Experience in Southeast Asia with cases of secretory

Transcription

1 GASTROENTEROLOGY 1999;116: Characterization of the Inhibitory Effect of Boiled Rice on Intestinal Chloride Secretion in Guinea Pig Crypt Cells CERI J. MATHEWS,* R. JOHN MACLEOD,, SHU XIAN ZHENG,* JOHN W. HANRAHAN,* HUGH P. J. BENNETT, and J. RICHARD HAMILTON, Departments of *Physiology, Medicine, and Pediatrics, McGill University, Montreal; and McGill University Montreal Children s Hospital Research Institute, Montreal, Quebec, Canada Background & Aims: When rice is incorporated into oral rehydration therapy for patients with secretory diarrhea, clinical outcomes improve. We have shown that a factor purified from boiled rice (RF) blocks the secretory response of intestinal crypt cells to adenosine 3,5 -cyclic monophosphate (camp). Now we report that the cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel is the cellular target for this rice inhibitor. Methods: We used RF, the same previously described extract prepared from boiled rice, to assess chloride channel activation in vitro, measuring (1) cell volume regulation of guinea pig intestinal crypt epithelial cell suspensions using standard Coulter counter technology, (2) transepithelial chloride current in monolayers of T 84 cells mounted in Ussing chambers, and (3) whole-cell and single-channel currents using the patch-clamp technique in cells transfected to express CFTR. Results: RF inhibited activation by camp of CFTR chloride channels in all experimental preparations; RF did not block volume-stimulated Cl secretion, suggesting that its effect might be specific for CFTR chloride channels. RF inhibited transepithelial camp-stimulated Cl current in T 84 cells and inhibited forskolin (i.e., camp)-induced current in cells transfected with CFTR. Excised patch and single-channel patch-clamp recordings supported the view that the response was a direct effect on CFTR rather than on camp signal transduction. Conclusions: RF exerts a specific inhibitory effect on CFTR chloride channels, blocking activation from the luminal surface of the cell and reversing established activation. Many major diarrheal states are based on camp-induced CFTR activation, leading to excessive gut secretion; our findings could have clinical relevance. Experience in Southeast Asia with cases of secretory diarrhea suggests that when rice is incorporated into oral rehydration solutions, clinical outcomes improve. 1 These observations from the field led us to in vitro studies in which intestinal crypt epithelial cells in suspension, when exposed to a factor extracted from boiled rice, failed to respond normally to the potent secretogogue adenosine 3,5 -cyclic monophosphate (camp). 2 This preliminary finding suggested that in addition to enhancing nutrientlinked absorption, rice might be capable of a direct antisecretory effect on the small intestine. In the present studies, we used a range of cellular and molecular techniques to characterize the inhibitory impact on intestinal secretion of a specific low-molecular-weight constituent extracted from boiled rice. To stimulate a secretory response in the intestinal epithelium, we continued to use in vitro models of ion secretion stimulated by camp, an important mediator of many human diarrheal states, including that occurring in cholera. Materials and Methods Preparation of Rice Factor This procedure was based on a previously published method. 2 The supernatant, obtained after boiling 275 g of unprocessed rice in 6Lofwater for 10 minutes, was first lyophilized; this material was then extracted in 1Lof50% (vol/vol) MeOH and 0.5% (vol/vol) trifluoroacetic acid, homogenized, and centrifuged for 15 minutes at 3000g. The supernatant was dried in a Speed-Vac, and the dried material was subjected to gel-filtration high-performance liquid chromatography. UV absorbance of the column eluates was monitored at 280 nm, and 1-minute fractions were collected. Fractions (corresponding to molecular weights of 1.5 kilodaltons), which contained most of the biological activity of interest, 2 were processed further, dried, and subjected to ion-exchange chromatography. The dried material was dissolved in 50 mmol/l Tris, ph 7.5, the ph was adjusted to 7.0 with 0.1N NaOH, and then they were loaded onto a QMA anion-exchange column (Pharmacia, Mississauga, Ontario, Canada). The column was eluted over 40 minutes with a gradient of 0 1 mol/l NaCl in 50 mmol/l Tris (ph 7.5), and 1-minute fractions were collected. Fractions were Abbreviations used in this paper: CFTR, cystic fibrosis transmembrane conductance regulator; Isc, short-circuit current; MgATP, magnesium adenosine triphosphate; PKA, protein kinase A; RF, rice factor; RVD, regulatory volume decrease; TES, N-tris(hydroxymethyl)- methyl-2-aminoethane sulfonic acid by the American Gastroenterological Association /99/$10.00

2 June 1999 ANTISECRETORY IMPACT OF RICE ON THE GUT 1343 pooled and subjected to reverse-phase high-performance liquid chromatography; these pooled fractions were acidified and loaded onto a Vydac C4 column (Scientific Product and Equipment Ltd., Concord, Ontario, Canada). The column was eluted over 1 hour with a gradient of 0% 48% (vol/vol) acetonitrile containing 0.1% (vol/vol) trifluoroacetic acid throughout. Column eluates were monitored for UV absorbance at 210 and 280 nm, and 1-minute fractions were collected. Fractions 19 42, which contained more than 90% of bioactivity in the crypt cell assay, 2 were pooled and dried. This highly purified rice factor (RF) was prepared in batches that were pooled as a single source for all the current experiments. In 6 of 6 of these batches, assays for biological activity using inhibition of camp-induced crypt cell volume regulation 2 found consistent 93% 95% inhibitory activity. The precise molecular nature of RF has not yet been determined; more than one component could be involved in the biological responses observed. The extract is hydrophobic because it is retained on octadecylsilyl-silica cartridges, it has a net negative charge because it is retained on the anion exchanger, and it has a molecular weight of less than 1.5 kilodaltons. Because of its small size and lack of amino sugars, it is neither a peptide nor a glycoprotein. 2 Inhibitory activity of the rice fraction was stable for several months when lyophilized and for at least 2 weeks at 4 C when in solution. Cell Volume Experiments We compared the impact of RF on regulatory volume decrease (RVD) in freshly isolated jejunal crypt epithelial cells exposed either to hypotonic dilution or to camp. In the intact small intestine, the role, if any, of villus cells in secretion is not clear, but crypt cells are known to respond to mediators of ion secretion such as camp. 3 We used established techniques to isolate fresh guinea pig crypt cells in suspension, to monitor their viability and purity, and to measure their volume electronically under various conditons. 4,5 Crypt cell volume was measured sequentially in response to (1) 0.7 isotonic dilution in an Na -free medium containing the cation ionophore gramicidin (0.2 µmol/l) and (2) in Na -containing isotonic medium with 8-Br-cAMP (0.5 mmol/l). The response to added RF (10 µg/ml) was assessed in these models of cell volume regulation. Short-Circuit Measurement of Chloride Secretion by Intestinal Cells Short-circuit current (Isc) provides a measure of net chloride movement during camp stimulation of T 84 monolayers. 5 Experiments were carried out in temperaturecontrolled (37 1 C) Ussing chambers modified to accept Millicell inserts (Millipore Canada, Toronto, Ontario, Canada). 6 Agar bridges (3 mol/l KCl), used to pass Isc and measure transepithelial potential, were connected through Ag/AgCl electrodes to conventional voltage clamps interfaced with a computer. We monitored resistance by applying voltage pulses (10 mv, 2-second duration) every 150 seconds. Correction was made for saline resistance. Digitized currents were saved to disk for subsequent analyses. To measure chloride secretion, monolayers were bathed on both sides with 115 mmol/l NaCl, 2.4 mmol/l K 2 HPO 4, 0.4 mmol/l KH 2 PO 4, 1.2 mmol/l MgCl 2, 1.2 mmol/l CaCl 2, 25 mmol/l NaHCO 3, and 10 mmol/l glucose and equilibrated with 95% O 2 /5% CO 2. A stimulant cocktail was added to increase the concentration of camp at the apical side of monolayers after basal Isc had stabilized. This stimulant mixture contained 50 mmol/l dibutyryl c-amp, 1 mmol/l forskolin, and 1 mmol/l 3-isobutyl-1-methylxanthine and was diluted 1:100 upon addition. To test RF in this system, it was dissolved in Ussing chamber buffer (114 µg/ml) and added to the apical side of cell monolayers after aspiration of the apical compartment. Control monolayers received similar treatment with fresh buffer added after aspiration of the apical chamber. No change in monolayer resistance was observed when solutions were changed. Patch-Clamp Electrophysiological Recordings From Cystic Fibrosis Transmembrane Conductance Regulator Expressing Cells Experiments were performed at room temperature (23 C) using patch-clamp electrophysiological techniques. 7 Whole-cell recordings. The standard solution in the recording chamber (volume, 150 µl) contained 150 mmol/l NaCl, 2 mmol/l MgCl 2, and 10 mmol/l N-tris(hydroxymethyl)methyl-2-aminoethane sulfonic acid (TES), ph This solution was supplemented by 50 mmol/l sucrose to prevent swelling-activated chloride currents during whole-cell recordings. The pipette (intracellular) solution used for recording whole-cell chloride currents contained 110 mmol/l N-methyl-D-glucamine-aspartate, 30 mmol/l N-methyl-Dglucamine-Cl, 1 mmol/l MgCl 2, 10 mmol/l TES, and 0.1 mmol/l 1,2-bis(o-aminophenoxy)ethane-N,N,N,N -tetraacetic acid tetrasodium salt (BAPTA; Biomol Research Laboratories Inc., Plymouth Meeting, PA), ph Magnesium adenosine triphosphate (MgATP, 1 mmol/l) was added to the pipette solution before it was used. Forskolin was added from stock solution in ME 2 SO (5 mmol/l). Data were low-pass filtered at 500 Hz using a three-pole Bessel filter and sampled at 1000 Hz. Voltage protocols and current acquisition were controlled by a personal computer running pclamp software (Version 6.0.3; Axon Instruments Inc., Foster City, CA). We held membrane potential (V m ) at 0 mv during experiments, and leak currents were not subtracted from traces. Single-channel and macropatch patch clamp recordings. Recordings were performed at room temperature using the excised inside-out configuration of the patch-clamp technique. 7 Pipette and bath solutions contained 150 mmol/l NaCl, 2 mmol/l MgCl 2, and 10 mmol/l TES, ph The catalytic subunit of protein kinase A (PKA), prepared by Dr. M. P. Walsh (University of Calgary, Alberta, Canada) 8 was added to patches at a final concentration of 180 nmol/l, and MgATP (1 mmol/l) was added from a 100 mmol/l stock in buffer. The probability of a channel being open (P o ) was calculated as the mean number of channels open during

3 1344 MATHEWS ET AL. GASTROENTEROLOGY Vol. 116, No. 6 segments of the record (NP o ) divided by the maximum number of channels seen in the patch (N). V m was held at 30 mv throughout, and outward flow of chloride was shown as an upward deflection in records. Data were low-pass filtered at 150 Hz using an eight-pole Bessel filter and acquired by personal computer at 500 Hz. Results Cell Volume Regulation Suspended in an Na -free hypotonic (0.7 isotonic), gramicidin-containing medium, crypt cells first swelled (relative volume, ; n 6); then over a 10-minute period they shrank back to their initial volume (relative volume, ; n 6; P 0.001). In a previous report, 8 we have characterized this complete RVD as caused by activation of a volumesensitive Cl channel. RF (10 µg/ml) added to the cells suspended in hypotonic medium failed to block this RVD response, which was identical whether RF was present in the medium or not. When the same Na -free 0.7 isotonic, gramicidin-containing medium was used with RF, the cells swelled initially (relative volume, 1.21 and 1.19) and shrank back to their initial volume (relative volume, 1.03 and 1.02) at 10 minutes in two separate experiments. When the same cells were centrifuged and resuspended in Na -containing isotonic medium without RF, 8-Br-cAMP (0.5 mmol/l) added to this suspension caused significant cell shrinkage compared with untreated controls (final relative volume, vs ; P 0.001; n 5). When added to the cells resuspended in isotonic medium, RF blocked 8-Br-cAMP induced cell shrinkage, as reported previously (final relative volume, 0.90 and 0.92). 2 These observations suggest a specific effect of RF on the cystic fibrosis transmembrane conductance regulator (CFTR) because this chloride channel, but not the volume-sensitive channel, is known to be responsive to camp. 9,10 Response of camp-stimulated Chloride Flux Across Monolayers of T84 Cells We tested RF on human intestinal cells (T 84 ) grown on permeable supports, measuring Isc as a reflection of net chloride current 7 during camp-induced secretion (Figure 1). Before stimulation, the basal Isc values of the T 84 monolayers were low, µa/cm 2 (n 8). After addition of the stimulant camp cocktail, mean Isc was µa/cm 2 (n 8), a level that was maintained in controls for minutes (mean decline, 5.1% 1.9%; n 4). Addition of RF to the apical surface of test monolayers caused a sharp decrease in Isc, whereas Isc in unexposed controls remained stable. Compared with control monolayers, those exposed to Figure 1. Effect of RF on camp-dependent Isc across monolayers of T 84 intestinal cells. (A) Typical control ( ) experiment. In response to camp, mixture Isc (measure of net chloride transport) increases to plateau at 20 minutes. (B) Arrow indicates RF (, inhibitor) addition to apical solution (final concentration, 114 µg/ml); Isc inhibited 37.6% from peak. inhibitor had significantly decreased Isc (23.8% 5.2%; P 0.02; n 4); thus RF placed at the apical surface of the cell inhibited basolateral-to-apical seretory chloride flux in an intestinal epithelial layer. Patch-Clamp Experiments Because camp-stimulated anion secretion in T 84 cells has been shown to be mediated by CFTR chloride channels, we studied the effects of RF on CFTR chloride channels in cells stably transfected with a plasmid directing the expression of wild-type CFTR. 9,13 The channel was stimulated via camp-dependent PKA-

4 June 1999 ANTISECRETORY IMPACT OF RICE ON THE GUT 1345 mediated phosphorylation in the presence of MgATP, and its activity was measured electrophysiologically using patch-clamp techniques. 7 We used BHK and CHO cells for these experiments; both are more suitable than T 84 cells for patch clamp studies because they readily form highly resistant seals with the glass pipette. Initially the effect of RF was tested on whole-cell chloride currents in CFTR-expressing BHK cells after forskolin stimulation (10 µmol/l). CFTR is expressed at very high levels in these cells, and it is the only channel stimulated significantly by PKA, making it possible to record whole-cell currents without contamination by other channels. 14 After RF addition (final concentration, 114 µg/ml), whole-cell chloride current, measured at V m 60 mv, was reduced by 71% 2% (n 2). In the representative experiment shown in Figure 2, current was reduced from 2.83 to 0.61 na. These findings support the hypothesis that the cellular target of the inhibitor RF is CFTR, 14 although an indirect effect on the camp signal transduction cascade could not be ruled out on the basis of these data. To determine whether the inhibitor acted directly or indirectly, we investigated its effect on inside-out membrane patches excised from CFTR-expressing cells. Macroscopic chloride current, stimulated in patches before addition of inhibitor by exposure to catalytic subunit of PKA and MgATP, was assumed to be almost exclusively mediated by CFTR channels. 14 Addition of inhibitor (60 µg/ml) resulted in an instantaneous reduction in current amplitude (Figure 3A). The decrease in current in 2 of 2 experiments, performed on patches excised from separate BHK cells and measured at V m 30 mv, was 60.4% 1.1% (n 2; PKA-activated current, pa; current remaining in the presence of inhibitor, pa). This finding supports an action of the inhibitor directly on the channel. Additional evidence for a direct action of RF comes from single-channel experiments on membrane patches (n 2) excised from CHO cells, which express a low level of CFTR, making them convenient for single-channel experiments. In a representative patch, CFTR channel activity, which was steady over a 6-minute control period, was almost abolished within seconds of addition of rice inhibitor (Figure 3B). The single-channel activity (measured as open probability; P o ) was reduced by 78% in the presence of inhibitor (to 0.05 from for the control recording). Current inactivation by the inhibitor in these excised patches was much faster and more complete than that observed after removal of PKA, 15 consistent with a direct action of RF on CFTR rather than a secondary effect via channel dephosphorylation after PKA inhibition. Discussion Our findings that a low-molecular-weight, hydrophobic, negatively charged component of boiled rice consistently inhibited camp-induced intestinal epithelial Cl secretion concur with and expand our earlier observations. 2 We have again used cell volume measurements to show that the RF that inhibited camp-induced cell shrinkage failed to inhibit shrinkage secondary to cell swelling caused by suspension of cells in a hyposmolar medium. Intestinal crypt epithelial cells possess a volumeactivated Cl channel, but to regulate volume in a hyposmolar medium they require a cation ionophore to compensate for their lack of a volume-activated K channel. 8 Because cell volume regulation remained intact in the presence of RF under these conditions, we conclude that the volume-sensitive Cl channel was not inhibited by RF. These observations, together with our earlier Figure 2. RF reduces wholecell chloride currents in cells stably transfected with plasmid, directing expression of wild-type CFTR chloride channels. (A) Whole-cell currents recorded from CFTR-expressing BHK cells (i) under control conditions, (ii) with stimulation by forskolin (10 µmol/l), and (iii) after addition of RF (final concentration, 114 µg/ml) plus forskolin (10 µmol/l). (B) Steady-state current-voltage relationships for whole-cell currents shown in A., Control;, stimulated;, inhibitor. Negative reversal potential after forskolin stimulation confirms chloride selectivity of whole-cell current.

5 1346 MATHEWS ET AL. GASTROENTEROLOGY Vol. 116, No. 6 Figure 3. Effect of RF on CFTR chloride channel activity. Patches excised from cells stably transfected to express high or low levels of CFTR protein (BHK and CHO cells, respectively). (A) (i) PKA (60 nm) added to inside-out patches excised from a BHK cell in the presence of MgATP (1 mmol/l) activates CFTR chloride channels; a macroscopic current developed over several minutes. RF (60 µg/ml; arrow) caused rapid, strong inhibition of this current. (ii) Expanded trace, showing openings of single CFTR channels shortly after PKA addition. Segment shown is indicated by a horizontal line in i.(b) Effect of RF on single-channel activity of CFTR chloride channels. (i) CFTR activity in an inside-out patch excised from a CHO cell. RF (60 µg/ml; arrow) caused immediate, marked inhibition of channel activity. Note the small ( 0.8 pa) decrease in baseline current after addition of RF. PKA (180 nmol/l) and MgATP (1 mmol/l) were present throughout. Two channels were observed in this patch, and the closed-channel current level is indicated by the thin line to the left of the trace. (ii) Bar graph illustrating 78% reduction in single-channel P o in the presence of RF (0.05; duration of record analyzed, 90 seconds) compared with control (0.226; duration of record analyzed, 130 seconds). published evidence that RF inhibited 8-Br-cAMP stimulated 36 Cl efflux from fresh crypt epithelial cells 2 led us to evaluate the CFTR chloride channel as a target for the rice inhibitor. By using monolayers of cultured T 84 human intestinal epithelial cells, we could assess transepithelial ion flux in a homogeneous population of polarized intestinal cells; these cultured cells are analogous to mammalian intestinal crypt cells in that they lack the features of differentiated villus cells. 16 In the presence of a potent camp-induced secretory drive, RF rapidly inhibited basolateral-to-apical flux of chloride. It was effective from the apical surface of the monolayer, and it reversed activated secretion. By applying patch-clamping techniques to transfected BHK cells or CHO cells expressing CFTR, we could evaluate CFTR in the absence of second messengers and other cytosolic variables. The observed capacity of RF to block activation of Cl transport in these patch-clamp experiments was consistent with data from our cell volume and Ussing chamber studies. Initially in whole cells, RF inhibited Cl current after the current had been stimulated by camp-dependent PKA-mediated phosphorylation, pointing either to a direct specific effect on CFTR or to inhibition of camp activation. The latter possibility was excluded by study of excised patches in the presence of a catalytic subunit of PKA and MgATP, in which instantaneous current inhibition could only be attributable to a direct effect. Furthermore, singlechannel activity was immediately and almost completely abolished by the rice extract. The effect of the rice extract on single-channel open probability and kinetics is compatible with the compound being a slow channel blocker. A slow blocker is one that has relatively high affinity for its receptor on the channel and therefore a relatively long residency time. 17 Fast blockers have low affinity and dissociate rapidly from the channel. They do not produce discrete blocked events, although single-channel current amplitudes appear reduced. Intermediate blockers have moderate affinity and cause obvious interuptions in the

6 June 1999 ANTISECRETORY IMPACT OF RICE ON THE GUT 1347 current. Several fast 18 and intermediate 19,20 blockers have been described previously, but a putative slow CFTR blocker has not been reported. We conclude that a low-molecular-weight substance extracted from boiled rice can inhibit CFTR chloride channels. The specificity of this ion-channel activity favors the view that we are dealing with a single pharmacological class of compound. Because many major diarrheal states are based on camp-induced gut secretion, 3 the issue of the clinical relevance of these cellular findings arises. We have shown a specific target for an inhibitor that is effective when placed at the apical surface of the intestinal cell. In addition to blocking the Cl efflux response to camp, RF can actually reverse established secretion. Our findings show that a component of a widely available food is consistently capable of altering a major secretory response in crypt cells that constitute the secretory compartment of the intestine. 3 These in vitro experiments used a highly concentrated rice extract; it remains to be determined whether RF has an effect on secretion in the intact intestine. Given the high prevalence of diarrheal illness in the world, and the vast experience with rice as an abundant safe food and a key component of oral rehydration solutions, further studies seem to be indicated. Such a research effort would be greatly expedited by a precise molecular definition of this rice-based factor. References 1. Gore SM, Fontaine O, Pierce NF. Impact of rice-based oral rehydration solution on stool output and duration of diarrhea: meta-analyses of 13 clinical trials. BMJ 1992; MacLeod RJ, Bennett HPJ, Hamilton JR. Inhibition of intestinal secretion by rice. Lancet 1995;346: Kaunitz JD, Barrett KE, McRoberts JA. Electrolyte secretion and absorption: small intestine and colon. In: Yamada T, ed. Textbook of gastroenterology. 2nd ed. Philadelphia: Lipincott, 1995: MacLeod RJ, Lembessis P, Hamilton JR. Isotonic volume reduction associated with cyclic- AMP stimulation of 36 Cl efflux from jejunal crypt epithelial cells. Am J Physiol 1994;267:G387 G Cartwright CA, McRoberts JA, Mandel KG, Dharmsathaporn K. Synergistic action of cyclic adenosine monophosphate and calcium-mediated chloride secretion in a colonic epithelial cell line. J Clin Invest 1985;76: Tabcharani JA, Harris RA, Boucher A, Eng JWL, Hanrahan JW. Basolateral K channel activated by carbachol in the epithelial cell line T84. J Membr Biol 1994;142: Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch 1981;391: MacLeod RJ, Lembessis P, Hamilton JR. Effect of osmotic swelling on K conductance in jejunal crypt epithelial cells. Am J Physiol 1992;262:G1021 G Tabcharani JA, Chang B, Riordan JR, Hanrahan JW. Phosphorylationregulated Cl channel in CHO cells stably expressing the cystic fibrosis gene. Nature 1991;352: Kubo M, Okada Y. Volume-regulatory Cl channel currents in cultured human epithelial cells. J Physiol 1992;456: Tabcharani JA, Low W, Elie D, Hanrahan JW. Low-conductance chloride channel activated by camp in the epithelial cell line T 84. FEBS Lett 1990;270: Anderson MP, Sheppard DN, Berger HA, Welsh MJ. Chloride channels in the apical membrane of normal and cystic fibrosis airway and intestinal epithelia. Am J Physiol 1992;263:L1 L Seibert FS, Tabcharani JA, Chang X-B, Dulhanty AM; Mathews C, Hanrahan JW, Riordan JR. camp-dependent protein kinase mediated phosphorylation of cystic fibrosis transmembrane conductance regulator residue ser-753 and its role in channel activation. J Biol Chem 1995;270: Linsdell P, Hanrahan JW. Disulphonic stilbene block of cystic fibrosis transmembrane conductance regulator Cl channels expressed in a mammalian cell line and its regulation by a critical pore residue. J Physiol (Lond) 1996;496: Anderson MP, Berger HA, Rich DP, Gregory RJ, Smith AE, Welsh MJ. Nucleoside triphosphates are required to open the CFTR chloride channel. Cell 1991;67: Madara JL, Stafford J, Dharmsathaphorn K, Carlson S. Structural analysis of a human intestinal epithelial cell line. Gastroenterology 1987;92: Hille B. Ionic channels of excitable membranes. 2nd ed. Sunderland, MA: Sinauer Associates, Linsdell P, Hanrahan JW. Flickery block of single CFTR chloride channels by intracellular anions and osmolytes. Am J Physiol 1996; 271:C628 C 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 1993;102: Shultz BD, DeRoos ADG, Venglarik CJ, Singh AK, Frizzell RA, Bridges RJ. Glibenclamide blockade of CFTR chloride channels. Am J Physiol 1996;271:L192 L200. Received September 11, Accepted February 23, Address requests for reprints to: Richard Hamilton, M.D., Department of Pediatrics, Montreal Children s Hospital, 2300 Tupper Street, Montreal, Quebec H3H 1P3, Canada. Fax: (514) Supported by grants from the Medical Research Council of Canada to the three laboratories participating in these studies (to J.W.H., H.P.J.B., and J.R.H.). The initial stimulus for these studies arose from discussions with colleagues at the International Centre for Diarrhoeal Disease Research, Bangladesh. The authors thank Susan James and Peter Lembessis for expert assistance in the conduct of these experiments and Carolyn Mandel for help in preparing the manuscript.