Intermediate Conductance Ca2+-Activated K+ Channels Modulate Human Placental Trophoblast Syncytialization

Regulation of human placental syncytiotrophoblast renewal by cytotrophoblast migration, aggregation/fusion and differentiation is essential for successful pregnancy. In several tissues, these events are regulated by intermediate conductance Ca2+-activated K+ channels (IKCa), in part through their ability to regulate cell volume. We used cytotrophoblasts in primary culture to test the hypotheses that IKCa participate in the formation of multinucleated syncytiotrophoblast and in syncytiotrophoblast volume homeostasis. Cytotrophoblasts were isolated from normal term placentas and cultured for 66 h. This preparation recreates syncytiotrophoblast formation in vivo, as mononucleate cells (15 h) fuse into multinucleate syncytia (66 h) concomitant with elevated secretion of human chorionic gonadotropin (hCG). Cells were treated with the IKCa inhibitor TRAM-34 (10 µM) or activator DCEBIO (100 µM). Culture medium was collected to measure hCG secretion and cells fixed for immunofluorescence with anti-IKCa and anti-desmoplakin antibodies to assess IKCa expression and multinucleation respectively. K+ channel activity was assessed by measuring 86Rb efflux at 66 h. IKCa immunostaining was evident in nucleus, cytoplasm and surface of mono- and multinucleate cells. DCEBIO increased 86Rb efflux 8.3-fold above control and this was inhibited by TRAM-34 (85%; p<0.0001). Cytotrophoblast multinucleation increased 12-fold (p<0.05) and hCG secretion 20-fold (p<0.05), between 15 and 66 h. Compared to controls, DCEBIO reduced multinucleation by 42% (p<0.05) and hCG secretion by 80% (p<0.05). TRAM-34 alone did not affect cytotrophoblast multinucleation or hCG secretion. Hyposmotic solution increased 86Rb efflux 3.8-fold (p<0.0001). This effect was dependent on extracellular Ca2+, inhibited by TRAM-34 and 100 nM charybdotoxin (85% (p<0.0001) and 43% respectively) but unaffected by 100 nM apamin. In conclusion, IKCa are expressed in cytotrophoblasts and their activation inhibits the formation of multinucleated cells in vitro. IKCa are stimulated by syncytiotrophoblast swelling implicating a role in syncytiotrophoblast volume homeostasis. Inappropriate activation of IKCa in pathophysiological conditions could compromise syncytiotrophoblast turnover and volume homeostasis in pregnancy disease.


Introduction
The syncytiotrophoblast is the transporting epithelium of the human placenta being the interface between maternal and fetal blood. This highly specialized epithelial cell also performs a number of other functions including hormone production and secretion. Syncytiotrophoblast has a short life span and is renewed by cellular turnover in a tightly regulated process where proliferative mononucleate cytotrophoblasts exit the cell cycle, differentiate and fuse with the overlying syncytial layer [1]; both apoptosis and autophagy have been hypothesized to play a role in completing turnover [2,3].
In vitro models have been used to study some of the features of syncytiotrophoblast turnover. These include cytotrophoblasts isolated from normal term placenta and maintained in primary culture [4,5]. After 15-18 h of culture, cytotrophoblasts are predominantly mononucleate and secrete small amounts of human chorionic gonadotropin (hCG). Over 24-66 h they migrate, aggregate and fuse to become multinucleated, a process reminis-cent of syncytiotrophoblast formation in vivo [5][6][7]. This morphological differentiation is associated with a several-fold increase in the production and secretion of hCG. hCG, which is synthesized and secreted by terminally differentiated syncytiotrophoblast [8], is one key regulator of cytotrophoblast biology and acts in an autocrine/paracrine manner to facilitate syncytiotrophoblast renewal by promoting cytotrophoblast differentiation and fusion [9].
The importance of syncytiotrophoblast renewal for the progression of normal pregnancy is highlighted by the fact that its dysregulation is linked to pregnancy complications associated with maternal and/or fetal morbidity and mortality, in particular pre-eclampsia [10][11][12], fetal growth restriction [11][12][13] and maternal obesity [14]. In pre-eclampsia there is elevated cytotrophoblast proliferation [11,15,16] and apoptosis [11,[17][18][19], and a greater number of syncytial nuclear aggregates [20], compared to normal pregnancy. Furthermore, there is evidence to suggest that a rate-limiting step for syncytiotrophoblast formation, cytotrophoblast fusion, is reduced in pre-eclampsia [12,21].
Cytotrophoblasts isolated from placentas of women with preeclampsia have a lower rate of syncytialization than those of normal pregnancy [22]. Expression of syncytin-1 [23] and syncytin-2 [24], envelope fusogenic proteins that induce syncytium formation [23,25,26] is downregulated both in isolated cytotrophoblasts and placental villous tissue from pregnancies complicated with pre-eclampsia [22,24,27,28]. Syncytiotrophoblast expression of other fusogenic proteins, for example e-cadherin [16], is also reduced in pre-eclampsia. Collectively, dysregulation of the processes contributing to syncytiotrophoblast renewal culminates in a decrease in the total volume of syncytiotrophoblast in pregnancies complicated by pre-eclampsia and fetal growth restriction [29]. This has implications for nutrient delivery to the fetus as syncytiotrophoblast volume correlates with fetal weight [30]. However, the intracellular and extracellular signals that trigger and regulate cytotrophoblast fusion to form syncytiotrophoblast are not well understood.
In non-placental tissues, cellular proliferation, fusion and apoptosis can be regulated by members of the Ca 2+ -activated K + channel (K Ca ) family, in particular by intermediate conductance Ca 2+ -activated K + channels (IK Ca ; K Ca 3.1; single channel conductance 50-200 pS). IK Ca s are voltage-insensitive and are strongly activated by increased concentrations of intracellular Ca 2+ ([Ca 2+ ] i ; 300-700 nM) [31,32]. IK Ca mRNA was shown to be highly expressed by human placenta over 15 years ago [33] but the functions of IK Ca in the placenta have not been explored.
A major function of IK Ca is to regulate cellular volume [34][35][36][37][38]. IK Ca activation induces K + efflux from cells, which both lowers intracellular K + concentration and promotes the loss of water by osmosis to induce cell shrinkage [39]. Appropriate adjustment of cell volume and/or intracellular K + concentration is essential for cells to undergo proliferation, migration, fusion and apoptosis [40]. Indeed, in non-placental tissues, IK Ca has been shown to contribute to tissue homeostasis by regulating proliferation [31,[41][42][43], differentiation/fusion [44,45], cell migration [46][47][48] and apoptosis [49]. The ability of IK Ca to regulate cell volume has been revealed experimentally by exposing cells to an osmotic challenge [34,35,37,50]. When placed in hyposmotic solutions, cells initially swell but then restore their volume by a process of regulatory volume decrease (RVD). In many cells hyposmotic cell swelling elevates intracellular Ca 2+ which activates IK Ca , promotes K + efflux and water follows to achieve RVD [34]. However, a role for IK Ca in regulating renewal of syncytiotrophoblast and/or syncytiotrophoblast volume has yet to be explored.
We tested the hypotheses that IK Ca participates in the formation of multinucleate syncytiotrophoblast and that IK Ca has a role in syncytiotrophoblast volume regulation. Using isolated cytotrophoblasts in primary culture we confirmed IK Ca protein expression and tested the effects of IK Ca modulators on 86 Rb efflux, the formation of multinucleate syncytia and the secretion of hCG. To investigate whether IK Ca participate in syncytiotrophoblast RVD, cells were exposed to hyposmotic solutions and 86 Rb efflux measured in the presence and absence of IK Ca modulators.

Materials
Unless otherwise stated, all chemicals were from Sigma-Aldrich (Poole, UK).

Ethics Statement
Human placentas used in this study were obtained from St. Mary's Hospital Maternity Unit (Manchester, UK) following written informed consent as approved by the Local Research Ethics Committee (North West (Haydock Park) Research Ethics Committee (Ref: 08/H1010/55)). Placentas were collected at term (37-42 weeks) following uncomplicated pregnancy and delivery of a healthy baby by vaginal or Caesarean section. Exclusion criteria were body mass index .30 (measured at booking), pregnancy hypertension/pre-eclampsia, fetal growth restriction, gestational diabetes. The investigation conforms to the principles outlined in the Declaration of Helsinki.
Cytotrophoblasts were obtained using an adaptation of the method used by Kliman et al. [5], as previously described [4]. Briefly, full thickness placenta samples (,2 cm 3 ) were taken within 30 min of delivery and placed into sterile saline. Placental villous tissue was further dissected from each sample after removal of the chorionic plate and decidua. ,30 g of villous tissue were obtained and submitted to digestion 3 times in Hank's balanced salt solution containing 2.5% trypsin and 0.2 mg/ml deoxyribonuclease (DNAse I) for 30 min at 37uC (in agitation). After each digestion, 100 ml of supernatant were obtained, layered onto 5 ml newborn calf serum and spun for 10 min at 2200 rpm (10006g) at 20uC. Afterwards, pellets were resuspended in 1 ml Dulbecco's modified Earle's medium (DMEM; Invitrogen, Paisley, UK) and centrifuged for 10 min at 2200 rpm. The supernatant was discarded and the pellet resuspended in 6 ml DMEM and layered onto a discontinuous Percoll density gradient and centrifuged for 30 min at 2800 rpm (15006g). The bands between 35-55% Percoll were obtained and mixed with cell culture medium (DMEM: Ham's F-12 Nutrient Mixture (Invitrogen, Paisley, UK) 1:1, 10% fetal calf serum (heat inactivated), 1% gentamicin, 0.2% benzylpenicillin, 0.2% streptomycin, 0.6% glutamine), before centrifugation at 2200 rpm for 10 min. The final pellet was resuspended in 2 ml of cell culture medium. Cells were plated onto 35 mm culture dishes (Nunc, Fisher Scientific, Loughborough, UK) in cell culture medium, or 16 mm coverslips in 12-well culture plates, at densities of 1-1.3610 6 /ml and 1610 6 /ml respectively at 37uC in a humidified incubator (95% air/5% CO 2 ).
At 15, 42 and 66 h of culture, cell culture medium was collected and stored at 220uC for measurement of b-hCG (hCG b-subunit; produced by terminally differentiated syncytiotrophoblast, used to assess cytotrophoblast biochemical differentiation [51]). Coverslips were placed into 1 ml 0.3 M NaOH, cells scraped and the cell lysate stored at 4uC. These samples were used to measure protein content (mg) with Bio-Rad Protein Assay, based on the Bradford method (Bio-Rad Laboratories, Hempstead, UK).
In addition, at 15, 42 and 66 h of culture, cells were fixed in absolute methanol (permeabilizing fixative; to detect intracellular immunostaining) for 20 min at 220uC or in 4% paraformaldehyde (PFA; non-permeabilizing fixative; to detect immunostaining associated with cellular surface) for 15 min at room temperature and stored in PBS at 4uC prior to immunofluorescence staining.

Measurement of Cytotrophoblast hCG Secretion
The b-subunit of hCG is secreted by terminally differentiated syncytiotrophoblast and was used as an indicator of cytotrophoblast differentiation in culture [51]. b-hCG was assayed in cellconditioned culture medium at 15, 42 and 66 h of culture by ELISA (DRG Diagnostics, Marburg, Germany). Thawed samples were used following the instructions of the manufacturer. Optical density was measured at 450 nm using a VersaMax microplate reader (Molecular Devices, CA, USA). hCG secretion was expressed as mIU/ml/mg protein.

Immunofluorescent Staining
Methanol and PFA-fixed cells on 16 mm coverslips were washed in tris-buffered saline (TBS). Block of non-specific binding was performed for 30 min with 4% bovine serum albumin (BSA) in TBS. Cells were incubated for 1 h at room temperature with mouse monoclonal antibody to desmoplakin I+II (clone 2Q400; Abcam, Cambridge, UK), diluted 1:100 in TBS or mouse monoclonal antibody to IK Ca (K Ca 3.1; clone 6C1; extracellular epitope; Alomone labs, Jerusalem, Israel), diluted 1:50 in 1% BSA in TBS. Negative control was obtained by omission of the primary antibody. Cells were washed with TBS and the secondary antibody, FITC-polyclonal rabbit anti-mouse immunoglobulin (Dako, Cambridgeshire, UK) diluted 1:50 in TBS, was applied and cells incubated for 1 h at room temperature in the dark. After washing with TBS, coverslips were mounted using Vectashield mounting medium with propidium iodide nuclear counterstain (PI; Vector labs, Peterborough, UK). Immunofluorescent images were captured using a Zeiss AxioObserver Inverted Microscope (magnification 4006).

Analysis of Cytotrophoblast Multinucleation
Microscope images of cytotrophoblasts stained for desmoplakin and nuclei were used to assess multinucleation as a measurement of cytotrophoblast morphological differentiation. Based on a previously published method [51,55], 2-3 observers counted the total number of nuclei per given field and the number of nuclei in syncytium (multinucleated cell defined as $3 nuclei within desmoplakin boundaries) using ImageJ 1.45 software (National Institutes of Health, USA). The number of multinucleated cells was expressed as a percentage of the total number of nuclei within a given field (% of nuclei in multinucleate cells).
After 15 min, the cells were lysed in 0.3 M NaOH for ,1 h and scraped in order to release intracellular 86 Rb which was then counted in the supernatant to give a measure of total 86 Rb remaining in the cells at the end of the experiment (cellular 86 Rb). Effluxed and cellular 86 Rb was measured in a gamma-counter (Packard Cobra II Auto Gamma, CA, USA). All counts recorded were at least 10 times higher than background counts.
The time course of percentage (%) 86 Rb efflux was calculated at each time point as (( 86 Rb effluxed/ 86 Rb in cells) x100). The efflux rate constant was also determined making the assumption that 86 Rb efflux at steady state reflects the loss of 86 Rb from a single compartment (syncytiotrophoblast) limited by the K + permeability of the plasma membrane. Consequently, the loss of 86 Rb was measured by a first-order rate constant which was calculated over 10 min experimental period as (l n ( 86 Rb in cell at time t/ 86 Rb in cell at t 0 )) where t 0 is the cellular 86 Rb at the start of the experiment.

Expression of Results and Statistics
Statistical analysis was performed using GraphPad Prism version 5 software. hCG secretion and multinucleation from control untreated cytotrophoblasts was expressed as mean 6 standard error (SE) with n as the number of placentas. hCG secretion and multinucleation in TRAM-34 and DCEBIO-treated cells was expressed as median 6 interquartile range (IQR) and analyzed with Friedman's test with Dunn's post hoc test. The relationship between 86 Rb efflux and extracellular fluid osmolality was analyzed comparing control vs. each experimental osmolality using ANOVA with Turkey Kramer multicomparison post hoc test. Each value was expressed as mean 6 SE. % 86 Rb efflux from multinucleated cytotrophoblasts was expressed as mean 6 SE for each time point. The effects of all treatments on 86 Rb efflux were assessed for statistical significance by comparing the differences in the slopes and intercepts of the rate constants using least squares linear regression analysis. In all cases, a p value less than 0.05 was considered statistically significant.

Expression of IK Ca in Cytotrophoblasts
IK Ca protein expression was confirmed in mono ( Figure 1A) and multinucleated ( Figure 1B, C) cytotrophoblasts using immunofluorescent staining with a specific antibody which detects an extracellular site in the pore forming domain (S5-6) of human IK Ca (K Ca 3.1). IK Ca immunostaining was detected in cells fixed with methanol (intracellular staining; Figures 1A, B) or with PFA (associated with cytotrophoblast surface; Figure 1C).
At 15 h, IK Ca staining (green) was evident in the nucleus (red; nuclear counterstain) of mononucleate cells, but also in the cytoplasm and surface of cell aggregates ( Figure 1A). At 66 h, IK Ca was associated to both the cytoplasm ( Figure 1B) and cell surface ( Figure 1C) of multinucleated cytotrophoblasts. Arrows indicate specific areas were the staining was associated to the cell surface. Figure 1D corresponds to a representative negative control showing that non-specific staining was not observed.
Functional expression of IK Ca was confirmed by measuring 86 Rb efflux, an indirect assessment of K + permeability, in  Figure 1E and F. Basal % 86 Rb efflux in control cytotrophoblasts showed a stable steady state over 13 min ( Figure 1E; black circles). DCEBIO, an IK Ca activator, caused a marked rapid increase (8.3-fold) in 86 Rb efflux which was completely blocked by TRAM-34 (85%), an IK Ca inhibitor ( Figure 1E). Rate constants, taken as the slopes of the regression lines fitted over the experimental period (10 min), were calculated and for all treatments the data could be fitted by a single exponential ( Table 1). The fall in intracellular 86 Rb (slope) was significantly greater with DCEBIO compared to DCEBIO+ TRAM-34 and controls. TRAM-34 had no effect on basal 86 Rb efflux ( Figure 1F). The increase in 86 Rb efflux with DCEBIO confirms the functional expression of IK Ca in multinucleated cytotrophoblasts.

Effect of IK Ca Modulators on Cytotrophoblast hCG Secretion
Compared to controls at 66 h, DCEBIO reduced b-hCG secretion by 80% (19.5 7.1/19.5; Figure 4A). This inhibition of differentiation was not associated with a fall in total cell protein ( Figure 4B), a proxy measure of cell number, suggesting that DCEBIO did not have a generalized toxic effect. On the contrary, DCEBIO caused a transient increase in cell protein at 42 h (148.8 134.8/157.1; Figure 4B). TRAM-34 did not affect cytotrophoblast hCG secretion ( Figure 4A) or total cell protein ( Figure 4B). In addition, the total number of nuclei was unaffected by the treatment with TRAM-34; however, treatment with DCEBIO caused a transient increase in the total number of nuclei at 15 h of culture ( Figure 4C).

Effect of IK Ca Inhibitor on Swelling-activated K + Efflux from Cytotrophoblasts
A role for IK Ca in regulating syncytiotrophoblast volume was explored using multinucleated cytotrophoblasts. We investigated the participation of IK Ca in syncytiotrophoblast RVD by experimentally exposing cytotrophoblasts to a hyposmotic solution and measuring 86 Rb efflux as a marker of syncytiotrophoblast K + permeability. Figure 5A shows the relationship between 86 Rb efflux and extracellular fluid osmolality (ranging from 283-138 mOsm/ kgH 2 O). Total 86 Rb efflux over 10 min (experimental period) was plotted against the reciprocal value for the osmolality of the fluid bathing the cytotrophoblasts after 66 h of culture. A reduction in osmolality to 218 mOsm/kgH 2 O (77% of control), Reducing extracellular osmolality to 183 and 138 mOsm/kgH 2 O (65 and 49% of control respectively) progressively stimulated 86 Rb efflux over control. Consequently, the minimum extracellular osmolality required to trigger 86 Rb efflux from multinucleated cytotrophoblasts is between 77-65% isotonic. Therefore, the remaining experiments were performed using a hyposmotic solution with an osmolality of 145 mOsm/kgH 2 O.
In agreement with previous results in placental villous tissue [57], exposure of multinucleated cytotrophoblasts to a hyposmotic solution markedly increased 86 Rb efflux (3.8-fold; Figure 5B). The rate constant (Table 1) for 86 Rb efflux was significantly greater in cytotrophoblasts exposed to the hyposmotic solution than controls. In addition, swelling-activated 86 Rb efflux was Ca 2+ -dependent, as removal of Ca 2+ from the hyposmotic solution abolished the activation of 86 Rb efflux at 66 h of culture ( Figure 5B; Table 1). Figure 5C shows that swelling-activated 86 Rb efflux was blocked by IK Ca inhibitor, TRAM-34 (85%; Figure 5C). In parallel, rate constant analysis shows a significant difference between hyposmotic solution and TRAM-34, indicating that cytotrophoblast cell swelling activates IK Ca (Table 1). Swelling-activated 86 Rb efflux is mediated specifically by IK Ca as exposure to the SK Ca inhibitor apamin did not affect the stimulated 86 Rb efflux. In contrast, exposing cytotrophoblasts to IK Ca /BK Ca inhibitor ChTx, almost completely inhibited swelling-activated 86 Rb efflux ( Figure 5D; Table 1), suggesting that the regulation of cytotrophoblast cell volume status is through IK Ca .

Discussion
This study shows that IK Ca protein is expressed by mono-and multinucleate cytotrophoblasts in vitro. Multinucleate cells show

IK Ca Expression and Function in Cytotrophoblasts from Term Placentas
Immunofluorescent staining of cytotrophoblasts confirmed the expression of IK Ca protein in mononuclear, aggregated and multinucleated cells. IK Ca staining was associated with the nucleus, cytoplasm and cytotrophoblast cell surface regardless of differentiation stage. Other K + channels, such as K V s [58] and K Ca s [59,60] have been localized to the cell nucleus in various cell types; it has been suggested K Ca s could control Ca 2+ release and mobilization within the cell nucleus [59]. In addition, there is evidence of intracellular localization of K Ca s which may be associated with different cellular functions in non-placental cell types, e.g. in mitochondria [61], intracellular trafficking [62]. Therefore, the heterogeneous localization of IK Ca could be related to diverse functions that these channels might have in cytotrophoblasts during differentiation.
The functional expression of IK Ca was assessed using 86 Rb efflux as a tracer of K + efflux. The results indicate that multinucleated cytotrophoblasts express functional IK Ca as exposure to the IK Ca activator DCEBIO, significantly increased 86 Rb efflux. DCEBIO was specific for IK Ca since this increase in efflux was completely blocked by TRAM-34. However, in a quiescent state IK Ca are inactive as TRAM-34 did not affect basal 86 Rb efflux. This opens the possibility that different stimuli can activate IK Ca in cytotrophoblasts under physiological/pathophysiological conditions but this remains to be determined.

Role of IK Ca in Cytotrophoblast Multinucleation
Cytotrophoblasts isolated from term placentas subjected to trypsin-DNAse digestion and Percoll gradient separation are enriched in trophoblast markers and lack contamination from other placental cell types such as, endothelial cells, smooth muscle cells, fibroblasts, or macrophages [4,5]. After isolation and during the first hours, these cells, which are mitotically inactive, remain mononucleated and secrete small amounts of hCG. After 24 h in culture, they migrate, aggregate and syncytialize by a process of fusion. By 66 h, cytotrophoblasts are predominantly multinucleated syncytial-like cells which secrete high levels of hCG reminiscent of the syncytiotrophoblast in vivo [4,5]. The loss of desmoplakin immunostaining was used to indicate cytotrophoblast fusion and there was, a progressive increase in the formation of multinucleated cytotrophoblasts ($3 nuclei) after 42 h in culture. Cytotrophoblast differentiation was impaired when IK Ca was activated over 42-66 h. DCEBIO did not alter aggregation but inhibited cytotrophoblast morphological and biochemical differentiation in vitro by reducing multinucleation and hCG secretion respectively. These effects were not related to toxicity as total protein was unaffected; indeed, protein levels and the total number of nuclei were higher with DCEBIO at 15-42 h compared to control and this might indicate a transient improved cell viability. Conversely, TRAM-34 treatment did not affect cytotrophoblast syncytialization or hCG secretion. This IK Ca inhibitor was also without effect on 86 Rb efflux indicating little or no IK Ca activity in syncytiotrophoblast under basal conditions.
In non-placental cell types IK Ca is associated with the regulation of processes that contribute to the maintenance of tissue homeostasis including proliferation [31,[41][42][43], differentiation/ fusion [44,45], cell migration [46][47][48] and apoptosis [49]. Particularly, a ChTx (IK Ca inhibitor)-sensitive K + channel activity is necessary for keratinocyte differentiation [44]. Here we showed that pharmacological activation of IK Ca markedly reduced cytotrophoblast syncytialization implying that IK Ca activation inhibits cytotrophoblast-syncytiotrophoblast fusion. In addition, this evidence suggests IK Ca function could change with cytotrophoblast differentiation and therefore chronically activating these channels could lead to abnormal cytotrophoblast-syncytiotrophoblast fusion and dysregulated turnover. A reduced trophoblast fusion leading to altered syncytiotrophoblast turnover has been proposed in pregnancy complications such as pre-eclampsia as fusogenic proteins are downregulated [22,24,27,28]. However, the specific role of IK Ca in this process, and the cellular signals acting in conjunction to co-ordinate trophoblast fusion, need to be addressed in future.
Despite the well-established role for IK Ca s in facilitating cell migration [46][47][48], it is unlikely that they have a similar role in cytotrophoblast migration in vitro as cell aggregation, although not assessed quantitatively, did not appear to be affected by openers/ inhibitors of IK Ca when applied from 3 h after cell isolation.

Role of IK Ca in Syncytiotrophoblast Endocrine Secretion
K + channels participate in endocrine secretion [63][64][65][66] and hCG secretion by placental syncytiotrophoblast is modulated by voltage-gated K + channels (K V ) [51]. hCG is synthesized and secreted by terminally differentiated trophoblast but the mechanism of secretion is still not fully understood. It is evident that hCG secretion is under autocrine/paracrine regulation by hCG which itself promotes cytotrophoblast cell differentiation and further hCG secretion [67]. K V s regulate the secretory process rather than hormone production [51]. Here we showed that the chronic activation of IK Ca significantly reduced hCG secretion by cytotrophoblasts, suggesting that IK Ca could inhibit the mechanism of hCG secretion. However, there is little evidence to link IK Ca function with endocrine secretion, and IK Ca action on hormone secretion is restricted to the central nervous system [68]. We speculate that the primary effect of IK Ca is to inhibit cytotrophoblast cell fusion/terminal differentiation and, as a result, hCG production and secretion is reduced.

Role of IK Ca in Syncytiotrophoblast Volume Regulation
In many cell types, restoration of cell volume in the presence of a hyposmotic stimulus (RVD) is mediated by K + channels, including IK Ca , in conjunction with swelling-activated anion channels [39]. In the current study, exposing multinucleated cytotrophoblasts to a hyposmotic solution increased 86 Rb efflux ,3.8-fold and this activated efflux was dependent on extracellular Ca 2+ , blocked (.80%) by the IK Ca inhibitors TRAM-34 and ChTx but was unaffected by the SK Ca inhibitor apamin. These data implicate IK Ca in cytotrophoblast RVD. Lowering extracellular osmolality also stimulated 86 Rb efflux from placental villous tissue [57] and caused a Ba 2+ -sensitive hyperpolarization of the syncytiotrophoblast microvillous membrane [69]; however, the identities of the K + channels underlying the resting conductance, or the change with cell swelling, remain unknown.
Exposing cells to a hyposmotic solution is an experimental maneuver often used to mimic the cell swelling which takes place secondary to a rise in intracellular osmolality as can occur following nutrient uptake [39,69,70]. In this case, activation of K + channels is a homeostatic process to promote water loss to restore the concentration of cytoplasmic constituents and to shrink cells back to their original size. On the other hand, in the absence of hyposmotic swelling, the activation of K + channels to promote water loss effects a cell volume change, and/or fall in intracellular K + , that is essential for a variety of processes that maintain tissue homeostasis such as cell proliferation, migration, differentiation/ fusion and cell death [71]. It is possible that dynamic changes in cell volume are required for normal cytotrophoblast fusion and that, in the current study, chronically activating IK Ca channels induced an inappropriate change in cell volume which inhibited fusion. Cytotrophoblast fusion may be altered by promoting IK Ca activity and consequently inducing water loss which alters the concentration of cytoplasmic factors that regulate fusion. These proposals need to be investigated in future, and in particular elucidate whether the primary effect of activation of IK Ca is on fusion.

Conclusions
The primary stimuli for IK Ca activation is an elevation in [Ca 2+ ] i and therefore factors that increase [Ca 2+ ] i will activate cytotrophoblast IK Ca . To date there are relatively few studies of the regulation of [Ca 2+ ] i in syncytiotrophoblast; however, preliminary evidence indicates that hyposmotic swelling increases [Ca 2+ ] i in multinucleated cytotrophoblasts, predominantly by entry from extracellular fluid [72].
Consequently, activation of IK Ca could regulate syncytiotrophoblast volume, which may change dynamically following solute uptake and/or cytotrophoblast cell fusion, an essential homeostatic mechanism to maintain nutrient transport and endocrine function respectively. In addition, we have previously shown that cytotrophoblast [Ca 2+ ] i is elevated following activation of purinergic receptors, including P2X4, by extracellular nucleotides and that this promotes 86 Rb efflux which is inhibited by ChTx, implicating activation of IK Ca . These findings might be of relevance to the etiology of pre-eclampsia, a disease of pregnancy characterized by abnormal cytotrophoblast fusion and renewal of syncytiotrophoblast. Indeed, the expression of P2X4 by the placenta is elevated in pre-eclampsia compared to normal pregnancy [73]. It is also proposed that hypoxia/elevated reactive oxygen species release nucleotides from the trophoblast in pre-eclampsia to elevate local concentrations in the extracellular fluid [73,74]. As a result, increased activation of P2X4 would elevate [Ca 2+ ] i and activate IK Ca s. The inappropriate activation of IK Ca could compromise cell volume homeostasis, and impact on cytotrophoblast cell fusion and syncytiotrophoblast renewal, endocrine function and nutrient transport in pre-eclampsia.