Maintaining placental syncytiotrophoblast, a specialized multinucleated transport epithelium, is essential for normal human pregnancy. Syncytiotrophoblast continuously renews through differentiation and fusion of cytotrophoblast cells, under paracrine control by syncytiotrophoblast production of human chorionic gonadotropin (hCG). We hypothesized that K+ channels participate in trophoblast syncytialization and hCG secretion in vitro. Two models of normal-term placenta were used: 1) isolated cytotrophoblast cells and 2) villous tissue in explant culture. Cells and explants were treated with K+ channel modulators from 18 h, and day 3, onward, respectively. Culture medium was analyzed for hCG, to assess secretion, as well as for lactate dehydrogenase (LDH), to indicate cell/tissue integrity. hCG was also measured in cytotrophoblast cell lysates, indicating cellular production. Syncytialization of cytotrophoblast cells was assessed by immunofluorescent staining of desmosomes and nuclei. Over 18–66 h, mononucleate cells fused to form multinucleated syncytia, accompanied by a 28-fold rise in hCG secretion. 1 mM Ba2+ stimulated cytotrophoblast cell hCG secretion at 66 h compared with control, whereas 5 mM tetraethylammonium (TEA) inhibited hCG secretion by >90%. 0.1–1 mM 4-aminopyridine (4-AP) reduced cytotrophoblast cell hCG secretion and elevated cellular hCG; without altering cellular integrity or syncytialization. In villous explants, hCG secretion was not altered by 1 mM Ba2+ but inhibited by 5 mM 4-AP and 5/10 mM TEA, without affecting LDH release. Anandamide, pinacidil, and cromakalim were without effect in either model. In conclusion, 4-AP- and TEA-sensitive K+ channels (e.g., voltage-gated and Ca2+-activated) regulate trophoblast hCG secretion in culture. If these K+ channels participate in hCG secretion in situ, they may regulate trophoblast turnover in health and disease.
- human chorionic gonadotropin
- voltage-gated K+ channel
during human pregnancy, the placenta facilitates the transport of nutrients to the fetus, eliminates waste products of fetal metabolism and produces hormones, which are required to support the fetoplacental unit. These functions are performed by the syncytiotrophoblast, a highly specialized multinucleated epithelial cell layer that surrounds the placental villi containing the fetal capillaries. The syncytiotrophoblast forms a physical, immunological, and endocrine interface between the maternal and fetal circulations, and it has a microvillous plasma membrane in contact with maternal blood and a basal plasma membrane facing the fetal capillaries. Consequently, the growth and well-being of the fetus are dependent on appropriate syncytiotrophoblast development and function throughout gestation.
Syncytiotrophoblast has a short life span and must be continuously renewed throughout pregnancy from proliferative mononucleate cytotrophoblast cells, which exit the cell cycle, differentiate and fuse with the overlying syncytial layer (26). Following fusion, nuclear material undergoes apoptosis and is sequestered into syncytial knots, which are shed into the maternal blood (25). Appropriate cell turnover to maintain syncytiotrophoblast is necessary for nutrient transfer to the fetus, and for the production of hormones that regulate fetal and placental growth. Human chorionic gonadotropin (hCG), produced by terminally differentiated syncytiotrophoblast, is particularly important in maintaining the placenta, having an autocrine/paracrine effect on cytotrophoblast cells to promote their differentiation and fusion, thereby regulating syncytiotrophoblast renewal (23, 47, 51). Syncytiotrophoblast volume is correlated with birth weight (13), and disruption of its renewal is a proposed underlying pathology of gestational diseases such as preeclampsia (PE) and intrauterine growth restriction (IUGR); these conditions are associated with altered cytotrophoblast proliferation and apoptosis with reduced syncytiotrophoblast volume compared with normal pregnancy (36). Thus, appropriately regulated syncytiotrophoblast cellular turnover, involving proliferation, differentiation, fusion, and apoptosis, is essential for successful pregnancy.
In many organs, cellular processes that are required for tissue development, maintenance and repair, including proliferation, differentiation, fusion, and apoptosis, are regulated by potassium (K+) channels. Members of each of the five main functional families of K+ channels, namely voltage-gated (KV), calcium-activated (KCa), ATP-sensitive (KATP), inwardly rectifying (KIR), and 2-pore domain (K2P), participate in the regulation of cell volume, proliferation, and apoptosis (30, 32, 45), cell fusion (11, 34), and tissue repair (50). K+ channels are also important therapeutic targets to control unregulated proliferation and apoptosis in disease (40) and to promote apoptosis in cancer (15, 49). Furthermore, K+ channels play a key role in regulating hormone secretion (33). However, the importance of K+ channels in regulating syncytiotrophoblast cell renewal and endocrine function has yet to be explored.
In comparison with other epithelia, relatively little is known of the functional expression of K+ channels in placental syncytiotrophoblast. Using fragments of placental villi, we demonstrated a Ba2+-sensitive K+ conductance in the syncytiotrophoblast microvillous membrane (18) and an increase in K+ conductance induced by hyposmotic swelling (4, 5). Although the molecular identities of the channels underlying these conductances have not been identified, basal 86Rb+ efflux from multinucleated cytotrophoblast cells in vitro is reduced by anandamide, an inhibitor of the K2P channel TASK-1 (3). In addition, ATP or volume-activated 86Rb+ efflux is dependent on Ca2+ and is inhibited by iberiotoxin, a blocker of intermediate conductance Ca2+-activated K+ channels (IKCa) (8, 16). Single-channel patch-clamp studies identified an intermediate conductance K+-selective channel and a large-conductance Ca2+-activated K+ channel (BKCa) in cytotrophoblast cells (19). Using whole cell recording, we demonstrated the presence of a Ba2+-sensitive strong inwardly rectifying K+ current with the characteristics of KIR2.1 in cytotrophoblast cells, the incidence of which increased with cell multinucleation (7). This implies that, in common with skeletal muscle (34), KIR2.1 might have a role in cytotrophoblast cell fusion.
Two in vitro models have been used to represent some of the features of trophoblast cell turnover in situ. In the first model, cytotrophoblast cells are isolated from human normal-term placenta (19, 29); when maintained in primary culture, isolated mononucleate cytotrophoblast cells are mitotically inactive and over the 24 h, they aggregate, differentiate, and then fuse to form multinucleate cells by day 3 or 4, which is reminiscent of the syncytiotrophoblast in vivo. These multinucleate aggregates produce and secrete ∼25 times more hCG than undifferentiated mononucleate cytotrophoblast cells (14, 20). 2) Normal-term placental villous tissue in explant culture. Villous explants can be maintained in culture for up to 11 days (48). The syncytiotrophoblast is shed over the first 48 h and thereafter regenerates by a process resembling trophoblast renewal during pregnancy, to form a multinucleate layer reminiscent of the syncytiotrophoblast in situ. hCG secretion is low at day 2, following loss of syncytiotrophoblast, and then increases to peak at day 4 and 5, coincident with new growth of differentiated syncytiotrophoblast (48). hCG secretion is often used as a marker of trophoblast biochemical differentiation in both cytotrophoblast cells and villous explants.
We hypothesized that K+ channels are involved in the regulation of trophoblast differentiation and hCG secretion. To test this hypothesis, we examined the effect of K+ channel modulators, chosen to target the major families of K+ channels, on the biochemical and morphological differentiation of trophoblast using cytotrophoblast cells and placental villous explants prepared from normal-term placenta.
MATERIALS AND METHODS
Unless stated otherwise, all materials used were obtained from Sigma-Aldrich (Poole, UK).
Cell Isolation and Culture
Normal-term placentas (>37 wk) from uncomplicated pregnancies were obtained following vaginal delivery or Caesarean section with informed patient consent. Cytotrophoblast cells were isolated and separated on a Percoll gradient using an adaptation of the method used by Kliman et al. (29), with modifications previously described by Greenwood et al. (19). Cells were plated onto 35-mm culture dishes (Nunc) or 16-mm glass coverslips, at densities of 3 × 106 and 1.5 × 106, respectively, and maintained in culture medium [Dulbecco's modified Eagle's medium (DMEM) and Ham's F-12 1:1, 10% FCS (heat inactivated), 1% gentamicin, 0.6% glutamine, 0.2% penicillin, 0.2% streptomycin] at 37°C in a humidified incubator (95% air/5% CO2).
Cytotrophoblast cells were maintained in culture for 66 h. At 18 h, cultures were washed 3 times with warm PBS to remove nonadherent cells, and the culture medium was replaced. Thereafter, medium was replaced daily with or without the following four K+ channel inhibitors: BaCl2 (0.1 and 1 mM), chosen as it blocks many K+ channels and has effects on placental syncytiotrophoblast, depolarizing the microvillous membrane (18) and inhibiting KIR2.1 currents in multinucleated cytotrophoblast cells (7); 4-aminopyridine (4-AP: 0.01–5 mM), which blocks KV channels (21); tetraethylammonium (TEA; 0.1–10 mM), an inhibitor of KV and KCa channels; anandamide (1, 10, and 100 μM), which blocks the K2P channel TASK-1, and reduced basal 86Rb+ efflux from cytotrophoblast cells (3). Cells were also treated with the KATP channel openers pinacidil and cromakalim (0.1–10 μM) and nifedipine (0.1–100 μM), an inhibitor of L-type voltage-gated Ca2+ channels. Diluents for anandamide and nifedipine (ethanol and DMSO, respectively) were added to their corresponding controls (final concentration: ethanol, 0.1%, and DMSO, 0.1 and 1%).
Placental Explant Culture
Six 1-cm3 sections of normal-term villous placental tissue were taken from different areas of each placenta midway between the chorionic and basal plate and placed in sterile PBS. Villous tissue was further dissected into explants of ∼3 mm3. Explants were placed in 74-μm polyester mesh Netwells, in 15-mm insert 12-well plates; 3 explants and 1.5-ml culture medium (10% CMRL-1066, 100 μg/ml streptomycin sulfate, 100 IU/ml penicillin-G, 0.1 μg/ml hydrocortisone, 1 μg/ml insulin, 0.1 μg/ml retinol acetate, 5% FCS; pH 7.2) were placed into each well, with the tissue supported on the mesh at the liquid-gas interface.
Villous explants were maintained in culture at 20% oxygen (37°C in a humidified incubator in 95% air/5% CO2) for 6 days (i.e., day placenta collected = day 0; day of harvest = day 6). Culture medium was collected from all wells daily. Explants were treated with the following ion channel modulators on days 3, 4, and 5 (following the collection of old media and the replacement with new): BaCl2 (1 mM), 4-AP (1 and 5 mM), TEA (5 and 10 mM), anandamide (100 μM), cromakalim (10 μM), and nifedipine (10−4M). Following collection, medium was frozen at −20°C and subsequently was analyzed for hCG and lactate dehydrogenase (LDH). On day 6, explant tissue was dissolved in 0.3 M NaOH (37°C for 24 h) for subsequent protein analysis.
Measurement of LDH and hCG
Culture medium was collected at 18 and 66 h of cytotrophoblast cell culture, (18 and 24 h collection, respectively), and daily from villous explants, and stored in aliquots at −20°C. The medium was analyzed for LDH, which is released from necrotic cells and used as a marker of cell viability, using a cytotoxicity detection kit (Roche Diagnostics, Mannheim, Germany), following the manufacturer's instructions. Culture medium was also assayed for secreted hCG using radioimmunoassay (cells: MP Biomedicals, Solon, OH) or ELISA (explants: DRG Diagnostics, DRG International, Marburg, Germany). In addition to secreted hCG, the cellular production of hCG from cytotrophoblast cells was assessed by lysing cells after 66 h of culture in 0.5 ml water for 30 min. Lysed cells were then scraped from the culture dish, the lysate centrifuged to pellet cellular protein, and the supernatant assayed for cellular hCG.
LDH release and hCG secretion/production from cells and explants were expressed as arbitrary units per hour per milligram protein and milli-International Units per hour per milligram protein and respectively. Cell and explant protein were determined using the Bradford method (Bio-Rad Laboratories, Munich, Germany).
Cytotrophoblast cells were fixed in methanol at −20°C for 25 min and then stored at 4°C in PBS prior to immunofluorescent staining. Following a 37°C incubation with blocking solution (2% FCS, 2% BSA, and 0.1% Tween × 20 in PBS) to reduce nonspecific binding, cells were incubated with primary antibody for 30 min at 37°C in the dark. Primary antibodies were monoclonal mouse anti-desmosomal protein (D1286, mouse IgG1 isotope, 1:200 dilution), cytokeratin monoclonal mouse anti-human cytokeratin-7 (DAKO, Glostrup, Denmark; 1:100 dilution) or monoclonal mouse anti-vimentin (V6630, 1:100 dilution). The cells were then washed with PBS and the secondary antibody, goat anti-mouse IgG FITC (F2012, 1:100 dilution) applied for 30 min at 37°C in the dark. After further washing with PBS, coverslips were mounted onto glass microscope slides using Vectashield mounting medium for fluorescence with propidium iodide. Immunofluorescent images were taken using an Olympus IX70 confocal microscope (×400 magnification).
Analysis of Cytotrophoblast Cell Multinucleation
Confocal images of cytotrophoblast cells stained for desmosomes and nuclei were used to assess multinucleation as a measure of morphological differentiation. Using a previously published method (28), three observers, blinded to the identity of the images, counted the total number of nuclei in a given field and the number of nuclei in syncytia. For each image, the nuclei in syncytia were expressed as a percent of the total number of nuclei. A cell was classified as multinucleate if it contained ≥3 nuclei.
Expression of Results and Statistics
Protein content (μg), hCG secretion (mIU/ml/h/mg protein), and LDH release (arbitrary units/h/mg protein) from control (untreated) cytotrophoblast cells isolated from the same placenta and cultured for 18 and 66 h are expressed as mean ± SE (n = number of placentas). Differences between these variables at 18 and 66 h of culture were analyzed using a paired t-test. The effect of K+ channel inhibitors on cellular protein, LDH release, cellular and secreted hCG, and on the number of nuclei in syncytia, at 66 h is expressed as a % of control (untreated cytotrophoblast cells at 66 h of culture) and plotted as median and interquartile range. A nonparametric ANOVA (Kruskal Wallace) followed by a Wilcoxon Signed Rank Test was used to compare the effect of the K+ channel inhibitors to the untreated control (100%) for the corresponding cell isolate.
hCG secretion and LDH release from control explants, and following treatment with K+ channel blockers, are expressed as means ± SE (n = no of placentas). The effect of K+ channel inhibitors on hCG secretion and LDH release was compared with control using a 2-way ANOVA with Bonferroni post tests. P < 0.05 was considered significant.
Characteristics of isolated cytotrophoblast cells.
In untreated cytotrophoblast cells, hCG secretion increased 28-fold from 18 to 66 h of culture (4 to 110 mIU·ml−1·h−1·mg protein−1, respectively), consistent with the biochemical differentiation that is associated with the formation of syncytia (51). Cell protein content and LDH release fell significantly with time in culture (Fig. 1A). Dual immunofluorescent staining for desmosomes and nuclei showed cytotrophoblast cell morphological differentiation: cells were mononucleate at 18 h, aggregated, and then fused to form syncytia by 66 h (Fig. 1B, a and b). Purity was assessed by staining cells at 66 h of culture with the epithelial and mesenchymal intermediate filament markers cytokeratin-7 and vimentin, respectively. Almost all cells stained positive for cytokeratin-7, confirming that they were trophoblast cells of epithelial origin (Fig. 1B, c). Very occasionally, vimentin-positive cells were observed (Fig. 1B, d) and although the incidence was not quantified in the present study, we previously reported a maximum contamination of 4% (vimentin-positive cells as a % of the total number of cell nuclei) using the same isolation method (6). To test whether the substrate might alter the phenotype of primary cell cultures, hCG secretion and LDH release was assessed in culture medium collected from cells that were maintained on glass coverslips (and subsequently used for immunofluorescent staining). hCG secretion (not corrected for cell protein) increased 30-fold from 0.32 to 9.96 mIU·ml−1·h−1, and LDH fell from 0.064 to 0.012 arbitrary units/h, between 18 and 66 h of culture (P < 0.007: paired t-test: n = 10 placentas). These data confirm that cell integrity and biochemical differentiation were similar when cells were maintained on glass coverslips and plastic dishes.
Ba2+, 4-AP, and TEA: effect on cytotrophoblast cell differentiation.
Treatment of cytotrophoblast cells, beginning at 18 h and continuing for 48 h, with 0.1 mM BaCl2 had no effect on any of the variables measured at 66 h (Fig. 2). However, BaCl2 at 1 mM significantly increased hCG secretion compared with control (Fig. 2C), without altering cell protein or LDH release. This increase in hCG secretion was associated with a rise in cellular hCG (Fig. 2D). Neither concentration of BaCl2 altered morphological differentiation; the number of nuclei in syncytia, expressed as a % of the total number of nuclei, was unaltered compared with untreated cells at 66 h of culture (data not shown).
Treatment with 0.01, 0.1, and 1 mM 4-AP increased cell protein and reduced hCG secretion compared with control cells at 66 h (Figs. 3, A and C). At these concentrations, cellular integrity was maintained (Fig. 3B). At 0.01 and 0.1 mM 4-AP, the reduced hCG secretion was associated with an increase in cellular hCG. Despite these changes in hCG, 4-AP at 0.01–1 mM did not affect the ability of the cells to undergo morphological differentiation and form multinucleated syncytia (Fig. 3E).
Treatment of cells for 48 h with 2–5 mM 4-AP significantly reduced cell protein, increased LDH release (indicating necrosis), and inhibited hCG secretion (Fig. 3, B and C). Immunofluorescent staining of desmosomes and nuclei following treatment with 5 mM 4-AP showed that the cells were severely compromised; there was reduced cytoplasm with shrunken nuclei and almost complete absence of multinucleation (Fig. 3, E and F).
TEA at 0.1 to 1 mM did not alter cytotrophoblast cell protein, hCG secretion, or LDH release (Fig. 4, A–C). In contrast, at 5 and 10 mM, TEA markedly reduced hCG secretion (Fig. 4C). The inhibition of hCG secretion with 5 mM TEA occurred in the absence of any change in LDH release and was accompanied by a fall in cellular hCG (Fig. 4D). However, at 10 mM, the inhibition of hCG secretion was associated with an increase in LDH release, although it was less than that LDH release following maximum cell necrosis induced by treatment with valinomycin (382%: n = 8 placentas, data not shown).
Analysis of multinucleation showed that TEA did not alter the number of nuclei in syncytia at any concentration used (Fig. 4E). However, 10 mM TEA caused a reduction in cell cytoplasm and nuclear shrinkage (Fig. 4F, c), which, in association with the increase in LDH release (Fig. 4B), indicates cellular necrosis at this concentration.
Anandamide, pinacidil, cromakalim, and nifedipine: no effect on cytotrophoblast cell differentiation.
Anandamide, which will block TASK-1 channels, the KATP channel openers pinacidil and cromakalim and the L-type voltage-gated Ca2+ channel blocker nifedipine, administered over 18–66 h of culture, did not affect cell protein, LDH release, or hCG secretion at 66 h of culture compared with untreated controls from the same placentas (n = 3 placentas: data not shown). Examination of the cells with phase contrast microscopy following treatment with these ion channel modulators did not show any gross changes in morphology, and analysis of multinucleation was not undertaken.
Placental Villous Explants
Characteristics of placental villous tissue in explant culture.
Fig. 5 shows that hCG secretion and LDH release from villous explants had a time course similar to that previously reported (48). Medium hCG and LDH levels were high on day 1, probably due to release from damaged/degenerating syncytiotrophoblast. hCG secretion fell on days 2 and 3, when syncytiotrophoblast is shed, and then increased over days 4–6 consistent with the syncytiotrophoblast regeneration previously reported (48). LDH release also fell on day 2 and then remained stable throughout the remaining culture period, indicating maintenance of tissue viability.
Ba2+, 4-AP, and TEA: effect on villous explant hCG secretion and LDH release.
Neither 1 mM BaCl2 (Fig. 6A) nor 1 mM 4-AP (Fig. 6B) affected hCG secretion on days 4 and 5 of culture. However, 5 mM 4-AP caused a reduction in hCG secretion following addition on day 3, which was significantly lower than the corresponding control on days 5 and 6.
TEA decreased hCG secretion on days 5 and 6 of culture (Fig. 6C). At 5 mM, the reduction in hCG secretion with TEA was significantly lower than the corresponding control at day 6, but with 10 mM, the reduction in secretion was significantly reduced at both days 5 and 6.
LDH release from placental explants was not affected by Ba2+, 4-AP, or TEA (Fig. 6), indicating that tissue viability was not compromised by treatment with the K+ channel blockers.
Anandamide, cromakalim, and nifedipine.
Administered days 3, 4, and 5, did not affect hCG secretion or LDH release compared with untreated controls from the same placentas (n = 4 or 5 placentas; data not shown).
This study aimed to determine the role of K+ channels in trophoblast cell biochemical (hCG production/secretion) and morphological (multinucleation) differentiation in vitro. Although mRNA and/or protein for members of the KCa (BKCa and IKCa), KV, KATP (KIR6.1), KIR (KIR2.1), and K2P (TASK-1 and TASK-2) K+ channel families are expressed by placenta (2, 21, 27, 31), the functional roles of the channels are poorly understood. Therefore, we used K+ channel blockers with broad specificity to reveal functionally important K+ channel families in the regulation of trophoblast hCG secretion and multinucleation. Initially, we confirmed that we could reproduce the characteristic morphological and biochemical differentiation of cytotrophoblast cells isolated from normal-term placenta and maintained in culture, reported previously by ourselves (6, 20) and others (e.g., 29). We showed that mononucleate cytotrophoblast cells aggregated and fused to form multinucleated syncytia over 18–66 h and that this was accompanied by an increase in hCG secretion. In villous explants, we reproduced the time course of hCG secretion associated with syncytiotrophoblast degeneration/regeneration that we have reported previously (48).
Effects of BaCl2
Many K+ channels can be blocked by Ba2+. We have previously reported that acute application of 5 mM BaCl2 depolarizes the resting microvillous membrane potential of the syncytiotrophoblast (18) and inhibits both basal and swelling-activated 86Rb+ efflux from villous explants and multinucleated cytotrophoblast cells (17, 48), implicating the presence of Ba2+-sensitive channels in the trophoblast. In patch-clamp studies, we identified a strong inwardly rectifying K+ current, typical of KIR2.1; this current had a higher incidence in multinucleate than mononucleate cells and was inhibited by 5 mM BaCl2 at the extracellular face (7). As KIR2.1 regulates the fusion of myoblasts to form myotubes (11), the only syncytial human tissue other than placenta, we speculated that KIR2.1 might be important for cytotrophoblast cell fusion and the generation of multinucleated syncytiotrophoblast (7). However, in the present study, treatment of cytotrophoblast cells over 18–66 h with 0.1 and 1 mM BaCl2 did not alter the formation of syncytial cells. On the assumption that BaCl2 at 1 mM was sufficient to block KIR2.1 (higher concentrations of BaCl2 formed an insoluble precipitate in culture medium), the results suggest that the expression of KIR2.1 is a consequence, rather than a cause, of multinucleation.
Although there was no change in morphological differentiation, 1 mM BaCl2 increased cytotrophoblast cell hCG secretion compared with control. In support of this, 1 mM BaCl2 showed a tendency to elevate hCG secretion from villous explants on days 4 and 5 of culture (P < 0.06 at day 4: % change from control) with an increase in secretion observed in four out of five placentas. The mechanism of hCG secretion by trophoblast is not completely understood. It is thought that basal secretion involves constitutive release from vesicles without storage and that regulated release in response to secretagogues, such as gonadotropin-releasing hormone (GnRH), involves exocytosis of dense-core large secretory granules (39, 41, 52). Several studies have shown that hCG release from normal-term placental villous tissue is modulated by extracellular Ca2+ (35), consistent with hCG secretion by excitation-secretion coupling. It has been suggested that L-type voltage-gated Ca2+ channels (VGCC), which are expressed by placenta (38), permit Ca2+ entry into trophoblast cells, thereby regulating intracellular Ca2+ concentration ([Ca2+]i) and exocytosis. Thus, raising extracellular K+, to depolarize the cell membrane, stimulated hCG secretion in a nifedipine-sensitive manner (37). Nifedipine, a blocker of VGCC, also inhibited secretion induced by GnRH but had no effect on basal hCG release (46). In the present study, it is likely that 1 mM BaCl2 inhibited several K+ conductances in cytotrophoblast cells, causing cell membrane depolarization (18). Thus, if the depolarization were sufficient to reach the activation threshold for VGCC, Ca2+ entry into the cells would elevate [Ca2+]i and stimulate hCG release by exocytosis of secretory granules. In our hands, nifedipine did not alter hCG secretion from cytotrophoblast cells or placental villous explants, suggesting that VGCC are not involved in basal release.
Effects of 4-Aminopyridine
In contrast to BaCl2 and anandamide, the KV channel blocker 4-AP had marked effects on trophoblast in culture; 10 μM-1 mM 4-AP significantly raised cytotrophoblast cell protein compared with control at 66 h, by preventing the fall in protein that is usually observed between 18 and 66 h of culture (Fig. 1A). It is possible that cell survival was enhanced following K+ channel inhibition with 4-AP, particularly if apoptosis contributes to the protein loss over 18–66 h. Mononucleate cells undergo apoptosis in culture more readily than multinucleated cells (12), and blockade of K+ channels can inhibit apoptosis in nonplacental tissues (45). We did not examine apoptosis in the current study, but if 4-AP did prevent early apoptosis, the surviving cells did not contribute to hCG secretion, as this was significantly lower with 10 μM-1 mM 4-AP compared with control. Indeed, any cells protected from apoptosis must have remained mononucleate and undifferentiated, as the number of nuclei in syncytia was unchanged. Another possible explanation for the rise in cellular protein and concurrent drop in hCG secretion, is that blocking K+ channels with 4-AP inhibits the exocytosis of secretory vesicles, leading to accumulation of hCG in the cell (although it is not possible to rule out an additional effect of 4-AP to stimulate hCG synthesis). hCG, and perhaps other peptide hormones, retained within secretory vesicles would contribute to the increase in cell protein with 10 μM-1 mM 4-AP. To test this possibility, we assessed cellular hCG and found that treatment with 10 and 100 μM 4-AP significantly elevated intracellular hCG, supporting the view that blocking KV channels with 4-AP inhibited hCG secretion, leading to cellular accumulation. We propose that block of KV channels with 4-AP at 10 μM-1 mM could depolarize membrane potential sufficiently to inhibit Ca2+ entry through Ca2+-permeable nonselective cation channels (NSCC), which are functionally expressed by the placental trophoblast (9). hCG secretion can be regulated by Ca2+ entry through NSCC, as blockers of these channels inhibit its secretion from villous fragments (35). Thus, a reduction in Ca2+ entry through NSCC induced by KV channel-dependent membrane depolarization, which is below the threshold to stimulate Ca2+ entry through VGCC, could result in a decrease in [Ca2+]i and reduced hCG secretion.
At concentrations greater than 1 mM, 4-AP markedly inhibited cytotrophoblast cell hCG secretion, reduced cellular protein, and increased LDH release, indicating loss of cell viability. Following treatment with 5 mM 4-AP, desmosome and nuclear staining revealed a loss of cell cytoplasm and shrunken nuclei, indicating cellular necrosis. Assuming that the effects of 4-AP are selective for inhibition of K+ channels, these data indicate the importance of 4-AP-sensitive K+ channels in regulating endocrine secretion and maintaining cytotrophoblast cell integrity. Furthermore, the effects of 4-AP at nontoxic concentrations show that an increase in hCG expression is not obligatory for cytotrophoblast cell multinucleation (and vice versa), in agreement with previous reports that biochemical and morphological differentiation can be independent events (14).
4-AP at 1 mM did not alter hCG secretion from villous explants but at 5 mM dramatically reduced secretion without affecting LDH release. The difference in the efficacy of 4-AP (as well as efficacy of 1 mM Ba2+, which stimulated hCG secretion from cells but not tissue) on tissue, and cells might relate to the effective concentration of the blocker at the microvillous membrane; explants are suspended at the air/medium interface and it is probable that, in the absence of flow, unstirred layers develop in the vicinity of the microvilli. However, the data from villous explants support the contention that 4-AP can inhibit hCG secretion from viable trophoblast.
Effects of Tetraethylammonium
TEA did not have a major effect on cytotrophoblast cell biochemical and morphological differentiation. However, at 5 mM, TEA inhibited hCG secretion, and cellular hCG production but did not alter LDH release or multinucleation, suggesting a selective effect of TEA on hCG synthesis or release. Again, these results demonstrate that hCG production and/or secretion and cytotrophoblast cell multinucleation can be separated. Indeed, TEA at 10 mM almost completely prevented hCG secretion but did not alter morphological differentiation, as assessed by the number of nuclei in syncytia. However, immunofluorescent staining of cell isolates (3 out of 5: see example shown in Fig. 4F, c) revealed a large number of small syncytia, containing shrunken nuclei, compared with control cells, or cells treated with 0.1–5 mM TEA. LDH release was also moderately elevated with 10 mM TEA, indicating some loss of cellular integrity. Overall, assuming that the effects of TEA are specific to the blockade of K+ channels, the data imply that TEA-sensitive K+ channels participate in hCG secretion and are important to maintain cytotrophoblast cell viability and morphological differentiation.
Five and ten millimoles TEA significantly reduced hCG secretion from villous explants, without altering LDH release. Taken together with the effects on cultured cells, these data suggest that TEA-sensitive channels regulate hCG secretion from trophoblast.
Putative Roles for KV Channels in Trophoblast
Anandamide had no effect on cytotrophoblast cell biochemical and morphological differentiation, or hCG secretion for villous explants, suggesting that the anandamide-sensitive basal 86Rb+ efflux from cytotrophoblast cells, attributed to TASK-1 (3), does not play an obligatory role in hCG secretion and syncytialization. Furthermore, the lack of effect of pinacidil and cromakalim indicates that opening KATP channels does not modulate trophoblast cell hCG secretion or cytotrophoblast multinucleation in vitro. However, the effects of 4-AP and TEA implicate a role for KV and KCa channels in regulating trophoblast function.
4-AP is used as a relatively nonspecific blocker of KV channels, and TEA inhibits KCa (∼1 mM) and KV channels at 5–10 mM (10). As TEA reduced hCG secretion from trophoblast at 5–10 mM, it is likely that the effects were due to inhibition of KV channels. We propose that KV channels are important for regulating hCG production/secretion from syncytiotrophoblast and that altered KV channel activity will disrupt the autocrine/paracrine regulation of trophoblast cell turnover by hCG.
Eleven families of KV channels have been identified and classified according to sequence homology (21). 4-AP at 0.1–1 mM predominantly targets the KV1 (Shaker) family, in particular, KV1.3, 1.5 and 1.7 (21). As the cell membrane is permeable to 4-AP, it is possible that application of the blocker for 48 h at 2–5 mM achieves concentrations in the cell cytoplasm that are high enough to inhibit other KV channels, e.g., KV2.1, from the intracellular face (22). mRNA for KV1.5, 1.7, 6.1, 7.1, 7.2, and 7.4 has been detected in the placenta (31) but, as far as we are aware, there are no studies of KV channel protein expression, localization, or function in the trophoblast, and this remains to be explored.
Perspectives and Significance
Our knowledge of placental electrophysiology in general, and K+ channel function, in particular, is rudimentary compared with other tissues. K+ channels are likely to participate in many aspects of placental biology, including trophoblast development (cell proliferation, migration, fusion, and apoptosis), homeostasis (regulation of Em, intracellular volume and ions), and endocrine secretion, as well as regulating nutrient, ion, and water delivery to the developing fetus. Our results, showing that the KV channel blockers 4-AP and TEA inhibit hCG secretion from trophoblast, implicate an important role for these channels in regulating placental endocrine function. Recently, it has become evident that KV channel families, ubiquitously expressed by excitable tissue, are also widely distributed in nonexcitable cells, including epithelia, where they regulate cell volume, cell kinetics, electrolyte/nutrient transport, and hormone secretion (43). Furthermore, KV channel expression and function can be inhibited by hypoxia (1, 43) and oxidative stress (22). As PE and IUGR are associated with altered oxygenation and elevated reactive oxygen species in the placenta (24, 42, 44), it is possible that inhibition of trophoblast KV channels disrupts the regulation of trophoblast turnover by hCG in these pregnancy complications. However, further studies are necessary to identify the functionally relevant KV channels in trophoblasts of normal pregnancy and to explore their role in the pathogenesis of altered placental development and function in PE and IUGR.
The authors wish to thank the staff and patients at the Central Delivery Unit at St. Mary's Hospital, Manchester, UK, for their assistance. This work was supported by Tommy's, the baby charity, Action Research, and The University of Manchester.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society