| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Received 8 June 1998; accepted after revision 13 August 1998.
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
cm2) and potential difference (15·3 mV), with an inward short-circuit current (Isc; 20·5 µA cm-2).
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
An appropriate ionic environment within the uterus is of central importance to reproductive function. For example, it has been suggested that sperm viability and fertilizing ability (Roblero et al. 1990) and the cleavage rate of early embryos (Roblero & Riffo, 1986) are sensitive to the external K+ concentration. However, little is known about the mechanisms involved in regulation of the intrauterine environment, particularly in the human. In the rat, a transuterine potential difference is associated with active absorption of Na+ and active secretion of K+, Cl- and HCO3- (Levin & Edwards, 1968, Levin & Scargill, 1987). The net electrogenic result of these transport mechanisms has been measured as an inward short circuit current (Isc) (Levin & Phillips, 1983; Levin & Scargill, 1987). Similarly, a transuterine potential difference has been noted in humans in vivo (Duncan & Levin, 1976). More recently, amiloride-sensitive Na+ absorption has been reported in porcine (Vetter & O'Grady, 1996) and murine (Chan et al. 1997) endometria. Human uterine fluid provides an environment which is high in K+ (Clemetson et al. 1973; Casslén & Nilsson, 1984) and, compared with serum, relatively low in Na+ and Ca2+ (Casslén & Nilsson, 1984). The endometrial functions which give rise to these concentration differentials are sustained in vitro, since human endometrial epithelial cells cultured on permeable tissue culture supports modify the apical (luminal) culture medium by active absorption of Na+ and Ca2+ and passive secretion of K+ (Matthews et al. 1993a). In this study, we have investigated the properties of epithelial cells isolated from endometrium and cultured on porous tissue culture supports and used such cultures to assess the nature of electrogenic ion transport across these cells. We have identified an apical amiloride-sensitive conductance in these cells which is likely to be responsible for the reduced Na+ concentration of the intrauterine environment.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Tissue samples
Samples of endometrium were obtained, with written, informed consent, from patients undergoing gynaecological surgery, either by dilatation and curettage or by use of a 'Pipelle' endometrial sampling device (Eurosurgical Ltd, Cranleigh, UK). Tissue samples were placed in Hanks' balanced salt solution without calcium and magnesium (Life Technologies, Paisley, UK) containing 20 mM Hepes, 100 i.u. ml-1 penicillin, 100 µg ml-1 streptomycin and 5 µg ml-1 fungizone (Life Technologies) for transfer to the laboratory. Ethical approval for the study was obtained from the Newcastle Health Authority-University of Newcastle upon Tyne Joint Ethical Committee.
Isolation of epithelial cells
Tissue samples were digested with collagenase as previously described (Matthews et al. 1992b). Isolated glands were separated from single stromal cells using nylon mesh filters or by allowing glands to sediment out of suspensions of glands and stromal cells.
Cell culture
Isolated fragments of glands were cultured on 10 mm Anopore (Nunc, Roskilde, Denmark), 12 mm Millicell PCF polycarbonate (Millipore, Watford, UK), 12 mm Snapwell polycarbonate or 6·5 mm Transwell Clear polyester (Costar, High Wycombe, UK) tissue culture supports coated with the basement membrane preparation Matrigel (Collaborative Biomedical Products, Becton Dickinson UK Ltd, Cowley, UK). The type of support used for different experimental procedures was determined by the requirements of the apparatus or the methods used. Ussing-type chambers were suitable for Anopore and Millicell PCF supports, while Snapwell chambers, which are simple to dismantle and wash, were selected for 22Na+ flux experiments. Transmission electron microscopy requires a culture support which is easy to cut, and 6·5 mm Transwell Clear supports have the advantage that larger numbers of replicate cultures can be established. There were no functional differences between cells grown on these different supports. The culture medium consisted of a 1:1 mixture of Ham's F12 and Dulbecco's modified Eagle's medium supplemented with 2 % v/v Ultroser G (a serum substitute containing oestrogens), 1 % v/v non-essential amino acids, 10 mM glutamine, 20 mM Hepes, 100 i.u. ml-1 penicillin, 100 µg ml-1 streptomycin, 5 µg ml-1 fungizone, 50 µg ml-1 gentamicin (all from Life Technologies) and 10 µg ml-1 insulin (Sigma). The cultures were maintained in a humidified atmosphere of 5 % CO2 in air at 37°C. The culture medium was changed every 2-3 days. The development of transepithelial resistance (RT) on individual culture supports was monitored using a combined voltmeter and constant current source (Evom, World Precision Instruments, Owslebury, Hampshire, UK) coupled to Ag-AgCl 'chopstick' electrodes.
Electron microscopy
Endometrial epithelial layers which produced high RT values ( > 400 cm2) on Millicell PCF culture inserts were prepared for transmission electron microscopy (TEM). Cell layers were fixed in 2 % glutaraldehyde in Sorensen's phosphate buffer and post-fixed in OsO4. Sections of 70 nm of epoxy resin-embedded cultures were transferred to Formvar-coated copper grids, stained with uranyl acetate and lead citrate and examined in a JEOL 100S electron microscope (JEOL (UK) Ltd, Welwyn Garden City, UK).
Cell layers with RT values over 400 cm2 grown on Anopore membranes were prepared for scanning electron microscopy. Cell layers were fixed in glutaraldehyde, dehydrated, dried, coated with 15 nm gold and examined with a Cambridge S240 scanning electron microscope (SEM; Leo Electron Microscopy Ltd, Cambridge, UK).
Electrophysiological measurements
Confluent epithelial cell layers maintained on Anopore or Millicell PCF tissue culture inserts were mounted in Ussing-type chambers maintained at 37°C, connected to an automatic voltage clamp (DVC 1000, World Precision Instruments) via 3 M KCl-3 % w/v agar salt bridges and reversible electrodes (Ag-AgCl for current passage, calomel for voltage sensing). Measurements of open-circuit electrical potential difference (VT) and Isc were made within 5-10 min in modified Krebs solutions. RT was calculated, using Ohm's law, from the voltage evoked by a standard current (5 µA). Calculated RT values were not significantly different from those obtained by Evom during culture, as is usually the case for culture inserts with a small growth area. During the course of experiments Isc was measured and RT was monitored by the passage of regular 1 s current pulses every 5·5 s. Experiments were commenced once the Isc was stable.
Solutions
The standard, modified Krebs solution contained (mM): NaCl, 137; KCl, 5·4; MgSO4, 0·99; KH2PO4, 0·34; NaH2PO4, 0·30; Tris base, 14; CaCl2, 4·17; glucose, 10; adjusted to pH 7·4 with HCl. In chloride-free solutions, Cl- was replaced with methanesulphonate and in nominally sodium-free solutions, Na+ was replaced with choline. In solutions containing intermediate sodium concentrations, isotonicity was maintained with choline. Phosphates were omitted from solutions for Ba2+ experiments and MgSO4 was replaced by MgCl2.
Pharmacological reagents
Amiloride hydrochloride (Sigma) and 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB; SmithKline Beecham, Welwyn Garden City, UK; stock solution 10 mM in DMSO) were added to modified Krebs solution as blockers of sodium and chloride channels, respectively. Ouabain (Sigma) was used as an inhibitor of Na+-K+-ATPase. BaCl2 (Sigma) was used as an inhibitor of K+ channels.
Measurement of 22Na+ flux
Epithelial cell layers maintained on Snapwell culture supports were mounted in Snapwell diffusion chambers (Costar) and bathed in modified Krebs solution at 37°C. VT, RT and Isc were measured as described above. To measure apical to basolateral Na+ flux (JNaab), 2 µCi of 22Na+ (as NaCl; Amersham International plc, Little Chalfont, UK) were added to the apical solution. Samples of 20 µl were taken in duplicate from this solution to determine the initial isotope concentration. Samples of 500 µl were taken from the basolateral solution at 10, 20 and 30 min, this volume being replaced with modified Krebs solution. During this 30 min incubation period monolayers were maintained under short circuit conditions and Isc was constantly monitored. VT and Isc were measured and RT calculated at the end of the incubation. The diffusion chambers and cell layers were extensively washed to remove the isotope, and VT, Isc and RT were again recorded to assess epithelial integrity and function. A 500 µl aliquot of the apical bathing solution was removed for measurement of any residual 22Na+ in this compartment, and replaced with modified Krebs solution. To measure basolateral to apical flux (JNaba), 2 µCi of 22Na+ was added to the basolateral solution, and 20 µl samples were taken as before. Samples of 500 µl were taken from the apical solution at 10, 20 and 30 min. 22Na+ in all samples was assessed using an appropriately gated channel in an LKB 1282 Compugamma gamma counter (LKB Pharmacia Ltd, Milton Keynes, UK). Cell layers in which RT fell below 700 cm2 following washing of the cell layer were excluded from the analyses.
Cell layers on Transwell Clear supports were used to assess the effects of amiloride on 22Na+ flux. Separate groups of cultures were used for apical to basolateral and basolateral to apical flux measurements under open-circuit conditions over a 30 min incubation period in the presence or absence of 10-4 M apical amiloride. Evom measurements of VT and RT were made before and after a 20 min incubation with amiloride prior to 22Na+ addition.
Data analyses and statistics
Values are expressed as means ± 1 standard error of the mean (of n cell layers, from N tissue samples). Two-sample comparisons were performed using Student's t test for paired or unpaired data or the Mann-Whitney U test as appropriate. Apparent Ki values were obtained using non-linear regression analysis (GraphPad Prism, GraphPad Software, Inc., San Diego, CA, USA).
Sodium fluxes are expressed in micromoles per square centimetre per hour. In calculations of basolateral to apical fluxes, any residual 22Na+ remaining in the apical compartment was taken into account. Predictions of the magnitude of sodium flux based on mean Isc values were made on the basis of 1 µA cm-2 = 36 nmol cm-2 h-1. Wilcoxon's rank sum test was used to determine whether the difference between measured and predicted fluxes was significantly different from zero.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cell culture
Human endometrial epithelial cells cultured on tissue culture supports generated significant RT values as early as 1 day, but more commonly within 2-6 days of seeding (Fig. 1A). Occasionally, RT developed after longer periods in culture. Early development of RT was not an indicator of final RT values and may simply reflect the initial seeding density. In many cultures, resistance could be maintained for several weeks (Fig. 1B). Epithelial layers with high RT values were prepared using both Anopore and polycarbonate supports and neither type of membrane proved superior in this respect. Cultures on Anopore membranes could be visualized under the light microscope and were found to contain areas which varied in the density of cell packing, presumably as a result of the sites of attachment of the individual glandular fragments. Occasional small glandular fragments remained.
Confluent epithelial cell layers were generated from endometrial samples obtained at all stages of the menstrual cycle. The majority of these successful cultures, however, were established in the central portion of the cycle (mid-proliferative to mid-secretory).
 |
View larger version [in this window] [in a new window] |
|
|
A, development of RT of four epithelial layers. Individual layers differ in both the time taken for RT to develop and the final RT reached. B, maintenance of high RT by an epithelial layer over an extended incubation period. Many layers have high RT for several weeks. | ||
Morphology of endometrial cultures
Human endometrial epithelial cells formed simple monolayers and only rarely were multilayered areas observed. The cells displayed features characteristic of columnar epithelia (Fig. 2). Cell nuclei were in a basal position and areas of contact between adjacent cells were joined by tight junctions, desmosomes and lateral membrane interdigitations. The apical surfaces of the cells bore microvilli which varied in diameter and length and ciliated cells were also present.
The apical surface of confluent epithelial layers was heterogeneous, with individual cells occupying varying areas of the epithelial surface (Fig. 3). Surface microvilli were clearly visible. Ciliated cells, which appeared to be relatively small and bore densely packed, longer surface projections, were interspersed among the other epithelial cells.
 |
View larger version [in this window] [in a new window] |
|
|
Transmission electron micrograph of human endometrial epithelial cells cultured on a polycarbonate (Millicell PCF) support (scale bars, 1 µm). A and B, epithelium and culture support; C, intercellular junctional complex; D and E, apical microvilli. | ||
 |
View larger version [in this window] [in a new window] |
|
|
Scanning electron micrograph of human endometrial epithelial cells cultured on an Anopore membrane (scale bars = 5 µm). A and B, cells occupy varying areas of the epithelial surface and most intercellular junctions are well defined. C and D, detail of cell surface illustrating microvilli and cilia. | ||
Electrophysiological measurements
Epithelial cultures which produced Evom measurements in excess of twice that of the culture support alone were mounted in Ussing-type chambers. Based upon this criterion, 26 % of tissue samples produced epithelial layers suitable for use in these experiments and the majority of these (94 %) demonstrated RT > 200 cm2. A total of 494 cell layers, derived from ninety-seven tissue samples, were used after 3-42 days (mean, 12 days) in culture. The initial measurements of VT, inward Isc and RT made within 5 min of mounting the tissue culture inserts in the Ussing chamber were 15·3 ± 0·6 mV (maximum, 63·9 mV), 20·5 ± 0·7 µA cm-2 (maximum, 119·0 µA cm-2) and 750 ± 22 cm2 (maximum, 3683 cm2), respectively. In general, a high VT accompanied a high Isc and these high values were most frequently associated with average to high RT values. No clear correlation was observed between any of these variables and the stage of the menstrual cycle during which the tissue sample was obtained or the length of time in culture.
Effects of ion substitutions on Isc
Na+. Bilateral replacement of modified Krebs solution with nominally Na+-free solution (choline substitution) largely abolished the Isc, which reached a stable level after 2-3 min and remained low. Mean Isc decreased from 19·2 ± 1·8 to 0·9 ± 0·4 µA cm-2 (n = 46, N = 16; P < 0·0001), a 92·1 ± 2·5 % reduction. In some epithelial layers (n = 11, N = 8), a small outward current was observed in Na+-free buffer. Incubation of endometrial epithelial cell layers in buffers containing 137, 100, 50 and 0 mM Na+ revealed a linear relationship between external Na+ concentration and the magnitude of the Isc (Fig. 4).
 |
View larger version [in this window] [in a new window] |
|
|
Isc, expressed as a percentage of the control Isc in standard buffer, is linearly related to the external Na+ concentration. Results are expressed as means ± 1 | ||
Cl-. Bilateral replacement of modified Krebs solution with Cl--free buffer (methanesulphonate replacement) resulted in a transient decrease in mean Isc, from 19·6 ± 2·2 to minimum values after 1-4 min of 11·8 ± 1·4 µA cm-2 (n = 51, N = 14; P < 0·0001), a reduction of 37·9 ± 3·8 %. While in the majority of cell layers, Isc decreased in Cl--free solutions, in a small number of cases (n = 4, N = 3) an increase was observed. In contrast to cell layers incubated in Na+-free solutions, following prolonged incubation in Cl--free buffer the initial decrease in Isc was followed by a gradual recovery to pre-treatment levels within 8-20 min. For example, in a group of layers with an initial mean Isc of 19·4 ± 2·0 µA cm-2, the change to Cl--free solution was associated with a decrease in Isc to 8·0 ± 0·6 µA cm-2, with subsequent recovery and overshoot of Isc to 22·8 ± 1·5 µA cm-2 (n = 3, N = 1) after 30 min; this Isc was then maintained over extended incubation periods (> 2 h).
Effects of pharmacological agents on Isc
Amiloride. Amiloride, used as a blocker of Na+ channels, was added to either the apical or basolateral bathing solutions. Measurements reached new, stable levels within 2 min. The presence of 10-5 M amiloride in the apical bathing solution caused a significant reduction in Isc from 30·3 ± 3·7 to 7·3 ± 1·0 µA cm-2 (P < 0·0001), a 73·4 ± 6·1 % decrease from pretreatment values (n = 19, N = 6). Amiloride at 10-4 M reduced the Isc from 24·9 ± 3·9 to 3·2 ± 0·6 µA cm-2 (P < 0·0001), an 83·8 ± 2·8 % reduction (n = 19, N = 7). In addition, short term treatment (3-5 min) of the apical surface of cell layers with 10-4 M amiloride was associated with a significant increase in RT from 687 ± 91 to 1000 ± 170 cm2 (P < 0·01; n = 8, N = 1), together with a decrease in transepithelial conductance (GT) from 1·753 ± 0·333 to 1·345 ± 0·316 mS cm-2 (P < 0·0001). In contrast, basolateral addition of 10-4 M amiloride caused a much smaller, but nevertheless significant, reduction in Isc from 29·7 ± 7·3 to 25·3 ± 6·7 µA cm-2 (18·4 ± 4·1 % reduction; P = 0·02; n = 8, N = 4).
In order to characterize the effect of amiloride on the inward Isc, the effect of a range of amiloride concentrations, added to the apical bathing solution, on Isc measured in modified Krebs buffers containing 137 or 50 mM Na+ (replaced by choline) was assessed. The Isc values were normalized as percentages of the control values in the absence of amiloride to control for variations in initial Isc between cell layers (Fig. 5). In the standard buffer containing 137 mM Na+, all concentrations of amiloride tested (10-7 to 10-4 M) reduced the basal Isc. In the presence of 10-4 M amiloride the Isc was largely abolished (90·3 ± 2·0 % reduction, n = 6, N = 1). In buffer containing 50 mM Na+ the concentration-response curve to amiloride was shifted to the left and under these conditions the current was largely abolished by 10-5 M amiloride (90·4 ± 6·5 % reduction, n = 5, N = 3). Apparent Ki values calculated from these concentration-response curves were 1·26 µM in the presence of 137 mM Na+ and 0·29 µM in buffer containing 50 mM Na+.
 |
View larger version [in this window] [in a new window] |
|
|
Isc of epithelial layers bathed in 137 mM ( | ||
NPPB. NPPB at 10-5 M was added to the apical bathing solution to investigate a possible contribution of Cl- secretion to the Isc. The vehicle, dimethyl sulphoxide, alone did not influence Isc. Any effects were apparent within 1 min. Addition of NPPB most commonly caused a small decrease in Isc, but in some layers (n = 6, N = 4) there was no effect and occasionally (n = 3, N = 3) a small increase in current occurred. Overall, a relatively small but significant decrease in Isc was observed in the presence of NPPB, from 13·1 ± 2·6 to 11·8 ± 2·3 µA cm-2(n = 26, N = 12; P = 0·01), an 8·4 ± 3·1 % decrease.
Ouabain. Layers on Millicell PCF supports were mounted in a modified Ussing chamber and Isc and VT were monitored continuously. Basolateral 10-4 M ouabain was applied and the Isc was compared with that in a group of control cultures. In ouabain-treated layers, 83·8 ± 2·1 % (n = 10, N = 5) of original Isc was lost during a 10 min incubation, while in control cell layers 35·6 ± 3·6 % (n = 15, N = 5) was lost over the same time scale. After correction for the reduction in Isc in control cell layers, the ouabain-induced reduction in Isc after 10 min was 74·8 %.
Ba2+. Addition of 10 mM Ba2+ to the basolateral solution caused an immediate decrease in Isc, representing a loss of 36·6 ± 2·6 % of initial Isc (n = 10; N = 5 ; P < 0·0001). By contrast, in the majority of epithelial layers, apical 10 mM Ba2+ gave slight effects (loss of < 10 % of Isc) indistinguishable from those seen following equivalent osmolarity changes induced by the addition of mannitol.
22Na+ flux measurements
In seven epithelial layers (N = 2) RT was maintained at > 700 cm2 throughout the experimental procedures (overall, RT decreased by 33·4 ± 6·3 %) and these layers were included in the analyses. Mean Isc values for these epithelia throughout the incubation period were 13·1 ± 1·0 µA cm-2. JNaab exceeded JNaba, resulting in net absorptive fluxes (JNanet) in all epithelial layers (e.g. Fig. 6). RT increased from 813 ± 44 to 1047 ± 91 cm2 during the second incubation period. While there was variation between individual epithelial layers and the data were not normally distributed, JNanet values (median, 0·534 µmol cm-2 h-1) were not significantly different from the flux rates predicted from mean Isc values calculated from the readings taken every minute during the incubation period (median, 0·443 µmol cm-2 h-1; median difference, 0·101 mmol cm-2 h-1; Wilcoxon's rank sum test for (observed) - (predicted) = 0, P = 0·353).
 |
View larger version [in this window] [in a new window] |
|
|
Cumulative transepithelial flux of Na+ in an illustrative epithelial layer, determined from bidirectional 22Na+ fluxes. | ||
Effect of amiloride on 22Na+ flux
In a group of layers, apical preincubation with 10-4 M amiloride caused a 68·8 ± 1·9 % reduction in VT (n = 33, N = 31). Basolateral to apical flux of Na+ was not significantly altered in the presence of amiloride (JNaba: control, 0·297 ± 0·019 µmol cm-2 h-1 (n = 18, N = 16); amiloride, 0·325 ± 0·025 µmol cm-2 h-1 (n = 18, N = 17); P = 0·0896), while apical to basolateral flux was significantly reduced (JNaab: control, 0·842 ± 0·056 µmol cm-2 h-1 (n = 18, N = 15); amiloride, 0·475 ± 0·030 µmol cm-2 h-1 (n = 15, N = 14); P < 0·0001). Net Na+ transport, in the apical to basolateral direction, was reduced by 77·4 % compared with control, non-amiloride treated cultures.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Control of the electrolyte composition of uterine fluid is essential for normal reproductive function. Using primary monolayer cultures of human endometrial epithelia on porous supports, we have demonstrated an inward Isc which is explained by the presence of an apical amiloride-sensitive Na+ conductance. This conductance, with characteristics similar to other Na+-absorbing epithelia, accounts for the Na+-absorptive ability of the human endometrium and establishes an electrochemical gradient favouring K+ secretion. The endometria of the pig (Vetter & O'Grady, 1996) and the mouse (Chan et al. 1997) contain similar Na+-conductive pathways. Conservation of such a function between species suggests that it may be essential to the creation of a suitable intrauterine environment.
Epithelial cells cultured as polarized monolayers provide an appropriate experimental system for studying epithelial function and transport processes. Such cultures are of particular value in the study of epithelia which are largely inaccessible in situ, such as those within the human endometrium. Polarized epithelial cultures were established from tissue samples obtained from all stages of the menstrual cycle, extending the preliminary observations of Schatz et al. (1990) on successful culture of two samples of human proliferative phase endometrium for 14 days. The majority of successful cultures, however, were established during the central part of the cycle. This may reflect lower cell yields earlier in the cycle, while tissue degeneration prior to menstruation may reduce success for samples from later stages of the cycle. The development of a significant RT as early as 1 day, and most frequently 2-6 days, after initiation of the cultures demonstrates that the epithelial cells within glandular fragments readily form a confluent monolayer. This was not solely a feature of proliferative phase tissue. Confluent endometrial cell monolayers displayed typical epithelial morphology with prominent apical tight junctional complexes. The heterogeneity of cell types within the monolayers, including ciliated cells interspersed among more numerous cells bearing microvilli, resembles closely the endometrial surface in situ (Healy, 1991). Hence, these cultures provide an appropriate morphological correlate of the endometrial epithelium in vivo.
These morphologically polarized endometrial cells also display clear evidence of polarization of function, as evidenced by the large inward Isc indicative of a significant capacity for electrogenic transport. A number of lines of evidence suggest that this Isc is the result of amiloride-sensitive Na+ absorption. In the absence of Na+ the Isc was virtually abolished and, over the concentration range 0-137 mM, the relationship between the Isc and the external [Na+] was linear. The lack of saturation of Na+ absorption over this concentration range demonstrates the large capacity of the Na+ transport system. That Na+ absorption accounts for the Isc was confirmed by measurement of bidirectional flux of 22Na+ across individual cell layers. The experimentally determined Na+ movement accounted for the current, consistent with the view that the current is due solely to Na+ absorption.
Apical amiloride inhibited the absorptive flux of Na+ across endometrial epithelia by approximately 80 %, while basolateral amiloride had a relatively small effect on Isc, which may be a non-specific consequence of disturbance of the cell layer during solution changes and/or amiloride diffusion across the epithelium. The reduction in Isc caused by apical amiloride was accompanied by a significant increase in RT and decrease in GT, highlighting the amiloride-induced inhibition of a Na+ conductance in the apical membrane. The apparent Ki for amiloride in 137 mM Na+ was 1·26 µM, suggesting a specific effect on a Na+ channel. A reduction in the external [Na+] from 137 to 50 mM resulted in a reduction in apparent Ki, consistent with competitive antagonism of the effect of amiloride by Na+. Amiloride has additional inhibitory effects on the Na+-H+ exchanger, on the Na+-Ca2+ exchanger and on Na+-coupled solute transport (Benos, 1988; Benos et al. 1992; Barbry & Hofman, 1997). However, it is not yet known whether these are expressed in human endometrial epithelial cells, while apparent Ki values differ significantly between epithelial Na+ channels and these other amiloride-sensitive transport mechanisms. The apparent Ki for human endometrial epithelium is similar to values reported for other Na+-absorbing epithelia (Smith & Benos, 1991; Garty & Palmer, 1997). Hence, the Isc is largely accounted for by an amiloride-inhibitable Na+ conductance at the apical cell surface. This conductance is fundamental to endometrial ion transport pathways since inhibition of the apical Na+ conductance with amiloride severely attenuates the ability of epithelial layers to modify the K+ and Cl-, as well as the Na+, composition of the apical and basolateral culture media (Matthews et al. 1993c).In contrast to the large effects of treatments interfering with Na+ transport, treatments capable of inhibiting Cl- transport had comparatively modest effects on the Isc. Removal of Cl- had only a temporary inhibitory effect on the Isc and the presence of NPPB, a Cl- channel blocker, in the apical bathing solution caused only a small (< 10 %) decrease in Isc, which may be due to relatively non-specific effects of NPPB on other ion transport systems, the respiratory chain, cAMP synthesis and intracellular Ca2+. Thus, while the generation of the Isc is dependent upon external Na+ and an amiloride-inhibitable Na+ conductance, the role of Cl- under these unstimulated conditions, if any, is minor. The situation in the pig is different, however, and in this species Cl- secretion is involved in the basal and the stimulated Isc (Deachapunya & O'Grady, 1998).
The inhibitory effects of basolateral ouabain and Ba2+ are consistent with basolateral expression of Na+-K+-ATPase and K+-conductive pathways, respectively. The failure of 10 mM Ba2+ to modify Isc when applied to the apical surface provides no evidence for involvement of an apical K+ conductive pathway in the generation of endometrial K+ secretion (Matthews et al. 1993c). Thus, a model outlining the features of human endometrial ion transport can be proposed (Fig. 7). Basolateral Na+-K+-ATPase generates an electrochemical gradient for Na+ absorption via apical amiloride-sensitive conductive pathways. The Na+-K+-ATPase pump activity is maintained by basolateral K+ conductance. In contrast, K+ secretion may be explained by paracellular transport driven by the Na+ absorption-generated transepithelial potential. Alternatively, a Ba2+-insensitive apical K+ conductance could explain K+ secretion by human endometrial epithelia. An apical Ca2+-activated K+ conductance has been identified in pig endometrial epithelium (Vetter & O'Grady, 1996).
 |
View larger version [in this window] [in a new window] |
|
|
Evidence for an apical amiloride-sensitive Na+ conductance and basolateral Na+-K+-ATPase and K+ conductance are presented in this study. The mechanism of K+ secretion has yet to be identified. | ||
The ability of the endometrial epithelium to elevate intrauterine [K+] may have a beneficial influence on sperm function and embryo growth (Roblero & Riffo, 1986; Roblero et al. 1990). Whether the active Na+ absorption pathway identified in the present study operates solely to provide the driving force for passive K+ secretion across the endometrial epithelium, or if the consequent lowering of intrauterine [Na+] per se is also functionally significant, has yet to be elucidated.
Endometrial ion transport may be under the control of steroid hormones since, in the rat, progesterone increases uterine fluid [Na+], while at the same time decreasing [K+] (Nordenvall et al. 1989). In the present study no functional changes occurring with the stage of the menstrual cycle at the time of tissue sampling were identified. However, any differences may have been lost in culture and we cannot exclude the possibility that endometrial ion transport in vivo may vary during the menstrual cycle. Such alterations are unlikely, however, since no significant changes in human uterine fluid [Na+] were found during the cycle (Casslén & Nilsson, 1984).
The endometrium, lining the uterine cavity, is ideally situated to influence the intrauterine environment via absorptive and secretory processes occurring across its epithelia. This study investigated the basic mechanisms of these processes in the human and provides the first direct evidence for an amiloride-inhibitable Na+ conductance in human endometrial epithelial cells. Electrogenic ion transport across human endometrial epithelial cells is influenced by adrenaline (Matthews et al. 1992a), bradykinin (Matthews et al. 1993a) and gastrin-releasing peptide (Matthews et al. 1993b) and such factors may, therefore, have the ability to influence the intrauterine environment. In some forms of infertility or pathology, the defect may be absent or inappropriate control of endometrial ion transport. An improved understanding of endometrial transport processes may, thus, provide an explanation for some cases of menstrual dysfunction or 'unexplained' infertility.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| Barbry, P. & Hofman, P. (1997). Molecular biology of Na+ absorption. American Journal of Physiology 273, G571-585 | [Medline] |
| Benos, D. J. (1988). Amiloride: chemistry, kinetics, and structure-activity relationships. In Na+/H+ Exchange, ed. Grinstein, S., pp. 121-136. CRC Press, Boca Raton, FL, USA. | |
| Benos, D. J., Warnock, D. G. & Smith, J. B. (1992). Amiloride-sensitive transport mechanisms. In Membrane Transport in Biology, ed. Schafer, J. A., Kristensen, P., Ussing, H. H. & Giebisch, G. H., vol. 5, pp. 166-232. Springer Verlag, Berlin. | |
| Casslén, B. G. & Nilsson, B. (1984). Human uterine fluid, examined in undiluted samples for osmolarity and the concentrations of inorganic ions, albumin, glucose, and urea. American Journal of Obstetrics and Gynecology 150, 887-881. | |
| Chan, H. C., Liu, C. Q., Fong, S. K., Law, S. H., Leung, P. S., Leung, P. Y., Fu, W. O., Cheng Chew, S. B. & Wong, P. Y. (1997). Electrogenic ion transport in the mouse endometrium: Functional aspects of the cultured epithelium. Biochimica et Biophysica Acta 1356, 140-148 | [Medline] |
| Clemetson, C. A. B., Kim, J. K., de Jesus, T. P. S., Mallikarjuneswara, V. R. & Wilds, J. H. (1973). Human uterine fluid potassium and the menstrual cycle. Journal of Obstetrics and Gynaecology of the British Commonwealth 80, 553-561 | [Medline] |
| Deachapunya, C. & O'Grady, S. M. (1998). Regulation of chloride secretion across porcine endometrial epithelial cells by prostaglandin E2. The Journal of Physiology 508, 31-47 | [Abstract/Full Text] |
| Duncan, S. L. B. & Levin, R. J. (1976). Transuterine, transendocervical and transvaginal potential differences in conscious women measured in situ. Journal of Physiology 259, 27P. | |
| Garty, H. & Palmer, L. G. (1997). Epithelial sodium channels: function, structure, and regulation. Physiological Reviews 77, 359-398 | [Medline] |
| Healy, D. L. (1991). Endometrial prolactin and implantation. Baillieres Clinical Obstetrics and Gynaecology 5, 95-105 | [Medline] |
| Levin, R. J. & Edwards, F. (1968). The transuterine endometrial potential difference, its variation during the oestrous cycle and its relation to uterine secretion. Life Sciences 7, 1019-1036 | [Medline] |
| Levin, R. J. & Phillips, J. C. (1983). Rat endometrial bioelectric activity in vivo and in vitro: effects of adrenaline. The Journal of Physiology 336, 465-478 | [Abstract] |
| Levin, R. J. & Scargill, M. (1987). Role of bicarbonate and chloride ions in rat endometrial electrogenic secretory activity. The Journal of Physiology 391, 21P. | |
| Matthews, C. J., McEwan, G. T. A., Redfern, C. P. F., Thomas, E. J. & Hirst, B. H. (1992a). Electrical properties of primary monolayer cultures of human endometrial glandular epithelial cells. The Journal of Physiology 446, 97P. | |
| Matthews, C. J., McEwan, G. T. A., Redfern, C. P. F., Thomas, E. J. & Hirst, B. H. (1993a). Bradykinin stimulation of electrogenic ion transport in epithelial layers of cultured human endometrium. Pflügers Archiv 422, 401-403 | [Medline] |
| Matthews, C. J., Redfern, C. P. F., Hirst, B. H. & Thomas, E. J. (1992b). Characterization of human purified epithelial and stromal cells from endometrium and endometriosis in tissue culture. Fertility and Sterility 57, 990-997 | [Medline] |
| Matthews, C. J., Redfern, C. P. F., Thomas, E. J. & Hirst, B. H. (1993b). Bombesin and gastrin-releasing peptide stimulate electrogenic ion transport in cultured human endometrial epithelial cell layers. Experimental Physiology 78, 715-718 | [Medline] |
| Matthews, C. J., Thomas, E. J., Redfern, C. P. F. & Hirst, B. H. (1993c). Ion transport by human endometrial epithelia in vitro. Human Reproduction 8, 1570-1575 | [Medline] |
| Nordenvall, M., Ulmsten, U. & Ungerstedt, U. (1989). Influence of progesterone on the sodium and potassium concentrations of rat uterine fluid investigated by microdialysis. Gynecologic and Obstetric Investigation 28, 73-77 | [Medline] |
| Roblero, L. S., Fernandez, E., Guadarrama, Z., Zegers-Hochschild, F. & Ortiz, M. E. (1990). High potassium concentration and the cumulus corona oocyte complex stimulate the fertilizing capacity of human spermatozoa. Fertility and Sterility 54, 328-332 | [Medline] |
| Roblero, L. & Riffo, M. D. (1986). High potassium concentration improves preimplantation development of mouse embryos in vitro. Fertility and Sterility 45, 412-416 | [Medline] |
| Schatz, F., Gordon, R. E., Laufer, N. & Gurpide, E. (1990). Culture of human endometrial cells under polarizing conditions. Differentiation 42, 184-190 | [Medline] |
| Smith, P. R. & Benos, D. J. (1991). Epithelial Na+ channels. Annual Review of Physiology 53, 509-530 | [Medline] |
Vetter, A. E. & O'Grady, S. M. (1996). Mechanisms of electrolyte transport across the endometrium. I. Regulation by PGF2 and cAMP. American Journal of Physiology 270, C663-672 |
[Medline] |
We are grateful to the medical and nursing staff of the gynaecology operating theatres in Newcastle General Hospital and Royal Victoria Infirmary, Newcastle upon Tyne, for their assistance. We thank the Biomedical Electron Microscopy Unit, University of Newcastle upon Tyne, for the handling of EM samples. This work was supported by The Wellcome Trust (grants 033271/Z/91/Z and 041743/Z/94/Z).
Corresponding author
B. H. Hirst: Department of Physiological Sciences, University of Newcastle upon Tyne, Medical School, Newcastle upon Tyne NE2 4HH, UK.
Email: barry.hirst{at}ncl.ac.uk
Authors' present addresses
G. T. A. McEwan: Department of Biomedical Sciences, Institute of Medical Sciences, University of Aberdeen, Aberdeen AB2 2ZD, UK.
E. J. Thomas: Department of Obstetrics and Gynaecology, University of Southampton, Princess Anne Hospital, Coxford Road, Southampton SO16 5YA, UK.
This article has been cited by other articles:
|
|
 |
|
  A. E. Vetter and S. M. O'Grady Sodium and anion transport across the avian uterine (shell gland) epithelium J. Exp. Biol., February 1, 2005; 208(3): 479 - 486. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
; n = 6, N = 1) or 50 mM Na+ (
; n = 5, N = 3) buffers as a function of amiloride concentration. Apparent Ki for amiloride inhibition of Isc is 1·26 µM (
, basolateral to apical; 
, net flux.