The Vertebrate 7S K RNA Separates Hagfish (Myxine glutinosa) and Lamprey (Lampetra fluviatilis)

7S RNA sequences from the hagfish (Myxiniformes) and lamprey (Petromyzontiformes) were cloned and analyzed. In both species, 7S L RNA (also designated SRP RNA, since it represents the RNA constituent of the signal recognition particle) was clearly detectable. The sequence similarity between the two species was 86%, compared with about 75% similarity between either of these species and mammals. 7S K RNA was also cloned from the lamprey. The similarity between the 7S K RNA of the lamprey and that of mammals was 68%. Interestingly, several interspersed elements were found with nearly 100% similarity compared with mammals. In contrast to the lamprey, no 7S K RNA-related sequences were detectable among hagfish RNA, neither in northern blots nor with the PCR assay. In view of the significant conservation between the 7S K RNA of lamprey and that of mammals (human), this unexpected result clearly separates lamprey and hagfish. In addition, the lack of detectable 7S K RNA sequences in an outgroup, such as amphioxus, indicates that these results do not reflect an autapomorphy of hagfish. Therefore, our data provide additional support to the notion of a sister group relationship between Petromyzontiformes and gnathostomous vertebrates to the exclusion of Myxiniformes. Received: 24 September 1999 / Accepted: 9 February 2000     

Effects of Permeant Cations on K+ Channel Gating in Nerve Axons Revisited

An increase in extracellular potassium ion concentration, K _o , significantly slows the potassium channel deactivation rate in squid giant axons, as previously shown. Surprisingly, the effect does not occur in all preparations which, coupled with the voltage independence of this result in preparations in which it does occur, suggests that it is mediated at a site outside of the electric field of the channel, and that this site is accessible to potassium ions in some preparations, but not in others. In other words, the effect does not appear to be related to occupancy of the channel by potassium ions. This conclusion is supported by a four-barrier, three-binding site model of single file diffusion through the channel in which one site, at most, is unoccupied by a potassium ion (single-vacancy model). The model is consistent with current-voltage relations with various levels of K _o , and, by definition, with multiple occupancy by K⁺. The model predicts that occupancy of any given site is essentially independent of K _o (or K _i ). The effects of extracellular Rb⁺ and Cs⁺ on gating are strongly voltage dependent, and they were observed in all preparations investigated. Consequently, the mechanism underlying these results would appear to be different from that which underlies the effect of K⁺ on gating. In particular, the effect of Rb⁺ on gating is reduced by strong hyperpolarization, which in the context of the occupancy hypothesis, is consistent with the voltage dependence of the current-voltage relation in the presence of Rb⁺. The primary, novel, finding in this study is that the effects of Cs⁺ are counterintuitive in this regard. Specifically, the slowing of channel deactivation rate by Cs⁺ is also reduced by hyperpolarization, similar to the Rb⁺ results, whereas blockade is enhanced, which is seemingly inconsistent with the concept that occupancy of the channel by Cs⁺ underlies the effect of this ion on gating. This result is further elucidated by barrier modeling of the current-voltage relation in the presence of Cs⁺. Received: 19 December 1995/Revised: 10 June 1996     

Characterization of K+ Channels in the Basolateral Membrane of Rat Tracheal Epithelia

To study K⁺ channels in the basolateral membrane of chloride-secreting epithelia, rat tracheal epithelial monolayers were cultured on permeable filters and mounted into an Ussing chamber system. The mucosal membrane was permeabilized with nystatin (180 μg/ml) in the symmetrical high K⁺ (145 mm) Ringer solution. During measurement of the macroscopic K⁺ conductance properties of the basolateral membrane under a transepithelial voltage clamp, we detected at least two types of K⁺ currents: one is an inwardly rectifying K⁺ current and the other is a slowly activating outwardly rectifying K⁺ current. The inwardly rectifying K⁺ current is inhibited by Ba²⁺. The slowly activating K⁺ current was potentiated by cAMP and inhibited by clofilium, phorbol 12-myristae 13-acetate (PMA) and lowering temperature. This is consistent with the biophysical characteristics of I _SK channel. RT-PCR analysis revealed the presence of I _SK cDNA in the rat trachea epithelia. Although 0.1 mm Ba²⁺ only had minimal affect on short-circuit current (I _sc) induced by cAMP in intact epithelia, 0.1 mm clofilium strongly inhibited it. These results indicate that I _SK might be important for maintaining cAMP-induced chloride secretion in the rat trachea epithelia. Received: 1 March 1996/Revised: 5 August 1996     

Single K+ Channels in Embryonic Leech Ganglion Cells

We investigated the properties of single K⁺ channels in the soma membrane of embryonic leech ganglion cells using the patch-clamp technique. We compared these K⁺ channels with the K⁺ channels found previously in Retzius neurons of the adult leech. In ganglion cells of 9- to 15-day-old embryos we characterized eight different types of K⁺ channels with mean conductances of 21, 55, 84, 111, 122, 132, 149 and 223 pS. The 55 pS and 84 pS channels showed flickering and were active for less than 2 min after excising the patch. The 111 pS channel was an outward rectifier, and the open state probability (p _o ) decreased in the inside-out configuration when the Ca²⁺ concentration was raised from pCa 7 to pCa 3. The 122 pS channel also showed outward rectification. This type of channel was activated after changing from the cell-attached to the inside-out configuration and it did not inactivate during more than 30 min. The p _o was Ca²⁺- and voltage-insensitive. One hundred μm glibenclamide reversibly reduced p _o . The 132 pS channel was an outward rectifier and was Ca²⁺-insensitive. The 149 pS channel inactivated in the inside-out configuration. The 149- and the 223 pS channel showed inward rectification. The 111 pS channel had similar properties to the Ca²⁺-dependent K⁺ channel and the 122 pS channel resembled the ATP-inhibited K⁺ channel found previously in Retzius neurons of the adult leech. Received: 20 April 1995/Revised: 18 January 1996     

Flow-Dependent Activation of Maxi K+ Channels in Apical Membrane of Rabbit Connecting Tubule

The Ca²⁺-activated maxi K⁺ channel was found in the apical membrane of everted rabbit connecting tubule (CNT) with a patch-clamp technique. The mean number of open channels (NP _o ) was markedly increased from 0.007 ± 0.004 to 0.189 ± 0.039 (n= 7) by stretching the patch membrane in a cell-attached configuration. This activation was suggested to be coupled with the stretch-activation of Ca²⁺-permeable cation channels, because the maxi K⁺ channel was not stretch-activated in both the cell-attached configuration using Ca²⁺-free pipette and in the inside-out one in the presence of 10 mm EGTA in the cytoplasmic side. The maxi K⁺ channel was completely blocked by extracellular 1 μm charybdotoxin (CTX), but was not by cytoplasmic 33 μm arachidonic acid (AA). On the other hand, the low-conductance K⁺ channel, which was also found in the same membrane, was completely inhibited by 11 μm AA, but not by 1 μm CTX. The apical K⁺ conductance in the CNT was estimated by the deflection of transepithelial voltage (ΔV _t ) when luminal K⁺ concentration was increased from 5 to 15 mEq. When the tubule was perfused with hydraulic pressure of 0.5 KPa, the ΔV _t was only −0.7 ± 0.4 mV. However, an increase in luminal fluid flow by increasing perfusion pressure to 1.5 KPa markedly enhanced ΔV _t to −9.4 ± 0.9 mV. Luminal application of 1 μm CTX reduced the ΔV _t to −1.3 ± 0.6 mV significantly in 6 tubules, whereas no significant change of ΔV _t was recorded by applying 33 μm AA into the lumen of 5 tubules (ΔV _t =−7.2 ± 0.5 mV in control vs.ΔV _t =−6.7 ± 0.6 mV in AA). These results suggest that the Ca²⁺-activated maxi K⁺ channel is responsible for flow-dependent K⁺ secretion by coupling with the stretch-activated Ca²⁺-permeable cation channel in the rabbit CNT. Received: 21 August 1997/Revised: 20 March 1998     

The Ach-evoked Ca2+-activated K+ Current in Mouse Mandibular Secretory Cells. Single Channel Studies

Although acetylcholine (ACh) is able to activate voltage- and Ca²⁺-sensitive K⁺ (BK) channels in mouse mandibular secretory cells, our recent whole cell studies have suggested that these channels, like those in sheep parotid secretory cells, do not contribute appreciably to the conductance that carries the ACh-evoked whole cell K⁺ current. In the present study, we have used cell-attached patch clamp methods to identify and characterize the K⁺ channel type responsible for carrying the bulk of this current. When the cells were bathed in a NaCl-rich solution the predominant channel type activated by ACh (1 μmol/l or 50 nmol/l) had a conductance only of 40 pS; it was not blocked by TEA but it was sensitive to quinine and it conducted Rb⁺ to an appreciable extent. BK channels, which could be seen in some but not all patches from resting cells, also showed increased activity when ACh was added to the bath, but they were much less conspicuous during ACh stimulation than the 40-pS channels. When the cells were bathed in a KCl-rich rather than a NaCl-rich solution, a small-conductance K⁺ channel, sensitive to quinine but not to TEA, was still the most conspicuous channel to be activated by ACh although its conductance was reduced to 25 pS. Our studies confirm that the ACh-evoked whole-cell K⁺ current is not carried substantially by BK channels and show that it is carried by a small-conductance K⁺ channel with quite different properties. Received: 28 September 1995/Revised: 26 December 1995     

Small Inward Rectifying K+ Channels in Coleoptiles: Inhibition by External Ca2+ and Function in Cell Elongation

Plant growth requires a continuous supply of intracellular solutes in order to drive cell elongation. Ion fluxes through the plasma membrane provide a substantial portion of the required solutes. Here, patch clamp techniques have been used to investigate the electrical properties of the plasma membrane in protoplasts from the rapid growing tip of maize coleoptiles. Inward currents have been measured in the whole cell configuration from protoplasts of the outer epidermis and from the cortex. These currents are essentially mediated by K⁺ channels with a unitary conductance of about 12 pS. The activity of these channels was stimulated by negative membrane voltage and inhibited by extracellular Ca²⁺ and/or tetraethylammonium-CI (TEA). The kinetics of voltage- and Ca²⁺-gating of these channels have been determined experimentally in some detail (steady-state and relaxation kinetics). Various models have been tested for their ability to describe these experimental data in straightforward terms of mass action. As a first approach, the most appropriate model turned out to consist of an active state which can equilibrate with two inactive states via independent first order reactions: a fast inactivation/activation by Ca²⁺-binding and -release, respectively (rate constants >>10³ sec⁻¹) and a slower inactivation/activation by positive/negative voltage, respectively (voltage-dependent rate constants in the range of 10³ sec⁻¹). With 10 mm K⁺ and 1 mm Ca²⁺ in the external solution, intact coleoptile cells have a membrane voltage (V) of −105 ± 7 mV. At this V, the density and open probability of the inward-rectifying channels is sufficient to mediate K⁺ uptake required for cell elongation. Extracellular TEA or Ca²⁺, which inhibit the K⁺ inward conductance, also inhibit elongation of auxin-depleted coleoptile segments in acidic solution. The comparable effects of Ca²⁺ and TEA on both processes and the similar Ca²⁺ concentration required for half maximal inhibition of growth (4.3 mm Ca²⁺) and for conductance (1.2 mm Ca²⁺) suggest that K⁺ uptake through the inward rectifier provides essential amounts of solute for osmotic driven elongation of maize coleoptiles. Received: 6 June 1995/Revised: 12 September 1995     

Protection from Cell Death by mcl-1 Is Mediated by Membrane Hyperpolarization Induced By K+ Channel Activation

Mcl-1, a member of the Bcl-2 family, has been identified as an inhibitor of apoptosis induced by anticancer agents and radiation in myeloblastic leukemia cells. The molecular mechanism underlying this phenomenon, however, is not yet understood. In the present study, we report that hyperpolarization of the membrane potential is required for prevention of mcl-1 mediated cell death in murine myeloblastic FDC-P1 cells. In cells transfected with mcl-1, the membrane potential, measured by the whole-cell patch clamp, was hyperpolarized more than −30 mV compared with control cells. The membrane potential was repolarized by increased extracellular K⁺ concentration (56 mV per 10-fold change in K⁺ concentration). Using the cell-attached patch-clamp technique, K⁺ channel activity was 1.7 times higher in mcl-1 transfected cells (NP _o = 22.7 ± 3.3%) than control cells (NP _o = 13.2 ± 1.9%). Viabilities of control and mcl-1 transfected cells after treatment with the cytotoxin etoposide (20 μg/ml), were 37.9 ± 3.9% and 78.2 ± 2.0%, respectively. Suppression of K⁺ channel activity by 4-aminopyridine (4-AP) before etoposide treatment significantly reduced the viability of mcl-1 transfected cells to 49.0 ± 4.6%. These results indicate that as part of the prevention of cell death, mcl-1 causes a hyperpolarization of membrane potential through activation of K⁺ channel activity. Received: 30 March 1999/Revised: 20 July 1999     

Temperature Sensitivity of the Tonoplast Ca2+-Activated K+ Channel in Chara: The Influence of Reversing the Sign of Membrane Potential

A detailed temperature dependence study of a well-defined plant ion channel, the Ca²⁺-activated K⁺ channel of Chara corallina, was performed over the temperature range of their habitats, 5–36°C, at 1°C resolution. The temperature dependence of the channel unitary conductance at 50 mV shows discontinuities at 15 and 30°C. These temperatures limit the range within which ion diffusion is characterized by the lowest activation energy (E _a = 8.0 ± 1.6 kJ/mol) as compared to the regions below 15°C and above 30°C. Upon reversing membrane voltage polarity from 50 to −50 mV the pattern of temperature dependence switched from discontinuous to linear with E _a = 13.6 ± 0.5 kJ/mol. The temperature dependence of the effective number of open channels at 50 mV showed a decrease with increasing temperature, with a local minimum at 28°C. The mean open time exhibited a similar behavior. Changing the sign of membrane potential from 50 to −50 mV abolished the minima in both temperature dependencies. These data are discussed in the light of higher order phase transitions of the Characean membrane lipids and corresponding change in the lipid-protein interaction, and their modulation by transmembrane voltage. Received: 14 June 2000/Revised: 20 September 2000     

Na+ Sensitivity of ROMK1 K+ Channel: Role of the Na+/H+ Antiporter

To examine the extracellular Na⁺ sensitivity of a renal inwardly rectifying K⁺ channel, we performed electrophysiological experiments on Xenopus oocytes or a human kidney cell line, HEK293, in which we had expressed the cloned renal K⁺ channel, ROMK1 (Kir1.1). When extracellular Na⁺ was removed, the whole-cell ROMK1 currents were markedly suppressed in both the oocytes and HEK293 cells. Single-channel ROMK1 activities recorded in the cell-attached patch on the oocyte were not affected by removal of Na⁺ from the pipette solution. However, macro-patch ROMK1 currents recorded on the oocyte were significantly suppressed by Na⁺ removal from the bath solution. A blocker of Na⁺/H⁺ antiporters, amiloride, largely inhibited the Na⁺ removal-induced suppression of whole-cell ROMK1 currents in the oocytes. The pH-insensitive K80M mutant of ROMK1 was much less sensitive to Na⁺ removal. Na⁺ removal was found to induce a significant decrease in intracellular pH in the oocytes using H⁺-selective microelectrodes. Coexpression of ROMK1 with NHE3, which is a Na⁺/H⁺ antiporter isoform of the kidney apical membrane, conferred increased sensitivity of ROMK1 channels to extracellular Na⁺ in both the oocytes and HEK293 cells. Thus, it is concluded that the ROMK1 channel is regulated indirectly by extracellular Na⁺, and that the interaction between NHE transporter and ROMK1 channel appears to be involved in the mechanism of Na⁺ sensitivity of ROMK1 channel via regulating intracellular pH. Received: 13 April 1999/Revised: 15 July 1999     

K+ Channels and the Intracellular Calcium Signal in Human Melanoma Cell Proliferation

K⁺ channels, membrane voltage, and intracellular free Ca²⁺ are involved in regulating proliferation in a human melanoma cell line (SK MEL 28). Using patch-clamp techniques, we found an inwardly rectifying K⁺ channel and a calcium-activated K⁺ channel. The inwardly rectifying K⁺ channel was calcium independent, insensitive to charybdotoxin, and carried the major part of the whole-cell current. The K⁺ channel blockers quinidine, tetraethylammonium chloride and Ba²⁺ and elevated extracellular K⁺ caused a dose-dependent membrane depolarization. This depolarization was correlated to an inhibition of cell proliferation. Charybdotoxin affected neither membrane voltage nor proliferation. Basic fibroblast growth factor and fetal calf serum induced a transient peak in intracellular Ca²⁺ followed by a long-lasting Ca²⁺ influx. Depolarization by voltage clamp decreased and hyperpolarization increased intracellular Ca²⁺, illustrating a transmembrane flux of Ca²⁺ following its electrochemical gradient. We conclude that K⁺ channel blockers inhibit cell-cycle progression by membrane depolarization. This in turn reduces the driving force for the influx of Ca²⁺, a messenger in the mitogenic signal cascade of human melanoma cells. Received: 9 May 1995/Revised: 30 January 1996     

The Calculation of Intracellular Ion Concentrations and Membrane Potential from Cell-attached and Excised Patch Measurements. Cytosolic K+ Concentration and Membrane Potential in Vicia faba Guard Cells

Ion channel activity in cell-attached patch recordings shows channel behavior under more physiological conditions than whole-cell and excised patch measurements. Yet the analysis of cell-attached patch measurements is complicated by the fact that the system is ill defined with respect to the intracellular ion activities and the electrical potential actually experienced by the membrane patch. Therefore, of the several patch-clamp configurations, the information that is obtained from cell-attached patch measurements is the most ambiguous. The present study aims to achieve a better understanding of cell-attached patch measurements. Here we describe a method to calculate the intracellular ion concentration and membrane potential prevailing during cell-attached patch recording. The first step is an analysis of the importance of the input resistance of the intact cell on the cell-attached patch measurement. The second step, and actual calculation, is based on comparison of the single channel conductance and reversal potential in the cell-attached patch and excised patch configurations. The method is demonstrated with measurements of membrane potential and cytosolic K⁺ concentrations in Vicia faba guard cells. The approach described here provides an attractive alternative to the measurement of cytosolic ion concentrations with fluorescent probes or microelectrodes. Received: 3 April 1998/Revised: 6 August 1998     

Enhanced Closed-state Inactivation in a Mutant Shaker K+ Channel

Many mutations that shift the voltage dependence of activation in Shaker channels cause a parallel shift of inactivation. The I2 mutation (L382I in the Shaker B sequence) is an exception, causing a 45 mV activation shift with only a 9 mV shift of inactivation midpoint relative to the wildtype (WT) channel. We compare the behavior of WT and I2 Shaker 29-4 channels in macropatch recordings from Xenopus oocytes. The behavior of WT channels can be described by both simple and detailed kinetic models which assume that inactivation proceeds only from the open state. The behavior of I2 channels requires that they inactivate from closed states as well, a property characteristic of voltage-gated sodium channels. A detailed ``multiple-state inactivation' model is presented that describes both activation and inactivation of I2 channels. The results are consistent with the view that residue L382 is associated with the receptor for the inactivation particles in Shaker channels. Received: 16 December 1996/Revised: 5 February 1997     

Clustered Distribution of Calcium Sensitivities: An Indication of Hetero-tetrameric Gating Components in Ca2+-activated K+ Channels Reconstituted from Avian Nasal Gland Cells

High-conductance, Ca²⁺-activated K⁺ channels from the basolateral membrane of rabbit distal colon epithelial cells were reconstituted into planar phospholipid bilayers to examine the effect of Mg²⁺ on the single-channel properties. Mg²⁺ decreases channel current and conductance in a concentration-dependent manner from both the cytoplasmic and the extracellular side of the channel. In contrast to other K⁺ channels, Mg²⁺ does not cause rectification of current through colonic Ca²⁺-activated K⁺ channels. In addition, cytoplasmic Mg²⁺ decreases the reversal potential of the channel. The Mg²⁺-induced decrease in channel conductance is relieved by high K⁺ concentrations, indicating competitive interaction between K⁺ and Mg²⁺. The monovalent organic cation choline also decreases channel conductance and reversal potential, suggesting that the effect is unspecific. The inhibition of channel current by Mg²⁺ and choline most likely is a result of electrostatic screening of negative charges located superficially in the channel entrance. But in addition to charge, other properties appear to be necessary for channel inhibition, as Na⁺ and Ba²⁺ are no (or only weak) inhibitors. Mg²⁺ and possibly other cations may play a role in the regulation of current through these channels. Received: 25 August 1995/Revised: 16 November 1995     

Potassium Uptake Through the TOK1 K+ Channel in the Budding Yeast

The current through TOK1 (YKC1), the outward-rectifying K⁺ channel in Saccharomyces cerevisiae, was amplified by expressing TOK1 from a plasmid driven by a strong constitutive promoter. TOK1 so hyper-expressed could overcome the K⁺ auxotrophy of a mutant missing the two K⁺ transporters, TRK1 and TRK2. This trk1Δtrk2Δ double mutant hyperexpressing the TOK1 transgene had a higher internal K⁺ content than one expressing the empty plasmid. We examined protoplasts of these TOK1-hyperexpressing cells under a patch clamp. Besides the expected K⁺ outward current activating at membrane potential (V _m ) above the K⁺ equilibrium potential (E _K+ ), a small inward current was consistently observed when the V _m was slightly below E _K+ . The inward and the outward currents are similar in their activation rates, deactivation rates, ion specificities and Ba²⁺ inhibition, indicating that they flow through the same channel. Thus, the yeast outwardly rectifying K⁺ channel can take up K⁺ into yeast cells, at least under certain conditions. Received: 1 October 1998/Revised: 9 December 1998     

Cloning and Function of the Rat Colonic Epithelial K+ Channel KVLQT1

K_VLQT1 (KCNQ1) is a voltage-gated K⁺ channel essential for repolarization of the heart action potential that is defective in cardiac arrhythmia. The channel is inhibited by the chromanol 293B, a compound that blocks cAMP-dependent electrolyte secretion in rat and human colon, therefore suggesting expression of a similar type of K⁺ channel in the colonic epithelium. We now report cloning and expression of K_VLQT1 from rat colon. Overlapping clones identified by cDNA-library screening were combined to a full length cDNA that shares high sequence homology to K_VLQT1 cloned from other species. RT-PCR analysis of rat colonic musoca demonstrated expression of K_VLQT1 in crypt cells and surface epithelium. Expression of rK_VLQT1 in Xenopus oocytes induced a typical delayed activated K⁺ current, that was further activated by increase of intracellular cAMP but not Ca²⁺ and that was blocked by the chromanol 293B. The same compound blocked a basolateral cAMP-activated K⁺ conductance in the colonic mucosal epithelium and inhibited whole cell K⁺ currents in patch-clamp experiments on isolated colonic crypts. We conclude that K_VLQT1 is forming an important component of the basolateral cAMP-activated K⁺ conductance in the colonic epithelium and plays a crucial role in diseases like secretory diarrhea and cystic fibrosis. Received: 17 July 2000/Revised: 25 October 2000     

Halide and Alkyl Phenols Block Volume-Sensitive Chloride Channels in Human Glial Cells (U-138MG)

Swelling of cells in hypotonic media activates a volume-sensitive Cl channel with well-known characteristics, but its structure and its regulation are still largely undetermined. It also has many inhibitors and most of them are also blocking other types of Cl channels. The numerous inhibitors of Cl channels have apparently no structural relationship among them. The purpose of this study was to try to determine the most simple molecules that could block these channels and identify some common properties among inhibitors. From the 37 new molecules that were studied, it was found that simple halide phenols like trichloro and triiodophenols could block these channels in the micromolar range. Also alkyl phenols, like butylphenols, are very sensitive blockers, comparable to other well-known blockers. But acidic halide phenols or nitrophenols are poor blockers. Also neutral polyphenols are more sensitive than acidic polyphenols. All these results indicate that the common basis for blocking these Cl channels is a phenol with hydrophobic groups, like short alkyl chains or an additional phenyl ring, attached to some of its sites, preferably sites 3-4-5. These results identify a new family of Cl channel blockers and hopefully improve our understanding of the blocking mechanism. Received: 28 August 1997/Revised: 12 December 1997     

K+-Sensitive Gating of the K+ Outward Rectifier in Vicia Guard Cells

The effect of extracellular cation concentration and membrane voltage on the current carried by outward-rectifying K⁺ channels was examined in stomatal guard cells of Vicia faba L. Intact guard cells were impaled with double-barrelled microelectrodes and the K⁺ current was monitored under voltage clamp in 0.1–30 mm K⁺ and in equivalent concentrations of Rb⁺, Cs⁺ and Na⁺. From a conditioning voltage of −200 mV, clamp steps to voltages between −150 and +50 mV in 0.1 mm K⁺ activated current through outward-rectifying K⁺ channels (I _{K, out}) at the plasma membrane in a voltage-dependent fashion. Increasing [K⁺] _o shifted the voltage-sensitivity of I _{K, out} in parallel with the equilibrium potential for K⁺ across the membrane. A similar effect of [K⁺] _o was evident in the kinetics of I _{K, out} activation and deactivation, as well as the steady-state conductance- (g _K ⁻) voltage relations. Linear conductances, determined as a function of the conditioning voltage from instantaneous I-V curves, yielded voltages for half-maximal conductance near −130 mV in 0.1 mm K⁺, −80 mV in 1.0 mm K⁺, and −20 mV in 10 mm K⁺. Similar data were obtained with Rb⁺ and Cs⁺, but not with Na⁺, consistent with the relative efficacy of cation binding under equilibrium conditions (K⁺≥ Rb⁺ > Cs⁺ > > Na⁺). Changing Ca²⁺ or Mg²⁺ concentrations outside between 0.1 and 10 mm was without effect on the voltage-dependence of g _K or on I _{K, out} activation kinetics, although 10 mm [Ca²⁺] _o accelerated current deactivation at voltages negative of −75 mV. At any one voltage, increasing [K⁺] _o suppressed g _K completely, an action that showed significant cooperativity with a Hill coefficient of 2. The apparent affinity for K⁺ was sensitive to voltage, varying from 0.5 to 20 mm with clamp voltages near −100 to 0 mV, respectively. These, and additional data indicate that extracellular K⁺ acts as a ligand and alters the voltage-dependence of I _{K, out} gating; the results implicate K⁺-binding sites accessible from the external surface of the membrane, deep within the electrical field, but distinct from the channel pore; and they are consistent with a serial 4-state reaction-kinetic model for channel gating in which binding of two K⁺ ions outside affects the distribution between closed states of the channel. Received: 27 November 1996/Revised: 4 March 1997