The Na+/H+ exchanger (NHE) has a key role in intracellular pH ([pH]i) regulation of the syncytiotrophoblast in the human placenta and may have a role in the life cycle of this cell. In other cells the NHE (actually a family of up to 9 isoforms) is regulated by a variety of factors, but its regulation in the syncytiotrophoblast has not been studied. Here, we tested the hypotheses that EGF and sphingosine-1-phosphate (S1P), both of which affect trophoblast apoptosis and, in other cell types, NHE activity, stimulate syncytiotrophoblast NHE activity. Villous fragments from term human placentas were loaded with the pH-sensitive dye, BCECF. NHE activity was measured by following the recovery of syncytiotrophoblast [pH]i following an imposed acid load, in the presence and absence of EGF, S1P, and specific inhibitors of NHE activity. Both EGF and S1P caused a dose-dependent upregulation of NHE activity in the syncytiotrophoblast. These effects were blocked by amiloride 500 μM (a nonspecific NHE blocker) and HOE694 100 μM (NHE blocker with NHE1 and 2 isoform selectivity). Effects of EGF were also reduced by the NHE3 selective blocker S3226 (used at 1 μM). These data provide the first evidence that both EGF and S1P stimulate NHE activity in the syncytiotrophoblast; they appear to do so predominantly by activating the NHE1 isoform.
- epidermal growth factor
- Na+/H+ exchanger
the exchange barrier between maternal and fetal circulations in the human placenta consists, essentially, of two cellular layers, the syncytiotrophoblast, and the fetal capillary endothelium. The syncytiotrophoblast, a multinucleated true syncytium, is the transporting epithelium of the placenta. Transport activity of the syncytiotrophoblast is essential for a supply of a range of solutes required for fetal growth, as well as homeostasis of the cell itself (30). Accumulated evidence suggests that in abnormal pregnancies complicated by intrauterine growth restriction (IUGR), where fetal growth is less than expected from its genetic potential, the activity of a variety of syncytiotrophoblast transporters is reduced or abnormal (31). One of the transporters affected by IUGR is the Na+/H+ exchanger (NHE) in the microvillous (maternal facing) plasma membrane (MVM) of the syncytiotrophoblast (16).
NHEs are ubiquitously found in cells and are involved in a variety of processes, including regulation of intracellular pH ([pH]i) and cell volume (25); as a result, they can affect a large array of cellular functions, including apoptosis (42). There is now good evidence that NHEs are very important for [pH]i regulation of the syncytiotrophoblast in the human placenta (8, 33) and, from work in the rat, transplacental Na+ transfer (1). There are nine members of the NHE protein family (25), and at least three NHE isoforms have been identified in the human placenta, localized to both the microvillous (MVM, maternal facing) and basal (fetal facing) plasma membranes (BM) of the syncytiotrophoblast. Reported NHE isoform distribution to these membranes varies between studies (14, 16, 26, 33), but we have shown by immunoblotting that NHE 1 protein is localized to the MVM and BM, while NHE 3 is only found on the MVM (33). However, NHE 1 activity predominates on MVM, with no evidence of NHE 3 activity under our normal conditions of experimentation (33).
NHE activity has been shown to be regulated by a range of molecules and conditions in other cell types: generally, NHE3 differs from NHE1 and NHE2 in that regulation is by alteration in Vmax due to increased cycling of the receptor from the endoplasmic reticulum rather than increased intracellular affinity for H+ (6). It is not clear how NHE activity is regulated in the placenta; however, we have shown previously that in cultured trophoblast cells, both EGF and the signaling lipid sphingosine-1-phosphate (S1P) inhibit apoptosis (18). As activation of NHEs is also known to inhibit apoptosis, and EGF is a regulator of NHE activity in other cell types (9, 20, 21, 43), we hypothesized that EGF and S1P may upregulate NHE activity in the syncytiotrophoblast.
The actions of the antiapoptotic signaling lipid S1P are less clearly established than those of EGF; however, it appears that most if not all the actions of S1P are the result of its interaction with G-protein linked S1P receptors (13, 28, 34). There is only one previous publication on the effect of S1P on the NHE family. Tornquist (38) demonstrated that S1P acutely activated NHE1, in a dose-responsive, pertussis-dependent fashion in modified thyroid cells. This is supported by another publication from the same group showing a similar response to another similar bioactive antiapoptotic sphingolipid, sphingosylphosphorylcholine (39).
In this study, we tested our hypothesis by examining the effect of EGF and S1P on the ability of NHEs to respond to acute acidification in term villous trophoblast. We find that both have the ability to increase the NHE-dependent rate of [pH]i recovery from an acid load, in a dose-dependent manner. We also demonstrate that increased NHE activity is primarily the result of increased NHE1 activation.
A NHE blocker with NHE1 and 2 isoform selectivity (HOE 694) and the NHE3 selective blocker S3226 were kindly donated by Aventis (Aventis Pharma, Frankfurt/Main, Germany). Ethanol, magnesium chloride, methanol, potassium chloride, potassium hydroxide, sodium chloride, and sodium hydroxide were purchased from Merck (Poole, Dorset, UK). Amiloride, ammonium chloride, carbenoxolone, choline (chloride), DMSO, glucose, hydrocortisone, 3-N-(morpholino) propane-sulfonic acid (MOPS), nigericin, and poly-l-lysine were purchased from Sigma (Poole, Dorset, UK). 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (BCECF AM; Molecular Probes, Cambridge Bioscience, Cambridgeshire, UK). EGF was purchased from Calbiochem (Merck Biosciences, Nottingham, UK). S1P was purchased from Biomol (Exeter, UK).
Placental tissue was obtained with informed consent and the approval of the local research ethics committee, from patients admitted to the Central Delivery Unit at St. Mary's Hospital, Manchester. Placentas were collected at term, following vaginal delivery or caeserean section (due to breech or for previous section), from women who had uncomplicated pregnancies.
Tissue Handling and Preparation and Visualization
Samples of placental villous tissue were taken from midway between the chorionic and basal plate of the placenta and finely chopped to produce small villous fragments, as described previously (33) and stored at 37°C in Tyrode's buffer (pH 7.4) until experimentation for no longer than 4 h post partum (32).
Fragments were immobilized, loaded with the pH-sensitive dye, 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (BCECF AM), and NHE activity was measured as the recovery from an acid load, as described previously (33). Briefly, fragments were immobilized onto poly l-lysine (Sigma)-coated glass cover slips and loaded with 1 μM BCECF at 37°C, pH 7.4 for 5 min in control Tyrode's buffer (containing in mM: 135 NaCl, 5 KCl, 1.8 CaCl2, 1 MgCl2, 10 MOPS, and 5.6 glucose). BCECF-loaded villous fragments were washed with BCECF-free Tyrode's buffer for 5 min before experimentation. The single placental villi were visualized using a Nikon TE300 inverted microscope and excited using a light from a xenon arc lamp passed through an Optoscan Monochromator at 440 nm and 490 nm (Cairn Research, Faversham, Kent, UK), and images for analysis were acquired every 5 s using a C-Coolsnap HQ Cooled digital charge-coupled device camera (Roper Photometrics via Cairn Research). Emission data at 530 nm from five selected areas of syncytiotrophoblast around a single terminal villous were acquired using MetaFlour software (Universal Imaging, Downington, PA, USA) and visualized through Excel (Microsoft, Redmond, WA, USA) and PRISM graphics packages (ver. 3.0, GraphPad Software, San Diego, CA).
Measurement of Intracellular pH [pH]i in Human Placental Syncytiotrophoblast
Calibration of the BCECF fluorescence ratio (490:440 nm) to obtain [pH]i was performed using a high K+/nigericin (10 μM) method, as described previously (33, 36). The activity of NHE was measured in intact syncytiotrophoblast by monitoring the recovery from an acid load imposed by an ammonium chloride (20 mM) pulse; this is both Na+ dependent and blockable by amiloride (33), as expected if it is predominantly due to the activity of the exchanger. Experiments were performed using control Tyrode's buffer, ammonium chloride Tyrode's buffer (containing in mM: 115 NaCl, 20 NH4Cl, 5 KCl, 1.8 CaCl2, 1 MgCl2, 10 MOPS, 5.6 glucose), and Na+-free Tyrode's buffer (containing in mM: 135 CholineCl, 5 KCl, 1.8 CaCl2, 1 MgCl2, 10 MOPS, 5.6 glucose) at 37°C and pH 7.4. The experimental protocol involved a 5-min ammonium chloride pulse, followed by a 3-min washout of ammonium chloride in Na+-free Tyrode's buffer, followed by a 5-min recovery phase in the presence of Na+ (control Tyrode's buffer). EGF (1–100 ng/ml), S1P (10–10,000 nM dissolved in NaOH 10 μM), Amiloride [500 μM is a nonspecific inhibitor of NHEs (23)], HOE 694 [at 100 μM inhibits NHE 1 or 2 (7, 33)] and S3226 [1 μM inhibits NHE 3 only and not NHE 1 (29, 33)] in 1% DMSO were dissolved in the Na+-free and control Tyrode's buffer, so the tissue could be exposed to them during the Na+-free washout period and during the recovery phase, respectively. At the end of each phase of the protocol, the bath solution change was completed within 20 s. The rate of recovery from acidification was quantified by fitting the initial portion (first 30 s) of the [pH]i recovery with a linear regression (27, 33), and the data were expressed as pH recovery in pH units/s.
All data are expressed as means ± SE; n = number of placentas. Statistical analysis was by one-way ANOVA followed by Bonferroni post hoc test, and differences were considered significant if *P < 0.05 or greater.
EGF Acutely Stimulates [pH]i Recovery
Treating term villous trophoblast fragments acutely with incrementally increasing doses of EGF resulted in a dose-dependent increase in rate of [pH]i recovery following acidification (Fig. 1), and 10 μM NAOH (⇓Fig. 3) had no effect on baseline pHi or NHE activity. The effect was maximal at an EGF concentration of 10 ng/ml, which caused a 100% increase in rate of [pH]i recovery compared with control. This dose was used in subsequent experiments in this study.
Rate of recovery of [pH]i above control by EGF is abolished by adding the NHE inhibitors amiloride, S3226, and HOE694 (Fig. 2). DMSO (data not shown) alone had no effect on baseline pHi or NHE activity. The presence of amiloride and HOE694 both results in [pH]i recovery levels that are significantly lower than control (Fig. 2).
EGF has no Effect on Baseline [pH]i
Activation of NHE3 has been shown to alkalinize cells at physiological pH without prior acidic challenge (3). Under control conditions, more than 20 min baseline [pH]i in villous trophoblasts hardly varied, gradually increasing by 0.00032 ± 0.00007 pH units/s from physiological starting [pH]i (which was 7.45 ± 0.07 in these experiments). EGF addition had no significant effect on baseline pH over 20 min compared with control: the rate of increase in the presence of the growth factor was 0.00026 ± 0.00005 pH units/s.
S1P Acutely Stimulates [pH]i Recovery
S1P increased rate of [pH]i recovery in a dose-dependent manner (Fig. 3). Stimulation of activity was maximal between 500 nM and 10 μM. S1P is dissolved in NaOH (10 μM), which could potentially change [pH]i independently of an S1P effect. However, when tested in isolation without S1P, there was no significant difference between rate of recovery between presence and absence of NaOH (Fig. 3).
S1P has no Effect on Baseline [pH]i
Like EGF, S1P had no further effect on the gradual rise in [pH]i seen over 20-min control incubation: with S1P, the rise was 0.00034 ± 0.00012 pH units.
Here, we show that both EGF and S1P significantly stimulate the rate of [pH]i recovery from an acid load in syncytiotrophoblast of intact human placental villous fragments. The inhibitory effects of specific blockers suggest that this stimulation was due to activation of NHE in the syncytiotrophoblast. Previous data also showed that recovery of syncytiotrophoblast [pH]i from an acid load is predominantly due to NHE activity (8).
Significant EGF upregulation of NHE activity occurred with concentrations of 10 ng/ml and 100 ng/ml, which is in line with EGF effects on cultured trophoblasts in regard to other end points such as apoptosis (11). Similar EGF concentrations are found in maternal urine (41); however, maternal serum EGF concentrations vary widely (40), and locally produced concentrations in the maternal villous space are unknown. In other cell types, EGF has been shown to stimulate NHEs by phosphorylation (25), calcium release (4), and increased cycling of the exchanger to the cell surface (5). In trophoblasts EGF increased activation of ERK (18), PI3K (18, 22), and PKC (37), which would potentially link EGF to activation of NHE1, 2, or 3. The stimulatory effect of EGF is completely blocked by amiloride at 500 μM, which preferentially blocks the activation of NHE1 but also potentially blocks activation of NHE2 and 3 (23). NHE2 activity is not present in the syncytiotrophoblast (33); however, NHE2 protein is present in the syncytiotrophoblast (14), and mRNA is present in the placenta throughout gestation (19), and absence of its activity cannot be confirmed without a specific inhibitor able to distinguish between NHE1 and NHE2. This is currently not available. In the results presented here, HOE694 does not reduce NHE activity to near zero like amiloride, which may be due to the lack of HOE694 action on NHE3 (33). S3226 at 1 μM reduces EGF-stimulated NHE activation to the level of control and predominantly acts on NHE3 [EC50 = 0.3 μM (29)], therefore suggesting increased NHE3 activity in response to EGF. However, like HOE694, S3226 does have some interaction with NHE1, though usually at higher concentrations than those used here (29, 33), and confirmation of NHE3 activation will require further investigation. The effect of amiloride, HOE694, and S3226 has previously shown a similar pattern of response on NHE in the human syncytiotrophoblast without the stimulation seen with EGF and S1P (33). We conclude that the EGF effect on NHE activity is likely to be predominantly through the NHE1 isoform but with possible contribution of NHE3. The data at present do not allow us to produce firm conclusions, and experiments in cell culture models with selective knockdown of expression of combinations of the NHE isoforms with siRNA may allow firm conclusions as to the importance of each isoform.
It has been shown previously that EGF is capable of stimulating rises in basal cell pH by phosphorylation of the exchanger (3). However, activation of NHE3 in this manner is not a universal finding (12). In the experiments reported here, EGF had no effect on basal syncytiotrophoblast pH after a 20-min incubation, a finding that repeats the findings of our own laboratory with other NHE agonists (P. F. Speake, personal communication). It is therefore likely that some other part of the complex, and not yet fully understood, NHE3 regulatory pathway is inhibiting NHE3 activation in syncytiotrophoblast at physiological pH.
The effect of S1P on [pH]i recovery became statistically significant at 500 nM, though it was more robust at 1 μM. In vivo S1P serum levels are not yet fully quantified but are thought to vary between 0.5 μM and 1 μM (15). The pattern of inhibition produced by the NHE inhibitors amiloride, HOE694, and S3226 was strikingly similar to that produced with EGF, but the effect of S3226 on S1P-induced NHE activation was not statistically significant. Although it cannot be concluded that there is any S1P activation of NHE3 in syncytiotrophoblast at this time, the trend suggests a weak link may be being concealed by the detection sensitivity of the NHE activity assay. Information on the signaling pathways stimulated by S1P in syncytiotrophoblast is limited; however, data have demonstrated the presence of mRNA for S1P receptors (17). Gαi/S1P receptor coupling could potentially influence NHE activity in two ways. First, Gαi inhibits adeylate cyclase and reduces intracellular cAMP. NHE1 activity is independent of cAMP concentrations (2, 10), but there is still a possibility of NHE2 activity as discussed above. Second, Gαi activates the MAPK, ERK (24), which has been shown to increase NHE1 activity via p90RSK (35). S1P could also activate NHE1 and 3 through Gαq and Gα12/13 coupling to S1P2&3. As with EGF, determining which of these is specifically involved in syncytiotrophoblast will require further investigation. S1P also had no effect on basal NHE activity, presumably for the same reason discussed in regard to EGF.
In summary, the data presented in this study are the first to show that EGF and S1P acutely upregulate NHE activity in the human placental syncytiotrophoblast. Further investigation is now required to examine whether this effect is related to the physiological role of NHE in controlling [pH]i of the syncytiotrophoblast and/or that activation of NHEs by EGF or S1P is part of the antiapoptotic response of this cell.
HOE 694 and S3226 were kindly donated by Dr. Jurgen Punter, Aventis Pharma. We would like to thank the midwives of the Central Delivery Unit of St. Mary's Hospital, Manchester, for their help in obtaining placentas.
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 © 2007 the American Physiological Society