- Email: [email protected]
Oxidation induces ClC-3-dependent anion channels in human leukaemia cells
FEBS Letters 581 (2007) 5407–5412
Oxidation induces ClC-3-dependent anion channels in human leukaemia cells Ravi S. Kasinathan, Michael Fo¨ller, Camelia Lang, Saisudha Koka, Florian Lang, Stephan M. Huber* Department of Physiology I, Eberhard-Karls-University Tu¨bingen, Gmelinstraße 5, D-72076 Tu¨bingen, Germany Received 27 June 2007; revised 17 October 2007; accepted 17 October 2007 Available online 30 October 2007 Edited by Vladimir Skulachev
Abstract To test for redox regulation of anion channels in erythroid cells, we exposed K562 cells to oxidants and measured changes in transmembrane Cl currents using patch-clamp, and in intracellular Cl content using the Cl selective dye MQAE. Oxidation with tert-butylhydroperoxide or H2O2 produced a plasma membrane anion permeability with a permse lectivity of NO 3 > lactate > gluconate . The permeability increase was paralleled by insertion of ClC-3 protein into the plasma membrane as evident from immunofluorescence microscopy and surface biotinylation. Down-regulation of ClC-3 protein by RNA interference as assessed by immunoblotting decreased the oxidation-stimulated permeability. In conclusion, oxidation induces surface expression of ClC-3 and activation of a ClC-3-dependent anion permeability in K562 cells. Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: K562; Clcn3; Cl channels; MQAE-fluorescence; Flow-cytometry; RNA interference; Patch-clamp recording; Oxidative stress; Surface expression
sensitive organic osmolyte and anion channel [7]. In contrast to this assumption, anion channel activity is seemingly normal in cells from gene targeted mice lacking functional ClC-3 (Clcn3/ mice) [8]. Moreover, ClC-4 and -5 have recently been shown to function as electrogenic Cl/H+ exchangers [9,10] and such a function has also been suggested for ClC-3 [11]. The present study aimed to characterize redox-sensitive organic osmolyte and anion channels by use of whole-cell patch-clamp recording, flow cytometry, immunofluorescence microscopy and immunoblotting. The K562 myeloid leukaemia cell line shares many characteristics with erythrocytes [12]. In particular, its expression pattern of ion channels in the plasma membrane is similar to that of human erythrocytes [13–15]. K562 cells were, therefore, chosen as an erythroblastlike cell model to study redox-sensitive changes in plasma membrane permeability with specific emphasis on ClC-3 as a candidate for a redox-sensitive anion channel/regulator.
2. Materials and methods 1. Introduction Modulation of the redox potential, e.g. through formation of ROS regulates numerous cellular processes through oxidative interference with signaling molecules such as tyrosine kinases, tyrosine phosphatases, MAP kinases, through activation of transcription factors [1], or through direct modification of effector proteins [2]. Anion channels regulated by redox modulation accomplish Cl secretion [3] or volume changes of malaria-infected erythrocytes by mediating the efflux of Cl [4] and/or organic osmolytes [5]. ClC-3, a member of the voltage-gated ClC family of Cl channels [6], has been suggested initially to be a cell volume-
* Corresponding author. Fax: +49 (0) 7071 29 3073. E-mail address: [email protected] (S.M. Huber).
Abbreviations: ClC-3, member 3 of the cloned chloride channel family ClC; EGTA, ethylene glycol-bis(b-aminoethyl ether)-N,N,N 0 ,N 0 -tetraacetic acid; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; MAP, mitogen-activated protein; MQAE, N-(ethoxycarbonylmethyl)-6-methoxyquinolinium bromide; NMDG, N-methyl-D -glucamine; PBS, phosphate-buffered saline; ROS, reactive oxygen species; SDS, sodium dodecyl sulfate; siRNA, small interference RNA; t-BHP, tert-butylhydroperoxide
2.1. Cell culture and oxidation of K562 Cells K562 human myeloid leukaemia cells were grown in RPMI1640 medium containing 10% FCS. For oxidation, cells were washed, resuspended in NaCl solution containing (in mM) 125 NaCl, 32 HEPES/ NaOH, pH 7.4, 5 KCl, 5 D -glucose, 1 MgSO4, 1 CaCl2, and incubated in the presence or absence of 1 mM t-BHP for 30 min at 37 °C. In few patch-clamp experiments, cells were oxidized with H2O2 (1 mM for 30 min at 37 °C). Oxidation was stopped by washing the cells in NaCl solution. 2.2. siRNA based silencing of Clcn3 siRNA expression cassettes (H1 promoter) were prepared by a PCRbased method with the Silencerä Express Kit from Ambion (Cambridgeshire, UK). The following PCR primers (in parenthesis) were used for the constructs Clcn3-1 (sense: AGCTACACAAACTGCTTGACCTATGATTAACCGGTGTTTCGTCCTTTCCACAAG; antisense: CGGCGAAGCTTTTTCCAAAAAATTAATCATAGGTCAAGCAGCTACACAAACTGC; the targeted sequence as indicated by the bold letters is at position 1095–1113, accession number: NM_ 001829), Clcn3-2 (sense: CCTCTACACAAAAGGTTAAAGAGTTGTGCTACCGGTTCTTCCTTTCCACAAG; antisense: CGGC GAAGCTTTTTCCAAAAAATAGCACAACTCTTTAACCTCTACACAAAAGGT; the targeted sequence is at position: 3019–3037, NM_001829), and a nonsense construct (sense: CATCTACACAAAATGTGAAGAAGGGTTTATCCGGTGTTTCGTCCTTTCCACAAG; antisense: CGGCGAAGCTTTTTCCAAAAAAGATAAACCCTTCTTCACATCTACACAAAATGT). K562 cells (50 000 cells in 1 ml medium) were transfected for 48 h by adding siRNA expression cassettes (1 lg) mixed with siPORT XP-1 transfection agent (7 ll; Ambion) in Opti Mem Medium (200 ll; Gibco).
0014-5793/$32.00 Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.10.042
5408
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
Fig. 1. Whole-cell Cl currents of control and oxidized K562 cells. (A) Current traces recorded from a control (left panels) and an oxidized (right panels) K562 cell first with NMDG-Cl (left) and then NMDG-gluconate in the bath (right). (B,C) Mean current–voltage relationships (n = 7–8) of control (B) and oxidized K562 cells (C) recorded as in (A) with NMDG-Cl (open circles) and NMDG-gluconate (closed triangles) bath solution. Voltages refer to the intracellular membrane face with respect to the earthed bath solution. (D) Mean slope conductances (n = 7–8) as calculated from (B–C) for the outward currents of control (left) and oxidized K562 cells (right) in MMDG-Cl (open bars) and NMDG-gluconate (closed bars). (E) FACS histograms showing the Cl-specific MQAE-fluorescence of oxidized K562 cells incubated in K-gluconate (left) and KNO3 solution (right) for 0 min (black line) and 45 min (red line). (F) Mean normalized MQAE-fluorescence intensity (n = 5–12) of control (closed bars) and oxidized K562 cells (open bars) incubated for 45 min in K+ salts of different anions. Oxidized K562 cells show a chloride efflux which is promoted by bath anions with a permeability sequence of NO 3 > lactate > gluconate . 2.3. Patch-clamp experiments Currents were recorded in fast whole-cell mode as described [5] under conditions, where the cells did not swell. A pipette solution containing (in mM) 140 NMDG-Cl (pH 7.4), 40 mannitol, 10 HEPES/NMDG, 1 MgCl2, 1 Mg-ATP, 0.5 EGTA, was combined with a bath solution containing 180 NMDG-Cl (pH 7.4), 10 HEPES/ NMDG, 1 CaCl2, 1 MgCl2. In an additional bath solution (NMDGgluconate) chloride was replaced isosmotically by D -gluconate. Liquid junction potentials were recorded according to [16] and the data corrected. Prior to compensation, slow capacitive currents were evoked by a l0-mV square pulse, the capacitive current transients fitted by exponential regression and the whole-cell membrane capacitances calculated as described previously [17]. The recorded membrane capacitances did not differ between control (8.9 ± 0.5 pF; n = 8) and oxidized cells (8.0 ± 0.9 pF, n = 13).
pended in Cl-free solutions in which KCl was isosmotically replaced by KNO3, K-L -lactate, or K-D -gluconate. The solution further contained a cation channel blocker cocktail (50 lM clotrimazole to inhibit SK4 K+ channels [15], 50 lM ethylisopropylamiloride, EIPA, to inhibit non-selective cation channels [13,14,18,19], 50 lM 4-aminopyridine and 1 mM BaCl2 to inhibit further K+ channels) to avoid Clefflux-driven cell shrinkage by parallel loss of cellular K+ and H2O. Under these conditions, Cl efflux was limited by, and, therefore, a direct function of the uptake of the bath anion. Cl efflux was assessed after 45 min of incubation in Cl-free media. Cl efflux was indicated in flow cytometry (FACS-Calibur from Becton Dickinson, Heidelberg, Germany) by an increase in the Cl-specific MQAE fluorescence intensity as recorded in fluorescence channel FL-1 (excitation wavelength of 488 nm, emission wavelength of 530 nm). The geometrical mean of the fluorescence intensity was analyzed by CellQuest software.
2.4. Anion permeability and Cl-dependent fluorescence Cells were loaded with the Cl-selective fluorescence dye (MQAE; Molecular Probes, Leiden, Netherlands) in KCl solution (1 mM for 30 min) containing (in mM) 130 KCl, 32 HEPES/KOH, pH 7.4, 5 D -glucose, 1 MgSO4, 1 CaCl2, washed in KCl solution, and resus-
2.5. Immunoblotting Lysis, SDS–PAGE, and immuno blotting of K562 cells transfected with siRNA expression cassettes and control cells were performed as described [5] using a rabbit polyclonal anti-ClC-3 antibody against the C-terminus (Alomone Labs, Munich, Germany, 1:200 dilution in
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
5409
Fig. 2. ClC-3 RNA interference downregulates ClC-3 protein and blunts the oxidation-stimulated anion permeability. (A,B) Downregulation of ClC-3 protein expression by ClC-3-specific RNA interference. Immunoblot (A) and quantitative analysis (B; n = 5) of ClC-3 protein expression following 48 h treatment with ClC-3-specific siRNA (constructs 1 and 2), or with vehicle (control). (C) Current tracings recorded from an oxidized control cell (transfected with the non-specific construct siNEG; upper line) and an oxidized siClC-3-1-transfected K562 cell (lower line) with NMDGCl (left) and NMDG-gluconate bath solution (right) combined with NMDG-Cl pipette solution. (D) Mean slope conductance of siNeg- (left; n = 4) and siClC-3-transfected (right; n = 6) oxidized K562-cells recorded as in (C, left) with NMDG-Cl bath solution. (E) FACS histogram showing the MAQE-fluorescence in K-lactate solution of oxidized K562-cells transfected with a ClC-3 siRNA construct (red line) or a negative control construct (blue line) and of non-transfected control cells (black line). (F) Mean normalized MAQE-fluorescence (n = 12) of oxidized K562 cells transfected with two different ClC-3-specific siRNA constructs 1 and 2, with a non-specific construct (siNEG), and with a vehicle control.
PBS/0.1% Tween/1% BSA) and secondary donkey anti-rabbit antibody (1: 2000 dilution) conjugated with horseradish peroxidase (Amersham, Freiburg, Germany). For the experiments shown in Fig. 3D, control and oxidized K562 cells were washed in NaCl solution and then incubated for 30 min with the membrane-impermeable BAC-SulphoNHS (Sigma Aldrich, Germany) at room temperature under slow agitation. The biotinylation was stopped by washing (three times in NaCl solution). Cells were lysed as described [5], diluted 1:3 with PBS, and centrifugated (2 min at 13 000 rpm). The supernatant was incubated for 5 h at 4 °C with streptavidin–agarose beads (Sigma Aldrich). The beads were washed (three times in lysis buffer), resuspended in 2-fold Laemmli buffer, and the protein was released from the beats by heating (10 min for 100 °C). 2.6. Immunofluorescence microscopy K562 cells were fixed with acetone/methanol, washed four times with PBS, blocked with 1% BSA in PBS (60 min), incubated with a rabbit polyclonal anti-ClC-3 antibody against the C-terminus (Alomone; 1:200 dilution in PBS/1% BSA) overnight at 4 °C, washed with PBS, and incubated with a secondary donkey anti-rabbit antibody (1:2000 dilution in PBS) conjugated with Alexa488 (Amersham) for 1 h at room temperature.
2.7. Statistics Data are arithmetic means ± S.E. *, **, and *** indicate significant difference with P 6 0.05, P 6 0.01, and P 6 0.001, respectively, as tested by two-tailed Welch corrected t-test or ANOVA, as appropriate.
3. Results To test for an oxidation-stimulated anion conductance, untreated and oxidized K562 cells were whole-cell recorded in voltage-clamp mode using Cl and the large impermeable cation NMDG+ in pipette and bath solution. Untreated cells exhibited very low whole-cell anion currents as shown in the original tracings in Fig. 1A (1st tracings) and the current–voltage (I/V) relationship in Fig. 1B (open circles and Fig. 1D, 1st bar). Replacement of Cl by gluconate in the bath (NMDGgluconate) decreased the outward currents (Fig. 1A, 2nd tracings and Fig. 1D, 2nd bar) and shifted the reversal potential of the I/V curve in the direction of the change of the Cl
5410
equilibrium potential (ECl > 100 mV) to about 50 mV (Fig. 1B, closed triangles) confirming the Cl selectivity of the recorded currents. Oxidation (1 mM t-BHP for 30 min) stimulated an increase in whole cell conductance (Fig. 1A, 3rd tracings, Fig. 1C, open circles, and Fig. 1D, 3rd bar). Similar to the untreated cells, replacement of bath Cl by gluconate decreased the outward current of oxidized cells (Fig. 1A, 4th tracings and Fig. 1D, 4th bar) and shifted the reversal potential (Fig. 1C, closed triangles) indicating an oxidation-stimulated anion current. Oxidation with an additional oxidant (1 mM H2O2 for 30 min) similarly increased the whole cell conductance to 117 ± 16 pS/pF (n = 6) when recorded with NMDG-Cl in bath and pipette solution (data not shown). To further characterize the oxidation-stimulated anion conductance, K562 cells were loaded with the Cl selective dye MQAE and re-suspended in Cl free solutions containing the K+ salt of NO 3 , lactate or gluconate . The decrease of the cytosolic Cl concentration (as indicated by an increase in MQAE fluorescence intensity) was a direct function of the influx of the bath anion and was used to estimate the relative permeabilities of the anions administered in the bathing solutions. Suspending untreated K562 cells in Cl-free solution did not result in a fluorescence increase, and the fluorescence intensities did not differ between the various bathing solutions (Fig. 1F, closed bars). Oxidation of K562 cells (1 mM t-BHP for 30 min), in contrast, led to a fluorescence increase in the majority of K562 cells upon suspension in Cl-free solutions (Figs. 1E and F, open bars). This increase was significantly larger in nitrate and lactate as compared to the gluconate solu-
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
tion (Figs. 1E and F, open bars) suggesting a permselectivity of the oxidation-stimulated permeability of NO 3 > lactate > gluconate . To study a putative role of ClC-3 in the oxidation-stimulated anion permeability, ClC-3 protein expression was downregulated by RNA interference. To this end, cells were transfected for 48 h with two linear DNA constructs each (constructs 1 and 2) encoding for hair pinned double-stranded RNAs specific for all known splice variants of ClC-3 mRNA. ClC-3 RNA interference significantly down-regulated ClC-3 protein expression (Figs. 2A and B) and blunted the oxidation-stimulated increase in whole-cell conductance (Figs. 2C and D) and MQAE fluorescence intensity of K562 cells (Figs. 2E and F), while transfection with a non-targeting control construct (siNEG) had no effect (Fig. 2F and compare 1st bar of Fig. 2D with 3rd bar of Fig. 1D). RNA interference decreased the whole cell conductance by about 55%, (Figs. 2D and E). This incomplete inhibition is probably due to the intermediate efficacy of the ClC-3 RNA interference as deduced from the only 40% downregulation of total ClC-3 protein content (Fig. 2B). The cellular distribution of ClC-3 protein was analyzed by immunofluorescence microscopy in untreated and oxidized (1 mM t-BHP for 30 min) K562 cells. In untreated cells, a diffuse ClC-3 immunofluorescence was limited to the cytoplasm as shown by the micrograph in Fig. 3A (left) and by the cross-section fluorescence profile in Fig. 3B (black line). Oxidation of K562 cells, however, induced a prominent immunofluorescence staining of the plasma membrane area in the micrographs (arrows in Fig. 3A, right) and the corresponding fluorescence maxima at both margins of the cross section profile in Fig. 3B (green line). Quantitative analysis of the
Fig. 3. Oxidation induces the translocation of ClC-3 into the plasma membrane. (A) ClC-3-specific immunofluorescence micrographs of control (left) and oxidized K562 cells (right). (B) Cross-section fluorescence profile of an untreated (black line) and an oxidized K562 cell (green line) as obtained by line scan of the fluorescence images. (C) Mean ratio of the ClC-3-specific fluorescence density (n = 8–10) between plasma membrane and cytoplasm of untreated (closed bars) and oxidized K562 cells. (D) Immunoblot of biotinylated surface proteins in non-oxidized (left lane) and oxidized (right lane) K562 cells. The biotinylated proteins were separated by SDS gel electrophoresis and probed against ClC-3.
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
cellular immunofluorescence distribution in untreated and oxidized K562 cells revealed a comparable staining of the cytoplasm but a more than fivefold higher staining of the plasma membrane area in oxidized than in untreated cells (Fig. 3C). To distinguish between a ClC-3 localization underneath the plasma membrane and an insertion into the membrane, intact control and oxidized (1 mM t-BHP for 30 min) cells were biotinylated, the biotinylated proteins enriched, separated by SDS–PAGE, blotted, and probed against ClC-3. The immunoblot in Fig. 3D clearly indicates oxidation-stimulated surface expression of ClC-3.
5411
membrane [15]. An oxidation-induced co-activation of the ClC-3-dependent anion permeability results in loss of KCl (and osmotically obliged H2O) and thus in cell shrinkage suggesting that the ClC-3-dependent anion permeability may generate apoptotic volume decrease. In addition, ClC-3 has been suggested to conduct superoxide anions and to be involved in paracrine ROS signaling [26]. Oxidation-stimulated surface expression of ClC-3 might amplify this signaling according to a feed-forward process. In conclusion, the present study provides several lines of evidence of an oxidation-stimulated anion permeability in human K562 cells which is dependent on the surface expression of ClC-3 protein.
4. Discussion In the present study, down-regulation of ClC-3 protein expression by RNA interference was paralleled by a decrease of the oxidation-stimulated whole-cell anion currents of about 50%. Probably, the inefficient inhibition of the current was due to the low efficacy of the ClC-3 protein down-regulation. However, the data cannot exclude the oxidation-stimulated activation of further, ClC-3-independent anion permeabilities. Nevertheless, the data suggest that ClC-3 is involved in the induction or generation of at least a fraction of the oxidation-stimulated anion permeability. Until now, plasma membrane expression and anion channel function of ClC-3 are discussed controversially. In several reports, ClC-3 has been shown to reside intracellularly (e.g. [8]). Other studies, in contrast, reported ClC-3 plasma membrane expression (e.g. [20]). Recently, ClC-3 protein has been demonstrated to traffic through the plasma membrane from where it is rapidly endocytosed [21]. In the present work, fluorescence microscopy revealed an intracellular localization of ClC-3 protein in K562 cells. Oxidation, however, increased the expression of ClC-3 protein in the plasma membrane suggesting that plasma membrane-inserted ClC-3 most probably exerted the observed effect on the oxidation-stimulated anion current. Several studies, imply a functional significance of ClC-3 for cell volume-sensitive anion currents (e.g. [22]) as evident from the inhibition of these currents by intracellulary dialyzed antiClC-3 antibodies or by ClC-3 mRNA targeting (RNA interference or antisense). For such ClC-3-dependent currents, permselectivities of Cl > I [23] as well as I > Cl (e.g. [24]) have been reported. These conflicting results clearly indicate that the presumed ClC-3 currents of the different studies were generated by different channel proteins which do not necessarily include ClC-3. Most strikingly, albeit identical biophysical properties, cell volume-sensitive anion currents of cells from Clcn3+/+ and Clcn3/ mice differ in their sensitivity to intracellular dialyzed anti-ClC-3 antibodies [25]. The fact that ClC-3-deficient cells express anion currents which are not dependent on ClC-3, while the anion currents in wild-type cells are dependent on ClC-3 [25] might suggest that ClC-3 is rather a channel regulator than a channel itself and that its function is replaced by another regulator protein in ClC-3-deficient cells. Oxidation not only affects anion currents in K562 cells but also activates the non-selective cation channel TRPM2 resulting in an increase of intracellular free Ca2+ concentration which may trigger apoptosis [19]. In K562 cells, elevated free Ca2+ concentrations activate SK4 K+ channels in the plasma
Acknowledgement: This study was supported by the Deutsche Forschungsgemeinschaft, Nr. La315/11-1/2 and Hu781/4-3.
References [1] Paravicini, T.M. and Touyz, R.M. (2006) Redox signaling in hypertension. Cardiovasc. Res. 71, 247–258. [2] Ulens, C. and Tytgat, J. (2000) Redox state dependency of HERGS631C channel pharmacology: relation to C-type inactivation. FEBS Lett. 474, 111–115. [3] Cowley, E.A. and Linsdell, P. (2002) Oxidant stress stimulates anion secretion from the human airway epithelial cell line Calu-3: implications for cystic fibrosis lung disease. J. Physiol. 543, 201– 209. [4] Huber, S.M. et al. (2004) Plasmodium induces swelling-activated ClC-2 anion channels in the host erythrocyte. J. Biol. Chem. 279, 41444–41452. [5] Tanneur, V. et al. (2006) Purinoceptors are involved in the induction of an osmolyte permeability in malaria-infected and oxidized human erythrocytes. Faseb J. 20, 133–135. [6] Dutzler, R. (2004) Structural basis for ion conduction and gating in ClC chloride channels. FEBS Lett. 564, 229–233. [7] Duan, D., Winter, C., Cowley, S., Hume, J.R. and Horowitz, B. (1997) Molecular identification of a volume-regulated chloride channel. Nature 390, 417–421. [8] Stobrawa, S.M. et al. (2001) Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to a loss of the hippocampus. Neuron 29, 185–196. [9] Scheel, O., Zdebik, A.A., Lourdel, S. and Jentsch, T.J. (2005) Voltage-dependent electrogenic chloride/proton exchange by endosomal CLC proteins. Nature 436, 424–427. [10] Dutzler, R. (2007) A structural perspective on ClC channel and transporter function. FEBS Lett. 581, 2839–2844. [11] Jentsch, T.J. (2007) Chloride and the endosomal–lysosomal pathway: emerging roles of CLC chloride transporters. J. Physiol. 578, 633–640. [12] Andersson, L.C., Jokinen, M. and Gahmberg, C.G. (1979) Induction of erythroid differentiation in the human leukaemia cell line K562. Nature 278, 364–365. [13] Foller, M., Kasinathan, R.S., Duranton, C., Wieder, T., Huber, S.M. and Lang, F. (2006) PGE2-induced apoptotic cell death in K562 human leukaemia cells. Cell Physiol. Biochem. 17, 201– 210. [14] Foller, M., Kasinathan, R.S., Koka, S., Lang, C., Shumilina, E., Birnbaumer, L., Lang, F. and Huber, S.M. (2008). TRPC6 contributes to the Ca2+ leak of human erythrocytes. Cell Physiol. Biochem., in press. [15] Lang, P.A., Kaiser, S., Myssina, S., Wieder, T., Lang, F. and Huber, S.M. (2003) Role of Ca2+-activated K+ channels in human erythrocyte apoptosis. Am. J. Physiol. Cell Physiol. 285, C1553–C1560. [16] Barry, P.H. and Lynch, J.W. (1991) Liquid junction potentials and small cell effects in patch-clamp analysis. J. Membr. Biol. 121, 101–117. [17] Lindau, M. and Neher, E. (1988) Patch-clamp techniques for time-resolved capacitance measurements in single cells. Pflugers Arch. 411, 137–146.
5412 [18] Duranton, C., Huber, S., Tanneur, V., Lang, K., Brand, V., Sandu, C. and Lang, F. (2003) Electrophysiological properties of the Plasmodium falciparum-induced cation conductance of human erythrocytes. Cell Physiol. Biochem. 13, 189–198. [19] Zhang, W. et al. (2006) TRPM2 is an ion channel that modulates hematopoietic cell death through activation of caspases and PARP cleavage. Am. J. Physiol. Cell Physiol. 290, C1146– C1159. [20] Huang, P., Liu, J., Di, A., Robinson, N.C., Musch, M.W., Kaetzel, M.A. and Nelson, D.J. (2001) Regulation of human CLC-3 channels by multifunctional Ca2+/calmodulin-dependent protein kinase. J. Biol. Chem. 276, 20093–20100. [21] Zhao, Z., Li, X., Hao, J., Winston, J.H. and Weinman, S.A. (2007) The ClC-3 chloride transport protein traffics through the plasma membrane via interaction of an N-terminal dileucine cluster with clathrin. J. Biol. Chem. 282, 29022–29031. [22] Zhou, J.G., Ren, J.L., Qiu, Q.Y., He, H. and Guan, Y.Y. (2005) Regulation of intracellular Cl-concentration through volume-
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
[23]
[24]
[25]
[26]
regulated ClC-3 chloride channels in A10 vascular smooth muscle cells. J. Biol. Chem. 280, 7301–7308. Li, X., Shimada, K., Showalter, L.A. and Weinman, S.A. (2000) Biophysical properties of ClC-3 differentiate it from swellingactivated chloride channels in Chinese hamster ovary-K1 cells. J. Biol. Chem. 275, 35994–35998. Jin, N.G. et al. (2003) Fundamental role of ClC-3 in volumesensitive Cl- channel function and cell volume regulation in AGS cells. Am. J. Physiol. Gastrointest. Liver Physiol. 285, G938– G948. Yamamoto-Mizuma, S. et al. (2004) Altered properties of volume-sensitive osmolyte and anion channels (VSOACs) and membrane protein expression in cardiac and smooth muscle myocytes from Clcn3/ mice. J. Physiol. 557, 439–456. Hawkins, B.J., Madesh, M., Kirkpatrick, C.J. and Fisher, A.B. (2007) Superoxide flux in endothelial cells via the chloride channel-3 mediates intracellular signaling. Mol. Biol. Cell 18, 2002–2012.
Oxidation induces ClC-3-dependent anion channels in human leukaemia cells Ravi S. Kasinathan, Michael Fo¨ller, Camelia Lang, Saisudha Koka, Florian Lang, Stephan M. Huber* Department of Physiology I, Eberhard-Karls-University Tu¨bingen, Gmelinstraße 5, D-72076 Tu¨bingen, Germany Received 27 June 2007; revised 17 October 2007; accepted 17 October 2007 Available online 30 October 2007 Edited by Vladimir Skulachev
Abstract To test for redox regulation of anion channels in erythroid cells, we exposed K562 cells to oxidants and measured changes in transmembrane Cl currents using patch-clamp, and in intracellular Cl content using the Cl selective dye MQAE. Oxidation with tert-butylhydroperoxide or H2O2 produced a plasma membrane anion permeability with a permse lectivity of NO 3 > lactate > gluconate . The permeability increase was paralleled by insertion of ClC-3 protein into the plasma membrane as evident from immunofluorescence microscopy and surface biotinylation. Down-regulation of ClC-3 protein by RNA interference as assessed by immunoblotting decreased the oxidation-stimulated permeability. In conclusion, oxidation induces surface expression of ClC-3 and activation of a ClC-3-dependent anion permeability in K562 cells. Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: K562; Clcn3; Cl channels; MQAE-fluorescence; Flow-cytometry; RNA interference; Patch-clamp recording; Oxidative stress; Surface expression
sensitive organic osmolyte and anion channel [7]. In contrast to this assumption, anion channel activity is seemingly normal in cells from gene targeted mice lacking functional ClC-3 (Clcn3/ mice) [8]. Moreover, ClC-4 and -5 have recently been shown to function as electrogenic Cl/H+ exchangers [9,10] and such a function has also been suggested for ClC-3 [11]. The present study aimed to characterize redox-sensitive organic osmolyte and anion channels by use of whole-cell patch-clamp recording, flow cytometry, immunofluorescence microscopy and immunoblotting. The K562 myeloid leukaemia cell line shares many characteristics with erythrocytes [12]. In particular, its expression pattern of ion channels in the plasma membrane is similar to that of human erythrocytes [13–15]. K562 cells were, therefore, chosen as an erythroblastlike cell model to study redox-sensitive changes in plasma membrane permeability with specific emphasis on ClC-3 as a candidate for a redox-sensitive anion channel/regulator.
2. Materials and methods 1. Introduction Modulation of the redox potential, e.g. through formation of ROS regulates numerous cellular processes through oxidative interference with signaling molecules such as tyrosine kinases, tyrosine phosphatases, MAP kinases, through activation of transcription factors [1], or through direct modification of effector proteins [2]. Anion channels regulated by redox modulation accomplish Cl secretion [3] or volume changes of malaria-infected erythrocytes by mediating the efflux of Cl [4] and/or organic osmolytes [5]. ClC-3, a member of the voltage-gated ClC family of Cl channels [6], has been suggested initially to be a cell volume-
* Corresponding author. Fax: +49 (0) 7071 29 3073. E-mail address: [email protected] (S.M. Huber).
Abbreviations: ClC-3, member 3 of the cloned chloride channel family ClC; EGTA, ethylene glycol-bis(b-aminoethyl ether)-N,N,N 0 ,N 0 -tetraacetic acid; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; MAP, mitogen-activated protein; MQAE, N-(ethoxycarbonylmethyl)-6-methoxyquinolinium bromide; NMDG, N-methyl-D -glucamine; PBS, phosphate-buffered saline; ROS, reactive oxygen species; SDS, sodium dodecyl sulfate; siRNA, small interference RNA; t-BHP, tert-butylhydroperoxide
2.1. Cell culture and oxidation of K562 Cells K562 human myeloid leukaemia cells were grown in RPMI1640 medium containing 10% FCS. For oxidation, cells were washed, resuspended in NaCl solution containing (in mM) 125 NaCl, 32 HEPES/ NaOH, pH 7.4, 5 KCl, 5 D -glucose, 1 MgSO4, 1 CaCl2, and incubated in the presence or absence of 1 mM t-BHP for 30 min at 37 °C. In few patch-clamp experiments, cells were oxidized with H2O2 (1 mM for 30 min at 37 °C). Oxidation was stopped by washing the cells in NaCl solution. 2.2. siRNA based silencing of Clcn3 siRNA expression cassettes (H1 promoter) were prepared by a PCRbased method with the Silencerä Express Kit from Ambion (Cambridgeshire, UK). The following PCR primers (in parenthesis) were used for the constructs Clcn3-1 (sense: AGCTACACAAACTGCTTGACCTATGATTAACCGGTGTTTCGTCCTTTCCACAAG; antisense: CGGCGAAGCTTTTTCCAAAAAATTAATCATAGGTCAAGCAGCTACACAAACTGC; the targeted sequence as indicated by the bold letters is at position 1095–1113, accession number: NM_ 001829), Clcn3-2 (sense: CCTCTACACAAAAGGTTAAAGAGTTGTGCTACCGGTTCTTCCTTTCCACAAG; antisense: CGGC GAAGCTTTTTCCAAAAAATAGCACAACTCTTTAACCTCTACACAAAAGGT; the targeted sequence is at position: 3019–3037, NM_001829), and a nonsense construct (sense: CATCTACACAAAATGTGAAGAAGGGTTTATCCGGTGTTTCGTCCTTTCCACAAG; antisense: CGGCGAAGCTTTTTCCAAAAAAGATAAACCCTTCTTCACATCTACACAAAATGT). K562 cells (50 000 cells in 1 ml medium) were transfected for 48 h by adding siRNA expression cassettes (1 lg) mixed with siPORT XP-1 transfection agent (7 ll; Ambion) in Opti Mem Medium (200 ll; Gibco).
0014-5793/$32.00 Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.10.042
5408
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
Fig. 1. Whole-cell Cl currents of control and oxidized K562 cells. (A) Current traces recorded from a control (left panels) and an oxidized (right panels) K562 cell first with NMDG-Cl (left) and then NMDG-gluconate in the bath (right). (B,C) Mean current–voltage relationships (n = 7–8) of control (B) and oxidized K562 cells (C) recorded as in (A) with NMDG-Cl (open circles) and NMDG-gluconate (closed triangles) bath solution. Voltages refer to the intracellular membrane face with respect to the earthed bath solution. (D) Mean slope conductances (n = 7–8) as calculated from (B–C) for the outward currents of control (left) and oxidized K562 cells (right) in MMDG-Cl (open bars) and NMDG-gluconate (closed bars). (E) FACS histograms showing the Cl-specific MQAE-fluorescence of oxidized K562 cells incubated in K-gluconate (left) and KNO3 solution (right) for 0 min (black line) and 45 min (red line). (F) Mean normalized MQAE-fluorescence intensity (n = 5–12) of control (closed bars) and oxidized K562 cells (open bars) incubated for 45 min in K+ salts of different anions. Oxidized K562 cells show a chloride efflux which is promoted by bath anions with a permeability sequence of NO 3 > lactate > gluconate . 2.3. Patch-clamp experiments Currents were recorded in fast whole-cell mode as described [5] under conditions, where the cells did not swell. A pipette solution containing (in mM) 140 NMDG-Cl (pH 7.4), 40 mannitol, 10 HEPES/NMDG, 1 MgCl2, 1 Mg-ATP, 0.5 EGTA, was combined with a bath solution containing 180 NMDG-Cl (pH 7.4), 10 HEPES/ NMDG, 1 CaCl2, 1 MgCl2. In an additional bath solution (NMDGgluconate) chloride was replaced isosmotically by D -gluconate. Liquid junction potentials were recorded according to [16] and the data corrected. Prior to compensation, slow capacitive currents were evoked by a l0-mV square pulse, the capacitive current transients fitted by exponential regression and the whole-cell membrane capacitances calculated as described previously [17]. The recorded membrane capacitances did not differ between control (8.9 ± 0.5 pF; n = 8) and oxidized cells (8.0 ± 0.9 pF, n = 13).
pended in Cl-free solutions in which KCl was isosmotically replaced by KNO3, K-L -lactate, or K-D -gluconate. The solution further contained a cation channel blocker cocktail (50 lM clotrimazole to inhibit SK4 K+ channels [15], 50 lM ethylisopropylamiloride, EIPA, to inhibit non-selective cation channels [13,14,18,19], 50 lM 4-aminopyridine and 1 mM BaCl2 to inhibit further K+ channels) to avoid Clefflux-driven cell shrinkage by parallel loss of cellular K+ and H2O. Under these conditions, Cl efflux was limited by, and, therefore, a direct function of the uptake of the bath anion. Cl efflux was assessed after 45 min of incubation in Cl-free media. Cl efflux was indicated in flow cytometry (FACS-Calibur from Becton Dickinson, Heidelberg, Germany) by an increase in the Cl-specific MQAE fluorescence intensity as recorded in fluorescence channel FL-1 (excitation wavelength of 488 nm, emission wavelength of 530 nm). The geometrical mean of the fluorescence intensity was analyzed by CellQuest software.
2.4. Anion permeability and Cl-dependent fluorescence Cells were loaded with the Cl-selective fluorescence dye (MQAE; Molecular Probes, Leiden, Netherlands) in KCl solution (1 mM for 30 min) containing (in mM) 130 KCl, 32 HEPES/KOH, pH 7.4, 5 D -glucose, 1 MgSO4, 1 CaCl2, washed in KCl solution, and resus-
2.5. Immunoblotting Lysis, SDS–PAGE, and immuno blotting of K562 cells transfected with siRNA expression cassettes and control cells were performed as described [5] using a rabbit polyclonal anti-ClC-3 antibody against the C-terminus (Alomone Labs, Munich, Germany, 1:200 dilution in
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
5409
Fig. 2. ClC-3 RNA interference downregulates ClC-3 protein and blunts the oxidation-stimulated anion permeability. (A,B) Downregulation of ClC-3 protein expression by ClC-3-specific RNA interference. Immunoblot (A) and quantitative analysis (B; n = 5) of ClC-3 protein expression following 48 h treatment with ClC-3-specific siRNA (constructs 1 and 2), or with vehicle (control). (C) Current tracings recorded from an oxidized control cell (transfected with the non-specific construct siNEG; upper line) and an oxidized siClC-3-1-transfected K562 cell (lower line) with NMDGCl (left) and NMDG-gluconate bath solution (right) combined with NMDG-Cl pipette solution. (D) Mean slope conductance of siNeg- (left; n = 4) and siClC-3-transfected (right; n = 6) oxidized K562-cells recorded as in (C, left) with NMDG-Cl bath solution. (E) FACS histogram showing the MAQE-fluorescence in K-lactate solution of oxidized K562-cells transfected with a ClC-3 siRNA construct (red line) or a negative control construct (blue line) and of non-transfected control cells (black line). (F) Mean normalized MAQE-fluorescence (n = 12) of oxidized K562 cells transfected with two different ClC-3-specific siRNA constructs 1 and 2, with a non-specific construct (siNEG), and with a vehicle control.
PBS/0.1% Tween/1% BSA) and secondary donkey anti-rabbit antibody (1: 2000 dilution) conjugated with horseradish peroxidase (Amersham, Freiburg, Germany). For the experiments shown in Fig. 3D, control and oxidized K562 cells were washed in NaCl solution and then incubated for 30 min with the membrane-impermeable BAC-SulphoNHS (Sigma Aldrich, Germany) at room temperature under slow agitation. The biotinylation was stopped by washing (three times in NaCl solution). Cells were lysed as described [5], diluted 1:3 with PBS, and centrifugated (2 min at 13 000 rpm). The supernatant was incubated for 5 h at 4 °C with streptavidin–agarose beads (Sigma Aldrich). The beads were washed (three times in lysis buffer), resuspended in 2-fold Laemmli buffer, and the protein was released from the beats by heating (10 min for 100 °C). 2.6. Immunofluorescence microscopy K562 cells were fixed with acetone/methanol, washed four times with PBS, blocked with 1% BSA in PBS (60 min), incubated with a rabbit polyclonal anti-ClC-3 antibody against the C-terminus (Alomone; 1:200 dilution in PBS/1% BSA) overnight at 4 °C, washed with PBS, and incubated with a secondary donkey anti-rabbit antibody (1:2000 dilution in PBS) conjugated with Alexa488 (Amersham) for 1 h at room temperature.
2.7. Statistics Data are arithmetic means ± S.E. *, **, and *** indicate significant difference with P 6 0.05, P 6 0.01, and P 6 0.001, respectively, as tested by two-tailed Welch corrected t-test or ANOVA, as appropriate.
3. Results To test for an oxidation-stimulated anion conductance, untreated and oxidized K562 cells were whole-cell recorded in voltage-clamp mode using Cl and the large impermeable cation NMDG+ in pipette and bath solution. Untreated cells exhibited very low whole-cell anion currents as shown in the original tracings in Fig. 1A (1st tracings) and the current–voltage (I/V) relationship in Fig. 1B (open circles and Fig. 1D, 1st bar). Replacement of Cl by gluconate in the bath (NMDGgluconate) decreased the outward currents (Fig. 1A, 2nd tracings and Fig. 1D, 2nd bar) and shifted the reversal potential of the I/V curve in the direction of the change of the Cl
5410
equilibrium potential (ECl > 100 mV) to about 50 mV (Fig. 1B, closed triangles) confirming the Cl selectivity of the recorded currents. Oxidation (1 mM t-BHP for 30 min) stimulated an increase in whole cell conductance (Fig. 1A, 3rd tracings, Fig. 1C, open circles, and Fig. 1D, 3rd bar). Similar to the untreated cells, replacement of bath Cl by gluconate decreased the outward current of oxidized cells (Fig. 1A, 4th tracings and Fig. 1D, 4th bar) and shifted the reversal potential (Fig. 1C, closed triangles) indicating an oxidation-stimulated anion current. Oxidation with an additional oxidant (1 mM H2O2 for 30 min) similarly increased the whole cell conductance to 117 ± 16 pS/pF (n = 6) when recorded with NMDG-Cl in bath and pipette solution (data not shown). To further characterize the oxidation-stimulated anion conductance, K562 cells were loaded with the Cl selective dye MQAE and re-suspended in Cl free solutions containing the K+ salt of NO 3 , lactate or gluconate . The decrease of the cytosolic Cl concentration (as indicated by an increase in MQAE fluorescence intensity) was a direct function of the influx of the bath anion and was used to estimate the relative permeabilities of the anions administered in the bathing solutions. Suspending untreated K562 cells in Cl-free solution did not result in a fluorescence increase, and the fluorescence intensities did not differ between the various bathing solutions (Fig. 1F, closed bars). Oxidation of K562 cells (1 mM t-BHP for 30 min), in contrast, led to a fluorescence increase in the majority of K562 cells upon suspension in Cl-free solutions (Figs. 1E and F, open bars). This increase was significantly larger in nitrate and lactate as compared to the gluconate solu-
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
tion (Figs. 1E and F, open bars) suggesting a permselectivity of the oxidation-stimulated permeability of NO 3 > lactate > gluconate . To study a putative role of ClC-3 in the oxidation-stimulated anion permeability, ClC-3 protein expression was downregulated by RNA interference. To this end, cells were transfected for 48 h with two linear DNA constructs each (constructs 1 and 2) encoding for hair pinned double-stranded RNAs specific for all known splice variants of ClC-3 mRNA. ClC-3 RNA interference significantly down-regulated ClC-3 protein expression (Figs. 2A and B) and blunted the oxidation-stimulated increase in whole-cell conductance (Figs. 2C and D) and MQAE fluorescence intensity of K562 cells (Figs. 2E and F), while transfection with a non-targeting control construct (siNEG) had no effect (Fig. 2F and compare 1st bar of Fig. 2D with 3rd bar of Fig. 1D). RNA interference decreased the whole cell conductance by about 55%, (Figs. 2D and E). This incomplete inhibition is probably due to the intermediate efficacy of the ClC-3 RNA interference as deduced from the only 40% downregulation of total ClC-3 protein content (Fig. 2B). The cellular distribution of ClC-3 protein was analyzed by immunofluorescence microscopy in untreated and oxidized (1 mM t-BHP for 30 min) K562 cells. In untreated cells, a diffuse ClC-3 immunofluorescence was limited to the cytoplasm as shown by the micrograph in Fig. 3A (left) and by the cross-section fluorescence profile in Fig. 3B (black line). Oxidation of K562 cells, however, induced a prominent immunofluorescence staining of the plasma membrane area in the micrographs (arrows in Fig. 3A, right) and the corresponding fluorescence maxima at both margins of the cross section profile in Fig. 3B (green line). Quantitative analysis of the
Fig. 3. Oxidation induces the translocation of ClC-3 into the plasma membrane. (A) ClC-3-specific immunofluorescence micrographs of control (left) and oxidized K562 cells (right). (B) Cross-section fluorescence profile of an untreated (black line) and an oxidized K562 cell (green line) as obtained by line scan of the fluorescence images. (C) Mean ratio of the ClC-3-specific fluorescence density (n = 8–10) between plasma membrane and cytoplasm of untreated (closed bars) and oxidized K562 cells. (D) Immunoblot of biotinylated surface proteins in non-oxidized (left lane) and oxidized (right lane) K562 cells. The biotinylated proteins were separated by SDS gel electrophoresis and probed against ClC-3.
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
cellular immunofluorescence distribution in untreated and oxidized K562 cells revealed a comparable staining of the cytoplasm but a more than fivefold higher staining of the plasma membrane area in oxidized than in untreated cells (Fig. 3C). To distinguish between a ClC-3 localization underneath the plasma membrane and an insertion into the membrane, intact control and oxidized (1 mM t-BHP for 30 min) cells were biotinylated, the biotinylated proteins enriched, separated by SDS–PAGE, blotted, and probed against ClC-3. The immunoblot in Fig. 3D clearly indicates oxidation-stimulated surface expression of ClC-3.
5411
membrane [15]. An oxidation-induced co-activation of the ClC-3-dependent anion permeability results in loss of KCl (and osmotically obliged H2O) and thus in cell shrinkage suggesting that the ClC-3-dependent anion permeability may generate apoptotic volume decrease. In addition, ClC-3 has been suggested to conduct superoxide anions and to be involved in paracrine ROS signaling [26]. Oxidation-stimulated surface expression of ClC-3 might amplify this signaling according to a feed-forward process. In conclusion, the present study provides several lines of evidence of an oxidation-stimulated anion permeability in human K562 cells which is dependent on the surface expression of ClC-3 protein.
4. Discussion In the present study, down-regulation of ClC-3 protein expression by RNA interference was paralleled by a decrease of the oxidation-stimulated whole-cell anion currents of about 50%. Probably, the inefficient inhibition of the current was due to the low efficacy of the ClC-3 protein down-regulation. However, the data cannot exclude the oxidation-stimulated activation of further, ClC-3-independent anion permeabilities. Nevertheless, the data suggest that ClC-3 is involved in the induction or generation of at least a fraction of the oxidation-stimulated anion permeability. Until now, plasma membrane expression and anion channel function of ClC-3 are discussed controversially. In several reports, ClC-3 has been shown to reside intracellularly (e.g. [8]). Other studies, in contrast, reported ClC-3 plasma membrane expression (e.g. [20]). Recently, ClC-3 protein has been demonstrated to traffic through the plasma membrane from where it is rapidly endocytosed [21]. In the present work, fluorescence microscopy revealed an intracellular localization of ClC-3 protein in K562 cells. Oxidation, however, increased the expression of ClC-3 protein in the plasma membrane suggesting that plasma membrane-inserted ClC-3 most probably exerted the observed effect on the oxidation-stimulated anion current. Several studies, imply a functional significance of ClC-3 for cell volume-sensitive anion currents (e.g. [22]) as evident from the inhibition of these currents by intracellulary dialyzed antiClC-3 antibodies or by ClC-3 mRNA targeting (RNA interference or antisense). For such ClC-3-dependent currents, permselectivities of Cl > I [23] as well as I > Cl (e.g. [24]) have been reported. These conflicting results clearly indicate that the presumed ClC-3 currents of the different studies were generated by different channel proteins which do not necessarily include ClC-3. Most strikingly, albeit identical biophysical properties, cell volume-sensitive anion currents of cells from Clcn3+/+ and Clcn3/ mice differ in their sensitivity to intracellular dialyzed anti-ClC-3 antibodies [25]. The fact that ClC-3-deficient cells express anion currents which are not dependent on ClC-3, while the anion currents in wild-type cells are dependent on ClC-3 [25] might suggest that ClC-3 is rather a channel regulator than a channel itself and that its function is replaced by another regulator protein in ClC-3-deficient cells. Oxidation not only affects anion currents in K562 cells but also activates the non-selective cation channel TRPM2 resulting in an increase of intracellular free Ca2+ concentration which may trigger apoptosis [19]. In K562 cells, elevated free Ca2+ concentrations activate SK4 K+ channels in the plasma
Acknowledgement: This study was supported by the Deutsche Forschungsgemeinschaft, Nr. La315/11-1/2 and Hu781/4-3.
References [1] Paravicini, T.M. and Touyz, R.M. (2006) Redox signaling in hypertension. Cardiovasc. Res. 71, 247–258. [2] Ulens, C. and Tytgat, J. (2000) Redox state dependency of HERGS631C channel pharmacology: relation to C-type inactivation. FEBS Lett. 474, 111–115. [3] Cowley, E.A. and Linsdell, P. (2002) Oxidant stress stimulates anion secretion from the human airway epithelial cell line Calu-3: implications for cystic fibrosis lung disease. J. Physiol. 543, 201– 209. [4] Huber, S.M. et al. (2004) Plasmodium induces swelling-activated ClC-2 anion channels in the host erythrocyte. J. Biol. Chem. 279, 41444–41452. [5] Tanneur, V. et al. (2006) Purinoceptors are involved in the induction of an osmolyte permeability in malaria-infected and oxidized human erythrocytes. Faseb J. 20, 133–135. [6] Dutzler, R. (2004) Structural basis for ion conduction and gating in ClC chloride channels. FEBS Lett. 564, 229–233. [7] Duan, D., Winter, C., Cowley, S., Hume, J.R. and Horowitz, B. (1997) Molecular identification of a volume-regulated chloride channel. Nature 390, 417–421. [8] Stobrawa, S.M. et al. (2001) Disruption of ClC-3, a chloride channel expressed on synaptic vesicles, leads to a loss of the hippocampus. Neuron 29, 185–196. [9] Scheel, O., Zdebik, A.A., Lourdel, S. and Jentsch, T.J. (2005) Voltage-dependent electrogenic chloride/proton exchange by endosomal CLC proteins. Nature 436, 424–427. [10] Dutzler, R. (2007) A structural perspective on ClC channel and transporter function. FEBS Lett. 581, 2839–2844. [11] Jentsch, T.J. (2007) Chloride and the endosomal–lysosomal pathway: emerging roles of CLC chloride transporters. J. Physiol. 578, 633–640. [12] Andersson, L.C., Jokinen, M. and Gahmberg, C.G. (1979) Induction of erythroid differentiation in the human leukaemia cell line K562. Nature 278, 364–365. [13] Foller, M., Kasinathan, R.S., Duranton, C., Wieder, T., Huber, S.M. and Lang, F. (2006) PGE2-induced apoptotic cell death in K562 human leukaemia cells. Cell Physiol. Biochem. 17, 201– 210. [14] Foller, M., Kasinathan, R.S., Koka, S., Lang, C., Shumilina, E., Birnbaumer, L., Lang, F. and Huber, S.M. (2008). TRPC6 contributes to the Ca2+ leak of human erythrocytes. Cell Physiol. Biochem., in press. [15] Lang, P.A., Kaiser, S., Myssina, S., Wieder, T., Lang, F. and Huber, S.M. (2003) Role of Ca2+-activated K+ channels in human erythrocyte apoptosis. Am. J. Physiol. Cell Physiol. 285, C1553–C1560. [16] Barry, P.H. and Lynch, J.W. (1991) Liquid junction potentials and small cell effects in patch-clamp analysis. J. Membr. Biol. 121, 101–117. [17] Lindau, M. and Neher, E. (1988) Patch-clamp techniques for time-resolved capacitance measurements in single cells. Pflugers Arch. 411, 137–146.
5412 [18] Duranton, C., Huber, S., Tanneur, V., Lang, K., Brand, V., Sandu, C. and Lang, F. (2003) Electrophysiological properties of the Plasmodium falciparum-induced cation conductance of human erythrocytes. Cell Physiol. Biochem. 13, 189–198. [19] Zhang, W. et al. (2006) TRPM2 is an ion channel that modulates hematopoietic cell death through activation of caspases and PARP cleavage. Am. J. Physiol. Cell Physiol. 290, C1146– C1159. [20] Huang, P., Liu, J., Di, A., Robinson, N.C., Musch, M.W., Kaetzel, M.A. and Nelson, D.J. (2001) Regulation of human CLC-3 channels by multifunctional Ca2+/calmodulin-dependent protein kinase. J. Biol. Chem. 276, 20093–20100. [21] Zhao, Z., Li, X., Hao, J., Winston, J.H. and Weinman, S.A. (2007) The ClC-3 chloride transport protein traffics through the plasma membrane via interaction of an N-terminal dileucine cluster with clathrin. J. Biol. Chem. 282, 29022–29031. [22] Zhou, J.G., Ren, J.L., Qiu, Q.Y., He, H. and Guan, Y.Y. (2005) Regulation of intracellular Cl-concentration through volume-
R.S. Kasinathan et al. / FEBS Letters 581 (2007) 5407–5412
[23]
[24]
[25]
[26]
regulated ClC-3 chloride channels in A10 vascular smooth muscle cells. J. Biol. Chem. 280, 7301–7308. Li, X., Shimada, K., Showalter, L.A. and Weinman, S.A. (2000) Biophysical properties of ClC-3 differentiate it from swellingactivated chloride channels in Chinese hamster ovary-K1 cells. J. Biol. Chem. 275, 35994–35998. Jin, N.G. et al. (2003) Fundamental role of ClC-3 in volumesensitive Cl- channel function and cell volume regulation in AGS cells. Am. J. Physiol. Gastrointest. Liver Physiol. 285, G938– G948. Yamamoto-Mizuma, S. et al. (2004) Altered properties of volume-sensitive osmolyte and anion channels (VSOACs) and membrane protein expression in cardiac and smooth muscle myocytes from Clcn3/ mice. J. Physiol. 557, 439–456. Hawkins, B.J., Madesh, M., Kirkpatrick, C.J. and Fisher, A.B. (2007) Superoxide flux in endothelial cells via the chloride channel-3 mediates intracellular signaling. Mol. Biol. Cell 18, 2002–2012.