depolarization with a hyper-potassium solution had no effect. Pharmacological studies using the imidazole derivatives SK&F96365 (30 μm) and LOE 908 (10 μm) partially inhibited the ET1-evoked Ca²⁺ response, thus providing evidence for the presence of both store-operated Ca²⁺ channels and non-selective cationic channels in the human ST.

In the human placenta, the chorionic villi, which are in direct contact with the maternal blood, consist of trophoblasts surrounding a core of connective tissue that includes the fetal vessels, fibroblasts and Hofbauer cells. The villous trophoblast differentiates from the fusion of cytotrophoblastic (CT) cells into a multinucleated true syncytium, the syncytiotrophoblast (ST). The ST is a transporting epithelium engaged in numerous placental functions required for fetal growth and development (different forms of exchange, metabolism and production of biologically active substances).

Ca²⁺ ion transfer across the ST is essential for normal fetal development. Its transport takes place in an active manner against a concentration gradient, as the Ca²⁺ concentration in the fetal circulation is considerably higher than that of the mother. This active transfer is carried out by the ST, for which the first step is that of membrane-gated Ca²⁺ entry. It has been suggested that the influx of Ca²⁺ across the ST microvillous membrane is performed by a facilitated diffusion process (Kamath et al. 1992). Recently, the presence of Ca²⁺ transporter types 1 and 2 was demonstrated in cultured trophoblastic cells (Moreau et al. 2002a). Furthermore, Ca²⁺ is required in multiple cellular functions that include secretion, ionic conductance, cell-cycle regulation and programmed cell death (Berridge et al. 2000). However, the nature of Ca²⁺ conductance in the ST, and the mechanisms by which it is regulated, are poorly understood. Previous studies have demonstrated an increase in intracellular Ca²⁺ in the ST following exposure to various biologically active substances acting in both autocrine and paracrine fashions: GnRH (Currie et al. 1993), ATP and UTP (Petit & Belisle, 1995), ATP and angiotensin II (Karl et al. 1997), endothelin (Cronier et al. 1999), ATP (Clarson et al. 2002). However, the nature of Ca²⁺ channel activation in the ST remains controversial, with voltage-operated Ca²⁺ channels (VOCC; Meuris et al. 1994; Petit & Belisle, 1995; Robidoux et al. 2000), non-selective cationic channels (NSCC; Grosman and Reisin, 2000; Llanos et al. 2002; Long & Clarson, 2002), store-operated Ca²⁺ channels (SOCCs; Clarson et al. 2003) and receptor-operated Ca²⁺ channels (ROCCs; Bax et al. 1994) all having been described.

We previously demonstrated the presence of endothelin (ET) receptors A and B on the human trophoblastic membrane (Malassiné et al. 1993b). With the major signal transduction pathway of endothelin-1 (ET1) acting via phospholipase C (PLC) and the mobilization of intracellular Ca²⁺, we have thus used cytofluorimetry and pharmacological analysis to determine the nature of ET1-mediated Ca²⁺ entry into ST cells in primary culture.

Methods

Materials

ET1, trypsin, DNase I, NiCl₂, Indo-1 AM and nifedipine were purchased from Sigma (St Louis, MO, USA), BQ123 and BQ788 were from Neosystem (Strasbourg, France), monoclonal anti-human leucocytic antigen-A, B and C antibody (W6-32HL) was from Sera Laboratory (Crawley Down, UK) and SK&F96365 was from Calbiochem (VWR International, Fontenay-sous-bois, France). LOE 908 was kindly provided by Boehringer Ingelheim (Ingelheim, Germany). All other reagents were from standard suppliers.

Trophoblastic cell culture

Human placentas were obtained after caesarean section from mothers with uncomplicated pregnancies. The use of human placentas for this study was approved by the ethics committee of the Clinique du Fief de Grimoire (Poitiers, France). CT cells were isolated using a previously described method (Malassiné et al. 1993a), which was adapted from that of Kliman et al. (1986). Briefly, after several sequential trypsin/DNase digestions followed by Percoll gradient centrifugation, cells were further purified by means of W6-32HL. CT cells were diluted to a final concentration of 0.5 × 10⁶ ml⁻¹ in minimum essential medium (MEM) containing 10% fetal calf serum (FCS), 25 mm glucose and 50 μg ml⁻¹ gentamicin. Cells were plated onto glass coverslips in 35 mm plastic dishes (Nunclon, Nunc, Roskilde, Denmark) and incubated for 2 days at 37 °C in 5% CO₂. The culture medium was renewed daily. Cytokeratin 07 immunocytochemistry (clone OV.TL12/30, Dako, Denmark) was performed to confirm the cytotrophoblastic nature of the attached cells. After the purification procedure, 95% of the cells stained positively for cytokeratin.

Recording of [Ca²⁺]_i transients

Intracellular free Ca²⁺ concentrations ([Ca²⁺]_i) were measured by means of the ratiometric method with an inverted epifluorescence microscope (Olympus IX 70). Briefly, the Ca²⁺ indicator Indo-1 was used, for which fluorescence emissions of the Ca²⁺-free (485 nm) and Ca²⁺-bound (405 nm) forms of the indicator were collected using a dichroic filter and two photomultiplier tubes (excitation wavelength 355 nm). The Ca²⁺ activity was estimated as the ratio of the 405/485 nm fluorescence emission intensities. For loading of the probe, trophoblastic cells were incubated for 45 min in the dark in Tyrode solution (mm: 144 NaCl, 5.4 KCl, 2.5 CaCl₂, 1 MgCl₂, 0.3 NaH₂PO₄, 5 Hepes, 5.6 glucose, pH 7.4) containing the lipophilic form of the dye (Indo-1 AM dissolved in DMSO 0.3%) at a concentration of 3 μm. After carefully washing off the unincorporated fluorogenic dye, cells were incubated in Tyrode solution for a further 15 min in the dark to obtain complete de-esterification of the dye. STs with between six and eight nuclei accumulated in a central nuclei mount were identified with the aid of the inverted epifluorescence microscope. Variations of [Ca²⁺]_i with time were measured in a defined area located approximately in the centre of the trophoblastic cells. By means of a home-made gravity-based microperfusion system, test solutions were applied rapidly onto the ST under investigation by using a streamline flow directed from the opening of a stainless steel capillary tube (internal diameter 50 μm) positioned in the bath. Switching between different solutions was performed with electrovalves controlling different juxtaposed capillaries. Treatments were performed by perfusion of Tyrode solution containing ET1 or pharmacological agents. All experiments were conducted at room temperature (20 ± 1 °C).

Ratio analyses and statistics

Indo-1 signals were not calibrated in terms of absolute values since this was not necessary for the monitoring of variations of Ca²⁺ levels. Intracellular Ca²⁺ concentration changes were expressed as changes in the ratio of the 405/485 nm fluorescence emissions of the dye. Reported data represent the mean ± s.e.m. of the percentage difference in ratio between the basal level and peak or plateau levels, with n being the number of STs tested. One-way analysis of variance followed by a Dunnett's post hoc test was used to compare peak ratio values, while paired Student's t tests were used for statistical comparisons of baseline and plateau Ca²⁺ levels for different treatments.

Results

ST formation

When purified CT cells are cultured in the presence of 10% FCS, after adhesion and flattening, cells make initial contacts by pseudopodia with neighbouring cells, transform into cellular aggregates and fuse to form STs. This differentiation process was monitored by the immunostaining of cells for desmoplakin and βhCG secretion as previously described (Cronier et al. 1994, 2003; Frendo et al. 2003). Under the experimental conditions used here, a large proportion of mononuclear cells had differentiated into ST after 48 h of culture. Only STs with between six and eight nuclei amassed in a central mount were selected for further study.

Effect of ET1 on [Ca²⁺]_i in STs

Previous studies have demonstrated that, in our experimental conditions, ET1 (100 nm) is effective in inducing a Ca²⁺ response in around 75% of investigated cells (Cronier et al. 1999). As shown in Fig. 1, a stable resting [Ca²⁺]_i level is followed by a rapid increase in fluorescence ratio upon the addition of 100 nm ET1. This was followed by a sustained prolongation of [Ca²⁺]_i in 41% of ET1-responding cells. The effect of ET3 on Ca²⁺ activity presented an identical profile (data not shown).

Figure 1. Effect of endothelin-1 (ET1) on [Ca²⁺]_i in cultured syncytiotrophoblastic cells.

A, Ca²⁺ change in response to exposure of a syncytiotrophoblast to ET1 (10⁻⁷ m). After a rapid rise to peak of the fluorescence ratio, a decrease in [Ca²⁺] to a level higher than that of the basal level was observed (plateau). Arrows indicate the time of measurement for calculation of the percentage change in ratio expressed in B. B, averaged data for the percentage increase in fluorescence ratio (F₄₀₅/F₄₈₅). Data are means ± s.e.m.; peak n = 49; plateau n = 20.

The nature of ET receptor(s) implicated in the ET1-induced change in [Ca²⁺]_i was subsequently investigated using specific antagonists. As shown in Fig. 2A and also Fig. 4, prior and concomitant perfusion of cells with BQ123 (10⁻⁶ m), a specific ETA receptor antagonist, did not change the Ca²⁺ response to ET1 (n = 24), while the presence of BQ788 (10⁻⁶ m; Figs 2B and 4), a specific ETB receptor antagonist, completely inhibited the ET1-stimulated increase in Ca²⁺ (n = 13).

Figure 2. Analysis of ET1-induced Ca²⁺ response.

A and B, the effects of BQ123 (A; 10⁻⁶ m) and BQ788 (B; 10⁻⁶ m) on the [Ca²⁺]_i increase induced by ET1 (10⁻⁷ m). C, recording of the Ca²⁺ response to ET1 (10⁻⁷ m) for a syncytiotrophoblast (ST) bathed in Ca²⁺-free extracellular medium. D, Ca²⁺ response to ET1 (10⁻⁷ m) after pre-incubation of the ST for 15 min in the presence of the phospholipase C (PLC) inhibitor, U73122 (2 μm). Arrows in A–D indicate the time of measurement used for the calculation of the percentage changes in ratios expressed in Fig. 4.

Figure 4. Average of percentage variation of fluorescence ratio during the spike obtained after various pharmacological treatments.

Data represent means ± s.e.m.; ET1 alone (n = 49); ET1 + BQ123 (n = 24); ET1 + BQ788 (n = 13); ET1 +0 Ca²⁺ (n = 10); ET1 + U73122 (n = 12); ET1 + Ni²⁺ (n = 13); ET1 + nifedipine (n = 20); 100 mm K⁺ (n = 15). One-way analysis of variance revealed significant differences between treatments (F = 39.1; P < 0.0001; d.f. = 7). Dunnett's post hoc test showed a non-significant difference between ET1 alone and ET1 + BQ123, and that BQ788, 0 Ca²⁺, U73122 and Ni²⁺ significantly inhibited ET1-mediated effects on Ca²⁺ activity (***P < 0.01). On the other hand, nifedipine (Nif) slightly increased the ET1 response (**P < 0.05). Furthermore, hyper-potassium (HypK⁺) perfusion induced a slight decrease of the baseline Ca²⁺ level (two-tailed paired Student's t test; **P < 0.05).

The signal transduction pathway involved in the Ca²⁺ response in human STs could be produced by the mobilization of intracellular Ca²⁺ following the ET-induced activation of PLC and/or by an influx of extracellular Ca²⁺. Therefore, the relative contribution of extracellular Ca²⁺ was evaluated. In Ca²⁺-free medium (Figs 2C and 4), the resting [Ca²⁺]_i was not affected and ET1 was able to induce a spike in [Ca²⁺]_i in 67% of STs tested. However, under these conditions, the Ca²⁺ spike was less than that observed in control STs, and the plateau phase was not obtained (n = 10). Therefore an influx of extracellular Ca²⁺ represents a large portion of the ET1-induced elevation of cytosolic Ca²⁺ under control conditions.

Coupling of the ETB receptor to PLC was examined pharmacologically using U73122, a membrane-permeable inhibitor which inhibits PLC by disrupting its coupling to G protein. As shown in Figs 2D and 4, pre-incubation of cells with 2 μm U73122 completely abolished the ET1-induced [Ca²⁺]_i increase (spike and plateau; n = 12) indicating that the mobilization of Ca²⁺ from inositol 1,4,5-trisphosphate (IP₃)-sensitive intracellular Ca²⁺ stores is a pre-requisite for the ET1-induced Ca²⁺ response.

Characterization of channels involved in Ca²⁺ entry

To characterize the nature of Ca²⁺ channels involved in ET1-evoked Ca²⁺ entry in STs, a panel of Ca²⁺ channel inhibitors was tested. Ni²⁺ is an inorganic, non-specific inhibitor of Ca²⁺ channels, and it has been demonstrated to block Ca²⁺ entry in other cell types. As illustrated in Fig. 3A and Fig. 4, 1 mm Ni²⁺ significantly reduced the ET1-evoked initial Ca²⁺ entry by 70% and inhibited the plateau phase (n = 13). The inhibitory effects of Ni²⁺ were reversible.

Figure 3. Effects of Ni²⁺, nifedipine and hyper-potassium solution on the ET1-induced Ca²⁺ response in a ST.

Ca²⁺ response to ET1 (10⁻⁷ m) in the presence of 1 mm Ni²⁺ (A) or nifedipine (10⁻⁶ m; B). C, representative recording of the Ca²⁺ response to 100 mm K⁺ perfusion and subsequent response to ET1 perfusion (10⁻⁷ m). Arrows in A–C indicate the time of measurement used for the calculation of the percentage changes in ratios expressed in Fig. 4.

The ET-induced Ca²⁺ entry could occur via voltage-operated or voltage-insensitive Ca²⁺ channels. The dihydropyridine compound nifedipine is an L-type Ca²⁺ channel blocker frequently used to demonstrate the presence of voltage-operated channels. As illustrated in Figs 3B and 4, perfusion with 1 mm nifedipine did not block ET1-induced Ca²⁺ entry (n = 20). Moreover, trophoblastic membrane depolarization induced by the perfusion of cells with a hyper-potassium solution did not induce an influx of Ca²⁺ into cells (Figs 3C and 4), thereby confirming the absence of voltage-operated Ca²⁺ channels. It should be noted that the antagonists and blockers used here (BQ123, BQ788, U73122 and nifedipine) had no significant effect on baseline Ca²⁺ levels (two-tailed paired Student's t test).

It is possible, therefore, that the ET-induced Ca²⁺ entry occurs via voltage-insensitive Ca²⁺ channels. Recently, two types of organic compounds have been used as pharmacological tools to block L-type Ca²⁺ channel blocker-insensitive Ca²⁺ influx mechanisms, they are the imidazole derivatives SK&F96365 and LOE 908. Importantly, NSCCs and SOCCs can be distinguished in terms of their sensitivities to these compounds (Kawanabe et al. 2001, 2002). NSCCs are sensitive to LOE 908, whereas SOCCs are resistant to LOE 908 and sensitive to SK&F96365. As shown in Fig. 5A and B, perfusion of STs with 30 μm SK&F96365 during the plateau phase decreased the ET1-evoked Ca²⁺ entry by 80% (n = 10). However, SK&F96365 did not totally abolished the plateau phase. Perfusion of STs with LOE 908 during the plateau phase induced a dose-dependent decrease of the ET1-evoked Ca²⁺ entry without totally abolishing the ET1 response (Fig. 5C and D).

Figure 5. Effects of SK&F96365 and LOE 908 on the ET1-induced calcium plateau phase.

A and C, recordings of Ca²⁺ changes induced in response to ET1 (10⁻⁷ m) in the presence of 30 μm SK&F96365 (A) and 0.1, 1 and 10 μm LOE 908 (C) during the plateau phase. Arrows indicate the time of measurement used for the calculation of the percentage changes in ratios expressed in B and D. B, average percentage change in fluorescence ratio (F₄₀₅/F₄₈₅) for results of SK&F96365 treatment presented in A. Data represent means ± s.e.m.; ET1 plateau (n = 10); ET1 plateau + SK&F96365 (n = 10); (**P < 0.01; one-tailed paired Student's t test). D, average of percentage increases in fluorescence ratio (F₄₀₅/F₄₈₅) for results of LOE 908 treatment presented in A. Data represent means ± s.e.m.; ET1 plateau (n = 10); ET1 plateau + LOE 908 0.1 μm (n = 6); ET1 plateau + LOE 908 1 μm (n = 6); ET1 plateau + LOE 908 10 μm (n = 6); (*P < 0.05; one-tailed paired Student's t test).

These ET1-evoked Ca²⁺ responses in the presence of SK&F96365 or LOE 908 suggest the participation of two types of voltage-insensitive Ca²⁺ channels, i.e. NSCCs and SOCCs.

Discussion

In the present study, an attempt has been made to pharmacologically characterize the nature of channels involved in Ca²⁺ entry in human trophoblasts in culture. Owing to its position in the human chorionic villi, bathed with maternal blood in the intervillous space, the ST is the site of numerous placental functions including various forms of exchange, as well as metabolism and the synthesis of hormones required for fetal growth and development (Benirschke & Kaufmann, 2000). It is now well established that the villous trophoblast differentiates from the fusion of CT cells into a ST, and that culture of trophoblastic cells provides a useful way to study trophoblast physiology (Kliman et al. 1986; Cronier et al. 1994; Malassiné & Cronier, 2002).

The results presented here indicate that ET1 stimulated a biphasic (transient and sustained) increase in [Ca²⁺]_i in trophoblastic cells. This Ca²⁺ response is mediated by the ETB receptor, since BQ788 totally abolished the response. ETB receptors were previously demonstrated in the term trophoblastic microvillous membrane (Malassiné et al. 1993b), in first trimester trophoblastic cells (Cervar et al. 1997) and in a human extravillous trophoblast cell line (Chakraborty et al. 2003). In various cell types, ETB receptors are protein Gq-coupled receptors activating PLC to produce IP₃ as well as 1,2-diacylglycerol (DAG). IP₃ stimulates the release of Ca²⁺ from IP₃-sensitive stores and subsequently the activation of SOCCs. Furthermore, it was recently demonstrated that a non-metabolizable analogue of DAG (OAG) could directly activate members of the transient receptor potential (TRP) superfamily, thereby inducing an increase of intracellular Ca²⁺ in placental explants (Clarson et al. 2003). Abrogation of the ET1-mediated Ca²⁺ response after pre-treatment with U73122, a PLC inhibitor, indicates firstly that in human trophoblastic cells ETB receptors are coupled to PLC via Gq protein. The abrogation of both the spike and plateau phases indicates that the inhibition of PLC could also induce inhibition of other functional Ca²⁺ channels such as SOCC, NSCC or other members of the TRP superfamily. The persistence of the rapid transient rise in [Ca²⁺]_i in Ca²⁺-free extracellular medium confirms the release of Ca²⁺ from intracellular Ca²⁺ stores in response to ET1 stimulation. Furthermore, abolition of the sustained increase in [Ca²⁺]_i in Ca²⁺-free extracellular medium argues for the entry of extracellular Ca²⁺ during the plateau phase. Inhibition of the plateau phase by Ni²⁺ confirms the existence of an ET1-induced Ca²⁺ entry.

Based on the findings presented here, several candidate Ca²⁺ channel types could mediate Ca²⁺ entry. These include voltage-operated channels, store-operated Ca²⁺ channels, receptor-operated channels and non-selective cationic channels. Abrogation of the Ca²⁺ response in the presence of U73122 argues for the absence of a receptor-operated channel stimulation induced by ET1. No evidence for the presence of voltage-operated channels was demonstrated during ET1 stimulation, since nifedipine did not reduce the ET1-induced Ca²⁺ response, and depolarization with a hyper-potassium solution had no effect. These results confirm findings from a previous electrophysiological study by this group (Cronier et al. 1999). Other studies have suggested the presence of nifedipine-sensitive channels in the ST of term placenta (Polliotti et al. 1994; Meuris et al. 1994; Cemerikic et al. 1998; Robidoux et al. 2000). However, Bax et al. (1994) using Ca²⁺ measurements could not demonstrate the presence of these voltage-operated channels in cultured trophoblastic cells. Furthermore, it was reported that in a trophoblastic cell line (BeWo cells), Ca²⁺ uptake was not influenced by L-type Ca²⁺ channel modulators (Moreau et al. 2001). This insensitivity towards blockers of voltage-operated Ca²⁺ channels was also observed with other experimental models including placental perfusion (Stulc et al. 1994), ST membrane vesicles (Kamath et al. 1992) and placental explants (Long & Clarson, 2002; Clarson et al. 2003). It should be pointed out that physiological studies require a clear identification of the investigated cells. Under the experimental conditions employed here, the procedure for trophoblast isolation prevents contamination by other placental cells such as macrophages, fibroblasts, endothelial or smooth muscle cells.

Recently, SK&F96365 (Merrit et al. 1990) and LOE 908 (Encabo et al. 1996) have been used as pharmacological tools to analyse L-type Ca²⁺ channel blocker-insensitive Ca²⁺ influx mechanisms. These compounds have permitted characterization of Ca²⁺ entry routes in excitable and non-excitable cells. Indeed, NSCCs are sensitive to LOE 908, whereas SOCCs are resistant to LOE 908 and sensitive to SK&F96365 (Kawanabe et al. 2001, 2002). From the pharmacological experiments reported here, the presence of both SOCC and NSCC activation during the ET1-induced Ca²⁺ response in cultured ST cells has been demonstrated.

SOCCs serve as an important class of Ca²⁺ entry channel which are activated by depletion of intracellular Ca²⁺ stores upon stimulation of G-protein-coupled receptors (Berridge et al. 2000; Peng et al. 2003). In electrically non-excitable cells, SOCCs serve as one of the main routes for the entry of extracellular Ca²⁺. The presence of SOCCs in cultured cytotrophoblastic cells was previously suggested in response to stimulation by various ligands (Petit & Belisle, 1995; Karl et al. 1997; Cronier et al. 1999; Clarson et al. 2002). Recently, using placental explants, Clarson et al. (2003) demonstrated pharmacologically the presence of SOCC in term placenta, the expression of transient receptor potential canonical (TRPC) mRNA in first trimester and term placentas, and the immunolocalization of TRPC3, 4 and 6 in term ST cells. They concluded that store-operated Ca²⁺ entry occurs in human term placenta and that it may be gestationally regulated. Here we have demonstrated the activation of SOCCs by ET1 in cultured human term trophoblastic cells. The molecular identity of this SOCC needs to be identified, but based on previous studies, CaT1 (Moreau et al. 2002a), TRPC (Clarson et al. 2003) and polycystin-2 (Ong et al. 1999) could be candidates for this function. In situ hybridization studies have demonstrated the presence of CaT1 in STs (Peng et al. 2001; Wissenbach et al. 2001). Its expression seems to be correlated to Ca²⁺ uptake activity and hCG secretion (Moreau et al. 2002a). Moreover, another member of the TRP superfamily, polycystin-2, is present in term human STs (Gonzalez-Perret et al. 2001).

NSCC form a mixed group of ion channels that include ligand-gated, mechanosensitive and hyperpolarization- or stress-activated channels. NSCCs are widely distributed in numerous tissue types and could be permeable to Ca²⁺ ions. Many of their biophysical and regulatory properties have been described (for review, see Nilius, 2003; Clapham, 2003), but channel functions remain unknown in many cases. The presence of NSCCs in the human placenta has been previously considered by means of electrophysiological studies on placental CT cells (Clarson et al. 1999), after fusion of microvillous membranes with planar lipid bilayers (Grosman & Reisin, 2000) or after reconstitution of brush border membranes into giant liposomes (Riquelme et al. 1995; Llanos et al. 2002). Moreover, in excitable (Van Renterghem et al. 1988; Enoki et al. 1995; Minowa et al. 1997) and non-excitable cells (Enoki et al. 1995; Lee et al. 1999), ET1 was able to activate NSCCs.

SOCCs and NSCCs appear to be essential in replenishing ST Ca²⁺ stores (Putney, 1999, 2001) and serve as an important means of Ca²⁺ entry in trophoblastic cells (Shennan & Boyd, 1987; Illsley & Sellers, 1992). Moreover, in non-excitable cell types, activation of NSCCs and SOCCs leading to Ca²⁺ influx seems to play a role in processes that include secretion, cell proliferation, gene transcription and cell death (Berridge et al. 2000). In the human trophoblast, transient intracellular Ca²⁺ variations could affect these processes. It has been demonstrated that in purified human trophoblastic cells, raising [Ca²⁺]_i induces an increase of K⁺ and Cl⁻ efflux (Kibble et al. 1996; Turner et al. 1999; Clarson et al. 2002). As the placental transfer of maternal calcium is carried out in vivo by the ST, SOCCs and Ca²⁺-permeable NSCCs could also represent regulated modes of Ca²⁺ entry. Furthermore, Ca²⁺ ions could serve as mediators implicated in gap junctional intercellular communication during trophoblastic fusion (Cronier et al. 1994, 1999, 2003).

Various bioactive substances have been demonstrated to induce an increase in intracellular Ca²⁺ in isolated trophoblastic cells. These include GnRH (Currie et al. 1993), ATP and UTP (Petit & Belisle, 1995), ATP and angiotensin II (Karl et al. 1997), ET (Cronier et al. 1999) and ATP (Clarson et al. 2002). The human placenta appears to be an important source of ETs; cultured trophoblastic cells have been shown to release ET1 and to express pre-pro-ET1 and pre-pro-ET3 mRNA (Malassiné et al. 1993a; Robert et al. 1996). Since the first description of ETs, it is evident that these peptides display a multitude of biological functions controlling cellular ion fluxes, cell-to-cell communication, hormone release, cell chemokinesis, cell proliferation, cell differentiation, and the growth and progression of various tumours (Nelson et al. 2003). We have previously demonstrated that ET1 impairs trophoblast differentiation (Cronier et al. 1999). Furthermore, ET1 stimulates the invasion of first trimester trophoblastic cells (Cervar et al. 1997) and the migration of a human extravillous trophoblast cell line (Chakraborty et al. 2003). Further studies are required to determine the implications of ET1-induced Ca²⁺ increases in the human trophoblast.