Induction of a Ca(2+)-activated K+ channel in human erythrocytes by mechanical stress.

Mechanical deformation of normal ATP-replete human erythrocytes increased their permeability to Ca2+ sufficiently to turn on the Ca(2+)-activated K+ channel (the Gardos channel). When Ca2+ is absent, mechanical deformation of normal erythrocytes induces an equivalent increase the permeability of both Na+ and K+, In the presence of 0.1 to 1 mM Ca2+, a further increase in the K+ efflux rate was seen. There was no increase in Na+ flux above that induced by deformation itself. The involvement of the Ca(2+)-activated H channel was verified by showing the specific inhibitors of the channel, quinine and charybdotoxin, prevent the Ca(2+)-induced increase in K+ efflux. These results are consistent with a model of sickle cell dehydration proposed by Bookchin et al. ((1987) Prog. Clin. Biol. Res. 240, 193-200). The estimated rate of Ca2+ entry under these conditions (37 degrees C, 1000 dyne/cm2, and laminar shear) was about 1 mmol/loc per h.     

Functional effects of auxiliary beta4-subunit on rat large-conductance Ca(2+)-activated K(+) channel

Large-conductance calcium-activated potassium (BK(Ca)) channels are composed of the pore-forming alpha-subunit and the auxiliary beta-subunits. The beta4-subunit is dominantly expressed in the mammalian central nervous system. To understand the physiological roles of the beta4-subunit on the BK(Ca) channel alpha-subunit (Slo), we isolated a full-length complementary DNA of rat beta4-subunit (rbeta4), expressed heterolgously in Xenopus oocytes, and investigated the detailed functional effects using electrophysiological means. When expressed together with rat Slo (rSlo), rbeta4 profoundly altered the gating characteristics of the Slo channel. At a given concentration of intracellular Ca(2+), rSlo/rbeta4 channels were more sensitive to transmembrane voltage changes. The activation and deactivation rates of macroscopic currents were decreased in a Ca(2+)-dependent manner. The channel activation by Ca(2+) became more cooperative by the coexpression of rbeta4. Single-channel recordings showed that the increased Hill coefficient for Ca(2+) was due to the changes in the open probability of the rSlo/rbeta4 channel. Single BK(Ca) channels composed of rSlo and rbeta4 also exhibited slower kinetics for steady-state gating compared with rSlo channels. Dwell times of both open and closed events were significantly increased. Because BK(Ca) channels are known to modulate neuroexcitability and the expression of the beta4-subunit is highly concentrated in certain subregions of brain, the electrophysiological properties of individual neurons should be affected profoundly by the expression of this second subunit.     

Competition for block of a Ca2(+)-activated K+ channel by charybdotoxin and tetraethylammonium

Single high-conductance Ca2(+)-activated K+ channels were incorporated into planar lipid bilayers, and the discrete block by charybdotoxin (CTX), a protein inhibitor of this channel, was studied. In particular, the effect of externally added tetraethylammonium (TEA) on CTX blocking kinetics was investigated. TEA decreases the on-rate of CTX in exact proportion to its blocking of the single-channel current. The CTX off-rate is unaffected by TEA. The results demonstrate that TEA and CTX are mutually exclusive in their binding to the channel. Since the site of TEA binding is known to be located on the external side of the conduction pore, this result further strengthens the proposal that the CTX binding site is located in the external mouth of the channel.     

Nitrendipine is a potent inhibitor of the Ca(2+)-activated K+ channel of human erythrocytes.

Nitrendipine, a classical blocker of L-type Ca2+ channels, is shown to be a potent inhibitor of the Ca(2+)-activated K+ channel of human erythrocytes. In erythrocytes suspended in a solution with physiological Na+ and K+ concentrations and in which the channel was activated using the Ca2+ ionophore ionomycin, nitrendipine inhibited K+(86Rb+) influx with an I50 of around 130 nM. Similar results were obtained for K+(86Rb+) efflux, and for K+(86Rb+) influx into cells suspended in a high-K+ medium.     

Activation of the human intermediate-conductance Ca(2+)-activated K(+) channel by 1-ethyl-2-benzimidazolinone is strongly Ca(2+)-dependent.

Modulation of the cloned human intermediate-conductance Ca(2+)-activated K(+) channel (hIK) by the compound 1-ethyl-2-benzimidazolinone (EBIO) was studied by patch-clamp technique using human embryonic kidney cells (HEK 293) stably expressing the hIK channels. In whole-cell studies, intracellular concentrations of free Ca(2+) were systematically varied, by buffering the pipette solutions. In voltage-clamp, the hIK specific currents increased gradually from 0 to approximately 300 pA/pF without reaching saturation even at the highest Ca(2+) concentration tested (300 nM). In the presence of EBIO (100 microM), the Ca(2+)-activation curve was shifted leftwards, and maximal currents were attained at 100 nM Ca(2+). In current-clamp, steeply Ca(2+)-dependent membrane potentials were recorded and the cells gradually hyperpolarised from -20 to -85 mV when Ca(2+) was augmented from 0 to 300 nM. EBIO strongly hyperpolarised cells buffered at intermediate Ca(2+) concentrations. In contrast, no effects were detected either below 10 nM (no basic channel activation) or at 300 nM Ca(2+) (V(m) close to E(K)). Without Ca(2+), EBIO-induced hyperpolarisations were not obtainable, indicating an obligatory Ca(2+)-dependent mechanism of action. When applied to inside-out patches, EBIO exerted a Ca(2+)-dependent increase in the single-channel open-state probability, showing that the compound modulates hIK channels by a direct action on the alpha-subunit or on a closely associated protein. In conclusion, EBIO activates hIK channels in whole-cell and inside-out patches by a direct mechanism, which requires the presence of internal Ca(2+).     

ATP inhibits smooth muscle Ca2(+)-activated K+ channels

There has been much recent interest in the roles played by smooth-muscle K+ channels in protecting cells against ischemic and anoxic insults and in therapeutic vaso- and bronchodilation (Buckingham 1990; Longmore & Weston 1990). A K+ channel, which is uniquely sensitive to cytoplasmic ATP (KATP), has been identified as a likely candidate for mediating these important functions (Standen et al. 1989). We now show, by using electrophysiological techniques in three different types of smooth muscle, that a large-conductance voltage and Ca2(+)-sensitive channel, otherwise indistinguishable from the the large-conductance Ca2(+)-activated K+ channel (BK channel), is also sensitive to cytoplasmic ATP and cromakalim. ATP, in a dose-dependent manner, decreased the probability of channel opening (Po) of rabbit aortic, rabbit tracheal and pig coronary artery BK channels with a Ki of 0.2-0.6 mM. Cromakalim, 10 microM, partially reversed the ATP induced inhibition and increased Po. Our observations raise the possibility that the ubiquitous BK channel may play a role during pathophysiological events.     

Functional coupling of the beta(1) subunit to the large conductance Ca(2+)-activated K(+) channel in the absence of Ca(2+). Increased Ca(2+) sensitivity from a Ca(2+)-independent mechanism

Coexpression of the beta(1) subunit with the alpha subunit (mSlo) of BK channels increases the apparent Ca(2+) sensitivity of the channel. This study investigates whether the mechanism underlying the increased Ca(2+) sensitivity requires Ca(2+), by comparing the gating in 0 Ca(2+)(i) of BK channels composed of alpha subunits to those composed of alpha+beta(1) subunits. The beta(1) subunit increased burst duration approximately 20-fold and the duration of gaps between bursts approximately 3-fold, giving an approximately 10-fold increase in open probability (P(o)) in 0 Ca(2+)(i). The effect of the beta(1) subunit on increasing burst duration was little changed over a wide range of P(o) achieved by varying either Ca(2+)(i) or depolarization. The effect of the beta(1) subunit on increasing the durations of the gaps between bursts in 0 Ca(2+)(i) was preserved over a range of voltage, but was switched off as Ca(2+)(i) was increased into the activation range. The Ca(2+)-independent, beta(1) subunit-induced increase in burst duration accounted for 80% of the leftward shift in the P(o) vs. Ca(2+)(i) curve that reflects the increased Ca(2+) sensitivity induced by the beta(1) subunit. The Ca(2+)-dependent effect of the beta(1) subunit on the gaps between bursts accounted for the remaining 20% of the leftward shift. Our observation that the major effects of the beta(1) subunit are independent of Ca(2+)(i) suggests that the beta(1) subunit mainly alters the energy barriers of Ca(2+)-independent transitions. The changes in gating induced by the beta(1) subunit differ from those induced by depolarization, as increasing P(o) by depolarization or by the beta(1) subunit gave different gating kinetics. The complex gating kinetics for both alpha and alpha+beta(1) channels in 0 Ca(2+)(i) arise from transitions among two to three open and three to five closed states and are inconsistent with Monod-Wyman-Changeux type models, which predict gating among only one open and one closed state in 0 Ca(2+)(i).     

Ca(2+)-activated K+ channel inhibition by reactive oxygen species

We studied the effect of H(2)O(2) on the gating behavior of large-conductance Ca(2+)-sensitive voltage-dependent K(+) (K(V,Ca)) channels. We recorded potassium currents from single skeletal muscle channels incorporated into bilayers or using macropatches of Xenopus laevis oocytes membranes expressing the human Slowpoke (hSlo) alpha-subunit. Exposure of the intracellular side of K(V,Ca) channels to H(2)O(2) (4-23 mM) leads to a time-dependent decrease of the open probability (P(o)) without affecting the unitary conductance. H(2)O(2) did not affect channel activity when added to the extracellular side. These results provide evidence for an intracellular site(s) of H(2)O(2) action. Desferrioxamine (60 microM) and cysteine (1 mM) completely inhibited the effect of H(2)O(2), indicating that the decrease in P(o) was mediated by hydroxyl radicals. The reducing agent dithiothreitol (DTT) could not fully reverse the effect of H(2)O(2). However, DTT did completely reverse the decrease in P(o) induced by the oxidizing agent 5,5'-dithio-bis-(2-nitrobenzoic acid). The incomplete recovery of K(V,Ca) channel activity promoted by DTT suggests that H(2)O(2) treatment must be modifying other amino acid residues, e.g., as methionine or tryptophan, besides cysteine. Noise analysis of macroscopic currents in Xenopus oocytes expressing hSlo channels showed that H(2)O(2) induced a decrease in current mediated by a decrease both in the number of active channels and P(o).     

Role of the beta1 subunit in large-conductance Ca(2+)-activated K(+) channel gating energetics. Mechanisms of enhanced Ca(2+) sensitivity

Over the past few years, it has become clear that an important mechanism by which large-conductance Ca²⁺-activated K⁺ channel (BK_Ca) activity is regulated is the tissue-specific expression of auxiliary β subunits. The first of these to be identified, β1, is expressed predominately in smooth muscle and causes dramatic effects, increasing the apparent affinity of the channel for Ca²⁺ 10-fold at 0 mV, and shifting the range of voltages over which the channel activates −80 mV at 9.1 μM Ca²⁺. With this study, we address the question: which aspects of BK_Ca gating are altered by β1 to bring about these effects: Ca²⁺ binding, voltage sensing, or the intrinsic energetics of channel opening? The approach we have taken is to express the β1 subunit together with the BK_Ca α subunit in Xenopus oocytes, and then to compare β1''s steady state effects over a wide range of Ca²⁺ concentrations and membrane voltages to those predicted by allosteric models whose parameters have been altered to mimic changes in the aspects of gating listed above. The results of our analysis suggest that much of β1''s steady state effects can be accounted for by a reduction in the intrinsic energy the channel must overcome to open and a decrease in its voltage sensitivity, with little change in the affinity of the channel for Ca²⁺ when it is either open or closed. Interestingly, however, the small changes in Ca²⁺ binding affinity suggested by our analysis (K_c 7.4 μM → 9.6 μM; K_o = 0.80 μM → 0.65 μM) do appear to be functionally important. We also show that β1 affects the mSlo conductance–voltage relation in the essential absence of Ca²⁺, shifting it +20 mV and reducing its apparent gating charge 38%, and we develop methods for distinguishing between alterations in Ca²⁺ binding and other aspects of BK_Ca channel gating that may be of general use.     

Ca(2+) influx and opening of Ca(2+)-activated K(+) channels in muscle fibers from control and mdx mice

Using the patch-clamp technique, we demonstrate that, in depolarized cell-attached patches from mouse skeletal muscle fibers, a short hyperpolarization to resting value is followed by a transient activation of Ca(2+)-activated K(+) channels (K(Ca)) upon return to depolarized levels. These results indicate that sparse sites of passive Ca(2+) influx at resting potentials are responsible for a subsarcolemmal Ca(2+) load high enough to induce K(Ca) channel activation upon muscle activation. We then investigate this phenomenon in mdx dystrophin-deficient muscle fibers, in which an elevated Ca(2+) influx and a subsequent subsarcolemmal Ca(2+) overload are suspected. The number of Ca(2+) entry sites detected with K(Ca) was found to be greater in mdx muscle. K(Ca) activity reflecting subsarcolemmal Ca(2+) load was also found to be independent of the activity of leak channels carrying inward currents at negative potentials in mdx muscle. These results indicate that the sites of passive Ca(2+) influx newly described in this study could represent the Ca(2+) influx pathways responsible for the subsarcolemmal Ca(2+) overload in mdx muscle fibers.     

Intracellular Mg(2+) enhances the function of BK-type Ca(2+)-activated K(+) channels.

BK channels modulate neurotransmitter release due to their activation by voltage and Ca(2+). Intracellular Mg(2+) also modulates BK channels in multiple ways with opposite effects on channel function. Previous single-channel studies have shown that Mg(2+) blocks the pore of BK channels in a voltage-dependent manner. We have confirmed this result by studying macroscopic currents of the mslo1 channel. We find that Mg(2+) activates mslo1 BK channels independently of Ca(2+) and voltage by preferentially binding to their open conformation. The mslo3 channel, which lacks Ca(2+) binding sites in the tail, is not activated by Mg(2+). However, coexpression of the mslo1 core and mslo3 tail produces channels with Mg(2+) sensitivity similar to mslo1 channels, indicating that Mg(2+) sites differ from Ca(2+) sites. We discovered that Mg(2+) also binds to Ca(2+) sites and competitively inhibits Ca(2+)-dependent activation. Quantitative computation of these effects reveals that the overall effect of Mg(2+) under physiological conditions is to enhance BK channel function.     

The mitochondrial Ca(2+)-activated K(+) channel contributes to cardioprotection by limb remote ischemic preconditioning in rat

AimsTo investigate the role of the mitochondrial Ca²⁺-activated K⁺ channel in cardioprotection induced by limb remote ischemic preconditioning.Main methodsMale Sprague–Dawley rats (250–300 g) were randomized into control, ischemia/reperfusion (I/R), remote ischemic preconditioning (RPC), NS1619 (a specific mitochondrial Ca²⁺-activated K⁺ channel opener), and RPC + paxilline (a specific mitochondrial Ca²⁺-activated K⁺ channel inhibitor) groups. RPC was induced by 4 cycles of 5 min of ligation followed by 5 min of reperfusion of the left femoral artery. Myocardial I/R was achieved by ligation of the left anterior descending coronary artery for 30 min, followed by 120 min of reperfusion. Infarct size was determined by 2,3,5-triphenyltetrazolium chloride staining, the hemodynamics were monitored, and lactate dehydrogenase (LDH) levels in the coronary effluent, manganese superoxide dismutase (Mn-SOD) content in mitochondria and mitochondrial membrane potential were measured spectrophotometrically. The ultrastructure of cardiomyocyte mitochondria was assessed by electron microscopy.Key findingsNS1619 (10 μM) improved heart function, decreased infarct size, reduced LDH release, maintained mitochondrial structural integrity and mitochondrial membrane potential, and increased the mitochondrial content of Mn-SOD to the same degree as RPC treatment. However, paxilline (1 μM) eliminated the cardioprotective effect conferred by RPC.SignificanceThe mitochondrial Ca²⁺-activated K⁺ channel participates in the myocardial protection by limb remote ischemic preconditioning.     

Functional triads consisting of ryanodine receptors, Ca(2+) channels, and Ca(2+)-activated K(+) channels in bullfrog sympathetic neurons. Plastic modulation of action potential

Fluorescent ryanodine revealed the distribution of ryanodine receptors in the submembrane cytoplasm (less than a few micrometers) of cultured bullfrog sympathetic ganglion cells. Rises in cytosolic Ca(2+) ([Ca(2+)](i)) elicited by single or repetitive action potentials (APs) propagated at a high speed (150 microm/s) in constant amplitude and rate of rise in the cytoplasm bearing ryanodine receptors, and then in the slower, waning manner in the deeper region. Ryanodine (10 microM), a ryanodine receptor blocker (and/or a half opener), or thapsigargin (1-2 microM), a Ca(2+)-pump blocker, or omega-conotoxin GVIA (omega-CgTx, 1 microM), a N-type Ca(2+) channel blocker, blocked the fast propagation, but did not affect the slower spread. Ca(2+) entry thus triggered the regenerative activation of Ca(2+)-induced Ca(2+) release (CICR) in the submembrane region, followed by buffered Ca(2+) diffusion in the deeper cytoplasm. Computer simulation assuming Ca(2+) release in the submembrane region reproduced the Ca(2+) dynamics. Ryanodine or thapsigargin decreased the rate of spike repolarization of an AP to 80%, but not in the presence of iberiotoxin (IbTx, 100 nM), a BK-type Ca(2+)-activated K(+) channel blocker, or omega-CgTx, both of which decreased the rate to 50%. The spike repolarization rate and the amplitude of a single AP-induced rise in [Ca(2+)](i) gradually decreased to a plateau during repetition of APs at 50 Hz, but reduced less in the presence of ryanodine or thapsigargin. The amplitude of each of the [Ca(2+)](i) rise correlated well with the reduction in the IbTx-sensitive component of spike repolarization. The apamin-sensitive SK-type Ca(2+)-activated K(+) current, underlying the afterhyperpolarization of APs, increased during repetitive APs, decayed faster than the accompanying rise in [Ca(2+)](i), and was suppressed by CICR blockers. Thus, ryanodine receptors form a functional triad with N-type Ca(2+) channels and BK channels, and a loose coupling with SK channels in bullfrog sympathetic neurons, plastically modulating AP.     

Bimodal control of a Ca(2+)-activated Cl(-) channel by different Ca(2+) signals

Ca(2+)-activated Cl(-) channels play important roles in a variety of physiological processes, including epithelial secretion, maintenance of smooth muscle tone, and repolarization of the cardiac action potential. It remains unclear, however, exactly how these channels are controlled by Ca(2+) and voltage. Excised inside-out patches containing many Ca(2+)-activated Cl(-) channels from Xenopus oocytes were used to study channel regulation. The currents were mediated by a single type of Cl(-) channel that exhibited an anionic selectivity of I(-) > Br(-) > Cl(-) (3.6:1.9:1.0), irrespective of the direction of the current flow or [Ca(2+)]. However, depending on the amplitude of the Ca(2+) signal, this channel exhibited qualitatively different behaviors. At [Ca(2+)] < 1 microM, the currents activated slowly upon depolarization and deactivated upon hyperpolarization and the steady state current-voltage relationship was strongly outwardly rectifying. At higher [Ca(2+)], the currents did not rectify and were time independent. This difference in behavior at different [Ca(2+)] was explained by an apparent voltage-dependent Ca(2+) sensitivity of the channel. At +120 mV, the EC(50) for channel activation by Ca(2+) was approximately fourfold less than at -120 mV (0.9 vs. 4 microM). Thus, at [Ca(2+)] < 1 microM, inward current was smaller than outward current and the currents were time dependent as a consequence of voltage-dependent changes in Ca(2+) binding. The voltage-dependent Ca(2+) sensitivity was explained by a kinetic gating scheme in which channel activation was Ca(2+) dependent and channel closing was voltage sensitive. This scheme was supported by the observation that deactivation time constants of currents produced by rapid Ca(2+) concentration jumps were voltage sensitive, but that the activation time constants were Ca(2+) sensitive. The deactivation time constants increased linearly with the log of membrane potential. The qualitatively different behaviors of this channel in response to different Ca(2+) concentrations adds a new dimension to Ca(2+) signaling: the same channel can mediate either excitatory or inhibitory responses, depending on the amplitude of the cellular Ca(2+) signal.     

Endothelial K(+) channel lacks the Ca(2+) sensitivity-regulating beta subunit.

Hyperpolarizing large-conductance, Ca(2+)-activated K(+) channels (BK) are important modulators of vascular smooth muscle and endothelial cell function. In vascular smooth muscle cells, BK are composed of pore-forming alpha subunits and modulatory beta subunits. However, expression, composition, and function of BK subunits in endothelium have not been studied so far. In patch-clamp experiments we identified BK (283 pS) in intact endothelium of porcine aortic tissue slices. The BK opener DHS-I (0.05-0.3 micromol/l), stimulating BK activity only in the presence of beta subunits, had no effect on BK in endothelium whereas the alpha subunit selective BK opener NS1619 (20 micromol/l) markedly increased channel activity. Correspondingly, mRNA expression of the beta subunit was undetectable in endothelium, whereas alpha subunit expression was demonstrated. To investigate the functional role of beta subunits, we transfected the beta subunit into a human endothelial cell line (EA.hy 926). beta subunit expression resulted in an increased Ca(2+) sensitivity of BK activity: the potential of half-maximal activation (V(1/2)) shifted from 73.4 mV to 49.6 mV at 1 micromol/l [Ca(2+)](i) and an decrease of the EC(50) value for [Ca(2+)](i) by 1 microM at +60 mV was observed. This study demonstrates that BK channels in endothelium are composed of alpha subunits without association to beta subunits. The lack of the beta subunit indicates a substantially different channel regulation in endothelial cells compared to vascular smooth muscle cells.     

Proton modulation of a Ca(2+)-activated K+ channel from rat skeletal muscle incorporated into planar bilayers

The effect of pH on the activation of a Ca-activated K+ [K(Ca)] channel from rat skeletal muscle incorporated into planar lipid bilayers was studied. Experiments were done at different intracellular Ca2+ and proton concentrations. Changes in pH modified channel kinetics only from the Ca-sensitive face of the channel. At constant Ca2+ concentration, intracellular acidification induced a decrease in the open probability (Po) and a shift of the channel activation curves toward the right along the voltage axis. The displacement was 23.5 mV per pH unit. This displacement was due to a change in the half saturation voltage (Vo) and not to a change in channel voltage dependence. The shifts in Vo induced by protons appeared to be independent of Ca2+ concentration. The slope of the Hill plot of the open-closed equilibrium vs. pH was close to one, suggesting that a minimum of one proton is involved in the proton-driven channel closing reaction. The change in Po with variations in pH was due to both a decrease in the mean open time (To) and an increase in the mean closed time (Tc). At constant voltage, the mean open time of the channel was a linear function of [Ca2+] and the mean closed time was a linear function of 1/[Ca2+]2. Changes in the internal pH modified the slope, but not the intercept of the linear relations To vs. [Ca2+] and Tc vs. 1/[Ca2+]2. On the basis of these results an economical kinetic model of the effect of pH on this channel is proposed. It is concluded that protons do not affect the open-closed reaction, but rather weaken Ca2+ binding to all the conformational states of the channel. Moreover, competitive models in which Ca2+ and H+ cannot bind to the same open or closed state are inconsistent with the data.     

Subconductance behavior in a maxi Ca2(+)-activated K+ channel induced by dendrotoxin-I

Toxin I (DTX-I), a 60-residue peptide belonging to the dendrotoxin family of Mamba snake neurotoxins, is a potent inhibitor of various types of voltage-gated K+ currents. To investigate the sensitivity of another major class of K+ channels to DTX-I, the effect of this toxin was studied on single Ca2(+)-activated K+ channels from rat skeletal muscle incorporated into planar bilayers. Internal (intracellular) DTX-I was found to induce reversibly a long-lived (tau = 40 s), inwardly rectifying subconductance state with 66% of the normal open-state current at +20 mV. Analysis of the kinetics of substate formation and the current-voltage behavior of the substate suggest that binding of DTX-I modifies conduction of K+ ions through the pore without affecting the Ca2+ dependence or voltage dependence of gating. These results identify a unique internal binding site for DTX-I (Kd = 90 nM in 50 mM KCl) on a ubiquitous class of high-conductance, Ca2(+)-activated K+ channels.     

Activation of red cell Ca2(+)-activated K+ channel by Ca2+ involves a temperature-dependent step

We found that vanadate-induced 45Ca2+ uptake by red cells is maximal at 25 degrees C. At this temperature, the Cai-induced increase of the K+ permeability (the Gárdos effect) shows a lag (up to 8 min) which is not observed at 37 degrees C. This cannot be explained by the lack of availability of Ca2+ for the Ca2(+)-activated K+ channel, and suggests that its activation by Ca2+ is mediated by a temperature-dependent mechanism which remains unknown so far. The lag is not observed when the Gárdos effect was initiated by propranolol. This shows that the putative temperature-dependent step is different from chloride transport.