Exposure of erythrocytes to the Ca2+ ionophore ionomycin has recently been shown to induce cell shrinkage, cell membrane blebbing, and breakdown of phosphatidylserine asymmetry, all features typical of apoptosis of nucleated cells. Although breakdown of phosphatidylserine asymmetry is thought to result from activation of a Ca2+-sensitive scramblase, the mechanism and role of cell shrinkage have not been explored. The present study was performed to test whether ionomycin-induced activation of Ca2+-sensitive Gardos K+ channels and subsequent cell shrinkage participate in ionomycin-induced breakdown of phosphatidylserine asymmetry of human erythrocytes. According to on-cell patch-clamp experiments, ionomycin (1 μM) induces activation of inwardly rectifying K+-selective channels in the erythrocyte membrane. Fluorescence-activated cell sorter analysis reveals that ionomycin leads to a significant decrease of forward scatter, reflecting cell volume, an effect blunted by an increase of extracellular K+ concentration to 25 mM and exposure to the Gardos K+ channel blockers charybdotoxin (230 nM) and clotrimazole (5 μM). As reflected by annexin binding, breakdown of phosphatidylserine asymmetry is triggered by ionomycin, an effect again blunted, but not abolished, by an increase of extracellular K+ concentration and exposure to charybdotoxin (230 nM) and clotrimazole (5 μM). Similar to ionomycin, glucose depletion leads (within 55 h) to annexin binding of erythrocytes, an effect again partially reversed by an increase of extracellular K+ concentration and exposure to charybdotoxin. K-562 human erythroleukemia cells similarly respond to ionomycin with cell shrinkage and annexin binding, effects blunted by antisense, but not sense, oligonucleotides against the small-conductance Ca2+-activated K+ channel isoform hSK4 (KCNN4). The experiments disclose a novel functional role of Ca2+-sensitive K+ channels in erythrocytes, i.e., their participation in regulation of erythrocyte apoptosis.
- cell volume
treatment of erythrocytes with the Ca2+ ionophore ionomycin has recently been shown to induce erythrocyte shrinkage, membrane blebbing, and breakdown of cell membrane phosphatidylserine asymmetry (3, 9, 11, 28), all typical features of apoptosis in other cell types (18, 19). The breakdown of phosphatidylserine asymmetry results from activation of a scramblase by an increase of cytosolic Ca2+ activity (12, 38). Inasmuch as macrophages are equipped with receptors specific for phosphatidylserine (15, 20, 33), erythrocytes exposing phosphatidylserine at their surface will be rapidly recognized, engulfed, and degraded (5, 14). Thus an increase of cytosolic Ca2+ activity could trigger apoptotic death and clearance of erythrocytes.
Ca2+ may enter erythrocytes through Ca2+-permeable cation channels (13, 22), which are activated after osmotic shock by an increase of extracellular osmolarity, oxidative stress by addition of t-butylhydroxyperoxide, and energy depletion by removal of extracellular glucose (28). Activation of these channels triggers breakdown of phosphatidylserine asymmetry and subsequent erythrocyte death (28).
The mechanism underlying cell shrinkage has not been addressed in those studies. Inasmuch as erythrocytes express Ca2+-sensitive K+ channels (6, 10), activation of these channels is expected to participate in the shrinking effect of ionomycin. The present experiments have been performed to test for activation of these channels by an increase of cytosolic Ca2+ activity and to explore whether these channels participate in ionomycin-induced cell shrinkage and breakdown of phosphatidylserine asymmetry. To this end, cellular K+ loss has been inhibited by an increase of extracellular K+ concentration and inhibition of the K+ channels by charybdotoxin or clotrimazole (2, 25). To identify the channel involved, additional experiments have been carried out in K-562 erythroleukemia cells with or without prior treatment with antisense oligonucleotides against hSK4 (KCNN4), the main isoform of the Ca2+-activated K+ channel in human red blood cells (21).
Cells and solutions. Erythrocytes were drawn from healthy volunteers. Experiments were performed at 37°C in Ringer solution containing 125 mM NaCl, 5 mM KCl, 1 mM MgSO4, 32 mM HEPES, 5 mM glucose, and 1 mM CaCl2 (pH 7.4). Where indicated, glucose was removed from the medium, and Na+ was partially replaced by 1 μM K+ and/or ionomycin (Sigma, Munich, Germany), 230 nM charybdotoxin (Sigma), and/or 5 μM clotrimazole. The final concentration of the solvent dimethylsulfoxide was <0.2%.
K-562 cells (kind gift of Dr. M. Topp, Dept. of Hematology, Oncology, Immunology, and Rheumatology, University Medical Center, University of Tuebingen, Germany) were grown in RPMI 1640 medium supplemented with 10% fetal calf serum, 0.56 g/l L-glutamine, 100,000 U/l penicillin, and 0.1 g/l streptomycin.
Antisense treatment of K-562 cells. Sense (5′-ATG GGC GGG GAT CTG GTG CT-3′) and antisense (5′-AGC ACC AGA TCC CCG CCC AT-3′) probes for hSK4 (KCNN4), synthesized with phosphorothioate bases, were purchased from InVitrogen (Karlsruhe, Germany). For treatment, K-562 cells were incubated in culture medium containing 1% fetal calf serum. The antisense or sense probes (2 μM) were added to the culture medium 48 h before the experiments, as described previously (37).
Patch clamp. The procedure used for patch-clamp recording has been described by Huber et al. (22). Single channels were recorded at room temperature from freshly obtained human erythrocytes. Records were obtained in on-cell, voltage-clamp mode with the use of patch pipettes made of borosilicate glass (model 150 TF-10, Clark Medical Instruments) with resistances of 8-12 MΩ. Currents were recorded and low-pass filtered at 3 kHz with an EPC-9 amplifier (Heka, Lambrecht, Germany) using Pulse software (Heka) and an ITC-16 interface (Instrutech, Port Washington, NY). Pipette solutions containing (in mM) 140 X-d-gluconate, 10 X-Cl, 5 HEPES/X-OH (pH 7.4), and 1 MgCl2 (with X = Na or K) were combined with a bath solution containing (in mM) 125 NaCl, 5 KCl, 1 MgSO4, 5 glucose, 1 CaCl2, and 32 HEPES/NaOH (pH 7.4). Ca2+-activated (Gardos) channel activity was stimulated by addition of the Ca2+ ionophore ionomycin (1 μM) to the bath solution. The applied voltages refer to the cytoplasmic face of the membrane with respect to the extracellular space. The inward currents, defined as flow of positive charge from the pipette into the cell, are negative currents and depicted as downward deflections of the original current traces.
Fluorescent-activated cell sorter analysis. Fluorescent-activated cell sorter (FACS) analysis was performed essentially as described previously (1, 28). After incubation, cells were washed in annexin-binding buffer containing 125 mM NaCl, 10 mM HEPES (pH 7.4), and 5 mM CaCl2. Erythrocytes were stained with annexin-Flous (Boehringer Mannheim) at a 1:100 dilution. After 15 min, samples were diluted 1:5 and measured by flow cytometric analysis (FACS-Calibur, Becton Dickinson, Heidelberg, Germany). The fluorescence intensity of annexin was measured in fluorescence channel FL-1. Cell volume was determined by forward- and side-scatter analysis of unstained cells.
Statistics. Values are means ± SE, and statistical analysis was performed by paired or unpaired t-test, where appropriate. P ≤ 0.05 was considered statistically significant.
To test for involvement of the Ca2+-sensitive Gardos channels during Ca2+ ionophore-induced cell shrinkage and phosphatidylserine exposure, human erythrocyte activity was recorded in on-cell mode during stimulation with 1 μM ionomycin. Non-treated cells exhibited almost no basal Gardos channel activity (Fig. 1A, left). Addition of 1 μM ionomycin to the bath solution induced channel activity within seconds (Fig. 1A, right). Comparison of records obtained with K+ (Fig. 1A, bottom) and Na+ (Fig. 1A, top) as principal cations in the pipette solution indicated K+ selectivity of the Ca2+ ionophore-induced channels. Addition of ionomycin induced activity of multiple channels in the sealed membrane patch (Fig. 1, B and E). When recorded with K+-containing pipette solution, these channels exhibited a slightly inwardly rectifying current-voltage relation (Fig. 1, B and C) with a unitary conductance of ∼10 and 30 pS at positive and negative voltages, respectively (Fig. 1, C and D).
As evident from the forward scatter in FACS analysis (Fig. 2), exposure of erythrocytes to ionomycin led to rapid, sustained cell shrinkage. Addition of 125 mM K+ to replace equimolar Na+ did not significantly modify the forward scatter in the absence of ionomycin but virtually abolished the decrease of cell volume after addition of 1 μM ionomycin. As illustrated in Fig. 2B,a gradual increase of extracellular K+ concentration led to a corresponding increase of forward scatter in the presence of ionomycin.
As reflected by annexin binding detected in FACS analysis, addition of ionomycin led to rapid breakdown of phosphatidylserine asymmetry across the erythrocyte cell membrane (Fig. 3), an effect that was blunted in the presence of 125 mM extracellular K+. An increase of extracellular K+ concentration to 25 mM was sufficient to significantly decrease the number of annexin-binding cells in the presence of ionomycin (Fig. 3). A further increase of extracellular K+ concentration to 125 mM, however, did not exert a significant additional inhibition of annexin binding (Fig. 3).
As illustrated in Fig. 4, the ionomycin (40 min)-induced cell shrinkage was similarly reversed in the presence of the K+ channel blockers charybdotoxin (230 nM) and clotrimazole (5 μM).
Pharmacological inhibition of the K+ channels further blunted the breakdown of phosphatidylserine asymmetry after exposure to ionomycin (Fig. 5). The number of annexin-binding cells after 40 min of exposure to 1 μM ionomycin significantly decreased in the presence of 230 nM charybdotoxin or 5 μM clotrimazole (Fig. 5).
Similar to ionomycin, glucose depletion (55 h) led to cell shrinkage (Fig. 6). Addition of 125 mM K+ to replace equimolar Na+ did not significantly modify the forward scatter in the presence of glucose but virtually abolished the decrease of cell volume after removal of glucose. As illustrated in Fig. 6, a gradual increase of extracellular K+ concentration led to a corresponding increase of forward scatter in glucose-depleted erythrocytes.
The effect of increased K+ concentration (125 mM) on cell volume of glucose-depleted erythrocytes was paralleled by inhibition of annexin binding in those cells (Fig. 7). An increase of extracellular K+ concentration to 75 mM significantly decreased the number of annexin-binding cells (Fig. 7). Further increase of extracellular K+ concentration led to a moderate further decrease of the number of annexin-binding cells (Fig. 7).
To explore the molecular identity of the Ca2+-regulated K+ channels involved in ionomycin-induced shrinkage and annexin binding, we used antisense oligonucleotides against hSK4 (KCNN4), which was most recently shown to represent the Gardos channel in red blood cells (21). As shown in Fig. 10, ionomycin significantly decreased the forward scatter in control cells, an effect reversed in the presence of 5 μM clotrimazole. In antisense-treated, but not sense-treated, K-562 cells, the effect of ionomycin was significantly blunted, and the effect of 5 μM clotrimazole was virtually abolished (Fig. 10). The decrease of forward scatter after ionomycin treatment was paralleled by a marked increase of annexin binding, an effect significantly blunted in the presence of 5 μM clotrimazole (Fig. 11). Again, treatment with antisense, but not sense, oligonucleotides significantly blunted the effect of ionomycin and virtually abolished the effect of clotrimazole (Fig. 11).
The present study confirms the previous observations that exposure of erythrocytes to ionomycin leads to cell shrinkage and annexin binding (28). Our observations further provide a mechanism contributing to the observed cell shrinkage. It is shown that the erythrocytes express Ca2+-activated, charybdotoxin- and clotrimazole-sensitive K+ channels and that the cell shrinkage is inhibited by an increase of extracellular K+ concentration and addition of the K+ channel blockers charybdotoxin and clotrimazole (2, 25). The sensitivity of cell shrinkage to antisense oligonucleotides against the hSK4 (KCNN4) isoform of the Ca2+-activated K+ channel points to the involvement of this channel, which has most recently, indeed, been shown to represent the Gardos channel of erythrocytes (21). By stimulating the K+ channels, increasing cytosolic Ca2+ concentrations obviously leads to erythrocyte hyperpolarization, which drives Cl- out of the cell. The loss of KCl with the osmotically obliged water then leads to the observed cell shrinkage. An increase of extracellular K+ concentration limits the hyperpolarization by decreasing the zero-current potential for K+ and, thus, blunts the ionomycin-induced cell shrinkage. Similarly, inhibition of the K+ channels by charybdotoxin and clotrimazole prevents the hyperpolarization and subsequent KCl loss.
Surprisingly, inhibition of cellular K+ loss by an increase of extracellular K+ concentration or addition of the K+ channel blockers charybdotoxin and clotrimazole not only blunted the ionomycin-induced cell shrinkage but also the ionomycin-induced annexin binding. Possibly, the hyperpolarization following the activation of Ca2+-sensitive K+ channels drives the positively charged Ca2+ into the cell, thus augmenting the increase of cytosolic Ca2+. Moreover, cell shrinkage may have contributed to the triggering of the scramblase. Osmotic shock has been shown to stimulate apoptosis in a wide variety of cells (29, 36). In erythrocytes, the annexin binding following osmotic shock was only partially blunted in the absence of Ca2+, pointing to an additional mechanism participating in the induction of apoptosis (28). In some cell types, loss of cellular K+ is by itself considered a requirement of apoptosis (7, 8, 17, 23, 24, 34, 35). In any case, at least in erythrocytes, the activation of Ca2+-sensitive K+ channels serves as an enhancing element in Ca2+-induced cell death. Similar to ionomycin-induced cell shrinkage, the annexin binding of K-562 was blunted by antisense oligonucleotides against hSK4 (KCNN4), again pointing to the involvement of this channel.
Activation of the Ca2+-sensitive K+ channels is particularly important in cell shrinkage and deformation of deoxygenized sickle cells (6, 10, 16, 26, 31), leading to a severe increase of blood viscosity (26). Most recent studies revealed that sickle cells are more susceptible to several triggers of erythrocyte apoptosis, including ionomycin (30). The increase of viscosity and the accelerated clearance of sickle cells may thus be at least in part due to enhanced exposure of phosphatidylserine at the cell surface (32).
The parallel stimulation of erythrocyte scramblase and Ca2+-sensitive K+ channels by an increase of cytosolic Ca2+ activity presumably serves to circumvent hemolysis of defective erythrocytes. Leakage of the cell membrane or impairment of Na+-K+-ATPase is expected to result in gain of NaCl, cell swelling, and, eventually, cell lysis (27). Even in normal erythrocytes, cytosolic Cl- activity is high, and the potential difference across the cell membrane is low (4). Thus the gain of Na+ and loss of K+ must be excessive to compromise volume constancy. Prior to that the leaky cell membrane increases cytosolic Ca2+ activity, which activates the erythrocyte scramblase, leading to breakdown of phosphatidylserine asymmetry. The phosphatidylserine-exposing erythrocytes are detected by the macrophages (15, 20, 33) and, thus, cleared from the circulation before hemolysis (5, 14). The parallel activation of the Ca2+-sensitive K+ channels, hyperpolarization, and cellular loss of K+ and Cl- counteract cell swelling and, thus, may serve to prevent premature lysis of the erythrocytes.
In conclusion, enhanced entry of Ca2+ or energy depletion lead to activation of the erythrocyte Gardos channel. The subsequent loss of K+ and Cl- leads to cell shrinkage, which participates in triggering of erythrocyte apoptosis.
This study was supported by Deutsche Forschungsgemeinschaft Grants La 315/4-3, La 315/6-1, and La 315/11-1, Bundesministerium für Bildung, Wissenschaft, Forschung, und Technologie (Center for Interdisciplinary Clinical Research) Grant 01 KS-9602, and European Union Biomedical Program Grant BMH4-CT96-0602.
We acknowledge the technical assistance of E. Faber and the meticulous preparation of the manuscript by Lejla Subasic and Tanja Loch.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society