| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DEVELOPMENTAL PHYSIOLOGY AND PREGNANCY
1Division of Human Development, St. Mary's Hospital, The Medical School, and 2School of Biological Sciences, University of Manchester, Manchester, United Kingdom; and 3Division of Development Growth and Function, Rowett Research Institute, Bucksburn, Aberdeen, United Kingdom
Submitted 24 August 2005 ; accepted in final form 15 December 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
calcium; potassium; cytotrophoblast
In nonplacental tissues, the extracellular nucleotides ATP and UTP, and their metabolites ADP and UDP, regulate a variety of cell functions, including electrolyte transport, volume homeostasis, cytokine secretion, and apoptosis (3, 6, 8, 16), by activating purinergic receptors and raising intracellular Ca2+ concentration ([Ca2+]i). Purinergic receptors can be divided into two groups based on their molecular structure and signal transduction mechanisms (7). The P2X receptor subgroup comprises seven members (P2X1–P2X7) that are ligand-gated Ca2+-permeable ion channels that open after the binding of an extracellular nucleotide (5). The P2Y receptors are G protein coupled, and agonist binding activates phosphoinositide breakdown to mobilize Ca2+ from intracellular stores (7). Fifteen members of the P2Y family have been reported (P2Y1–P2Y15), although some are only related through weak homology and several (P2Y5, P2Y7, P2Y9, P2Y10) do not function as receptors that raise [Ca2+]i (44).
In nonplacental epithelia, cellular nucleotides are released constitutively (16, 62), but secretion can be enhanced by stimuli such as mechanical or shear stress, cell swelling, hypoxia, inflammation, or platelet activation (21, 31). In the kidney and intestine, secretion of cellular nucleotides raises their concentration in extracellular fluid sufficient to activate purinergic receptors and modulate electrolyte transport (34, 55). Ectonucleotidases on the apical plasma membrane of these tissues hydrolyze the secreted nucleotides and control their local concentrations (64). Although nucleotide secretion has not been measured from the human placenta, it is possible that nucleotides are released constitutively or that secretion is stimulated by mechanical/shear stress generated by the flow of maternal blood that makes direct contact with the syncytiotrophoblast microvillous membrane; consistent with this, ectonucleotidases have been localized to this plasma membrane (9, 37). Furthermore, nucleotide secretion may be enhanced in diseases of pregnancy, such as preeclampsia or intrauterine growth restriction, which are associated with fetal hypoxia, abnormal placental oxygenation, and platelet aggregation (20, 38). It is therefore plausible that extracellular nucleotides may modulate syncytiotrophoblast solute transport physiology and pathophysiology.
Under primary culture conditions, mononuclear cytotrophoblast cells isolated from human term placenta mimic the behavior of cytotrophoblast cells in vivo by fusing and differentiating to form multinucleate syncytiotrophoblast-like cells. Accordingly, these cells have been widely used as an in vitro model of the syncytiotrophoblast and a system in which to examine trophoblast cell differentiation from the mono- to multinucleate phenotype (11, 22). Our group (12) previously demonstrated that extracellular ATP elevates [Ca2+]i and promotes K+ efflux from multinucleate cytotrophoblast cells, although the receptors involved were not identified. In the present study, we tested the hypothesis that multiple purinergic receptor subtypes are expressed by mono- and multinucleate cytotrophoblast cells and that their activation by extracellular nucleotides modulates ion (K+) efflux and [Ca2+]i. We studied the effect of a panel of purinergic receptor agonists on K+ (86Rb) efflux rate and [Ca2+]i and determined the expression of mRNA and protein for P2X and P2Y receptors by cytotrophoblast cells at early (mononucleated) and later (multinucleated) stages of culture.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-methyleneadenosine 5'-triphosphate (-meATP), ADP, ATP, 2',3'-O-(4-benzoyl-benzoyl)adenosine 5'-triphosphate (BzATP), UDP, and UTP were purchased from Sigma-Aldrich (Poole, UK), and fura-2 AM was purchased from Molecular Probes Europe (Leiden, The Netherlands). All other reagents were analytical grade and purchased from standard suppliers.
Cell isolation and culture.
All studies were performed in accordance with the Declaration of Helsinki, and the local ethics committee approved all procedures. Cytotrophoblast cells were isolated using the method described previously in detail (11, 22), modified from the procedure originally described by Kliman et al. (29). Briefly, placentas were collected within 30 min of delivery, and villous tissue was dissected, roughly minced, and washed with 0.9% saline. The tissue was digested in trypsin (Invitrogen, Paisley, UK) with DNase (Sigma-Aldrich) in a HEPES-buffered saline solution, and isolated cells were separated by centrifugation through a discontinuous Percoll gradient. Cytotrophoblast cells were plated at a density of 
4.5 x 106 cells per 35-mm culture dish for efflux experiments or 1.5 x 106 cells per flamed 16-mm diameter glass coverslip for microfluorimetry and maintained in a humidified incubator at 37°C in 5% CO2-95% air for up to 3 days of culture. Investigators in our laboratory (11, 22) have previously shown that during this time, cytotrophoblast cells undergo both morphological and biochemical differentiation to form multinuclear syncytiotrophoblast-like cells. To examine whether the activity and expression of purinergic receptors altered with differentiation, we studied cytotrophoblast cells at two time points in culture: after 18 h, when the majority of cells are mononuclear, and after 66 h, when the majority of cells are multinucleate and resemble the syncytiotrophoblast in situ (11, 22).
86Rb efflux.
86Rb efflux was assessed from cytotrophoblast cells under control conditions and in response to a panel of P2Y and P2X receptor agonists (2-MeS-ADP, 2-MeS-ATP, 5-BrUTP, -meATP, ADP, ATP, BzATP, UDP, and UTP) using the methods previously described (12). Briefly, cytotrophoblast cells were loaded for 2 h with a 3 µCi/ml stock solution of 86Rb in control Tyrode's solution (135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5.6 mM glucose, and 10 mM HEPES, adjusted to pH 7.4 with 10 M NaOH) at room temperature. After loading, cells were washed for 3 min with 50 ml of control Tyrode's solution to remove extracellular isotopes. Fresh solution (1 ml) was then applied and collected at 1-min intervals over a 10-min experimental period. Tyrode's solution containing agonist was applied each minute from 5 min onward. At the end of the experiment, the cells were lysed for 2 h in 0.3 M NaOH. Cell lysates and efflux samples were counted for 86Rb activity with a Packard gamma counter.
Experimental groups. In an initial screening study, each agonist was applied at a single concentration (100 µM) to cells isolated from the same placenta, to eliminate variability between placentas [our group previously found that 100 µM ATP produced maximal 86Rb efflux from cytotrophoblast cells (12)]. From these data, the rate constant of the fall in cellular isotope with time in controls was compared with that in the presence of each agonist on cells from the same isolate. A concentration-response study was subsequently performed for each agonist over the concentration range 1 µM-1 mM.
Contamination of commercial ADP and UDP with ATP and UTP, respectively, was assessed by treating the stock solutions with hexokinase. Treatment with hexokinase is a standard approach that has been used previously to decontaminate commercial preparations of ADP (36) and UDP (39). A 10-mM stock solution of ADP or UDP was prepared in hexokinase buffer (150 mM NaCl, 10 mM HEPES, 2.5 mM KCl, 1 mM MgCl2, 2.5 mM CaCl2, and 22 mM glucose, adjusted to pH 8.0 with NaOH) with 3 U/ml hexokinase (Roche Diagnostics) and was incubated for 1 h at 37°C before use (stock diluted 1:20). This procedure reduces the triphosphate contaminants to insignificant concentrations (36).
Calculation of 86Rb efflux results. The 86Rb efflux rate constants were calculated from plots of ln [Rbi(t)/Rbi(t = 0)] as a function of time, where Rbi(t = 0) are the counts associated with the cells at the start of the time course and Rbi(t) are the counts remaining in the cells at time t.
The time course of 86Rb efflux (%/min) was calculated as (86Rb in 1-min efflux sample cell/86Rb at start of 1-min collection) x 100.
Microfluorimetry.
Cells were loaded with 4 µM fura-2 AM (prepared from a 1 mM stock solution in DMSO) in control Tyrode's solution containing 1% BSA for 30 min at 37°C. After loading, the coverslip was transferred to a perfusion chamber (Warner Instruments, Hamden, CT). The perfusion chamber was mounted on a Nikon Diaphot inverted microscope, and cells were superfused with control Tyrode's solution at 2 ml/min. A field of view was selected, and six cell-containing regions were defined for analysis of [Ca2+]i. Mononuclear cytotrophoblast cells were chosen for experiments at 18 h of culture, and multinucleated areas (3 or more nuclei per cell) were selected at 66 h.
After a 2-min equilibration period, the cells were stimulated by the addition of ATP, BzATP, 2-MeS-ATP, -meATP, ADP, UDP, UTP, or 5-BrUTP for 2 min. The concentration of agonist used was determined (empirically) from the 86Rb efflux experiments as that inducing the maximum stimulation of 86Rb efflux. For those agonists that are thought to act at P2Y receptors (ADP, UDP, UTP, 5-BrUTP) along with ATP and BzATP (P2Y and P2X agonists), the effect of the agonist on [Ca2+]i also was examined in Ca2+-free conditions; an increase in [Ca2+]i in response to agonists in the absence of extracellular Ca2+ can be attributed to mobilization of Ca2+ from intracellular stores following the activation of P2Y receptors. Experiments were repeated in Ca2+-free Tyrode's solution (as described in 86Rb efflux but with no added CaCl2 and containing 0.5 mM EGTA), which was present for 6 min before agonist application, for the duration of agonist exposure (2 min), and for a further minute after agonist removal. All experiments were carried out at room temperature.
The fluorescence imaging system was controlled by a personal computer running the Kinetic Imaging (Nottingham, UK) AQM Advance software package, driving an Optoscan monochromator (Cairn Research, Faversham, UK) and a Hamamatsu cooled charge-coupled device camera (model OrcaER) to allow image acquisition at 340 and 380 nm. Ratio images were constructed off-line, and six cells in each image were selected for analysis. The 340- to 380-nm fluorescence ratio was calculated as a measure of [Ca2+]i. Two time points were selected for analysis of the agonist-induced effect on [Ca2+]i: the time of maximal increase in ratio (peak) and 2 min after the peak, which has been referred to as the plateau. The control ratio was measured in the 2 min preceding agonist application. In a few experiments with -meATP and 2-MeS-ATP, a well-defined peak increase in [Ca2+]i was not observed in mononucleated cells. In these experiments, the ratio was measured 10 s after addition of the agonist, because this corresponded to the average time that a peak increase was detected in responsive cells.
RT-PCR. Gene-specific primers for the purinergic receptor subtypes P2X1, P2X2, P2X3, P2X4, P2X5, P2X7, P2Y1, P2Y2, P2Y4, P2Y6, and P2Y11 were designed for use in RT-PCR. Primers of 20 base pairs (bp) in length were selected to amplify partial cDNA from each sequence (Table 1). Total RNA, prepared from cytotrophoblast cells at 18 and 66 h of culture (n = 3 placental preparations), was reverse transcribed (Invitrogen) for use in PCR by using a Thermocycler (Perkin Elmer). For PCR amplification, samples were denatured at 95°C for 2.5 min, followed by a protocol consisting of 35 cycles of 1 min at 95°C, 1 min of annealing at Tm, where Tm is the primer-specific annealing temperature (Table 1), and 1 min of extension at 72°C. A further 5 min at 72°C ensured cDNA extension. PCR products were separated on a 1.2% agarose gel containing ethidium bromide for visualization over UV light. Identity of PCR products was determined by restriction enzyme digest and sequencing. Negative controls omitted cDNA; positive control amplifications used cDNA derived from tissues known to express the receptor subtype of interest.
|
-actin) with PBS containing 5% milk protein (Marvel) and 0.025% sodium azide.
Nitrocellulose membranes were incubated with affinity-purified polyclonal antibodies to P2X4 (1:100), P2X7, P2Y2, and P2Y4 (1:50; all antibodies from Alomone Laboratories, Jerusalem, Israel) for 2 h at room temperature in PBS containing 5% milk protein and 0.025% sodium azide. Membranes also were incubated with a mouse monoclonal antibody to -actin (1:2,500; Sigma-Aldrich) for 1 h at room temperature in Tris-buffered saline (TBS)-0.1% Tween (polyoxyethylene sorbitan monolaurate; Sigma-Aldrich) containing 5% milk protein. To confirm specificity, we also incubated membranes in the absence of primary antibody (negative control) or after antibody preabsorption with the specific blocking (immunizing) peptide (Alomone Laboratories). For preabsorption, P2X4, P2X7, P2Y2, P2Y4, and P2Y6 antibodies were incubated with 1 µg of peptide per 1 µg of antibody for 1 h at room temperature before incubation with the membrane as described above.
After primary antibody incubation, the membranes were incubated with either horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (Sigma-Aldrich; purinergic receptors and negative control) diluted 1:8,000 in PBS containing 5% milk protein or horseradish peroxidase-conjugated sheep anti-mouse secondary antibody (Sigma-Aldrich; -actin) diluted 1:1,000 in TBS-0.1% Tween containing 5% milk protein. Signals were detected with an enhanced chemiluminescence detection system (ECL; Amersham Biosciences) and detected on photosensitive film (Hyperfilm-ECL; Amersham Biosciences).
Data analysis and statistics. The data are expressed as means ± SE, with n representing the number of placentas from which cells were isolated. Least-squares linear regression was used to analyze the decrease in cellular 86Rb with time {ln [Rbi(t)/Rbi(t = 0)]; see Calculation of 86Rb efflux results}, and the rate constants for 86Rb efflux were taken as the negative slope of the regression line fitted over the experimental period (6–10 min). To analyze the agonist concentration-response data, we combined the %86Rb efflux/min for minutes 6 and 7, i.e., the first 2 min of agonist application (see Fig. 1) (corresponding to the period of agonist application in microfluorimetry studies), to give %86Rb efflux/2 min and plotted the data as percentage increase above control. Statistical analyses of the differences in 86Rb efflux in control conditions and in response to agonists were performed on the raw data by using repeated-measures ANOVA with a Dunnett's multicomparison post hoc test.
|
The effects of 2-MeS-ATP and -meATP on [Ca2+]i were assessed using a paired Student's t-test to compare the percentage change in 340/380 ratio at the peak and plateau response. For the remaining agonists, peak and plateau were compared in both Ca2+-containing and Ca2+-free conditions by using repeated-measures ANOVA and a Bonferroni post hoc test comparing all groups. The changes in [Ca2+]i in response to agonists in cytotrophoblast cells at 18 and 66 h of culture were compared using an unpaired Student's t-test. P values <0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-meATP at 100 µM were initially examined on cells isolated from the same placenta (n = 3 placentas). The rate constants, calculated as the slopes of the least-squares linear regression lines fitted over the period of agonist application and corresponding control period (6–10 min), are given in Table 2. Under both control and experimental conditions, a single exponential gave a statistically significant fit, indicating that 86Rb efflux from mono- and multinucleated cytotrophoblast cells in response to each agonist is predominantly through a single exit pathway or, less likely, several pathways with identical rate constants.
|
|
|
|
UTP had a potent effect on mononucleated cytotrophoblast cells, inducing a significant increase in 86Rb efflux at all concentrations from 1 to 500 µM, but in multinucleated cells, only 500 µM UTP elevated 86Rb efflux significantly above control (Fig. 2D). UDP had no significant effect on 86Rb efflux at early stages of culture but raised efflux in differentiated cells at 100 and 500 µM (Fig. 2E). Neither 2-MeS-ATP nor 2-MeS-ADP had any effect on 86Rb efflux (Fig. 3, A and B). Finally, -meATP did not alter efflux from cells at 18 h, but a small response was induced at 500 µM after 66 h of culture (Fig. 3C).
Contamination of commercial ADP and UDP sources. Previous reports have suggested that nucleotide diphosphates, such as ADP and UDP, may be contaminated with their triphosphate counterparts and that this contamination may provide a false-positive result. Thus, to determine whether the responses induced by ADP and UDP were specific or attributable to contamination with ATP and UTP, respectively, we treated ADP and UDP stocks with hexokinase before use in 86Rb efflux experiments. Hexokinase breaks down any contaminating triphosphates with no effect on the diphosphate content. At early and later stages of culture, the increase in 86Rb efflux with 500 µM ADP was slightly, but significantly, reduced after hexokinase treatment (Fig. 4, A and B). For mononucleated cells, the reduction in total efflux was 22% (measured over the 4 min of agonist treatment), whereas for multinucleated cells, the reduction was 30%. The increase in 86Rb efflux from cytotrophoblast cells in response to 500 µM UDP was unaffected by hexokinase treatment (Fig. 4, C and D).
Agonist stimulated [Ca2+]i.
ATP (100 µM), BzATP (500 µM), 2-MeS-ATP (500 µM), -meATP (1 mM), ADP (500 µM), UDP (500 µM), UTP (10 µM), and 5-BrUTP (500 µM) induced a peak increase in [Ca2+]i, followed by a plateau. For all agonists, the 340/380 ratio values at the peak and plateau were significantly higher than control in the presence and absence of extracellular Ca2+ (P < 0.008; Wilcoxon matched-pairs test; data not shown).
Figures 5 and 6 show the percent change in [Ca2+]i above control after a 2-min application of purinergic receptor agonists. All agonists induced a peak increase in [Ca2+]i that was significantly higher than the plateau in mono- and multinucleated cells in both the presence and absence of extracellular Ca2+. The only exception was the response to BzATP in multinucleated (66 h) cells, where the initial peak increase in [Ca2+]i was maintained during the plateau phase in the presence, but not the absence, of extracellular Ca2+ (Fig. 6B, peak +Ca vs. plateau +Ca not significant), consistent with a sustained entry of Ca2+ through P2X receptors.
|
|
-meATP and 2-MeS-ATP induced a significant increase in [Ca2+]i as indicated by Wilcoxon matched-pairs analysis. However, the overall response to these agonists was relatively small (Figs. 5, C and D, and 6, C and D; note the difference in scale on the y-axis), and a peak increase in [Ca2+]i was not induced by -meATP and 2-MeS-ATP in all cells (see example traces, Figs. 5D and 6C). In mononucleated cells, the increase in [Ca2+]i in response to ATP and ADP (Fig. 5, A and E) was unaffected by Ca2+-free conditions, showing that the response was dominated by intracellular Ca2+ release and thus implicating activation of P2Y receptors. However, both the peak and plateau rise in [Ca2+]i following BzATP, UDP, UTP, and 5-BrUTP (Fig. 5, B, F, G, and H) were significantly reduced in 0 Ca2+, suggesting that these agonists can activate both P2X and P2Y receptors.
In contrast to mononucleated cells, the peak and plateau increases in [Ca2+]i evoked by ATP and ADP in multinucleated cells were significantly reduced in 0 Ca2+ (Fig. 6, A and E), consistent with their stimulating both P2X and P2Y receptors (peak in 0 Ca2+ with ADP mono- vs. multinucleated cells, P < 0.001). The responses to UDP and UTP (Fig. 6, F and G) also differed with stage of cell culture: UTP elicited a larger peak increase in [Ca2+]i in multinucleated cells (P < 0.05 vs. mononucleated), and the peak and plateau [Ca2+]i with UDP and UTP were unaffected by 0 Ca2+ in multinucleated cells. This suggests a predominant activation of P2Y receptors by these agonists as cytotrophoblast cells differentiate (peak in 0 Ca2+ with UTP mono- vs. multinucleated cells, P < 0.001). Finally, in contrast to mononucleated cells, the peak increase in [Ca2+]i with BzATP was unaffected by 0 Ca2+ in multinucleated cells (Fig. 6B) (activation of P2Y), and in the presence of extracellular Ca2+, the peak and plateau [Ca2+]i did not differ, possibly indicating activation of P2X receptors. However, the changes in [Ca2+]i in response to 5-BrUTP were similar at both stages of differentiation (Figs. 5H and 6H).
In general, the increase in [Ca2+]i induced by the agonists (Figs. 5 and 6) was consistent with their ability to promote 86Rb efflux (Figs. 2 and 3). Accordingly, -meATP and 2-MeS-ATP, which had weak effects on [Ca2+]i, failed to stimulate 86Rb efflux, and for the other agonists, the rise in [Ca2+]i was proportionally similar to the stimulation of 86Rb efflux.
Expression of purinergic receptor mRNA. PCR products of the expected size were obtained for P2X1, P2X2, P2X4, and P2X7 and for P2Y1, P2Y2, P2Y4, P2Y6, and P2Y11 in cytotrophoblast cells at both stages of differentiation (Figs. 7 and 8). P2X3 and P2X5 mRNA could not be detected (Fig. 7). In all cases, product identification was confirmed by restriction enzyme digests and/or sequencing (data not shown).
|
|
-Actin was detected in each cytotrophoblast cell sample, confirming that the protein samples were intact, and representative examples are shown in Fig. 9A.
|
65 kDa in cells at both 18 and 66 h of culture (n = 3). All bands disappeared after preincubation with the antigenic peptide (Fig. 9B). Two products were identified after probing with the P2X7 antibody: in rat brain, a band was detected at 76 kDa (n = 5), whereas in human brain and the cytotrophoblast cells, the product was 145 kDa (n = 3). Again, all bands disappeared after preabsorption with the P2X7-specific antigen (Fig. 9C). Use of the antibody to P2Y2 revealed a product in rat brain at a size of 56 kDa, whereas in cytotrophoblast cells, the product was 65 kDa (Fig. 9D). Despite the discrepancy between product sizes, all bands disappeared after preabsorption with the P2Y2-specific antigen. The specific band for P2Y2 (65 kDa) was observed in four of six mononucleated, and in four of six multinucleated, cytotrophoblast cell preparations. Use of the antibody to P2Y4 revealed a band of the expected size of 91 kDa in two positive control tissues, rat (n = 4) and human brain (n = 2) (Fig. 9E). P2Y4 was not detected in cytotrophoblast cells at either 18 (n = 2) or 66 h of culture (n = 2). Use of the antibody to P2Y6 revealed weak bands at the expected size of 110 kDa in the positive control tissue, rat brain, and in cytotrophoblast cell samples at both 18 (n = 3) and 66 h of culture (n = 3) (Fig. 9F). All negative control blots were clear with no bands detected (data not shown). |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Work in our laboratory (12) previously showed that K+ efflux from cytotrophoblast cells promoted by ATP was completely inhibited by charybdotoxin but unaffected by iberiotoxin or apamin, implicating intermediate-conductance Ca2+-activated K+ channels in the efflux pathway. In the present study, the agonists that stimulated K+ efflux also raised [Ca2+]i over the same time course, and it is probable that Ca2+-activated K+ channels mediated the raised K+ efflux from cytotrophoblast cells after activation of purinergic receptors.
Cytotrophoblast cell differentiation and purinergic receptors. The basal rate of K+ efflux was greater in mono- than in multinucleated cells. This decrease in passive K+ permeability/conductance is consistent with the depolarization of membrane potential difference that is associated with cytotrophoblast cell differentiation (22). Because of this difference in basal efflux, the percent increase in K+ loss promoted by purinergic receptor agonists was greater in the more differentiated cells. However, the concentration dependence and time course of K+ efflux (data not shown) in response to agonists did not differ markedly with the stage of culture. In accord with this, we did not observe any substantial differences in the expression of mRNA or protein for purinergic receptors (P2X4, P2X7, P2Y2, or P2Y6) by cytotrophoblast cells at early and later stages of culture.
In general, the increase in [Ca2+]i induced by the agonists was consistent with their ability to promote 86Rb efflux. However, there were some differences in the effects of agonists on [Ca2+]i between mono- and multinucleated cytotrophoblast cells. For example, UTP induced a greater rise in [Ca2+]i in multi- than in mononucleated cells, whereas BzATP activated both P2Y and P2X receptors with a more marked effect on P2X in differentiated cells. ATP and ADP activation of P2Y receptors was greater in mono- than in multinucleated cells, whereas the opposite was true for UTP and UDP, where P2Y receptor activation was more readily apparent in differentiated cells. These differences might reflect a difference in the functional expression of P2X and P2Y receptor subtypes with differentiation to the multinucleate phenotype, although it also is possible that downstream signaling events are affected by differentiation. However, relatively high concentrations of agonists, some with poor selectivity, had to be used in these experiments to reveal functional receptors, and the significance of the differences observed in relation to activation of the receptor subtypes by endogenous nucleotides remains to be determined.
Evidence for P2X receptors in cytotrophoblast cells.
Extracellular ATP activates all P2X subtypes but is a potent activator of P2X4 (40). In the present study, we have confirmed previous findings of a potent effect of ATP on [Ca2+]i and K+ efflux from cytotrophoblast cells (12), and in addition, we have demonstrated mRNA and protein expression for P2X4. The effect of ATP at lower concentrations is consistent with activation of P2X4, although other receptors may be stimulated at higher concentrations. -MeATP is selective for P2X1 and P2X3 (2, 10) and is a partial agonist at P2X4 receptors at high concentrations. The lack of effect of -meATP at concentrations <500 µM suggests that P2X1 and P2X3 are not active in cytotrophoblast cells, and indeed, mRNA for P2X3 was not detected. Our PCR data for P2X5 confirm a previous report (33) that this receptor subtype is not expressed in trophoblasts.
2-MeS-ATP is a partial agonist for P2X2 and has been reported to activate P2X4 at higher concentrations (23). In this study, we demonstrated a small, yet significant effect of 500 µM 2-MeS-ATP on [Ca2+]i, also supporting a role for P2X4, but not P2X2, in modulating trophoblast function. In the future, it would be useful to examine the effect of zinc, which is an agonist at P2X4 receptors (1) but an antagonist for P2X7 (60).
BzATP was initially reported to be selective for P2X7 (45), but it is now evident that it also activates P2Y1 and P2Y11 receptors (40). BzATP raised [Ca2+]i and stimulated K+ efflux from cytotrophoblast cells, but only at relatively high concentrations (100 µM and higher). This may reflect activation of receptors other than P2X7, because the latter is typically activated by low concentrations of BzATP [e.g., 15 µM (19)], and the EC50 value for ATP is 10-fold greater than that for BzATP at P2X7 (45). Indeed, in cytotrophoblast cells BzATP elevated [Ca2+]i in the absence of extracellular Ca2+, consistent with activation of P2Y receptors and the subsequent release of Ca2+ from intracellular stores. However, the peak increase in [Ca2+]i induced in the presence of Ca2+ was significantly greater than the peak in Ca2+-free conditions, which suggests additional activation of P2X receptors by BzATP. These effects of BzATP are consistent with the findings of Divald et al. (17) that BzATP signals through P2X7 to activate phospholipase D in human placental cytotrophoblast cells.
P2X7 products were detected at two different sizes in the Western blots, with a band in rat brain at the expected size of 70 kDa and a band at 145 kDa in the cytotrophoblast cell samples and human brain. Both the 70- and 145-kDa bands disappeared when the P2X7 primary antibody was preabsorbed with the specific antigen. Because the cell product was approximately twice the expected size, it is likely that the P2X7 dimer is expressed in cytotrophoblast cells. P2X7 can form multimers, and differential assembly of P2X7 subunits has been demonstrated (28), although the functional implications are unknown. However, both the P2X7 mRNA and protein expression appeared weak, although the results were not quantified precisely, suggesting that this receptor is not abundantly expressed in the plasma membrane of cytotrophoblast cells.
Evidence for P2Y1 and P2Y11. 2-MeS-ADP activates P2Y1 (30), but this agonist did not affect 86Rb efflux from cytotrophoblast cells at concentrations up to 1 mM. ADP is a potent P2Y1 agonist in nonplacental tissue (25, 61), and the large response to ADP in cytotrophoblast cells was greater than anticipated considering the published potencies and the lack of response to 2-MeS-ADP. We initially examined the possibility that the source of ADP was contaminated with ATP, using the method of degradation of triphosphate contaminants by incubation with hexokinase. Although we did not measure ATP levels in the present study, hexokinase treatment is reported to reduce such contamination to insignificant levels (36). Treatment of 500 µM ADP with hexokinase significantly reduced 86Rb efflux compared with untreated ADP, indicating that commercial sources of ADP are contaminated with ATP. Nonetheless, there remained a substantial elevation of 86Rb efflux after treatment that may be attributable to ADP activation of P2Y1. However, the possibility that ADP is acting at a different and as-yet unidentified receptor cannot be dismissed.
Obtaining functional evidence for P2Y11 is hampered by the lack of selective agonists for this receptor subtype. As mentioned previously, there is evidence that BzATP can activate both P2Y1 and P2Y11 receptors (40). Because the increase in [Ca2+]i induced by BzATP in cytotrophoblast cells had both an extracellular (Ca2+ entry) and intracellular (Ca2+ release) component, it is plausible that P2Y1 and/or P2Y11 are functionally expressed in trophoblast cells. P2Y1 mRNA has shown to be highly expressed by human placental tissue (35), and human P2Y11 was first isolated after screening of a placental cDNA library (13). We demonstrated the expression of mRNA for both P2Y1 and P2Y11 in cytotrophoblast cells, but because no commercial antibodies to these receptors were available at the time of the study, further investigation is needed to determine whether P2Y1 or P2Y11 are functionally active in the human placenta.
Evidence for P2Y2/P2Y4/P2Y6. The extracellular uridine nucleotides UTP and UDP are selective for the uridine nucleotide-preferring receptor subtypes P2Y2, P2Y4, and P2Y6. UTP activates P2Y2 and P2Y4, whereas UDP and the synthetic UTP analog 5-BrUTP are both selective for P2Y6 (39). UTP had a potent effect on cytotrophoblast cells, stimulating 86Rb efflux maximally at 10 µM, which suggests a functional role for P2Y2 or P2Y4. Similarly, 10 µM UTP induced a significant rise in [Ca2+]i that was still evident in the absence of extracellular Ca2+, consistent with activation of a P2Y receptor. A previous study by Petit and Belisle (43) also implicated a functional role for P2Y2 in cultured cytotrophoblast cells. Typically, UTP and ATP act at this receptor in an equipotent manner (39, 43). Although direct comparison between the effects of agonists is difficult in cells that express multiple receptors, both UTP and ATP were potent at low concentrations in cytotrophoblast cells. Addition of 1 µM UTP induced a significantly elevated 86Rb efflux from mononucleated cytotrophoblast cells (Fig. 2D), whereas 10 µM ATP stimulated efflux from both mono- and multinucleated cells (Figs. 2A and 3A). Furthermore, our group (12) previously reported that 1 µM ATP can increase [Ca2+]i in cytotrophoblast cells. RT-PCR revealed expression of mRNA for both P2Y2 and P2Y4. Western blotting was unable to detect P2Y4 in the cytotrophoblast cell samples, although this may reflect poor sequence homology between rat and human P2Y4 in the region to which the antibody is directed. These data together suggest that the UTP can regulate both [Ca2+]i and K+ efflux in cytotrophoblast cells via activation of the P2Y2 receptor, although P2Y4 cannot yet be ruled out.
UDP and 5-BrUTP, specific agonists for the P2Y6 subtype (39), both induced a significant elevation of [Ca2+]i and a relatively small elevation in 86Rb efflux from mono- and multinucleated cytotrophoblast cells. Human P2Y4 and P2Y6 were first isolated after screening of a placental cDNA library (14, 15), and Somers et al. (58), using Northern hybridization, reported high levels of P2Y6 mRNA in the syncytiotrophoblast and cytotrophoblast. In this study, we have demonstrated mRNA and protein expression for P2Y6 in cytotrophoblast cells at early and later stages of differentiation. Overall, our data suggest that nucleotide activation of P2Y6 can modulate cytotrophoblast cell ion transport physiology.
Role of purinergic receptors in syncytiotrophoblast of human placenta.
This study has demonstrated effects of ATP, UTP, ADP, and UDP on cytotrophoblast cell physiology. Although the function of purinergic receptors in the syncytiotrophoblast remains to be confirmed, we have preliminary data demonstrating the expression of P2X4, P2X7, and P2Y2 protein in both first-trimester and term placental membrane fraction homogenates (unpublished observations), and P2X4 is immunolocalized to the syncytiotrophoblast of term placenta (47). However, nothing is known of the release of nucleotides from placental cells or of their concentration in placental extracellular fluid. In nonplacental epithelia, it is thought that ATP concentrations at the site of release can reach 1 mM, whereas circulating amounts are <1 µM (21). If the local extracellular ATP concentrations in placenta are similar, and the expression and/or sensitivity of purinergic receptor subtypes in the syncytiotrophoblast resembles those we identified in cytotrophoblast cells, then constitutive nucleotide release may be sufficient to activate trophoblast purinergic receptors with high affinity for ATP (P2X4/P2Y2). Constitutive release of UTP is reported to achieve an extracellular concentration of 1 µM in nonplacental tissue (32), and we have demonstrated an effect of UTP on [Ca2+]i and 86Rb efflux from cytotrophoblast cells at this concentration, probably mediated by P2Y2 receptors.
Although we have demonstrated increased K+ loss from cytotrophoblast cells following the elevation of [Ca2+]i in response to purinergic receptor activation, it is unlikely that altered K+ efflux is the only consequence of a transient increase in trophoblast [Ca2+]i. For example, placental trophoblast cells express Ca2+-activated Cl– channels (27, 59), and Cl– efflux could accompany K+ loss following purinergic receptor activation; this would also be expected to result in cell shrinkage. A role for ATP and UTP (via P2X and P2Y receptors) in modifying Cl– secretion has been shown in various epithelial cell types (56, 57).
In addition to constitutive release, ATP is secreted into the extracellular environment under circumstances of cell stress/damage (21, 55) leading to prolonged elevation of extracellular ATP and/or higher concentrations of this nucleotide. In brain, exposure to ATP released after ischemic damage can induce P2X7 pore formation, which promotes Ca2+ entry and K+ loss and lowers intracellular K+ concentration. These events are associated with necrosis/apoptosis and enhanced cytokine release (63). In this regard, it is of interest that our laboratory (54) recently demonstrated an increase in IL-6 production by extracellular ATP from placental villous fragments in the presence of a proinflammatory stimulus. Furthermore, in preeclampsia, a disease of pregnancy associated with elevated cytokines and conditions known to stimulate nucleotide release from cells (abnormal placental blood flow, ischemia, platelet activation; Refs. 20, 38), there is altered trophoblast cell turnover and increased apoptosis (24). It is also intriguing that maxi-anion channels, candidate proteins for mediating ATP release from cardiomyocytes under ischemic or hypoxic conditions (18), are expressed in the microvillous membrane of the syncytiotrophoblast (4), and their open probability is reported to increase in preeclampsia (46).
In the absence of specific agonists, it is difficult to definitively identify functional receptors in a tissue such as the placenta with multireceptor expression. However, together, our data indicate a functional role for P2X7, P2X4, P2Y2, and P2Y6 in raising [Ca2+]i and promoting K+ loss from cytotrophoblast cells. The significance of these purinergic receptors in the control of electrolyte transport, trophoblast apoptosis, or cytokine release in normal and diseased pregnancy awaits the measurement of extracellular nucleotides in the placenta and the localization of receptor proteins to the plasma membranes of the syncytiotrophoblast.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Present address of V. H. J. Roberts: Department of Obstetrics and Gynecology, Medical Sciences Building, University of Cincinnati, OH 45267-0526.
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
and IL-1 from murine macrophages. J Immunol 170: 3029–3036, 2003.
, Control (basal) efflux; 
, efflux in response to untreated agonist; and 
, efflux in response to hexokinase-treated agonist. Values are means ± SE, n = 3 placentas. *P < 0.05; **P < 0.01, ADP vs. hexokinase-treated ADP (ANOVA with Bonferroni's post hoc test).
