Functional expression of rat brain cloned α1E calcium channels in COS-7 cells

 The properties of the rat brain α1E Ca²⁺ channel subunit and its modulation by accessory rat brain α2-δ and β1b subunits were studied by transient transfection in a mammalian cell line in order to attempt to reconcile the debate as to whether α1E forms a low-voltage-activated (LVA) or high-voltage-activated (HVA) Ca²⁺ channel and to examine its pharmacology in detail. α1E alone was capable of forming an ion-conducting pore in COS-7 cells. The properties of heteromultimeric α1E/α2-δ/β1b channels were largely dictated by the presence of the β1b subunit, which increased current density and tended to produce a hyperpolarizing shift in the voltage dependence of activation and inactivation. α1E/α2-δ/β1b channels did not appear to be regulated by Ca²⁺-induced inactivation. α1E was shown to exhibit a unique pharmacological profile. ω-Agatoxin IVA blocked the current in a dose-dependent manner with an IC₅₀ of approximately 50 nM and a maximum inhibition of about 80%, whilst ω-conotoxin MVIIC was without effect. The 1,4-dihydropyridine (DHP) antagonist nicardipine (1 μM) produced an inhibition of 51 ± 7%, whereas the DHP agonist S-(–)BAY K 8644 was without effect. Our findings suggest a re-evaluation of the classification of the α1E Ca²⁺ channel subunit; we propose that rat brain α1E forms a novel Ca²⁺ channel with properties more similar to a subtype of LVA than HVA Ca²⁺ current. Received: 30 August 1996 / Received after revision and accepted: 28 October 1996     

Blockade of HERG channels expressed in Xenopus oocytes by external H+

 We have investigated the effect of external H⁺ concentration ([H⁺]_o)on the human-ether-a-go-go-related gene (HERG) current (I _HERG), the molecular equivalent of the cardiac delayed rectifier potassium current (I _Kr), expressed in Xenopus oocytes, using the two-microelectrode voltage-clamp technique. When [H⁺]_o was increased, the amplitude of the I _HERG elicited by depolarization decreased, and the rate of current decay on repolarization was accelerated. The activation curve shifted to a more positive potential at lower external pH (pH_o) values (the potential required for half-maximum activation, V _1/2, was: –41.8 mV, –38.0 mV, –33.7 mV, –26.7 mV in pH_o 8.0, 7.0, 6.6, 6.2, respectively). The maximum conductance (g _max) was also affected by [H⁺]_o: a reduction of 7.9%, 14.6%, and 22.8% was effected by decreasing pH_o from 8.0 to 7.0, 6.6, and 6.2, respectively. We then tested whether this pH effect was affected by the external Ca²⁺ concentration, which is also known to block HERG channels. When the extracellular Ca²⁺ concentration was increased from 0.5 mM to 5 mM, the shift in V _1/2 caused by increasing [H⁺]_o was attenuated, suggesting that these two ions compete for the same binding site. On the other hand, the decrease in g _max caused by increasing [H⁺]_o was not significantly affected by changing external Ca²⁺ levels. The results indicate that HERG channels are inhibited by [H⁺]_o by two different mechanisms: voltage-dependent blockade (shift of V _1/2) and the decrease in g _max. With respect to the voltage-dependent blockade, the interaction between H⁺ and Ca²⁺ is competitive, whereas for the decreasing g _max, their interaction is non-competitive. Received: 12 January 1999 / Received after revision: 15 February 1999 / Accepted: 16 February 1999     

The effects of nucleotides and potassium channel openers on the SUR2A/Kir6.2 complex K+ channel expressed in a mammalian cell line, HEK293T cells

 The effects of potassium channel opening drugs and intracellular nucleotides on the ATP-sensitive K+ (K_ATP) channel composed of SUR2A and Kir6.2 in HEK293T cells were examined using the patch-clamp technique. The SUR2A/Kir6.2 channel was activated effectively by pinacidil, marginally by nicorandil but not by diazoxide. The pinacidil-activated channel currents were inhibited by glibenclamide with a K _i value of 160 nM. Upon formation of inside-out (I-O) patches, spontaneous openings of the channels appeared, which were inhibited by intracellular ATP (ATP_i) equipotently in the presence and in the absence of intracellular Mg²⁺ (Mg²⁺ _i). The channel activity ran-down gradually in I-O patches. The run-down channels could be reactivated by ATP_i only in the presence of Mg²⁺ _i. Uridine 5’-diphosphate (UDP) antagonized the ATP_i-mediated inhibition of the channel activity before run-down. After run-down, UDP activated the channel without antagonizing ATP_i-mediated channel inhibition. Thus, the SUR2A/Kir6.2 reproduced the major properties of the native cardiac K_ATP channel well in terms of nucleotide regulation and pharmacology, and therefore can be a useful tool with which to elucidate the molecular mechanisms characterizing the K_ATP channel. Received: 24 October 1997 / Received after revision and accepted: 4 December 1997     

Channel-specific effects of n-alkyl sulphate anions on three Shaker-related potassium channels expressed in Xenopus oocytes

 n-Alkyl sulphate anions have been shown to reversibly affect the functioning of voltage-gated ion channels in a variety of preparations. They are suggested to exert their effects by increasing the magnitude of the negative surface potential at the external face of the membrane. Here we report the effects of n-octyl sulphate (OS^–) and n-dodecyl sulphate (DDS^–) on RCK1 (Kv1.1), RCK4 (Kv1.4) and Shaker B potassium channels exogenously expressed in Xenopus oocytes. Both OS^–and DDS^–produced a hyperpolarising shift in the activation voltage dependence of all three channels, consistent with an increased negative external surface potential. Similar kinetic changes were also observed, the kinetics of both activation and inactivation being accelerated in the presence of OS^–and DDS^–. However, we also found that while 10 mM OS^–and 50 μM DDS^–significantly increased the maximum conductance of RCK1 and Shaker B channels, 5 mM OS^–and 15 μM DDS^–produced a large decrease in RCK4 conductance; the possible involvement of RCK4 residue K533 in this effect is discussed. Our data indicate that n-alkyl sulphate anions can perturb ion channel function in a variety of ways and that their effects are complex and channel specific. Received: 12 December 1996 / Received after revision and accepted: 21 January 1997     

Characteristics of rat downregulated in adenoma (rDRA) expressed in HEK 293 cells

Studies with apical membrane vesicles have shown that two distinct and separate anion exchange processes are present in rat distal colon, 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS)-sensitive exchange, and DIDS-resistant Cl⁻–OH⁻ exchange. These studies proposed that anion exchanger (AE)-1 isoform encodes the former as both apical membrane DIDS-sensitive exchange, and AE1 specific mRNA are present only in surface cells and are downregulated in Na-depleted rats, whereas downregulated in adenoma (DRA) encodes the latter as both DIDS-resistant Cl⁻–OH⁻ exchange, and DRA-specific proteins are present in apical membranes of both surface and crypt cells and are not altered in Na⁺-depleted rats. Studies were, therefore, initiated to identify the function of rat DRA (rDRA) in vitro. rDRA cDNA isolated from rat distal colon encodes a 757-amino-acid protein which has 96 and 81% homology with mDRA and hDRA, respectively. rDRA-specific mRNA expression was detectable only in specific segments of the digestive tract (duodenum, ileum, cecum, proximal colon, and distal colon) but not in the stomach, jejunum, or in the kidney, brain, heart, and lung. HEK 293 cells stably transfected with rDRA exhibited DIDS-insensitive and intracellular acid pH (pH_i 6.5)-sensitive Cl uptake that: (1) was significantly stimulated by outward Cl⁻, , isobutyrate, and possibly OH⁻ gradients; (2) was saturated as a function of increasing extracellular Cl concentrations with an apparent K _m for Cl of 2.9 ± 0.3 mM; and (3) was inhibited competitively by extracellular oxalate but not by . A high rate of DIDS-insensitive Cl influx at pH 6.5 was also present under physiological Cl⁻ concentration. Our observations that rDRA mediates DIDS-insensitive, acid pH-dependent Cl⁻ uptake are consistent with prior observations that rDRA does not mediate DIDS-sensitive exchange in rat distal colon. We speculate that, in addition to mediating pH-sensitive Cl⁻ uptake, rDRA may function as a modifier of other anion transport proteins.     

Small (SKCa) Ca2+-activated K+ channels in cultured rat hippocampal pyramidal neurones

 Small (SK_Ca) Ca²⁺-activated K⁺ channels were identified in membrane patches excised from cultured CA1-CA3 pyramidal neurones of the neonatal rat hippocampus. When recorded in low-K⁺ extracellular solution ([K⁺]_o=2.5 mM), SK_Ca channels had a low conductance (@3 pS at 0 mV), were activated by ≥175 nM Ca²⁺ (P _o=0.54 at 500 nM Ca²⁺) and there were two open-time components (2.1 and @70 ms) to their activity. These properties of single SK_Ca channels are similar to those of slow after-hyperpolarization channels (sAHP) previously inferred from fluctuation analysis of the sAHP current. It is concluded that the SK_Ca channel reported here may be the channel that generates the sAHP in hippocampal pyramidal neurones. Received: 9 July 1998 / Received after revision: 5 October 1998 / Accepted: 7 October 1998     

KVLQT channels are inhibited by the K+ channel blocker 293B

Previous data have indicated that the chromanol 293B blocks a cAMP activated K⁺ conductance in the colonic crypt, a K⁺ conductance in pig cardiac myocytes and the K⁺ conductance induced by IsK protein expression in Xenopus oocytes. We have also shown that cAMP-activated cystic fibrosis transmembrane conductance regulator (CFTR) up-regulates, apart from the typical Cl^–current, a 293B- inhibitable K⁺ current. Very recently it has been shown that the IsK protein interacts with K_VLQT subunits to produce a K⁺ channel. These data have prompted us to ask the following questions: Is the 293B-inhibitable current in oocytes expressing CFTR and activated by cAMP caused by an endogenous Xenopus K_VLQT (XK_VLQT), and is mouse K_VLQT (mK_VLQT) expressed in oocytes inhibited by 293B? Antisense and sense probes for XK_VLQT were coinjected with CFTR cRNA into oocytes. After 3–4 days the oocytes were examined by two electrode voltage clamp. It was found that in control oocytes expressing CFTR and stimulated by isobutylmethylxanthine (IBMX, 1 mmol/l) 293B (10 μmol/l) reduced the conductance (G_m). In oocytes coinjected with the sense probe for XK_VLQT and pretreated with IBMX 293B still reduced G_m, whilst the 293B-inhibitable G_m was almost completely absent in oocytes coinjected with XK_VLQT antisense. In another series a full length clone for mK_VLQT was generated by PCR techniques and the cRNA was injected into oocytes. After several days these oocytes, unlike water injected ones, were found to be strongly hyperpolarized and their G_m was increased significantly. The oocytes were depolarized significantly and their G_m was reduced reversibly by 10 μmol/l 293B. These data indicate that CFTR activation by IBMX indeed co-activates an endogenous oocyte XK_VLQT channel and that this channel is inhibited by a new class of channel blockers, of which 293B is the prototype.     

Influence of outer pore residue K533 on the inhibition of Kv1.4 potassium channels by n-alkyl sulphate anions

 We have previously shown that although n-octyl sulphate (OS^–) and n-dodecyl sulphate (DDS^–) anions had similar effects on the kinetics and activation voltage dependence of RCK1 (Kv1.1), RCK4 (Kv1.4) and Shaker B channels expressed in Xenopus oocytes, both compounds produced a large decrease in the maximum conductance of RCK4 channels while significantly increasing the conductance of RCK1 and Shaker B. We suggested that this channel-specific inhibition might depend on the nature of the amino-acid residue corresponding to position 533 in RCK4. We now present data on the effects of n-alkyl sulphates on an RCK4 mutant in which the wild-type lysine at position 533 was changed to the corresponding tyrosine residue in RCK1. At a concentration of 15 μM, DDS^– caused a 48% reduction in the wild-type current at 50 mV but a 32% increase in the mutant current. n-Hexyl sulphate and OS^– had similar differential effects. The activation and inactivation kinetics of the mutant current were still accelerated by n-alkyl sulphates and 15 μM DDS^– moved the conductance/voltage curves of both wild-type and mutant channels some 24 mV in the hyperpolarizing direction. The K533Y mutation thus had a selective effect on current inhibition by n-alkyl sulphates. Received: 6 April 1998 / Accepted: 18 May 1998     

Molecular basis for different pore properties of potassium channels from the rat brain Kv1 gene family

 Members of the rat brain Kv1 family of cloned potassium channels are structurally highly homologous, but have diverse conductance and pharmacological characteristics. Here we present data on the effects of mutating residues K533 in the P-region and H471 in the S4-S5 linker of Kv1.4 to their equivalent residues in Kv1.1 and Kv1.6 on single-channel conductance and sensitivity to external tetraethylammonium cations (TEA⁺) and internal Mg²⁺. Exchange of residue K533 for its equivalent residue (Y) in Kv1.1 and Kv1.6 increased the single-channel conductance at both negative and positive potentials. This mutation is known to reduce the IC₅₀ for external TEA⁺ from > 100 mM to 0.6 mM, almost identical to that for Kv1.1 (0.53 mM). We have now found that the additional exchange of residue H471 for the equivalent residue (K) in Kv1.6 increased the IC₅₀ for external TEA⁺ from 0.6 mM (Kv1.4K533Y) to 2.39 mM; this is very close to that for wild-type Kv1.6 channels (2.84 mM). The mutation H471K alone was ineffective. We thus provide evidence that the S4-S5 linker does contribute to the channel’s inner-pore region. Data on the block of Kv1 channels by internal Mg²⁺ indicate that while the binding site is probably situated within the deep-pore region, its exact location may be channel specific. Received: 25 February 1997 / Received after revision: 4 June 1997 / Accepted: 10 June 1997     

Modulation of cloned human neuronal voltage-gated potassium channels (hKv1.1 and hKv2.1) by neurosteroids

 During Ca²⁺-activated C^– current oscillations a mechanical deformation of the Xenopus laevis oocyte by a fluid stream evokes transient inward currents of high amplitude (stream evoked inward current, I _i,st). This current can be observed either in native or RNA-injected oocytes expressing ligand-controlled ion channels from rat brain. I _i,st reversed at the equilibrium potential of chloride and was blocked by 9-anthracene carboxylic acid (2 mM). Power spectral analysis of the oscillations did not reveal a correlation between the features of the oscillations and the amplitude of I _i,st. Antagonists of stretch-activated cation channels [gadolinium (100 μM) and lanthanum (1mM)] did not block I _i,st. Calcium channel blockers [cobalt and manganese (10 mM)] did not inhibited I _i,st and I _i,st could also be elicited in calcium-free medium. Preloading oocytes with pertussis toxin (PTX) for 17 h prevented current oscillations and I _i,st caffeine (10 mM), an antagonist of the liberation of calcium from intracellular stores, inhibited I _i,st. Our results proride evidence for modulation of the mechanosensitivity of chloride currents by activation of intracellular second messenger cascades. Received: 28 Oktober 1997 / Received after revision: 5 January 1998 / Accepted: 6 January 1998     

Regulation of stably expressed and native BK channels from human myometrium by cGMP- and cAMP-dependent protein kinase

 The cloned BK channel α subunit from human myometrium was stably expressed in Chinese hamster ovary cells, either alone (CHOα cells) or in combination with the auxiliary β subunit (CHOα+β cells). We studied basic channel properties and the effects of cGMP- and cAMP-dependent protein kinases on the BK channel activity. Coexpression of α and β subunits enhanced the Ca²⁺ and voltage sensitivity of the BK channel, and decreased the inhibitory potency of iberiotoxin. Blocking and stimulating effects on BK channel activity by charybdotoxin and nitric oxide, respectively, were independent of the β subunit. The cGMP kinase Iα and cAMP kinase failed to affect BK channel activity in CHOα and CHOα+β cells at different [Ca²⁺]_i and voltages. In contrast, BK channels in freshly isolated myometrial cells from postmenopausal women responded to cAMP kinase and cGMP kinase with a fourfold and twofold decrease in their open probability (NP _o), respectively. These effects could be reversed by alkaline phosphatase and remained unaffected by the phosphatase inhibitor okadaic acid (100 nM). In 28% of myometrial cells, however, cAMP and cGMP kinases increased NP _o 2-fold and 3.5-fold, respectively. This stimulation was enhanced rather than reversed by alkaline phosphatase and was abolished by 100 nM okadaic acid. The results suggest that in stably transfected CHO cells the expressed BK channel is not regulated by cAMP kinase and cGMP kinase. However, in native myometrial cells stimulatory and inhibitory regulation of BK channels by cAMP kinase and cGMP kinase was observed, suggesting that channel regulation by the protein kinases requires factors that are not provided by CHO cells. Alternatively, failure of regulation may have been due to the primary structure of the myometrial BK channel protein used in this study. Received: 17 March 1998 / Received after revision: 15 May 1998 / Accepted: 28 May 1998     

Inhibition of the L-type calcium channel by the five muscarinic receptors (m1–m5) expressed in NIH 3T3 cells

 Modulation of L-type calcium channels by the five cloned muscarinic receptors was studied by expression of the receptors in NIH 3T3 cells. Application of acetylcholine (ACh) to cells transfected with m1–m5 resulted in a reduction in the L-type calcium current amplitude. Elevations in intracellular cAMP concentrations induced by 8-bromo-cAMP or forskolin resulted in no discernible change in the L-type calcium current. In addition, treatment with Rp-adenosine 3′,5′-cyclic monophosphothioate triethylamine (Rp-cAMPS), a protein kinase A (PKA) inhibitor, had no effect on the L-type currents. Conversely, application of phorbol dibutyrate, an activator of protein kinase C (PKC) or 8-bromo-cGMP, an activator of cGMP-dependent protein kinase (PKG), reduced the calcium currents. Incubation of the cells with KT5823, an inhibitor of PKG, resulted in a reduction of the response to 8-bromo-cGMP. The ACh-induced depression of L-type calcium current amplitude was sensitive to pertussis toxin (PTX) in cells transfected with the m2 or m4 receptor subtype. The m2-muscarinic-receptor-induced inhibition of the L-type calcium current was attenuated by preincubation of the cells with 8-bromo-cAMP and was unaffected by KT5823 or by calphostin C. The m1-muscarinic-receptor-induced inhibition of the L-type calcium conductance was insensitive to PTX treatment. However, the m1-induced response was blocked by preincubation of the cells with calphostin C. The present data indicate that the m2 (and possibly also the m4) muscarinic receptors inhibit the L-type calcium conductance by a reduction in cAMP concentration and that the m1 (and possibly also the m3 and m5) muscarinic receptors inhibit the L-type calcium channel via activation of PKC. Received: 2 September 1996 / Received after revision and accepted: 15 October 1996     

cAMP-mediated long-term modulation of voltage-dependent K+ channels in cultured colliculi neurons

 Previously, we have described prolonged cAMP-induced inhibition of a K⁺ current in cultured colliculi neurons. The aim of the present study was to characterize the channel responsible for this cAMP-dependent effect. We detected the presence of a non-inactivating voltage-dependent 16-pS K⁺ channel that displayed long-lasting inhibition upon a brief application of cAMP and greater sensitivity to tetraethylammonium than to 4-aminopyridine. In addition to this channel, colliculi neurons express two other voltage-sensitive, non-inactivating K⁺ channels (8 and 49 pS) whose activity is facilitated by a brief application of cAMP, the effect of which is also long-lasting. These results suggest the presence of common sustained cAMP-dependent processes responsible for both up- and down-regulation of these channels in the neurons studied. They indicate that the 16-pS, but not the 8-pS or the 49-pS channels, participates in the cAMP-inhibited macroscopic K⁺ current. Received: 21 April 1998 / Received after revision: 10 July 1998 / Accepted: 5 August 1998     

Voltage-dependent gating of veratridine-modified RIIA Na+ channel α subunit expressed heterologously in CHO cells

The voltage-dependent kinetics of veratridine-modified RIIA Na⁺ channel α subunit expressed heterologously in CHO cells were studied using the whole-cell patch-clamp technique. The activation and deactivation kinetics are well described by double exponential functions but poorly by a monoexponential function. Unlike the slow component, the fast time constant and associated amplitude factor depended steeply on the potential. The steady-state activation of veratridine-modified channels is described by a Boltzmann function with a V _1/2 of –131.9 mV and a slope of 9.41 mV. A two-state model is proposed for the fast component that explains the kinetics of veratridine’s mechanism of action. Received: 11 December 1998 / Received after revision: 23 March 1999 / Accepted: 7 April 1999     

Endosomes: another extra-mitochondrial location of type-1 porin/voltage-dependent anion-selective channels (VDAC)

 Endocytotic vesicles (EV) isolated from rat renal cortex were subjected to SDS-polyacrylamide gel electrophoresis and Western blotting. A monoclonal antibody against human type-1 porin (31 kDa) detected a strong band of 31 kDa. The same antibody has been used as the primary antibody in indirect immunocytochemistry. Light microscopy of cryostat sections of rat renal cortex showed a heavy staining of EV underneath the brush-border membrane. Electron microscopy was performed by ”preembedding immunogold staining” of rat renal cortex, the sections of which showed an extensive labelling of EV with gold particles. These results demonstrate that the expression of type-1 porin is not restricted to outer mitochondrial membranes. The biological function of endosomal type-1 porin has as yet to be ascertained. Received: 10 February 1998 / Accepted: 3 April 1998     

Electrophysiological study on the high K+ sensitivity of rat glomerulosa cells

 Elevation of extracellular potassium concentration by as little as some tenth of mM activates rat adrenal glomerulosa cells. In the present study some factors responsible for this high K⁺ sensitivity were examined. Using whole-cell voltage-clamp technique we found that both T-type and L-type voltage-dependent Ca²⁺ channels have very low threshold potential (–71 and –58 mV, resp.). By means of patch-clamp technique combined with single-cell fluorimetry we also provided evidence that the activation of I_gl, a K⁺-activated inward rectifying current is associated with Ca²⁺ influx. Both the low activation threshold of voltage-dependent Ca²⁺ channels and the function of I_gl contribute to the exceptional K⁺ sensitivity of the glomerulosa cells. Received: 30 September 1997 / Accepted: 4 November 1997     

N-Terminal deletions of rKv1.4 channels affect the voltage dependence of channel availability

Rat Kv1.4 potassium channels undergo rapid inactivation, which is mediated by the N-terminal structure of the polypeptide. This inactivation can be removed by N-terminal deletion of about 20 residues. However, more substantial deletion (e.g. 37 residues) restores inactivation suggesting a second inactivating domain [Kondoh et al. J Biol Chem 272:19333–19338, 1997]. Here we provide evidence that this inactivation shares all properties with N-type inactivation. Pore mutations, which are supposed to affect C-type inactivation, have no effect. In addition, the redox regulation of inactivation, which is typical for Kv1.4 channels, can be conferred to the inactivation of the deleted constructs by incorporation of an N-terminal cysteine residue. The most remarkable feature of this secondary inactivation is the existence of two components in the steady-state voltage dependence of inactivation. For mutant rKv1.4Δ2–37 about 90% of the channels only activate when the holding membrane potential is more negative than about –120 mV; the remaining 10% show the typical threshold at –60 mV. Mutagenesis of the truncated channel affected the relative amplitudes of these two components, but not the voltage dependence. The results suggest that the secondary ball structures of rKv1.4 channels interact with the protein structures responsible for activation. Received: 3 February 1999 / Accepted: 15 March 1999     

Potassium channels of adult locust (Schistocerca gregaria) muscle

Two types of K⁺ channels have been identified in patches of plasma membrane of metathoracic extensor tibiae muscle fibres of adult locust, Schistocerca gregaria. One channel had a maximum conductance of 170 pS, fast open-closed kinetics, and a linear current/ voltage relationship. In inside-out patches it was activated by ‘‘internally applied’’ Ca²⁺, but at unexpectedly low levels (between 10⁻¹⁰ and 10⁻⁹M). The other channel had a maximum conductance of 35 pS, slower open-closed kinetics, and was not activated by Ca²⁺. In cell-attached patches, its channel conductance measured in symmetrical salines was about three times greater for hyperpolarisations than for depolarisations. This inward rectification was proved to be due to block by intracellular Mg²⁺. For both channels, open probability (P _o) and mean open time increased during depolarisations and decreased during hyperpolarisations, resulting in outward rectifications in terms of net current (I _n , product of the single-channel current and P _o). For both channels, the K⁺ conductance was 10 times greater than that for Na⁺. Internally applied tetraethylammonium or tetramethylammonium ions blocked both channels. Received: 12 June 1995/Received after revision and accepted: 30 January 1996     

Secretory apical Cl– channels in A6 cells: possible control by cell Ca2+ and cAMP

Distal kidney cells (A6) from Xenopus laevis were cultured to confluency on porous supports. Tissues were mounted in Ussing-type chambers to measure short-circuit current (I _sc), transepithelial conductance and capacitance, and to analyse the fluctuation in I _sc. In the absence of apical NaCl, but with normal basolateral NaCl Ringer’s solution, extracellular addition of ATP, oxytocin, a membrane-permeant cAMP derivative, and forskolin produced a transient increase of the electrical parameters. Noise analysis revealed a spontaneous Lorentzian component. All responses depend strictly on the presence of basolateral Cl^– and are caused by the activation of an apical (CFTR type) Cl^– permeability. Repetitive treatment with ATP (or oxytocin) resulted in refractoriness. ATP and oxytocin acted antagonistically, whereas cAMP and ATP had additive effects. Incubation with the vesicular Ca²⁺ pump inhibitor thapsigargin or application of the Ca²⁺ channel blocker nifedipine elicited finite but variable Cl^– channel activity. After treatment with nifedipine or thapsigargin, the response to oxytocin was severely impaired. We speculate that not only cAMP but also cell Ca²⁺ plays a crucial role in the activation of CFTR in A6. Ca²⁺ may be multifunctional but the rise in capacitance (apical area) observed with all stimulants strongly suggests its involvement in, and contribution to, exocytosis in the process of the CFTR-mediated transcellular Cl^– movements. Received: 30 November 1998 / Received after revision: 23 February 1999 / Accepted: 12 March 1999