- Email: [email protected]
Transient increases in extracellular K+ produce two pharmacological distinct cytosolic Ca2+ transients
Brain Research 1031 (2005) 174 – 184 www.elsevier.com/locate/brainres
Research report
Transient increases in extracellular K+ produce two pharmacological distinct cytosolic Ca2+ transients Alexandra Corrales, Jose´ V. Montoya G., Jhon-Jairo Sutachan, Genoveve Cornillez-Ty, Zayra Garavito-Aguilar, Fang Xu, Thomas J.J. Blanck, Esperanza Recio-Pinto* Anesthesiology Department, New York University School of Medicine, 550 First Avenue, RR605, New York, NY 10016, USA Accepted 28 October 2004 Available online 16 December 2004
Abstract Transient increases in extracellular K+ are observed under various conditions, including repetitive neuronal firing, anoxia, ischemia and hypoglycemic coma. We studied changes in cytoplasmic Ca2+ ([Ca2+]cyt) evoked by pulses of KCl in human neuroblastoma SH-SY5Y cells and rat dorsal root ganglia (DRG) neurons at 37 8C. A bpulseQ of KCl evoked two transient increases in [Ca2+]cyt, one upon addition of KCl (K+on) and the other upon removal of KCl (K+off). The K+on transient has been described in many cell types and is initiated by the activation of voltage-dependent Ca2+ channels followed by Ca2+-evoked Ca2+ release from intracellular Ca2+ stores. The level of KCl necessary to evoke the K+off transient depends on the type of neuron, in SH-SY5Y cells it required 100 mM KCl, in most (but not all) of dorsal root ganglia neurons it could be detected with 100–200 mM KCl and in a very few dorsal root ganglia neurons it was detectable at 20–50 mM KCl. In SHSY5Y cells, reduction of extracellular Ca2+ inhibited the K+on more strongly than the K+off and slowed the decay of K+off. Isoflurane (1 mM) reduced the K+on- but not the K+off-peak. However, isoflurane slowed the decay of K+off. The nonspecific cationic channel blocker La3+ (100 AM) had an effect similar to that of isoflurane. Treatment with thapsigargin (TG) at a concentration known to only deplete IP3-sensitive Ca2+ stores did not affect K+on or K+off, suggesting that Ca2+ release from the IP3-sensitive Ca2+ stores does not contribute to K+on and K+off transients and that the thapsigargin-sensitive Ca2+ ATPases do not contribute significantly to the rise or decay rates of these transients. These findings indicate that a pulse of extracellular K+ produces two distinct transient increases in [Ca2+]cyt. D 2004 Elsevier B.V. All rights reserved. Theme: C Topic: Other ion channels Keywords: Cytoplasmic Ca2+; Extracellular K+; Neuroblastoma; Dorsal root ganglia neurons
1. Introduction The aim of this study was to identify and characterize the changes in cytoplasmic Ca2+ ([Ca2+]cyt) produced by a pulse of extracellular potassium ([K+]e). In vivo, neurons are exposed to transient increases of [K+]e during intensive repetitive electrical firing [9,12,13,34]. These increases range between 7 and 17 mM and appear to enhance excitability of normal and regenerating nerves [11,26].
* Corresponding author. Fax: +1 212 2636139. E-mail address: [email protected] (E. Recio-Pinto). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.10.031
Under some pathological conditions such as anoxia, focal ischemia and hypoglycemic coma, [K+]e can locally increase up to 60–80 mM [2,15,27,32,42], and such large increase in [K+]e appears to be responsible for the initiation of the phenomenon known as cortical spreading depression that consists in a propagating transient suppression of electrical activity associated with cellular depolarization [6,24,25,27,30,36,42,45]. During cortical spreading depression, there can be periods of transient repolarization accompanied by transient decreases in [K+]e [15]. Experimentally, spreading depression can be evoked by the application of high K+ either to the surface of the tissue or by injection into the nervous tissue [40].
A. Corrales et al. / Brain Research 1031 (2005) 174–184
In previous studies, we observed that neuronal cells displayed two transient increases in [Ca2+]cyt when exposed to pulses of high KCl: one during the onset of KClapplication (K+on) and a second following removal of KCl (K+off) [10]. K+on has been described in many cell types, and it is triggered by a K+-mediated depolarization that leads to the activation of voltage-dependent Ca2+ channels and hence Ca2+-entry through these channels followed by Ca2+-evoked Ca2+ release from intracellular Ca2+ stores [5,20,41,46]. K+on also has been shown to be sensitive to volatile anesthetics [10,46]. This study has characterized, for the first time, the physiology and pharmacology of the K+off. We found that the trigger and pharmacology of K+on- and K+off-transients are different. Such differences should be taken into consideration when designing pharmacological approaches for reducing the pathological effects of transient increases in [K+]e observed in vivo.
2. Materials and methods 2.1. Cell culture and solutions SH-SY5Y human neuroblastoma cells were cultured in RPMI 1640 medium with l-glutamine, supplemented with penicillin (50 U/ml), streptomycin (50 Ag/ml), and 12% fetal bovine serum, at 37 8C, in a humidified atmosphere containing 5% CO2. All cell culture components were Gibco BRL products purchased from Life Technologies (Rockville, MD, USA). Experiments were performed on monolayer of cells as previously reported [10]. Cells were plated on glass coverslips (25-mm diameter) at a density of 2– 4 104 cells/ml (2 ml cell suspension/35 mm culture dish) and used when they formed a confluent monolayer (~10 –16 days after plating). Isolated adult rat dorsal root ganglia (DRG) sensory neurons were cultured on glass coverslips (25-mm diameter placed in 35 mm culture dishes) in DMEM high glucose medium supplemented with l-glutamine (2.2 mM), penicillin (100 U/ml), streptomycin (100 Ag/ml) and 10% heat-inactivated fetal bovine serum, at 37 8C, in a humidified atmosphere containing 5% CO2. The glass coverslips were pretreated with Poly-l-lysine (2 ml of 12.5 Ag/ml). Cervical DRG were isolated (1 rat/preparation) and dissociated by incubating them first in medium containing collagenase type I (2 mg/ml) and dispase (5 mg/ml) for 1 h at 37 8C, and then in medium containing only dispase (5 mg/ml) for 30 min at 37 8C. Mechanical dissociation of the cells was carried out by passing the enzymatically treated DRG several times through a Pasteur pipette. The cell suspension was centrifuged and the pellet resuspended in 600 Al of the culture medium; 10 Al of cell suspension was placed on each glass coverslip. The total volume of medium was 2 ml/dish. During experiments, the DRG cells were continuously perfused with a HEPES buffer containing (in mM) 140
175
NaCl, 5 KCl, 5 NaHCO3, 10 HEPES, 1 MgCl2, 1.5 CaCl2, and 10 glucose (pH 7.4). For SH-SY5Y cells, the HEPES buffer also contained 1 mM ATP. Experiments were performed at 37 8C and the temperature was controlled with a Dual Heater controller TC-344A and an inline heater SH-27B (Warner Instruments, Hamden, CT, USA). The exchange of the solution was carried out with a manifold. The solutions containing KCl (20–200 mM), with and without 100 AM La3+ and 100 nM thapsigargin, were prepared using the HEPES buffer. Saturated isoflurane (Ohmeda Caribe, Guayana, PR) solutions were prepared in HEPES buffer 24 h in advance in gas-tight containers and diluted to the final concentration (1 mM) immediately before used as previously described [10]. 2.2. Ca2+ measurements Cells were loaded with the fluorescent Ca2+ indicator Fura-2 by incubating the cells attached on coverslips in the culture medium containing 5 AM of the acetoxymethyl ester of the dye (Fura-2 AM; Molecular Probes, Eugene, OR, USA) for 30 min under culture conditions [10]. After loading, cells were washed three times with the HEPES buffer, and the coverslips were placed into the perfusion chamber (with volume of 250 Al) and perfused (250 Al/min) for 30 min with the HEPES buffer at 37 8C before being exposed to the various drugs. The HEPES buffer with or without the drugs was perfused at a speed of 250 Al/min. For SH-SY5Y cells, the perfusion chamber was set on an inverted microscope (DIAPHOT 300; Nikon, Melville, NY, USA), equipped with a 40 Fluor oil-immersion objective (N.A.1.30 Nikon). The microscope was connected to a highspeed multiwavelength illuminator (DeltaRAM V; Photon Technology International, PTI, Lawrenceville, NJ, USA). In most experiments, the Ca2+-sensitive excitation wavelengths for Fura-2 (340 and 380 nm), and in some experiments an additional Ca2+-insensitive excitation wavelength for Fura-2 (358 nm) were alternately (every 0.02 s) generated by a monochromator. The total emitted fluorescence (upon the alternated excitation at 340, 380 and 358 nm) from 15 to 20 cells was filtered with the fluorescence barrier filter BA 515 nm, collected with a photomultiplier (PMT01-710, Photon Technology International, PTI), and digitized at 50 Hz. For DRG neurons, the perfusion chamber was set on an inverted microscope (Axiovert S100; Zeiss, Thornwood, NY, USA), equipped with a 40 Fluar oilimmersion objective (N.A. 1.3 Zeiss). The microscope was connected to a Attofluor RatioVision real-time digital fluorescence analyzer (Zeiss). The excitation wavelengths for Fura-2 (334 and 380 nm) were alternately (every 0.55 s) generated by an Attofluor 4 position wavelength selection device (ATTO Instruments, Rockville, MD, USA) equipped with 334 and 380 nm filters (Chroma Technology, Rockingham, VT, USA) using a transmitted light source generated by an AttoArc2 HBO unit (Zeiss). The emitted florescence from individual DRG cells was
176
A. Corrales et al. / Brain Research 1031 (2005) 174–184
filtered with the fluorescence barrier filter BA 475 nm (Chroma Technology) and collected with an Intensified CCD camera (ATTO Instruments). Standard Ca2+ calibration solutions (Molecular Probes) contained (in mM) :100 KCl, 10 Mops pH 7.2, and 10 CaEGTA+EGTA. The amount of CaEGTA and EGTA varied for achieving various levels of free Ca2+ (for 0 Ca2+ = 10 mM EGTA only; for 17 nM Ca2+ = 9 mM EGTA + 1 mM CaEGTA; for 38 nM Ca2+ = 8 mM EGTA + 2 mM CaEGTA; for 150 nM Ca2+ = 5 mM EGTA + 5 mM CaEGTA; for 225 nM Ca2+ = 4 mM EGTA + 6 mM CaEGTA; for 351 nM Ca2+ = 3 mM EGTA + 7 mM CaEGTA; for 602 nM Ca2+ = 2 mM EGTA + 8 mM CaEGTA; for 1350 nM Ca2+ = 1 mM EGTA + 9 mM CaEGTA; and for 39,800 nM Ca2+ = 10 mM CaEGTA only). A 20-Al drop of each standard Ca2+ buffer containing Fura2Na+ (180 nM) was placed on the middle of a glass coverslip and Ca2+-fluorescence at 515 nm was measured upon excitation at 340 and 380 nm using the PTI system. 2.3. Data analysis Data collection and analysis were carried out using the software Felix (version 1.42a, Photon Technology International) for SH-SY5Y cells and the software Attofluor RatioVision (version 6.10p, ATTO Instruments) for DRG neurons. Analysis was also aided with the software Clampfit (Pclamp 8, Axon Instruments, Foster City, CA, USA) and Graphpad Prism (Graphpad Software, San Diego, CA, USA). For each treatment, experiments were done under different conditions on sister cultures plated on the same day. The averaged traces shown in the figures were obtained by lining up the K+on and the K+off peaks. In figures, the data represent the delta ratio (D ratio) of the emission of Fura-2 at 515 nm (or at N475 nm) generated by excitation at 340 nm (or 334 nm) and 380 nm (ratio 340/380). Comparison between different groups was performed using unpaired two-tailed t-test when there was only one treatment and one-way analysis of variance test with Newman–Keuls post hoc test when there was more than one treatment by using the Graphpad Prism software (Graphpad Software).
3. Results 3.1. In neuronal cells pulses of high KCl evoke two transient [Ca2+]cyt-increases one upon the addition and the other upon removal of high KCl Fig. 1A shows [Ca2+]cyt-changes in SH-SY5Y cells during exposure to two consecutive pulses of 200 mM KCl. In response to application of KCl, the [Ca2+]cyt rapidly increased followed by a decay to a plateau level as previously reported [46]. Upon removal of KCl, there was a slower and lower increase in [Ca2+]cyt that was also
Fig. 1. Pulses of KCl produce two transient increases in [Ca2+]cyt in SHSY5Y cells. (A) Cells were exposed to two consecutive KCl pulses (200 mM) for 2 min. K+on is the transient increase in [Ca2+]cyt observed upon the addition of KCl; K+plt is the new steady-state [Ca2+]cyt observed in the continual presence of KCl; K+off is the transient increase in [Ca2+]cyt observed upon removal of KCl. (B) Cells were exposed to consecutive KCl pulses (2 min each) of increasing levels of KCl. * Indicates K+off transients; # indicates a spontaneous transient increase in [Ca2+]cyt following the exposure to 500 mM KCl.
followed by a decay to a plateau level approaching the preKCl [Ca2+]cyt level. In this study, we define the [Ca2+]cyttransient observed upon the KCl addition as K+on, the steady-state (plateau) level of [Ca2+]cyt observed in the continual presence of KCl as K+plt and the [Ca2+]cyt transient observed upon KCl removal as K+off (Fig. 1A). The K+-evoked responses could be evoked repetitively in these cells (Fig. 1A). The magnitudes of K+on, K+plt and K+off were comparable between the first and second KCl-pulse, as long as the time between the KCl pulses was not shorter than 10 min. As the concentration of KCl increased, the decay from K+on became faster, and the magnitudes of K+on and the K+off increased (Fig. 1B). The magnitude of the K+on was always larger than that of the K+off (Fig. 1B). In SHSY5Y cells, at KCl concentrations above 20 mM clear phases of K+on and K+plt were observed, while the K+off only became detectable when the KCl level was higher than 80 mM (Fig. 1B). In some experiments, the cells started to display bspontaneousQ transient increases in [Ca2+]cyt upon removal of 500 mM KCl (Fig. 1B, indicated with #). A similar response to KCl was observed in some cultured rat dorsal root ganglia (DRG) neurons. As in SH-SY5Y cells, in most DRG neurons tested the K+on and K+plt became apparent at lower KCl levels than the K+off and the decay of K+on became faster with increasing KCl
A. Corrales et al. / Brain Research 1031 (2005) 174–184
177
concentration (Fig. 2A). In contrast to SH-SY5Y cells, the KCl concentration required for evoking a detectable K+on and an apparent K+off varied among DRG neurons. Most, but not all, DRG neurons displayed a K+off when exposed to 100–200 mM KCl (Fig. 2A) and a few DRG neurons (b5%) displayed K+off when exposed to 20–50 mM KCl (Fig. 2B), probably reflecting the heterogeneous nature of these neurons. We have chosen SH-SY5Y cells to further characterize the K+off, since their response was highly reproducible, reflecting the homogeneous nature of this cloned neuronal cells. In addition, we selected 200 mM KCl since it evoked a relative high Koff transient which facilitated the study of how various conditions/drugs affected the Koff magnitude. 3.2. The K+on and the K+off have different sensitivity to isoflurane Under the same experimental conditions, the K+on peak has been shown to be strongly reduced by 1 mM isoflurane [10]. We investigated whether the K+plt and K+off were affected by isoflurane. As expected, we found that isoflurane at 1 mM strongly inhibited the K+on peak, in addition, isoflurane also inhibited the K+plt but had no significant effect on the K+off peak (Fig. 3).
Fig. 3. Isoflurane (1 mM) reduced K+on and K+plt, but not K+off. The application of isoflurane was started 10 min before the 2-min stimulation with 200 mM KCl. (A) The averaged KCl-evoked [Ca2+]cyt responses in the absence and presence of isoflurane. Differences between dashed lines indicate the measurements for K+on ( ) isoflurane (1), K+on (+) isoflurane (2), K+off ( ) isoflurane (3), and K+off (+) isoflurane (4). (5) and (6) stand for the K+plt-levels without and with isoflurane, respectively. (B) Mean F S.E.M. for K+on, K+plt and K+off. The number of experiments for each condition is indicated on the top of each bar. Statistically significant difference between control and isoflurane groups: +P b 0.01 (unpaired two-tailed t-test).
3.3. The K+on and the K+off have different triggers
Fig. 2. Pulses of KCl produce two transient increases in [Ca2+]cyt in cultured rat DRG neurons. (A) Response of a single DRG neuron (31-Am diameter) exposed to consecutive 2-min pulses of increasing levels of KCl. In this neuron, the K+off transient became apparent at 100 mM KCl. (B) The response of two neurons displaying K+off transient at lower levels of KCl: a 2-min exposure to 20 mM KCl (top, 31-Am diameter) and a 5-min exposure to 50 mM KCl (bottom, 26-Am diameter).
In SH-SY5Y cells, the trigger for K+on is Ca2+-entry through the voltage-dependent Ca2+ channels on the plasma membrane, which in turn leads to a Ca2+-induced Ca2+ release from the caffeine-sensitive Ca2+ stores through activation of the ryanodine-sensitive Ca2+ release channels [10,46]. Since removal of KCl repolarizes the cells (towards the resting potential), it is then possible that K+off was triggered by a Ca2+ tail current through the voltage-dependent Ca2+ channels. In order to investigate this possibility, we tested the KCl response at low extracellular Ca2+ ([Ca2+]e). We found that lowering the [Ca2+]e from 1.5 mM to 150 AM reduced more strongly the K+on-peak than the K+off-peak (by 96% vs. 57%) (Fig. 4A). Part of this reduction may reflect partial Ca2+depletion of intracellular Ca2+ stores when the cells were perfused with buffer containing low [Ca2+]. Therefore, these experiments were repeated after overloading the cells with Ca2+ by exposing them to high [Ca2+]e (37 mM) for 20 min. Under these conditions, reducing [Ca2+]e to 150 AM led to a much greater reduction of the K+on-peak than of the K+off-peak (by 88% vs. 16%) (Fig. 4B). In both
178
A. Corrales et al. / Brain Research 1031 (2005) 174–184
Ca2+-entry through the plasma membrane appears to influence the rise and decay rates of the K+off transient. 3.4. Thapsigargin-sensitive intracellular Ca2+ stores do not contribute to any of the components of the KCl-evoked [Ca2+]cyt response We have previously shown that in SH-SY5Y cells 100 nM thapsigargin (TG) did not affect the KCl-evoked [Ca2+]cyt-transient (corresponding to the K+on peak), while it completely inhibited the carbachol-evoked [Ca2+]cyttransient [14,46]. The carbachol-evoked [Ca2+]cyt-transient mostly results from Ca2+-release from the IP3-sensitive Ca2+ stores, while the KCl-evoked [Ca2+]cyt-transient mostly results from Ca2+ release from the caffeine-sensitive Ca2+ stores [14,46]. In this study, we have investigated whether TG-sensitive Ca2+ stores contributed to K+plt and K+off. As expected, 100 nM TG did not affect the K+on-peak. In addition, we found that 100 nM TG did not affect the K+plt or the K+off as well (Fig. 7). TG-sensitive Ca2+ stores do not appear to significantly contribute to any of the KCl-evoked [Ca2+]cyt responses. 3.5. The KCl-evoked Koff transient does not reflect changes in Fura-2 affinity or in cell volume
Fig. 4. Low [Ca2+]e reduces the K+on more strongly than the K+off. After loading the cells with Fura, cells were equilibrated for 30 min in the experimental chamber (see Materials and methods) in the presence of either physiological level (1.5 mM) of [Ca2+]e (A), or high level (37 mM) of [Ca2+]e (B). In both cases, the [Ca2+]e was subsequently decreased to 150 AM prior to exposure to a 200 mM KCl pulse for 2 min. The cells remained in 150 AM [Ca2+]e for a total period of 10 min. Then the [Ca2+]e was increased to that used initially (1.5 and 37 mM for A and B, respectively). After an equilibration period for 10 min, the cells were exposed to a 200 mM KCl pulse for 2 min. The magnitudes (D[Ca2+]cyt) of K+on, K+plt and K+off observed at each [Ca2+]e are listed in the inset tables in A and B.
cases, with and without Ca2+ overloading, the rates of the rise and decay of K+off were slower when evoked in the presence of low [Ca2+]e (150 AM) than in the presence of normal (1.5 mM) or high (37 mM) [Ca2+]e (Fig. 4). We also used La3+, a nonspecific cationic channel blocker that blocks Ca2+-entry through all voltage-dependent Ca2+ channels and other non-voltage-dependent cationic channels [3,4,23]. As in the case of isoflurane, La3+ reduced the K+on-peak and the K+plt, but not the K+off-peak (Fig. 5). However, both La3+ and isoflurane decreased the rate of decay of K+off (Fig. 6). In the presence of La3+, isoflurane did not produce an additional blocking effect on K+on or K+plt, but showed a tendency to reduce the K+off, although this tendency was statistically not significant (Fig. 5). The data with low [Ca2+]e and with La3+ suggest that the trigger for the K+off-transient is different to that for the K+ontransient and that Ca2+-entry through voltage-dependent Ca2+ channels does not appear to be involved. However,
Since the affinity of Fura-2 is known to be affected by the ionic strength [16], we investigated how the Fura-2 signal (fluorescence ratio upon excitation at 340 and 380 nm; refer as to ratio 340/380) was affected by changes in ionic strength evoked during the 2-min pulses of 200 mM KCl. Fig. 8A shows that a 400 mosmol increase in the solution (evoked by the addition of 200 mM KCl) reduced the Fura-2 signal measured between 150 and 602 nM free Ca2+. To investigate how these changes in Fura-2 signal could have affected the KCl-evoked [Ca2+]cyt response in cells (e.g. Fig. 8B, bcellsQ). The cell-free chamber was perfused with a solution containing Fura-2, 100 mM KCl and 225 nM of free Ca2+, resembling the intracellular ionic levels at rest. Then 2-min pulses of the same solution, but with an increase of 200 mM in the KCl concentration (i.e. 300 mM KCl in total), were applied following the protocol used for experiments with cells. These 2-min pulses resulted in small reversible decreases in the Fura-2 signal (ratio 340/380) (Fig. 8B, bcell-freeQ) as it was expected from the observed decrease in Fura-2 signal during an increase in ionic strength (Fig. 8A). However, the magnitude of this change was too small to account for or to even contribute significantly to the KCl-evoked [Ca2+]cyt changes. This can be seen more easily when the cell response to 200 mM KCl is superimposed on the changes in the Fura-2 signal due solely to increase in the ionic strength by 200 mM KCl (Fig. 8C). Since addition of 200 mM KCl would increase the osmotic strength of the solution, we also investigated whether the Koff reflected a response to change in cell
A. Corrales et al. / Brain Research 1031 (2005) 174–184
179
Fig. 5. The nonselective cationic channel blocker La3+ reduced K+on and K+plt, but not K+off. The application of La3+ (100 AM) and isoflurane (1 mM) was started 10 and 15 min before the 200 mM KCl stimulation, respectively. The KCl-stimulation was for 2 min. (A) shows how the magnitudes (D ratio) of K+on, K+plt and K+off were measured. (B) The averaged KCl-evoked [Ca2+]cyt-response in the absence and presence of La3+. (C) The averaged KCl-evoked [Ca2+]cyt response in the presence of La3+ and of La3+ plus isoflurane. In this group of cells, we also measured the KCl-evoked [Ca2+]cyt-response in the presence of isoflurane (1 mM) while in the absence of La3+. Since this response overlapped with that in the presence of only La3+, we only show the measurements in D– F. The mean F S.E.M. for K+on (D), K+plt (E) and K+off (F) are shown. The number of experiments for each condition is indicated on the top of each bar in panel D. Statistically significant difference between the treated group and the control group: +P b 0.01, *P b 0.001. No statistical difference was found between treated groups (one-way Anova plus Newman–Keuls test).
volume. We measured simultaneously both [Ca2+]cytchanges and volume changes by using Fura-2. Changes in volume were measured using the Ca2+-insensitive (isosbestic) excitation wavelength [1,29]. The isosbestic wavelength of Fura-2 in our system was found to be 358 nm (Fig. 9A). Fig. 9B shows the Fura-2 fluorescence after excitation at 340, 380 and 358 nm (S340, S380 and S358) at various free Ca2+ concentrations. As expected, the S340 increased while the S380 decreased with increasing the free Ca2+ concentration. On the other hand, S358 did not
change with increases in free Ca2+. Since the Fura-2 fluorescence is proportional to the concentration of Fura-2, an increase in S358 will reflect cell shrinkage and a decrease in S358 will reflect cell swelling. Fig. 9C shows that during the pulse of 200 mM KCl the cells display a reversible decrease in volume (reversible increase in S358). Moreover, the maximum decrease in volume (S358 peak) occurred between the KCl-evoked Kon and Koff peaks. We also exposed the cells for 2 min to a solution with equivalent ionic strength as the HEPES
180
A. Corrales et al. / Brain Research 1031 (2005) 174–184
during exposure to solution A the cells swelled as indicated by the decrease in S358. The reason for the cells swelling during exposure to solution A is because the plasma membrane is highly K+ permeable, so that KCl does not produce as strong an osmotic pressure as NaCl or sucrose do. In fact, addition of 200 or 400 mM sucrose to the HEPES buffer produced significantly larger decreases in cell areas/volume than addition of 200 mM KCl (data not shown). Then Koff is observed when the cells are exposed to
Fig. 6. The rate of decay of K+off is decreased by La3+ and isoflurane. The application of La3+ (100 AM) and isoflurane (1 mM) was started 10 and 15 min before the 200 mM KCl (2 min) stimulation, respectively. (A) While the rate of decay of the K+on can be well described by a single exponential function, the rate of the rise and decay of the K+off appear to be linear vs. time. Values (mean F S.E.M.) shown in (B), (C) and (D) represent the decay rate of K+on, the decay rate of K+off and the rise rate of K+off, respectively. The number of experiments for each condition is indicated on the top of each bar in (B). Statistically significant difference vs. the control group: *P b 0.001. No statistical difference was found between treated groups (one-way Anova plus Newman–Keuls test).
buffer (see Materials and methods) but with high K+ and low Na+ and Ca2+ (solution A in mM: 145 KCl, 5 NaHCO3, 10 HEPES, 1 MgCl2, 10 glucose and 490 nM Ca2+, pH 7.4). Fig. 9D shows that exposing the cells to solution A (high K+, low Na+ and Low Ca2+) also evoked Kon (as expected by the KCl-evoked depolarization) and Koff; moreover,
Fig. 7. Thapsigargin (100 nM) did not affect K+on, K+plt or K+off. (A) Cells were treated with thapsigargin (TG, 100 nM) for 28–35 min before the 200 mM KCl stimulation (2 min). In parallel control, cells were exposed to two consecutive pulses of 200 mM KCl, the time between the two KClapplications was also 28–35 min. The KCl-evoked [Ca2+]cyt-responses at corresponding times (indicated with the box) were compared. TG was continuously present in four experiments and removed at the arrow in the other four experiments. Since TG binding is irreversible [47] and the response was the same for both groups, the data was pooled. (B) The averaged KCl-evoked [Ca2+]cyt-response in the absence and presence of TG. (C) Mean F S.E.M. of measurements for K+on, K+plt and K+off. The number of experiments for each condition is indicated on the top of each bar. No statistically significant differences were found between control and TG groups (unpaired two-tailed t-test).
A. Corrales et al. / Brain Research 1031 (2005) 174–184
181
Fig. 8. The KCl-evoked [Ca2+]cyt response does not reflect changes in Fura-2 affinity. (A) The Fura-2 signal (ratio 340/380) at various concentrations of free Ca2+ was measured at two ionic strengths. Standard Ca2+ calibration solutions containing 100 mM KCl and the following levels of free Ca2+: (in nM) 0, 17, 38, 150, 225, 351, 602, 1350 were used. To 100 Al of a given standard Ca2+ buffer was added either 5 Al of H2O (open symbols) or 5 Al of 4 M KCl (filled symbols) that increased the KCl level to 300 mM. (B) The first panel (cells) displays a response of cells to a 2 min KCl pulse (200 mM). The second panel (cell-free) displays the change in the Fura-2 signal (ratio 340/380) in a solution during a 2 min KCl pulse (200 mM). In this case, the chamber was perfused with a solution containing a level of free Ca2+ of 225 nM, that resembles that found in cells at rest, and Fura-2 (Fura-2 Na+ salt, 18 nM). Then 2-min pulses of the same solution containing an additional 200 mM KCl were applied following the protocol used for experiments with cells. (C) The cell response to a 2 min KCl pulse (200 mM) (corresponding to part B bcellsQ) is superimposed with the changes in the Fura-2 signal due solely to increasing the ionic strength by 200 mM KCl (corresponding to part B bcell-freeQ).
pulses of high K regardless of the osmotic strength of the solution (cell swelling or shrinking).
4. Discussion In this study, we found that pulses of [K+]e have a complex effect on [Ca2+]cyt, namely they evoke two distinct transient increases in [Ca2+]cyt. One during the onset of high KCl-application (K+on), which corresponds to the one triggered by depolarization-evoked opening of voltagedependent Ca2+ channels followed by Ca2+-evoked Ca2+release from intracellular Ca2+ stores [5,20,41,46], and a second novel transient increase in [Ca2+]cyt observed upon withdrawal of high KCl-application (K+off). 4.1. Triggers and intracellular Ca2+ stores involved in K+on and K+off are different It is well known that the trigger for K+on is the opening of voltage-dependent Ca2+ channels followed by Ca2+ entry through these channels [5,20,41,46]. In contrast, experiments with low extracellular Ca2+ and with the nonspecific cationic channel blocker La3+, in the present study, indicated that the trigger for K+off does not involve Ca2+ entry through voltage-dependent Ca2+ channels or through any other cation selective channel on the plasma membrane. Ca2+-entry, however, appears to speed up the decay of K+off. These results suggest that the rate of the decay of K+off is modulated by Ca2+-entry, while the magnitude of K+off is not affected by Ca2+-entry. Since the Ca2+ pumps and transporters on the plasma membrane are modulated by [Ca2+]e and Ca2+-entry [33,44], they may contribute to defining the rate of decay of K+off. Based on their sensitivity to the Ca2+-ATPase blocker thapsigargin (TG), most neuronal cells have two types of
intracellular Ca2+ stores, a TG-sensitive and a TGinsensitive. In SH-SY5Y cells, the IP3-sensitive Ca2+ store is TG-sensitive, while the caffeine-sensitive Ca2+ store is TG-insensitive [14]. We found that TG-sensitive Ca2+ stores, including the IP3-sensitive Ca2+ store, do not contribute to the magnitude of K+on, K+plt or K+off. The data with 100 nM TG also indicate that TG-sensitive Ca2+ATPases do not appear to contribute to the rate of decay of the K+on or K+off. In these cells, the K+on and K+plt have been attributed mainly to Ca2+-entry and Ca2+-release from the TG-insensitive caffeine-sensitive Ca2+ stores. It is unlikely that the TG-insensitive caffeine-sensitive store is responsible for K+off because the magnitude of K+off was the same whether this store was partially depleted (in the continual presence of KCl after K+on and its decay) or not (in the presence of La3+ that blocked K+on and K+plt and prevented Ca2+-discharge from the caffeine-sensitive store). Moreover, isoflurane, which partially depletes the caffeinesensitive Ca2+ stores, reduced the K+on and K+plt but not the K+off. However, more experiments are required to determine whether and which other TG-insensitive Ca2+ stores, such as the mitochondria and nuclear envelope, may contribute to the K+off. 4.2. The generation of K+off -transients depends on the extracellular K+ level and on the neuronal cell type Due to their clonal origin, SH-SY5Y cells represent a homogenous neuronal cell population. As expected, their [Ca2+]cyt-responses to transient increases in [K+]e are highly reproducible. We found that in the heterogeneous population of rat dorsal root ganglia (DRG) neurons, transient increases in [K+]e also evoked complex changes in [Ca2+]cyt. The major difference between DRG neurons and SH-SY5Y cells was in the K+off response. In SHSY5Y cells, K+off was always evoked by high levels of
182
A. Corrales et al. / Brain Research 1031 (2005) 174–184
KCl (100–200 mM). In DRG neurons, high levels of KCl (100–200 mM) did not always evoke a K+off and low levels of KCl (20–50 mM) evoked a K+off in only a few DRG neurons. The separation of K+on and K+off was clear when the cells were exposed to KCl for at least 2 min. Previous studies have not reported K+off in DRG neurons, the reason could be that in those studies neurons were exposed to only the lower KCl levels (20–50 mM) and the KCl exposure was brief (from a few seconds up to 1 min compared to 2–5 min in the present study) [17,37,38,39]. Neurons can be exposed to high [K+]e during intensive repetitive firing in the absence of cell damage, and to even higher [K+]e in cases with cell damage. Then K+off may have a physiological role in some DRG neurons. 4.3. Physiological implications
Fig. 9. The KCl-evoked [Ca2+]cyt response does not reflect cell volume changes. (A) Excitation scans of solutions containing free Ca2+ in nM: 0, 300, 600 and 39,800 are shown. With the optics of our system, the isosbestic wavelength (Ca2+-insensitive) of Fura-2 was found to be 358 nm. (B) shows the Fura-2 signals after stimulating at 340, 360 and 358 nm at the indicated concentrations of free Ca2+. (C) shows the cytoplasmic Ca2+ changes (D ratio 340/380 nm) and the volume changes (358 nm) upon exposure to two consecutive 2 min KCl pulses (200 mM). (D) shows the cytoplasmic Ca2+ changes (D ratio 340/380) and the volume changes (358 nm) upon exchanging for 2 min the bath HEPES-solution with a high K+ and low Na+ and Ca2+ solution (solution A = Sol. A) and upon a 2 min KCl pulse (200 mM).
In vivo there are certain conditions, in which [K+]e is increased, such as during repetitive firing, anoxia, ischemia, hypoglycemic coma and migraine aura [18,24,27,42,43,45]. The repetitive firing-evoked increase in [K+]e results from a combination of K+-release from voltage- and Ca2+-dependent K+ conductances and opening of ligand-gated postsynaptic channels [43]. In case of tissue damage, a further increase in [K+]e results from a decrease in energy that leads to a decrease in the Na+/K+ ATPase activity as well as from cell breakdown. In this case, in addition to an increase in [K+]e there is also a decrease in [Na+]e and in [Ca2+]e [40]. Small increases in [K+]e have been shown to be associated with increases in neuronal activity, which can improve nerve regeneration [26,46] and probably contribute to the sensation of deafferentation pain [22]. In focal ischemia, there is a dense ischemic core region composed of irreversibly damaged tissue named dfocusT and a less dense ischemic perifocal zone named dpenumbraT in which cells are at risk of becoming irreversibly damaged. Although the primary threat to the cells in the penumbra is the underperfusion, it has been reported that irregularly occurring depolarizations and ionic transients of the spreading depression type jeopardize the survival of neurons in the penumbra [7,8,15,19,28]. A large increase in [K+]e appears to be responsible for the initiation of spreading depression observed not only in focal ischemia but in many other in vivo conditions [30]. In these in vivo conditions, the increases and fluctuations in [K+]e are gradual (second-minute range) [40]. In our experiments, the change in [K+]e was also gradual (it takes at least 1 min to complete the exchange of buffer in the recording chamber). At this time, we do not know how the magnitude and the relative contribution of Kon and Koff are affected by different rates of change in [K+]e. In the presence of cell damage, there is not only an elevation of [K+]e but also a decrease in [Na+]e and [Ca2+]e [40]. Therefore, exposure to higher levels of K+ alone will mimic more closely the changes in ionic environment during excessive repetitive firing, while the exposure to high K+ in combination with low Na+ and low Ca2+ as shown in Fig. 9, will
A. Corrales et al. / Brain Research 1031 (2005) 174–184
mimic more closely the changes in ionic environment during cell damage as during ischemia [40]. However, in both cases transient increases in [K+]e evoke both [Ca2+]cyt transients (Kon and Koff), then excessive increases in [K+]e can produce excessive and prolonged increases in [Ca2+]cyt, which in turn are known to change neuronal activity and even lead to neuronal death. The neuroprotective effects during ischemia of both voltage-dependent Ca2+ channel blockers and volatile anesthetics have in part been attributed to their capacity to reduce neuronal Ca2+ accumulation. Interestingly, repetitive spreading depression that is evoked by high [K+]e is ameliorated but not eliminated by voltage-dependent Ca2+ channel blockers [30,35] and by isoflurane and other volatile anesthetics [21,31]. Our data indicate that Ca2+ channel blockers and isoflurane can only reduce some of the K+-evoked increases in [Ca2+]cyt (K+on and K+plt, but not K+off). The identification and use of pharmacological agents, alone or in combination, which completely block the K+-evoked increases in [Ca2+]cyt (K+on, K+plt, K+off), should provide more effective neuronal protection in situations where the tissue is exposed to large and transient increases of [K+]e. In summary, we found that transient increases in [K+]e produce two distinct transient increases in [Ca2+]cyt referred as to K+on and K+off. The triggers and pharmacology and the possible source of Ca2+ for these two K+-evoked cytoplasmic Ca2+-responses are different. Therefore, these differences should be taken into consideration when designing pharmacological approaches for reducing the pathological effects resulting from [Ca2+]cyt-increases evoked by transient increases in [K+]e.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14]
[15]
[16]
Acknowledgments We are grateful to June Biedler PhD for providing the SH-SY5Y human neuroblastoma cells (Sloan-Kettering Institute for Cancer Research, Rye, NY, USA at the providing time; Dr. Biedler is currently a Distinguished Resident Scientist, Fordham University, Bronx, NY, USA).
[17]
[18] [19]
[20]
References [1] J. Altamirano, M.S. Brodwick, F.J. Alvarez-Leefmans, Regulatory volume decrease and intracellular Ca2+ in murine neuroblastoma cells studied with fluorescent probes, J. Gen. Physiol. 112 (1998) 145 – 160. [2] J. Astrup, K. Norberg, Potassium activity in cerebral cortex in rats during progressive severe hypoglycemia, Brain Res. 103 (1976) 418 – 423. [3] B.M. Block, W.C. Stacey, S.W. Jones, Surface charge and lanthanum block of calcium current in bullfrog sympathetic neurons, Biophys. J. 74 (1998) 2278 – 2284. [4] A. Bouron, Activation of a capacitative Ca2+ entry pathway by store depletion in cultured hippocampal neurons, FEBS Lett. 470 (2000) 269 – 272. [5] D. Bowman, S. Alexander, D. Lodge, Pharmacological characterization of the calcium channels coupled to the plateau phase of KCl-
[21]
[22]
[23]
[24] [25] [26]
183
induced intracellular free Ca2+ elevation in chicken and rat synaptosomes, Neuropharmacology 32 (1993) 1195 – 1202. J. Bures, O. Buresova, J. Krivanek, The Mechanism and Application of Leao’s Spreading Depression of Electroencephalographic Activity, Academic Press, New York, 1974, 410 pp. E. Busch, M.L. Gyngell, M. Eis, M. Hoehn-Berlage, K.A. Hossmann, Potassium-induced cortical spreading depressions during focal cerebral ischemia in rats: contribution to lesion growth assessed by diffusion-weighted NMR and biochemical imaging, J. Cereb. Blood Flow Metab. 16 (1996) 1090 – 1099. Q. Chen, M. Chopp, G. Bodzin, H. Chen, Temperature modulation of cerebral depolarization during focal cerebral ischemia in rats: correlation with ischemic injury, J. Cereb. Blood Flow Metab. 13 (1993) 389 – 394. B.W. Connors, B.R. Ransom, D.M. Kunis, M.J. Gutnick, Activitydependent K+ accumulation in the developing rat optic nerve, Science 216 (1982) 1341 – 1343. A. Corrales, F. Xu, Z. Garavito-Aguilar, T.J. Blanck, E. Recio-Pinto, Isoflurane reduction of carbachol-evoked cytoplasmic calcium transients is dependent on caffeine-sensitive calcium stores, Anesthesio logy 99 (2003) 882 – 888. D.L. Eng, J.D. Kocsis, Activity-dependent changes in extracellular potassium and excitability in turtle olfactory nerve, J. Neurophysiol. 57 (1987) 740 – 754. J. Forstl, M. Galvan, G. ten Bruggencate, Extracellular K+ concentration during electrical stimulation of rat isolated sympathetic ganglia, vagus and optic nerves, Neuroscience 7 (1982) 3221 – 3229. B. Frankenhaeuser, A.L. Hodgkin, The after-effects of impulses in the giant nerve fibres of Loligo, J. Physiol. 131 (1956) 341 – 376. Z.V. Garavito-Aguilar, E. Recio-Pinto, A.V. Corrales, T.J.J. Blanck, F. Xu, Different Intracellular Ca2+ stores display different sensitivities to thapsigargin in human SH-SY5Y neuroblastoma cells, Brain Res. 1011 (2004) 177 – 186. G. Gido, T. Kristian, B.K. Siesjo, Extracellular potassium in a neocortical core area after transient focal ischemia, Stroke 28 (1997) 206 – 210. G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 1985 (1985 Mar 25) 3440 – 3450. K.E. Hall, A.A. Sima, J.W. Wiley, Opiate-mediated inhibition of calcium signaling is decreased in dorsal root ganglion neurons from the diabetic BB/W rat, J. Clin. Invest. 97 (1996) 1165 – 1172. C. Iadecola, From CSD to headache: a long and winding road, Nat. Med. 8 (2002) 110 – 112. T. Iijima, G. Mies, K.A. Hossmann, Repeated negative DC deflections in rat cortex following middle cerebral artery occlusion are abolished by MK-801: effect on volume of ischemic injury, J. Cereb. Blood Flow Metab. 12 (1992) 727 – 733. B.C. Kimball, D.I. Yule, M.W. Mulholland, Caffeine- and ryanodinesensitive Ca2+ stores in cultured guinea pig myenteric neurons, Am. J. Physiol. 270 (1996) G594 – G603. Y. Kitahara, K. Taga, H. Abe, K. Shimoji, The effects of anesthetics on cortical spreading depression elicitation and c-fos expression in rats, J. Neurosurg. Anesthesiol. 13 (2001) 26 – 32. N. Kriz, R. Rokyta, Effect of unilateral deafferentation on extracellular potassium concentration levels in rat thalamic nuclei, Neuroscience 96 (2000) 101 – 108. J.B. Lansman, Blockade of current through single calcium channels by trivalent lanthanide cations. Effect of ionic radius on the rates of ion entry and exit, J. Gen. Physiol. 95 (1990) 679 – 696. M. Lauritzen, Pathophysiology of the migraine aura. The spreading depression theory, Brain 117 (1994) 199 – 210. A.A.P. Leao, Spreading depression of activity in the cerebral cortex, J. Neurophysiol. 7 (1944) 359 – 390. P.A. Low, Endoneurial potassium is increased and enhances spontaneous activity in regenerating mammalian nerve fibers—implications for neuropathic positive symptoms, Muscle Nerve 8 (1985) 27 – 33.
184
A. Corrales et al. / Brain Research 1031 (2005) 174–184
[27] A. Mayevsky, T. Zeuthen, B. Chance, Measurements of extra cellular potassium, ECoG and pyridine nucleotide levels during cortical spreading depression in rats, Brain Res. 76 (1974) 347 – 349. [28] G. Mies, T. Iijima, K.A. Hossmann, Correlation between peri-infarct DC shifts and ischaemic neuronal damage in rat, NeuroReport 4 (1993) 709 – 711. [29] S. Muallem, B.X. Zhang, P.A. Loessberg, R.A. Star, Simultaneous recording of cell volume changes and intracellular pH or Ca2+ concentration in single osteosarcoma cells UMR-106-01, J. Biol. Chem. 267 (1992) 17658 – 17664. [30] T.P. Obrenovitch, E. Zilkha, High extracellular potassium, and not extracellular glutamate, is required for the propagation of spreading depression, J. Neurophysiol. 73 (1995) 2107 – 2114. [31] P.M. Patel, J.C. Drummond, D.J. Cole, P.J. Kelly, M. Watson, Isoflurane and pentobarbital reduce the frequency of transient ischemic depolarizations during focal ischemia in rats, Anesth. Analg. 86 (1998) 773 – 780. [32] D. Pelligrino, H. Yokoyama, M. Ingvar, B.K. Siesjo, Moderate arterial hypotension reduces cerebral cortical blood flow and enhances cellular release of potassium in severe hypoglycemia, Acta Physiol. Scand. 115 (1982) 511 – 513. [33] K.D. Philipson, D.A. Nicoll, Sodium–calcium exchange: a molecular perspective, Annu. Rev. Physiol. 62 (2000) 111 – 133. [34] D.A. Prince, H.D. Lux, E. Neher, Measurement of extracellular potassium activity in cat cortex, Brain Res. 50 (1973) 489 – 495. [35] F. Richter, A. Ebersberger, H.G. Schaible, Blockade of voltage-gated calcium channels in rat inhibits repetitive cortical spreading depression, Neurosci. Lett. 334 (2002) 123 – 126. [36] G.G. Somjen, Mechanisms of spreading depression and hypoxic spreading depression-like depolarization, Physiol. Rev. 81 (2001) 1065 – 1096. [37] K.G. Sutton, D.J. Martin, R.D. Pinnock, K. Lee, R.H. Scott,
[38]
[39]
[40] [41]
[42]
[43]
[44]
[45] [46]
[47]
Gabapentin inhibits high-threshold calcium channel currents in cultured rat dorsal root ganglion neurones, Br. J. Pharmacol. 135 (2002) 257 – 265. N. Svichar, A. Shmigol, A. Verkhratsky, P. Kostyuk, InsP3-induced Ca2+ release in dorsal root ganglion neurones, Neurosci. Lett. 227 (1997) 107 – 110. N. Svichar, V. Shishkin, P. Kostyuk, Mitochondrial participation in modulation of calcium transients in DRG neurons, NeuroReport 10 (1999) 1257 – 1261. E. Sykova, Extracellular K+ accumulation in the central nervous system, Prog. Biophys. Mol. Biol. 42 (1983) 135 – 189. N.V. Voitenko, E.P. Kostyuk, I.A. Kruglikov, P.G. Kostyuk, Changes in calcium signalling in dorsal horn neurons in rats with streptozotocin-induced diabetes, Neuroscience 94 (1999) 887 – 890. F. Vyskocil, N. Kritz, J. Bures, Potassium-selective microelectrodes used for measuring the extracellular brain potassium during spreading depression and anoxic depolarization in rats, Brain Res. 39 (1972) 255 – 259. K.D. Walton, M. Chesler, Activity-related extracellular potassium transients in the neonatal rat spinal cord: an in vitro study, Neuroscience 25 (1988) 983 – 995. K.K. Wang, A. Villalobo, B.D. Roufogalis, The plasma membrane calcium pump: a multiregulated transporter, Trends Cell Biol. 2 (1992) 46 – 52. K.M. Welch, G.L. Barkley, N. Tepley, N.M. Ramadan, Central neurogenic mechanisms of migraine, Neurology 43 (1993) S21 – S25. F. Xu, J. Zhang, E. Recio-Pinto, T.J. Blanck, Halothane and isoflurane augment depolarization-induced cytosolic CA2+ transients and attenuate carbachol-stimulated CA2+ transients, Anesthesiology 92 (2000) 1746 – 1756. H.S. Young, C. Xu, P. Zhang, D.L. Stokes, Locating the thapsigarginbinding site on Ca(2+)-ATPase by cryoelectron microscopy, J. Mol. Biol. 308 (2001) 231 – 240.
Research report
Transient increases in extracellular K+ produce two pharmacological distinct cytosolic Ca2+ transients Alexandra Corrales, Jose´ V. Montoya G., Jhon-Jairo Sutachan, Genoveve Cornillez-Ty, Zayra Garavito-Aguilar, Fang Xu, Thomas J.J. Blanck, Esperanza Recio-Pinto* Anesthesiology Department, New York University School of Medicine, 550 First Avenue, RR605, New York, NY 10016, USA Accepted 28 October 2004 Available online 16 December 2004
Abstract Transient increases in extracellular K+ are observed under various conditions, including repetitive neuronal firing, anoxia, ischemia and hypoglycemic coma. We studied changes in cytoplasmic Ca2+ ([Ca2+]cyt) evoked by pulses of KCl in human neuroblastoma SH-SY5Y cells and rat dorsal root ganglia (DRG) neurons at 37 8C. A bpulseQ of KCl evoked two transient increases in [Ca2+]cyt, one upon addition of KCl (K+on) and the other upon removal of KCl (K+off). The K+on transient has been described in many cell types and is initiated by the activation of voltage-dependent Ca2+ channels followed by Ca2+-evoked Ca2+ release from intracellular Ca2+ stores. The level of KCl necessary to evoke the K+off transient depends on the type of neuron, in SH-SY5Y cells it required 100 mM KCl, in most (but not all) of dorsal root ganglia neurons it could be detected with 100–200 mM KCl and in a very few dorsal root ganglia neurons it was detectable at 20–50 mM KCl. In SHSY5Y cells, reduction of extracellular Ca2+ inhibited the K+on more strongly than the K+off and slowed the decay of K+off. Isoflurane (1 mM) reduced the K+on- but not the K+off-peak. However, isoflurane slowed the decay of K+off. The nonspecific cationic channel blocker La3+ (100 AM) had an effect similar to that of isoflurane. Treatment with thapsigargin (TG) at a concentration known to only deplete IP3-sensitive Ca2+ stores did not affect K+on or K+off, suggesting that Ca2+ release from the IP3-sensitive Ca2+ stores does not contribute to K+on and K+off transients and that the thapsigargin-sensitive Ca2+ ATPases do not contribute significantly to the rise or decay rates of these transients. These findings indicate that a pulse of extracellular K+ produces two distinct transient increases in [Ca2+]cyt. D 2004 Elsevier B.V. All rights reserved. Theme: C Topic: Other ion channels Keywords: Cytoplasmic Ca2+; Extracellular K+; Neuroblastoma; Dorsal root ganglia neurons
1. Introduction The aim of this study was to identify and characterize the changes in cytoplasmic Ca2+ ([Ca2+]cyt) produced by a pulse of extracellular potassium ([K+]e). In vivo, neurons are exposed to transient increases of [K+]e during intensive repetitive electrical firing [9,12,13,34]. These increases range between 7 and 17 mM and appear to enhance excitability of normal and regenerating nerves [11,26].
* Corresponding author. Fax: +1 212 2636139. E-mail address: [email protected] (E. Recio-Pinto). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.10.031
Under some pathological conditions such as anoxia, focal ischemia and hypoglycemic coma, [K+]e can locally increase up to 60–80 mM [2,15,27,32,42], and such large increase in [K+]e appears to be responsible for the initiation of the phenomenon known as cortical spreading depression that consists in a propagating transient suppression of electrical activity associated with cellular depolarization [6,24,25,27,30,36,42,45]. During cortical spreading depression, there can be periods of transient repolarization accompanied by transient decreases in [K+]e [15]. Experimentally, spreading depression can be evoked by the application of high K+ either to the surface of the tissue or by injection into the nervous tissue [40].
A. Corrales et al. / Brain Research 1031 (2005) 174–184
In previous studies, we observed that neuronal cells displayed two transient increases in [Ca2+]cyt when exposed to pulses of high KCl: one during the onset of KClapplication (K+on) and a second following removal of KCl (K+off) [10]. K+on has been described in many cell types, and it is triggered by a K+-mediated depolarization that leads to the activation of voltage-dependent Ca2+ channels and hence Ca2+-entry through these channels followed by Ca2+-evoked Ca2+ release from intracellular Ca2+ stores [5,20,41,46]. K+on also has been shown to be sensitive to volatile anesthetics [10,46]. This study has characterized, for the first time, the physiology and pharmacology of the K+off. We found that the trigger and pharmacology of K+on- and K+off-transients are different. Such differences should be taken into consideration when designing pharmacological approaches for reducing the pathological effects of transient increases in [K+]e observed in vivo.
2. Materials and methods 2.1. Cell culture and solutions SH-SY5Y human neuroblastoma cells were cultured in RPMI 1640 medium with l-glutamine, supplemented with penicillin (50 U/ml), streptomycin (50 Ag/ml), and 12% fetal bovine serum, at 37 8C, in a humidified atmosphere containing 5% CO2. All cell culture components were Gibco BRL products purchased from Life Technologies (Rockville, MD, USA). Experiments were performed on monolayer of cells as previously reported [10]. Cells were plated on glass coverslips (25-mm diameter) at a density of 2– 4 104 cells/ml (2 ml cell suspension/35 mm culture dish) and used when they formed a confluent monolayer (~10 –16 days after plating). Isolated adult rat dorsal root ganglia (DRG) sensory neurons were cultured on glass coverslips (25-mm diameter placed in 35 mm culture dishes) in DMEM high glucose medium supplemented with l-glutamine (2.2 mM), penicillin (100 U/ml), streptomycin (100 Ag/ml) and 10% heat-inactivated fetal bovine serum, at 37 8C, in a humidified atmosphere containing 5% CO2. The glass coverslips were pretreated with Poly-l-lysine (2 ml of 12.5 Ag/ml). Cervical DRG were isolated (1 rat/preparation) and dissociated by incubating them first in medium containing collagenase type I (2 mg/ml) and dispase (5 mg/ml) for 1 h at 37 8C, and then in medium containing only dispase (5 mg/ml) for 30 min at 37 8C. Mechanical dissociation of the cells was carried out by passing the enzymatically treated DRG several times through a Pasteur pipette. The cell suspension was centrifuged and the pellet resuspended in 600 Al of the culture medium; 10 Al of cell suspension was placed on each glass coverslip. The total volume of medium was 2 ml/dish. During experiments, the DRG cells were continuously perfused with a HEPES buffer containing (in mM) 140
175
NaCl, 5 KCl, 5 NaHCO3, 10 HEPES, 1 MgCl2, 1.5 CaCl2, and 10 glucose (pH 7.4). For SH-SY5Y cells, the HEPES buffer also contained 1 mM ATP. Experiments were performed at 37 8C and the temperature was controlled with a Dual Heater controller TC-344A and an inline heater SH-27B (Warner Instruments, Hamden, CT, USA). The exchange of the solution was carried out with a manifold. The solutions containing KCl (20–200 mM), with and without 100 AM La3+ and 100 nM thapsigargin, were prepared using the HEPES buffer. Saturated isoflurane (Ohmeda Caribe, Guayana, PR) solutions were prepared in HEPES buffer 24 h in advance in gas-tight containers and diluted to the final concentration (1 mM) immediately before used as previously described [10]. 2.2. Ca2+ measurements Cells were loaded with the fluorescent Ca2+ indicator Fura-2 by incubating the cells attached on coverslips in the culture medium containing 5 AM of the acetoxymethyl ester of the dye (Fura-2 AM; Molecular Probes, Eugene, OR, USA) for 30 min under culture conditions [10]. After loading, cells were washed three times with the HEPES buffer, and the coverslips were placed into the perfusion chamber (with volume of 250 Al) and perfused (250 Al/min) for 30 min with the HEPES buffer at 37 8C before being exposed to the various drugs. The HEPES buffer with or without the drugs was perfused at a speed of 250 Al/min. For SH-SY5Y cells, the perfusion chamber was set on an inverted microscope (DIAPHOT 300; Nikon, Melville, NY, USA), equipped with a 40 Fluor oil-immersion objective (N.A.1.30 Nikon). The microscope was connected to a highspeed multiwavelength illuminator (DeltaRAM V; Photon Technology International, PTI, Lawrenceville, NJ, USA). In most experiments, the Ca2+-sensitive excitation wavelengths for Fura-2 (340 and 380 nm), and in some experiments an additional Ca2+-insensitive excitation wavelength for Fura-2 (358 nm) were alternately (every 0.02 s) generated by a monochromator. The total emitted fluorescence (upon the alternated excitation at 340, 380 and 358 nm) from 15 to 20 cells was filtered with the fluorescence barrier filter BA 515 nm, collected with a photomultiplier (PMT01-710, Photon Technology International, PTI), and digitized at 50 Hz. For DRG neurons, the perfusion chamber was set on an inverted microscope (Axiovert S100; Zeiss, Thornwood, NY, USA), equipped with a 40 Fluar oilimmersion objective (N.A. 1.3 Zeiss). The microscope was connected to a Attofluor RatioVision real-time digital fluorescence analyzer (Zeiss). The excitation wavelengths for Fura-2 (334 and 380 nm) were alternately (every 0.55 s) generated by an Attofluor 4 position wavelength selection device (ATTO Instruments, Rockville, MD, USA) equipped with 334 and 380 nm filters (Chroma Technology, Rockingham, VT, USA) using a transmitted light source generated by an AttoArc2 HBO unit (Zeiss). The emitted florescence from individual DRG cells was
176
A. Corrales et al. / Brain Research 1031 (2005) 174–184
filtered with the fluorescence barrier filter BA 475 nm (Chroma Technology) and collected with an Intensified CCD camera (ATTO Instruments). Standard Ca2+ calibration solutions (Molecular Probes) contained (in mM) :100 KCl, 10 Mops pH 7.2, and 10 CaEGTA+EGTA. The amount of CaEGTA and EGTA varied for achieving various levels of free Ca2+ (for 0 Ca2+ = 10 mM EGTA only; for 17 nM Ca2+ = 9 mM EGTA + 1 mM CaEGTA; for 38 nM Ca2+ = 8 mM EGTA + 2 mM CaEGTA; for 150 nM Ca2+ = 5 mM EGTA + 5 mM CaEGTA; for 225 nM Ca2+ = 4 mM EGTA + 6 mM CaEGTA; for 351 nM Ca2+ = 3 mM EGTA + 7 mM CaEGTA; for 602 nM Ca2+ = 2 mM EGTA + 8 mM CaEGTA; for 1350 nM Ca2+ = 1 mM EGTA + 9 mM CaEGTA; and for 39,800 nM Ca2+ = 10 mM CaEGTA only). A 20-Al drop of each standard Ca2+ buffer containing Fura2Na+ (180 nM) was placed on the middle of a glass coverslip and Ca2+-fluorescence at 515 nm was measured upon excitation at 340 and 380 nm using the PTI system. 2.3. Data analysis Data collection and analysis were carried out using the software Felix (version 1.42a, Photon Technology International) for SH-SY5Y cells and the software Attofluor RatioVision (version 6.10p, ATTO Instruments) for DRG neurons. Analysis was also aided with the software Clampfit (Pclamp 8, Axon Instruments, Foster City, CA, USA) and Graphpad Prism (Graphpad Software, San Diego, CA, USA). For each treatment, experiments were done under different conditions on sister cultures plated on the same day. The averaged traces shown in the figures were obtained by lining up the K+on and the K+off peaks. In figures, the data represent the delta ratio (D ratio) of the emission of Fura-2 at 515 nm (or at N475 nm) generated by excitation at 340 nm (or 334 nm) and 380 nm (ratio 340/380). Comparison between different groups was performed using unpaired two-tailed t-test when there was only one treatment and one-way analysis of variance test with Newman–Keuls post hoc test when there was more than one treatment by using the Graphpad Prism software (Graphpad Software).
3. Results 3.1. In neuronal cells pulses of high KCl evoke two transient [Ca2+]cyt-increases one upon the addition and the other upon removal of high KCl Fig. 1A shows [Ca2+]cyt-changes in SH-SY5Y cells during exposure to two consecutive pulses of 200 mM KCl. In response to application of KCl, the [Ca2+]cyt rapidly increased followed by a decay to a plateau level as previously reported [46]. Upon removal of KCl, there was a slower and lower increase in [Ca2+]cyt that was also
Fig. 1. Pulses of KCl produce two transient increases in [Ca2+]cyt in SHSY5Y cells. (A) Cells were exposed to two consecutive KCl pulses (200 mM) for 2 min. K+on is the transient increase in [Ca2+]cyt observed upon the addition of KCl; K+plt is the new steady-state [Ca2+]cyt observed in the continual presence of KCl; K+off is the transient increase in [Ca2+]cyt observed upon removal of KCl. (B) Cells were exposed to consecutive KCl pulses (2 min each) of increasing levels of KCl. * Indicates K+off transients; # indicates a spontaneous transient increase in [Ca2+]cyt following the exposure to 500 mM KCl.
followed by a decay to a plateau level approaching the preKCl [Ca2+]cyt level. In this study, we define the [Ca2+]cyttransient observed upon the KCl addition as K+on, the steady-state (plateau) level of [Ca2+]cyt observed in the continual presence of KCl as K+plt and the [Ca2+]cyt transient observed upon KCl removal as K+off (Fig. 1A). The K+-evoked responses could be evoked repetitively in these cells (Fig. 1A). The magnitudes of K+on, K+plt and K+off were comparable between the first and second KCl-pulse, as long as the time between the KCl pulses was not shorter than 10 min. As the concentration of KCl increased, the decay from K+on became faster, and the magnitudes of K+on and the K+off increased (Fig. 1B). The magnitude of the K+on was always larger than that of the K+off (Fig. 1B). In SHSY5Y cells, at KCl concentrations above 20 mM clear phases of K+on and K+plt were observed, while the K+off only became detectable when the KCl level was higher than 80 mM (Fig. 1B). In some experiments, the cells started to display bspontaneousQ transient increases in [Ca2+]cyt upon removal of 500 mM KCl (Fig. 1B, indicated with #). A similar response to KCl was observed in some cultured rat dorsal root ganglia (DRG) neurons. As in SH-SY5Y cells, in most DRG neurons tested the K+on and K+plt became apparent at lower KCl levels than the K+off and the decay of K+on became faster with increasing KCl
A. Corrales et al. / Brain Research 1031 (2005) 174–184
177
concentration (Fig. 2A). In contrast to SH-SY5Y cells, the KCl concentration required for evoking a detectable K+on and an apparent K+off varied among DRG neurons. Most, but not all, DRG neurons displayed a K+off when exposed to 100–200 mM KCl (Fig. 2A) and a few DRG neurons (b5%) displayed K+off when exposed to 20–50 mM KCl (Fig. 2B), probably reflecting the heterogeneous nature of these neurons. We have chosen SH-SY5Y cells to further characterize the K+off, since their response was highly reproducible, reflecting the homogeneous nature of this cloned neuronal cells. In addition, we selected 200 mM KCl since it evoked a relative high Koff transient which facilitated the study of how various conditions/drugs affected the Koff magnitude. 3.2. The K+on and the K+off have different sensitivity to isoflurane Under the same experimental conditions, the K+on peak has been shown to be strongly reduced by 1 mM isoflurane [10]. We investigated whether the K+plt and K+off were affected by isoflurane. As expected, we found that isoflurane at 1 mM strongly inhibited the K+on peak, in addition, isoflurane also inhibited the K+plt but had no significant effect on the K+off peak (Fig. 3).
Fig. 3. Isoflurane (1 mM) reduced K+on and K+plt, but not K+off. The application of isoflurane was started 10 min before the 2-min stimulation with 200 mM KCl. (A) The averaged KCl-evoked [Ca2+]cyt responses in the absence and presence of isoflurane. Differences between dashed lines indicate the measurements for K+on ( ) isoflurane (1), K+on (+) isoflurane (2), K+off ( ) isoflurane (3), and K+off (+) isoflurane (4). (5) and (6) stand for the K+plt-levels without and with isoflurane, respectively. (B) Mean F S.E.M. for K+on, K+plt and K+off. The number of experiments for each condition is indicated on the top of each bar. Statistically significant difference between control and isoflurane groups: +P b 0.01 (unpaired two-tailed t-test).
3.3. The K+on and the K+off have different triggers
Fig. 2. Pulses of KCl produce two transient increases in [Ca2+]cyt in cultured rat DRG neurons. (A) Response of a single DRG neuron (31-Am diameter) exposed to consecutive 2-min pulses of increasing levels of KCl. In this neuron, the K+off transient became apparent at 100 mM KCl. (B) The response of two neurons displaying K+off transient at lower levels of KCl: a 2-min exposure to 20 mM KCl (top, 31-Am diameter) and a 5-min exposure to 50 mM KCl (bottom, 26-Am diameter).
In SH-SY5Y cells, the trigger for K+on is Ca2+-entry through the voltage-dependent Ca2+ channels on the plasma membrane, which in turn leads to a Ca2+-induced Ca2+ release from the caffeine-sensitive Ca2+ stores through activation of the ryanodine-sensitive Ca2+ release channels [10,46]. Since removal of KCl repolarizes the cells (towards the resting potential), it is then possible that K+off was triggered by a Ca2+ tail current through the voltage-dependent Ca2+ channels. In order to investigate this possibility, we tested the KCl response at low extracellular Ca2+ ([Ca2+]e). We found that lowering the [Ca2+]e from 1.5 mM to 150 AM reduced more strongly the K+on-peak than the K+off-peak (by 96% vs. 57%) (Fig. 4A). Part of this reduction may reflect partial Ca2+depletion of intracellular Ca2+ stores when the cells were perfused with buffer containing low [Ca2+]. Therefore, these experiments were repeated after overloading the cells with Ca2+ by exposing them to high [Ca2+]e (37 mM) for 20 min. Under these conditions, reducing [Ca2+]e to 150 AM led to a much greater reduction of the K+on-peak than of the K+off-peak (by 88% vs. 16%) (Fig. 4B). In both
178
A. Corrales et al. / Brain Research 1031 (2005) 174–184
Ca2+-entry through the plasma membrane appears to influence the rise and decay rates of the K+off transient. 3.4. Thapsigargin-sensitive intracellular Ca2+ stores do not contribute to any of the components of the KCl-evoked [Ca2+]cyt response We have previously shown that in SH-SY5Y cells 100 nM thapsigargin (TG) did not affect the KCl-evoked [Ca2+]cyt-transient (corresponding to the K+on peak), while it completely inhibited the carbachol-evoked [Ca2+]cyttransient [14,46]. The carbachol-evoked [Ca2+]cyt-transient mostly results from Ca2+-release from the IP3-sensitive Ca2+ stores, while the KCl-evoked [Ca2+]cyt-transient mostly results from Ca2+ release from the caffeine-sensitive Ca2+ stores [14,46]. In this study, we have investigated whether TG-sensitive Ca2+ stores contributed to K+plt and K+off. As expected, 100 nM TG did not affect the K+on-peak. In addition, we found that 100 nM TG did not affect the K+plt or the K+off as well (Fig. 7). TG-sensitive Ca2+ stores do not appear to significantly contribute to any of the KCl-evoked [Ca2+]cyt responses. 3.5. The KCl-evoked Koff transient does not reflect changes in Fura-2 affinity or in cell volume
Fig. 4. Low [Ca2+]e reduces the K+on more strongly than the K+off. After loading the cells with Fura, cells were equilibrated for 30 min in the experimental chamber (see Materials and methods) in the presence of either physiological level (1.5 mM) of [Ca2+]e (A), or high level (37 mM) of [Ca2+]e (B). In both cases, the [Ca2+]e was subsequently decreased to 150 AM prior to exposure to a 200 mM KCl pulse for 2 min. The cells remained in 150 AM [Ca2+]e for a total period of 10 min. Then the [Ca2+]e was increased to that used initially (1.5 and 37 mM for A and B, respectively). After an equilibration period for 10 min, the cells were exposed to a 200 mM KCl pulse for 2 min. The magnitudes (D[Ca2+]cyt) of K+on, K+plt and K+off observed at each [Ca2+]e are listed in the inset tables in A and B.
cases, with and without Ca2+ overloading, the rates of the rise and decay of K+off were slower when evoked in the presence of low [Ca2+]e (150 AM) than in the presence of normal (1.5 mM) or high (37 mM) [Ca2+]e (Fig. 4). We also used La3+, a nonspecific cationic channel blocker that blocks Ca2+-entry through all voltage-dependent Ca2+ channels and other non-voltage-dependent cationic channels [3,4,23]. As in the case of isoflurane, La3+ reduced the K+on-peak and the K+plt, but not the K+off-peak (Fig. 5). However, both La3+ and isoflurane decreased the rate of decay of K+off (Fig. 6). In the presence of La3+, isoflurane did not produce an additional blocking effect on K+on or K+plt, but showed a tendency to reduce the K+off, although this tendency was statistically not significant (Fig. 5). The data with low [Ca2+]e and with La3+ suggest that the trigger for the K+off-transient is different to that for the K+ontransient and that Ca2+-entry through voltage-dependent Ca2+ channels does not appear to be involved. However,
Since the affinity of Fura-2 is known to be affected by the ionic strength [16], we investigated how the Fura-2 signal (fluorescence ratio upon excitation at 340 and 380 nm; refer as to ratio 340/380) was affected by changes in ionic strength evoked during the 2-min pulses of 200 mM KCl. Fig. 8A shows that a 400 mosmol increase in the solution (evoked by the addition of 200 mM KCl) reduced the Fura-2 signal measured between 150 and 602 nM free Ca2+. To investigate how these changes in Fura-2 signal could have affected the KCl-evoked [Ca2+]cyt response in cells (e.g. Fig. 8B, bcellsQ). The cell-free chamber was perfused with a solution containing Fura-2, 100 mM KCl and 225 nM of free Ca2+, resembling the intracellular ionic levels at rest. Then 2-min pulses of the same solution, but with an increase of 200 mM in the KCl concentration (i.e. 300 mM KCl in total), were applied following the protocol used for experiments with cells. These 2-min pulses resulted in small reversible decreases in the Fura-2 signal (ratio 340/380) (Fig. 8B, bcell-freeQ) as it was expected from the observed decrease in Fura-2 signal during an increase in ionic strength (Fig. 8A). However, the magnitude of this change was too small to account for or to even contribute significantly to the KCl-evoked [Ca2+]cyt changes. This can be seen more easily when the cell response to 200 mM KCl is superimposed on the changes in the Fura-2 signal due solely to increase in the ionic strength by 200 mM KCl (Fig. 8C). Since addition of 200 mM KCl would increase the osmotic strength of the solution, we also investigated whether the Koff reflected a response to change in cell
A. Corrales et al. / Brain Research 1031 (2005) 174–184
179
Fig. 5. The nonselective cationic channel blocker La3+ reduced K+on and K+plt, but not K+off. The application of La3+ (100 AM) and isoflurane (1 mM) was started 10 and 15 min before the 200 mM KCl stimulation, respectively. The KCl-stimulation was for 2 min. (A) shows how the magnitudes (D ratio) of K+on, K+plt and K+off were measured. (B) The averaged KCl-evoked [Ca2+]cyt-response in the absence and presence of La3+. (C) The averaged KCl-evoked [Ca2+]cyt response in the presence of La3+ and of La3+ plus isoflurane. In this group of cells, we also measured the KCl-evoked [Ca2+]cyt-response in the presence of isoflurane (1 mM) while in the absence of La3+. Since this response overlapped with that in the presence of only La3+, we only show the measurements in D– F. The mean F S.E.M. for K+on (D), K+plt (E) and K+off (F) are shown. The number of experiments for each condition is indicated on the top of each bar in panel D. Statistically significant difference between the treated group and the control group: +P b 0.01, *P b 0.001. No statistical difference was found between treated groups (one-way Anova plus Newman–Keuls test).
volume. We measured simultaneously both [Ca2+]cytchanges and volume changes by using Fura-2. Changes in volume were measured using the Ca2+-insensitive (isosbestic) excitation wavelength [1,29]. The isosbestic wavelength of Fura-2 in our system was found to be 358 nm (Fig. 9A). Fig. 9B shows the Fura-2 fluorescence after excitation at 340, 380 and 358 nm (S340, S380 and S358) at various free Ca2+ concentrations. As expected, the S340 increased while the S380 decreased with increasing the free Ca2+ concentration. On the other hand, S358 did not
change with increases in free Ca2+. Since the Fura-2 fluorescence is proportional to the concentration of Fura-2, an increase in S358 will reflect cell shrinkage and a decrease in S358 will reflect cell swelling. Fig. 9C shows that during the pulse of 200 mM KCl the cells display a reversible decrease in volume (reversible increase in S358). Moreover, the maximum decrease in volume (S358 peak) occurred between the KCl-evoked Kon and Koff peaks. We also exposed the cells for 2 min to a solution with equivalent ionic strength as the HEPES
180
A. Corrales et al. / Brain Research 1031 (2005) 174–184
during exposure to solution A the cells swelled as indicated by the decrease in S358. The reason for the cells swelling during exposure to solution A is because the plasma membrane is highly K+ permeable, so that KCl does not produce as strong an osmotic pressure as NaCl or sucrose do. In fact, addition of 200 or 400 mM sucrose to the HEPES buffer produced significantly larger decreases in cell areas/volume than addition of 200 mM KCl (data not shown). Then Koff is observed when the cells are exposed to
Fig. 6. The rate of decay of K+off is decreased by La3+ and isoflurane. The application of La3+ (100 AM) and isoflurane (1 mM) was started 10 and 15 min before the 200 mM KCl (2 min) stimulation, respectively. (A) While the rate of decay of the K+on can be well described by a single exponential function, the rate of the rise and decay of the K+off appear to be linear vs. time. Values (mean F S.E.M.) shown in (B), (C) and (D) represent the decay rate of K+on, the decay rate of K+off and the rise rate of K+off, respectively. The number of experiments for each condition is indicated on the top of each bar in (B). Statistically significant difference vs. the control group: *P b 0.001. No statistical difference was found between treated groups (one-way Anova plus Newman–Keuls test).
buffer (see Materials and methods) but with high K+ and low Na+ and Ca2+ (solution A in mM: 145 KCl, 5 NaHCO3, 10 HEPES, 1 MgCl2, 10 glucose and 490 nM Ca2+, pH 7.4). Fig. 9D shows that exposing the cells to solution A (high K+, low Na+ and Low Ca2+) also evoked Kon (as expected by the KCl-evoked depolarization) and Koff; moreover,
Fig. 7. Thapsigargin (100 nM) did not affect K+on, K+plt or K+off. (A) Cells were treated with thapsigargin (TG, 100 nM) for 28–35 min before the 200 mM KCl stimulation (2 min). In parallel control, cells were exposed to two consecutive pulses of 200 mM KCl, the time between the two KClapplications was also 28–35 min. The KCl-evoked [Ca2+]cyt-responses at corresponding times (indicated with the box) were compared. TG was continuously present in four experiments and removed at the arrow in the other four experiments. Since TG binding is irreversible [47] and the response was the same for both groups, the data was pooled. (B) The averaged KCl-evoked [Ca2+]cyt-response in the absence and presence of TG. (C) Mean F S.E.M. of measurements for K+on, K+plt and K+off. The number of experiments for each condition is indicated on the top of each bar. No statistically significant differences were found between control and TG groups (unpaired two-tailed t-test).
A. Corrales et al. / Brain Research 1031 (2005) 174–184
181
Fig. 8. The KCl-evoked [Ca2+]cyt response does not reflect changes in Fura-2 affinity. (A) The Fura-2 signal (ratio 340/380) at various concentrations of free Ca2+ was measured at two ionic strengths. Standard Ca2+ calibration solutions containing 100 mM KCl and the following levels of free Ca2+: (in nM) 0, 17, 38, 150, 225, 351, 602, 1350 were used. To 100 Al of a given standard Ca2+ buffer was added either 5 Al of H2O (open symbols) or 5 Al of 4 M KCl (filled symbols) that increased the KCl level to 300 mM. (B) The first panel (cells) displays a response of cells to a 2 min KCl pulse (200 mM). The second panel (cell-free) displays the change in the Fura-2 signal (ratio 340/380) in a solution during a 2 min KCl pulse (200 mM). In this case, the chamber was perfused with a solution containing a level of free Ca2+ of 225 nM, that resembles that found in cells at rest, and Fura-2 (Fura-2 Na+ salt, 18 nM). Then 2-min pulses of the same solution containing an additional 200 mM KCl were applied following the protocol used for experiments with cells. (C) The cell response to a 2 min KCl pulse (200 mM) (corresponding to part B bcellsQ) is superimposed with the changes in the Fura-2 signal due solely to increasing the ionic strength by 200 mM KCl (corresponding to part B bcell-freeQ).
pulses of high K regardless of the osmotic strength of the solution (cell swelling or shrinking).
4. Discussion In this study, we found that pulses of [K+]e have a complex effect on [Ca2+]cyt, namely they evoke two distinct transient increases in [Ca2+]cyt. One during the onset of high KCl-application (K+on), which corresponds to the one triggered by depolarization-evoked opening of voltagedependent Ca2+ channels followed by Ca2+-evoked Ca2+release from intracellular Ca2+ stores [5,20,41,46], and a second novel transient increase in [Ca2+]cyt observed upon withdrawal of high KCl-application (K+off). 4.1. Triggers and intracellular Ca2+ stores involved in K+on and K+off are different It is well known that the trigger for K+on is the opening of voltage-dependent Ca2+ channels followed by Ca2+ entry through these channels [5,20,41,46]. In contrast, experiments with low extracellular Ca2+ and with the nonspecific cationic channel blocker La3+, in the present study, indicated that the trigger for K+off does not involve Ca2+ entry through voltage-dependent Ca2+ channels or through any other cation selective channel on the plasma membrane. Ca2+-entry, however, appears to speed up the decay of K+off. These results suggest that the rate of the decay of K+off is modulated by Ca2+-entry, while the magnitude of K+off is not affected by Ca2+-entry. Since the Ca2+ pumps and transporters on the plasma membrane are modulated by [Ca2+]e and Ca2+-entry [33,44], they may contribute to defining the rate of decay of K+off. Based on their sensitivity to the Ca2+-ATPase blocker thapsigargin (TG), most neuronal cells have two types of
intracellular Ca2+ stores, a TG-sensitive and a TGinsensitive. In SH-SY5Y cells, the IP3-sensitive Ca2+ store is TG-sensitive, while the caffeine-sensitive Ca2+ store is TG-insensitive [14]. We found that TG-sensitive Ca2+ stores, including the IP3-sensitive Ca2+ store, do not contribute to the magnitude of K+on, K+plt or K+off. The data with 100 nM TG also indicate that TG-sensitive Ca2+ATPases do not appear to contribute to the rate of decay of the K+on or K+off. In these cells, the K+on and K+plt have been attributed mainly to Ca2+-entry and Ca2+-release from the TG-insensitive caffeine-sensitive Ca2+ stores. It is unlikely that the TG-insensitive caffeine-sensitive store is responsible for K+off because the magnitude of K+off was the same whether this store was partially depleted (in the continual presence of KCl after K+on and its decay) or not (in the presence of La3+ that blocked K+on and K+plt and prevented Ca2+-discharge from the caffeine-sensitive store). Moreover, isoflurane, which partially depletes the caffeinesensitive Ca2+ stores, reduced the K+on and K+plt but not the K+off. However, more experiments are required to determine whether and which other TG-insensitive Ca2+ stores, such as the mitochondria and nuclear envelope, may contribute to the K+off. 4.2. The generation of K+off -transients depends on the extracellular K+ level and on the neuronal cell type Due to their clonal origin, SH-SY5Y cells represent a homogenous neuronal cell population. As expected, their [Ca2+]cyt-responses to transient increases in [K+]e are highly reproducible. We found that in the heterogeneous population of rat dorsal root ganglia (DRG) neurons, transient increases in [K+]e also evoked complex changes in [Ca2+]cyt. The major difference between DRG neurons and SH-SY5Y cells was in the K+off response. In SHSY5Y cells, K+off was always evoked by high levels of
182
A. Corrales et al. / Brain Research 1031 (2005) 174–184
KCl (100–200 mM). In DRG neurons, high levels of KCl (100–200 mM) did not always evoke a K+off and low levels of KCl (20–50 mM) evoked a K+off in only a few DRG neurons. The separation of K+on and K+off was clear when the cells were exposed to KCl for at least 2 min. Previous studies have not reported K+off in DRG neurons, the reason could be that in those studies neurons were exposed to only the lower KCl levels (20–50 mM) and the KCl exposure was brief (from a few seconds up to 1 min compared to 2–5 min in the present study) [17,37,38,39]. Neurons can be exposed to high [K+]e during intensive repetitive firing in the absence of cell damage, and to even higher [K+]e in cases with cell damage. Then K+off may have a physiological role in some DRG neurons. 4.3. Physiological implications
Fig. 9. The KCl-evoked [Ca2+]cyt response does not reflect cell volume changes. (A) Excitation scans of solutions containing free Ca2+ in nM: 0, 300, 600 and 39,800 are shown. With the optics of our system, the isosbestic wavelength (Ca2+-insensitive) of Fura-2 was found to be 358 nm. (B) shows the Fura-2 signals after stimulating at 340, 360 and 358 nm at the indicated concentrations of free Ca2+. (C) shows the cytoplasmic Ca2+ changes (D ratio 340/380 nm) and the volume changes (358 nm) upon exposure to two consecutive 2 min KCl pulses (200 mM). (D) shows the cytoplasmic Ca2+ changes (D ratio 340/380) and the volume changes (358 nm) upon exchanging for 2 min the bath HEPES-solution with a high K+ and low Na+ and Ca2+ solution (solution A = Sol. A) and upon a 2 min KCl pulse (200 mM).
In vivo there are certain conditions, in which [K+]e is increased, such as during repetitive firing, anoxia, ischemia, hypoglycemic coma and migraine aura [18,24,27,42,43,45]. The repetitive firing-evoked increase in [K+]e results from a combination of K+-release from voltage- and Ca2+-dependent K+ conductances and opening of ligand-gated postsynaptic channels [43]. In case of tissue damage, a further increase in [K+]e results from a decrease in energy that leads to a decrease in the Na+/K+ ATPase activity as well as from cell breakdown. In this case, in addition to an increase in [K+]e there is also a decrease in [Na+]e and in [Ca2+]e [40]. Small increases in [K+]e have been shown to be associated with increases in neuronal activity, which can improve nerve regeneration [26,46] and probably contribute to the sensation of deafferentation pain [22]. In focal ischemia, there is a dense ischemic core region composed of irreversibly damaged tissue named dfocusT and a less dense ischemic perifocal zone named dpenumbraT in which cells are at risk of becoming irreversibly damaged. Although the primary threat to the cells in the penumbra is the underperfusion, it has been reported that irregularly occurring depolarizations and ionic transients of the spreading depression type jeopardize the survival of neurons in the penumbra [7,8,15,19,28]. A large increase in [K+]e appears to be responsible for the initiation of spreading depression observed not only in focal ischemia but in many other in vivo conditions [30]. In these in vivo conditions, the increases and fluctuations in [K+]e are gradual (second-minute range) [40]. In our experiments, the change in [K+]e was also gradual (it takes at least 1 min to complete the exchange of buffer in the recording chamber). At this time, we do not know how the magnitude and the relative contribution of Kon and Koff are affected by different rates of change in [K+]e. In the presence of cell damage, there is not only an elevation of [K+]e but also a decrease in [Na+]e and [Ca2+]e [40]. Therefore, exposure to higher levels of K+ alone will mimic more closely the changes in ionic environment during excessive repetitive firing, while the exposure to high K+ in combination with low Na+ and low Ca2+ as shown in Fig. 9, will
A. Corrales et al. / Brain Research 1031 (2005) 174–184
mimic more closely the changes in ionic environment during cell damage as during ischemia [40]. However, in both cases transient increases in [K+]e evoke both [Ca2+]cyt transients (Kon and Koff), then excessive increases in [K+]e can produce excessive and prolonged increases in [Ca2+]cyt, which in turn are known to change neuronal activity and even lead to neuronal death. The neuroprotective effects during ischemia of both voltage-dependent Ca2+ channel blockers and volatile anesthetics have in part been attributed to their capacity to reduce neuronal Ca2+ accumulation. Interestingly, repetitive spreading depression that is evoked by high [K+]e is ameliorated but not eliminated by voltage-dependent Ca2+ channel blockers [30,35] and by isoflurane and other volatile anesthetics [21,31]. Our data indicate that Ca2+ channel blockers and isoflurane can only reduce some of the K+-evoked increases in [Ca2+]cyt (K+on and K+plt, but not K+off). The identification and use of pharmacological agents, alone or in combination, which completely block the K+-evoked increases in [Ca2+]cyt (K+on, K+plt, K+off), should provide more effective neuronal protection in situations where the tissue is exposed to large and transient increases of [K+]e. In summary, we found that transient increases in [K+]e produce two distinct transient increases in [Ca2+]cyt referred as to K+on and K+off. The triggers and pharmacology and the possible source of Ca2+ for these two K+-evoked cytoplasmic Ca2+-responses are different. Therefore, these differences should be taken into consideration when designing pharmacological approaches for reducing the pathological effects resulting from [Ca2+]cyt-increases evoked by transient increases in [K+]e.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] [14]
[15]
[16]
Acknowledgments We are grateful to June Biedler PhD for providing the SH-SY5Y human neuroblastoma cells (Sloan-Kettering Institute for Cancer Research, Rye, NY, USA at the providing time; Dr. Biedler is currently a Distinguished Resident Scientist, Fordham University, Bronx, NY, USA).
[17]
[18] [19]
[20]
References [1] J. Altamirano, M.S. Brodwick, F.J. Alvarez-Leefmans, Regulatory volume decrease and intracellular Ca2+ in murine neuroblastoma cells studied with fluorescent probes, J. Gen. Physiol. 112 (1998) 145 – 160. [2] J. Astrup, K. Norberg, Potassium activity in cerebral cortex in rats during progressive severe hypoglycemia, Brain Res. 103 (1976) 418 – 423. [3] B.M. Block, W.C. Stacey, S.W. Jones, Surface charge and lanthanum block of calcium current in bullfrog sympathetic neurons, Biophys. J. 74 (1998) 2278 – 2284. [4] A. Bouron, Activation of a capacitative Ca2+ entry pathway by store depletion in cultured hippocampal neurons, FEBS Lett. 470 (2000) 269 – 272. [5] D. Bowman, S. Alexander, D. Lodge, Pharmacological characterization of the calcium channels coupled to the plateau phase of KCl-
[21]
[22]
[23]
[24] [25] [26]
183
induced intracellular free Ca2+ elevation in chicken and rat synaptosomes, Neuropharmacology 32 (1993) 1195 – 1202. J. Bures, O. Buresova, J. Krivanek, The Mechanism and Application of Leao’s Spreading Depression of Electroencephalographic Activity, Academic Press, New York, 1974, 410 pp. E. Busch, M.L. Gyngell, M. Eis, M. Hoehn-Berlage, K.A. Hossmann, Potassium-induced cortical spreading depressions during focal cerebral ischemia in rats: contribution to lesion growth assessed by diffusion-weighted NMR and biochemical imaging, J. Cereb. Blood Flow Metab. 16 (1996) 1090 – 1099. Q. Chen, M. Chopp, G. Bodzin, H. Chen, Temperature modulation of cerebral depolarization during focal cerebral ischemia in rats: correlation with ischemic injury, J. Cereb. Blood Flow Metab. 13 (1993) 389 – 394. B.W. Connors, B.R. Ransom, D.M. Kunis, M.J. Gutnick, Activitydependent K+ accumulation in the developing rat optic nerve, Science 216 (1982) 1341 – 1343. A. Corrales, F. Xu, Z. Garavito-Aguilar, T.J. Blanck, E. Recio-Pinto, Isoflurane reduction of carbachol-evoked cytoplasmic calcium transients is dependent on caffeine-sensitive calcium stores, Anesthesio logy 99 (2003) 882 – 888. D.L. Eng, J.D. Kocsis, Activity-dependent changes in extracellular potassium and excitability in turtle olfactory nerve, J. Neurophysiol. 57 (1987) 740 – 754. J. Forstl, M. Galvan, G. ten Bruggencate, Extracellular K+ concentration during electrical stimulation of rat isolated sympathetic ganglia, vagus and optic nerves, Neuroscience 7 (1982) 3221 – 3229. B. Frankenhaeuser, A.L. Hodgkin, The after-effects of impulses in the giant nerve fibres of Loligo, J. Physiol. 131 (1956) 341 – 376. Z.V. Garavito-Aguilar, E. Recio-Pinto, A.V. Corrales, T.J.J. Blanck, F. Xu, Different Intracellular Ca2+ stores display different sensitivities to thapsigargin in human SH-SY5Y neuroblastoma cells, Brain Res. 1011 (2004) 177 – 186. G. Gido, T. Kristian, B.K. Siesjo, Extracellular potassium in a neocortical core area after transient focal ischemia, Stroke 28 (1997) 206 – 210. G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 1985 (1985 Mar 25) 3440 – 3450. K.E. Hall, A.A. Sima, J.W. Wiley, Opiate-mediated inhibition of calcium signaling is decreased in dorsal root ganglion neurons from the diabetic BB/W rat, J. Clin. Invest. 97 (1996) 1165 – 1172. C. Iadecola, From CSD to headache: a long and winding road, Nat. Med. 8 (2002) 110 – 112. T. Iijima, G. Mies, K.A. Hossmann, Repeated negative DC deflections in rat cortex following middle cerebral artery occlusion are abolished by MK-801: effect on volume of ischemic injury, J. Cereb. Blood Flow Metab. 12 (1992) 727 – 733. B.C. Kimball, D.I. Yule, M.W. Mulholland, Caffeine- and ryanodinesensitive Ca2+ stores in cultured guinea pig myenteric neurons, Am. J. Physiol. 270 (1996) G594 – G603. Y. Kitahara, K. Taga, H. Abe, K. Shimoji, The effects of anesthetics on cortical spreading depression elicitation and c-fos expression in rats, J. Neurosurg. Anesthesiol. 13 (2001) 26 – 32. N. Kriz, R. Rokyta, Effect of unilateral deafferentation on extracellular potassium concentration levels in rat thalamic nuclei, Neuroscience 96 (2000) 101 – 108. J.B. Lansman, Blockade of current through single calcium channels by trivalent lanthanide cations. Effect of ionic radius on the rates of ion entry and exit, J. Gen. Physiol. 95 (1990) 679 – 696. M. Lauritzen, Pathophysiology of the migraine aura. The spreading depression theory, Brain 117 (1994) 199 – 210. A.A.P. Leao, Spreading depression of activity in the cerebral cortex, J. Neurophysiol. 7 (1944) 359 – 390. P.A. Low, Endoneurial potassium is increased and enhances spontaneous activity in regenerating mammalian nerve fibers—implications for neuropathic positive symptoms, Muscle Nerve 8 (1985) 27 – 33.
184
A. Corrales et al. / Brain Research 1031 (2005) 174–184
[27] A. Mayevsky, T. Zeuthen, B. Chance, Measurements of extra cellular potassium, ECoG and pyridine nucleotide levels during cortical spreading depression in rats, Brain Res. 76 (1974) 347 – 349. [28] G. Mies, T. Iijima, K.A. Hossmann, Correlation between peri-infarct DC shifts and ischaemic neuronal damage in rat, NeuroReport 4 (1993) 709 – 711. [29] S. Muallem, B.X. Zhang, P.A. Loessberg, R.A. Star, Simultaneous recording of cell volume changes and intracellular pH or Ca2+ concentration in single osteosarcoma cells UMR-106-01, J. Biol. Chem. 267 (1992) 17658 – 17664. [30] T.P. Obrenovitch, E. Zilkha, High extracellular potassium, and not extracellular glutamate, is required for the propagation of spreading depression, J. Neurophysiol. 73 (1995) 2107 – 2114. [31] P.M. Patel, J.C. Drummond, D.J. Cole, P.J. Kelly, M. Watson, Isoflurane and pentobarbital reduce the frequency of transient ischemic depolarizations during focal ischemia in rats, Anesth. Analg. 86 (1998) 773 – 780. [32] D. Pelligrino, H. Yokoyama, M. Ingvar, B.K. Siesjo, Moderate arterial hypotension reduces cerebral cortical blood flow and enhances cellular release of potassium in severe hypoglycemia, Acta Physiol. Scand. 115 (1982) 511 – 513. [33] K.D. Philipson, D.A. Nicoll, Sodium–calcium exchange: a molecular perspective, Annu. Rev. Physiol. 62 (2000) 111 – 133. [34] D.A. Prince, H.D. Lux, E. Neher, Measurement of extracellular potassium activity in cat cortex, Brain Res. 50 (1973) 489 – 495. [35] F. Richter, A. Ebersberger, H.G. Schaible, Blockade of voltage-gated calcium channels in rat inhibits repetitive cortical spreading depression, Neurosci. Lett. 334 (2002) 123 – 126. [36] G.G. Somjen, Mechanisms of spreading depression and hypoxic spreading depression-like depolarization, Physiol. Rev. 81 (2001) 1065 – 1096. [37] K.G. Sutton, D.J. Martin, R.D. Pinnock, K. Lee, R.H. Scott,
[38]
[39]
[40] [41]
[42]
[43]
[44]
[45] [46]
[47]
Gabapentin inhibits high-threshold calcium channel currents in cultured rat dorsal root ganglion neurones, Br. J. Pharmacol. 135 (2002) 257 – 265. N. Svichar, A. Shmigol, A. Verkhratsky, P. Kostyuk, InsP3-induced Ca2+ release in dorsal root ganglion neurones, Neurosci. Lett. 227 (1997) 107 – 110. N. Svichar, V. Shishkin, P. Kostyuk, Mitochondrial participation in modulation of calcium transients in DRG neurons, NeuroReport 10 (1999) 1257 – 1261. E. Sykova, Extracellular K+ accumulation in the central nervous system, Prog. Biophys. Mol. Biol. 42 (1983) 135 – 189. N.V. Voitenko, E.P. Kostyuk, I.A. Kruglikov, P.G. Kostyuk, Changes in calcium signalling in dorsal horn neurons in rats with streptozotocin-induced diabetes, Neuroscience 94 (1999) 887 – 890. F. Vyskocil, N. Kritz, J. Bures, Potassium-selective microelectrodes used for measuring the extracellular brain potassium during spreading depression and anoxic depolarization in rats, Brain Res. 39 (1972) 255 – 259. K.D. Walton, M. Chesler, Activity-related extracellular potassium transients in the neonatal rat spinal cord: an in vitro study, Neuroscience 25 (1988) 983 – 995. K.K. Wang, A. Villalobo, B.D. Roufogalis, The plasma membrane calcium pump: a multiregulated transporter, Trends Cell Biol. 2 (1992) 46 – 52. K.M. Welch, G.L. Barkley, N. Tepley, N.M. Ramadan, Central neurogenic mechanisms of migraine, Neurology 43 (1993) S21 – S25. F. Xu, J. Zhang, E. Recio-Pinto, T.J. Blanck, Halothane and isoflurane augment depolarization-induced cytosolic CA2+ transients and attenuate carbachol-stimulated CA2+ transients, Anesthesiology 92 (2000) 1746 – 1756. H.S. Young, C. Xu, P. Zhang, D.L. Stokes, Locating the thapsigarginbinding site on Ca(2+)-ATPase by cryoelectron microscopy, J. Mol. Biol. 308 (2001) 231 – 240.