Chapter 8 Chloride Channels in Epithelial Cells

The Role of Ion Channels in Cell Proliferation

Significant progress has been made in identifying physiologically important growth fac tors, receptors, and signal transduction pathways involved in the control of normal and malignant cell proliferation. More recent studies suggest that changes in the proliferative state of a cell are coupled to specific changes in membrane properties, which suggests that there are links between ion channel activity and cell proliferation in both neuronal and non-neuronal cells. Changes in ion channel expression may be necessary to permit cell cycle progression and, ultimately, cell proliferation. The potential mechanisms in volved in the translation of ion channel activity into changes in gene expression are discussed. NEUROSCIENTIST 5:70-73, 1999

Placental chloride channels: a review

The human placental syncytiotrophoblast (hSTB) is a polarized epithelial structure, that forms the main barrier to materno-fetal exchange. The chloride (Cl(-)) channels in other epithelial tissues contribute to several functions, such as maintenance of the membrane potential, volume regulation, absorption and secretion. Additionally, the contributions of Cl(-) channels to these functions are demonstrated by certain diseases and knock-out animal models. There are multiple lines of evidence for the presence of Cl(-) channels in the hSTB, which could contribute to different placental functions. However, both the mechanism by which these channels are involved in the physiology of the placenta, and their molecular identities are still unclear. Furthermore, a correlation between altered Cl(-) channels functions and pathological pregnancies is beginning to emerge. This review summarizes recent developments on conductive placental chloride transport, and discusses its potential implications for placental physiology.

Secretory activation of basolateral membrane Cl- channels in guinea pig distal colonic crypts

Cell-attached recordings revealed Cl(-) channel activity in basolateral membrane of guinea pig distal colonic crypts isolated from basement membrane. Outwardly rectified currents ((gp)Cl(or)) were apparent with a single-channel conductance (gamma) of 29 pS at resting membrane electrical potential; another outward rectifier with gamma of 24 pS was also observed ( approximately 25% of (gp)Cl(or)). At a holding potential of -80 mV gamma was 18 pS for both (gp)Cl(or) currents, and at +80 mV gamma was 67 and 40 pS, respectively. Identity as Cl(-) channels was confirmed in excised patches by changing bath ion composition. From reversal potentials, relative permeability of K(+) over Cl(-) (P(K)/P(Cl)) was 0.07 +/- 0.03, with relative permeability of Na(+) over Cl(-) (P(Na)/P(Cl)) = 0.08 +/- 0.04. A second type of Cl(-) channel was seen with linear current-voltage (I-V) relations ((gp)Cl(L)), having subtypes with gamma of 21, 13, and 8 pS. Epinephrine or forskolin increased the number of open (gp)Cl(or) and (gp)Cl(L). Open probabilities (P(o)) of (gp)Cl(or), (gp)Cl(L21), and (gp)Cl(L13) were voltage dependent in cell-attached patches, higher at more positive potentials. Kinetics of (gp)Cl(or) were more rapid with epinephrine activation than with forskolin activation. Epinephrine increased P(o) at the resting membrane potential for (gp)Cl(L13). Secretagogue activation of these Cl(-) channels may contribute to stimulation of electrogenic K(+) secretion across colonic epithelium by increasing basolateral membrane Cl(-) conductance that permits Cl(-) exit after uptake via Na(+)-K(+)-2Cl(-) cotransport.

Conductive pathways for chloride and oxalate in rabbit ileal brush-border membrane vesicles

To evaluate the possibility that an apical membrane conductive pathway for oxalate is present in the rabbit distal ileum, we studied oxalate ([14C]oxalate) and chloride (36Cl) uptake into brush-border membrane vesicles enriched 15- to 18-fold in sucrase activity. Voltage-sensitive pathways for oxalate and chloride were identified by the stimulation of uptake provided by an inwardly directed potassium diffusion potential in the presence of valinomycin. Additionally, outwardly directed oxalate (or chloride) gradients stimulated [14C]oxalate (or 36Cl) uptake to a greater degree in the absence of valinomycin (when intracellular and extracellular potassium are equal) than in the presence of valinomycin. Voltage-dependent anion uptake was poorly saturable: apparent affinity constants were 141 +/- 17 and 126 +/- 8 mM for chloride and oxalate, respectively. Activation energies for the voltage-dependent uptake processes were low: 4.7 and 6.3 kcal/mol for chloride and oxalate, respectively. Sensitivity profiles of voltage-dependent chloride and oxalate uptake to anion transport inhibitors were similar. We conclude that an anion conductance is present in the apical membranes of ileal enterocytes and that this conductance is a candidate pathway for oxalate efflux from the enterocyte during transepithelial oxalate secretion.

Modulation of chloride secretion across the pigmented rabbit conjunctiva

The purpose of the present study was to investigate whether active Cl- secretion in the pigmented rabbit conjunctiva was subject to cAMP, Ca2+ and protein kinase C (PKC) modulation. The excised pigmented rabbit conjunctivas were mounted in the modified Ussing-type chambers for measurement of unidirectional 36Cl fluxes under the open-circuit condition and of the short-circuit current (Isc), potential difference, and transconjunctival electrical resistance. The results indicate that Cl- secretion across the conjunctiva was abolished by mucosal application of 1 mM N-phenylanthranilic acid and was reduced by 40% by serosal application of 10 microM bumetanide. Net Cl- flux was stimulated by 133% by 1 mM 8-Br cAMP, 107% by 10 microM A23187, and 87% by 1 microM phorbol 12-myristate-13-acetate (PMA), suggesting that cAMP, Ca2+, and PKC all modulated active Cl- secretion, respectively. There existed a linear correlation between measured changes in net Cl- flux and observed changes in Isc (r2=0.99). The serial treatment of the conjunctiva with (a) 1 mM 8-Br cAMP and 10 microM A23187 and (b) 10 microM A23187 and 1 microM PMA resulted in sequence-independent, additive stimulation of Isc. In the case of 1 mM 8-Br cAMP and 1 microM PMA, additive stimulation of Isc was observed only when 1 mM 8-Br cAMP was added prior to 1 microM PMA. These results suggest that a given pharmacological agent may affect more than one channel type and that there might be a possible connection among the channels at the signal transduction level. In summary, Cl- appears to enter the pigmented rabbit conjunctiva from the serosal fluid via Na+-(K+)-2Cl- cotransport process and exit to the mucosal fluid via channels, resulting in active Cl- secretion. Active Cl- secretion in the pigmented rabbit conjunctiva appears to be modulated by cAMP, Ca2+, and PKC.

Functional expression of a truncated Ca(2+)-activated Cl- channel and activation by phorbol ester

We have isolated a niflumic acid-insensitive, Ca(2+)-activated Cl- channel (CaCC) from bovine trachea that migrates at 38 kDa (p38) on reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis. However, a cloned CaCC isolated from a tracheal cDNA expression library by screening with an antibody raised against p38 has a primary cDNA transcript of 2712 base pairs that codes for a 100-kDa protein and is not susceptible to dithiothreitol reduction. To test the hypothesis that the functional channel may be a much smaller posttranslationally processed form of the 100-kDa protein, we generated a mutant construct (CaCCX, 42.5-kDa protein) truncated at the NH2 and COOH termini. The whole cell currents of wild-type (wt) CaCC and CaCCX expressed in Xenopus oocytes were 10-fold higher than those of water-injected oocytes and were further increased by ionomycin or A-23187 and inhibited by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid and dithiothreitol. Whole cell currents in wtCaCC- and CaCCX-expressing oocytes could also be activated by phorbol 12-myristate 13-acetate and could be inhibited by chelerythrine chloride, suggesting that the cloned CaCC is regulated by protein kinase C. These results suggest that a smaller form of the full-length CaCC can form a functional channel.

Comparison of the intestinal secretory response to 5-hydroxytryptamine in the rat jejunum and ileum in-vitro

A secretory response to 5-hydroxytryptamine (5-HT) is observed throughout the intestinal tract; this investigation has compared the nature of this response in the jejunum and ileum of the rat in-vitro. Different basal electrical activity was observed for jejunal and ileal sheets of rat small intestine. In both intact and stripped preparations the basal short-circuit current (SCC) was greater and the tissue resistance lower in the jejunum than in the ileum. 5-HT caused concentration-dependent increases in SCC in intact and stripped preparations of both regions. EC50 values were similar in the jejunum and ileum, stripped sheets from both regions showing greater sensitivity. In the ileum the maximum increase in SCC induced by 5-HT was similar in intact and stripped sheets, but in the jejunum the response was greater in intact preparations. The jejunal response to 5-HT was reduced in the absence of bicarbonate but unaffected by lack of chloride, whereas the ileal response was inhibited by removal of chloride but unaltered in bicarbonate-free conditions. In intact sheets the tetrodotoxin-sensitive neural component was greater in the jejunum. In stripped sheets a neural component could still be detected in the ileum, but not in the jejunum. There are, therefore, fundamental differences in the way in which the jejunum and ileum respond to 5-HT stimulation--the jejunal response is primarily a result of stimulation of bicarbonate secretion whereas chloride secretion predominates in the ileum. The myenteric plexus appears to play a more prominent role in the jejunum; in the ileum other neural elements also contribute to the response.

Molecular events in neutrophil transepithelial migration

Neutrophil transepithelial migration is a central component of many inflammatory diseases of the gastrointestinal, respiratory and urinary tracts, and correlates with disease symptoms. In vitro modeling with polarized intestinal epithelial monolayers has shown that neutrophil transepithelial migration can influence crucial epithelial functions, ranging from barrier maintenance to electrolyte secretion. Studies have also demonstrated a dynamic involvement of the epithelium in modulating neutrophil transepithelial migration. Characterization of the molecular interactions between neutrophils and epithelial cells has revealed that transepithelial migration is dependent on the neutrophil beta 2 integrin CD11b/CD18, and does not appear to involve adhesive interactions with the selectins or intercellular adhesion molecule-1. Recent studies have implicated another transmembrane glycoprotein, CD47, as a crucial component of the transepithelial migration response. While the precise function of CD47 is not known, current evidence suggests that CD47-dependent events occur after CD11b/CD18-mediated neutrophil adhesion to the epithelium. This review will highlight key features of the current understanding of the molecular events important in neutrophil migration across epithelial surfaces.

Kinetic and pharmacological properties of the calcium-activated chloride-current in macrovascular endothelial cells

We have studied the kinetic and pharmacological properties of the Ca(2+)-activated Cl- current (ICl,Ca) in cultured cell pulmonary artery endothelial (CPAE) cells by means of combined patch clamp and Fura-2 micro fluorescence measurements. The current was activated by loading the cells via the patch pipettes with Ca(2+)-buffered solutions. Currents activated slowly at positive potentials, and decayed rapidly at negative potentials. The time constant of activation decreased at more positive membrane potentials and more elevated intracellular Ca2+ concentrations ([Ca2+]i). The time constant of deactivation was Ca(2+)-independent and decreased at more negative potentials. Steady-state currents showed strong outward rectification, but the instantaneous current-voltage relationship was almost ohmic. The calmodulin antagonists trifluoperazine (TFP) and calmidazolium inhibit ICl,Ca. Half maximal block for TFP occurred at 5.7 +/- 2.1 microM (n = 16). GTP gamma S did not activate ICl,Ca, but activated a Cl- current similar to the volume-activated Cl- current (ICl,vol). [Ca2+]i for half maximal activation of ICl,Ca was voltage-dependent, and suggests that the apparent binding constant for Ca2+ decreases with depolarization. Its value at 0 mV is 430 nM, and the binding site is 12% within the electrical field from the cytoplasmic side. The Hill-coefficient, nH, of the binding was larger than 1 and increased with depolarization. The maximal Cl- conductance at saturating [Ca2+]i did not depend on the membrane potential. RT-PCR experiments did not provide any evidence that the endothelial Ca(2+)-activated Cl- channel might be identical with a recently cloned Ca(2+)-sensitive Cl- channel (CaCC).

Luteinizing hormone activates chloride currents in hen ovarian granulosa cells

Luteinizing hormone (LH) induces progesterone production in hen ovarian granulosa cells, and this induction is inhibited when chloride ions are removed from the culture medium. This suggests that chloride channels may be involved in the signal transduction pathway responsible for the LH-induced progesterone production. In this report, we examined effects of LH on plasma membrane ion currents in single granulosa cells isolated from the largest preovulatory follicle (Fl) of the hen (Gallus domesticus). Using the perforated patch whole cell voltage clamp technique, we found that addition of LH rapidly activated a chloride current in these cells. This chloride current was present at all voltages tested (-90 to +50 mV), showed outward rectification and showed no obvious time or voltage dependence. Its magnitude was 3.5-fold that of the total resting membrane current measured before LH treatment. LH is known to elevate cyclic AMP in these cells. We found that addition of the cAMP analog Sp-cAMPS mimicked LH in inducing chloride currents in these cells. We conclude that LH can activate a chloride conductance in granulosa cells, and that this action may be mediated by cAMP.

Identification of a new outwardly rectifying Cl- channel that belongs to a subfamily of the ClC Cl- channels

A new outwardly rectifying Cl- channels (ORCC) that belongs to ClC Cl- channel family has been identified from rat kidney and designated as ClC-5. ClC-5 cDNA encodes a polypeptide of 746 amino acids, which is indicated by hydrophobicity analysis to have structural features that are common of the ClC family. However, the amino acid sequence was weakly homologous to those of other ClC Cl- channels except for ClC-3, which we recently identified as a Ca2+-sensitive ORCC. Northern blot analysis of rat tissues showed that ClC-5 mRNA was predominantly expressed in the kidney and colon. To characterize the functional properties of ClC-5 by whole cell patch-clamp technique, we established the stably transfected CHO-K1 cell line using intranuclear microinjection technique. The transfected cells induced outwardly rectifying and 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid-sensitive Cl- currents on whole cell configuration. Following the identification of two highly homologous ORCCs, ClC-3 and ClC-5, a new subfamily encoding ORCC has emerged in the ClC family. Furthermore, ClC-5 was almost identical to a partial sequence of human cDNA that is related to Dent's disease. The molecular structure and functional properties of ClC-5 will provide an important insight into ORCCs and the pathogenesis of Dent's disease.

Muscarinic receptors in the prenatal mouse embryo. Comparison of M35-immunohistochemistry with [3H]quinuclidinyl benzylate autoradiography

Muscarinic cholinergic receptors are widespread in nervous tissue and smooth muscle or paracrine epithelial cells of various organs. In the embryo, muscarinic receptors are transitorily expressed in the early blastoderm and later on in blastemic tissues during morphogenesis. Recently, a monoclonal antibody (M35) against muscarinic receptor from calf brain became available. In the present study the use of M35-immunohistochemistry is compared to autoradiographic localization of muscarinic binding sites in the mouse embryo. The aim of the study is to test the suitability of the antibody for localization of muscarinic receptors in embryonic tissues. For autoradiography whole-body sagittal cryostat sections of the 17- and 18-day mouse embryo were covered with LKB-Ultrofilm after incubation with the radioactive ligand [3H] quinuclidinyl benzylate (QNB). For immunohistochemistry cryostat sections of formalin fixed tissues were used. In general, all tissues exhibiting ligand binding were also recognized by the antibody. M35-immunohistochemistry resulted in higher spatial resolution of receptor localization than [3H]QNB autoradiography. Definitive muscarinic receptors were observed in smooth muscle and the epithelial lining of the vasuclar, intestinal, respiratory and urinary system, in the brain, spinal cord and peripheral nerves. The embryonic type of the muscarinic receptor was detected in the mesothelium of lung and liver, in the nephrogenic blastema of the metanephros, and in lung mesenchyme. A large amount of embryonic muscarinic receptors was found in the remnants of the notochord and in the nucleus pulposus of the developing vertebral column. A function in morphogenesis is discussed of the embryonic muscarinic receptor.

Forskolin- but not ionomycin-evoked Cl- secretion in colonic epithelia depends on intact microtubules

Several transport proteins are known to be trafficked to the cell membrane in response to appropriate secretagogues. In several cases, the response has been shown to be dependent on the cytoskeleton. We tested the hypothesis that the forskolin- and/or ionomycin-sensitive Cl- secretory response in colonic epithelia is dependent on an intact cytoskeleton. Using 125I- efflux as an assay for Cl- transport in the colonic epithelial cell line T84, we found that preincubation of the tissue for 3 h with either of two inhibitors of microtubule polymerization, nocodazole or colchicine, disrupted the cellular tubulin architecture and also reduced the forskolin- but not the ionomycin-evoked I- efflux. In contrast, brief exposure (4 min) to nocodazole was without effect on the forskolin-sensitive efflux, suggesting that the drug is not acting to block the stimulus-response pathway. An inactive structural analogue of colchicine, beta-lumicolchicine, had no inhibitory effect on either the forskolin-sensitive efflux or on microtubular structure. In a second model of Cl- secretion, the stripped rat colon, both colchicine and nocodazole reduced the forskolin-dependent short-circuit current by an average of 30-40%, suggesting a similar mechanism for insertion of Cl- channels into the plasma membrane. These findings suggest that the Cl- secretory response is dependent on microtubules and has a physiological role in the adenosine 3',5'-cyclic monophosphate-dependent, but not the Ca(2+)-dependent, Cl- secretion in colonic epithelia.

Niflumic and flufenamic acids are potent inhibitors of chloride secretion in mammalian airway

Effects of niflumic acid (NFA) and flufenamic acid (FFA), the two nonsteroid anti-inflammatory agents recently reported to inhibit Cl- current in Xenopus oocytes, were examined in cultured monolayers of dog and cow trachea. Both agents showed potent inhibition to the short-circuit current (Isc), an index of magnitude of transepithelial Cl- secretion, with values of Ki of 0.02 (for NFA) and 0.06 (for FFA) mM, respectively. The sensitivity sequence of Isc to the Cl- channel inhibitors tested was NFA > FFA > diphenylamine-2-carboxylate (DPC) >> anthracene-9-carboxylate (A9C). Thus, NFA and FFA are so far the most potent commercially available Cl- channel inhibitors tested in Cl(-)-secreting epithelia. The sensitivity sequence of 36Cl uptake to the above Cl- channel inhibitors in Xenopus laevis oocytes was found to be identical to the cultures of trachea. This seems to imply that the membrane Cl- channels of Xenopus oocytes are functionally similar to that identified in mammalian Cl(-)-secreting epithelia.

Probing the basic defect in cystic fibrosis

The concurrent developments in electrophysiology studies and the identification of the cystic fibrosis transmembrane conductance regulator (CFTR) gene has provided a unique opportunity to probe the basic cellular defect underlying cystic fibrosis. Various properties of the CFTR protein have been deduced from its primary sequence, the variety of mutations in patients and genotype-phenotype correlations, as well as the results of more recent DNA transfection studies. The most exciting observation is the fact that CFTR acts like a cAMP-regulated Cl- channel.