The existence and properties of volume-activated Cl- currents were studied in 15 different cell types (endothelium: human umbilical vein, human aorta, bovine pulmonary artery; fibroblasts: Swiss 3T3, L, C3H 10T1/2 and COS-1; epithelium: KB3, HeLa and A6; blood cells: RBL-2H3 and Jurkat; endothelioma cells derived from both subcutaneous and thymic hemangiomas; skin: IGR1 melanoma). Volume-activated Cl- currents with common characteristics, i.e. small conductance, outward rectification, higher permeability for iodide than for chloride and sensitivity to block by 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) could be elicited in all cells. The block of this current by tamoxifen and dideoxyforskolin is different for the various cell types, as well as the time course and the amplitude of the responses induced by repetitive applications of hypotonicity. Volume-activated Cl- channels with similar biophysical properties are therefore wide-spread among mammalian cells. This may reflect either a single Cl- channel that is ubiquitously expressed or a family of functionally related Cl- channels with cell specific expression patterns.

Possible interactions of cytoskeletal elements with mechanically induced membrane currents and Ca2+ signals were studied in human endothelial cells by using a combined patch-clamp and Fura II technique. For mechanical stimulation, cells were exposed to hypotonic solution (HTS). The concomitant cell swelling activates a Cl- current, releases Ca2+ from intracellular stores and activates Ca2+ influx. To interfere with the cytoskeleton, cells were loaded either with the F-actin-stabilizing agent phalloidin (10 mumol/l), or the F-actin-depolymerizing substance cytochalasin B (50 mumol/l). These were administered either in the bath or the pipette solutions. The tubulin structure of the endothelial cells was modulated by taxol (50 mumol/l), which supports polymerization of tubulin, or by the depolymerizing agent colcemid (10 mumol/l) both applied to the bath. Immunofluorescence experiments show that under the chosen experimental conditions the cytoskeletal modifiers employed disintegrate the F-actin and microtubuli cytoskeleton. Neither of these cytoskeletal modifiers influenced the HTS-induced Cl- current. Ca2+ release was not affected by cytochalasin B, taxol or colcemid, but was suppressed if the cells were loaded with phalloidin. Depletion of intracellular Ca2+ stores by thapsigargin renders the intracellular [Ca2+] sensitive to the extracellular [Ca2+], which is indicative of a Ca2+ entry pathway activated by store depletion. Neither cytochalasin B nor phalloidin affected this Ca2+ entry. We conclude that F-actin turnover or depolymerization is necessary for Ca2+ release by mechanical activation. The tubulin network is not involved. The Ca2+ release- activated Ca2+ entry is not modulated by the F-actin cytoskeleton.

1 We have studied the permeation and pharmacological properties of a recently described volumeactivated, calcium-insensitive, small-conductance Cl--channel 4 Extracellular ATP induced a low-affinity block of the current, that showed a small voltagedependence (K, = 4.9 mmol 1-at + 80 mV and K, = 8.2 mmol 1`at -80 mV). 5 Extracellularly applied arachidonic acid (10 imol 1-) irreversibly blocked the current in 5 out of 9 cells. This block seems to be non-specific, because other ionic currents, e.g. inwardly rectifying K+ currents, were blocked as well.6 Tamoxifen induced a high affinity block of the current (K, = 2.9 pmol 1`). Block and reversal of block were however much slower than with NPPB. 7 Cytotoxic compounds, which are substrates of the P-glycoprotein multidrug transporter, loaded into endothelial cells via the patch pipette, exerted only minor effects on the volume-activated current. Vinblastine and colcemid did not affect the volume-activated current, whereas daunomycin and vincristine induced a slow 'run-down' of the current. 8 The similarity between permeation and pharmacological properties of volume-activated Cl--currents in endothelial cells and those in many other cell types may suggest that they all belong to the same family of volume-activated small-conductance Cl--channels. Evidence that they belong to the class of P-glycoprotein associated Cl--channels is however only marginal, whereas their biophysical characteristics differ significantly from those of the CIC-2 volume-activated Cl--channels.

1. We have measured changes in cell volume, membrane potential and ionic currents in distal nephron A6 cells following a challenge with hypotonic solutions (HTS). 2. The volume increase induced by HTS is compensated by a regulatory volume decrease (RVD), which is inhibited by both 5-nitro-2-(3-phenylpropylamino)-benzoate (NPPB) and quinine. Quinine (500 uM) completely blocked RVD, whereas 100 /M NPPB delayed and attenuated RVD.3. The resting potential in A6 cells was -52-3 + 4.8 mV (n = 53), and shifted to -35-1 + 2-2 mV (n = 33) during HTS. 4. Resting membrane current in A6 cells was 0 35 + 0-12 pA pF-1 at -80 mV and 0-51 + 0-16 pA pF-1 at +80 mV (n = 5). During cell swelling these values increased to 11.5 + 1 1 and 29-3 + 2-8 pA pF-1 (n = 29), respectively. 5. Quinine (500 /zM) completely blocked the HTS-activated current at -15 mV, the reversal potential for Cl-currents, but exerted only a small block at -100 mV (K+ equilibrium potential). NPPB (100 /M) inhibited the current at both potentials almost to the same extent. The HTS-induced net current reversed at -41 + 2-5 mV (n = 15), which is close to the measured resting potential during HTS. 6. The quinine-insensitive current reversed near the Cl-equilibrium potential. The quininesensitive current reversed near the K+ equilibrium potential. The respective conductances activated by HTS at the zero-current potential were 2-1 + 0 7 nS for K+ and 5-2 + 1-3 nS for Cl (n = 15). 7. Single channel analysis unveiled activation of at least two different channels during HTS.A 36 pS channel reversing at the Cl-equilibrium potential showed increased open probability at depolarized potentials. HTS also activated a K+ channel with a 29 pS conductance in high-K+ extracellular solutions (130 mM) or 12 pS in 2-5 mm K+.8. This coactivation of K+ and Cl-channels shifts the membrane potential towards a value between EK and Eci (the reversal potentials for K+ and CF), where a net efflux of Cl-(Clinward current) and K+ (K+ outward current) under zero-current conditions occurs. Block of either the K+ or the Cl-conductance will shift the zero-current potential towards the equilibrium potential of the unblocked channel, preventing net efflux of osmolytes and RVD. This coactivation of K+ and Cl-currents causes a shift of osmolytes out of the cells, which almost completely accounts for the observed RVD.Processes involved in the regulation of cell volume have We have recently shown that a challenge with hyposmotic been widely studied during the last 10 years (for reviews, solutions was not followed by a regulatory volume decrease see McCarty

We have measured ionic currents and changes in intracellular Ca2+ concentration ([Ca2+]i) induced by extracellular ATP in single epithelial cells of the distal nephron from toad (A6 cells). ATP increased [Ca2+]i and concomitantly activated ionic currents. The ATP concentration for half-maximal increase in [Ca2+]i was approximately 10 microM. Current activation and elevation of [Ca2+]i also occurred in Ca(2+)-free bath solutions but were abolished by loading the cells via the patch pipette with 10 mM 1,2-bis(2-aminophenoxy)ethane N,N,N',N'-tetraacetic acid (BAPTA) or by preincubating the cells with 10 microM BAPTA-acetoxymethyl ester for 120 min. ATP-activated currents reversed at -53.9 +/- 1.9 mV (n = 22). Tetraethylammonium (TEA, 25 mM), a K+ channel blocker, partially blocked this current but did not affect the Ca2+ transients. The TEA-insensitive component of the current reversed close to Cl- equilibrium potential. 5-Nitro-2-(3-phenylpropylamino) benzoic acid, a putative Cl- channel blocker (100 microM), abolished nearly completely the ATP-activated current. Suramin (100 microM), a P2-purinergic receptor antagonist, strongly attenuated both Ca2+ transients and currents. In cell-attached patches, single channel currents activated by ATP could be observed, i.e., an inwardly rectifying K+ channel with a slope conductance for inward currents of approximately 32 pS and an ohmic Cl- channel with a conductance of 34 pS. It is concluded that ATP activates both Cl- and K+ channels in distal nephron epithelial cells by a Ca(2+)-dependent mechanism.