The Ca-activated K channel of human red cells: all or none behaviour of the Ca2+-gating mechanism

Osmotic Vesicle Collapse of Sealed Inside-Out Membrane Vesicles From Red Blood Cells

The preparation of plasma membrane vesicles from a large variety of cells has contributed a wealth of information on the identity and vectorial properties of membrane transporters and enzymes. Vesicles from red blood cell (RBC) membranes are generated in media of extremely low tonicity. For functional studies, it is required to suspend the vesicles in higher tonicity media in order to bring the concentrations of the substrates of transporters and enzymes under investigation within the physiological ranges. We investigated the effects of hypertonic transitions on the vesicle morphology using transmission electron microscopy. The results show that hypertonic transitions cause an irreversible osmotic collapse of sealed membrane vesicles. Awareness of the collapsed condition of vesicles during functional studies is critical for the proper interpretation of experimental results.

Highly co-operative Ca2+ activation of intermediate-conductance K+ channels in granulocytes from a human cell line

1. To study Ca(2+)-activated K+ currents in dimethyl sulphoxide (DMSO)-differentiated HL-60 cells (HL-60 granulocytes), we have combined the patch clamp technique with microfluorimetric measurements of the cytosolic free Ca2+ concentration ([Ca2+]i). 2. Elevations of [Ca2+]i induced by the receptor agonist N-formyl-L-methionyl-L-phenylalanine (f-MLP), by cellular spreading or by the Ca2+ ionophore ionomycin, activated whole-cell currents. The kinetics of the current elevations closely paralleled the kinetics of the elevations in [Ca2+]i. Cellular spreading induced oscillations in [Ca2+]i and parallel oscillatory changes in the amplitude of the recorded currents. 3. The reversal potential of the Ca(2+)-activated current was a function of the extracellular K+ concentration (56.1 mV per log [K+]), demonstrating that the underlying conductance was selective for K+. 4. The current was blocked by charybdotoxin, but insensitive to apamin. 5. The whole-cell current was inwardly rectifying. No time-dependent activation or inactivation of the current could be observed within the range of voltages tested (-100 to +100 mV). 6. The dependence of the current amplitude on the measured [Ca2+]i revealed a half-maximal activation at approximately 350 nM [Ca2+]i, and a highly co-operative activation by [Ca2+]i with an apparent Hill coefficient of approximately 8. Neither the half-maximal activation by [Ca2+]i nor the apparent Hill coefficient depended on the voltage, and they were identical for Ca2+ elevations caused by the ionophore and the receptor agonist. 7. Analysis of Ca(2+)-activated single-channel events in cell-attached recordings revealed an inwardly rectifying K+ channel with a slope conductance of 35 pS. Fluctuation analysis of the Ca(2+)-activated whole-cell current suggested an underlying single-channel conductance of a similar size (28 pS). 8. In summary, we describe a charybdotoxin-sensitive, intermediate-conductance Ca(2+)-activated K+ channel in HL-60 granulocytes. The characteristics of the Ca2+ activation of this current (i.e. sensitivity to submicromolar [Ca2+]i, high co-operativity and voltage independence) are similar to the Ca2+ activation of the apamin-sensitive small-conductance K+ channel. Our results also suggest that [Ca2+]i elevations are the predominant, if not the only, activators of this channel during physiological stimulation of HL-60 granulocytes.

All or none cell responses of Ca2+-dependent K channels elicited by calcium or lead in human red cells can be explained by heterogeneity of agonist distribution

We have studied the all or none cell response of Ca2+-dependent K+ channels to added Ca in human red cells depleted of ATP by incubation with iodoacetate and inosine. A procedure was used which allows separation and differential analysis of responding and nonresponding cells. Responding (H for heavy) cells incubated in medium containing 5 mM K lose KCl and water and increase their density to the point of sinking on diethylphthalate (specific gravity = 1.12) on centrifugation. Nonresponding (L for light) cells do not lose KCl at all. There is no intermediate behavior. Increasing the Ca concentration in the medium increases the fraction of cells which become H. No differences in the sensitivity to Ca2+ of the individual K+ channels were detected in inside-out vesicles prepared either from H or from L cells. The Ca content of H cells was higher than that of L cells. Cells depleted of ATP by incubation with iodoacetate and inosine sustain pump-leak Ca fluxes of about 15 mumol/liter cells per hour. ATP seems to be resynthesized in these cells at the expense of cell 2,3-diphosphoglycerate stores at a rate of about 150 mumol/liter cells per hour. Inhibition of 2,3-diphosphoglycerate phosphatase by tetrathionate increased 6-8 times the measured rate of uptake of external 45Ca. This was accompanied by an increase in the fraction of H cells. All or none cell responses of Ca2+-dependent K channels have also been evidenced in intact human red cells on addition of Pb. They have the same characteristics as those in responding and nonresponding cells. The detailed study of the kinetics of Pb-induced shrinkage of red cells suspended in medium containing 5 mM K showed that changes of Pb concentration changed not only the fraction of H cells but also the rate of shrinkage of responding cells. H cells generated by Pb treatment contained significantly more lead than L cells. The above results suggest that the two all or none cell responses studied here can be explained by heterogeneity of agonist distribution among cells. Since pump-leak fluxes exist in both cases, differences of agonist distribution could be generated by heterogeneity of pumping among cells. This interpretation turns interest from K channels to Ca pumps to explain the heterogeneous behavior of red cells in response to a uniform stimulus.

Magnitude of calcium influx required to induce dehydration of normal human red cells

Activation by [Ca2+]i of Ca2+-sensitive K+ channels has long been known to cause dehydration of red cells suspended in low-K, plasma-like media. However, the fundamental question of the extent to which Ca influx must be increased to trigger dense cell formation in conditions likely to arise in the circulation has not been established. We report here that in ionophore permeabilized red cells, increasing Ca influx above 0.7 mmol/litre cells per h induces the formation of subpopulations of dehydrated cells within 1-2 hours. The presence or absence of glycolytic substrates had little effect suggesting that ATP depletion was not large enough to significantly inhibit the pump within that period. Below maximal dehydrating Ca influxes of about 1.2 mmol/litre cells per h, the trend was for the fraction of dense cells formed to remain steady in time. As Ca influx was increased, both the rate of dense cell formation and the fraction of dense cells formed increased. These results are analyzed in relation to mechanisms and to possible states of increased Ca2+ permeability in physiological and physiopathological conditions.

Analysis of the all or nothing behaviour of Ca-dependent K channels in one-step inside-out vesicles from human red cell membranes

The all or nothing behaviour of Ca2+-dependent K+ channels has been analyzed in one-step inside-out vesicles. There is a threshold for Ca2+ below which the K+ channels remain silent, and which ranges between the 10(-6) and 10(-8) M for different vesicles under the experimental conditions tested, in the absence of Mg2+. The increase of Ca2+ concentration within this range recruits a larger fraction of the vesicles to the active (permeable to 86Rb+) state. The apparent rate of 86Rb+ transport through each individual channel was found to increase, however, with Ca2+ concentration. This finding is not an artefact due to size heterogeneity of the vesicle population, and it is consistent with the variations of the mean open time of the channels with Ca2+ concentration reported previously in patch-clamp experiments. The electron donor system ascorbate + phenazine-methosulphate increases the rate of 86Rb+ transport through the channels whereas oxidized cytochrome c has the opposite effect.

Inhibition of Ca2+-dependent K+ channels by lead in one-step inside-out vesicles from human red cell membranes

Pb2+ modified the apparent threshold sensitivity to Ca2+ of individual K+ channels with a biphasic time-course. At first, the sensitivity to Ca2+ was lowered with the result of a decrease of the fraction of activated vesicles at a given Ca2+ concentration. Later, Pb2+ increased the sensitivity to Ca2+ and the fraction of activated vesicles. The increase of Pb2+ concentration increased the extent of the initial inhibition but decreased its duration. The inhibitory effect was not observed when the addition of Ca2+ preceded the addition of Pb2+. The presence of Mg2+ in the incubation medium was also required. In the absence of Mg2+, Pb2+ decreased the rate of uptake of 86Rb, but no decrease in the fraction of activated vesicles could be demonstrated.

The all-or-none role of innervation in expression of apamin receptor and of apamin-sensitive Ca2+-activated K+ channel in mammalian skeletal muscle

The long-lasting after-hyperpolarization(s) (AHP) that follows the action potential in rat myotubes differentiated in culture is due to Ca2+-activated K+ channels. These channels have the property to be specifically blocked by the bee venom toxin apamin at low concentrations. Apamin has been used in this work to analyze, by electrophysiological and biochemical techniques, the role of innervation in expression of these important channels. The main results are as follows: (i) Long-lasting AHP that follows the action potential in rat myotubes in culture disappears when myotubes are cocultured with nerve cells from the spinal cord under the conditions of in vitro innervation. (ii) Extensor digitorum longus muscles from adult rats have action potentials that are not followed by AHP but AHP are systematically recorded after muscle denervation and they are blocked by apamin. (iii) Specific 125I-labeled apamin binding is undetectable in innervated muscle fibers but it becomes detectable 2-4 days after muscle denervation to be maximal 10 days after denervation. (iv) Apamin receptors detected with 125I-labeled apamin are present at fetal stages with biochemical characteristics identical to those found in myotubes in culture. The receptor number decreases as maturation proceeds and 125I-labeled apamin receptors completely disappear after the first week of postnatal life, in parallel with the disappearance of multi-innervation. All these results taken together strongly suggest an all-or-none effect of innervation on the expression of apamin-sensitive Ca2+-activated K+ channels.