HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 294/4/C879    most recent 00490.2007v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lee, W.-K. Articles by Thévenod, F. Search for Related Content PubMed PubMed Citation Articles by Lee, W.-K. Articles by Thévenod, F.

MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS

Evidence for KCNQ1 K⁺ channel expression in rat zymogen granule membranes and involvement in cholecystokinin-induced pancreatic acinar secretion

¹Department of Physiology and Pathophysiology, University of Witten/Herdecke, Witten, Germany; and ²Institute of Anatomy and Cell Biology, Department of Molecular Embryology, University of Freiburg, Freiburg, Germany

Submitted 17 October 2007 ; accepted in final form 18 January 2008

	ABSTRACT

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

 
Secretion of enzymes and fluid induced by Ca²⁺ in pancreatic acini is not completely understood and may involve activation of ion conductive pathways in zymogen granule (ZG) membranes. We hypothesized that a chromanol 293B-sensitive K⁺ conductance carried by a KCNQ1 protein is expressed in ZG membranes (ZGM). In suspensions of rat pancreatic ZG, ion flux was determined by ionophore-induced osmotic lysis of ZG suspended in isotonic salts. The KCNQ1 blocker 293B selectively blocked K⁺ permeability (IC₅₀ of 10 µM). After incorporation of ZGM into planar bilayer membranes, cation channels were detected in 645/150 mM potassium gluconate cis/trans solutions. Channels had linear current-voltage relationships, a reversal potential (E_rev) of –20.9 ± 0.9 mV, and a single-channel K⁺ conductance (g_K) of 265.8 ± 44.0 pS (n = 39). Replacement of cis 500 mM K⁺ by 500 mM Na⁺ shifted E_rev to –2.4 ± 3.6 mV (n = 3), indicating K⁺ selectivity. Single-channel analysis identified several K⁺ channel groups with distinct channel behaviors. K⁺ channels with a g_K of 651.8 ± 88.0 pS, E_rev of –22.9 ± 2.2 mV, and open probability (P_open) of 0.43 ± 0.06 at 0 mV (n = 6) and channels with a g_K of 155.0 ± 11.4 pS, E_rev of –18.3 ± 1.8 mV, and P_open of 0.80 ± 0.03 at 0 mV (n = 3) were inhibited by 100 µM 293B or by the more selective inhibitor HMR-1556 but not by the maxi-Ca²⁺-activated K⁺ channel (BK channel) inhibitor charybdotoxin (5 nM). KCNQ1 protein was demonstrated by immunoperoxidase labeling of pancreatic tissue, immunogold labeling of ZG, and immunoblotting of ZGM. 293B also inhibited cholecystokinin-induced amylase secretion of permeabilized acini (IC₅₀ of 10 µM). Thus KCNQ1 may account for ZG K⁺ conductance and contribute to pancreatic hormone-stimulated enzyme and fluid secretion.

exocytosis; acinar cell; secretory granule; planar bilayer

In pancreatic acinar cells, secretion of the primary fluid occurs by Na⁺-coupled secondary active Cl^– transport (41), which can be switched on and off depending on demand. This is accomplished by cytosolic Ca²⁺ signals elicited by neurotransmitter (e.g., ACh) or hormone interaction (e.g., CCK) with receptor sites on the surface of the acinar cell membrane (57). An increase of cytosolic Ca²⁺ concentration evoked by the secretagogues ACh or CCK also leads to the fusion of ZG with the apical plasma membrane followed by release of digestive enzymes into the lumen and affects Ca²⁺-sensitive docking and fusion proteins (24), which interact at different stages of the exocytotic process to promote ZG fusion with the plasma membrane and the release of the granule contents into the lumen.

In conjunction with exocytosis, granule membrane components are also inserted into the apical plasma membrane. Thus this process may also lead to the insertion of regulated ion channels expressed in the membrane of ZG into the apical plasma membrane. Experiments, which were carried out on digitonin-permeabilized rat pancreatic acini (14), showed that enzyme secretion evoked by ACh or CCK was critically dependent on the presence of Cl^– and K⁺ in the cytosol. Furthermore, enzyme secretion could be abolished by application of Cl^– and K⁺ channel blockers (14). These results led to the hypothesis that some of the physiological targets in the cascade of Ca²⁺-dependent events leading to secretion of primary fluid and enzymes are anion and cation channels residing in the ZG membrane (48). This model has gained further support by more recent electrophysiological studies on mechanisms of regulated exocytosis in pancreatic - (22) and β-cells (2).

In line with this hypothesis, ZG isolated from rat pancreatic acinar cells undergo osmotic swelling and lysis due to the presence of conductive pathways for both monovalent cations and anions in ZG membranes (for review, see Ref. 48). Several candidate anion channel proteins have been identified so far, which are expressed in the membrane of pancreatic ZG. A Ca²⁺-activated anion channel is permeated by HCO₃^– and blocked by DTT and 4,4'-diisothiocyanatodihydrostilbene-2,2'-disulfonate (DIDS) (52). Activation of this channel may promote enzyme release by increasing the dissolution of the condensed granule contents. In addition, a DIDS-sensitive ClC-2 Cl^– channel is also expressed in ZG membranes (7).

The molecular identity of two monovalent cation conductive pathways, which contribute to K⁺ fluxes into ZG, is less clear. A nonselective monovalent cation conductive pathway is equally permeable to K⁺ and Na⁺ and selectively blocked by flufenamate (49). A K⁺ conductive pathway is blocked by ATP and glyburide (49, 50) and thus appears to be similar to ATP-sensitive K⁺ channels (23, 54). KCNJ11/K_ir6.2, the pore-forming subunit of ATP-sensitive K⁺ channels, is not expressed in rat pancreatic acini (46), although KCNJ8/K_ir6.1 has been recently detected in rat pancreatic ZG by immunoblotting (25). KCNQ1/K_vLQT1 may be another candidate for the ZG K⁺ channel. KCNQ1 (K_VLQT1; Kv7.1) is a very low-conductance (<1.5 pS), voltage-gated K⁺ channel distributed widely in epithelial and nonepithelial tissues (for review, see Refs. 4, 36). KCNQ1 was found to be mutated in the hereditary cardiac disease "long QT syndrome 1." In the heart and inner ear, KCNQ1 coassembles with a β-subunit KCNE1 (IsK, minK) to form the outwardly rectifying and slowly activating, low-conductance K⁺ channel current (I_Ks). I_Ks are selectively blocked by chromanol 293B, which binds to KCNQ1 (45, 59). In situ hybridization studies have shown that KCNQ1 is expressed in rodent pancreatic acini (12). Moreover, KCNQ1, which is the apical K⁺ channel required for active K⁺/H⁺ exchange and stimulated HCl secretion by gastric parietal cells, has also been detected in the intracellular tubulovesicles of parietal cells (18). In epithelial tissues, KCNQ1 K⁺ channels are activated by cAMP or Ca²⁺ and are selectively inhibited by chromanol 293B (4, 12, 18, 20, 29, 30, 45, 56).

In the present study, we tested the hypothesis that a 293B-sensitive K⁺ conductance carried by a KCNQ1 protein is expressed in rat pancreatic ZG membranes.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

 
Materials. L--Phosphatidylcholine from soybean (asolectin) type II-S (P5638), protease inhibitor cocktail, Percoll, valinomycin, CCCP, collagenase (type III from Clostridium histolyticum, 790 U/mg), CCK octapeptide (CCK-OP), BSA (98–99%, essentially fatty-acid free), charybdotoxin, and goat anti-rabbit IgG coupled to 10-nm gold particles were purchased from Sigma (Deisenhofen, Germany). Trypsin inhibitor (from hen egg white) was obtained from Boehringer (Mannheim, Germany). Phadebas amylase kit was from Phadia (Freiburg, Germany). Digitonin was from Fluka (Seelze, Germany). Glutaraldehyde, osmiumtetroxyde, and Araldit resin were purchased from Plano (Wetzlar, Germany). Horseradish peroxidase (HRP)-conjugated donkey anti-rabbit IgG was from GE Healthcare (Little Chalfont, Buckinghamshire, UK), and goat anti-rabbit IgG coupled to HRP was obtained from Dako (Hamburg, Germany). Chromanol 293B and HMR-1556 were gifts from Sanofi-Aventis, Germany. All other reagents were of the highest analytical grade available.

Isolation of ZG and purification of ZG membranes. ZG from rat exocrine pancreas were isolated as described in detail elsewhere (6, 50, 52). Briefly, male Wistar rats (180–200 g, fasted overnight) were humanely killed by CO₂ anesthesia according to the ethical guidelines of the German state on animal welfare (approval no. 50.8735.1 Nr99/4).

Pancreatic tissue was immediately removed, homogenized by nitrogen pressure cavitation, and centrifuged on a self-forming Percoll gradient. ZG at the bottom of the centrifuge tube were washed in isotonic buffer containing 50 mM sodium succinate for removal of mitochondrial contamination before use (51). To obtain ZG membranes, a standard protocol was applied (50, 51) with slight modifications: ZG were resuspended on ice in a hypotonic lysis buffer (50 mM HEPES-Tris·HCl, 0.1 mM MgSO₄, pH 7.0) containing a protease inhibitor cocktail. The ZG were disrupted at 0°C with the Branson Sonifier ultrasonic cell disruptor S250A using a microtip probe (3 x 5-s pulses at 10 A), lysed by freezing at –80°C followed by rapid thawing, and centrifuged for 30 min at 196,000 g. The membrane pellet was either resuspended in lysis buffer for immunoblotting or in 645 mM KCl for planar lipid bilayer experiments. Protein concentration was determined according to Bradford (5).

Measurement of ion conductive pathways of rat pancreatic ZG. Ion conductive pathways of rat pancreatic ZG were assayed according to previously reported protocols (49, 50). The assay for ion conductive pathways of pancreatic ZG relies on the measurement of osmotic lysis of ZG resuspended in buffered isotonic salt solutions, which occurs after maximal permeabilization of the ZG membranes with electrogenic ionophores for counterions. Lysis is therefore limited by ion fluxes through endogenous conductive pathways. Granule lysis causes a decrease in absorbance of the suspension, which is measurefd at wavelength 540 nm. Measurements were carried out at 37°C in a Beckman DU-640 spectrophotometer equipped with a Peltier constant-temperature chamber and an automatic six-unit sampler. Data were captured and converted with DU-WinConnection Suite software.

For measurement of Cl^– conductive pathways, ZG were suspended in 150 mM KCl, 5 mM MgSO₄, and 50 mM HEPES (adjusted to pH 7.0 with Tris). Granule lysis was measured after addition of 5 µM valinomycin, an electrogenic ionophore that maximally and selectively permeabilizes granule membranes for K⁺. The influx of salt and water causes ZG lysis and is limited under these conditions by the permeation of Cl^– through endogenous conductive pathways (10).

To determine K⁺ or Na⁺ conductive pathways (the latter being a measure of nonselective monovalent cation conductive pathways) (49), ZG were suspended in 150 mM potassium or sodium acetate containing 5 mM MgCl₂ and 1 mM EDTA and buffered with 50 mM imidazole (pH 7.0, adjusted with acetic acid). Because the intragranule pH is 6.5 (50), an inside-to-outside directed H⁺ concentration gradient of 0.5 pH units is generated across the granule membrane. After maximal permeabilization of the granule membrane to H⁺ by addition of the electrogenic protonophore CCCP (16 µM), the H⁺ concentration gradient is converted to an inside-negative H⁺ diffusion potential. This in turn drives K⁺ or Na⁺ influx through endogenous K⁺ and/or nonselective cation permeabilities (48, 49). Anion influx occurs through the uncharged molecule acetic acid, which permeates through the lipid membrane by nonionic diffusion and dissociates, thus continuously providing the intragranule space with protons for protonation of imidazole as well as for proton efflux from the acidic interior. Under these conditions, influx of monovalent cations through endogenous K⁺ and nonselective cation permeabilities is rate limiting for bulk salt influx into the intragranule space as well as for the resulting granule lysis.

Lysis rates were expressed as inverse half times of lysis, which were considered proportional to the lysis rate constants. Half time of granule lysis was estimated from the slope of decrease in absorbance with time between addition of ionophore and either experimental half time or the entire observation period if the half time was not reached. The slope of absorbance change with time was calculated by linear regression of the digitized data. At most, linear regression is an approximation of the actual data, but it has proved to be a very reliable and established approach for quantitative analysis of the kinetics of osmotic lysis in ZG, assuming that only one population of ZG is present. Stock solutions of chromanol 293B or HMR-1556 were dissolved in DMSO, which was added in the same concentration (0.1%) to the control cuvettes.

Isolation of pancreatic acini and assay for amylase release. The preparation of dispersed acini by collagenase digestion has been previously described (1). The pancreas from one male Wistar rat (130–160 g) was trimmed free of fat and connective tissue, minced to a fine paste, and suspended in 10 ml of digestion buffer containing (in mM) 130 NaCl, 2 MgCl₂, l CaCl₂, 5 KC1, 1.2 KH₂PO₄, 0.01% trypsin inhibitor (wt/vol), 0.2% BSA (wt/vol), 10 glucose, and 20 HEPES (pH 7.4). Collagenase digestion was carried out in two steps. Tissue was first treated with 600 U collagenase for 15 min and then with 900 U for 30 min at 37°C under a continuous supply of 100% O₂. After digestion, the tissue was pipetted through fire-polished 5-ml tips with decreasing end diameter. Acini were then purified by centrifugation in 4% BSA (wt/vol) and resuspended, and a final washing step was performed in the same buffer that is used for the secretion assay. Intact acini were incubated with 100% O₂ in a standard secretion buffer of the following composition (in mM): 135 KCl, 20 HEPES, 1.2 KH₂PO₄, 2 MgCl₂, 0.1 CaCl₂, 0.01% trypsin inhibitor (wt/vol), 0.2% BSA (wt/vol), and 10 glucose, pH titrated to 7.4 with KOH. To eliminate the contribution of plasma membrane monovalent cation transport pathways to amylase secretion, the plasma membrane was permeabilized by addition of 5 µg/ml digitonin to aliquots of acini in 2 ml of the standard secretion assay buffer for 10 min. Acini were subsequently incubated with a Ca²⁺-dependent secretagogue CCK-OP (1 nM). After permeabilization, test substances or solvents (0.1% vol/vol) were preincubated for 5 min before stimulation with CCK-OP. Samples of 200-µl acinar suspension were taken shortly before and 30 min after addition of CCK-OP and centrifuged for 30 s at 14,000 rpm in an Eppendorf microfuge. The supernatant was removed, and amylase release from acini was determined using the Phadebas amylase test kit. To determine total amylase activity, portions of the residual samples were used. Acini were lysed in "diluent solution" containing 10 mM Na₂HPO₄-10 mM NaH₂PO₄ buffer (pH 7.8), 0.1% SDS (wt/vol), and 0.1% BSA (wt/vol) for 1 h and further sonicated for 10 min. Amylase release was expressed as the percentage of total amylase activity present in the pancreatic acinar suspension. The percentage of amylase released during a period of 30 min in the absence of CCK-OP was subtracted for each value.

L--Phosphatidylcholine purification. L--Phosphatidylcholine for the planar lipid bilayer technique was purified from asolectin by solvent extraction and purification procedures (19). Asolectin (2 g) from soybean was dissolved in 50 ml of chilled chloroform and mixed for 1 h at 4°C. The asolectin was subsequently precipitated with a fivefold excess of ice-cold acetone and mixed for a further 1 h at 4°C. The purified lipid was aliquoted and collected by centrifugation at 1,100 g for 15 min at 4°C. Excess acetone was decanted, and L--phosphatidylcholine was stored at –80°C under N₂. Upon reconstitution, purified L--phosphatidylcholine was weighed, dried under N₂ to remove any remnants of acetone, weighed again, and dissolved to a final concentration of 25 mg/ml in n-decane.

Planar lipid bilayer technique. The planar lipid bilayer technique was set up as described elsewhere (19, 34). A planar lipid membrane was formed by spreading phospholipid dispersions on a 250-µm diameter hole, which separates two chambers (cis and trans with 1 ml of internal volume) (Hugo Sachs Elektronik, Harvard Apparatus, Hugstetten, Germany). The cis compartment, defined as the compartment to which ZG membrane vesicles were added, contained 645 mM KCl or potassium gluconate and 10 mM HEPES (pH 7.2 adjusted with Tris). The trans compartment, connected to the virtual ground of the amplifier, contained 150 mM KCl or potassium gluconate and 10 mM HEPES (pH 7.2). For ion selectivity experiments, 500 mM potassium gluconate was replaced by 500 mM sodium gluconate in the cis chamber only. Nonpolarizing electrodes (Ag/AgCl pellets) immersed in 3 M KCl were used to connect each side of the bilayer to the headstage of a BLM-120 bilayer clamp amplifier (BioLogic, Claix, France) through agar salt bridges saturated with 3 M KCl. The aperture was first thinly coated with lipid solution and air-dried before bilayer formation to improve membrane stability. Planar phospholipid bilayers were painted with purified L--phosphatidylcholine dissolved in n-decane at a concentration of 25 mg/ml. Formation and thinning of the bilayer were monitored by capacitance measurements. Only bilayers with a capacitance >100 pF were used for experiments. Frozen aliquots of ZG membrane vesicles isolated as described above and suspended in 645 mM KCl to a protein concentration of 2.5–5 mg/ml were thawed, and 2–5 µl of vesicle suspension were added to the cis chamber. Unless otherwise indicated, the fusion of vesicles was initiated by applying an osmotic gradient across the planar bilayer (645 mM K⁺ cis-150 mM K⁺ trans), as well as intermittent stirring of the vesicle suspension. Voltage was applied to the cis compartment of the chamber, and the trans compartment was grounded. All current measurements were carried out at 20°C.

Current-voltage relations were determined with a 10-mV voltage step protocol from –70 to +70 mV unless otherwise indicated. Between steps, the membrane potential was returned to 0 mV to ensure proper reopening of the channels. Unitary conductance was calculated from currents obtained by depolarizing voltage steps from –30 to +30 mV. For pharmacological experiments, single-channel currents were recorded at a holding potential varying from –50 to +50 mV.

Data analysis. Signals were filtered at a corner frequency of 0.5 kHz (low-pass Bessel filter 8-pole), digitized with a 12-bit AD/DA converter at a sampling rate of 5 kHz (PowerLab 2/25; ADInstruments, Spechbach, Germany), and transferred to a personal computer for off-line analysis with pCLAMP version 9.2 (Axon Instruments, Union City, CA). To determine single-channel conductance, all-points histograms were generated. The peak amplitude data were grouped into user-defined bins (amplitude steps) to elicit a histogram. Each histogram peak was fitted further by Gaussian distribution fit to determine the mean amplitude of each channel level. To account for the concentration gradient of permeable ions between the cis and trans chambers, channel conductance was determined with the modified current-voltage law (21): I_K = g_K(E – E_K), where I_K = K⁺ current, g_K = K⁺ conductance, E = holding potential, and E_K = reversal potential of K⁺ at 20°C.

For the majority of experiments, more than one channel was reconstituted into the bilayer. With the assumption that the different levels were multiple openings of the same channel type, the open probability (P_open) was computed as sum of total open time x level/total time/number of channels, where number of channels is the number of levels. This is referred to as "simple P_open" in Clampfit (pCLAMP, version 9.2, Axon Instruments; Excel XP, Microsoft, Unterschleißheim, Germany). To calculate accurately the total number of open channels (N), the channel activity from individual experiments was plotted out to allow visualization of open-closed channel transitions. The amplitude of these channel events was then determined manually. N was determined visually from 1-min recordings by counting the maximal number of simultaneously open channels with the same current amplitude at 0 mV. NP_open, the product of N and P_open, was measured using Clampfit. This then permitted derivation of NP_open/N, the single-channel P_open. In the presence of channel blockers, recordings lasted up to 3–4 min. The time constant () for open channel and closed channel transitions were determined by conducting a threshold search for each open channel level. The data were first smoothed using a sum of two exponentials before the time was determined to a single exponential function using the Chebyshev polynomial technique (Clampfit). To calculate , 10% of the trace was removed from both the peak and baseline ends of the rise and decay phases, and the remaining midsection was used for derivation of . Channel dwell times were analyzed by performing a single-channel search using Clampfit. When more than one channel level opening was present, this analysis was conducted on parts of traces when only one channel was open.

Data acquisition, unitary current, dwell time and gating measurements, statistical analysis, and data processing were performed by using commercially available software packages (pCLAMP, version 9.2; and Sigma Plot 8.0, SPSS, Chicago, IL).

Antibodies. The KCNQ1 antibody (BLE 2-1) was obtained by immunizing rabbits with a peptide (CPADLGPRPRVSLDPRVSIY) of the cytosolic NH₂ terminus (residues 67–94 of rat KCNQ1; accession no. CH473953.1) (18, 20). An additional antiserum against KCNQ1 was raised in rabbits against a peptide that represents a part of the COOH terminus of human KCNQ1 (CLTVPQTGPDEGS-OH; residues 658–669; accession no. AY114213.1) (11, 13). A commercial antibody (AB5587-50UL) was raised against a peptide corresponding to residues 661–676 of human KCNQ1 (accession no. P51787), which does not react with other QKT proteins. All KCNQ1 antibodies were affinity purified. Goat anti-rabbit IgG coupled to HRP for immunohistochemistry was used as secondary antibody, goat anti-rabbit IgG coupled to 10-nm gold particles for immunogold labeling, and a HRP-conjugated donkey-anti-rabbit IgG for immunoblotting.

Immunohistochemistry. Rat pancreas was perfusion fixed with 4% paraformaldehyde (wt/vol), cut in small pieces, cryo-protected, and frozen in liquid N₂. Immunohistochemistry was performed on 5-µm cryosections. Slides were treated with 1% SDS (wt/vol) for 5 min, blocked with 1% BSA (wt/vol)-PBS for 15 min, and incubated with rabbit polyclonal antibodies against human KCNQ1 at 1:20 dilution overnight at 4°C. After incubation with goat anti-rabbit IgG coupled to HRP (1:50), peroxidase reaction was visualized as described elsewhere (38).

Preembedding immunogold electron microscopy. Electron microscopy of isolated rat pancreatic ZG was performed as described earlier (52) with minor modifications. ZG were fixed in 2.5% glutaraldehyde in PBS for 1 h and adsorbed onto poly-L-lysine-coated glass slides. Granules were washed with PBS, treated sequentially with PBS containing 0.5% BSA, 0.5% gelatin, and 0.5% Tween 20, followed by PBS containing 0.05% BSA, 0.05% gelatin, and 0.05% Tween 20, and then incubated with BLE 2-1 (1:30 dilution) overnight at 4°C. Samples were incubated with goat anti-rabbit IgG coupled to 10-nm gold particles (1:10 dilution) for 1 h at room temperature, fixed with 1% glutaraldehyde in PBS for 30 min, and washed with water for 45 min. Samples were postfixed with 1% OsO₄ in 0.1 M sodium cacodylate, treated with aqueous 2% uranyl acetate, dehydrated in a graded series of ethanol, and embedded in Araldit resin. Thin sections were counterstained with 2% uranyl acetate and lead citrate and viewed with a Philips EM-10 electron microscope.

Immunoblotting. Twenty-five micrograms of protein of rat pancreas tissue homogenate or ZG membranes were mixed with 3x Laemmli buffer, incubated for 5 min at 95°C, and sonicated on ice. Membrane proteins were separated by SDS-PAGE on 9% acrylamide Laemmli minigels and transferred onto polyvinylidene difluoride membranes. After blocking with Tris-buffered saline containing 0.1% Tween 20 and 3% non-fat dry milk for 8 h, the membranes were incubated at 4°C with BLE 2-1 (1:1,000) overnight, followed by HRP-conjugated donkey anti-rabbit IgG (1:10,000) for 1 h. The blots were developed by Western Lighting Plus chemiluminescence reagents (Perkin-Elmer Life Sciences, Boston, MA), and signals were visualized on X-ray films.

Statistics. Representative data or means ± SE are shown. Statistical analysis using unpaired Student's t-test was carried out with Sigma Plot 8.0 (SPSS). For more than two groups, one-way ANOVA was used, assuming equality of variance with Levene's test and Tukey's post hoc test for pairwise comparison with SPSS 12.0. Results with P 0.05 were considered to be statistically significant.

	RESULTS

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

 
Differential inhibition of ZG ion conductive pathways by chromanol 293B. The membrane of ZG is equipped with at least two cation and two anion conductive pathways, which contribute to K⁺ and Cl^– fluxes into the granule matrix and promote the release of digestive enzymes (for review, see Ref. 48). The osmotic lysis assay of isolated ZG represents a simple and robust technique to measure macroscopic flux of ions across the ZG membrane. It also permits screening of drugs with high throughput but has the disadvantage of low sensitivity and specificity when it comes to identifying ion channels (58). As a first approach to determine whether KCNQ1-type K⁺ channels contribute to ZG K⁺-conductance pathways, we investigated the effect of the chromanol 293B on ZG K⁺ conductance. In the absence of the electrogenic protonophore CCCP (16 µM), the absorbance of ZG suspended in 150 mM potassium acetate and buffered with 50 mM imidazole to pH 7.0 remains stable throughout the recording period when K⁺ or other monovalent cations are used as major osmolytes (49, 50) (Fig. 1, A and B; data not shown). Addition of CCCP decreases the absorbance of the ZG suspension as a result of increased osmotic lysis of ZG (Fig. 1A). This is due to CCCP-induced influx of K⁺ and acetic acid, which is driven by the inside-negative H⁺ diffusion potential (pH of ZG matrix < pH of the buffer) (see EXPERIMENTAL PROCEDURES and Ref. 50). As shown in Fig. 1, A and D, 293B inhibited ZG lysis in K⁺ buffer in a concentration-dependent manner from 0.1 to 100 µM 293B. Inhibition of ZG lysis was half-maximal at 10 µM (Fig. 1D) and blocked 70–80% of ZG lysis at 100 µM 293B, but 20–30% of K⁺ conductive pathway was not affected by 293B (Fig. 1A). Moreover, HMR-1556 (0.1–100 µM), another specific inhibitor of KCNQ1-type K⁺ channels (15, 17, 32), also inhibited the ZG K⁺ conductive pathway with a similar potency (data not shown). When K⁺ is used as the major osmolyte, both K⁺-selective and a nonselective monovalent cation permeability pathway contribute to overall ZG lysis (49). Therefore, we investigated the effect of 293B on the nonselective cation pathway by replacing K⁺ with Na⁺, which equally permeates the nonselective cation conductive pathway (49). As shown in Fig. 1, B and D, 293B even at 100 µM had a very small inhibitory effect (20% inhibition) on ZG lysis, indicating that the 293B-insensitive ZG lysis in K⁺ buffer can be accounted for by the nonselective cation permeability pathway.

View larger version (14K): [in this window] [in a new window]   Fig. 1. Effect of 293B on K+, Na+, and Cl– conductive pathways of rat pancreatic zymogen granules (ZG) suspended in buffered isosmotic salt solution. A: protonophore-induced osmotic lysis of ZG suspended in 150 mM potassium acetate buffered with 50 mM imidazole (pH 7.0) and inhibition by different concentrations of 293B. ZG were suspended in the buffer (1st arrow) with or without (0.1% DMSO) the indicated concentrations of 293B. The protonophore CCCP (16 µM) was added (2nd arrow) to induce K+ conductance (for details of the lysis assay, see EXPERIMENTAL PROCEDURES). B: to assay nonselective monovalent cation conductive pathways, ZG were suspended in 150 mM sodium acetate buffered with 50 mM imidazole (pH 7.0). Otherwise, experimental conditions were identical to those in A. C: to assay Cl– permeability, ZG were resuspended in 150 mM KCl buffered with 50 mM HEPES (pH 7.0), and osmotic lysis was induced by adding 5 µM of the electrogenic potassium ionophore valinomycin (2nd arrow). Dotted traces represent control experiments in the absence of ionophore. All curves are typical for 3–5 experiments. D: dose-response curve for block of ion permeabilities by 293B in rat pancreatic ZG. Experiments were conducted in the absence or presence of different concentrations of the KCNQ1 blocker 293B. Control rates (100%) of granular lysis after addition of ionophores were measured in the absence of 293B but with 0.1% DMSO in the buffer and were estimated to be 10.6 ± 3.5 h–1 for K+ permeability, 4.8 ± 2.9 h–1 for Na+ permeability, and 3.3 ± 0.7 h–1 for Cl– permeability. Results are means ± SE of 3–5 different experiments.

Biophysical characterization of ZG membrane K⁺ channels incorporated into planar bilayers and pharmacological block by 293B and HMR-1556. From 11 individual preparations of rat ZG membranes, 472 experiments were performed. After incorporation of ZG membranes, channel events were frequently observed (i.e., in 20–25% of all experiments). However, in only 48 experiments were full-range current-voltage relationships available, allowing detailed biophysical and pharmacological analyses. Only these 48 experiments were included in the study. Of these, five single channels were observed with KCl buffer in both chambers. The reversal potential (E_rev) of these five single channels was –15.3 ± 4.5 mV (compared with a theoretical Cl^– potential of +36.8 mV or a theoretical E_K of –36.8 mV based on the Nernst equation with 645 mM KCl cis and 150 mM KCl trans). Obviously, the currents were more K⁺ than Cl^– selective. However, we could not exclude the superpositioning of currents from simultaneously incorporated Cl^– and K⁺ channels, which would affect E_rev toward nonselectivity. Clearly, in some of the KCl experiments, the estimated E_rev was about –30.0 mV, showing near "ideal" K⁺ selectivity. Apart from K⁺ channels, anion-selective channels were also occasionally recorded. In the two experiments where current-voltage relationships were obtained, the respective E_rev values were +33.6 and +34.5 mV, indicating near perfect Cl^– selectivity (data not shown). However, these channels were very rarely observed and were not investigated further. To prevent simultaneous recordings of Cl^– and K⁺ currents, we replaced Cl^– with gluconate ion to focus on the more frequently observed K⁺ channels (in addition, we sometimes observed that the bilayer itself displayed a nonspecific background permeability, which could have also affected E_rev in some experiments; data not shown).

Under potassium gluconate buffer conditions, the E_rev of single channels from all experiments recorded was –20.9 ± 0.9 mV (n = 39) (theoretical E_K = –36.8 mV). The average single-channel conductance of these K⁺-selective channels was 265.8 ± 44.0 pS when measured in the range of –30 to +30 mV with 645 mM/150 mM potassium gluconate cis/trans. The mean single-channel P_open was estimated at 0.49 ± 0.05 (Table 1). To ensure that the K⁺ currents were not due to nonselective cation channels [a flufenamate-sensitive nonselective monovalent cation permeability has been described in ZG with the osmotic lysis assay, which is equally selective to K⁺ and Na⁺ (48, 49); Fig. 1], we replaced 500 mM potassium gluconate in the cis chamber with 500 mM sodium gluconate. This gradient of 145/150 mM potassium gluconate cis/trans should shift E_rev of K⁺-selective channels toward more positive values (E_K = +0.9 mV), whereas E_rev of a nonselective monovalent cation channel would remain unaffected by this change of solution. As shown in Table 1, E_rev under 145/150 mM potassium gluconate cis/trans was –2.4 ± 3.6 mV (n = 3), and the single-channel conductance decreased to 105.3 ± 23.8 pS. This indicates that the ZG membrane channels observed are K⁺ selective.

View larger version (22K): [in this window] [in a new window]   Fig. 2. Single-channel activity and biophysical and pharmacological properties of K+ channels (K+ group I) in ZG membranes using planar lipid bilayer. ZG membrane vesicles were fused with planar lipid bilayers, and currents were recorded at different voltage steps in 645/150 mM potassium gluconate cis/trans. Channel activity was recorded for up to 1 min. Closed state (C) and different open-channel levels (O1–O4) are indicated. Scales are applicable to all traces for each channel type. A: K+ group I channels are large K+-conducting channels with frequent openings and closings and show multiple levels of opening (up to 4). Typical traces of 6 individual experiments are shown. B: respective all-points histogram at 0 mV. C: current-voltage relationship of K+ group I channels was fitted by linear regression. Results are means ± SE of 6 experiments. Reversal potential (Erev; –22.9 ± 2.2 mV) was determined from each fitted plot. D: to test the effect of HMR-1556 (100 µM), currents were recorded at 0 mV. Channel activity was recorded for 4 min after addition of the inhibitor. E: single-channel open probability (Popen) was derived from NPopen/N (where N is no. of open channels), assuming that the different levels represent multiple openings of the same channel type based on the data from the single-channel search analysis in Clampfit. Results are means ± SE of 6 experiments with HMR-1556. Statistical analysis using Student's paired t-test compared single Popen before and after inhibitor application.

View larger version (10K): [in this window] [in a new window]   Fig. 3. Single-channel activity of K+-selective channels (K+ group II) in ZG membranes using planar lipid bilayer. A: K+ group II channels are channels with a smaller conductance and were detected in 3 experiments. This K+-selective channel also shows multiple levels of opening, which were more prominent at voltages >20 mV. Representative recordings with up to 4 channel levels are shown. B: current-voltage relationship was fitted by linear regression. Results are means ± SE of 3 experiments. Erev (–22.1 ± 0.5 mV) was determined from each fitted plot. C: corresponding all-points histogram at +40 mV.

View larger version (22K): [in this window] [in a new window]   Fig. 4. Single-channel activity and biophysical and pharmacological properties of K+-selective channels (K+ group III) in ZG membranes using planar lipid bilayer. A: K+ group III channels have a characteristic "flickering" gating pattern. At positive voltages, the channel is rarely closed, whereas at negative voltages, the channel switches frequently between open and closed states. Typical traces of 3 different experiments are depicted. B: all-points histogram at –50 mV depicts the sublevel conductance of this channel. C: current-voltage plot was fitted linearly, and Erev = –18.3 ± 1.8 mV (n = 3) was determined. Results are means ± SE. D: block of K+ channel by 293B at –50 mV. The inhibitor 293B (100 µM) was applied to both cis and trans sides. Immediate inhibition of the larger K+ current revealed a smaller 293B-insensitive K+ channel with different gating properties and peak amplitude of 1 pA (inset, bottom right). E: single Popen before and after 293B application (n = 5). Statistical analysis was performed with Student's paired t-test.

View larger version (60K): [in this window] [in a new window]   Fig. 5. Expression of KCNQ1 in pancreatic ZG membranes. A: immunolocalization of KCNQ1 in ZG (arrows) of rat pancreatic acinar cells using an antibody raised against human KCNQ1 (1:20) and immunoperoxidase light microscopy (+Ab). No labeling was detectable when tissue sections were incubated with the secondary antibody alone (–Ab). B: preembedding immunogold electron microscopy of isolated rat pancreatic ZG with KCNQ1 antibody BLE 2-1 (1:30) directed against rodent KCNQ1 (+Ab). Immunogold staining was absent in controls, i.e., in the absence of primary antibody (–Ab). Bar = 0.2 µm. C: immunoblotting of rat pancreatic homogenate (Ho) and ZG membranes (ZGM). Membrane protein (25 µg) was probed with the polyclonal antibody BLE 2-1 raised against KCNQ1 (1:1,000). One representative immunoblot of 3 similar ones is shown. In the absence of primary antibody (–Ab), a nonspecific protein band of >50 kDa molecular mass (MM) was also detected in ZGM.

Inhibition of CCK-OP-induced amylase secretion by 293B in permeabilized pancreatic acini. To study the possible role of KCNQ1 in secretagogue activation of pancreatic enzyme secretion, we tested various concentrations of the inhibitor of KCNQ1, chromanol 293B, on basal and CCK-OP-induced amylase secretion of isolated pancreatic acini. Acini were permeabilized with 5 µg/ml digitonin before hormonal stimulation to exclude a possible contribution of KCNQ1/KCNE1 K⁺ channels expressed in the plasma membrane of rat pancreatic acinar cells (56). Amylase released into the medium within 30 min under basal conditions or after addition of a maximally stimulatory concentration of 1 nM CCK-OP was calculated as a percentage of the total amount of amylase in the cell suspension (1). Under basal conditions, amylase release was similar in buffers without and with 293B. In DMSO controls, 3.1 ± 0.4% of total amylase was released (n = 5), whereas in the presence of the maximal tested concentration of 100 µM 293B, the release amounted to 2.9 ± 0.6% of total amylase (n = 4) (data not shown). CCK-OP (1 nM) stimulated amylase release in control acini (3.7 ± 0.7% of total amylase above basal; n = 5). Quite unexpectedly, 293B (1–100 µM) did not affect CCK-OP-stimulated amylase release (e.g., 3.4 ± 0.4% of total amylase above basal at 100 µM 293B; n = 7) (see Fig. 6). However, both the K⁺ conductive and the nonselective monovalent cation conductive pathway of ZG are permeable to K⁺ (49). Therefore, even if 293B blocked the K⁺-selective channel, K⁺ could still enter the granule matrix via a nonselective channel and thereby enhance CCK-OP-evoked enzyme secretion. Flufenamic acid has been shown to specifically inhibit the nonselective monovalent cation permeabilities of ZG but has no effect on ZG K⁺ and Cl^– conductive pathways (Ref. 49 and data not shown). Hence, we repeated the experiments in the presence of 100 µM flufenamic acid. Basal amylase release was not affected by 100 µM flufenamate (3.3 ± 0.8% of total amylase, n = 3; data not shown). With 100 µM flufenamate, CCK-OP (1 nM)-stimulated amylase secretion was slightly increased (4.7 ± 0.7% of total amylase above basal; n = 5) but did not reach significance. In contrast, when 100 µM flufenamate was present, 293B inhibited CCK-OP-induced amylase secretion in a concentration-dependent manner with a half-maximal effect at 10 µM (Fig. 6). This suggests that both the nonselective and K⁺-selective cation conductances can mediate secretagogue-induced enzyme secretion.

View larger version (16K): [in this window] [in a new window]   Fig. 6. Effect of 293B on CCK octapeptide (CCK-OP)-induced amylase secretion of permeabilized rat pancreatic acini. Acini were permeabilized with 5 µg/ml digitonin, incubated in buffer with 135 mM KCl, and preincubated for 5 min with inhibitors or 0.1% solvent before addition of the secretagogue. The effect of different concentrations of 293B without or with 100 µM flufenamic acid on amylase released 30 min after stimulation with 1 nM CCK-OP is shown. Basal amylase release per 30 min, i.e, in the absence of secretagogue (3.1 ± 0.4% and 2.9 ± 0.6% for 0.1% DMSO and 100 µM 293B, respectively) was subtracted from the respective values with CCK-OP. Values are means ± SE of 3–7 experiments. *Significant differences (P < 0.01) between experiments without or with 293B using one-way ANOVA.

	DISCUSSION

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

293B inhibitor specificity and proposed site of action. The interpretation of the data in Fig. 1 is critically dependent on the relative permeability for K⁺ over Na⁺ of the K⁺ channel studied, the relative permeability for K⁺ over Na⁺ of the nonselective cation channel, and the efficacy of 293B to block the K⁺ channel as opposed to the nonselective cation channel. In fact, 293B was found to affect lysis rates in K⁺ buffer at much lower concentrations than lysis in Na⁺ buffer (Fig. 1), suggesting that the bulk of the lysis effect is due to influx via the K⁺ channel. It has been suggested that 293B might have the properties of a protonophore. The osmotic lysis assay for cation conductive pathways depends on an H⁺ gradient. Hence, it could be argued that the inhibitory effect of 293B on K⁺ permeability might be caused by dissipation of the H⁺ gradient across the ZG membrane rather than by specific inhibition of a KCNQ1 channel. However, ZG lysis in Na⁺ buffer was not affected by 293B (Fig. 1). Moreover, Grahammer et al. (18) have directly shown by BCECF fluorescence measurements of intact cells that 293B does not act as a protonophore. In addition, fusion of ZG membranes with planar bilayers resulted in the appearance of K⁺-selective channels (Figs. 2–4) that were inhibited by 293B or by HMR-1556, which specifically inhibits KCNQ1 channels (15, 17, 32). Finally, CCK-OP-induced secretion of amylase by permeabilized pancreatic acini was efficiently blocked by 293B with an IC₅₀ of 10 µM, when the inhibitor of the nonselective cation conductance flufenamate was also present in the medium (Fig. 6), indicating that ZG cation conductances are required for enzyme secretion to occur.

Characteristics of ZG K⁺ channels. Kelly et al. (25) were able to patch single pancreatic ZG and detected ion conductances with KCl in the bath and patch pipette that were partially inhibited by low concentrations (20–40 µM) of the Cl^– channel blocker DIDS and K⁺ channel blockers quinine and glyburide. Single-channel currents were also observed, which were nonselective but inhibited by 20 µM DIDS. Channels had an intermediate conductance of 40 pS and no voltage dependence. Immunoblots revealed expression of the Cl^– channel proteins ClC-2 and ClC-3 and of the K⁺ channel protein KCNJ8/K_ir6.1. This study was hampered by the use of relatively unspecific inhibitors but still represents the first direct electrophysiological evidence for ion channels in ZG membranes. In the present study, we provide evidence for the presence of a KCNQ1 K⁺ channel protein in ZG membranes (Fig. 5) in conjunction with K⁺-selective single channels incorporated into planar bilayers, which are blocked by the specific inhibitors of KCNQ1, 293B and HMR-1556 (Figs. 2–4).

Although the solution replacement experiments (Table 1) indicated that the monovalent cation channels observed were indeed K⁺ selective, the large conductances observed are unusual for KCNQ1 K⁺ channels (4, 36). This anomaly may well be accounted for by possible interaction of KCNQ1 with specific, yet unknown, KCNQ1 channel subunits, which could modify the properties of the channel (37) (see also below). However, this remains unproven. Another important fact that could account for the large K⁺ conductance is that the K⁺ concentration used in the study in the cis chamber (645 mM) is much higher than those used in published patch-clamp studies. In those studies (37, 43), 100–140 mM K⁺ was used and conductances between 7 and 16 pS were measured. Moreover, external (trans) K⁺ concentration may also affect KCNQ1 conductance. When KCNQ1 was expressed in oocytes, an increase of external K⁺ from 0 to 10 mM K⁺ resulted in an increase of single-channel conductance from 1.9 to 4.8 pS, i.e., the resulting apparent channel conductance increased with increasing external K⁺ concentration (60). These specific characteristics of our experimental conditions may have also contributed to the increased K⁺ conductance observed.

Taking these specific features into consideration, a still obvious candidate for large-conductance K⁺ channels would be the BK channels. However, charybdotoxin, a specific inhibitor of BK channels (35), had no effect (Table 1). In contrast, 293B and HMR-1556, specific blockers of KCNQ1 channels (15, 17, 32), inhibited K⁺ currents. Strikingly, the concentrations of 293B and HMR-1556 required to inhibit K⁺-selective channels of ZG membranes were high (100–200 µM) (Figs. 2D, 2E, 4D, and 4E), which differs from the low micromolar to submicromolar concentrations required to inhibit the typical I_Ks associated with KCNQ1 (17, 32). This is the case because KCNE1 has been shown to enhance the sensitivity of KCNQ1 to 293B and HMR-1556 (32). It could be argued that HMR-1556 and 293B are not absolutely specific blockers of KCNQ1, particularly at the high concentrations used in the present study. So far, however, only one study has been reported in which a non-KCNQ1 K⁺ channel has been shown to be blocked by both compounds in the concentration range used in our study, namely, the transient outward potassium current (I_to) associated with Kv4.3 (53). However, Kv4.3 is exclusively expressed in the heart and brain (42). In contrast, it is well established that KCNQ1 is expressed in pancreatic acini (12, 56). Hence, it is very likely that HMR-1556 and 293B block K⁺ currents of ZG membranes that are mediated by KCNQ1.

The specific biophysical and pharmacological properties of KCNQ1 are determined by its regulatory β-subunits KCNE1 (IsK, minK), KCNE2 (MiRP1), and KCNE3 (MiRP2), which are expressed in a tissue-specific manner to form the native K⁺ channel (36). In the heart and inner ear, KCNQ1 interacts with KCNE1 to produce a voltage-gated I_Ks (3, 39). In contrast, coassembly of KCNQ1 with KCNE3 yields currents that are nearly instantaneous and depend linearly on voltage (40). As shown in Figs. 2C and 4C, 293B- and HMR-1556-sensitive ZG membrane K⁺ channels incorporated into planar bilayers showed a linear current-voltage relationship, suggesting that KCNQ1 expressed in ZG membranes is not associated with KCNE1. Both KCNQ1 and KCNE1 are expressed in rodent pancreatic acinar cells (12, 29, 47), but KCNQ1 may be associated with other yet unknown regulatory subunits, which could account for activation of the K⁺ current by cAMP or Ca²⁺ (4). We have previously described the association of a regulatory subunit of ZG K⁺ conductance, ZG-16p, with the ZG membrane (6). This protein is exclusively expressed in colon, pancreas, duodenum, and stomach (9). Whether ZG-16p represents a novel β-subunit of KCNQ1 in gastrointestinal epithelia remains to be investigated. These considerations are important because they could account for the different biophysical properties of the channels incorporated into the planar bilayer (e.g., K⁺ channel groups I and III shown in Figs. 2 and 4) and could reflect the association of KCNQ1 with different β-subunits or the dissociation of β-subunits after incorporation of the channels into the bilayer. The relative heterogeneity of the biophysical properties of the K⁺ channels detected could also reflect to some extent contamination from other organelles, although it is unlikely because the preparation of ZG used yield membranes of high purity (51).

Finally, we noticed the occurrence of subconductance levels in all three K⁺ channel groups (groups I-III) but particularly in group III (Figs. 2A, 3A, and 4A). These short-lived subconductance levels were mostly visited when the channel gate moved between the fully open state and the closed state. So far, there has been no report in the literature referring to a possible molecular mechanism underlying these subconductance levels for KCNQ1 K⁺ channels. As for all other six-membrane-spanning K⁺ channels, it is believed that assembly of four KCNQ1 proteins is required to form a functional channel. Interestingly, it has been suggested for another six-membrane-spanning K⁺ channel (the K_v2.1) that, when channels move from the closed state to the fully open state, these sublevels result from heteromeric pore conformations, which are more frequently observed in partially activated channels, in which some but not all subunits have undergone voltage-dependent conformational changes required for channel opening (8). Hence, similar mechanisms could be operative in KCNQ1 K⁺ channels to account for subconductance levels.

Physiological significance of ZG K⁺ channel. There is a precedent for the expression of KCNQ1 in intracellular secretory vesicles in the gastrointestinal tract. Recently, KCNQ1 has been identified as a K⁺ channel located in intracellular tubulovesicles and apical membrane of parietal cells, where it colocalizes with the H⁺-K⁺-ATPase (18). Inhibition of KCNQ1 current by chromanol 293B abolished acid secretion. The β-subunits KCNE2 and KCNE3 were expressed in stomach; KCNE1, however, was not. This suggested that KCNQ1 is the pore-forming subunit of the K⁺ channel responsible for sustained HCl secretion (18). Thus, in parietal cells, KCNQ1 appears to have a dual subcellular distribution by trafficking between intracellular tubulovesicles and the apical plasma membrane.

In Cl^– secretory epithelia, such as the colon and pancreas, a 293B-sensitive small conductance (1–2 pS) Ca²⁺-activated K⁺ current that is located in the basolateral plasma membrane has been proposed to provide the driving force for luminal Cl^– secretion that may be mediated by KCNQ1 (26, 27, 29). KCNQ1 expression in the basolateral membrane of rodent pancreatic acini was subsequently confirmed by immunofluorescence microscopy using the rabbit polyclonal antibody BLE 2-1 (56) (see also Fig. 5A). It is conceivable that KCNQ1 K⁺ channels expressed in ZG membranes could also contribute to this plasma membrane K⁺ current after fusion of ZG with the apical plasma membrane. In a recent study, Lee et al. (31) tested the effect of 293B and HMR-1556 on fluid and enzyme secretion induced by ACh in the vascularly perfused rat pancreas. However, they were unable to observe any inhibitory effect of the KCNQ1 channels blockers on secretion and concluded that KCNQ1 is not essential for Ca²⁺-mediated secretion of rat pancreatic acini. These data are at odds with their former studies on Cl^– secretion in isolated acini (26, 27) and could hint at additional factors that modify acinar secretion in the intact pancreas. At first sight, they also appear to contradict our present results concerning the role of KCNQ1 in enzyme secretion (Fig. 6). However, from our previous (reviewed in Ref. 48) and present studies, it appears that the membrane of ZG is equipped with at least two cation and two anion conductive pathways, which contribute to K⁺ and Cl^– fluxes into the granules and promote release of digestive enzymes. This apparent "functional redundancy" of ion channels in secretory granules appears to be a more widespread phenomenon. "Knockout" (–/–) studies of ClC-3, a Cl^– channel, which promotes acidification of synaptic vesicles (44), show that ClC-3^(–/–) synaptic vesicles still acidify, although at a lower rate, suggesting the contribution of other Cl^– channels to vesicular acidification. Furthermore, late endosomes and lysosomes of osteoclasts from ClC-7^(–/–) mice show normal acidification rates, although the Cl^– channel ClC-7 is expressed in these organelles (28). Similarly, we found that CCK-stimulated enzyme secretion in permeabilized rat pancreatic acini is abolished only if the inhibitor of nonselective cation conductive pathway, flufenamate, is applied together with the KCNQ1 K⁺ channel inhibitor 293B (Fig. 6), which would reconcile our data with the results obtained by Lee et al. (31) for ACh-induced amylase secretion.

Previous studies have shown that CCK-OP- or carbachol-induced enzyme secretion of isolated rat pancreatic acini is absolutely dependent on Cl^–: omission of Cl^– in the cytosol or inhibition of ZG Cl^– channels abolished enzyme secretion (14). Flufenamate has been shown to block CFTR Cl^– channels and other channels (see for instance Ref. 33), but it does not block Cl^–-dependent enzyme secretion as shown in Fig. 6 (compare experiments without 293B in left and right). Moreover, in the osmotic lysis assay of pancreatic ZGs, flufenamate appears to be the only compound that specifically blocks the nonselective cation permeability but not K⁺ permeability (reviewed in Ref. 48) and in addition has no effect on ZG Cl^– conductive pathway at concentrations up to 100 µM (data not shown). Hence, flufenamate appears to be a useful pharmacological tool in our secretion experiments, which helps to unravel the fact that both nonselective and K⁺ conductance contribute to enzyme secretion. This observation therefore hints at a functional overlap of both monovalent cation conductive pathways with regard to stimulated enzyme secretion. Thus, at the present stage of our knowledge, we must also consider that under specific (yet undetermined) physiological conditions one or the other conductance will dominate the secretory process.

To conclude, in the present study, we tested the hypothesis that KCNQ1 channels, or channels containing KCNQ1 subunits, contribute to a charge-compensating K⁺-conducting pathway in the membrane of pancreatic ZGs. The data indicate that KCNQ1 protein is expressed in ZG membranes of the rat exocrine pancreas. KCNQ1 may account for K⁺-selective currents, which were detected after incorporation of ZG membranes into planar bilayer membranes and blocked by the specific blockers of KCNQ1-associated K⁺ currents, 293B and HMR-1556. However, the unusual properties of the K⁺ channels at the single-channel level and the relatively low potency of the drugs tested on these channels indicate that additional subunits may modify the channel characteristics. This aspect of the study therefore requires further investigation. We propose that the K⁺ channels associated with KCNQ1 may underlie ZG K⁺-selective monovalent K⁺ conductive pathways of ZG and contribute to secretagogue-induced enzyme secretion from pancreatic acini. Further investigations aimed at the electrophysiological and molecular characterization of additional cation and anion conductive pathways in ZGs are also required, which could lead to a better understanding of the physiology of exocrine pancreatic enzyme secretion.

	GRANTS

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

 
This work was supported by Deutsche Mukoviszidose e.V. (F04/04 to F. Thévenod) and start-up funds from the University of Witten/Herdecke.

	ACKNOWLEDGMENTS

 
We thank Drs. R. Warth (Institute of Physiology, University of Regensburg, Regensburg, Germany) and M. Bleich (Institute of Physiology, Christian-Albrechts-University, Kiel, Germany) for providing the BLE 2-1 antibody and Professor T. J. Jentsch (Leibniz Institute for Molecular Medicine/Max-Delbrück-Center for Molecular Medicine, Berlin, Germany) for providing the rabbit antiserum against the COOH terminus of human KCNQ1. HMR-1556 and 293B were a gift of Sanofi-Aventis Germany. We are indebted to Professor Patrik Rorsman (Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, Oxford, UK) for continuous interest in our work.

	FOOTNOTES

 

Address for reprint requests and other correspondence: F. Thévenod, Dept. of Physiology & Pathophysiology, Univ. of Witten/Herdecke, Stockumer Strasse 12 (Thyssenhaus), D-58448 Witten, Germany (e-mail: frank.thevenod{at}uni-wh.de)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

	REFERENCES

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION GRANTS REFERENCES

 
1. Anderie I, Thévenod F. Evidence for involvement of a zymogen granule Na⁺/H⁺ exchanger in enzyme secretion from rat pancreatic acinar cells. J Membr Biol 152: 195–205, 1996.[CrossRef][Web of Science][Medline]

2. Barg S, Huang P, Eliasson L, Nelson DJ, Obermuller S, Rorsman P, Thévenod F, Renstrom E. Priming of insulin granules for exocytosis by granular Cl(-) uptake and acidification. J Cell Sci 114: 2145–2154, 2001.[Abstract/Free Full Text]

3. Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M, Romey G. K(V)LQT1 and lsK (minK) proteins associate to form the I(Ks) cardiac potassium current. Nature 384: 78–80, 1996.[CrossRef][Web of Science][Medline]

4. Bleich M, Warth R. The very small-conductance K⁺ channel KvLQT1 and epithelial function. Pflügers Arch 440: 202–206, 2000.[Web of Science][Medline]

5. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254, 1976.[CrossRef][Web of Science][Medline]

6. Braun M, Thévenod F. Photoaffinity labeling and purification of ZG-16p, a high-affinity dihydropyridine binding protein of rat pancreatic zymogen granule membranes that regulates a K⁺-selective conductance. Mol Pharmacol 57: 308–316, 2000.[Abstract/Free Full Text]

7. Carew MA, Thorn P. Identification of ClC-2-like chloride currents in pig pancreatic acinar cells. Pflügers Arch 433: 84–90, 1996.[CrossRef][Web of Science][Medline]

8. Chapman ML, VanDongen AM. K channel subconductance levels result from heteromeric pore conformations. J Gen Physiol 126: 87–103, 2005.[Abstract/Free Full Text]

9. Cronshagen U, Voland P, Kern HF. cDNA cloning and characterization of a novel 16 kDa protein located in zymogen granules of rat pancreas and goblet cells of the gut. Eur J Cell Biol 65: 366–377, 1994.[Web of Science][Medline]

10. De Lisle RC, Hopfer U. Electrolyte permeabilities of pancreatic zymogen granules: implications for pancreatic secretion. Am J Physiol Gastrointest Liver Physiol 250: G489–G496, 1986.[Abstract/Free Full Text]

11. Dedek K, Waldegger S. Colocalization of KCNQ1/KCNE channel subunits in the mouse gastrointestinal tract. Pflügers Arch 442: 896–902, 2001.[CrossRef][Web of Science][Medline]

12. Demolombe S, Franco D, de Boer P, Kuperschmidt S, Roden D, Pereon Y, Jarry A, Moorman AF, Escande D. Differential expression of KvLQT1 and its regulator I_sK in mouse epithelia. Am J Physiol Cell Physiol 280: C359–C372, 2001.[Abstract/Free Full Text]

13. Estevez R, Boettger T, Stein V, Birkenhager R, Otto E, Hildebrandt F, Jentsch TJ. Barttin is a Cl^– channel β-subunit crucial for renal Cl^– reabsorption and inner ear K⁺ secretion. Nature 414: 558–561, 2001.[CrossRef][Medline]

14. Fuller CM, Eckhardt L, Schulz I. Ionic and osmotic dependence of secretion from permeabilised acini of the rat pancreas. Pflügers Arch 413: 385–394, 1989.[CrossRef][Web of Science][Medline]

15. Gerlach U, Brendel J, Lang HJ, Paulus EF, Weidmann K, Bruggemann A, Busch AE, Suessbrich H, Bleich M, Greger R. Synthesis and activity of novel and selective I(Ks)-channel blockers. J Med Chem 44: 3831–3837, 2001.[CrossRef][Web of Science][Medline]

16. Gimenez-Gallego G, Navia MA, Reuben JP, Katz GM, Kaczorowski GJ, Garcia ML. Purification, sequence, and model structure of charybdotoxin, a potent selective inhibitor of calcium-activated potassium channels. Proc Natl Acad Sci USA 85: 3329–3333, 1988.[Abstract/Free Full Text]

17. Gogelein H, Bruggemann A, Gerlach U, Brendel J, Busch AE. Inhibition of I_Ks channels by HMR 1556. Naunyn Schmiedebergs Arch Pharmacol 362: 480–488, 2000.[CrossRef][Web of Science][Medline]

18. Grahammer F, Herling AW, Lang HJ, Schmitt-Graff A, Wittekindt OH, Nitschke R, Bleich M, Barhanin J, Warth R. The cardiac K⁺ channel KCNQ1 is essential for gastric acid secretion. Gastroenterology 120: 1363–1371, 2001.[CrossRef][Web of Science][Medline]

19. Hanke W, Schlue WR. Planar Lipid Bilayers: Methods and Applications. London: Academic, 1993.

20. Heitzmann D, Grahammer F, von Hahn T, Schmitt-Graff A, Romeo E, Nitschke R, Gerlach U, Lang HJ, Verrey F, Barhanin J, Warth R. Heteromeric KCNE2/KCNQ1 potassium channels in the luminal membrane of gastric parietal cells. J Physiol 561: 547–557, 2004.[Abstract/Free Full Text]

21. Hille B. Ionic Channels of Excitable Membranes. Sunderland, MA: Sinauer Associates, 1984.

22. Hoy M, Olsen HL, Bokvist K, Buschard K, Barg S, Rorsman P, Gromada J. Tolbutamide stimulates exocytosis of glucagon by inhibition of a mitochondrial-like ATP-sensitive K⁺ (K_ATP) conductance in rat pancreatic A-cells. J Physiol 527: 109–120, 2000.[Abstract/Free Full Text]

23. Inagaki N, Gonoi T, Clement JPt Namba N, Inazawa J, Gonzalez G, Aguilar-Bryan L, Seino S, Bryan J. Reconstitution of IKATP: an inward rectifier subunit plus the sulfonylurea receptor. Science 270: 1166–1170, 1995.[Abstract/Free Full Text]

24. Jahn R, Lang T, Sudhof TC. Membrane fusion. Cell 112: 519–533, 2003.[CrossRef][Web of Science][Medline]

25. Kelly ML, Abu-Hamdah R, Jeremic A, Cho SJ, Ilie AE, Jena BP. Patch clamped single pancreatic zymogen granules: direct measurements of ion channel activities at the granule membrane. Pancreatology 5: 443–449, 2005.[CrossRef][Web of Science][Medline]

26. Kim SJ, Greger R. Voltage-dependent, slowly activating K⁺ current (I_Ks) and its augmentation by carbachol in rat pancreatic acini. Pflügers Arch 438: 604–611, 1999.[CrossRef][Web of Science][Medline]

27. Kim SJ, Kim JK, Pavenstadt H, Greger R, Hug MJ, Bleich M. Regulation of slowly activating potassium current (I_Ks) by secretin in rat pancreatic acinar cells. J Physiol 535: 349–358, 2001.[Abstract/Free Full Text]

28. Kornak U, Kasper D, Bosl MR, Kaiser E, Schweizer M, Schulz A, Friedrich W, Delling G, Jentsch TJ. Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104: 205–215, 2001.[CrossRef][Web of Science][Medline]

29. Kottgen M, Hoefer A, Kim SJ, Beschorner U, Schreiber R, Hug MJ, Greger R. Carbachol activates a K⁺ channel of very small conductance in the basolateral membrane of rat pancreatic acinar cells. Pflügers Arch 438: 597–603, 1999.[CrossRef][Web of Science][Medline]

30. Kunzelmann K, Hubner M, Schreiber R, Levy-Holzman R, Garty H, Bleich M, Warth R, Slavik M, von Hahn T, Greger R. Cloning and function of the rat colonic epithelial K⁺ channel KVLQT1. J Membr Biol 179: 155–164, 2001.[CrossRef][Web of Science][Medline]

31. Lee JE, Park HS, Uhm DY, Kim SJ. Effects of KCNQ1 channel blocker, 293B, on the acetylcholine-induced Cl- secretion of rat pancreatic acini. Pancreas 28: 435–442, 2004.[CrossRef][Web of Science][Medline]

32. Lerche C, Seebohm G, Wagner CI, Scherer CR, Dehmelt L, Abitbol I, Gerlach U, Brendel J, Attali B, Busch AE. Molecular impact of MinK on the enantiospecific block of I(Ks) by chromanols. Br J Pharmacol 131: 1503–1506, 2000.[CrossRef][Web of Science][Medline]

33. McCarty NA, McDonough S, Cohen BN, Riordan JR, Davidson N, Lester HA. Voltage-dependent block of the cystic fibrosis transmembrane conductance regulator Cl^– channel by two closely related arylaminobenzoates. J Gen Physiol 102: 1–23, 1993.[Abstract/Free Full Text]

34. Miller C. Ion Channel Reconstitution. New York: Plenum, 1986.

35. Miller C, Moczydlowski E, Latorre R, Phillips M. Charybdotoxin, a protein inhibitor of single Ca²⁺-activated K⁺ channels from mammalian skeletal muscle. Nature 313: 316–318, 1985.[CrossRef][Medline]

36. Robbins J. KCNQ potassium channels: physiology, pathophysiology, and pharmacology. Pharmacol Ther 90: 1–19, 2001.[CrossRef][Web of Science][Medline]

37. Romey G, Attali B, Chouabe C, Abitbol I, Guillemare E, Barhanin J, Lazdunski M. Molecular mechanism and functional significance of the MinK control of the KvLQT1 channel activity. J Biol Chem 272: 16713–16716, 1997.[Abstract/Free Full Text]

38. Roussa E, Bertram J, Berge KE, Labori KJ, Thévenod F, Raeder MG. Differential regulation of vacuolar H⁺-ATPase and Na⁺/H⁺ exchanger 3 in rat cholangiocytes after bile duct ligation. Histochem Cell Biol 125: 419–428, 2006.[CrossRef][Web of Science][Medline]

39. Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL, Keating MT. Coassembly of K(V)LQT1 and minK (I_sK) proteins to form cardiac I(Ks) potassium channel. Nature 384: 80–83, 1996.[CrossRef][Medline]

40. Schroeder BC, Waldegger S, Fehr S, Bleich M, Warth R, Greger R, Jentsch TJ. A constitutively open potassium channel formed by KCNQ1 and KCNE3. Nature 403: 196–199, 2000.[CrossRef][Medline]

41. Seow KT, Lingard JM, Young JA. Anionic basis of fluid secretion by rat pancreatic acini in vitro. Am J Physiol Gastrointest Liver Physiol 250: G140–G148, 1986.[Abstract/Free Full Text]

42. Serodio P, Vega-Saenz de Miera E, Rudy B. Cloning of a novel component of A-type K⁺ channels operating at subthreshold potentials with unique expression in heart and brain. J Neurophysiol 75: 2174–2179, 1996.[Abstract/Free Full Text]

43. Sesti F, Goldstein SA. Single-channel characteristics of wild-type I_Ks channels and channels formed with two minK mutants that cause long QT syndrome. J Gen Physiol 112: 651–663, 1998.[Abstract/Free Full Text]

44. Stobrawa SM, Breiderhoff T, Takamori S, Engel D, Schweizer M, Zdebik AA, Bosl MR, Ruether K, Jahn H, Draguhn A, Jahn R, Jentsch TJ. Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to a loss of the hippocampus. Neuron 29: 185–196, 2001.[CrossRef][Web of Science][Medline]

45. Suessbrich H, Bleich M, Ecke D, Rizzo M, Waldegger S, Lang F, Szabo I, Lang HJ, Kunzelmann K, Greger R, Busch AE. Specific blockade of slowly activating I(sK) channels by chromanols—impact on the role of I_sK channels in epithelia. FEBS Lett 396: 271–275, 1996.[CrossRef][Web of Science][Medline]

46. Suzuki M, Fujikura K, Kotake K, Inagaki N, Seino S, Takata K. Immuno-localization of sulphonylurea receptor 1 in rat pancreas. Diabetologia 42: 1204–1211, 1999.[CrossRef][Web of Science][Medline]

47. Takumi T, Ohkubo H, Nakanishi S. Cloning of a membrane protein that induces a slow voltage-gated potassium current. Science 242: 1042–1045, 1988.[Abstract/Free Full Text]

48. Thévenod F. Ion channels in secretory granules of the pancreas and their role in exocytosis and release of secretory proteins. Am J Physiol Cell Physiol 283: C651–C672, 2002.[Abstract/Free Full Text]

49. Thévenod F, Anderie I, Schulz I. Monoclonal antibodies against MDR1 P-glycoprotein inhibit chloride conductance and label a 65-kDa protein in pancreatic zymogen granule membranes. J Biol Chem 269: 24410–24417, 1994.[Abstract/Free Full Text]

50. Thévenod F, Chathadi KV, Jiang B, Hopfer U. ATP-sensitive K⁺ conductance in pancreatic zymogen granules: block by glyburide and activation by diazoxide. J Membr Biol 129: 253–266, 1992.[Web of Science][Medline]

51. Thévenod F, Haase W, Hopfer U. Large-scale purification of calf pancreatic zymogen granule membranes. Anal Biochem 202: 54–60, 1992.[CrossRef][Web of Science][Medline]

52. Thévenod F, Roussa E, Benos DJ, Fuller CM. Relationship between a HCO3- -permeable conductance and a CLCA protein from rat pancreatic zymogen granules. Biochem Biophys Res Commun 300: 546–554, 2003.[CrossRef][Web of Science][Medline]

53. Thomas GP, Gerlach U, Antzelevitch C. HMR 1556, a potent and selective blocker of slowly activating delayed rectifier potassium current. J Cardiovasc Pharmacol 41: 140–147, 2003.[CrossRef][Web of Science][Medline]

54. Trube G, Rorsman P, Ohno-Shosaku T. Opposite effects of tolbutamide and diazoxide on the ATP-dependent K⁺ channel in mouse pancreatic beta-cells. Pflügers Arch 407: 493–499, 1986.[CrossRef][Web of Science][Medline]

55. Tseng CJ, Wang YJ, Liang YC, Jeng JH, Lee WS, Lin JK, Chen CH, Liu IC, Ho YS. Microtubule damaging agents induce apoptosis in HL 60 cells and G2/M cell cycle arrest in HT 29 cells. Toxicology 175: 123–142, 2002.[CrossRef][Web of Science][Medline]

56. Warth R, Garcia Alzamora M, Kim JK, Zdebik A, Nitschke R, Bleich M, Gerlach U, Barhanin J, Kim SJ. The role of KCNQ1/KCNE1 K⁺ channels in intestine and pancreas: lessons from the KCNE1 knockout mouse. Pflügers Arch 443: 822–828, 2002.[CrossRef][Web of Science][Medline]

57. Williams JA. Intracellular signaling mechanisms activated by cholecystokinin-regulating synthesis and secretion of digestive enzymes in pancreatic acinar cells. Annu Rev Physiol 63: 77–97, 2001.[CrossRef][Web of Science][Medline]

58. Xu J, Wang X, Ensign B, Li M, Wu L, Guia A. Ion-channel assay technologies: quo vadis? Drug Discov Today 6: 1278–1287, 2001.[CrossRef][Web of Science][Medline]

59. Yang IC, Scherz MW, Bahinski A, Bennett PB, Murray KT. Stereoselective interactions of the enantiomers of chromanol 293B with human voltage-gated potassium channels. J Pharmacol Exp Ther 294: 955–962, 2000.[Abstract/Free Full Text]

60. Yang Y, Sigworth FJ. Single-channel properties of I_Ks potassium channels. J Gen Physiol 112: 665–678, 1998.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: 294/4/C879    most recent 00490.2007v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lee, W.-K. Articles by Thévenod, F. Search for Related Content PubMed PubMed Citation Articles by Lee, W.-K. Articles by Thévenod, F.