| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Academic Division of Obstetrics & Gynaecology,4 School of Human Development, University of Nottingham, Derby City General Hospital, Nottingham DE22 3NE, United Kingdom
Institute of Cell Signalling,5 University of Nottingham, Queens Medical Centre, Nottingham NG7 2UH, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
decidua, parturition, pregnancy, signal transduction, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The human decidua comprises a heterogeneous cell population. Flow cytometric analysis of human term decidual cell suspensions has revealed that more than half of the cells are of stromal origin, with the remainder expressing the leukocyte common antigen (CD45) and of hematopoietic lineage [6].
Ion channels are ubiquitously expressed in excitable and nonexcitable cells, where they mediate transmembrane fluxes. Membrane channels are ideally located to respond to chemical and electrical changes in the microenvironment, thereby coordinating the local signaling pathways. In the pregnant human uterus, the best-characterized channels are of the large-conductance, calcium-activated potassium (BKCa)-channel class in the myometrium [7–9]. Very little information is available regarding ion-channel expression and function in fetal tissues (amnion, chorion, and placenta) and maternal decidua. In view of the close apposition of the decidua to the myometrium, these two tissues likely regulate, either directly or indirectly, the physiological status of the other via the influence of their respective secretory outputs. Sartor et al. [10] originally described a T-type, voltage-dependent calcium current in human decidual cells that was linked to prolactin secretion. The authors were also able to demonstrate spontaneous electrical activity in these cells. With the exception of the latter, the electrophysiology of ion channels in decidual cells has not been described in any species.
The aim of the present study was to establish if differential activation of whole-cell potassium (K+)-channel subtypes occurs in the human decidua during pregnancy and parturition. We report here that subtle variations in the expression of K+ currents in an enriched stromal cell preparation derived from human term decidual tissues are observed following spontaneous vaginal delivery (SVD; i.e., labor) compared with tissue obtained at elective cesarean section (CS; i.e., nonlabor).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Physiological salt solution (PSS) consisted of NaCl (135 mM), KCl (5 mM), MgCl2 (1 mM), CaCl2 (1 mM), and Hepes (10 mM) and was pH adjusted to 7.2 with NaOH. The intracellular (pipette) solution contained KCl (140 mM), MgCl2 (1 mM), EGTA (1 mM), Hepes (10 mM), and CaCl2 (0–1 mM) depending on the required free intracellular Ca2+ concentration ([Ca2+]i) as determined using the ion-ligand binding program METLIG (Dr. R.M. Denton, University of Bristol, Bristol, U.K.). Collagenase A, Ficoll, Percoll, Igepal, BSA, iberotoxin, paxilline 4-aminopyridine (4-AP), tetraethylammonium chloride (TEA), E-4031, and verruculogen were purchased from Sigma, Ltd. (Poole, Dorset, U.K.). Dynabeads M-450 CD45 coated with EO1 monoclonal antibody were from Dynal Biotech (Warrington, U.K.). Anti-Thy-1 antibody was purchased from Serotec (Oxford, U.K.) and mouse immunoglobulin (Ig) G from DAKO (Glostrup, Denmark). Paxilline and verruculogen were dissolved in dimethyl sulfoxide to make a stock solution of 10 mM. The concentration was kept below 0.1% (v/v). The TEA, 4-AP, and E-4031 were prepared as 1 M, 5 M, and 10 mM stock solutions, respectively.
Collection and Dissociation of Decidual Tissue
Informed, written consent was obtained from patients undergoing either elective CS (nonlabor) or uncomplicated SVD (labor) between 37 and 42 wk of gestation. The CS was performed for maternal request, fetal distress, or breech presentation in the absence of any underlying disease. Patients with diabetes, preeclampsia, or hypertension were excluded from the present study, which was approved by the Ethics Committee of Queens Medical Center (Nottingham, U.K.).
Immediately after delivery of the placenta, decidua parietalis was stripped from the underlying chorion and transferred in cold PSS (see Solutions and Chemicals) to the laboratory within 10 min. Decidual tissue (20 g) was washed three times in Hepes-buffered, calcium (Ca2+)- and magnesium (Mg2+)-free Hanks balanced salt solution (HBSS), minced finely in HBSS supplemented with collagenase A (2 mg ml-1), and incubated in the same enzymatic solution for 1 h at 37°C with gentle trituration every 15 min. Following filtration through a sterile, 40-µm cell strainer, the cell suspension was carefully layered onto a 60% (v/v) Percoll gradient and centrifuged at 4°C, 800 x g, for 20 min and washed twice in PSS in readiness for experiments. The removal of leukocytes was carried out by the addition of CD45-coated Dynabeads that were washed in PSS containing 0.01% (v/v) BSA before use. The cell sample (1 ml) was added immediately to the Dynabeads, mixed gently, and incubated for 20 min at 2–8°C with gentle agitation. Sufficient PSS was added to dilute the sample 3-fold, and the tube was placed in the Dynal magnet for 2–3 min. The supernatant containing CD45-negative cells was transferred into a second tube for immunohistochemistry and electrophysiological study.
Anti-Thy-1 Immunofluorescence
Cytospins of isolated decidual cells were fixed in 2% (w/v) paraformaldehyde, permeabilized with 0.5% Igepal in 0.1 M PBS. Following blocking with 3% (w/v) BSA in PBS, cells were incubated with anti-Thy-1 (CD90) antibody (1:100) at 4°C overnight and washed (three times) in PBS followed by fluorescein isothiocyanate-conjugated goat anti-mouse secondary antibody (1:50). Corresponding negative controls were processed as described above but were incubated with mouse IgG instead of the primary antibody and visualized with a fluorescence microscope (Nikon Eclipse, TE200; Surrey, U.K.).
Patch-Clamp Electrophysiology
The conventional whole-cell configuration of the patch-clamp technique [11] was employed to record from freshly dispersed decidual stromal cells. The cells were placed directly in the cell chamber mounted on the stage of an inverted microscope (Nikon Eclipse 200), allowed to settle, and then washed with PSS at a constant speed (6 ml min-1). Pipettes were drawn from borosilicate glass on a Heka-puller (Heka, Lambrecht, Germany) and had resistances of 4–7 M
when filled with electrolyte. Automatic series-resistance compensation was performed routinely and monitored continuously. Recordings were terminated if the access resistance changed by more than 25% during the recording period.
All current and voltage pulses were delivered through an EPC-9 amplifier using Pulse software (V8.51; Heka). Cell resting membrane potentials were recorded in current-clamp mode immediately after a stable seal configuration had been established. Following this, membrane voltage responses and action potentials were evoked by current injection to deliver both hyperpolarizing and depolarizing pulses of varying stimulus intensity and duration. The effects of a variety of classical ion-channel blockers, tetrodotoxin (TTX), TEA, barium (Ba2+), and cobalt (Co2+) on action potential properties were tested under current-clamp.
Outward currents were recorded under whole-cell voltage-clamp. Step depolarizations were applied from various holding potentials (-80, -60, -40, and -20 mV) in 10-mV increments to a maximum voltage of +60 mV. The sensitivity of the evoked outward currents to a series of K+ channel blockers was assessed by gravity perfusion of the drugs directly into the bathing chamber. All experiments were carried out at room temperature (20–24°C).
Data and Statistical Analysis
Data acquisition and analysis were performed with the Pulse (V8.51) software and Microsoft Excel (Microsoft, Redmond, WA) on a Pentium III PC. Action potential duration (APD) was measured at 25%, 50%, and 90% levels of repolarization (APD25, APD50, and APD90), respectively. Leak subtraction, using an automated Positive/Negative (P/N) protocol, was performed on all currents. Steady-state or peak outward currents, in the absence or presence of test agents, were separated into resistant and sensitive components. Thus, sensitive current was obtained by digital subtraction of the resistant component from total current. Currents were filtered at 1 kHz (-3 dB down) and sampled at 2 kHz. Results are expressed as the mean ± SEM of n observations. Statistically significant differences were evaluated by using a paired or unpaired two-tailed Student t-test as appropriate. A P value of <0.05 was considered to be significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Current-Clamp Study
High-resistance (5–15 G) seals between the electrode and stromal cell membrane were obtained consistently. Resting membrane potentials and evoked electrical activity were recorded in current-clamp mode. Stromal cells typically had resting membrane potentials of -63.0 ± 4.7 mV (n = 14) for the SVD group and of -57.5 ± 4.8 mV (n = 16) for the CS group (P > 0.05). Approximately 80% of cells in both the SVD and CS groups were able to generate action potentials following current injection. The responses were graded and dependent on stimulus intensity. The current response was defined as an action potential when its overshoot exceeded zero and by its dependence on the threshold potential for action potential generation. For decidual cells, this threshold potential was found to be positive to -30 mV.
The actions of the respective Na+, K+, and Ca2+ ion-channel selective blockers TTX, TEA, and Co2+ as well as Ba2+ on action potential parameters were tested. Figure 1 illustrates the effects of these agents on SVD cells. Identical effects were observed on CS cells (data not shown). Thus, TEA (20 mM) caused a slight but significant (P < 0.05), reversible prolongation of the action potential compared with control (n = 7), with no effect on overshoot (Fig. 1A). The Ba2+ (5 mM) increased both the duration of the repolarization phase of the action potential as well as augmented the action potential overshoot (n = 7) for SVD (Fig. 1B) and CS cells (data not shown). The inorganic calcium-channel blocker Co2+ (5 mM) reduced repolarization of the action potential in both the SVD (n = 6) (Fig. 1C) and CS group (n = 7). A decreased overshoot was also associated with this effect (Fig. 1C and Table 1). Action potential duration was reduced when external Ca2+ solution was replaced with a nominally Ca2+-free solution (n = 7) (Fig. 1D). This effect was partially reversed when extracellular Ca2+ levels were restored. Action potentials of both groups were not significantly affected by TTX (n = 7). A summary of these results is presented in Table 1.
|
|
Whole-Cell K+ in Decidual Cells
Cell capacitance, an indicator of cell surface area, was measured under conventional whole-cell voltage clamp. Both SVD and CS cells were characterized by capacitance values of 98.4 ± 6.7 pF (n = 23) and 99.6 ± 6.9 pF (n = 31), respectively (P > 0.05). Outward K+ current (IK) was elicited by depolarizing the membrane to +60 mV for 200–400 msec from different holding potentials (-40, -60, -80, and -100 mV) without the addition of pharmacological agents. Under these conditions, all cells tested responded with outward currents, which activated and decayed to differing extents within the duration of the test pulse. In Figure 2, currents evoked after depolarization of the cell membrane from a Vh of either -40 or -80 mV were characterized by an early peak component that was more prominent in the SVD group (Fig. 2A), whereas currents from cells of the CS group appeared to inactivate less rapidly (Fig. 2B) than those of their SVD counterpart. The early peak component appears to resemble an IA-like current [12], whereas the current activated later in the test pulse has the features of the delayed rectifier (IKV) current (Fig. 2, A and B, top). Interestingly, macroscopic peak and steady-state current densities in the SVD group were significantly smaller compared with the CS group, as noted for individual current families and current-voltage relationships (Fig. 2, A–C). The steady-state IK current-voltage relationship (Fig. 2C) demonstrates clear evidence of outwardly rectifying currents at positive membrane voltages.
|
Comparison of the Effects of TEA on K+ Currents Between SVD and CS
Bath-applied TEA (10 mM) had a large inhibitory effect on IK at all voltages tested. In Figure 3, A and B, a representative TEA block of outward current is shown from a Vh of -80 to +60 mV in 10-mV steps where both peak and steady-state current components were inhibited. When total IK (Fig. 3C) was pharmacologically separated into TEA-resistant and TEA-sensitive components and current-voltage relationships for both groups compared, it was apparent that a relatively smaller TEA-resistant IK occurred in the SVD cells, yet a greater TEA-resistant IK contribution to total IK occurred in the CS groups. To define the voltage-dependence and degree of TEA block in both groups more clearly, TEA-sensitive IK was normalized with respect to total IK and percentage inhibition by TEA for both groups calculated (Fig. 4). From this analysis, approximately 40–60% of IK in SVD cells is TEA-sensitive (n = 6) in the voltage range of -10 to +50 mV, compared with a 10–35% TEA-sensitive component in CS cells (n = 9; P < 0.05).
|
|
Effect of E-4031 on K+ Currents in SVD and CS
The sensitivity of IK to E-4031, a blocker of the rapid delayed rectifier (IKr) [13], was tested in both the nonlabor CS and the labor SVD group. Figure 5 depicts a family of outward currents activated from a Vh of -40 in 10-mV test pulses to +60 mV. This Vh was chosen to eliminate contamination of whole-cell current with the fast, transient outward current. It is evident from this representative trace (Fig. 5A) that a considerable proportion of the SVD current is inhibited by E-4031 (n = 6) and is in contrast to the lack of an appreciable E-4031-sensitive current in the CS group (Fig. 5B). Contrary to its role as a blocker of IKr, the E-4031-sensitive current in decidua was unmasked as a slowly activating outward current (Fig. 5A, bottom trace). These findings are summarized in the current-voltage plots where steady-state current density during the test pulse, IKstep (Fig. 5, C–E), and tail current densities, IKtail (Fig. 5, F–H), are resolved into E-4031-resistant (Fig. 5, D and G) and E-4031-sensitive components (Fig. 5, E and H). Both analyses demonstrate that E-4031 sensitivity is considerably reduced in the CS group compared with the SVD group (P < 0.05). We investigated whether the E-4031 block observed could be caused by activation of a HERG (human ether-ago-go)-like conductance. By employing a characteristic pulse protocol designed to isolate IKr [14], we did not observe any such current (data not shown).
|
4-AP Sensitivity of IK Currents in SVD and CS Cells
Total peak outward current (Ipeak) in decidual stromal cells of the SVD (n = 8) and CS (n = 8) group was blocked by extracellular application of 4 mM 4-AP, a blocker of the fast, transient outward current (IA) [12], causing a rounding of the early peak current (Fig. 6, A and B). This inhibition by 4-AP was significant (P < 0.05) at depolarized potentials in both groups relative to control current (Fig. 6, C and D, upper traces). No significant effect (P > 0.05) of 4-AP on steady-state current density (IKV) was found in either the SVD or CS group (Fig. 6, C and D, upper traces; also see Fig. 8).
|
|
Calcium-Activated K+ Currents in SVD and CS Cells
The Ca2+-dependence of the whole-cell steady-state outward current was initially investigated by varying free [Ca2+]i levels. When whole-cell recordings were made in the presence of nominally free, 500 nM and, subsequently, 1 µM [Ca2+]i, no increase in current in either the SVD (n = 5) or CS group (n = 4) was noted (Fig. 7, A and B). We sought to confirm this by pharmacological blockade of both small (SKCa) and large-conductance Ca2+-activated K channels (BKCa). Under these conditions, [Ca2+]i was maintained at 1 µM to maximize Ca2+-dependent K+ current activation. The SKCa channel-blocker apamin (1 µM) and the BKCa channel-blockers iberiotoxin (100 nM), paxilline (100 nM), and verruculogen (100 nM) all failed to reveal any discernible effects on whole-cell currents in either the SVD (n = 12) (Fig. 8) or CS group (n = 17) (Fig. 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our interest in characterizing decidual stromal cells electrophysiologically stems from the fact that myometrial activation and stimulation is intricately controlled by factors emanating from the decidua, the secretion and signal transduction pathways of which would likely involve decidual current modulation. Moreover, despite their cellular origin, decidual stromal cells express high levels of 
-actin, a smooth muscle-associated contractile protein and may therefore be considered to be myofibroblasts [16]. This latter finding highlights the complexity of the decidua and its specialized biology, which to date has not been exploited for therapeutic intervention. Several studies have described methods that yield highly purified decidual cell preparations [17–19] and have been utilized to investigate immune [17, 20, 21] and endocrine [18] aspects of decidual biology but, hitherto, not their electrophysiology.
Only a few studies to date have characterized the currents present in stromal or fibroblast cells, with Rane [21] reporting the presence of an intermediate Ca2+-activated K+ conductance in stromal cells. From the present study, the prevalent current type recorded from decidual stromal cells is a Ca2+-insensitive delayed rectifier, apparently comprising distinct K+ conductances. When the Vh was altered, clear changes in associated whole-cell current waveforms were evident. Outward currents, predominantly of the delayed rectifier type, arise from the large voltage-gated (KV) K+ channel family and are found in a host of excitable and nonexcitable cell types. The KV currents have a varied phenotype generated principally by multiple genes encoding KV channels and posttranslational modifications to the channel assembly. Under our recording regimen, decidual stromal cells possess an IA-like current [12] but appear to lack Ca2+-activated K+ currents despite attempts at manipulating [Ca2+]i to maximize current flow or using pharmacological blockade.
Whereas no dramatic temporal changes in expression of K+ current subtypes were observed with the onset of labor, subtle differences were clearly noted. First, IK currents in cells of patients undergoing CS were generally much larger than those of cells from the SVD group. Because K+ current activation will tend to oppose cellular depolarization, the increased current density in nonlabor (CS) cells may be directed at maintaining a relaxed, inexcitable uterine environment for the fetus. Second, an enhanced sensitivity of IK to both TEA and E-4031 during labor was observed, with the latter in particular unmasking an E-4031-sensitive macroscopic and tail current component. These findings suggest that fine-tuning of K+ current expression occurs with the onset of labor to effect the shift from uterine quiescence to activation. Frequently, E-4031 is used to block IKr, the rapid delayed rectifier, encoded by the HERG (human ether-ago-go) gene [22], and in cardiac myocytes, it is recognizable by virtue of its unique activation profile. However, we conclude that the actions of E-4031 on decidual IK are in addition to those classically ascribed to this drug. Our results demonstrate that during labor, the enhanced sensitivity to TEA and E-4031 may result from the activation of specific, temporally activated K+ currents, most likely reflecting functional membrane adjustments to facilitate labor. At present, the precise function of K+ currents in decidual tissue is speculative. As in all mammalian cells, membrane integrity and permeability is maintained by individual ion conductances and, particularly, at rest by K+ channels. Ion channels also play key roles in secretory function. In light of the abundance of biologically active molecules produced by the decidua, K+ currents may be important in the secretory mechanism of this tissue, and inhibition or augmentation of these currents could potentially offer a novel means of regulating secretion.
Despite intense investigative effort, the triggers for the inititiation of human labor remain unidentified. In common with the myometrium, the decidua is quiescent throughout gestation until term, when this quiescence is reversed and is accompanied by an increased synthesis of contraction-associated proteins [23]. It is probable that in human decidua, some of these secreted products may determine the response of this tissue, in addition to that of the myometrium, by autocrine or paracrine influences. Indeed, the reduced release of ECCIa (an inhibitor of the L-type calcium channel) from fetal membranes with term gestation has been proposed as a novel mechanism whereby uterine quiescence is maintained through block of L-type Ca2+ current and, hence, myometrial contractility [24, 25].
It is difficult to stop the myometrium from contracting during established labor. Whereas decidual cells lack the dramatic changes of currents such as those seen in excitable tissues, a knowledge of decidual ion current expression and function could potentially provide a putative therapeutic target to treat preterm labor as an alternative to tocolytic drugs or as a supplement in combination with tocolytic agents currently prescribed to inhibit myometrial contractility. In fact, we have observed that oxytocin inhibits Ca2+ currents in human decidual stromal cells of patients undergoing SVD with greater potency than those of the CS group (unpublished observations). The latter finding implies that oxytocin may act in a negative feedback fashion by potentially inhibiting its own release following delivery, thereby allowing cessation of myometrial activity. In extension of this, if the signals that lead to a decreased density of potassium current can be deciphered, then these may be applied clinically to maintain the decidua in a prelabor state, thereby preventing myometrial activation. Furthermore, the significance of action potentials in term decidual tissue, and whether they are expressed in early pregnancy human decidua, is not clear, although they would facilitate cell-cell communication and could be linked to secretion. We propose that the reduced K+ current density of decidual cells, together with the switch in IK phenotype in CS compared with SVD cells, may correlate with the enhanced contractile status of the uterus overall. In parallel with earlier, more definitive diagnoses of preterm labor, it may eventually prove possible to prevent the onset of preterm labor by arresting the process in the decidua even before myometrial activation has occurred. The decidua is a complex, differentiated tissue. It has nutritional, immunological, and supportive roles throughout gestation, and it may be that many of these functions are, in part, subserved by ion channels. Further studies are underway in our laboratory to determine the molecular identity and modulation of these currents in bringing about the fine-tuning of K+ currents in this tissue.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: R.N. Khan, Academic Division of Obstetrics & Gynaecology, University of Nottingham, Derby City General Hospital, Uttoxeter New Road, Derby DE22 3NE, U.K. FAX: 1332 625634; raheela.khan{at}nottingham.ac.uk
3 Current address: Department of Obstetrics & Gynaecology, St. George's Hospital, Tooting, London, U.K
Received: 1 October 2002.
First decision: 8 November 2002.
Accepted: 21 January 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-smooth muscle actin and show ultrastructural similarities with myofibroblasts. Hum Reprod 1999 14:599-1605This article has been cited by other articles:
|
|
 |
|
  S. L. Pierce, J. D.K. Kresowik, K. G. Lamping, and S. K. England Overexpression of SK3 Channels Dampens Uterine Contractility to Prevent Preterm Labor in Mice Biol Reprod, June 1, 2008; 78(6): 1058 - 1063. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Liu, S. J. Hill, and R. N. Khan Oxytocin Inhibits T-Type Calcium Current of Human Decidual Stromal Cells J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4191 - 4197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and CS (•) decidual cells. Each point is the mean (± SEM) of at least 10 separate observations. *Significance when P < 0.05
