The molecular mechanisms controlling fluid secretion within the oviduct have yet to be determined. As in other epithelia, both secretory and absorptive pathways are likely to work in tandem to drive appropriate ionic movement to support fluid movement across the oviduct epithelium. This study explored the role of potassium channels in basolateral extracellular ATP (ATPe)-stimulated ion transport in bovine oviduct epithelium using the Ussing chamber short-circuit current (ISC) technique. Basal ISC in bovine oviduct epithelium comprises both chloride secretion and sodium absorption and was inhibited by treatment with basolateral K+ channel inhibitors tetrapentlyammonium chloride (TPeA) or BaCl2. Similarly, ATP-stimulated chloride secretion was significantly attenuated by pretreatment with BaCl2, tetraethylammonium (TEA), tolbutamide, and TPeA. Basolateral K+ current, isolated using nystatin-perforation technique, was rapidly activated by ATPe, and pretreatment of monolayers with thapsigargin or TPeA abolished this ATP-stimulated K+ current. To further investigate the type of K+ channel involved in the ATP response in the bovine oviduct, a number of specific Ca2+-activated K+ channel inhibitors were tested on the ATP-induced ΔISC in intact monolayers. Charbydotoxin, (high conductance and intermediate conductance inhibitor), or paxilline, (high conductance inhibitor) did not significantly alter the ATPe response. However, pretreatment with the small conductance inhibitor apamin resulted in a 60% reduction in the response to ATPe. The presence of small conductance family member KCNN3 was confirmed by RT-PCR and immunohistochemistry. Measurements of intracellular calcium using Fura-2 spectrofluorescence imaging revealed the ability of ATPe to increase intracellular calcium in a phospholipase C-inositol 1,4,5-trisphosphate pathway-sensitive manner. In conclusion, these results provide strong evidence that purinergic activation of a calcium-dependent, apamin-sensitive potassium conductance is essential to promote chloride secretion and thus fluid formation in the oviduct.
- oviduct epithelium
the fluid formed within the oviduct provides an essential and appropriate environment for gamete maturation, transport, fertilization, and early embryo development. Although the composition of this fluid has been well documented for some time (8, 40), the mechanisms underlying fluid formation surprisingly remain poorly understood.
As a general principle, fluid formation is secondary to and driven by the active transport of ions across an epithelial layer. The predominant electrolyte responsible for driving fluid secretion across most epithelia is chloride. Chloride secretion via ion channels is accompanied by the movement of sodium across the epithelial layer via paracellular pathways. The resulting luminal accumulation of sodium chloride generates an osmotic gradient pulling water into the lumen. In the oviduct, fluid secretion has been shown to be directly associated with chloride secretion (14, 15). Oviduct fluid formation is essential to hydrate and protect the mucosal surface and to provide an appropriate ionic environment to facilitate healthy reproductive processes (25). In many disease states, such as pelvic inflammatory disease or cystic fibrosis, dysregulated fluid formation occurs leading to reduced fertility in vivo and can also compromise in vitro fertilization efforts in women effected (1, 2). Thus there is both a fundamental gap in our knowledge and a clear clinical necessity for more investigation into the basis and regulation of chloride transport processes in the oviduct.
Chloride secretion is maintained by the concerted actions of a number of ion transporters and channels in the luminal epithelium. Luminal chloride exiting the epithelium occurs predominantly through the calcium-activated Cl− channel (CaCC) and the cAMP-activated CFTR Cl− channels (24). Epithelial Cl− secretion is also critically linked to basolateral K+ channel activity. In epithelia, generally, the potassium channel activation that serves to recycle potassium ions thus preventing cell depolarization and ultimately maintaining a driving force for chloride secretion at the apical pole of the cell (5, 30). Additionally, K+ channels are important in supporting sodium absorption (which also occurs in the bovine oviduct), through apically located epithelial sodium channels (ENaC) (24, 42). The identity and role of these potassium channels in reproductive tissues and particularly the oviduct has yet to be investigated. Over ten years ago the presence of a large conductance maxi-K+ channel type in the apical membranes of rabbit oviduct was reported (23). More recently the presence of calcium-activated potassium channels in the apical membrane of porcine endometrial epithelia, cells closely related to the oviduct, has been reported (34). In the majority of polarized epithelia, two principle types of basolateral K+ conductances have been described, activated by either an increase in cAMP or intracellular calcium providing an electrical driving force for chloride secretion (5, 6, 30). The possible involvement of these channels in generating a hyperpolarization and subsequent increase in driving force for chloride secretion and sodium absorption in response to extracellular ATP (ATPe) in the oviduct epithelial is unknown. Recently, we have shown that ATPe directly activates a P2Y, G protein-coupled two receptor on the basolateral membrane, which promotes chloride secretion through calcium-dependent (CaCC) and independent mechanisms (cystic fibrosis transmembrane conductance regulator, CFTR) (24). The supporting role and identity of the basolateral potassium channels in this process is as yet unknown and key to developing our understanding of fluid formation in the oviduct. Extracellular ATP may regulate these important transport processes by acting as an autocrine/paracrine agent. A variety of stimuli are capable of causing ATP release from epithelial cells, including hormonal, inflammatory, or mechanical stimuli. In the setting of inflammation, for example, the release of ATP may play an important role in the reproductive tract. ATPe could be involved in initiating nonspecific defense mechanisms whereby activation of chloride secretion, and the subsequent movement of water into the lumen, would help flush noxious substances from the lumen similar to ciliary clearance seen in the airways. This study reports the first known characterization of potassium channels associated with chloride secretion and sodium absorption across the oviduct epithelium. We have also identified the K+ channel subtypes activated in response to basolateral ATP challenge, which are essential in driving chloride secretion and thus fluid formation in the oviduct.
MATERIALS AND METHODS
ATP, barium chloride, tetraethylammonium (TEA), tetrapentlyammonium chloride (TPeA), and thapsigargin were purchased from Sigma Aldrich. Apamin, r-charbydotoxin, paxilline, and antibodies were purchased from Alomone. All general cell culture reagents were purchased from Sigma, with the exceptions of pancreatin, fungizone, and fetal calf serum (FCS), which were purchased from GIBCO.
Bovine oviducts at various stages of the estrous cycle were removed from cattle at a local abattoir within 5 min of slaughter. Epithelial cells were isolated according to the methods previously described for the isolation of rat uterine epithelium (15, 18). The oviducts were immediately separated from connective tissue, fat, and major blood vessels, and the fimbriae were removed. Oviducts were washed in a Ca2+- and Mg2+-free Hanks balanced salt solution (HBSS). Each oviduct was opened longitudinally to expose the epithelia and cut into 1-cm segments and incubated with 0.5% type I trypsin and 2.7% (wt:vol) pancreatin for 1 h at 4°C followed by 1 h at room temperature. The cellular suspension was vortexed for 30 s, followed by centrifugation at 500 g for 4 min. The resultant pellet was washed three times in Ca2+- and Mg2+-free HBSS. After washing was completed, cells were resuspended in prewarmed, pregassed culture medium.
Culture medium consisted of the nutrient mixture F12 Hams plus Dulbecco's modified Eagle's medium in a 1:1 ratio (vol/vol). The final culture medium also had the following additions: 0.1% bovine serum albumin, 150 IU/ml penicillin G, 150 μg/ml streptomycin sulfate, 1.25 μg/ml fungizone, 2 mM l-glutamine, and 10% FCS. Isolated cells from multiple oviducts were pooled and plated at a density of approximately 1 × 106 cells/ml.
Measurement of short-circuit current.
Cells were plated on snapwell inserts with a growth area of 12 mm (Corning Costar), and monolayers achieved confluence after approximately 4 days in culture. A volume of 250 μl of the cell suspension was added on top of the inserts, and 4 ml of culture medium were added underneath. The cells were incubated in a humidified incubator at 37°C and gassed with 5% CO2 in air. The medium above and below the filters was replaced every 48 h. The development of an appreciable transepithelial resistance (Rte) across these filters was used an indicator of the formation of a polarized monolayer, and this was measured daily using chopstick electrodes (EVOM voltage meter, World Precision Instruments, Sarasota, FL). All monolayers were used in a window of 6–7 days in culture and were only considered for use in electrophysiological studies once Rte ≥ 1 kΩ·cm2. Filters were placed in modified Ussing chambers (World Precision Instruments), with both surfaces of the cells bathed with normal Krebs-Ringer bicarbonate solution and gassed with 95% O2-5% CO2, at 37.5°C. The normal Krebs Ringer bicarbonate solution contained (im mM) 118 NaCl, 25 NaHCO3, 4.74 KCl, 1.19 MgSO4, 1.17 KH2PO4, 1.17 CaCl2, 1 glucose, and gassed with 95% O2-5% CO2.
The spontaneous transmembrane potential was measured using a voltage-clamp amplifier (DVC 1000, World Precision Instruments) and clamped to 0 mV by the application of a short-circuit current (ISCISC before beginning an experiment. Baseline values for transepithelial potential difference (Vte), ISC, and Rte were 6.5 ± 0.35 mV, 3.4 ± 0.3 μA/cm2, and 1.56 ± 0.08kΩ·cm2, respectively. ATP was prepared in normal Krebs-Ringer and added to the experimental chamber in 100-μl aliquots. All other agents/antagonists were prepared in DMSO (maximum final volume of 0.1%). Where ion channel inhibitors were used, cells were incubated in the presence of the inhibitor for 10 min before ATPe addition. Results are expressed as mean ISC (μA/cm2) or as a percentage of the control (normal) ATPe response.
Measurement of basolateral potassium currents.
To evaluate the basolateral K+ conductance of oviduct epithelia, monolayers were exposed to an apical to basolateral K+ gradient (21). NaCl in the apical bathing solution was replaced with equimolar K+-gluconate, and NaCl in the basolateral bathing solution was replaced with equimolar Na+-gluconate. The pore-forming antibiotic nystatin (500 IU/ml in <0.01% methanol) was added to the apical bath to permeabilize the apical membrane, and the transepithelial current was allowed to reach a steady state (26, 43). Complete permeabilization of the apical membrane with nystatin was confirmed by lack of effect of apical application of amiloride (100 μM), an inhibitor of ENaC located on the apical membrane of oviduct epithelial cells. Under these conditions the starting Rte was 1.588 ± 0.96 kΩ·cm2, and final Rte at termination of experiment was 1.31 ± 0.16 kΩ·cm2. In nystatin permeabilization experiments the basolateral membrane was bathed in a low-chloride Krebs solution of the following ionic composition (in mM): 100 sodium gluconate 20 NaCl, 25 NaHCO3, 11 glucose, 4.7 KCl, 2.5 CaCl2, 1.2 MgSO4, and 1.2 KH2PO4. pH was maintained at 7.4 by gassing with 95% O2-5% CO2. The apical membrane was bathed in potassium-rich solution of the following ionic composition (in mM): 120 potassium gluconate, 20 NaCl, 3 MgSO4, 1.2 KH2PO4, 11 glucose, 5 EDTA, and 6 HEPES. Nystatin forms cation-permeable pores in the apical membrane and isolates the basolateral membrane electrically (26). In the presence of a mucosa-to-serosa K+ gradient, the transepithelial current is generated by the K+ conductance of the basolateral membrane and K+ transport across the paracellular pathway. Any transport that may occur via a paracellular pathway is insensitive to K+ channel blockers. In these experiments the current remaining after K+ channel inhibition accounted for <10% of the total.
Measurement of intracellular calcium concentration.
Changes in intracellular calcium concentration ([Ca2+]i) were measured using the calcium-sensitive fluorescent probe fura-2 AM (Invitrogen). Confluent monolayers of epithelial cells grown on glass-bottomed dishes were loaded with fura-2 AM (10 μM in DMSO) for 30 min at 37°C, followed by 15 min at room temperature in the dark. After loading was completed, the cells were washed twice in a HEPES-buffered Krebs solution (in mM: 118 NaCl, 4.74 KCl, 1.19 MgSO4·7H2O, 1.17 KH2PO4, 1.17 CaCl2, 1.0 glucose, and 10 Na HEPES) before transfer to the stage of a Zeiss Axiovert inverted microscope. The microscope was equipped with a 10 position Orbit I filterwheel (Improvision), which had excitation filters for 340 and 380 nm. A built-in filter cube containing a 505-nm dichroic and a 510-nm long-pass selected the emission light, which was collected every second with an Orca 285 camera from Hamamatsu. The system was controlled by the Openlab system version 4.0 from Improvision run on a power mac G5. The software allowed for image collection and background subtraction, and ratio calculations were generated at every time point. The change in 340/380 fluorescence ratio was used to express changes in [Ca2+]i based on the Grynkiewicz equation (19). Rmin and Rmax are the values of ratio pairs in the presence of nominally zero and saturating Ca2+. Rmax was measured in the presence of 5 μM ionomycin with an external calcium concentration of 10 mmol/l and Rmin in nominally calcium-free solution buffered with 3 mmol/l EGTA.
RNA isolation and RT-PCR.
Briefly, oviduct epithelial cells (up to 5 × 106) cultured on snapwell inserts were collected by scraping the cells from the growing surface using a Pasteur pipette. The cellular suspension was then centrifuged and RNA isolation was achieved using a Nucleospin RNA II kit from Macherey-Nagel as per manufacturers instructions. The absorbance ratio at 260/280 nm of the resultant purified RNA was measured to determine quality and the RNA concentration from its absorption at 260 nm. RT-PCR analysis was performed using an enhanced avian HS RT-PCR kit from Sigma (cat no. HSRT-ZO) with primers designed to recognize the different K+ channel types (Table 1). The one-step RT-PCR reaction was carried as per the manufacturer's instruction with no modifications. The PCR products were stored immediately at −20°C and later analyzed by agarose (1.5%) gel electrophoresis and stained with ethidium bromide. Bands were visualized using an Alpha DigiDoc system. The primers used were confirmed positive for bovine transcripts by performing control experiments using total RNA from bovine CNS purchased from AMS Biotechnology (Abingdon, UK).
Isolated bovine oviduct epithelial cells were cultured for 5–7 days in glass-bottomed six-well plates to achieve confluency. Cells were washed three times in rinse buffer (20 mM Tris·HCl, 150 mM NaCl, 0.05% Tween-20), fixed with 4% paraformaldehyde (Sigma) for 20 min, and then permeabilized with 0.1% Triton X-100/PBS for 20 min at room temperature. Cells were then washed three times in rinse buffer, followed by a blocking solution (4% goat serum, Sigma) for 15 min, and stained with anti-KCNN3 (Alomone) for 1 h. Cells were washed three times with rinse buffer and incubated with a FITC-labeled goat anti-mouse IgM secondary for 60 min. Cells were then washed three times in rinse buffer and visualized on a Zeiss Axiovert 200 inverted microscope.
Results are expressed as means ± SE. Each cell isolation is an experimental block and each well/monolayer a replicate within the block. Thus each block contained cells pooled from two or more animals. For all treatments n indicates the number of wells/monolayers examined, and in all cases, replicates were examined over two or more experimental blocks. In all cases each treatment was paired with a vehicle control. The vehicle control was found to have no effect on basal or ATPe-stimulated ISC responses. All results were examined statistically by analysis of variance and, where appropriate, the Bonnferroni/Dunn post hoc test was used. P values <0.05 were considered to be statistically significant.
Baseline current in bovine oviduct epithelial cells.
Cultured monolayers of oviduct epithelial cells exhibited an average basal ISC of 3.4 ± 0.3 μA/cm2. Previously, we have shown that ∼50% of the baseline ISC is sensitive to amiloride, whereas ∼20% is sensitive to CFTR channel inhibitors (24). Thus the basal ISC in the bovine oviduct reflects predominately both Cl− secretion and Na+ absorption. To investigate the role of potassium channels on the basal ion transport, monolayers were treated with a range of potassium channel inhibitors of varying specificities: TEA (5 mM), BaCl2 (5 mM), tolbutamide (100 μM), or TPeA (100 μM) and the resultant percentage inhibition of ISC calculated (Table 2). TEA had no effect on the baseline ISC. Basolateral application of the nonspecific K+ channel blocker BaCl2 elicited a 46 ± 2% inhibition of the basal ISC. Tolbutamide, an inhibitor of KATP channels, and TPeA, a Ca2+-activated K+ channel inhibitor, induced larger decreases of 70 ± 2% and 71 ± 5%, respectively. In these experiments the presence or absence of bicarbonate had no effect on the level of current inhibition.
To investigate the contribution of each of the potassium channel types on basal sodium absorption in the oviduct, the magnitude of the inhibition of basal ISC by each of the K+ channel blockers before and after treatment with amiloride was also examined. In the presence of amiloride, the percentage inhibition of basal ISC by BaCl2, tolbutamide, and TPeA was 30.5 ± 2.5, 35.9 ± 2.8, 42.0 ± 7.2%, respectively, which was considerably lower compared with respective amiloride-untreated cells (Table 2). This highlights the importance of these basolateral K+ channels in maintaining luminal Na+ absorption in the oviduct epithelium. The effect of apical potassium channel inhibition on the basal ISC was also investigated by treating the cells on the apical side with tolbutamide (100 μM) or TPeA (100 μM), and the resultant percentage change in ISC was measured. Apical amiloride (100 μM) was present to prevent contamination with any sodium absorptive current. Apical TPeA and tolbutamide inhibited the baseline ISC by 39.8 ± 6.2% and 35.7 ± 3.0%, respectively, illustrating a role for apical K+ channels in the generation of the basal ISC in the bovine oviduct. This is of particular interest since this ion is present in the oviduct fluid at a concentration approximately three times that in plasma (8). Furthermore, apical K+ channel inhibition by either TPeA or tolbutamide in intact monolayers, in the presence of amiloride (100 μM) resulted in a significant reduction in the subsequent change in ISC in response to ATPe 26.1 ± 3.2% and 21.3 ± 5.8% (P < 0.0001), respectively, suggesting a role for apical potassium conductance in ATPe-mediated chloride secretion. Based on the current data, the mechanisms for the observed inhibition are unclear. We would suggest that the reduction in baseline current observed is most likely due to the inhibition of chloride secretion due to a depolarization of the apical cell membrane altering Va resulting in a decrease driving force for anion secretion.
Basolateral potassium movement involved in ATPe-stimulated chloride secretion.
In a number of tissues, basolateral potassium channels play an important role in maintaining the driving force for agonist-induced chloride secretion, and thus the effects of K+ channel blockers on the ATPe-induced chloride secretion were determined in monolayers of bovine oviductal epithelial cells (BOECs). Each K+ channel inhibitor was added to the basolateral bath 10 min before the application of ATPe in the presence of amiloride.
Pretreatment of the basolateral face of the monolayers with the inhibitors TEA or BaCl2 caused a significant reduction in the subsequent ISC response to ATPe by ∼9 ± 1.23 or 42% and 14 ± 1.38 μA/cm2 or 60% (P < 0.0001), respectively. Pretreatment with tolbutamide or TPeA resulted in the greatest level of inhibition of ISC in response to ATPe, by 18 ± 1.36 μA/cm2 or 80% and 21 ± 1.28 μA/cm2 or 90% (P < 0.0001), respectively (Fig. 1A). A representative trace of the effect of TPeA treatment is shown in Fig. 1B. TPeA is known to inhibit Ca2+-activated K+ channels in epithelia (41, 31) and has previously been shown to inhibit chloride secretory events in response to various agonists through inhibition of basolateral K+ channels (30).
Effect of ATPe on the basolateral potassium conductance in nystatin-perforated monolayers.
To further investigate the direct activation of basolateral K+ channels by ATPe, the nystatin perforation technique was used. This technique involves treatment of the apical membrane with nystatin, which has been previously shown to remove the electromotive force and resistance generated by the apical membrane thereby allowing investigation of the basolateral membrane in isolation (30).
In the presence of a basolaterally directed (apical to basolateral) potassium gradient, the addition of nystatin (500 IU) to the apical bath produced an immediate increase in transepithelial current of 38.1 ± 2.59 μA/cm2 and corresponding decrease in Rte (Fig. 2D). The K+ dependence of this current was tested by replacement of all potassium salts with sodium salts in the apical bath. Under these conditions nystatin failed to cause an increase in current, implying that the nystatin-induced transepithelial current under these conditions wholly reflects potassium transport across the basolateral membrane (IK) (data not shown). Subsequent basolateral addition of 100 μM ATPe resulted in a rapid transient increase in ISC of 35.56 ± 3.44 μA/cm2 and decrease in Rte (Fig. 2D) followed by a sustained decrease in IK (Fig. 2A), which was further decreased toward baseline by addition of TPeA. The current reduction in response to TPeA corresponded to an increase in Rte back toward perpermeabilization levels (Fig. 2D). The increase in current in response to ATPe was significantly inhibited in the presence of TPeA (P < 0.0001) (Fig. 2, B and C), illustrating the importance for basolateral Ca2+-dependent K+ channels. Tolbutamide, a KATP channel inhibitor, also greatly reduced the response to ATPe (P < 0.0001) (Fig. 2C). However, when both TPeA and tolbutamide were used together, no significant additive inhibitory effect was noted (Fig. 2C); therefore, it is possible that at the levels of tolbutamide used, other classes of potassium channel may also be inhibited (31). The sustained decrease observed after ATPe (Fig. 2A) is absent in the presence of TPeA pretreatment (Fig. 2B), suggesting that the inhibitory effect of ATPe is masked when K+ channels are already blocked and ATPe is directly effecting the movement of K+.
Effect of ATPe on intracellular calcium on intact monolayers.
We considered that calcium-activated potassium channels are implicated in the secretory response to ATPe, and we have previously shown that disruption of calcium signaling attenuates ATPe-stimulated chloride secretion in BOECs (24). We postulated that these effects were mediated by mobilization of intracellular calcium in response to ATPe challenge. Fura-2 spectrofluorescence imaging revealed that ATPe stimulated a rapid increase in [Ca2+]i (Fig. 3) characterized by a transient peak followed by a rapid return to basal levels. The absence of a maintained plateau suggested that the source of calcium solely involved Ca2+ mobilization from internal stores. This hypothesis was tested by examining the effects of ATPe on [Ca2+]i under external Ca2+-free conditions by treating cells with EGTA (2 mM) or in the presence of either thapsigargin (1 μM), 2-aminoethoxydiphenyl borate (2-APB, 150 μM), or BAPTA-AM (25 μM) to inhibit the internal release or limit internal availability of Ca2+ or in the presence of U73122 to block IP3 formation. Thapsigargin or BAPTA-AM inhibited the ATPe-stimulated increase in intracellular Ca2+ by ∼65% and 75%, respectively (Figs. 3 and 4). Pretreatment with the phospholipase C (PLC) inhibitor U73122 or inositol 1,4,5-trisphosphate (IP3) receptor antagonist 2-APB also significantly attenuated the ATPe response (Figs. 3 and 4). These results are consistent with the involvement of the PLC-IP3 pathway and the release of internal calcium in response to ATPe. Surprisingly, EGTA had a similar effect on the ATPe response, as P2RY2 receptor activation is normally associated with release of Ca2+ from intracellular Ca2+ stores (31). However, it was noted that in the presence of EGTA the basal calcium level was significantly reduced (data not shown). Thus we suggest that it is likely that the absence of extracellular Ca2+ caused depletion of internal stores thus rendering the cells refractory to subsequent stimulation with ATPe. Finally, both ATPe and UTPe activated increases in intracellular calcium, which were blocked to a similar extent (effect of suramin on ATPe action shown) by the P2 receptor antagonist suramin (Fig. 4), supporting our proposed role for P2RY2 in mediating secretory responses to ATPe (24).
Role of intracellular Ca2+ in activation of basolateral K+ current by ATPe under nystatin conditions.
To investigate the direct role for intracellular calcium in the activation of basolateral K+ current by ATPe, under nystatin-treated conditions, BOEC monolayers were pretreated with the Ca2+-ATPase inhibitor thapsigargin (1 μM) for 20 min. Under these conditions ATPe failed to activate the rapid transient component of the IK seen in Fig. 2A (P < 0.002); however, it is clear that only the rapid transient component is calcium dependent, while the sustained inhibitory effect is independent of calcium (Fig. 5, A and B). This result confirms the involvement of intracellular Ca2+ in the activation of a rapid transient basolateral potassium current activated by basolateral ATPe. Taken together, thus far these results demonstrate that basolateral purinoceptor activation by ATPe activates basolateral Ca2+-dependent K+ channels in bovine oviduct epithelial cells and a calcium independent inhibition of a K+ family type.
Type of Ca2+-activated K+ channel involved in the ATPe response.
As our previous work has highlighted the role of P2YR2 receptors and calcium in secretory response to ATPe in oviductal epithelial, we have focused on the rapid transient current component here. There are three main types of Ca2+-activated K+ channels: high conductance (BK), intermediate conductance (IK), and small conductance (SK) potassium channel. To further investigate the type of K+ channel involved, a number of specific Ca2+-activated K+ channel inhibitors were tested on the ATPe-induced ΔISC in intact monolayers.
Pretreatment of the cells basolaterally with CBTX, a BK and IK inhibitor, or paxilline, a BK inhibitor, did not significantly alter the baseline or ATPe response. However, pretreatment with apamin (P < 0.0001), a SK inhibitor while not effecting baseline current, caused a 60% reduction in the response to ATPe (Fig. 6), illustrating a role for the small conductance K+ channels in mediating the ATPe-activated response.
Expression of SK family members in BOEC monolayers.
With the use of RT-PCR, mRNA expression for each of the four known SK family members (KCNN1–4) was investigated. RT-PCR was performed on total RNA isolated from BOECs, which had been grown to confluence on Snapwell inserts under the same conditions used for standard electrophysiology experiments. Polymerase chain reaction products were derived using specific primers corresponding to published sequences for KCNN1, KCNN2, KCNN3, and KCNN4 (Table 1). Specific primers for 18S rRNA were used as housekeeping controls. Bands of predicted size were identified for KCNN3 (Fig. 7, lane 4) and were excised and sequenced commercially to confirmed identity, matching GenBank accession number XM_868686.2 and confirming the presence of KCNN3 at the mRNA level in BOECs. Furthermore, KCNN3 protein expression was examined in BOECs grown on glass coverslips until ∼85–90% confluent. Under these conditions cells stained positive for KCNN3 (Fig. 8). Taken together these results provide strong support for the electrophysiological data supporting the role of potassium channels in fluid formation and identifying KCNN3 (SK3) as a key player in basolateral ATPe-induced chloride secretory responses in oviduct epithelia.
The present study investigated the presence and role of potassium channels in supporting and maintaining basal and ATPe-activated secretion in the oviduct epithelium. Previously, we have shown that chloride secretion is the mainstay of basal and ATPe-stimulated ionic currents across the bovine oviduct epithelium (24). Furthermore, we showed that the observed ATPe effect is mediated at least in part through the purinergic receptor P2RY2 on the basolateral membrane of oviduct epithelial cells. The majority of the mechanistic elements involved in the chloride secretion pathway in the oviduct are unknown. The results presented here clearly demonstrate the presence of potassium channels in the basolateral domain of polarized BOECs. These potassium channels are active under basal conditions supporting both sodium absorption and chloride secretion by providing a driving force for the rapid transient increase in chloride secretion following stimulation with ATPe. Furthermore, we show that one of the specific downstream targets of P2YR2 activation is a small conductance calcium-activated potassium channel in the basolateral membrane that we propose facilities potassium cycling via the basolateral membrane providing the electrical driving force for chloride secretion into the lumen of the tube. Thus far the only report of potassium channels in the oviduct reported the presence of maxi-calcium-activated potassium channels using the patch clamp technique in isolated epithelial cells (23).
Transepithelial anion secretion is critically dependent on activation of basolateral K+ channels in a number of epithelia (4, 16, 38). This coordinated series of events has been described in detail in airway, colon epithelium, and pancreatic ducts. Airway epithelial cells like many secretory epithelia display electrogenic transepithelial anion secretion under basal conditions (10, 17, 28, 36) an activity that can be enhanced by increases in intracellular cAMP or Ca2+ concentrations (17, 37, 32). Anion secretion occurs predominately via cAMP-activated CFTR Cl− channels located at the apical membrane (7, 12, 20, 36) and is dependent on K+ channel activity. In most tissues studied thus far these channels allow for potassium recycling through the basolateral membrane with the ancillary effect of causing a significant hyperpolarization of the cell interior, thereby increasing the driving force for chloride secretion and sodium absorption across the apical pole (22, 42). Previously, we have shown that BOECs have anion secretory capability through CFTR and calcium-activated chloride channels found in the apical membrane. However, the possible involvement of potassium channels in the anion secretion underlying this process remains unknown. A similar mechanism to other epithelial cell types is likely to exist in the oviduct, however, to date there are no reports detailing the identity or function of potassium channels in oviduct epithelial cells.
Here we have presented data that support a role for potassium channels, in the generation of the basal ISC and furthermore are essential to the ATPe-induced Cl− secretory response (Fig. 1). The baseline ISC in BOECs is dependent on Na+ absorption via ENaC and Cl− secretion via CFTR (24). The substantial inhibitory effect of each of the various potassium channel inhibitors tested on the basolateral membrane on baseline ISC suggests that basolateral potassium channels have an important role in supporting basal ISC in BOECs most likely through increasing the driving force by generating a hyperpolarization of the cell. The inhibitory effects on basal ISC in the presence of amiloride (Table 2) illustrates that not only are potassium channels important in maintaining basal levels of chloride secretion but also that of sodium absorption in the resting state. Basolateral K+ conductance was also shown to be essential in generating the rapid increase in chloride secretion in response to ATPe (Fig. 1). Each of the K+ channel inhibitors exerted a significant inhibitory effect on the ATPe secretory response. TPeA almost completely blocked (>90%) the ATPe-induced Cl− secretion (Fig. 1) indicating an absolute requirement for activation of a basolateral K+ channels to promote the transient secretory response observed. Luminal oviduct K+ is normally higher than plasma thus an active K+ secretory mechanism is likely to exist as has been shown in the uterus (34). Here we show that apical K+ channel inhibition reduced basal and ATPe-activated current most likely due to depolarization of the apical cell membrane resulting in a reduced driving force for anion secretion and thus reduced secretory current. To investigate activation of basolateral potassium channels by ATPe, the basolateral membrane was isolated using nystatin, and in the presence of a basolaterally directed potassium gradient, the nystatin-induced current was identified as a potassium-dependent current. ATPe rapidly stimulated a transient increase followed by a sustained decrease in IK in nystatin-perforated monolayers, which in both cases could be attenuated by the potassium channels inhibitors TPeA and tolbutamide (Fig. 2). TPeA predominately inhibits Ca2+-activated K+ channels in epithelia (31) and has previously been shown inhibit Cl− secretory responses to various agonists through inhibition of basolateral K+ channels (30). Here the rapid transient activation of an IK is blocked by TPeA and tolbutamide (Fig. 2, B and inset) and furthermore this current is Ca2+ dependent (Fig. 5). Interestingly the subsequent sustained inhibition of IK has a different profile, it is not apparent in the presence of K+ channel blockers, and it is insensitive to thapsigargin, suggesting a different K+ channel or potentially the activation of alternate signaling pathways resulting in IK inhibition. Increases in intracellular calcium in response to basolateral ATPe was shown and examined using a variety of agents (Figs. 3 and 4). The ATPe-induced transient increase in K+ current is calcium dependent, confirmed by pretreatment of the monolayers with thapsigargin, which completely ablated the ATPe-stimulated increase in K+ current (Fig. 5). The exact identity of the Ca2+-activated K+ channel involved was investigated using specific inhibitors of the three known subtypes of Ca2+-activated K+ channels (BK, IK, and SK). Apamin was the only inhibitor tested to significantly reduce the ATPe-induced Cl− secretory response in intact monolayers, promoting a role for the small-conductance Ca2+-activated K+ channels (SK) (Fig. 6). Apamin, an 18 amino acid peptide from bee venom, is a widely used SK channel blocker (11, 39). It is a potent and selective inhibitor of KCNN1, KCNN2, and KCNN3 channels. Significantly ∼35% of the ATPe-stimulated secretory current is insensitive to apamin suggesting that other Ca2+-activated K+ channel subtypes may be present and are activated by ATPe. The presence of SK channel KCNN3 was confirmed by RT-PCR and immunohistochemistry (Figs. 7 and 8). Despite 35% of the current remaining postapamin, we found no evidence for expression of KCNN4. To the best of our knowledge these results present the first evidence for the expression and functional role for basolateral KCNN3 Ca2+-activated K+ channels in oviduct epithelial cells. Previously other groups have reported the presence of SK1 and SK4 mRNA in a human bronchial cell line and linked the activity of the apamin-insensitive SK4 gene product to calcium-activated chloride secretion in the apical and basolateral membranes of these cells (5). Our data clearly suggests two families of basolateral potassium channel are affected by ATPe, Ca2+-activated channels (apamin sensitive and insensitive), and Ca2+-insensitive K+ channel(s) inhibited by ATPe challenge. The existence of two distinct basolateral K+ channel families in polarized epithelia is not new: it has been reported in the airway (29) and colon (13, 27). However, the finding that basolateral KCNN3 is a likely candidate in supporting anion secretion in epithelial is novel and an area requiring further investigation. Clearly ATPe has a dual effect of rapid transient activation of K+ current (KCNN3 mediated) followed by a sustained inhibition of IK. A similar effect of ATPe has been reported previously in mouse β cells where interactions with different P2 receptors can result in different effects on intracellular signaling and the movement of K+ (33). ATPe binding to P2YR4 and/or P2YR6 can inhibit KATP channels (35), whereas binding to P2YR2 results in increased intracellular calcium (3). Previously, we have shown that BOECs express both P2YR2 and P2YR4 (24). Thus the different responses to ATPe may reflect distinct ATPe receptor subtypes in BOECs activating alternate signaling pathways.
The present study clearly identifies and emphasizes the central role of basolateral K+ channels in mediating Cl− secretion by oviduct epithelial cells and demonstrates that basolateral Ca2+-activated K+ channels are key targets of purinergic regulation of oviduct epithelial Cl− secretion. Understanding the specific mechanism of Cl− secretion is likely to have implications in disease states associated with dysregulated epithelial electrolyte and fluid transport and may prove useful in developing therapeutic treatment strategies across a range of epithelia. For example, pelvic inflammatory disease (PID), which is commonly caused by Chlamydia tachomatis infection, is associated with severe tubal damage, infertility, hydrosalpinx, and ectopic pregnancy. The production of hydrosalpinx fluid (HF) and its reflux into the uterine cavity has been reported to inhibit implantation. One of the most common symptoms seen in PID is tissue edema associated with inappropriate fluid production; however, the mechanism underlying HF formation and postinfection edema are to a large extent unknown. In uterine tissues, C. trachomatis infection upregulates CFTR expression and have also reported increased mRNA expression of CFTR in human hydrosalpinx (2). The mechanism involved in the unregulated CFTR expression and fluid secretion following infection has yet to be determined. However, it is likely the release of inflammatory mediators could account for the upregulation of CFTR. For example, tumor necrosis factor α has been previously shown to upregulate CFTR (9). This increase in anion channel expression and the release of the inflammatory mediator ATPe activating basolateral K+ channels could be a key element in the pathogenesis of this disease and furthermore provide a possible therapeutic target for sequalae associated with PID.
This work was partly funded by the Millennium Research Fund National University of Ireland Galway.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: N.K. and L.R.Q. conception and design of research; N.K. and L.R.Q. performed experiments; N.K. and L.R.Q. analyzed data; N.K. and L.R.Q. interpreted results of experiments; N.K. and L.R.Q. prepared figures; N.K. and L.R.Q. drafted manuscript; N.K. and L.R.Q. edited and revised manuscript; N.K. and L.R.Q. approved final version of manuscript.
- Copyright © 2012 the American Physiological Society