Molecular and functional identification of sodium channels in K562 cells

Modulations of ion channel activity underlie rapid changes in membrane transport of cations in various nonexcitable cells. Previously, in smooth muscle cells, macrophages, lymphocytes, carcinoma and leukemia cell lines, non-voltage-gated sodium (NVGS) channels have been found. The activity of NVGS channels was shown to be critically dependent on the organization of actin cytoskeleton. The molecular identity of NVGS channels remains unclear. The present work is focused on molecular and functional identification of NVGS channels in human myeloid leukemia K562 cells. Degenerin/epithelial Na⁺ channels (DEG/ENaC) can be considered as possible molecular correlates. By using RT-PCR, expression of ??-, ??-, and ??-hENaC subunits in the K562 cells was detected. Various modes of the patch-clamp method were used to examine functional properties of sodium channels??specifically, to test the effect of amiloride on single channel and integral currents. The biophysical characteristics of the NVSG channels were close to those of ENaC; the channels have unitary conductance of 12 pS (145 mM Na⁺) and were impermeable to divalent cations (Ca²⁺ and Mg²⁺). We found that amiloride did not inhibit NVGS channels. Importantly, no amiloride-blockable sodium current was detected in the plasma membrane of K562 cells. Taken together, our observations suggest that amiloride-insensitive sodium channels in the K562 cells belong to the ENaC family.     

Disruption of actin filaments increases the activity of sodium-conducting channels in human myeloid leukemia cells.

With the use of the patch clamp technique, the role of cytoskeleton in the regulation of ion channels in plasma membrane of leukemic K562 cells was examined. Single-channel measurements have indicated that disruption of actin filaments with cytochalasin D (CD) resulted in a considerable increase of the activity of non-voltage-gated sodium-permeable channels of 12 pS unitary conductance. Background activity of these channels was low; open probability (po) did not exceed 0.01-0.02. After CD, po grew at least 10-20 times. Cell-attached and whole-cell recordings showed that activation of sodium channels was elicited within 1-3 min after the addition of 10-20 micrograms/ml CD to the bath extracellular solution or in the presence of 5 micrograms/ml CD in the intracellular pipette solution. Preincubation of K562 cells with CD during 1 h also increased drastically the activity of 12 pS sodium channels. Whole-cell measurements confirmed that CD-activated channels were permeable to monovalent cations (preferentially to Na+ and Li+), but not to bivalent cations (Ca2+, Ba2+). Colchicine (1 microM), which affect microtubules, did not alter background channel activity. Our data indicate that actin filaments organization plays an important role in the regulation of sodium-permeable channels which may participate in providing passive Na+ influx in red blood cells.     

Store-operated cationic channels in human myeloid leukaemia K562 cells

Using the whole-cell patch clamp technique, single channels operated by intracellular Ca(2+)-store depletion were first revealed in human myeloid leukaemia cells K562. A single store-operated channel could be detected in divalent-free extracellular solutions with Na+ as a permeant ion, and intracellular solutions with strong Ca(2+)-helating agent with some delay after whole-cell formation. Addition of inositol-1,4,5-triphosphate to the pipette solution resulted in a significant decrease of this latency. These channels had a conductance of 29 pS, and were inhibited by low concentration of external Ca2+. Our results enable us to assume that the revealed channels are calcium release-activated calcium channels, operated by Ca2+ depletion of endoplasmic reticulum.     

Sodium channel activity in leukemia cells is directly controlled by actin polymerization

The actin cytoskeleton has been shown to be involved in the regulation of sodium-selective channels in non-excitable cells. However, the molecular mechanisms underlying the changes in channel function remain to be defined. In the present work, inside-out patch experiments were employed to elucidate the role of submembranous actin dynamics in the control of sodium channels in human myeloid leukemia K562 cells. We found that the application of cytochalasin D to the cytoplasmic surface of membrane fragments resulted in activation of non-voltage-gated sodium channels of 12 picosiemens conductance. Similar effects could be evoked by addition of the actin-severing protein gelsolin to the bath cytosol-like solution containing 1 microm [Ca(2+)](i). The sodium channel activity induced by disassembly of submembranous microfilaments with cytochalasin D or gelsolin could be abolished by intact actin added to the bath cytosol-like solution in the presence of 1 mm MgCl(2) to induce actin polymerization. In the absence of MgCl(2), addition of intact actin did not abolish the channel activity. Moreover, the sodium currents were unaffected by heat-inactivated actin or by actin whose polymerizability was strongly reduced by cleavage with specific Escherichia coli A2 protease ECP32. Thus, the inhibitory effect of actin on channel activity was observed only under conditions promoting rapid polymerization. Taken together, our data show that sodium channels are directly controlled by dynamic assembly and disassembly of submembranous F-actin.     

Formyl-peptide receptor like 1: a potent mediator of the Ca2+ release-activated Ca2+ current ICRAC

In electrically non-excitable cells, one major source of Ca²⁺ influx is through the store-operated (or Ca²⁺ release-activated Ca²⁺) channel by which the process of emptying the intracellular Ca²⁺ stores results in the activation of Ca²⁺ channels in the plasma membrane. Using both whole-cell patch-clamp and Ca²⁺ imaging technique, we describe the electrophysiology mechanism underlying formyl-peptide receptor like 1 (FPRL1) linked to intracellular Ca²⁺ mobilization. The FPRL1 agonists induced Ca²⁺ release from the endoplasmic reticulum and subsequently evoked I_CRAC-like currents displaying fast inactivation in K562 erythroleukemia cells which expresses FPRL1, but had almost no effect in K562 cells treated with FPRL1 RNA-interference and HEK293 cells which showed no FPRL1 expression. The currents were impaired after either complete store depletion by the sarco/endoplasmic reticulum Ca²⁺-ATPase inhibitor thapsigargin, or after inhibition of PLC by U73122. Our results present the first evidence that FPRL1 is a potent mediator in the activation of CRAC channels.     

Actin cytoskeleton disassembly affects conductive properties of stretch-activated cation channels in leukaemia cells

Mechanosensitive channels in various eucaryotic cells are thought to be functionally and structurally coupled to the cortical cytoskeleton. However, the results of electrophysiological studies are rather controversial and the functional impact of cytoskeleton assembly-disassembly on stretch-activated channel properties remains unclear. Here, the possible involvement of cytoskeletal elements in the regulation of stretch-activated Ca²⁺-permeable channels was studied in human leukaemia K562 cells with the use of agents that selectively modify the actin or tubulin system. F-actin disassembly resulted in a considerable reduction of the amplitude of stretch-activated currents without significant change in channel open probability. The effects of treatments with cytochalasins or latrunculin were principally similar, developed gradually and consisted a strong decrease of single channel conductance. Microtubule disruption did not affect stretch-activated channels. The data presented here are in principal agreement with the general conclusion that mechanosensitive channel functions are largely dependent on the integrity of the cortical actin cytoskeleton. Specifically, changes in conductive properties of the pore may provide an essential mechanism of channel regulation underlying functional modulation of membrane currents. Our results allow one to speculate that microfilament organization may be an important determinant in modulating biophysical characteristics of stretch-activated cation channels in cells of blood origin.     

Regulation of sodium channel activity by capping of actin filaments

Ion transport in various tissues can be regulated by the cortical actin cytoskeleton. Specifically, involvement of actin dynamics in the regulation of nonvoltage-gated sodium channels has been shown. Herein, inside-out patch clamp experiments were performed to study the effect of the heterodimeric actin capping protein CapZ on sodium channel regulation in leukemia K562 cells. The channels were activated by cytochalasin-induced disruption of actin filaments and inactivated by G-actin under ionic conditions promoting rapid actin polymerization. CapZ had no direct effect on channel activity. However, being added together with G-actin, CapZ prevented actin-induced channel inactivation, and this effect occurred at CapZ/actin molar ratios from 1:5 to 1:100. When actin was allowed to polymerize at the plasma membrane to induce partial channel inactivation, subsequent addition of CapZ restored the channel activity. These results can be explained by CapZ-induced inhibition of further assembly of actin filaments at the plasma membrane due to the modification of actin dynamics by CapZ. No effect on the channel activity was observed in response to F-actin, confirming that the mechanism of channel inactivation does not involve interaction of the channel with preformed filaments. Our data show that actin-capping protein can participate in the cytoskeleton-associated regulation of sodium transport in nonexcitable cells.     

Properties of Mg(2+)-dependent cation channels in human leukemia K562 cells

The endogenous Mg(2+)-inhibited cation (MIC) current was recently described in different cells of hematopoietic lineage and was implicated in the regulation of Mg2+ homeostasis. Here we present a single channel study of endogenously expressed Mg(2+)-dependent cation channels in the human myeloid leukemia K562 cells. Inwardly directed unitary currents were activated in cell-attached experiments in the absence of Ca2+ and Mg2+ in the pipette solution. The current-voltage (I-V) relationships displayed strong inward rectification and yielded a single channel slope conductance of approximately 30 pS at negative potentials. The I-V relationships were not altered by patch excision into divalent-free solution. Channel open probability (P(o)) and mean closed time constant (tau(C)) were strongly voltage-dependent, indicating that gating mechanisms may underlie current inward rectification. Millimolar concentrations of Ca2+ or Mg2+ applied to the cytoplasmic side of the membrane produced slow irreversible inhibition of channel activity. The Mg(2+)-dependent cation channels described in this study differ from the MIC channels described in human T-cells, Jurkat, and rat basophilic leukemia (RBL) cells in their I-V relationships, kinetic parameters and dependence on intracellular divalent cations. Our results suggested that endogenously expressed Mg(2+)-dependent cation channels in K562 cells and the MIC channels in other hematopoietic cells might be formed by different channel proteins.     

Modification of ATP-sensitive K+ channels by proteolysis in smooth muscle cells from pig urethra

Patch-clamp experiments have been performed to investigate the effects of endoproteases (such as trypsin, carboxypeptidase B) on both membrane currents and unitary currents in isolated smooth muscle cells from pig proximal urethra (conventional whole-cell configuration, cell-attached configuration, and inside-out patches). Application of either trypsin (1 mg/mL) or carboxypeptidase B (0.1 mg/mL) to the intracellular surface of the excised membrane patches stimulated the activity of a 2.1 pA K+ channel (in symmetrical 140 mM K+ conditions) at a holding potential of -50 mV. The trypsin-induced K+ channels in inside-out configuration exhibited the same amplitude and similar channel opening kinetics to the levcromakalim-induced ATP-sensitive K+ channel (i.e. K ATP channel) in cell-attached patches of the same membrane; however, the sensitivity of the channels to glibenclamide was greatly reduced after the trypsin-treatment. The activity of the trypsin-induced K+ channel was reversibly inhibited by cibenzoline in an inside-out configuration (Ki = 5 microM). It is concluded that trypsin and carboxypeptidase B reactivate the channel with an intact pore activity but the different pharmacological properties of the channels may reflect some change in the conformation in channel proteins after proteolysis.     

Functional properties of sodium channels in cholesterol-depleted K562 cells

In the present paper, functional properties of nonvoltage-gated sodium channels in K562 cells were studied after cholesterol depletion, i.e., under conditions of the destruction of microdomains (rafts). For cholesterol depletion, cells were incubated with methyl-beta-cyclodextrin (MbCD), an oligosaccharide that selectively binds sterols. Single currents through sodium channels were recorded in cell-attached and inside-out experiments using the patch-clamp technique. After incubation with MbCD (2.5 or 5 mM), the activation of sodium channels in response to cytochalasin B or D was observed in both native cells and membrane fragments. Biophysical characteristics of sodium channels in cholesterol-depleted K562 cells were close to those in control; unitary conductance was 12 pS. Inside-out experiments with the use of globular actin have indicated that filament assembly on cytoplasmic membrane side causes an inactivation of sodium channels in the modified cells. These data imply that sodium channels in K562 cells are not associated with cholesterol-rich membrane microdomains. Possible mechanisms of the interaction of the plasma membrane and the cortical cytoskeleton are discussed.     

Cell swelling activates cloned Ca(2+)-activated K(+) channels: a role for the F-actin cytoskeleton

Cloned Ca(2+)-activated K(+) channels of intermediate (hIK) or small (rSK3) conductance were expressed in HEK 293 cells, and channel activity was monitored using whole-cell patch clamp. hIK and rSK3 currents already activated by intracellular calcium were further increased by 95% and 125%, respectively, upon exposure of the cells to a 33% decrease in extracellular osmolarity. hIK and rSK3 currents were inhibited by 46% and 32%, respectively, by a 50% increase in extracellular osmolarity. Cell swelling and channel activation were not associated with detectable increases in [Ca(2+)](i), evidenced by population and single-cell measurements. In addition, inhibitors of IK and SK channels significantly reduced the rate of regulatory volume decrease (RVD) in cells expressing these channels. Cell swelling induced a decrease, and cell shrinkage an increase, in net cellular F-actin content. The swelling-induced activation of hIK channels was strongly inhibited by cytochalasin D (CD), in concentrations that caused depolymerization of F-actin filaments, indicating a role for the F-actin cytoskeleton in modulation of hIK by changes in cell volume. In conclusion, hIK and rSK3 channels are activated by cell swelling and inhibited by shrinkage. A role for the F-actin cytoskeleton in the swelling-induced activation of hIK channels is suggested.     

HERG1 Currents in Native K562 Leukemic Cells

The human ether-a-go-go related gene (HERG1) K⁺ channel is expressed in neoplastic cells, in which it was proposed to play a role in proliferation, differentiation and/or apoptosis. K562 cells (a chronic myeloid leukemic human cell line) express both the full-length (herg1a) and the N-terminally truncated (herg1b) isoforms of the gene, and this was confirmed with Western blots and coimmunoprecipitation experiments. Whole-cell currents were studied with a tail protocol. Seventy-eight percent of cells showed a HERG1-like current: repolarization to voltages negative to −40 mV produced a transient peak inward tail current, characteristic of HERG1 channels. Cells were exposed to a HERG-specific channel blocker, E4031. Half-maximal inhibitory concentration (IC₅₀) of the blocker was 4.69 nm. The kinetics of the HERG1 current in K562 cells resembled the rapid component of the native cardiac delayed rectifier current, known to be conducted by heterotetrameric HERG1 channels. Fast and slow deactivation time constants at −120 mV were 27.5 and 239.5 ms, respectively. Our results in K562 cells suggest the assembling of heterotetrameric channels, with some parameters being dominated by one of the isoforms and other parameters being intermediate. Hydrogen peroxide was shown to increase HERG1a K⁺ current in heterologous expression systems, which constitutes an apoptotic signal. However, we found that K562 HERG1 whole-cell currents were not activated by H₂O₂.     

Drug-transport and volume-activated chloride channel functions in human erythroleukemia cells: Relation to expression level of P-glycoprotein

The characteristics of volume-activated chloride currents, drug transport function and levels of P-glycoprotein (PgP) expression were compared between two human chronic erythroleukemia cell lines: a parental (K562) cell line and a derivative obtained by vinblastine selection (K562 VBL400). Parental K562 cells showed no detectable P-glycoprotein expression, measured at the protein level (immunofluorescence labeling with monoclonal antibodies), and had very low levels of MDR-1 mRNA expression (RT-PCR analysis), when compared with levels measured in K562 VBL400. Differences in Pgp-mediated transport were estimated by comparing the rates of Fluo3 accumulation. The higher drug-transport function of K562 VBL400 cells (e.g., lower Fluo3 accumulation) correlated with their elevated levels of MDR-1. The rate of dye transport was sensitive to verapamil but was not affected by the tonicity of the extracellular medium.In contrast to the clear differences in transport function, the characteristics of chloride currents induced by cell swelling were indistinguishable between the two cell lines. Currents measured in the whole-cell configuration were outwardly rectifying, had a higher permeability to iodide than to chloride (SCN^– > I^– > Cl^– > gluconate), were potently blocked by NPPB and were unresponsive to verapamil. The percentage of responding cells and the mean current density were nearly identical in both cell lines. In addition, activation of the volume-sensitive current was not prevented during whole-cell recordings obtained with pipettes containing high concentration of cytotoxic drugs (vincristine or vinblastine). These results do not lend support to the previously reported association between Pgp expression and volume-sensitive chloride channels, and suggest that a different protein is responsible for this type of chloride channel in K562 cells.The authors wish to thank Dr. Humbert de Smedt, Ms. Anja Florizoone and Ms. Marina Crabbe for assistance in the culturing of cells. F.V. was supported by a post-doctoral fellowship (EX93 36037569) from the Ministerio de Educatión y Ciencia (Spain). K.V.A. was supported by the Institute for Scientific Research in Agriculture and Industry (Belgium). J.E. is a postdoctoral fellow of the Belgian National Fund for Scientific Research (NFWO). C.D.G. and B.N. received support from the Max Planck Gesselschaft (Germany).     

Inhibiting and stimulating effect of amiloride on potential-dependent cation channels in K562 cells

Non-voltage-gated ion channels play an essential role in cellular signalling and ionic homeostasis in nonexcitable cells. The patch clamp method in cell-attached configuration was used to search for the effects of amiloride and gadolinium (Gd3+) exerted on two types of voltage-insensitive cationic channels in plasma membrane of human leukemia K562 cells: Na-selective channels activated by actin disassembly, and mechanosensitive channels. Here we demonstrate that amiloride in high concentrations (1 mM) caused a full inhibition of mechanosensitive channels in K562 cells similarly to Gd3+ effect in micromolecular concentration range. Na-selective channels controlled by actin dynamics were shown to be unaffected by Gd3+ similarly as by amiloride. We also found that application of amiloride to the extracellular surface of membrane patch resulted in a significant increase in the activity of sodium channels. This unexpected stimulatory effect of amiloride may represent an unknown mechanism of activation of non-voltage-gated sodium channels. The data show an essential difference of the activation and blockage of these types of cation-selective channels.     

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     

Active site conformational changes of prostasin provide a new mechanism of protease regulation by divalent cations

Prostasin or human channel‐activating protease 1 has been reported to play a critical role in the regulation of extracellular sodium ion transport via its activation of the epithelial cell sodium channel. Here, the structure of the extracellular portion of the membrane associated serine protease has been solved to high resolution in complex with a nonselective d‐FFR chloromethyl ketone inhibitor, in an apo form, in a form where the apo crystal has been soaked with the covalent inhibitor camostat and in complex with the protein inhibitor aprotinin. It was also crystallized in the presence of the divalent cation Ca⁺². Comparison of the structures with each other and with other members of the trypsin‐like serine protease family reveals unique structural features of prostasin and a large degree of conformational variation within specificity determining loops. Of particular interest is the S1 subsite loop which opens and closes in response to basic residues or divalent ions, directly binding Ca⁺² cations. This induced fit active site provides a new possible mode of regulation of trypsin‐like proteases adapted in particular to extracellular regions with variable ionic concentrations such as the outer membrane layer of the epithelial cell.     

Actin cytoskeleton disassembly affects conductive properties of stretch-activated cation channels in leukaemia cells

Mechanosensitive channels in various eucaryotic cells are thought to be functionally and structurally coupled to the cortical cytoskeleton. However, the results of electrophysiological studies are rather controversial and the functional impact of cytoskeleton assembly-disassembly on stretch-activated channel properties remains unclear. Here, the possible involvement of cytoskeletal elements in the regulation of stretch-activated Ca2+-permeable channels was studied in human leukaemia K562 cells with the use of agents that selectively modify the actin or tubulin system. F-actin disassembly resulted in a considerable reduction of the amplitude of stretch-activated currents without significant change in channel open probability. The effects of treatments with cytochalasins or latrunculin were principally similar, developed gradually and consisted a strong decrease of single channel conductance. Microtubule disruption did not affect stretch-activated channels. The data presented here are in principal agreement with the general conclusion that mechanosensitive channel functions are largely dependent on the integrity of the cortical actin cytoskeleton. Specifically, changes in conductive properties of the pore may provide an essential mechanism of channel regulation underlying functional modulation of membrane currents. Our results allow one to speculate that microfilament organization may be an important determinant in modulating biophysical characteristics of stretch-activated cation channels in cells of blood origin.     

Activation of mechanosensitive ion channels in plasma membrane of K562 cells

Patch clamp method in cell-attached configuration was used to search for mechanogated ion channels in plasma membrane of human myeloid leukemia K562 cells. A reversible activation of transmembrane currents in response to negative pressure applied to membrane patch was observed. Four types of mechanosensitive channels were identified in K562 cells: two main types were characterized with conductance values of 16 and 25 pS; while two others, showing higher conductance values (about 35 and 50 pS), were rarely met. In terms of gating, all channels described here could be assigned to the stretch-activated type. No inactivation of mechanosensitive channels at the sustained stimulation was observed. The activation of mechanosensitive channels in K562 cells was not dependent upon the presence of bivalent cations in the extracellular solution.     

The cytosolic inactivation domains of BKi channels in rat chromaffin cells do not behave like simple, open-channel blockers.

Most BK-type voltage- and Ca(2+)-dependent K+ channels in rat chromaffin cells exhibit rapid inactivation. This inactivation is abolished by brief trypsin application to the cytosolic face of membrane patches. Here we examine the effects of cytosolic channel blockade and pore occupancy on this inactivation process, using inside-out patches and whole-cell recordings. Occupancy of a superficial pore-blocking site by cytosolic quaternary blockers does not slow inactivation. Occupancy of a deeper pore-blocking site by cytosolic application of Cs+ is also without effect on the onset of inactivation. Although the rate of inactivation is relatively unaffected by changes in extracellular K+, the rate of recovery from inactivation (at -80 and -140 mV with 10 microM Ca2+) is faster with increases in extracellular K+ but is unaffected by the impermeant ion, Na+. When tail currents are compared after repolarization, either while channels are open or after inactivation, no channel reopening is detectable during recovery from inactivation. BK inactivation appears to be mechanistically distinct from that of other inactivating voltage-dependent channels. Although involving a trypsin-sensitive cytosolic structure, the block to permeation does not appear to occur directly at the cytosolic mouth or inner half of the ion permeation pathway.