Regulation of ICl,swell in neuroblastoma cells by G protein signaling pathways
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Am J Physiol Cell Physiol
281: C89–C98, 2001.
Regulation of ICl,swell in neuroblastoma cells
by G protein signaling pathways
ANA Y. ESTEVEZ, TAMARA BOND, AND KEVIN STRANGE
Departments of Anesthesiology and Pharmacology, Anesthesiology Research Division,
Laboratories of Cellular and Molecular Physiology, Vanderbilt University
Medical Center, Nashville, Tennessee 37232
Received 10 October 2000; accepted in final form 31 January 2001
Estevez, Ana Y., Tamara Bond, and Kevin Strange. with tachycardia-induced congestive heart failure
Regulation of ICl,swell in neuroblastoma cells by G pro- (CHF), suggesting that the channel contributes to elec-
tein signaling pathways. Am J Physiol Cell Physiol 281: trophysiological and contractile abnormalities of CHF
C89–C98, 2001.—Guanosine 5⬘-O-(3-thiotriphosphate) (11). Swelling-induced taurine release via the ICl,swell
(GTP␥S) activated the ICl,swell anion channel in N1E115
channel has been proposed to play a role in controlling
neuroblastoma cells in a swelling-independent manner.
GTP␥S-induced current was unaffected by ATP removal and osmotic regulation of vasopressin secretion in magno-
broadly selective tyrosine kinase inhibitors, demonstrating cellular neurons (13). The transformation of microglia
that phosphorylation events do not regulate G protein-depen- from an ameboid to a ramified shape is modulated by a
dent channel activation. Pertussis toxin had no effect on stretch-activated anion channel with biophysical char-
GTP␥S-induced current. However, cholera toxin inhibited acteristics similar to ICl,swell (20). Changes in cell vol-
the current ⬃70%. Exposure of cells to 8-bromoadenosine ume are postulated to play important signaling roles in
3⬘,5⬘-cyclic monophosphate did not mimic the effect of cholera cell metabolism, excitability, contraction, growth, pro-
toxin, and its inhibitory action was not prevented by treat- liferation, and apoptosis (32, 38, 47). Volume-induced
ment of cells with an inhibitor of adenylyl cyclase. These signaling may be mediated in part by changes in
results demonstrate that GTP␥S does not act through G␣i/o
ICl,swell activity.
GTPases and that G␣s/G␥ G proteins inhibit the channel
and/or channel regulatory mechanisms through cAMP-inde- Although volume-sensitive ion channels are ex-
pendent mechanisms. Swelling-induced activation of ICl,swell pressed ubiquitously and likely play important roles in
was stimulated two- to threefold by GTP␥S and inhibited by cellular physiology and pathophysiology, the molecular
10 mM guanosine 5⬘-O-(2-thiodiphosphate). The Rho GTPase identity of the channel responsible for ICl,swell is still
inhibitor Clostridium difficile toxin B inhibited both GTP␥S- unknown and the field is fraught with controversy (21,
and swelling-induced activation of ICl,swell. Taken together, 50, 56). In addition, the signaling mechanisms by
these findings indicate that Rho GTPase signaling pathways which cell swelling is transduced into channel activa-
regulate the ICl,swell channel via phosphorylation-indepen- tion are incompletely understood and may vary be-
dent mechanisms. tween different cell types. For example, some studies
cell volume regulation; Rho GTPase; anion channel suggest a requirement for serine/threonine or tyrosine
kinase phosphorylation (9, 12, 64) in ICl,swell activation,
whereas others have demonstrated that ATP hydroly-
THE ABILITY TO SENSE AND RESPOND to changes in volume is sis or phosphorylation events are not required (6, 49,
an essential and fundamental property of all cells (32, 59). In contrast, it has also been suggested that dephos-
47). In response to cell swelling, most vertebrate cells phorylation events mediate activation of the ICl,swell
activate an outwardly rectifying anion current termed channel (14, 61). The apparent variation in signaling
ICl,swell. The ICl,swell channel appears to play an impor- pathways that regulate the channel suggests three
tant role in regulating cell volume (reviewed in Refs. possibilities: 1) ICl,swell is due to the activity of more
43, 50, 57). than a single channel type, 2) channel regulation var-
In addition to its role in volume homeostasis, the ies between cell types, and/or 3) pharmacological and
ICl,swell channel may function in other physiological molecular disruption of signaling pathways has indi-
processes and may contribute to organ system patho- rect effects on channel activity.
physiology. For example, it has been suggested that the The uncertainty that exists over the signaling mech-
ICl,swell channel is a pathway for excitotoxic amino acid anisms that regulate ICl,swell and the molecular iden-
release in the central nervous system during cerebral tity of the channel(s) underscores the need for exten-
ischemia and trauma (4, 29, 51). ICl,swell is constitu- sive additional characterization of channel function
tively active in ventricular myocytes isolated from dogs and regulation. At present, it is known that ICl,swell can
Address for reprint requests and other correspondence: K. The costs of publication of this article were defrayed in part by the
Strange, Vanderbilt Univ. Medical Center, Anesthesiology Research payment of page charges. The article must therefore be hereby
Division, T-4202 Medical Center North, Nashville, TN 37232-2520 marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
(E-mail: kevin.strange@mcmail.vanderbilt.edu). solely to indicate this fact.
http://www.ajpcell.org 0363-6143/01 $5.00 Copyright © 2001 the American Physiological Society C89C90 REGULATION OF ICL,SWELL BY G PROTEINS
be activated by cell swelling (50, 57) and reduced in- idodiphosphate (AMP-PNP; Boehringer Mannheim, Ger-
tracellular ionic strength (8, 22, 44). Doroshenko and many) were added as sodium and lithium salts, respectively.
colleagues (15, 16) demonstrated that guanosine 5⬘-O- Metabolic inhibitors were added from concentrated stock
(3-thiotriphosphate) (GTP␥S) activates an outwardly solutions dissolved in DMSO. Final DMSO concentration in
the pipette solutions was 0.2%.
rectifying anion current with many of the properties of
Electrodes had direct current resistances of 3–5 M⍀. Cells
the ICl,swell channel in bovine chromaffin cells. More were used only if the series resistance was no greater than
recently, Nilius and co-workers (46, 64) have shown ⬃150% of the pipette resistance and the reversal potential
that activation of G proteins activates ICl,swell in endo- was within ⫾4 mV of the calculated value of ⫹14.7 mV for a
thelial cells. perfectly anion-selective channel. Reversal potentials signif-
The purpose of the present study was to investigate icantly below ⫹14.7 mV were taken as an indication of loss of
the role of G proteins in regulation of ICl,swell activation seal integrity.
in N1E115 neuroblastoma cells. Our results demon- An Axopatch 200A (Axon Instruments, Foster City, CA)
strate that GTP␥S activates ICl,swell in the absence of patch-clamp amplifier was used to voltage clamp N1E115
swelling. Current activation does not require phos- cells following gigaseal formation and attainment of whole
phorylation events and is insensitive to pertussis toxin. cell access. Command voltage generation, data digitization,
and data analysis were carried out on a Pentium II computer
However, cholera toxin and Clostridium difficile toxin using a DigiData 1200 AD/DA interface with pCLAMP 6
B significantly inhibited GTP␥S-induced current acti- software (Axon Instruments). Data were digitized at 5 kHz
vation. Swelling-induced activation of ICl,swell was and filtered at 0.5 kHz using an eight-pole Bessel filter
stimulated by GTP␥S and inhibited by guanosine 5⬘- (model 902; Frequency Devices, Haverhill, MA). Electrical
O-(2-thiodiphosphate) (GDPS) and toxin B. Taken connections to the amplifier were made using Ag-AgCl pellets
together, these results demonstrate that G␣s/G␥ and and 3 M KCl-agar bridges. Whole cell currents were mea-
Rho G protein signaling pathways are important reg- sured by varying membrane potential from ⫺80 to ⫹80 mV
ulators of the ICl,swell channel. at 80 mV/s every 5 s.
Measurement of relative cell volume changes. Whole cell
MATERIALS AND METHODS currents and volume changes were measured simultaneously
in single patch-clamped cells. Cells attached to the coverslip
Cell culture. N1E115 mouse neuroblastoma cells (Ameri- bottom of the patch-clamp bath chamber were visualized by
can Type Culture Collection, Manassas, VA) were cultured in video-enhanced differential interference contrast microscopy.
the presence of 5% CO2-95% air in high-glucose DMEM Optical sectioning (58) demonstrated that the cells main-
(GIBCO, Gaithersburg, MD) containing 25 mM HEPES, 10% tained a spherical morphology for at least 60 min after
fetal bovine serum, 50 U/ml penicillin and 50 g/ml strepto- attachment to the coverslip. Cells were routinely removed
mycin. Cells were used between passages 13 and 34. The from the bath chamber and replaced with fresh cells every
osmolality of the growth medium was 295–305 mosmol/ 30–45 min. Given that the cells have a spherical morphology,
kgH2O. relative cell volume change was determined as
Patch-clamp recordings. N1E115 cells were grown in
35-mm culture dishes and dissociated by brief treatment Relative cell volume ⫽ 共experimental CSA/control CSA兲3/2 (1)
with Ca2⫹- and Mg2⫹-free modified Hanks’ solution. Dissoci-
ated cells were allowed to reattach to the poly-L-lysine-coated where CSA is the cell cross-sectional area measured at a
coverslip bottom of a bath chamber (model R-26G; Warner single focal plane. In all CSA measurements described in this
Instrument, Hamden, CT) that was mounted onto the stage paper, we imaged cells at focal planes located at the point of
of a Nikon TE300 inverted microscope. Patch electrodes were maximum cell diameter.
pulled from 1.5-mm-outer diameter borosilicate glass micro- Cell images were recorded continuously throughout a
hematocrit tubes (Fisher Scientific, St. Louis, MO) that had patch-clamp experiment using a super VHS videocassette
been silanized with dimethyl-dichloro silane (Sigma Chemi- recorder (model SVO-2000; Sony Electronics, San Jose, CA)
cal, St. Louis, MO). Electrodes were not fire polished before and a Hamamatsu charge-coupled device camera (model
use. C2400; Hamamatsu Photonics, Hamamatsu City, Japan).
The bath solution contained (in mM) 70 N-methyl-D-gluca- CSAs of single cells were quantified by digitizing recorded
mine chloride, 5 MgSO4, 12 HEPES, 8 Tris, 5 glucose, 2 video images with an image-processing computer board (MV-
glutamine, 120 sucrose, and 0.4 or 1.3 CaCl2, (pH 7.4; osmo- 1000; MuTech, Woburn, MA) with 512 ⫻ 480 ⫻ 8-bit resolu-
lality ⫽ 300 mosmol/kgH2O). Bath osmolality was altered by tion and a Pentium II computer. Digitized images were
increasing or reducing sucrose concentration. displayed on the computer monitor, and cell borders were
Patch clamping was carried out using a pipette solution traced using a mouse and a computer-generated cursor. The
that contained (in mM) 125 CsCl, 10 HEPES, 10 1,2-bis(2- CSA of a traced region was determined by image analysis
aminophenoxy)ethane-N,N,N⬘,N⬘-tetraacetic acid (BAPTA; software (Optimas; Bioscan, Edmonds, WA). This image ac-
tetracesium salt; Molecular Probes, Eugene, OR), 1 MgSO4, quisition and analysis system allows detection of changes in
5 CsOH, 2 ATP, and 0.5 GTP or GTP␥S (tetralithium salt; CSA with an accuracy of ⫾2–3%.
Sigma; pH 7.2). In some pilot studies, 1 mM EGTA (Sigma) Data analysis. Whole cell currents were recorded within
was used instead of BAPTA. To prevent spontaneous cell 15–20 s after membrane rupture. The mean resting or base-
swelling, the osmolality of the pipette solution was hypotonic line current is defined as current measured before activation
(280 mosmol/kgH2O) with respect to the bath. by GTP␥S or cell swelling. Baseline current was subtracted
Experiments where the ATP requirement of channel acti- from all data points within a given record to correct for
vation was studied utilized a pipette solution containing 125 variability in resting current levels between different cells.
mM CsCl, 10 mM HEPES, 10 mM BAPTA, 1 mM EDTA Because of culture-to-culture variability in the response to
(Sigma), 5 mM CsOH, 0.5 mM GTP␥S, 40 M oligomycin, 5 GTP␥S, control measurements were performed in parallel
M iodoacetate, and 20 M rotenone. ATP or 5⬘-adenylylim- with all experimental treatments.REGULATION OF ICL,SWELL BY G PROTEINS C91
Rates of GTP␥S current activation and inactivation and
peak current were measured. Current activation is defined as
the point at which there is a continuous increase in current
amplitude above the baseline current (6). Rates of current
activation and inactivation were quantified by linear regres-
sion analysis.
Under control conditions, a small percentage (⬍10%) of
cells treated with GTP␥S showed no current activation. To
facilitate comparison with experimental treatments that may
have inhibited the GTP␥S response, nonactivating cells were
included when calculating the means ⫾ SE rate of GTP␥S-
induced current activation.
Whole cell anion current was also activated by cell swell-
ing. Rates of current activation and cell swelling were deter-
mined by linear regression analysis. For these studies, we
also quantified the cell volume set point of the channel. Cell
Fig. 2. Guanosine 5⬘-O-(2-thiodiphosphate) (GDPS) inhibits both
volume set point is defined as the relative cell volume at
the rate of activation and peak GTP␥S-induced current. For these
which current activation begins (6). experiments, the concentration of CsCl in the pipette solution was
Throughout the course of the experiments, a small per- reduced to 110 mM to osmotically compensate for the GDPS that
centage of cells exhibited bleb formation during current re- was added. The control pipette solution contained 110 mM CsCl and
cordings. These cells were excluded from our analyses. 10 mM Li3 citrate to control for the Li⫹ that was added with GDPS.
Statistical analysis. Data are presented as means ⫾ SE. Values are means ⫾ SE. *P ⬍ 0.05; ***P ⬍ 0.001. Number of
Statistical significance was determined using Student’s two- observations (n) is shown in parentheses.
tailed t-test for unpaired, independent means. When compar-
ing three or more groups, statistical significance was deter-
mined by one-way analysis of variance. Values of P ⬍ 0.05 RESULTS
were taken to indicate statistical significance.
GTP␥S activates ICl,swell in the absence of cell swell-
ing. Dialysis of N1E115 cells with 0.5 mM GTP␥S
activated an outwardly rectifying whole cell anion cur-
rent (Fig. 1A). Current activation began within 0.85 ⫾
0.08 min after the whole cell configuration was ob-
tained and reached a plateau within 5.3 ⫾ 0.8 min (n ⫽
19; Fig. 1B). Activation was transient, and current
levels returned to baseline 11.5 ⫾ 1.2 min (n ⫽ 16; Fig.
1B) after the plateau was reached. Cell swelling was
not observed during GTP␥S-induced current activation
(Fig. 1B). The mean relative cell volume at the peak
GTP␥S-induced current was 0.99 ⫾ 0.02 (n ⫽ 26).
G proteins cycle between an active GTP-bound state
and an inactive GDP-bound state. GDPS is a nonhy-
drolyzable analog of GDP that competes with GTP or
GTP analogs for the nucleotide binding sites on G
proteins, rendering them inactive. To examine further
the role of G proteins in whole cell anion current
activation, 10 mM GDPS was included in the pipette
solution. In the presence of 10 mM GDPS, only three
of six cells (50%) activated spontaneously with GTP␥S
compared with six of six cells in the paired control
group. GDPS also inhibited the rate of GTP␥S-
induced current activation and decreased the ampli-
tude of the peak current by 82% and 71%, respectively
(Fig. 2).
The outwardly rectifying current could be due to
activation of ICl,swell or Ca2⫹-dependent Cl⫺ channels
(ICl,Ca). G protein stimulation has been reported to
Fig. 1. Guanosine 5⬘-O-(3-thiotriphosphate) (GTP␥S) activates an
outwardly rectifying anion current in N1E115 neuroblastoma cells. activate ICl,Ca in several (28, 36), but not all (30, 45),
A: steady-state current-voltage (I-V) relationship of current acti- cell types. All experiments presented in this paper,
vated by GTP␥S in a single neuroblastoma cell. Calculated reversal however, were carried out using nominal Ca2⫹ in the
potential for a perfectly anion-selective channel is 14.7 mV. Mea- bath solution (0.4 mM) and a pipette solution contain-
sured reversal potential is 15.2 mV. B: example of simultaneous
current and volume measurements performed on a single patch-
ing the highly selective fast Ca2⫹ buffer BAPTA (10
clamped cell dialyzed with 0.5 mM GTP␥S. Time 0 refers to the time mM), demonstrating that ICl,Ca are not responsible for
at which recordings were initiated. the GTP␥S-induced current.C92 REGULATION OF ICL,SWELL BY G PROTEINS
Intracellular Ca2⫹ actually appeared to exert an
inhibitory effect on the GTP␥S current activation. Dur-
ing pilot studies, cells were patch clamped in the pres-
ence of 1.3 mM bath Ca2⫹ and a pipette solution con-
taining 1 mM EGTA instead of 10 mM BAPTA. Peak
GTP␥S-induced currents and rates of current activa-
tion and inactivation were unaffected by reduced Ca2⫹
buffering (data not shown). However, in the presence of
1 mM EGTA, GTP␥S-induced current activation oc-
curred in ⬍50% of patch-clamped cells. A much more
consistent activation of the GTP␥S current was ob-
served when 10 mM BAPTA was used to buffer intra-
cellular Ca2⫹. Current activation was detected in 93%
(n ⫽ 92) of control cells dialyzed with BAPTA-buffered
pipette solutions. The reason increased Ca2⫹ buffering
increases the frequency of current activation is un-
known. It is conceivable that a Ca2⫹-dependent process
antagonizes the stimulatory effect of GTP␥S.
To determine whether the GTP␥S-induced current is
due to the activity of the ICl,swell channel, we examined
its biophysical characteristics and volume sensitivity.
As shown in Table 1, the rectification ratio and relative
anion permeability of the channel responsible for the
GTP␥S current are the same as those observed for
ICl,swell. Furthermore, neither current exhibited signif- Fig. 3. Characteristics of GTP␥S- and swelling-induced anion cur-
icant voltage-dependent activation or inactivation rents. A: whole cell currents elicited by stepping membrane voltage
(Fig. 3A). from ⫺100 to ⫹100 mV in 20-mV steps from a holding potential of 0
The rate of spontaneous current inactivation (see mV. ICl,swell was activated by exposing cells to 0.5 mM GTP␥S or by
swelling with a hypotonic bath solution (200 mosmol/kgH2O). The
Fig. 1A) during GTP␥S stimulation was ⫺0.7 ⫾ 0.2 arrows indicate zero-current levels. B: cell shrinkage inhibits the
pA 䡠 pF⫺1 䡠 min⫺1 (n ⫽ 18). To determine whether the GTP␥S-induced current. Current and volume measurements shown
GTP␥S-induced current was volume sensitive, the cur- are for a single patch-clamped cell. Exposure of the cell to a hyper-
rent was allowed to reach a stable plateau level and tonic bath solution (400 mosmol/kgH2O or 400 mOsm) caused the
cells were then shrunken by exposure to a hypertonic GTP␥S-induced current to rapidly inactivate. The mean rate of
current inactivation for all experiments was ⬃8 times faster than
bath solution (400 mosmol/kgH2O). Cell shrinkage in- spontaneous inactivation observed in the absence of cell shrinkage
hibited the peak GTP␥S-induced current by 68 ⫾ 6% (see Fig. 1A).
(n ⫽ 5) at a rate of ⫺5.5 ⫾ 2.3 pA 䡠 pF⫺1 䡠 min⫺1 (n ⫽ 5;
see Fig. 3B). The rate of shrinkage-induced inactiva-
tion is nearly eight times faster than the rate of spon- GTP␥S-induced ICl,swell activation is not modulated
taneous current inactivation, demonstrating that the by ATP or phosphorylation reactions. Intracellular
GTP␥S-activated channel is sensitive to cell volume. ATP and nonhydrolyzable ATP analogs modulate but
On the basis of results shown in Figs. 1–3 and Table 1, are not essential for swelling-induced activation of
we conclude that stimulation of G proteins with GTP␥S ICl,swell in N1E115 cells (6). In a variety of cell types,
activates ICl,swell in the absence of cell swelling. nonhydrolyzable ATP analogs support normal ICl,swell
activity, indicating that phosphorylation events are not
involved in channel activation (6, 49). However, the
Table 1. Characteristics of Cl⫺ currents activated results of a number of studies have also suggested that
by GTP␥S and cell swelling protein kinases and phosphatases modulate channel
GTP␥S n Swelling n
activity (14, 61, 64). Recently, Voets et al. (64) proposed
that the stimulatory effect of GTP␥S on ICl,swell in
Rectification ratio 1.38 ⫾ 0.07 15 1.59 ⫾ 0.06 9 endothelial cells is mediated by tyrosine phosphoryla-
Anion permeability
PBr⫺/PCl⫺ 1.21 ⫾ 0.03 7 1.17 ⫾ 0.01 4
tion.
PF⫺/PCl⫺ 0.57 ⫾ 0.03 7 0.64 ⫾ 0.08* 5 Given these findings, we examined the effect of ty-
PI⫺/PCl⫺ 1.59 ⫾ 0.06 7 1.50 ⫾ 0.20* 4 rosine kinase inhibitors on GTP␥S-induced activation
PNO3⫺/PCl⫺ 1.40 ⫾ 0.05 7 1.27 ⫾ 0.20* 5 of ICl,swell in N1E115 cells. Cells were treated with the
Values are means ⫾ SE; n, no. of observations. Rectification ratio broadly selective tyrosine kinase inhibitors genistein
is the ratio of currents measured at ⫹60 and ⫺60 mV. Relative anion (100 M) or tyrphostin A51 (100 M). Two experimen-
permeabilities (PX/PCl) were calculated using the Goldman-Hodgkin- tal protocols were used. Cells were patch clamped with
Katz equation and measured changes in the reversal potential in- inhibitors present only in the pipette solution or with
duced by complete replacement of bath Cl⫺ with the test anion.
* Data from Bond et al. (6). Parameters measured for the guanosine
the inhibitors present in both the pipette and bath
5⬘-O-(3-thiotriphosphate) (GTP␥S)- and swelling-induced currents solutions. In the latter case, cells were preincubated for
were not significantly (P ⬎ 0.05) different. 7–21 min with inhibitors before patch clamping. Pi-REGULATION OF ICL,SWELL BY G PROTEINS C93
pette solutions were kept on ice and bath and pipette
solutions were remade every hour to minimize prob-
lems associated with breakdown of the inhibitors (64).
Results using the two protocols were not significantly
different for either genistein (P ⬎ 0.7) or tyrphostin
A51 (P ⬎ 0.3), and the data were therefore averaged
and are presented in Fig. 4. GTP␥S activated ICl,swell in
all cells treated with tyrphostin A51 and DMSO and in
9 of 10 cells treated with genistein. Neither inhibitor
significantly altered the rate of current activation or
peak current (Fig. 4).
To corroborate the inhibitor studies and to further
test for the involvement of phosphorylation events in
the GTP␥S-induced activation of ICl,swell, cells were Fig. 5. GTP␥S-induced current activation is not modulated by intra-
metabolically poisoned and patch clamped with Mg2⫹- cellular ATP or phosphorylation reactions. All cells were patch
free pipette solutions containing 1 mM EDTA. Cellular clamped with a Mg2⫹-free pipette solution containing the metabolic
ATP production was blocked by incubating cells for inhibitors oligomycin (40 M), rotenone (20 M), and iodoacetate (5
M) and were also preincubated for 10–30 min in a bath solution
10–30 min in bath solution containing 5 mM 2-deoxy- containing 2-deoxyglucose (5 mM) and rotenone (100 nM). ATP
glucose and 100 nM rotenone. In addition, the pipette removal or substitution with 5⬘-adenylylimidodiphosphate (AMP-
solution contained 40 M oligomycin, 20 M rotenone, PNP) had no significant (P ⬎ 0.1) effect on the rate of activation and
and 5 M iodoacetate. peak GTP␥S-induced current. Values are means ⫾ SE. Number of
observations (n) is shown in parentheses.
In metabolically poisoned cells, removal of ATP from
the pipette solution or replacement with 2 mM AMP-
PNP had no significant effect on the rate of GTP␥S- for the involvement of G␣i/o G proteins in ICl,swell reg-
induced ICl,swell activation or peak current (Fig. 5). ulation, N1E115 cells were preincubated with 100
These results demonstrate clearly that ATP hydrolysis ng/ml pertussis toxin for 6–10 h before patch-clamp
does not play a role in the GTP␥S signaling pathway. measurements were taken. Pretreatment with 1 g/ml
The number of metabolically poisoned cells in which of pertussis toxin for ⬎4 h is sufficient to completely
GTP␥S triggered current activation was 9 of 9 in the ADP-ribosylate the ␣-subunit of Gi in N1E115 cells (7).
presence of 2 mM ATP, 17 of 18 with 0 mM ATP, and 16 Pertussis toxin had no significant (P ⬎ 0.5) effect on
of 16 cells with 2 mM AMP-PNP in the pipette solution. ICl,swell activation. Rates of GTP␥S-induced current
Cholera toxin inhibits GTP␥S-induced activation of activation and peak currents (means ⫾ SE) in the
ICl,swell via cAMP-independent mechanisms. G proteins presence and absence of pertussis toxin were 1.8 ⫾ 0.4
are categorized into three families: heterotrimeric, low- pA 䡠 pF⫺1 䡠 min⫺1 (n ⫽ 8) and 4.9 ⫾ 0.8 pA/pF (n ⫽ 7),
molecular-weight (small) monomeric, and high-molec- and 1.4 ⫾ 0.4 pA 䡠 pF⫺1 䡠 min⫺1 (n ⫽ 10) and 5.0 ⫾ 0.9
ular-weight (large) monomeric (3). The heterotrimeric
G proteins are composed of three subunits termed ␣, ,
and ␥. Pertussis and cholera toxins are commonly used
to determine whether a heterotrimeric G protein fam-
ily is involved in a specific signaling pathway.
Pertussis toxin catalyzes the ADP-ribosylation and
inactivation of members of the G␣i/o subfamily. To test
Fig. 6. G␣s signaling pathways inhibit GTP␥S-induced activation of
ICl,swell via cAMP-independent mechanisms. Incubation of cells over-
night with 100 ng/ml cholera toxin (CTX) inhibited GTP␥S-induced
current activation ⬃70%. Exposure of cells to the adenylyl cyclase
Fig. 4. Tyrosine kinase inhibitors have no significant (P ⬎ 0.5) effect inhibitor 2⬘-5⬘-dideoxyadenosine (DDA; 100 M) had no effect on the
on GTP␥S-induced current activation. Genistein (100 M) and tyr- inhibitory action of CTX. Overnight incubation of cells with 500 M
phostin A51 (100 M) were dissolved in DMSO and added to the 8-bromoadenosine 3⬘,5⬘-cyclic monophosphate (8-BrcAMP) did not
patch pipette and bath solutions at a final DMSO concentration of mimic the effects of CTX treatment. Values are means ⫾ SE. *P ⬍
0.1%. Control solutions contained 0.1% DMSO. Values are means ⫾ 0.05; **P ⬍ 0.01; ***P ⬍ 0.001. Number of observations (n) is shown
SE. Number of observations (n) is shown in parentheses. in parentheses.C94 REGULATION OF ICL,SWELL BY G PROTEINS
pA/pF (n ⫽ 5), respectively. These results indicate that
that G␣i/o proteins do not mediate the effect of GTP␥S.
Cholera toxin catalyzes the ADP-ribosylation of the
G␣s subfamily of G proteins, rendering them constitu-
tively active. Overnight incubation with 100 ng/ml
cholera toxin reduced GTP␥S-induced ICl,swell activa-
tion and peak current by ⬃70% (Fig. 6).
Activation of G␣s activates adenylyl cyclase and ele-
vates intracellular cAMP. This suggests that cAMP
might mediate the inhibitory effect of cholera toxin. To
test whether inhibition occurred via a cAMP-depen-
dent mechanism, cells were incubated overnight with
cholera toxin and 2⬘-5⬘-dideoxyadenosine (DDA; 100
M), an inhibitor of adenylyl cyclase. As shown in Fig.
6, the inhibitory effect of cholera toxin was unaltered
by DDA.
In an effort to mimic the inhibitory action of cholera
toxin, we incubated cells overnight with 500 M 8-bro-
moadenosine 3⬘,5⬘-cyclic monophosphate (8-BrcAMP)
and included it in the patch pipette solution. Current
activation was not significantly different in 8-BrcAMP-
treated cells (Fig. 6). We conclude that G␣s inhibits
GTP␥S-induced activation of ICl,swell by directly inhib-
iting the channel and/or channel regulatory mecha-
nisms. Fig. 8. G protein signaling pathways regulate swelling-induced ac-
GTP␥S-induced activation of ICl,swell is mediated by tivation of ICl,swell. A: effect of GTP␥S treatment on swelling-induced
Rho GTPases. The low-molecular-weight monomeric G current activation in a single N1E115 cell. After decay of the GTP␥S
proteins include the Ras, Rho, Rab, Arf, and Ran fam- current, cell was swollen by exposure to a hypotonic (200 mosmol/
kgH2O) bath solution. Inset: mean rates of swelling-induced current
ilies (3). Ten classes of mammalian Rho GTPases have activation in cells dialyzed with 0.5 mM GTP or GTP␥S. GTP␥S
been identified (5): Rho (A, B, C isoforms), Rac, Cdc42, increased rate of swelling-induced current activation 2- to 3-fold. B:
Rnd1/Rho6, Rnd2/Rho7, Rnd3/RhoE, Rho D, RhoG, dialysis of cells with 10 mM GDPS or overnight exposure to 10
TC10, and TTF. To test for the involvement of Rho G ng/ml toxin B inhibited swelling-induced ICl,swell activation. For
experiments with GDPS, CsCl concentration in the pipette solution
proteins in ICl,swell regulation, N1E115 cells were incu- was reduced to 110 mM to maintain osmolality. The control pipette
bated with 1 ng/ml C. difficile toxin B for 19–24 h. solution contained 110 mM CsCl and 10 mM Li3 citrate to control for
Toxin B catalyzes the UDP-glucosylation of the Rho the Li⫹ that was added with GDPS. Values are means ⫾ SE. *P ⬍
0.03; ***P ⬍ 0.001. Number of observations (n) is shown in paren-
subfamily of monomeric G proteins including Rho, Rac, theses.
and Cdc42 (1). Incubation with toxin B inhibited the
rate of GTP␥S-induced current activation and peak
current ⬃70% (Fig. 7). These results demonstrate that Swelling-induced activation of ICl,swell is modulated
Rho GTPase signaling pathways regulate GTP␥S-in- by Rho signaling pathways. To determine whether G
duced activation of ICl,swell. protein signaling pathways modulate swelling-induced
activation of ICl,swell, cells were dialyzed with GTP␥S
and exposed to a hypotonic bath (100 mosmol/kgH2O
reduction in bath osmolality) after the GTP␥S-induced
current had inactivated (Fig. 8A). Spontaneous inacti-
vation of the GTP␥S-induced current did not preclude
further activation of ICl,swell with a swelling stimulus.
The mean ⫾ SE volume set points for current activa-
tion in the presence and absence of GTP␥S were 1.13 ⫾
0.02 (n ⫽ 22) and 1.18 ⫾ 0.03 (n ⫽ 9), respectively, and
were not significantly (P ⬎ 0.1) different. However,
GTP␥S stimulated the rate of swelling-induced current
activation two- to threefold (Fig. 8A).
To determine whether G proteins are required for
swelling-induced current activation, 10 mM GDPS
was included in the pipette solution in the absence of
Fig. 7. GTP␥S-induced activation of ICl,swell is regulated by Rho GTP␥S. GDPS inhibited swelling-induced ICl,swell ac-
GTPases. Incubation of cells overnight with 1 ng/ml C. difficile toxin
B inhibited the rate of current activation and peak current. Values tivation by ⬃80% (Fig. 8B) and significantly (P ⬍
are means ⫾ SE. *P ⬍ 0.05. Number of observations (n) is shown in 0.002) increased mean ⫾ SE channel volume set point
parentheses. from 1.1 ⫾ 0.02 (n ⫽ 4) to 1.22 ⫾ 0.01 (n ⫽ 4).REGULATION OF ICL,SWELL BY G PROTEINS C95
Overnight exposure of cells to 10 ng/ml toxin B inhib- and increased protein phosphorylation, whereas the
ited current activation ⬃70% (Fig. 8B) without altering stimulatory effect of cAMP occurs in a phosphoryla-
channel volume set point (control ⫽ 1.1 ⫾ 0.02, n ⫽ 7; tion-independent fashion. More recently, Shimizu et al.
toxin B ⫽ 1.1 ⫾ 0.03, n ⫽ 5). Taken together, these (55) demonstrated that cAMP enhances ICl,swell activa-
data suggest strongly that Rho GTPase signaling path- tion in Intestine 407 cells by a PKA-independent mech-
ways regulate swelling-induced activation of ICl,swell. anism.
We tested for the involvement of adenylyl cyclase
DISCUSSION and cAMP in mediating the effect of cholera toxin on
GTP␥S-induced activation of ICl,swell. The inhibitory
ICl,swell is a swelling-activated anion current that action of cholera toxin was not mimicked by overnight
appears to be expressed ubiquitously in mammalian exposure to 8-BrcAMP and was not blocked by the
cells (43, 50, 57). This current is outwardly rectifying, adenylyl cyclase inhibitor DDA (Fig. 6), indicating that
exhibits an Eisenman type I anion permeability se- G␣s functions through cAMP-independent pathways.
quence, and is inhibited by a wide variety of pharma- Recent studies have demonstrated that G␣s is capa-
cological agents. The ICl,swell channel plays an impor- ble of modulating ion channel activity in the absence of
tant role in cell volume regulation (reviewed in Refs. adenylyl cyclase and PKA function (e.g., Refs. 31, 37).
43, 50, 57) and may participate in the control of other G␣s may directly inhibit the ICl,swell channel and/or
physiological processes such as cell metabolism, mem- may act on signaling pathways that regulate channel
brane excitability, and cell growth, proliferation, and activation. G␥ may also inhibit ICl,swell in a cAMP-
apoptosis (32, 38, 47). independent manner. It has been demonstrated that
The molecular identity of the channel responsible for G␥ subunits can directly modulate ion channel activ-
ICl,swell is uncertain and controversial. P-glycoprotein ity in a stimulatory or inhibitory fashion (10, 52). It is
and pICln have both been suggested to function as the interesting to speculate that spontaneous inactivation
ICl,swell channel. However, most workers in the field no of ICl,swell (Fig. 1) may be mediated by GTP␥S stimula-
longer consider these proteins to be viable channel tion of G␣s or G␥ signaling mechanisms. Extensive
candidates (21, 50, 56). More recently, ICl,swell was molecular biological studies are required to fully deter-
proposed to be due to the activity of ClC-3, a member of mine which heterotrimeric subunit inhibits ICl,swell ac-
the ClC superfamily of voltage-gated anion channels tivity and to delineate the mechanism by which this
(19). The findings of Duan and co-workers (19) on ClC-3 inhibition occurs.
have not yet been reproduced by other laboratories, Regulation of ICl,swell is mediated at least in part by
and a variety of recent observations have begun to small monomeric Rho GTPases. C. difficile toxin B
raise doubts about a widespread role for this channel in inhibited both GTP␥S- and swelling-induced channel
ICl,swell function (35, 48, 65). activation (Figs. 7 and 8). Toxin B inhibits Rho, Rac,
Regulation of ICl,swell. Cell swelling and reduced in- and Cdc42 Rho GTPases (1). Recently, Nilius et al. (46)
tracellular ionic strength activate ICl,swell (8, 22, 44, 50, demonstrated that ICl,swell in endothelial cells is inhib-
57). The signal transduction mechanisms responsible ited by Clostridium C3 exoenzyme. The C3 exoenzyme
for channel activation are unclear. Recently, Nilius and is a selective inhibitor of Rho A, B, and C (1). C3
co-workers (46) demonstrated that ICl,swell in endothe- exoenzyme does not readily permeate plasma mem-
lial cells is activated transiently by GTP␥S in a pertus- branes, and, in our study, we were unable to ensure
sis toxin-insensitive manner. A similar outwardly rec- that it was loaded effectively into N1E115 cells. How-
tifying anion current exhibiting many of the basic ever, assuming that ICl,swell in endothelial and N1E115
properties of ICl,swell was originally shown by Dorosh- cells are controlled by similar mechanisms, our find-
enko and colleagues (15, 16) to be triggered by GTP␥S ings in conjunction with those of Nilius et al. (46) argue
in bovine chromaffin cells. that Rho A, B, and/or C are important regulators of this
ICl,swell in N1E115 neuroblastoma cells is also acti- current.
vated transiently by GTP␥S (Fig. 1). Activation of The molecular details of how Rho GTPase signaling
ICl,swell by swelling is dramatically stimulated by pathways regulate ICl,swell are unknown. Rho GTPases
GTP␥S (Fig. 8A) and inhibited by GDPS (Fig. 8B). have been implicated in a variety of cellular processes
Taken together, these results indicate that cell swell- including actin cytoskeletal organization, membrane
ing is transduced into channel activation at least in trafficking, transcriptional activation, cell growth, mo-
part via G protein signaling pathways. tility, and morphogenesis (24, 53, 63). Alterations in
GTP␥S current activation occurs via a pertussis tox- cytoskeletal organization regulated by Rho GTPases
in-insensitive mechanism. However, cholera toxin sig- have been studied extensively and include the forma-
nificantly inhibited GTP␥S-induced current develop- tion of focal adhesions, actin stress fibers, lamellipodia,
ment, suggesting that G␣s regulates ICl,swell (Fig. 6). membrane ruffles, and filopodia (24, 42, 53, 60).
Activation of G␣s stimulates adenylyl cyclase. Several Changes in the organization of the actin cytoskeleton
studies have recently demonstrated that ICl,swell is have long been implicated in regulating volume-sensi-
modulated by cAMP. Du and Sorota (17) observed both tive transport pathways (32, 41, 50, 57), including the
inhibitory and stimulatory effects of cAMP in dog atrial ICl,swell channel (34, 54, 62). Mechanical forces have
cells. They showed that inhibition of ICl,swell is due to been shown to directly modulate G protein activity (23)
cAMP-induced activation of protein kinase A (PKA) as well as cytoskeletal architecture (25, 26). It is at-C96 REGULATION OF ICL,SWELL BY G PROTEINS
tractive to postulate then that swelling-induced alter- has been shown for inhibitors of arachidonic acid me-
ations in Rho GTPase activity may alter cytoskeletal tabolism (39), or they may have other nonspecific ef-
structure, which in turn triggers ICl,swell activation. fects. Conclusions drawn from pharmacological studies
Alternatively, cell swelling may directly alter the orga- of phosphorylation-dependent regulation of ICl,swell
nization of the cytoskeleton. Cytoskeletal changes should be corroborated where possible by metabolic
could conceivably activate Rho G protein signaling inhibition and Mg2⫹ and ATP removal experiments
pathways that regulate ICl,swell. such as those shown in Fig. 5 and described by Bond et
Role of protein phosphorylation in ICl,swell regulation. al. (6).
Phosphorylation has emerged as an extremely con- Downstream effectors of Rho GTPases include vari-
founding variable in understanding how ICl,swell is reg- ous protein and lipid kinases (2, 5), a finding consistent
ulated. In numerous cell types, nonhydrolyzable ATP with the postulated role of Rho kinases in regulating
analogs support normal swelling-induced current acti- ICl,swell in endothelial cells (46). However, Rho GTPases
vation (6, 49, 50, 57), an observation that argues can also regulate cellular processes in a phosphoryla-
strongly against a role for phosphorylation events in tion-independent manner. For example, Rho GTPase-
channel regulation. However, serine/threonine phos- regulated actin polymerization and cross-linking in
phorylation (e.g., Refs. 9, 40), serine/threonine dephos- vitro occur in the absence of ATP and phosphorylation
phorylation (18, 19), tyrosine phosphorylation (e.g., reactions (27). The nonkinase effectors of Rho GTPases
Refs. 12, 33), and tyrosine dephosphorylation (14, 61) include various scaffolding proteins that play impor-
have been proposed to play roles in channel activation. tant roles in actin cytoskeletal organization (2, 5).
Even experiments from the same laboratory have Phosphorylation-independent activation of ICl,swell in
generated confounding results. Szücs et al. (59) failed N1E115 cells may be mediated by changes in the in-
to detect an inhibitory effect of the tyrosine kinase teraction of Rho-regulated scaffolding proteins with
inhibitor genistein on swelling-induced ICl,swell activa- the ICl,swell channel and/or associated regulatory ma-
tion in bovine endothelial cells. However, more re- chinery.
cently, Voets et al. (64) demonstrated that this com- To conclude, we have demonstrated that ICl,swell in
pound inhibited activation of the current by both neuroblastoma cells is regulated by G protein signaling
GTP␥S and swelling. Voets et al. (64) suggested that pathways. Swelling- and GTP␥S-induced channel acti-
the discrepant findings may have been due to low vation are mediated at least in part by Rho GTPases.
solubility and stability of genistein. Extensive additional studies utilizing molecular and
In our study, we were unable to detect any inhibitory electrophysiological approaches are required to fully
action of genistein or tyrphostin A51 on GTP␥S-in- elucidate the mechanisms by which G protein-depen-
duced activation of ICl,swell (Fig. 4). Because of potential dent regulation occurs.
problems associated with the use of these drugs (see
Ref. 64), we examined the combined effects of intracel- This work was supported by National Institutes of Health Grants
NS-30591 and DK-51610. A.Y. Estevez was supported by a National
lular ATP and Mg2⫹ removal on current activation. As Science Foundation postdoctoral fellowship. T. Bond was supported
shown in Fig. 5, ATP and Mg2⫹ removal or replace- by a Stroke Investigator Award from the Heart and Stroke Founda-
ment of ATP with the nonhydrolyzable analog AMP- tion of Ontario.
PNP in metabolically poisoned cells has no effect on
GTP␥S-induced current activation. Similarly, we have REFERENCES
shown previously that swelling-induced ICl,swell activa-
1. Aktories K. Rho proteins: targets for bacterial toxins. Trends
tion in neuroblastoma cells occurs normally in the Microbiol 5: 282–288, 1997.
absence of hydrolyzable ATP (6). Indeed, even dialysis 2. Aspenstrom P. Effectors for the Rho GTPases. Curr Opin Cell
of metabolically poisoned cells with Mg2⫹- and ATP- Biol 11: 95–102, 1999.
free pipette solutions containing AMP-PNP and alka- 3. Barritt GJ and Gregory RB. An evaluation of strategies avail-
able for the identification of GTP-binding proteins required in
line phosphatase to dephosphorylate proteins has no intracellular signalling pathways. Cell Signal 9: 207–218, 1997.
effect on current activation (6). On the basis of these 4. Basarsky TA, Feighan D, and MacVicar BA. Glutamate
results, we conclude that phosphorylation signaling release through volume-activated channels during spreading
pathways do not regulate ICl,swell in N1E115 neuroblas- depression. J Neurosci 19: 6439–6445, 1999.
toma cells during GTP␥S- or swelling-induced activa- 5. Bishop AL and Hall A. Rho GTPases and their effector pro-
teins. Biochem J 348: 241–255, 2000.
tion. 6. Bond T, Basavappa S, Christensen M, and Strange K. ATP
The requirement for phosphorylation observed in dependence of the ICl,swell channel varies with rate of cell swell-
other cell types may reflect the existence of distinct ing. Evidence for two modes of channel activation. J Gen Physiol
channel types. Alternatively, it may reflect the exis- 113: 441–456, 1999.
7. Bozou JC, de Nadai F, Vincent JP, and Kitabgi P. Neuro-
tence of multiple signaling/regulatory pathways in- tensin, bradykinin and somatostatin inhibit cAMP production in
volved in channel activation. These pathways could be neuroblastoma N1E115 cells via both pertussis toxin sensitive
cell specific, they may reflect the physiological status of and insensitive mechanisms. Biochem Biophys Res Commun
the cell, and/or they may be sensitive to experimental 161: 1144–1150, 1989.
parameters such as the mechanism or rate of cell 8. Cannon CL, Basavappa S, and Strange K. Intracellular ionic
strength regulates the volume sensitivity of a swelling-activated
swelling (see Ref. 6). It is also distinctly possible that anion channel. Am J Physiol Cell Physiol 275: C416–C422, 1998.
pharmacological agents used to inhibit kinases and 9. Chou CY, Shen MR, Hsu KS, Huang HY, and Lin HC.
phosphatases may directly block the channel, such as Involvement of PKC-␣ in regulatory volume decrease responsesREGULATION OF ICL,SWELL BY G PROTEINS C97
and activation of volume-sensitive chloride channels in human 32. Lang F, Busch GL, Ritter M, Volkl H, Waldegger S, Gul-
cervical cancer HT-3 cells. J Physiol (Lond) 512: 435–448, 1998. bins E, and Haussinger D. Functional significance of cell
10. Clapham DE and Neer EJ. G protein ␥ subunits. Annu Rev volume regulatory mechanisms. Physiol Rev 78: 247–306, 1998.
Pharmacol Toxicol 37: 167–203, 1997. 33. Lepple-Wienhues A, Szabo I, Laun T, Kaba NK, Gulbins E,
11. Clemo HF, Stambler BS, and Baumgarten CM. Swelling- and Lang F. The tyrosine kinase p56lck mediates activation of
activated chloride current is persistently activated in ventricular swelling-induced chloride channels in lymphocytes. J Cell Biol
myocytes from dogs with tachycardia-induced congestive heart 141: 281–286, 1998.
failure. Circ Res 84: 157–165, 1999. 34. Levitan I, Almonte C, Mollard P, and Garber SS. Modula-
12. Crepel V, Panenka W, Kelly ME, and MacVicar BA. Mito- tion of a volume-regulated chloride current by F-actin. J Membr
gen-activated protein and tyrosine kinases in the activation of Biol 147: 283–294, 1995.
astrocyte volume-activated chloride current. J Neurosci 18: 35. Li X, Shimada K, Showalter LA, and Weinman SA. Biophys-
1196–1206, 1998. ical properties of ClC-3 differentiate it from swelling-activated
13. Deleuze C, Duvoid A, and Hussy N. Properties and glial chloride channels in CHO-K1 cells. J Biol Chem. 275: 35944 –
origin of osmotic-dependent release of taurine from the rat su- 35998, 2000.
praoptic nucleus. J Physiol (Lond) 507: 463–471, 1998. 36. Loirand G, Pacaud P, Mironneau C, and Mironneau J.
14. Doroshenko P. Pervanadate inhibits volume-sensitive chloride GTP-binding proteins mediate noradrenaline effects on calcium
current in bovine chromaffin cells. Pflügers Arch 435: 303–309, and chloride currents in rat portal vein myocytes. J Physiol
1998. (Lond) 428: 517–529, 1990.
15. Doroshenko P and Neher E. Volume-sensitive chloride con- 37. Lu T, Lee HC, Kabat JA, and Shibata EF. Modulation of rat
ductance in bovine chromaffin cell membrane. J Physiol (Lond) cardiac sodium channel by the stimulatory G protein ␣ subunit.
449: 197–218, 1992. J Physiol (Lond) 518: 371–384, 1999.
16. Doroshenko P, Penner R, and Neher E. Novel chloride con- 38. Maeno E, Ishizaki Y, Kanaseki T, Hazama A, and Okada Y.
ductance in the membrane of bovine chromaffin cells activated Normotonic cell shrinkage because of disordered volume regula-
by intracellular GTP␥S. J Physiol (Lond) 436: 711–724, 1991. tion is an early prerequisite to apoptosis. Proc Natl Acad Sci
17. Du XY and Sorota S. Modulation of dog atrial swelling-induced USA 97: 9487–9492, 2000.
chloride current by cAMP: protein kinase A-dependent and -in- 39. McManus M, Serhan C, Jackson P, and Strange K. Keto-
dependent pathways. J Physiol (Lond) 500: 111–122, 1997. conazole blocks organic osmolyte efflux independently of its
18. Duan D, Cowley S, Horowitz B, and Hume JR. A serine effect on arachidonic acid conversion. Am J Physiol Cell Physiol
residue in ClC-3 links phosphorylation-dephosphorylation to 267: C266–C271, 1994.
chloride channel regulation by cell volume. J Gen Physiol 113: 40. Meyer K and Korbmacher C. Cell swelling activates ATP-
57–70, 1999. dependent voltage-gated chloride channels in M-1 mouse cortical
19. Duan D, Winter C, Cowley S, Hume JR, and Horowitz B. collecting duct cells. J Gen Physiol 108: 177–193, 1996.
41. Moustakas A, Theodoropoulos PA, Gravanis A, Hauss-
Molecular identification of a volume-regulated chloride channel.
inger D, and Stournaras C. The cytoskeleton in cell volume
Nature 390: 417–421, 1997.
regulation. Contrib Nephrol 123: 121–134, 1998.
20. Eder C, Klee R, and Heinemann U. Involvement of stretch-
42. Narumiya S, Ishizaki T, and Watanabe N. Rho effectors and
activated Cl⫺ channels in ramification of murine microglia.
reorganization of actin cytoskeleton. FEBS Lett 410: 68–72,
J Neurosci 18: 7127–7137, 1998.
1997.
21. Eggermont J, Buyse G, Voets T, Tytgat J, Droogmans G,
43. Nilius B, Eggermont J, Voets T, and Droogmans G. Volume-
and Nilius B. Is there a link between protein pICln and volume-
activated Cl⫺ channels. Gen Pharmacol 27: 1131–1140, 1996.
regulated anion channels? Biochem J 331: 347–349, 1998.
44. Nilius B, Prenen J, Voets T, Eggermont J, and Droogmans
22. Emma F, McManus M, and Strange K. Intracellular electro- G. Activation of volume-regulated chloride currents by reduction
lytes regulate the volume set point of the organic osmolyte/anion of intracellular ionic strength in bovine endothelial cells.
channel VSOAC. Am J Physiol Cell Physiol 272: C1766–C1775, J Physiol (Lond) 506: 353–361, 1998.
1997. 45. Nilius B, Prenen J, Voets T, Van den Bremt K, Eggermont
23. Gudi S, Nolan JP, and Frangos JA. Modulation of GTPase J, and Droogmans G. Kinetic and pharmacological properties
activity of G proteins by fluid shear stress and phospholipid of the calcium-activated chloride-current in macrovascular en-
composition. Proc Natl Acad Sci USA 95: 2515–2519, 1998. dothelial cells. Cell Calcium 22: 53–63, 1997.
24. Hall A. Rho GTPases and the actin cytoskeleton. Science 279: 46. Nilius B, Voets T, Prenen J, Barth H, Aktories K, Kaibuchi
509–514, 1998. K, Droogmans G, and Eggermont J. Role of Rho and Rho
25. Hamill O and McBride D. Molecular clues to mechanosensi- kinase in the activation of volume-regulated anion channels in
tivity. Biophys J 65: 17–18, 1993. bovine endothelial cells. J Physiol (Lond) 516: 67–74, 1999.
26. Ingber DE. Tensegrity: the architectural basis of cellular mech- 47. O’Neill WC. Physiological significance of volume-regulatory
anotransduction. Annu Rev Physiol 59: 575–599, 1997. transporters. Am J Physiol Cell Physiol 276: C995–C1011, 1999.
27. Katanaev VL and Wymann MP. GTP␥S-induced actin poly- 48. Obermuller N, Gretz N, Kriz W, Reilly RF, and Witzgall R.
merisation in vitro: ATP- and phosphoinositide-independent sig- The swelling-activated chloride channel ClC-2, the chloride
nalling via Rho-family proteins and a plasma membrane-associ- channel ClC-3, and ClC-5, a chloride channel mutated in kidney
ated guanine nucleotide exchange factor. J Cell Sci 111: 1583– stone disease, are expressed in distinct subpopulations of renal
1594, 1998. epithelial cells. J Clin Invest 101: 635–642, 1998.
28. Kibble JD, Greenwood SL, Clarson LH, and Sibley CP. A 49. Oike M, Droogmans G, and Nilius B. The volume-activated
Ca2⫹-activated whole-cell Cl⫺ conductance in human placental chloride current in human endothelial cells depends on intracel-
cytotrophoblast cells activated via a G protein. J Membr Biol lular ATP. Pflügers Arch 427: 184–186, 1994.
151: 131–138, 1996. 50. Okada Y. Volume expansion-sensing outward-rectifier Cl⫺
29. Kimelberg HK and Mongin AA. Swelling-activated release of channel: fresh start to the molecular identity and volume sensor.
excitatory amino acids in the brain: relevance for pathophysiol- Am J Physiol Cell Physiol 273: C755–C789, 1997.
ogy. Contrib Nephrol 123: 240–257, 1998. 51. Phillis JW, Song D, and O’Regan MH. Inhibition by anion
30. Kotera T and Brown PD. Cl⫺ current activation in choroid channel blockers of ischemia-evoked release of excitotoxic and
plexus epithelial cells involves a G protein and protein kinase A. other amino acids from rat cerebral cortex. Brain Res 758: 9–16,
Am J Physiol Cell Physiol 266: C536–C540, 1994. 1997.
31. Lader AS, Xiao YF, Ishikawa Y, Cui Y, Vatner DE, Vatner 52. Qin N, Platano D, Olcese R, Stefani E, and Birnbaumer L.
SF, Homcy CJ, and Cantiello HF. Cardiac Gs␣ overexpression Direct interaction of G␥ with a C-terminal G␥-binding domain
enhances L-type calcium channels through an adenylyl cyclase of the Ca2⫹ channel ␣1 subunit is responsible for channel inhi-
independent pathway. Proc Natl Acad Sci USA 95: 9669–9674, bition by G protein-coupled receptors. Proc Natl Acad Sci USA
1998. 94: 8866–8871, 1997.C98 REGULATION OF ICL,SWELL BY G PROTEINS
53. Ridley AJ. Rho: theme and variations. Curr Biol 6: 1256–1264, 60. Takai Y, Sasaki T, Tanaka K, and Nakanishi H. Rho as a
1996. regulator of the cytoskeleton. Trends Biochem Sci 20: 227–231,
54. Shen MR, Chou CY, Hsu KF, Hsu KS, and Wu ML. Modula- 1995.
tion of volume-sensitive Cl⫺ channels and cell volume by actin 61. Thoroed SM, Bryan-Sisneros A, and Doroshenko P. Protein
filaments and microtubules in human cervical cancer HT-3 cells. phosphotyrosine phosphatase inhibitors suppress regulatory vol-
Acta Physiol Scand 167: 215–225, 1999. ume decrease and the volume-sensitive Cl⫺ conductance in
55. Shimizu T, Morishima S, and Okada Y. Ca2⫹-sensing re- mouse fibroblasts. Pflügers Arch 438: 133–140, 1999.
ceptor-mediated regulation of volume-sensitive Cl⫺ channels 62. Tilly BC, Edixhoven MJ, Tertoolen LG, Morii N, Saitoh Y,
in human epithelial cells. J Physiol (Lond) 528: 457–472, Narumiya S, and de Jonge HR. Activation of the osmo-
2000. sensitive chloride conductance involves P21rho and is accompa-
56. Strange K. Molecular identity of the outwardly rectifying,
nied by a transient reorganization of the F-actin cytoskeleton.
swelling-activated anion channel: time to reevaluate pICln. J Gen
Mol Biol Cell 7: 1419–1427, 1996.
Physiol 111: 617–622, 1998.
57. Strange K, Emma F, and Jackson PS. Cellular and molecular 63. Van Aelst L and D’Souza-Schorey C. Rho GTPases and sig-
physiology of volume-sensitive anion channels. Am J Physiol naling networks. Genes Dev 11: 2295–2322, 1997.
Cell Physiol 270: C711–C730, 1996. 64. Voets T, Manolopoulos V, Eggermont J, Ellory C, Droog-
58. Strange K and Spring KR. Methods for imaging renal tubule mans G, and Nilius B. Regulation of a swelling-activated chloride
cells. Kidney Int 30: 192–200, 1986. current in bovine endothelium by protein tyrosine phosphorylation
59. Szücs G, Heinke S, De Greef C, Raeymaekers L, Egger- and G proteins. J Physiol (Lond) 506: 341–352, 1998.
mont J, Droogmans G, and Nilius B. The volume-activated 65. Wang L, Chen L, and Jacob TJ. The role of ClC-3 in volume-
chloride current in endothelial cells from bovine pulmonary activated chloride currents and volume regulation in bovine
artery is not modulated by phosphorylation. Pflügers Arch 431: epithelial cells demonstrated by antisense inhibition. J Physiol
540–548, 1996. (Lond) 524: 63–75, 2000.You can also read
