AJP - GI Watch the video to learn how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 292: G1683-G1694, 2007. First published March 15, 2007; doi:10.1152/ajpgi.00268.2006
0193-1857/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/6/G1683    most recent
00268.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moser, A. J.
Right arrow Articles by Bridges, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Moser, A. J.
Right arrow Articles by Bridges, R. J.

LIVER AND BILIARY TRACT

Electrogenic bicarbonate secretion by prairie dog gallbladder

A. James Moser,1,2 A. Gangopadhyay,1 N. A. Bradbury,3 K. W. Peters,2 R. A. Frizzell,2 and R. J. Bridges3

Departments of 1Surgery and 2Cell Biology and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; and 3Department of Physiology and Biophysics, Rosalind Franklin School of Medicine, North Chicago, Illinois

Submitted 19 June 2006 ; accepted in final form 19 February 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Pathological rates of gallbladder salt and water transport may promote the formation of cholesterol gallstones. Because prairie dogs are widely used as a model of this event, we characterized gallbladder ion transport in animals fed control chow by using electrophysiology, ion substitution, pharmacology, isotopic fluxes, impedance analysis, and molecular biology. In contrast to the electroneutral properties of rabbit and Necturus gallbladders, prairie dog gallbladders generated significant short-circuit current (Isc; 171 ± 21 µA/cm2) and lumen-negative potential difference (–10.1 ± 1.2 mV) under basal conditions. Unidirectional radioisotopic fluxes demonstrated electroneutral NaCl absorption, whereas the residual net ion flux corresponded to Isc. In response to 2 µM forskolin, Isc exceeded 270 µA/cm2, and impedance estimates of the apical membrane resistance decreased from 200 {Omega}·cm2 to 13 {Omega}·cm2. The forskolin-induced Isc was dependent on extracellular HCO3– and was blocked by serosal 4,4'-dinitrostilben-2,2'-disulfonic acid (DNDS) and acetazolamide, whereas serosal bumetanide and Cl– ion substitution had little effect. Serosal trans-6-cyano-4-(N-ethylsulfonyl-N-methylamino)-3-hydroxy-2,2-dimethyl-chroman and Ba2+ reduced Isc, consistent with the inhibition of cAMP-dependent K+ channels. Immunoprecipitation and confocal microscopy localized cystic fibrosis transmembrane conductance regulator protein (CFTR) to the apical membrane and subapical vesicles. Consistent with serosal DNDS sensitivity, pancreatic sodium-bicarbonate cotransporter protein pNBC1 expression was localized to the basolateral membrane. We conclude that prairie dog gallbladders secrete bicarbonate through cAMP-dependent apical CFTR anion channels. Basolateral HCO3– entry is mediated by DNDS-sensitive pNBC1, and the driving force for apical anion secretion is provided by K+ channel activation.

disease; ion transport; cystic fibrosis conductance membrane regulator; sodium-bicarbonate symporters

ALTHOUGH GALLBLADDER DISEASE is the most frequent indication for abdominal surgery in the United States (19), the factors that confer susceptibility to gallstones remain poorly understood. Prairie dogs have been widely studied as an experimental model of human cholelithiasis due to their unique propensity for developing gallstones on high-cholesterol chow (7). Prior studies in cholesterol-fed prairie dogs (4) have demonstrated that gallbladder salt and water transport are altered before gallstones form, a phenomenon that may cause cholesterol to precipitate in the gallbladder lumen. The potential for targeting gallbladder ion transport as a strategy for preventing cholelithiasis is underscored by data showing that amiloride inhibits the formation of gallstones in cholesterol-fed prairie dogs (56). To use the prairie dog as a model to investigate this event, we performed a detailed characterization of gallbladder ion transport in animals fed control chow.

Classic studies in Necturus and rabbit gallbladders established the paradigm for electroneutral ion transport across an electrically leaky epithelium but made no attempt to identify the role of the gallbladder in gallstone formation (9, 11, 44). In addition to low transepithelial resistance (Rt), both Necturus and rabbit gallbladders generated negligible short-circuit current (Isc < 20 µA/cm2) as a result of minimal ionic conductance at the apical membrane (Ga) (40, 41). Nominal Isc and lumen-positive potential difference in these species was attributed to the back-diffusion of Na+ across a cation-selective paracellular junction (32). Significant electrogenic movement of Na+, Cl–, and HCO3– was excluded by means of pharmacological blockers, ion substitution, and microelectrode impalements. In particular, transepithelial bicarbonate transport was mediated by electroneutral parallel Na+/H+, Cl–/HCO3– exchangers, although electrogenic HCO3– secretion (<60 µA/cm2) has been observed following maximal PGE2 simulation in guinea pigs (55).

Recent observations in human and primate gallbladders challenge the prevailing concept that the gallbladder is an electrically silent absorptive organ that concentrates bile between meals and suggest the intriguing possibility that bicarbonate transport in an important cause of gallstone susceptibility. Using in vivo methods to analyze bile composition, Igimi et al. (14) and Svanvik et al. (58) found that human and monkey gallbladders absorb electrolytes at night but secrete bicarbonate-rich fluid after meals. The association of bicarbonate secretion with high values for transepithelial potential difference and Isc reported in these species suggests that bicarbonate secretion might be electrogenic and introduces the possibility that bicarbonate secretion may be an important etiologic factor in gallstone susceptibility (12, 45). Moreover, a link between aberrant luminal pH and gallstones has been observed in humans (50) as well as cholesterol-fed prairie dogs (25). Our own electrophysiological observations in normal human (A. J. Moser, unpublished data) and prairie dog gallbladders (33) demonstrate basal Isc exceeding 130 µA/cm2, a value that is inconsistent with the established paradigm for electroneutral gallbladder ion transport. The ionic basis for the significant Isc and the shared susceptibility of human and prairie dog gallbladders to develop gallstones in response to dietary cholesterol have not been investigated.

We hypothesize that the gallbladder is an integral component of the biliary tree that maintains a nonlithogenic environment for the transport of bile and cholesterol into the digestive tract. Alterations in gallbladder concentrating capacity and biliary pH as a result of bicarbonate secretion may be critical factors causing gallstone susceptibility during periods of excess dietary cholesterol intake. As a prelude to studies in cholesterol-fed animals, we performed a detailed investigation of electrogenic ion transport in normal prairie dogs by means of transepithelial currents, isotopic fluxes, ion substitution, pharmacology, impedance analysis, and molecular biology.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Surgical procedures and animal care. All experimental procedures were approved by the Institutional Animal Care and Use Committee. Adult male prairie dogs were caught in the wild under an FDA Special Permit and were fed a nutritionally complete diet (http://www.labdiet.com) containing trace cholesterol (0.02%). After a 16-h fast with water ad libitum, animals were anesthetized with intramuscular ketamine (100 mg/kg) and xylazine (10 mg/kg), and cholecystectomy was performed. Gallbladders were opened longitudinally and were rinsed with warm PBS at pH 7.4 to remove adherent bile. Tissues were mounted in 5.0-ml Ussing chambers (Navicyte, San Diego, CA) with 0.64-cm2 apertures and were heated to 37°C. Chambers were sealed with silicone grease and were gassed continuously with 95% O2-5% CO2 or 100% O2. Parafilm gaskets were used to minimize edge damage. Glass condensers reduced evaporative losses.

Electrophysiological measurements. Intact gallbladders were used because the mucosa is damaged by stripping the thin serosa. Tissues were short-circuited with fluid resistance (Rsol) compensation by using automatic voltage clamps (Department of Bioengineering, University of Iowa, Iowa City, IA) connected to a computer interface (National Instrument) running Labview data-acquisition software. Voltage and Isc measurements mere made with glass electrodes (University of Pittsburgh Machine Shop) containing Ag wire electroplated with AgCl2 and filled with 1 M KCl (8). Rt was calculated every 20 s after a 2-mV bipolar pulse. Electrode drift was checked at the conclusion of experiments by adding 1 mM serosal ouabain. Data were excluded from statistical analysis when electrode drift exceeded 2 mV.

Unidirectional ion fluxes. After Isc stabilized, radioisotopes (22Na and 36Cl or 86Rb) were added to the appropriate bath. Fluxes were measured in only one direction during each experiment to minimize errors, i.e., mucosa to serosa (ms) or serosa to mucosa (sm). Samples (400 µl) were taken in duplicate every 15 min from the unlabeled side, and volume was replaced with fresh unlabeled solution. Samples were weighed in tared scintillation vials, and the calculated volumes were used to correct the chamber volume (8). For experiments using both 22Na and 36Cl, samples were analyzed in both beta-(liquid scintillation analyzer; Packard Instrument) and {gamma}-(Auto-Gamma counting system; Packard Instrument) counters. Activities of the individual isotopes were derived from the beta- and {gamma}-decay energies of 22Na and 36Cl and the efficiencies of the respective counters. Unidirectional fluxes were calculated by using standard equations (3), and mean unidirectional ion flux was calculated from pooled data. Net flux (Jneti) was calculated [Jneti = Jmsi – Jsmi, where i represents the given isotope] as long as mean Rt differed by <20% between the unidirectional groups. Net residual ion flux (JnetR) was calculated by: JnetR = Isc – (JnetNa – JnetCl + JnetRb).

Transepithelial impedance analysis. Impedance was measured by published methods using the same Ussing chambers and electrodes employed during flux studies (52). Total epithelial capacitance was measured at five selectable frequencies (2, 4.1, 8.2, 11.0, and 16.5 kHz) in response to 99 computerized sine waves, whereas the mean voltage across the tissue (Vt) was clamped to 0 V. The fundamental frequency was 1 Hz, and the frequency range was 1 Hz–22 kHz.

Curve-fitting routines derived from a one- or two-membrane equivalent electrical circuit were used to estimate individual membrane parameters from the acquired data (52). We modeled the intact gallbladder as a series arrangement of two independent resistor-capacitor elements representing the apical (Ra and Ca) and basolateral (Rb and Cb) membranes in parallel with a shunt resistance (Rp), representing the paracellular resistance. The term Rseries represented the sum of the bathing solution (Rsol) and subepithelial connective tissue (Rsub) resistances. This two-membrane equivalent circuit is the simplest morphologically correct model of epithelial structure and has been used previously for studies of human bronchial epithelial cells (20), T84 cells (52), and Necturus gallbladder (16, 47). Conversely, the one-membrane equivalent circuit modeled a single cellular membrane capacitance in parallel with a low-resistance shunt.

Estimation of Rp. The use of curve-fitting routines to analyze impedance data requires an independent estimate of one circuit element. Because mammalian gallbladder is notoriously difficult to impale with microelectrodes, we estimated Rp by linear regression analysis of the transepithelial conductance (Gt) and Isc data in each experiment (61). To confirm constant Rp, we first measured mucosa-to-serosa fluxes of [3H]mannitol under basal conditions and found <20% variability (n = 3; data not shown). We subsequently measured the passive unidirectional flux of 22Na (JsmNa) during the basal period (5.9 ± 0.6 µeq·cm–2·h–1; n = 8) and in response to both 10 µM indomethacin (6.2 ± 0.7 µeq·cm–2·h–1; n = 8) and 2 µM forskolin (6.4 ± 0.6 µeq·cm–2·h–1; n = 8) to obtain limiting estimates of Rp. Because the passive flux of Na+ is equal to the partial ionic conductance of Na+ (49), the maximum estimate of Rp was 169 ± 16 mS/cm (Rp= 1,000/JsmNa). Given the relationship between Rt and the cellular (Rcell) and shunt resistances (1/Rt = 1/Rcell + 1/Rp; equation 1), the minimum value of Rp in each experiment was equal to Rt. Although Isc and Rt changed dramatically in response to indomethacin and forskolin, mean JsmNa varied by <10%, suggesting that Rp remained constant.

Having defined the boundaries of Rp (169 ± 16 mS/cm2 > Rp > Rt), we plotted unique estimates of Rp during each experiment by linear regression analysis of the Gt and Isc data according to Gt = Isc/Vt + Gp (equation 2), where Gp is paracellular conductance, and used this value in curve-fitting analysis. Despite having low Rp, prairie dog gallbladder satisfied the major methodological requirements for estimating the individual membrane parameters by impedance techniques: 1) the plot of Gt vs. Isc was linear in each experiment, and r = 0.84 ± 0.05 (P < 0.0001) in 10 experiments; 2) Rp remained constant at different values of Isc as demonstrated by stable passive fluxes of [3H]mannitol and 22Na; and 3) the best-fit impedance function correlated well with measured impedance under all experimental conditions, as reflected by the low normalized error (0.047 ± 0.008) of the curve-fitting routine. The suitability of intact gallbladder to impedance measurements may be attributed to its homogeneous cell population, limited epithelial folding, and the minimal constant Rsub (6.2 ± 1.9 {Omega}·cm2). Rsub was calculated (Rsub = Rseries – Rsol) by measuring Rsol during instrument setup and Rseries during impedance measurements at different values of Isc.

Power law dependence and curve fitting parameters. Deviations of actual impedance from the best-fit ideal were assessed by using the power law factor ({gamma}) and the curve-fitting parameters normalized error (norm) and r. The {gamma} measured the Cole-Cole power law dependence of the membrane dielectric and described the divergence of the impedance locus from a circular arc (15, 52). The norm was a measure of the percent difference between the observed and fitted impedance at each frequency. Although we assumed that intact gallbladder would exhibit complex dielectric properties manifested by center suppression of the impedance locus as well as frequency-dependent dispersions of the complex capacitance from the best-fit ideal, actual impedance deviated minimally from the best-fit curves below 7 Hz and above 6 kHz (data not shown). This observation suggested minimal dispersion of the membrane dielectric throughout the frequency range, consistent with calculated values of {gamma} close to unity (see GoGoGoTable 4).


View this table:
[in this window]
[in a new window]

  		Table 1. Electrical parameters during the basal period and in response to 10 µM indomethacin and increasing doses of forskolin

 

View this table:
[in this window]
[in a new window]

  		Table 2. Effects of pharmacological blockers and ion substitution on forskolin-induced ISC

 

View this table:
[in this window]
[in a new window]

  		Table 3. Unidirectional fluxes of Na+ and Cl– during the basal period

 

View this table:
[in this window]
[in a new window]

  		Table 4. Impedance estimates of individual membrane resistance during the basal period and in response to both indomethacin and forskolin

 
Because we did not observe center suppression of the impedance locus or deviations of the high- and low-frequency data from the best-fit ideal indicating a capacitative property of the junctional complex, we concluded that the junctional pathway could be represented as a single resistive element in the equivalent circuit without oversimplifying the electrical behavior of the tight junction and lateral intercellular space (16, 36, 60). This observation permitted the classic two-membrane equivalent circuit to be used as a model of epithelial impedance. The low resistance of the paracellular shunt (Rp < 100 {Omega}·cm2) reduced the shape of the Nyquist plots toward a single semicircle. Because the leaky nature of the gallbladder required a wide range of frequencies to calculate impedance, we focused this report on estimates of the Ra and Rb, given the potential for capacitance to be frequency-dependent when Ra {cong} Rp Rb (36, 37). Unlike capacitance, individual membrane resistance is calculated from the real term of impedance (ZR) and is frequency-independent (1, 23, 29) given the relationship Z (j{omega}) = V /I, which simplifies to ZR = R (23).

Bathing solutions. The composition of the bicarbonate Ringer solution was (in mM): 114 NaCl, 25 NaHCO3, 4.8 KCl, 2.4 Na2HPO4, 0.6 KH2PO4, 1.2 MgCl2, 1.2 CaCl2, 10 mucosal mannitol, and 10 serosal glucose, pH 7.4. The solution was gassed with 95% O2-5% CO2 at 37°C. For bicarbonate-free experiments, 25 mM Na-gluconate was substituted for NaHCO3, 10 mM HEPES was added, and the solution was gassed with 100% O2. For Cl–-free experiments, equimolar gluconate salts of Na+, K+, and Mg2+ replaced NaCl, KCl, and MgCl2. Ca-gluconate (4 mM) replaced 1.2 mM CaCl2 to compensate for the Ca2+-buffering capacity of gluconate.

Chemicals. 4,4'-Dinitrostilben-2,2'-disulfonic acid (DNDS) was synthesized by Tokyo Kasei Kogyo (Tokyo, Japan). Trans-6-cyano-4-(N-ethylsulfonyl-N-methylamino)-3-hydroxy-2,2-dimethyl-chroman (293B) was a gift from Dr. Rainer Greger (Albert-Ludwigs-Universität, Freiberg, Germany). Forskolin and apamin were purchased from Calbiochem. Acetazolamide, amiloride, BaCl2, bumetanide, DIDS, indomethacin, ouabain, phloridzin, sodium taurocholate, and tetrodotoxin were purchased from Sigma. Amiloride, tetrodotoxin, phloridzin, sodium taurocholate, BaCl2, and ouabain were made as stock solutions in H2O. DNDS and DIDS were dissolved in buffer immediately prior to addition. Forskolin, acetazolamide, and 293B were dissolved in DMSO, whereas bumetanide was dissolved in ethanol. Charybdotoxin was obtained from Accurate Chemical and Scientific and was made as a 10 µM stock solution in buffer. All compounds prepared in ethanol or DMSO were made as 1,000-fold stock solutions to keep solvent concentration in the bath at <0.1%. 22Na, 36Cl, and 86Rb were purchased from NEN Life Science Products (Boston, MA).

RT-PCR. Total RNA was extracted from intact gallbladders by using Tri-Reagent (Sigma) and was treated with DNase 1 (Ambion). First-strand cDNA synthesis and PCR were performed by published methods (21) using a Robo ThermoCycler (Stratagene). Cystic fibrosis transmembrane conductance regulator (CFTR)-specific primers (forward: 5'-TTGGAATGCAGATGAGAATACC-3'; reverse: 5'-CCCTGAGAAGAAGAAGGCTGA-3') amplified a 499-bp region of the published prairie dog sequence (GenBank accession no. AF012893). Pancreatic sodium-bicarbonate cotransporter (pNBC1) primers (forward: 5'-GGATGAAGCTGTCCTGGACAG-3'; reverse: 5'-CCAAGAAGCTGGCATCAGTGGC-3') amplified a 1.7-kb region of the published prairie dog sequence (GenBank accession no. DQ431115). Annealing temperatures for the CFTR and pNBC1 reactions were 59°C and 63°C, respectively. Controls were performed by omitting reverse transcriptase or by adding only water to the reaction.

Immunoprecipitation and phosphorylation of CFTR. CFTR immunoprecipitation was performed on gallbladder tissue homogenates by using monoclonal anti-CFTR antibody, N-Terminus (L12B4; Upstate Biotechnology, Lake Placid, NY) as described (2, 13). Wild-type HEK-293 cells transfected with CFTR served as a positive control. Negative controls included untransfected HEK-293 cells in addition to prairie dog skeletal muscle and liver as a test of nonspecific binding. Immunoprecipitates were labeled by 32P phosphorylation, and the reaction products were resolved on 6% SDS-PAGE.

Immunofluorescence of CFTR and pNBC1. Freshly collected tissues were washed with ice-cold PBS and were fixed with 2% paraformaldehyde for 2 h at 4°C, followed by treatment with 30% sucrose overnight. Cryostat sections were blocked with 2% BSA for 40 min, followed by incubation with primary antibody in 0.5% BSA for 1 h. Sections were labeled with the following antibodies: 1) 1:100 dilution of rabbit polyclonal anti-pNBC1 antibody (21); and 2) 1:100 dilution of mouse monoclonal antibody against the NH2 terminus of CFTR (L12B4). After three washes with 0.5% BSA, the sections were further incubated with secondary antibody, goat anti-rabbit Alexa 488 (1:500; Molecular Probes) for pNBC1 or goat anti-mouse Alexa 488 (1:500; Molecular Probes) for CFTR. Samples were counterstained with rhodamine phalloidin (red) for 40 min at room temperature to label the apical membrane. Nonimmune isotype IgG followed by exposure to secondary antibody served as a control. Prairie dog heart was used as a second negative control for CFTR expression. Slides were visualized by using an Olympus Fluoview 500 confocal microscope.

Data analysis. Data are expressed as means ± SE; n indicates the number of experiments. Comparisons between groups were made by ANOVA followed by Fisher's exact test or by Student's paired t -test as appropriate by using StatView software. Significance was assumed when P < 0.05. For net flux data, experimental error was calculated by the method of error propagation. Curve-fitting of the impedance data was performed by using BLIMP (52) or a curve-fitting routine written for Micromath Scientist.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Basal Isc. We studied 72 gallbladders from prairie dogs fed nonlithogenic chow. Gallbladders were mounted in bicarbonate Ringer solution. Isc (79 ± 22 µA/cm2) and Gt (13.0 ± 0.8 mS/cm2) remained stable for an average of 7 min with a calculated lumen-negative Vt of –6.2 ± 1.8 mV (Table 1). Afterward, both Isc and Gt increased spontaneously for ~30 min (range, 4–46 min) before stabilizing at 171 ± 21 µA/cm2 (P < 0.01, ANOVA) and 16.3 ± 1.5 mS/cm2. Regression analysis of Gt vs. Isc during the spontaneous rise in Isc was a linear function with a slope Gp of 12.3 ± 1.0 mS/cm2, equal to 84 ± 3% of total Gt. Because stimulatory VIPergic neurons have been reported in guinea pig gallbladder submucosa (39), we added 3 µM tetrodotoxin to the serosal bathing solution (n = 3; data not shown) to block potential neuronal stimulation of Isc but observed no effect. Similarly, adding 10 mM glucose (n = 5), 100 µM phloridzin (n = 3), or 30 mM taurocholic acid (n = 4) to the mucosal solutions to probe for electrogenic Na+-coupled glucose or bile salt cotransport had no effect on Isc.

We therefore hypothesized that cholecystectomy triggered endogenous prostaglandin synthesis and added 10 µM indomethacin to the bathing solutions once Isc stabilized (33, 35). As expected, indomethacin reduced Isc and Gt to 77 ± 18 µA/cm2 (n = 10) and 11.6 ± 1.0 mS/cm2 (n = 10) over 26 min (range, 13–41 min), in keeping with the inhibition of a prostaglandin-dependent ionic conductance. To reproduce the stimulatory effects of prostaglandins on Isc, we used increasing doses of forskolin to raise cAMP production. Isc increased almost immediately, achieved a transient peak within 3 min, and then settled back to a sustained plateau. Forskolin (2 µM) was a maximal stimulus, raising Isc to 241 ± 21 µA/cm2 and Gt to 23.1 ± 2.0 mS/cm2 in 10 experiments (P < 0.01 vs. indomethacin, ANOVA). Forskolin (200 nM) had intermediate effects on Isc (124.6 ± 27.9 µA/cm2) and Gt (15.7 ± 2.3 mS/cm2) that were statistically significant (P = 0.05, ANOVA) compared with the indomethacin period. Because indomethacin and forskolin replicated the transport phenotypes observed before and after the spontaneous rise in Isc, we used 10 µM indomethacin and 2 µM forskolin in subsequent experiments to create stable conditions for mechanistic studies.

Ionic basis of the forskolin-induced Isc. To determine the ionic basis of the forskolin-induced current, we first tested a series of apical cation-channel blockers. Gallbladders were exposed to both low (10 µM) and intermediate doses (50 µM) of the Na+-channel blocker amiloride under four conditions: during the spontaneous rise in Isc, after 10 µM indomethacin, and after both 200 nM and 2 µM forskolin. Neither 10 µM nor 50 µM amiloride had any effect on Isc or Gt under any of the four experimental conditions (n = 9; data not shown). Mucosal application of Ba2+ (2 mM BaCl2), a nonspecific K+-channel blocker, also had no effect on Isc (n = 4).

Having excluded amiloride and Ba2+-sensitive apical cation channels, we hypothesized that forskolin stimulated cAMP-dependent anion secretion (Fig. 1). We used 20 µM serosal bumetanide to block the basolateral Cl–-entry component of apical Cl– secretion but observed minimal inhibition (26 ± 5 µA/cm2 or 11 ± 2%; n = 10) of the forskolin-induced current. Because Cl– secretion did not account for Isc, we next probed electrogenic HCO3– secretion as described in pancreatic duct (30) and Calu-3 cells (8). We used 3 mM serosal DNDS (46) to block the potential Cl–/HCO3– anion exchanger and Na+-HCO3– cotransporter mechanisms of basolateral HCO3– entry. Sequential treatment with serosal DNDS and 100 µM acetazolamide inhibited approximately half of the forskolin-induced rise in Isc. Serosal DNDS (3 mM) decreased Isc by 66 ± 15 µA/cm2, whereas100 µM acetazolamide inhibited an additional 52 ± 4.8 µA/cm2. Serosal DIDS (500 µM) had no extra effect in the presence of serosal DNDS (data not shown). We concluded that forskolin-induced Isc was mostly the result of transepithelial HCO3– secretion linked to DNDS-sensitive basolateral HCO3– entry and metabolic HCO3– production.


Figure 1
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 1. Current trace from a typical experiment using pharmacological blockers to inhibit the forskolin-stimulated short-circuit current (Isc). Gallbladders were pretreated with 10 µM indomethacin and 2 µM forskolin. Pharmacological inhibitors were added to the indicated solutions while Isc and transepithelial resistance (Rt) were measured. Ouabain was used to measure electrode drift. DNDS, 4,4'-dinitrostilben-2,2'-disulfonic acid; M, mucosal; S, serosal.

 
Despite adding blockers of Cl– and HCO3– secretion to the bathing solutions, significant Isc (118 ± 11 µA/cm2) persisted in forskolin-treated tissues. This finding suggested residual anion secretion insensitive to the pharmacological inhibitors. To test the hypothesis that residual Isc depended on basolateral membrane hyperpolarization caused by K+ efflux, we added 100 µM serosal 293B to block the cAMP/PKA-dependent K+ channel KvLQT1 (27). 293B reduced Isc significantly (38.8 ± 16.5 µA/cm2; n = 4; P = 0.05, paired t-test), whereas 20 nM serosal charybdotoxin (Ca2+-sensitive K+ channel blocker) and 100 nM serosal apamin (SK channel blocker) had no effect (data not shown). Serosal BaCl2 (2 mM) inhibited Isc transiently (24.4 ± 3.7 µA/cm2; n = 7; P = 0.05, ANOVA), a finding that we attributed to the precipitation of insoluble Ba2+ salts in phosphate- and HCO3–-buffered Ringer. All of the residual Isc was eliminated by 1 mM serosal ouabain (–13 ± 15 µA/cm2; P < 0.0001, ANOVA), confirming the dependence of Isc on serosal Na+-K+-ATPase activity (Fig. 1).

Effects of ion substitution on forskolin-induced Isc. To identify the transporters mediating basolateral anion entry, we repeated these pharmacological studies in HCO3–-free buffer. As shown in Table 2, the forskolin response in HCO3–-free buffer was reduced by >80% compared with standard conditions ({Delta}Isc = 26 ± 3.5 µA/cm2 vs. 170 ± 35 µA/cm2; n = 5; P < 0.001, ANOVA), and 20 µM serosal bumetanide blocked Isc completely. Neither 3 mM serosal DNDS nor 100 µM acetazolamide had any effect on Isc in the absence of HCO3–. We concluded that 1) forskolin activated a cAMP-dependent Cl– current dependent on basolateral Na-K-2Cl cotransport, 2) the inhibitory effect of 3 mM serosal DNDS required extracellular HCO3–, 3) metabolic HCO3– production did not contribute to Isc in the absence of exogenous CO2, and 4) removing HCO3– from the buffer solution inhibited the forskolin-induced Isc significantly more than the sequential additions of serosal DNDS and acetazolamide in standard buffer (P < 0.008, ANOVA). We inferred from this observation that DNDS and acetazolamide did not completely block HCO3– secretion under standard conditions, a finding that accounted for residual Isc at the conclusion of the blocker studies.

Because the magnitude of the forskolin-induced Isc decreased significantly in the absence of HCO3–, we hypothesized that cAMP stimulated HCO3– secretion in preference to Cl– under standard conditions. To determine the basolateral entry mechanism for HCO3–, we removed Cl– from the bath to block Cl–/HCO3– exchange as well as Na-K-2Cl cotransport. As shown in Table 2, removing Cl– from the bathing solution inhibited the forskolin-induced Isc (P < 0.02) significantly more than 20 µM serosal bumetanide did in standard buffer, suggesting that DNDS-sensitive Cl–/HCO3– exchange contributed to the forskolin-induced Isc in standard buffer. In Cl–-free solution, serosal bumetanide had no effect, whereas 3 mM serosal DNDS and 100 µM acetazolamide inhibited >50% of forskolin-induced Isc (P < 0.01, ANOVA). This inhibitor profile in HCO3–-free solution was consistent with transepithelial HCO3– secretion driven by carbonic anhydrase and Cl–-independent, serosal DNDS-sensitive HCO3– entry characteristic of basolateral Na+-HCO3– cotransport.

Unidirectional ion fluxes. To determine the ionic basis of Isc in standard buffer, we measured unidirectional 22Na, 36Cl, and 86Rb fluxes in 16 normal gallbladders (Table 3). After the spontaneous run-up in Isc, gallbladders generated stable current (166 ± 18 µA/cm2 or 6.2 ± 0.7 µeq·cm–2·h–1) with a calculated lumen-negative potential difference of –10.5 ± 0.9 mV. JmsNa was twice the rate of passive JsmNa under these conditions, generating net Na+ absorption at 6.5 ± 1.0 µeq·cm–2·h–1. Similarly, JmsCl significantly exceeded JsmClS, resulting in net Cl– absorption (4.3 ± 1.0 µeq·cm–2·h–1) nearly equal to the rate of net Na+ absorption. Subsequent measurements of unidirectional 86Rb fluxes excluded net 86Rb movement as a component of Isc (data not shown). Given minor differences between the rates of net Na+ and Cl– absorption, we attributed Isc primarily (70%) to the net flux of unmeasured ions across the epithelium (JnetR). Because HCO3– was the only unmeasured ion in abundance and JnetRb was not significant, we concluded that transepithelial HCO3– secretion accounted for JnetR. Moreover, the drop in Isc caused by adding DNDS and acetazolamide (115 ± 15 µA/cm2) was equal to the measured value of JnetR(4.0 ± 1.4 µeq·cm–2·h–1 or 107 ± 37 µA/cm2), further evidence that Isc was attributable to HCO3– secretion.

Transepithelial impedance analysis. To estimate the resistances of the individual membranes at varying levels of Isc, we performed impedance analysis during the basal period and in response to indomethacin and forskolin. Fourteen normal tissues were evaluated in standard buffer. After tissues were mounted, Isc rose spontaneously and stabilized within 40 min (Fig. 2A). The corresponding Nyquist plots (Fig. 2B) were shifted to the right of the origin of the real axis of impedance by a Rseries equal to the sum of Rsol and Rsub. The statistical correlation between measured impedance and the fitted Nyquist plots was significant at all time points (norm < 0.05), indicating that the classic two-membrane equivalent electrical circuit was a valid model of the intact gallbladder. The impedance locus at time 0 described a semicircle with slight flattening of the low-frequency arc and minimal center suppression ({gamma} = 0.976) compared with the ideal ({gamma} = 1.0, dashed line). The high-conductance shunt Rp magnified the impedance locus of the apical membrane relative to that of the higher-resistance basolateral membrane (36).


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Impedance data from a typical experiment using intact gallbladder. A: Isc tracing from the basal period and after response to 10 µM indomethacin (Indo) and increasing concentrations of forskolin (Forsk). After gallbladder was mounted at time 0, Isc rose spontaneously for 35 min and stabilized prior to the subsequent additions of indomethacin and forskolin. B: Nyquist plots of baseline impedance measurements at 0, 30, and 40 min after mounting. Symbols denote the acquired data, and solid lines show best-fit Nyquist plots. As Isc increased, the impedance loci evolved distinct apical and basolateral semicircles consistent with falling apical (Ra) and basal resistance (Rb) (inset). Rp, paracellular resistance; Zi, imaginary term of impedance; ZR, real term of impedance. C: impedance loci acquired 0, 20, and 35 min after 10 µM indomethacin. As Isc fell, the Nyquist plots reverted to the shape observed at the beginning of the basal period. Indomethacin increased the estimates of Ra and Rb significantly (inset). D: Nyquist plots following 200 nM and 2 µM forskolin. The 200 nM forskolin flattened the low-frequency impedance arc and reduced both total impedance and Rt. The 2 µM forskolin generated 2 intersecting loci of impedance with visually-identifiable characteristic frequencies for the apical and basolateral membranes (inset).

 
The spontaneous rise of Isc after mounting was accompanied by progressive flattening of the low-frequency impedance arc and falling Rt, where Rt was represented by the intersection of the low-frequency impedance data with the real axis. These changes in impedance were observed in all experiments and indicated that the Ga and Gb increased as Isc rose (Table 4). The change in conductance was most pronounced at the apical membrane, where Ra decreased from 203 ± 81 {Omega}·cm2 to 47 ± 9 {Omega}·cm2 (Table 4). Rising Gb (falling Rb) was consistent with basolateral K+ channel activation demonstrated by the inhibitory effects of serosal 293B and Ba2+ on Isc. Neither 10 nor 50 µM amiloride had any affect on Ra to indicate amiloride-sensitive Na+ channel expression (data not shown). Despite falling Rt, Rsub (6.2 ± 1.9 {Omega}·cm2) remained constant across the range of Isc, indicating that the subepithelial connective tissue was not affected by the prostaglandin stimulation associated with mounting tissues in the chamber.

These changes in transepithelial impedance were reversed by adding 10 µM indomethacin (Fig. 2C) to the bathing solutions. Indomethacin inhibited Isc and shifted the low-frequency data to higher impedance, consistent with rising Rt. The corresponding Nyquist plots were consistent with the two-membrane electrical circuit model for an average of 11 min in all experiments. In five tissues, the Nyquist plots adopted a single impedance locus as Isc approached its minimum value, meaning that the two-membrane equivalent electrical circuit could not be used to derive independent estimates of apical and basolateral membrane resistance. Because curve-fitting analysis could not be performed at nadir Isc in all experiments, we inferred that the mean values for Ra and Rb in Table 4 probably underestimated indomethacin's maximum effect. Nonetheless, Ra increased from 47 ± 9 {Omega}·cm2 to at least 157 ± 61 {Omega}·cm2, whereas Rb increased from 229 ± 73 {Omega}·cm2 to 1,102 ± 223 {Omega}·cm2. Because stable values of Rseries and JsmNa demonstrated that the Rp and Rsub remained constant after indomethacin, we concluded that indomethacin raised Rt and reduced Isc by inhibiting the prostaglandin-dependent Ga and Gb that carried the transepithelial anion current.

Conversely, forskolin stimulated Isc and restored the Nyquist plots to the shape observed during the spontaneous rise in Isc. These data demonstrate stimulation of the Ga and Gb by cAMP. Forskolin (200 nM) reduced total impedance, flattened the low-frequency arc, and decreased Rt. Forskolin (2 µM) was a maximum stimulus to Isc and generated two distinct impedance semicircles with visually identifiable characteristic frequencies (Fig. 2D) for the apical and basolateral membranes.

Because 2 µM forskolin reduced Rt by 40%, we next performed a range analysis to test the hypothesis that Rp was altered by forskolin as well. In proportion to the gallbladder's low baseline Rp, small changes in absolute Rp have a large relative effect on the individual membrane resistances estimated by curve-fitting algorithms. The assumption that Rp remained constant after 2 µM forskolin rested on the absence of a statistically significant change in the unidirectional fluxes of [3H]mannitol and JsmNa and the constant slope of the Gt-vs.-Isc plots. We therefore picked a range of potential Rp values that were statistically equivalent due to experimental errors in the unidirectional mannitol flux and JsmNa. Assuming a constant standard deviation (S = 1.7 µeq·cm–2·h–1) of mean JsmNa, a 40% drop in Rp (corresponding to the 40% drop in Rt after forskolin) would increase predicted JsmNa to 9.0 ± 1.7 µeq·cm–2·h–1 according to equation 2, a value that is significantly greater (P = 0.004) than observed JsmNa (6.4 ± 1.7 µeq·cm–2·h–1) after forskolin. Given this potentially significant change in JsmNa, we concluded that 40% was the maximum possible effect of forskolin on Rp. Our subsequent range analysis accounted for a potential 40% decrease in Rp in each experiment subject to the constraints of equation 1, which requires that measured Rt always be the minimum value of Rp. The analyzed range of Rp included all of the values: 85 {Omega}·cm2 > Rp > 51 {Omega}·cm2, where the standard deviation of mean Rp was 21 {Omega}·cm2.

Recalculating the impedance data in each experiment with this unique range of Rp values (Table 4) demonstrated that a 40% drop in Rp had no significant effect on the impedance estimates of Ra and produced a relatively narrow range of Rb estimates. These results are similar to the analysis of rabbit colon by Wills and Clausen (60), in which impedance estimates of Ra and Rb were reasonably robust against errors in Rp up to 2SD of mean Rp in magnitude. In prairie dog gallbladder, forskolin's effect predominated at the apical membrane, where Ra fell into a range of values between 12.6 ± 2.0 and 13.3 ± 2.2 {Omega}·cm2 as Isc exceeded 250 µA/cm2 (Table 4). The ranges of Ra and Rb values resulting from this analysis were significantly lower than estimates obtained in 10 µM indomethacin, indicating that cAMP activated highly significant ionic conductances in the apical and basolateral membranes. Furthermore, the estimated values of Ra at peak forskolin effect (13 ± 2 {Omega}·cm2) were commensurate with the remarkable rate of HCO3– secretion demonstrated by Isc (267 ± 36 µA/cm2; n = 10). Although estimates of Rb resulting from the Rp range analysis were less robust than Ra, changes in Rb were still statistically significant compared with the indomethacin values and were consistent with the activation of cAMP-dependent basolateral K+ channels sensitive to both 293B and serosal Ba2+.

Membrane localization of CFTR and pNBC1. Activation of cAMP-dependent Ga is consistent with the function of CFTR protein. RT-PCR analysis of total gallbladder RNA with CFTR-specific primers demonstrated a 499-bp cDNA with 90% sequence homology to human CFTR (Fig. 3A). Subsequent immunoprecipitation with monoclonal anti-CFTR antibody (L12B4) confirmed the synthesis of CFTR protein. As shown in Fig. 3B, both prairie dog gallbladder (lane 3) and CFTR-transfected HEK-293 cells (lane 1) expressed a 190-kDa protein corresponding to fully-glycosylated band C CFTR (13). The 190-kDa band was not observed in untransfected HEK-293 cells (lane 2) or the negative prairie dog controls. Given the specificity of the L12B4 monoclonal antibody for prairie dog CFTR, we performed confocal microscopy (Fig. 4) on intact tissues and observed CFTR immunoreactivity primarily at the apical membrane, with diffuse staining of the apical cytoplasm corresponding to the intracellular distribution of CFTR in subapical vesicles (59). The absence of apical signal with idiotypic antibody (Fig. 4C) on prairie dog skeletal muscle (Fig. 4D) excluded nonspecific antibody binding.


Figure 3
View larger version (64K):
[in this window]
[in a new window]

  		Fig. 3. Molecular detection of cystic fibrosis transmembrane conductance regulator (CFTR) in prairie dog gallbladder. A: amplified CFTR PCR product on 1.5% agarose gel stained with ethidium bromide. Lane 1, 100-bp DNA ladder; lane 2, water control; lane 3, no reverse transcriptase control; lane 4, CFTR PCR product at 499 bp. B: autoradiograph of 32P-labeled gallbladder immunoprecipitates using anti-CFTR L12B4 monoclonal antibody. Identical concentrations of extracted protein were loaded in each lane. Lane 1, wild-type HEK-293 cells transfected with CFTR showing fully glycosylated band C CFTR at 170–200 kDa (positive control); lane 2, untransfected HEK-293 cells (negative control); lane 3, intact prairie dog gallbladder showing the expected band at 170–200 kDa; lane 4, prairie dog skeletal muscle (negative control); lane 5, prairie dog liver (negative control).

 

Figure 4
View larger version (80K):
[in this window]
[in a new window]

  		Fig. 4. Confocal immunofluorescence microscopy of CFTR and beta-actin using double-labeled Alexa 488 conjugated anti-CFTR L12B4 monoclonal antibody (green) and rhodamine phalloidin (red). A: L12B4 labeling of the apical membrane and subapical vesicles of intact prairie dog gallbladder epithelium. Rhodamine phalloidin (red) staining has been digitally removed from this image. B: combined immunofluorescence image demonstrating phalloidin staining of the actin cytoskeleton along the apical aspect and green L12B4 labeling of apical CFTR. C: goat anti-mouse IgG control demonstrating faint background Alexa 488 fluorescence and apical rhodamine binding. D: double-label immunofluorescence of prairie dog skeletal muscle with L12B4 and rhodamine phalloidin (negative control).

 
To ascertain the mechanism for Cl–-independent, DNDS-sensitive HCO3– movement across the basolateral membrane, we next used RT-PCR to probe for pNBC1 expression. pNBC1-specific primers amplified a 1.7-kb cDNA with 94% homology to human pNBC1 but not the renal splice variant kNBC1 (Fig. 5A). As shown by immunofluorescence microscopy (Fig. 5, B–D), prairie dog gallbladder localized pNBC1 protein to the basolateral membrane in keeping with the site of DNDS sensitivity. The negative isotype IgG control confirmed that antibody binding was specific. Furthermore, the deduced amino acid sequence of the prairie dog pNBC1 cDNA was 100% identical to the 22-amino acid human pNBC1 polypeptide used to generate the rabbit polyclonal antibody.


Figure 5
View larger version (66K):
[in this window]
[in a new window]

  		Fig. 5. Expression and localization of pancreatic sodium-bicarbonate transporter (pNBC-1) in prairie dog gallbladder. A: pNBC1 PCR product on 1.5% agarose gel stained with ethidium bromide. Lane 1, DNA ladder; lane 2, water control; lane 3, no reverse transcriptase control; lane 4, 1.7-kb PCR product of pNBC1. B: detection of pNBC1 and beta-actin by confocal immunofluorescence microscopy using double-labeled Alexa 488 conjugated anti-pNBC1 polyclonal antibody (green) and rhodamine phalloidin (red). Anti-pNBC1 labeling of the basolateral membrane is clearly evident. Rhodamine phalloidin stains the actin cytoskeleton along the apical aspect. C: anti-pNBC1 labeling of the basolateral membrane. Rhodamine phalloidin (red) staining has been digitally removed to show the prominent basolateral Alexa 488 signal. D: goat anti-rabbit IgG control demonstrating very faint Alexa 488 background fluorescence.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Ion transport across prairie dog gallbladders differed significantly from the electroneutral model established by previous studies in rabbits, Necturus, and guinea pigs. Despite sharing high Gp (12.3 ± 1.0 mS/cm2) with the gallbladders of these other species, prairie dog gallbladders generated remarkable basal Isc (79 ± 22 µA/cm2), which rose spontaneously to 171 ± 21 µA/cm2 as Rt fell. Impedance analysis correlated rising Isc with progressive activation of ion channels in the apical and basolateral membranes, during which Ra decreased from ~200 {Omega}·cm2 to ~50 {Omega}·cm2 and Rb decreased from 800 {Omega}·cm2 to 200 {Omega}·cm2. The changes in Isc and membrane conductance were reversed by indomethacin and were replicated by forskolin, consistent with cAMP-dependent activation of the relevant ionic conductances. Tetrodotoxin had no effect on Isc to suggest VIPergic neuronal stimulation within the intact submucosa. Prostaglandin synthesis was linked to cholecystectomy and probably represented an inflammatory response to tissue manipulation.

Transepithelial HCO3– secretion. The combination of low apical membrane resistance and high resting Isc in prairie dog gallbladder is a unique finding in an epithelium formerly regarded as electrically silent (5, 6). On the basis of results obtained with pharmacological blockers and ion substitution, we attributed the forskolin-induced Isc primarily to HCO3– secretion (89%; 249 ± 27 µA/cm2) with a smaller rate of Cl– secretion (11%; 24 ± 4.8 µA/cm2). By itself, forskolin-sensitive Ga is not a novel finding in the gallbladder and has been reported in both Necturus (5) and guinea pigs (62). By comparison with the modest forskolin-induced current of 0.5 µeq·cm–2·h–1 in Necturus, resting HCO3– current in prairie dog gallbladder was 4.0 ± 1.4 µeq·cm–2·h–1 and increased to 9.3 ± 1.0 µeq·cm–2·h–1 after 2 µM forskolin (16–18). The cAMP response of the Ga pathway coupled with its apparent permeability to both HCO3– and Cl– in ion-substituted buffer was consistent with our molecular demonstration of CFTR protein and its localization to the apical membrane. Our findings corroborate previous reports of CFTR expression in human (10), mouse (38), and canine gallbladders (22). Although CFTR appears to be uniformly expressed among gallbladder epithelia, prairie dog gallbladders generated much larger forskolin-induced anion currents with lower Ra (13 ± 2 {Omega}·cm2) than other species. In the absence of patch-clamp data, our identification of the Ga as CFTR must be tempered by the observation that other channels, notably pCLCA1, can mediate cAMP-dependent anion secretion (26).

The observation of cAMP-dependent bicarbonate secretion across prairie dog gallbladder is more analogous to the function of pancreatic duct epithelium than the electrically silent gallbladders of Necturus and rabbits. The high rate of HCO3– secretion gives physiological relevance to our identification of three different mechanisms for HCO3– uptake, including basolateral Na+-HCO3– cotransport, Cl–/HCO3– exchange, and metabolic production by carbonic anhydrase. Given the magnitude of the DNDS-sensitive Isc in Cl–-free buffer, the majority of transepithelial HCO3– secretion was mediated by basolateral sodium-bicarbonate cotransporter activity, consistent with our demonstration of basolateral pNBC1 expression. pNBC1 is the pancreatic isoform of the Na+-HCO3– cotransporter that mediates electrogenic HCO3– uptake in pancreatic duct and Calu-3 cells, consistent with its role in transepithelial HCO3– secretion. Basolateral Na+-HCO3– cotransport has been reported in guinea pig gallbladder, where it generates electroneutral HCO3– secretion via apical Cl–/HCO3– exchange at 1.0 µeq·cm–2·h–1 (62). In addition to cotransporter activity, we also observed inhibition of the forskolin-induced Isc by Cl–-free buffer, suggesting Cl–-dependent HCO3– exchange across the basolateral membrane. This functional evidence for a second HCO3– uptake mechanism confirms our prior report that prairie dog gallbladder expresses anion exchanger 2 (33a), a basolateral isoform of the Cl–/HCO3– exchanger that forms membrane-bound complexes with carbonic anhydrase (57). Data showing that carbonic anhydrase is required for maximal Cl–/HCO3– exchanger activity in transfected HEK-293 cells (54) may explain the effects of acetazolamide on Isc in our experiments as well as the reduction of HCO3– secretion by acetazolamide in both murine (31) and mouse (34).

Impedance measurements of Ga. Previous studies correlating membrane resistance with electrogenic HCO3– secretion in rabbit and Necturus gallbladders reveal different limitations to conductive HCO3– exit across the apical membrane. For example, rabbit gallbladder demonstrated a considerable rate (2.8 µeq·cm–2·h–1) of stilbene-sensitive HCO3– uptake across the basolateral membrane but no electrogenic HCO3– secretion. Corresponding microelectrode measurements showed a high resting membrane resistance ratio (Ra/Rb = 1.9) in rabbit gallbladder that did not respond to 8-bromoadenosine-cAMP as required for conductive HCO3– exit (41). Similar studies in Necturus gallbladder showed negligible electrogenic HCO3– secretion under basal conditions, with a slight increase to {approx}0.5 µeq·cm–2·h–1 after 2 µM forskolin. Unlike in rabbits, the rate-limiting step for HCO3– secretion across Necturus gallbladder was not Ga. Using impedance techniques, Kottra et al. (17) demonstrated that Ra fell from its basal value of 2,850 {Omega}·cm2 in Necturus gallbladder to 370 {Omega}·cm2 within 3 min of exposure to 2 µM forskolin and to 20 ± 5 {Omega}·cm2 within 30 min. Likewise, Rb fell from 1,185 to 512 {Omega}·cm2, reducing Ra/Rb from 2.5 ± 0.3 to 0.1 ± 0.1. Considering the magnitude of the forskolin-induced Isc across Necturus gallbladder, the large change in the membrane resistance ratio indicated that driving forces limited the rate of transepithelial HCO3– secretion. In contrast to both rabbits and Necturus, gallbladders demonstrated significantly lower Ra ({cong}200 {Omega}·cm2) under resting conditions but similar Rb ({cong}800 {Omega}·cm2). In response to cAMP, Ra fell precipitously to the remarkable value of 13 ± 2 {Omega}·cm2 within 4 ± 1 min of forskolin addition, whereas the membrane resistance ratio fell from 0.19 ± 0.04 to 0.13 ± 0.03 (P < 0.04) despite a simultaneous fall in Rb. These remarkable conductance estimates are commensurate with the degree of apical CFTR expression observed by immunoprecipitation and immunofluorescence microscopy.

Although high Ga and redundant mechanisms for basolateral HCO3– uptake set the stage for HCO3– movement, the eventual rate of transepithelial secretion is determined by driving forces across the individual membranes. In response to forskolin, Ga exceeded 75 mS/cm2, suggesting that a driving force of {approx}3.6 mV could generate the observed anion current of 269 µA/cm2. Large changes in Rb after exposure to indomethacin and forskolin indicated that a cAMP-dependent ionic conductance regulated the potential of the basolateral membrane and the resulting driving force for anion secretion across the apical membrane. Inhibitory effects of serosal Ba2+ and 293B as well as the insensitivity to charybdotoxin and apamin demonstrated a basolateral K+ conductance consistent with a cAMP-dependent K+ channel like KvLQT1 rather than a Ca2+-sensitive K+ channel or SK type channel (28). Elimination of the residual Isc by serosal ouabain excluded systematic errors in the measurement of Isc in this leaky preparation and confirmed the dependence of anion transport on basolateral Na+-K+-ATPase activity.

We based our conclusions about the individual membrane resistances on the assumption that forskolin and indomethacin did not appreciably alter the Rp during impedance experiments. Neither forskolin nor indomethacin altered the passive fluxes of nontransported substances, and the plots of Gt vs. Isc remained linear across the range of Isc with a high correlation coefficient. These results supported the assumption that Rp remained constant over the range of observed Isc. Furthermore, recalculating impedance estimates of Ra and Rb over a range of potential Rp values did not alter our conclusion that the individual membrane conductances in prairie dog gallbladder were significantly lower than in rabbits and Necturus. Our assumption of constant Rp is strengthened by data obtained in Necturus gallbladder showing that cAMP had no effect on junctional resistance but gradually increased the resistance of the lateral intercellular spaces (18). The impact of including this resistance as a circuit element in a more complex, distributed equivalent electrical circuit of the prairie dog gallbladder has already been evaluated in Necturus. Although the distributed model reduced the magnitude of the deviation between the observed and fitted impedance functions compared with the classic equivalent electrical circuit used in our experiments, prior authors (16) observed no significant effect of the more complex circuit model on the magnitude of the calculated parameters like Ra and Rb derived from curve fitting.

A further test of the assumptions underlying our impedance analysis in this leaky preparation was the use of indomethacin to check for dielectric dispersions over the wide range of frequencies required during these experiments. Similar to the use of amiloride in frog skin, indomethacin blocked Ga and Gb in prairie dog gallbladder, raising the values of Ra and Rb relative to Rp (1, 36). Under conditions of Ra/Rb >> Rp caused by indomethacin, we observed no significant change in the Cole-Cole power law factor {gamma} to indicate that dielectric dispersions affected impedance measurements, lending further credibility to our fitted estimates of Ra and Rb. These data confirmed that impedance analysis could be successfully applied to prairie dog gallbladders despite their low Rp, confirming prior reports in Necturus gallbladder (15, 47) and human colon (48).

Significance of apical CFTR and basolateral pNBC1 expression. The combination of significant Ga and redundant mechanisms for basolateral HCO3– uptake constitute a coordinated mechanism for electrogenic HCO3– secretion into the lumen of the prairie dog gallbladder under resting conditions that is directed opposite to the absorptive flux of NaCl. The bumetanide sensitivity of Isc suggested that JnetCl was the sum of oppositely directed Cl– movements involving Cl– secretion mediated by basolateral Na-K-2Cl cotransport at 1.0 ± 0.2 µeq·cm–2·h–1 offset by electroneutral Cl– absorption at 5.3 µeq·cm–2·h–1. In the absence of amiloride-sensitive current or apical resistance, equal rates of net Na+ and Cl– absorption suggested that electroneutral parallel ion exchange or Na+-Cl– cotransport mediated NaCl absorption as described by Reuss (43) and Frizzell et al. (11) in Necturus and rabbit gallbladders.

On the basis of the accumulated evidence, we propose the following model (Fig. 6) of the prairie dog gallbladder: CFTR mediates Ga under resting conditions and is stimulated by cAMP and endogenous prostaglandins. The negligible value of bumetanide-sensitive Cl– secretion suggests that high resting Ga depolarizes the apical membrane potential and reduces it to a value nearly equal to the electrochemical equilibrium potential for Cl–, thereby limiting the driving force for Cl– movement. The majority of Isc therefore represents electrogenic HCO3– current maintained by three sources of HCO3– uptake: basolateral pNBC1, basolateral Cl–/HCO3– exchange, and metabolic production from CO2 via carbonic anhydrase. The driving force for HCO3– secretion is provided by the activation of cAMP-dependent basolateral K+ channels like KvLQT1 and is maintained by the ouabain-sensitive basolateral Na+-K+-ATPase. In prairie dog gallbladder, electrogenic HCO3– secretion thereby exists in parallel with electroneutral NaCl absorption and is regulated by the cAMP-dependent conductances of the apical and basolateral membranes and the uptake of HCO3– by multiple basolateral mechanisms.


Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 6. Proposed model of electrogenic anion secretion by prairie dog gallbladder. Under basal conditions, similar net fluxes of Na+ and Cl– indicate electroneutral NaCl absorption. Isc is amiloride-insensitive and corresponds primarily to net residual ion flux, with a small contribution from bumetanide-sensitive Cl– secretion. The apical anion conductance is permeable to both Cl– and HCO3– and is cAMP-dependent, consistent with CFTR expression as confirmed by confocal microscopy and immunoprecipitation. HCO3– crosses the basolateral membrane via Na+-HCO3– cotransport (pNBC1) as well as Cl–/HCO3– exchange and is generated from exogenous CO2 by acetazolamide-sensitive carbonic anhydrase. Cl– uptake at the basolateral membrane is mediated by Na-K-2Cl cotransport. In response to forskolin, Ra falls to negligible values and Isc approaches 10 µeq·cm–2·h–1 due to HCO3– and Cl– secretion. Trans-6-cyano-4-(N-ethylsulfonyl-N-methylamino)-3-hydroxy-2,2-dimethyl-chroman (293B)-sensitive K+ channels (KvLQT1) hyperpolarize the basolateral membrane and provide the driving force for anion secretion.

 
The physiological relevance of electrogenic bicarbonate secretion in the prairie dog model has not been addressed in this study but may flush cholesterol and mucus out of the gallbladder lumen between meals as observed in humans and primates. We further speculate that HCO3– secretion counteracts the fall in luminal pH resulting from the absorption of salt and water by the gallbladder epithelium and the resulting acidification of the lumen caused by rising bile acid concentration ( 42, 50 ). Defective HCO3– secretion may permit the juxtaepithelial mucus layer to undergo a pH-dependent phase shift to the gel state that has been associated with the nucleation of cholesterol crystals from saturated bile (53). Indirect evidence for this hypothesis is the observation that gallstone risk is increased 10-fold among patients with cystic fibrosis (24) and reported links between CFTR and mucus hypersecretion during gallstone formation (51). Future studies in cholesterol-fed prairie dogs will determine the role of altered gallbladder HCO3– secretion during the formation of gallstones.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants K08-DK-60485 (to A. J. Moser), 1-P30-DK-72506 (to R. A. Frizzell), 5-R01-DK-68196 (to R. A. Frizzell), and 7-RO1-DK-58782 (to R. J. Bridges) and by the Competitive Medical Research Foundation of the University of Pittsburgh (to A. J. Moser).


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the excellent technical assistance of Sean Alper for immunofluorescence microscopy, Matt Green for assistance with impedance measurements, and Jessica Robuck for assistance with the flux measurements. We are also indebted to Drs. Mike Butterworth and Dan Devor for helpful discussions regarding this manuscript.


   FOOTNOTES
 

Address for reprint requests and other correspondence: A. J. Moser, Univ. of Pittsburgh School of Medicine, 497 Scaife Hall; 3550 Terrace St.; Pittsburgh, PA 15261 (e-mail: moseraj{at}upmc.edu)

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
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Awayda MS, Van Driessche W, Helman SI. Frequency-dependent capacitance of the apical membrane of frog skin: dielectric relaxation processes. Biophys J 76: 219–232, 1999.[Web of Science][Medline]
  2. Bradbury NA, Clark JA, Watkins SC, Widnell CC, Smith HS 4th, Bridges RJ. Characterization of the internalization pathways for the cystic fibrosis transmembrane conductance regulator. Am J Physiol Lung Cell Mol Physiol 276: L659–L668, 1999.[Abstract/Free Full Text]
  3. Bridges RJ, Rummel W, Simon B. Forskolin induced chloride secretion across the isolated mucosa of rat colon descendens. Naunyn Schmiedebergs Arch Pharmacol 323: 355–360, 1983.[CrossRef][Web of Science][Medline]
  4. Conter RL, Roslyn JJ, Porter-Fink V, DenBesten L. Gallbladder absorption increases during early cholesterol gallstone formation. Am J Surg 151: 184–191, 1986.[CrossRef][Web of Science][Medline]
  5. Copello J, Heming TA, Segal Y, Reuss L. cAMP-activated apical membrane chloride channels in Necturus gallbladder epithelium. Conductance, selectivity, and block. J Gen Physiol 102: 177–199, 1993.[Abstract/Free Full Text]
  6. Cremaschi D, Vallin P, Porta C. Hydrochlorothiazide action on the apical Cl–, Ca2+ and K+ conductances in rabbit gallbladder epithelium. Presence of an apamin-sensitive, Ca2+-activated K+ conductance. J Membr Biol 147: 159–171, 1995.[Web of Science][Medline]
  7. DenBesten L, Safaie-Shirazi S, Connor WE, Bell S. Early changes in bile composition and gallstone formation induced by a high cholesterol diet in prairie dogs. Gastroenterology 66: 1036–1045, 1974.[Web of Science][Medline]
  8. Devor DC, Singh AK, Lambert LC, DeLuca A, Frizzell RA, Bridges RJ. Bicarbonate and chloride secretion in Calu-3 human airway epithelial cells. J Gen Physiol 113: 743–760, 1999.[Abstract/Free Full Text]
  9. Diamond JM. The reasborptive function of the gall-bladder. J Physiol 161: 442–473, 1962.[Free Full Text]
  10. Dray-Charier N, Paul A, Veissiere D, Mergey M, Scoazec JY, Capeau J, Brahimi-Horn C, Housset C. Expression of cystic fibrosis transmembrane conductance regulator in human gallbladder epithelial cells. Lab Invest 73: 828–836, 1995.[Web of Science]
  11. Frizzell RA, Dugas MC, Schultz SG. Sodium chloride transport by rabbit gallbladder. Direct evidence for a coupled NaCl influx process. J Gen Physiol 65: 769–795, 1975.[Abstract/Free Full Text]
  12. Gelarden RT, Rose RC. Electrical properties and diffusion potentials in the gallbladder of man, monkey, dog, goose and rabbit. J Membr Biol 19: 37–54, 1974.[CrossRef][Web of Science][Medline]
  13. Gregory RJ, Cheng SH, Rich DP, Marshall J, Paul S, Hehir K, Ostedgaard L, Klinger KW, Welsh MJ, Smith AE. Expression and characterization of the cystic fibrosis transmembrane conductance regulator. Nature 347: 382–386, 1990.[CrossRef][Medline]
  14. Igimi H, Yamamoto F, Lee SP. Gallbladder mucosal function: studies in absorption and secretion in humans and in dog gallbladder epithelium. Am J Physiol Gastrointest Liver Physiol 263: G69–G74, 1992.[Abstract/Free Full Text]
  15. Kottra G, Fromter E. Rapid determination of intraepithelial resistance barriers by alternating current spectroscopy. I. Experimental procedures. Pflügers Arch 402: 409–420, 1984.[CrossRef][Web of Science][Medline]
  16. Kottra G, Fromter E. Rapid determination of intraepithelial resistance barriers by alternating current spectroscopy. II. Test of model circuits and quantification of results. Pflügers Arch 402: 421–432, 1984.[CrossRef][Web of Science][Medline]
  17. Kottra G, Fromter E. Tight-junction tightness of Necturus gall bladder epithelium is not regulated by cAMP or intracellular Ca2+. II. Impedance measurements. Pflügers Arch 425: 535–545, 1993.[CrossRef][Web of Science][Medline]
  18. Kottra G, Haase W, Fromter E. Tight-junction tightness of Necturus gall bladder epithelium is not regulated by cAMP or intracellular Ca2+. I. Microscopic and general electrophysiological observations. Pflügers Arch 425: 528–534, 1993.[CrossRef][Web of Science][Medline]
  19. Kozak LJ, Owings MF, Hall MJ. National Hospital Discharge Survey: 2002 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13: 1–199, 2005.
  20. Kreindler JL, Jackson AD, Kemp PA, Bridges RJ, Danahay H. Inhibition of chloride secretion in human bronchial epithelial cells by cigarette smoke extract. Am J Physiol Lung Cell Mol Physiol 288: L894–L902, 2005.[Abstract/Free Full Text]
  21. Kreindler JL, Peters KW, Frizzell RA, Bridges RJ. Identification and membrane localization of electrogenic sodium bicarbonate cotransporters in Calu-3 cells. Biochim Biophys Acta 1762: 704–710, 2006.[Medline]
  22. Kuver R, Klinkspoor JH, Osborne WR, Lee SP. Mucous granule exocytosis and CFTR expression in gallbladder epithelium. Glycobiology 10: 149–157, 2000.[Abstract/Free Full Text]
  23. Lewis SA, Clausen C, Wills NK. Impedance analysis of epithelia. In: Epithelial Transport: A Guide to Methods and Experimental Analysis, edited by Wills NK, Reuss L, Lewis SA. London: Chapman & Hall, 1996, p. 118–145.
  24. L'Heureux PR, Isenberg JN, Sharp HL, Warwick WJ. Gallbladder disease in cystic fibrosis. AJR Am J Roentgenol 128: 953–956, 1977.[Abstract]
  25. Lillemoe KD, Magnuson TH, High RC, Peoples GE, Pitt HA. Caffeine prevents cholesterol gallstone formation. Surgery 106: 400–407, 1989.[Web of Science][Medline]
  26. Loewen M, Bekar L, Gabriel S, Walz W, Forsyth G. pCLCA1 becomes a cAMP-dependent chloride conductance mediator in Caco-2 cells. Biochem Biophys Res Commun 298: 531–536, 2002.[CrossRef][Web of Science][Medline]
  27. Lohrmann E, Burhoff I, Nitschke RB, Lang HJ, Mania D, Englert HC, Hropot M, Warth R, Rohm W, Bleich M, Greger R. A new class of inhibitors of cAMP-mediated Cl– secretion in rabbit colon, acting by the reduction of cAMP-activated K+ conductance. Pflügers Arch 429: 517–530, 1995.[CrossRef][Web of Science][Medline]
  28. Loussouarn G, Charpentier F, Mohammad-Panah R, Kunzelmann K, Baro I, Escande D. KvLQT1 potassium channel but not IsK is the molecular target for trans-6-cyano-4-(N-ethylsulfonyl-N-methylamino)-3-hydroxy-2,2-dimethyl-chromane. Mol Pharmacol 52: 1131–1136, 1997.[Abstract/Free Full Text]
  29. Margineanu DG, Van Driessche W. Effects of millimolar concentrations of glutaraldehyde on the electrical properties of frog skin. J Physiol 427: 567–581, 1990.[Abstract/Free Full Text]
  30. Marino CR, Jeanes V, Boron WF, Schmitt BM. Expression and distribution of the Na+-HCO3– cotransporter in human pancreas. Am J Physiol Gastrointest Liver Physiol 277: G487–G494, 1999.[Abstract/Free Full Text]
  31. Martin LC, Hickman ME, Curtis CM, MacVinish LJ, Cuthbert AW. Electrogenic bicarbonate secretion in mouse gallbladder. Am J Physiol Gastrointest Liver Physiol 274: G1045–G1052, 1998.[Abstract/Free Full Text]
  32. Moreno JH, Diamond JM. Cation permeation mechanisms and cation selectivity in "tight junctions" of gallbladder epithelium. Membranes 3: 383–497, 1975.[Medline]
  33. Moser AJ, Abedin MZ, Morgenstern KE, Abedin ZR, Roslyn JJ. Endogenous prostaglandins modulate chloride secretion by prairie dog gallbladder. J Lab Clin Med 135: 82–88, 2000.[CrossRef][Web of Science][Medline]
  34. Moser AJ, Graf K, Abedin ZR, Morgenstern KE, Smith PR, Roslyn JJ, Albedin MZ. Anion exchanger AE-2 mediates chloride transport by gallbladder epithelium. Ann Mtg Am Gastrointest Assoc 99th New Orleans, LA, 1998.
  35. Nilsson B, Valantinas J, Hedin L, Friman S, Svanvik J. Acetazolamide inhibits stimulated feline liver and gallbladder bicarbonate secretion. Acta Physiol Scand 174: 117–123, 2002.[CrossRef][Web of Science][Medline]
  36. O'Leary DP, LaMorte WW, Scott TE, Booker ML, Stevenson J. Inhibition of prostaglandin synthesis fails to prevent gallbladder mucin hypersecretion in the cholesterol-fed prairie dog. Gastroenterology 101: 812–820, 1991.[Web of Science][Medline]
  37. Paunescu TG, Helman SI. cAmp activation of apical membrane Cl– channels: theoretical considerations for impedance analysis. Biophys J 81: 838–851, 2001.[Web of Science][Medline]
  38. Paunescu TG, Helman SI. PGE2 activation of apical membrane Cl– channels in A6 epithelia: impedance analysis. Biophys J 81: 852–866, 2001.[Web of Science][Medline]
  39. Peters RH, French PJ, van Doorninck JH, Lamblin G, Ratcliff R, Evans MJ, Colledge WH, Bijman J, Scholte BJ. CFTR expression and mucin secretion in cultured mouse gallbladder epithelial cells. Am J Physiol Gastrointest Liver Physiol 271: G1074–G1083, 1996.[Abstract/Free Full Text]
  40. Petersen KU, Goergen R, Hofken F, Macherey HJ, Sprakties G. Electrogenic bicarbonate secretion in gallbladder: induction by barium via neuronal, possibly VIP-ergic pathways. Naunyn Schmiedebergs Arch Pharmacol 348: 526–535, 1993.[Web of Science][Medline]
  41. Petersen KU, Reuss L. Cyclic AMP-induced chloride permeability in the apical membrane of Necturus gallbladder epithelium. J Gen Physiol 81: 705–729, 1983.[Abstract/Free Full Text]
  42. Petersen KU, Wehner F, Winterhager JM. Transcellular bicarbonate transport in rabbit gallbladder epithelium: mechanisms and effects of cyclic AMP. Pflügers Arch 416: 312–321, 1990.[CrossRef][Web of Science][Medline]
  43. Plevris JN, Bouchier IA. Defective acid base regulation by the gall bladder epithelium and its significance for gall stone formation. Gut 37: 127–131, 1995.[Free Full Text]
  44. Reuss L. Independence of apical membrane Na+ and Cl– entry in Necturus gallbladder epithelium. J Gen Physiol 84: 423–445, 1984.[Abstract/Free Full Text]
  45. Reuss L, Finn AL. Electrical properties of the cellular transepithelial pathway in Necturus gallbladder. I. Circuit analysis and steady-state effects of mucosal solution ionic substitutions. J Membr Biol 25: 115–139, 1975.[CrossRef][Web of Science][Medline]
  46. Rose RC, Gelarden RT, Nahrwold DL. Electrical properties of isolated human gallbladder. Am J Physiol 224: 1320–1326, 1973.[Free Full Text]
  47. Salhany J. Mechanism of competition between chloride and stilbenedisulfonates for binding to human erythrocyte band 3 (AE1). Biochem Cell Biol 76: 715–722, 1998.[CrossRef][Web of Science][Medline]
  48. Schifferdecker E, Fromter E. The AC impedance of Necturus gallbladder epithelium. Pflügers Arch 377: 125–133, 1978.[CrossRef][Web of Science][Medline]
  49. Schmitz H, Barmeyer C, Fromm M, Runkel N, Foss HD, Bentzel CJ, Riecken EO, Schulzke JD. Altered tight junction structure contributes to the impaired epithelial barrier function in ulcerative colitis. Gastroenterology 116: 301–309, 1999.[CrossRef][Web of Science][Medline]
  50. Schultz SG, Zalusky R, Gass AE Jr. Ion transport in isolated rabbit ileum. 3. Chloride fluxes. J Gen Physiol 48: 375–378, 1964.[Abstract/Free Full Text]
  51. Shiffman ML, Sugerman HJ, Moore EW. Human gallbladder mucosal function. Effect of concentration and acidification of bile on cholesterol and calcium solubility. Gastroenterology 99: 1452–1459, 1990.[Web of Science][Medline]
  52. Shoda J, Kano M, Asano T, Irimura T, Ueda T, Iwasaki R, Furukawa M, Kamiya J, Nimura Y, Todoroki T, Matsuzaki Y, Tanaka N. Secretory low-molecular-weight phospholipases A2 and their specific receptor in bile ducts of patients with intrahepatic calculi: factors of chronic proliferative cholangitis. Hepatology 29: 1026–1036, 1999.[CrossRef][Web of Science]
  53. Singh AK, Singh S, Devor DC, Frizzell RA, van Driessche W, Bridges RJ. Transepithelial impedance analysis of chloride secretion. Methods Mol Med 70: 129–142, 2002.[Medline]
  54. Smith BF. Gallbladder mucin as a pronucleating agent for cholesterol monohydrate crystals in bile. Hepatology 12: 183S–188S, 1990.
  55. Sterling D, Reithmeier RA, Casey JR. A transport metabolon. Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J Biol Chem 276: 47886–47894, 2001.[Abstract/Free Full Text]
  56. Stewart CP, Winterhager JM, Heintze K, Petersen KU. Electrogenic bicarbonate secretion by guinea pig gallbladder epithelium: apical membrane exit. Am J Physiol Cell Physiol 256: C736–C749, 1989.[Abstract/Free Full Text]
  57. Strichartz SD, Abedin MZ, Abdou MS, Roslyn JJ. The effects of amiloride on biliary calcium and cholesterol gallstone formation. Ann Surg 209: 152–156, 1989.[Web of Science][Medline]
  58. Stuart-Tilley A, Sardet C, Pouyssegur J, Schwartz MA, Brown D, Alper SL. Immunolocalization of anion exchanger AE2 and cation exchanger NHE-1 in distinct adjacent cells of gastric mucosa. Am J Physiol Cell Physiol 266: C559–C568, 1994.[Abstract/Free Full Text]
  59. Svanvik J, Allen B, Pellegrini C, Bernhoft R, Way L. Variations in concentrating function of the gallbladder in the conscious monkey. Gastroenterology 86: 919–925, 1984.[Web of Science][Medline]
  60. Webster P, Vanacore L, Nairn AC, Marino CR. Subcellular localization of CFTR to endosomes in a ductal epithelium. Am J Physiol Cell Physiol 267: C340–C348, 1994.[Abstract/Free Full Text]
  61. Wills NK, Clausen C. Transport-dependent alterations of membrane properties of mammalian colon measured using impedance analysis. J Membr Biol 95: 21–35, 1987.[CrossRef][Web of Science][Medline]
  62. Wills NK, Lewis SA, Eaton DC. Active and passive properties of rabbit descending colon: a microelectrode and nystatin study. J Membr Biol 45: 81–108, 1979.[CrossRef][Web of Science][Medline]
  63. Winterhager JM, Stewart CP, Heintze K, Petersen KU. Electroneutral secretion of bicarbonate by guinea pig gallbladder epithelium. Am J Physiol Cell Physiol 250: C617–C628, 1986.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/6/G1683    most recent
00268.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moser, A. J.
Right arrow Articles by Bridges, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Moser, A. J.
Right arrow Articles by Bridges, R. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.