|
|
||||||||
Articles |
a Department of Physiology, Faculty of Medicine, Chinese University of Hong Kong, Shatin, N.T. Hong Kong, China
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Rat caudal epididymal fluid contains a high concentration (40 mM) of K+ ions [5, 6]. High luminal K+ concentration has been shown to inhibit sperm motility during sperm storage and maturation [7]. This high K+ content is mainly caused by the absorption of water in the epididymis and probably in part by the active secretion of K+ from the apical membrane [5]. More direct study showed that adrenaline stimulated 86Rb+ efflux across the apical membrane of monolayer cultures of the rat epididymis [8]. K+ channels have been identified on the apical membrane of rat epididymis [1] and human vas deferens [4] in primary culture. Ion channels were also recorded in the apical membrane of the isolated rat caudal epididymis [9]. In addition, K+ secretion may involve apical K+ channels in other epithelia, e.g., mammalian salivary and lacrimal glands [10] and renal cortical collecting duct [1113].
In the present study, we used the patch-clamp technique to record K+ channels directly from the apical surface of the epithelial cells in the freshly isolated caudal epididymal duct. We investigated the general characteristics of the K+ channels of ~54 pS and demonstrated that extracellular tetraethylammonium ions (TEA+) and intracellular ATP, but not glibenclamide, inhibited channel activity. The tissue preparation employed in this study offers one important advantage compared to primary cell culture, i.e., a KCa channel can be precisely identified from the apical membrane of epithelia. The existence of KCa channels and possibly other K+ channels implies that they might play a role in maintaining a high K+ level along the epididymal tubule in a regulated manner. This high content of K+ is required for sperm storage in the epididymis [7].
| MATERIALS AND METHODS |
|---|
|
|
|---|
Male Sprague-Dawley rats (400 g; Animal Resource Center, Canning Vale, Australia) were anesthetized with pentobarbital (40 mg/kg BW). The lower part of the abdomen was cut open to expose the vas deferens and caudal epididymis. The content of the caudal epididymis was flushed out with Krebs-Henseleit solution by cannulation of the caudal tubule and vas as previously described [14, 15]. A segment 5 mm in length was dissected from the caudal epididymal duct and slit open longitudinally. The flat tissue was mounted on a perspex platform (made by the Technical Support Unit, Chinese University of Hong Kong) with the luminal surface facing upward and was affixed in place with two small platinum bars. Recordings were made from the luminal side of the epithelial layer.
Single-Channel Recording
Single ionic channels were investigated with the patch-clamp technique [16] using the cell-attached and inside-out configurations. Patch pipettes were prepared from 1.5-mm o.d. borosilicate glass capillaries on a micropipette puller (Sutter Instrument Co., Novato, CA), coated with Sylgard elastomer, and fire-polished. The pipettes were filled with electrolyte, and tip resistance ranged between 10 and 20 M
. Seals were over 10 G
. Single channels were recorded using Axopatch 1D (Axon Instruments Inc., Foster City, CA) amplifier, stored in either a VCR or computer using pClamp software (5.51 version; Axon Instruments). According to convention, membrane voltages were defined as inside relative to outside, and outward K+ current was represented by an upward deflection of current when membrane potential was held positive to the equilibrium potential estimated from the Nernst equation.
Recording Solutions
For experiments using the inside-out configuration, the epithelial layer was bathed in a K+-rich solution comprising (in mM) 10 NaCl, 140 KCl, 1 MgCl2, 5 EGTA, 10 Hepes, pH = 7.2, giving a desired free Ca2+ concentration with appropriate concentration of CaCl2 (calculated by EQCAL; Biosoft, Cambridge, UK). The pipette-filling solution contained (in mM) 140 NaCl, 5.5 KCl, 2 MgCl2, 2.5 CaCl2, 0.5 KH2PO4, 4.2 NaHCO3, 10 Hepes, 11.1 glucose, pH = 7.4. For the ion-replacement experiments, internal K+ was replaced with an equimolar concentration of Na+ to maintain a constant ionic strength. The osmolarity of bath and pipette solutions was adjusted to 300 mOsm/L with mannitol.
Data Collection and Analysis
Single-channel data were stored on videotape and then played back at a filtering rate of 1 kHz before computer acquisition at 10 kHz. Data analyses were performed on pClamp software. Amplitude histograms were made after separation of open and closed states of the channels using the threshold set at half of the open-channel amplitude. A nonlinear Levenberg-Marquardt least-squares curve-fitting procedure was used for fitting Gaussian curves. The open state probability (Popen) was calculated as the total open time divided by the total observation time. NPopen (N, the number of the channels in the patch estimated from the maximum number of simultaneous openings; Popen, the open state probability of an individual channel) was calculated over 90 sec as
Nj=1 tjj/TN, where tj is the time spent with j = 1,2, ...N channels open.
Since EK and ENa are opposite in polarity and ECl is approximately zero in the symmetric Cl- solutions used in the present study, the relative permeability (PK/PNa) was estimated using one of two limiting theoretical formulas [17]:
![]()
where PK and PNa are the permeabilities of a channel to K+ and Na+. RT/F = 25.43 mV at 22°C.
To examine the inhibitory effect of internal ATP, the concentration dependence of ATP-induced inhibition of the channel activity can be estimated using the following equation:
![]()
where Pc and Pb are the NPopen in the absence and presence of internal ATP. KD is the dissociation constant for ATP inhibition, and n is the Hill coefficient.
Chemicals and Drugs
The following drugs were used in the present study: ATP (Na2-ATP), tetraethylammonium ions, diazoxide, and glibenclamide (Sigma Chemical Co., St. Louis, MO). Diazoxide and glibenclamide were dissolved in dimethyl sulfoxide. ATP stock was prepared in an intracellular K+-rich solution containing 10 µM Ca2+ and pH was adjusted to 7.2.
| RESULTS |
|---|
|
|
|---|
). A single group of ion channels with intermediate conductance was detected. However, we were unable to make either whole-cell current recordings or outside-out patches from the apical membrane of the freshly isolated epithelial sheet in the rat caudal epididymis. Intermediate-Conductance Ca2+-Activated K+ Channels
Active channels were seldom seen in the cell-attached patches. After excision, several channels were usually active in the patches with the cytoplasmic side of the membrane facing 1 µM Ca2+. Figure 1A shows single-channel current traces recorded at a range of membrane holding potentials. In this patch the potential at which channel opening first appeared was -20 mV. Figure 1B shows the current-voltage relationship of the single channel. The single-channel conductance at 0 mV was 54.2 ± 2.8 pS (n = 11), assuming that the reversal potential was -80 mV estimated from the Nernst equation.
|
The K+ selectivity of the channel in inside-out patches was examined by changing the K+ concentration in the bath. Figure 2 shows the current-voltage relationship with two different K+ concentrations in the bath. For these experiments KCl was replaced with an equimolar amount of NaCl, and [Cl-] was symmetric. The conductance at 0 mV at 140 mM KClin/6 mM KClout was 51.6 ± 4.6 pS (n = 4), and the reversal potential of -86.3 ± 5.7 mV (n = 4) was extrapolated from current-voltage relation fitted to the Goldman-Hodgkin-Katz equation. When the K+ concentration in the bath was reduced by half (70 mM), the conductance at 0 mV was 29.4 ± 4.2 pS (n = 4) and the reversal potential was -61.7 ± 4.8 mV (n = 4). The reversal potentials were close to the calculated values from the Nernst equation at the given K+ concentration gradients. A mean value for PK/PNa of 171 ± 24 (n = 4) was calculated from equation 1.
|
Internal Ca2+ Sensitivity of the Channel
The K+ channels recorded from the apical membrane of the isolated rat epididymal tubule displayed modest sensitivity to internal Ca2+. This is demonstrated in original current recordings shown in Figure 3A. Occasional openings of a channel in this patch were seen at 10 nM [Ca2+]i (NPopen = 0.0005). When [Ca2+]i was increased to 10 µM, NPopen increased to 0.951. The Ca2+ sensitivity was reversible. Figure 3B shows the effect of membrane potential on NPopen of the channels at various [Ca2+]i. NPopen shows a sigmoid voltage dependence with increasing NPopen upon depolarization. NPopen at low [Ca2+]i (10 nM) in the inside-out patches was similar to that measured in the cell-attached patches. Increase in [Ca2+]i caused a leftward shift in the NPopen-voltage curve. At high [Ca2+]i (10 µM), NPopen increased even at negative potentials and reached a maximum at 0 mV but decreased at potential more positive than +20 mV.
|
Effect of External TEA+
The blocking effect of external TEA+ was investigated using an inside-out configuration. Figure 4A shows single-channel currents recorded at 0 mV with 10 µM [Ca2+]i in the absence and presence of external TEA+. TEA+ at 0.1 mM, when included in the pipette solution, induced a flickery block of the single K+ channel current. It is apparent that TEA+ frequently interrupted long openings. TEA+ blocked the channel by reducing the single-channel current amplitude (4.52 pA in control and 2.83 pA at 0.1 mM TEA+, Fig. 4B), and open-channel noise increased in comparison with that of the control trace. The increased noise level in the open channel was caused by the partial time resolution of the channel activity. The mean current-voltage relationship in control patches and in patches with 0.1 mM TEA+ is shown in Figure 4C. In four experiments with 0.5 mM TEA+ included in the pipette solution, the single-channel current amplitude at 0 mV was reduced from 4.32 ± 0.19 pA (n = 5) in the control to 1.45 ± 0.13 pA (n = 4). The remaining channel activity in the presence of TEA+ was completely eliminated by addition of 0.5 mM EGTA. By contrast, when included in the pipette solution, glibenclamide (10 µM) did not affect the activity of K+ channels with 10 µM [Ca2+]i. Mean NPopen and single-channel current amplitude at 0 mV were 1.42 ± 0.32 and 4.82 ± 0.43 pA (n = 3), respectively. These values were not different from control values (NPopen: 1.34 ± 0.18, n = 5; mean amplitude: 4.50 ± 0.23 pA, n = 11; p > 0.05 in unpaired data).
|
Effect of Internal ATP on the Channel
Figure 5 shows the inhibition of these intermediate-conductance K+ channels by internal ATP. In the inside-out patch shown in Figure 5A, ATP induced a concentration-dependent reduction of NPopen by reducing the mean open time and number of active channels in the patch. Diazoxide at 100 µM reactivated the channel that had been inhibited by ATP. In four patches, 2 mM ATP decreased NPopen (2 min measurement) from a mean control value of 1.589 ± 0.252 to 0.432 ± 0.21, and diazoxide (100 µM) increased the NPopen to 1.18 ± 0.28. However, the single-channel conductance was unaffected (4.21 pA in control, 4.08 pA in 2 mM ATP, and 4.43 pA in diazoxide, measured for the first opening level at 0 mV). Figure 5B shows NPopen as a function of concentration of intracellular ATP and diazoxide. In another inside-out patch containing only one active channel, mean NPopen was calculated at 10-sec intervals and plotted against the time in seconds shown in Figure 5C. In the control at 10 µM [Ca2+]i, NPopen at 0 mV was close to 1. ATP at 2 mM completely inhibited the channel activity when applied to the bath. A lag of 12 min for ATP inhibition was probably due to a slow diffusion process, and a similar delay was seen in Ca2+ activation of the channel. Figure 6 shows the dependence of Pb/Pc on ATP concentration at 0 mV. The points are closely fitted by equation 2 with a KD of 1.188 ± 0.342 mM (fitted by a least-squares algorithm to data points, n = 4). This KD value is much greater than that for ATP-sensitive K+ channels in rat ß-pancreatic cells (IC50 = 15 µM [18]), in frog skeletal muscle (IC50 = 135 µM [18]), and in rat central neurones (KD < 1 mM, [19]).
|
|
| DISCUSSION |
|---|
|
|
|---|
TEA+ has been shown to inhibit KCa channels in a number of epithelial cells [3, 21]. TEA+ when included in recording pipettes caused a flickery block of single-channel current through K+ channels in the apical surface of the epididymal epithelium. The blocking and unblocking events were too fast to be resolved. The external TEA+ block appeared as a reduction in open-channel unitary current in a range of holding membrane potentials. TEA+ at 100 µM reduced amplitude of the channel current by approximately 43%, while the KD values for a 50% reduction in single-channel current for external TEA+ block of KCa channels in chromaffin cells and smooth muscle range between 100 and 300 µM [22, 23].
Internal ATP inhibited the intermediate-conductance K+ channels in a concentration-dependent fashion. ATP reduced the NPopen without effect on the single-channel current amplitude. ATP has been found to be an effective inhibitor for the ATP-sensitive K+ channels in many types of cells including epithelia [2, 18, 24, 25]. In non-epithelial cells, the ATP-sensitive K+ channels and KCa channels have distinct pharmacological properties [23]. We did not detect any difference in the single-channel current amplitude and voltage dependence of ATP-inhibitable channels from the K+ channel activated by [Ca2+]i. In view of the important role of intracellular ATP in ion transport on epithelial membranes, it is possible that intracellular ATP may be a modulator for the K+ channel in epithelia. Similar findings were reported in airway smooth muscle, where cytoplasmic ATP inhibited the large-conductance KCa channel with Ki of 0.2 mM [26]. In three recordings, diazoxide, a K+ channel opener, reactivated the channel previously inhibited by ATPi. Diazoxide has been found previously to promote activity of KCa channels [27]. Furthermore, glibenclamide, a selective blocker for ATP-sensitive K+ channels, failed to influence channel activity. These results indicate that internal ATP might not act at ATP-sensitive K+ channels.
We recorded the intermediate-conductance K+ channels in about 6% of a total of 432 membrane patches excised from the apical membrane of freshly isolated epithelia from the rat caudal epididymis. This indicates that there is a low density of active K+ channels situated on the apical membrane and that the activity of apical K+ channels may be too low to make a significant contribution to membrane K+ conductance and basal K+ secretion in unstimulated cells. In contrast, a higher success rate of channel detection was reported in cultured epithelial cells from the rat epididymis and human vas deferens [1, 4]. This discrepancy in channel density might be a result of differing treatment of the epithelia. However, the K+ channel observed in this study is modestly activated by increasing [Ca2+]i above 100 nM. Many intracellular Ca2+-mobilizing agonists would readily raise [Ca2+]i above this level and promote channel activity.
In a general model for electrogenic Cl- secretion in the epithelia, the excess intracellular K+ brought in by Na-K-2Cl symport must be removed to maintain an environment favorable for Cl- exit through the apical membrane down the electrochemical gradient. Both basolateral and apical K+ channels should provide effective routes for K+ efflux if they are subjected to similar regulation. Marty et al. [10] proposed a model of fluid secretion in mammalian salivary and lacrimal glands based on their work on the stimulatory effect of Ca2+-mobilizing agonist on large-conductance KCa channels. During the initial phase of secretion, both Ca2+-activated K+ and Cl- channels are activated, and K+ and Cl- ions leave the cell on both apical and basolateral membranes through the respective channels. During the steady state phase, ion extrusion occurs only on the apical side because Cl- efflux through the basolateral membrane is inverted and basolateral K+ channels are assumed to be inhibited. At the apical membrane, entry of NaCl leads to activation of the Na-K pump and subsequent entry of KCl into the cell [10]. This model clearly indicates a significant role of apical K+ channels in fluid secretion, and it also explains the K+-rich secretions found both in salivary [28, 29] and lacrimal [30] glands in response to low levels of agonist stimulation. Moreover, a positive role for the apical K+ channels in Cl- secretion has been indicated through computer modeling of the epithelial cells [31]. Placing 1020% of the total cell K+ conductance on the apical membrane would increase the fluid secretion driven by secondary active transport of Cl- [31]. K+ secretion may be another function of the apical K+ channel. The model presented by Cook and Young [31] predicts that apical K+ channels do not produce an especially K+-rich primary fluid and that the luminal K+ concentration is not much higher than 40 mM. This value is in agreement with that calculated by Marty et al. [10]. During sustained fluid secretion in exocrine glands, the likely luminal K+ concentration is 40 mM. Passive excretion of KCl requires that intracellular K+ concentration remain high (140 mM) in order for EK to be more negative than ECl on the apical membrane [10]. Actually, the primary fluid K+ concentrations are all in this range in micropuncture experiments on various exocrine glands [32] and in rat caudal epididymis [15, 33]. In view of the high luminal K+ concentration (40 mM) measured at the caudal portion of the rat epididymis [15, 33], it is possible that the K+ channel may play a part in K+ secretion through the apical membrane, assuming that the amount of K+ released from spermatozoa is insignificant. Indeed, adrenaline caused an increase in the rate of 86Rb+ efflux from the apical membrane of cultured rat epididymal epithelium [8] and activated the apical K+ channels in primary culture [1]. In addition, noradrenaline increased intracellular Ca2+ concentration in the same preparation [34]. However, K+ secretion through the apical membrane is likely to be regulated by secretagogues, since activity of the channel is extremely low at the resting membrane potential.
In summary, we detected an intermediate-conductance KCa channel on the apical membrane of the rat caudal epididymis. The K+ channels were activated by internal Ca2+ and membrane depolarization but inhibited by external TEA+ and internal ATP. These channels may be involved in both fluid secretion and K+ secretion. The functional role of these channels needs to be further investigated.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence. FAX: 852 2603 5022; yu-huang{at}cuhk.edu.hk ![]()
Accepted: February 1, 1999.
Received: July 30, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. W. Carlin, R. L. Sedlacek, R. R. Quesnell, F. Pierucci-Alves, D. M. Grieger, and B. D. Schultz PVD9902, a porcine vas deferens epithelial cell line that exhibits neurotransmitter-stimulated anion secretion and expresses numerous HCO3- transporters Am J Physiol Cell Physiol, June 1, 2006; 290(6): C1560 - C1571. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-X. Wang, M. Ikeda, and W. B. Guggino The Cytoplasmic Tail of Large Conductance, Voltage- and Ca2+-activated K+ (MaxiK) Channel Is Necessary for Its Cell Surface Expression J. Biol. Chem., January 17, 2003; 278(4): 2713 - 2722. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |