Cellular Signalling 15 (2003) 197 – 207

www.elsevier.com/locate/cellsig

Endothelin-1 stimulated capacitative Ca2+ entry through ETA receptors

of a rat brain-derived type-1 astrocyte cell line, IA-1g1
You Jing Ju a, Chia-Mei Wang a,1, Amos C. Hung a,1, Jun-Chih Lo a,
Hung-Jung Lin b, Synthia H. Sun a,*
a
Institute of Neuroscience, College of Life Science, National Yang Ming University, #155, Section 2, Li-Non Street, Shi-Pai, Taipei, Taiwan, ROC
b
Emergency Department, Chi-Mei Medical Center, Tainan, Taiwan, ROC
Received 7 January 2002; accepted 7 August 2002

Abstract

The present study demonstrated that endotheline-1 (ET-1) stimulated a biphasic (transient and sustained) increase in [Ca2 +]i and signaling
was blocked by BQ123 and inhibited by BQ788. RT-PCR analysis revealed that ETA was expressed more than ETB mRNA—suggesting that
ETA is the major receptor. Simply reintroducing Ca2 + in the buffer stimulated a sustained increase in [Ca2 +]i and the effect was inhibited by
U73122, thapsigargin (TG), miconazole and SKF96365. When measured in Ca2 +-free buffer, the ET-1-stimulated Ca2 + transient decreased
by 73% and the reintroduction of Ca2 + induced a large sustained increase in [Ca2 +]i. These effects were not affected by nifedipine, but were
inhibited by miconazole and SKF96365—indicating that the sustained increase in [Ca2 +]i mediated by ET-1 was mostly due to capacitative
Ca2 + entry (CCE). The ET-1-induced CCE was inhibited by phorbol ester (PMA) but was enhanced by GF109203X; it was also enhanced by
8-bromo-cyclic AMP (8-Br-cAMP) but was inhibited by H89. Thus, protein kinase C (PKC) negatively regulated and cAMP-dependent
protein kinase (PKA) positively regulated the ET-1-mediated CCE in these cells.
D 2003 Elsevier Science Inc. All rights reserved.

Keywords: Capacitative Ca2+ entry; Endothelin-1; ETA receptor; ETB receptor; Protein kinase C PKC; cAMP-dependent protein kinase PKA; Type-1 astrocyte

1. Introduction DNA synthesis, proliferation, secretion of neurotrophic

Endothelin is a potent vasoconstricting peptide, initially
purified, cloned and named by Yanagisawa et al. [52,53].
Three distinct genes encoding three endothelin isoforms,
namely ET-1, ET-2 and ET-3 [20] and two receptor sub-
types, ETA and ETB, have been identified in mammals
[1,39]. Increasing evidence indicates that endothelins play
various roles in central nervous system. Brain injury- and
inflammation-enhanced endothelin production has been
identified in astrocytes and microglia [47,51,55,56]; endo-
thelins modulate the function of astrocytes by stimulating
Abbreviations: CCE, capacitative Ca2+ entry; ET-1, endothelin-1; ETA,
endothelin A subtype receptor; ETB, endothelin B subtype receptor; [Ca2+]i,
intracellular Ca2+ concentration; PKA, cAMP-dependent protein kinase;
PKC, protein kinase C; RT-PCR, reverse transcription-polymerase chain
reaction; TG, thapsigargin; VSCC, voltage sensitive Ca2+ channel.
* Corresponding author. Tel.: +886-2-28267103; fax: +886-2-
28200259.
E-mail address: shsun@ym.edu.tw (S.H. Sun).
1
These authors have contributed equally in this study.
factors and cytoskeletal organization [6,10,17,27,44]; and
a recent study of enzyme and receptor expression implied
that endothelin is also a signal mediator between neurons
and astrocytes in the brain [31].
ETB, but not ETA mRNA, was detected in rat cerebral
astrocytes, and ETB levels were enhanced by treating cells
with dibutyryl cylic AMP [16]. Activation of ETB recep-
tors was also identified in reactive astrocytosis in striatum
[2], and in pathological response-activated astrocytes
[24,30,42], suggesting that ETB receptors are functionally
important in astrocytes. However, predominant expression
of ETA receptor was found in human astrocytoma U138MG
cells [50], and alteration of ETA receptor was found in
dibutyryl cAMP (dB-cAMP)-treated proliferating C6
glioma cells [40]. Therefore, the expression and the function
of endothelin receptor subtypes of astrocytes need to be
further examined.
The major signal transduction pathway of endothelin is
via activation of phospholipase C (PLC) and stimulation of
the inositol triphosphates (IP3)-sensitive Ca2 + release by

0898-6568/03/$ - see front matter D 2003 Elsevier Science Inc. All rights reserved.
PII: S 0 8 9 8 - 6 5 6 8 ( 0 2 ) 0 0 0 7 9 - 7
198 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207
binding to the G-protein-coupled receptors. A biphasic
increase in ETA-mediated Ca2 + signaling was observed in
vascular smooth muscle, with the initial transient peak at-
tributed to Ca2 + release from IP3-sensitive intracellular stores
and the sustained phase to Ca2 + influx [23,26]. The sustained
phase was later found to involve the voltage sensitive Ca2 +
channel (VSCC) in endothelin-stimulated vascular smooth
muscle [14,22,23] and ventricular myocytes [18].
The agonist-stimulated Ca2 + release from intracellular
stores may also induce a subsequent sustained increase in
[Ca2 +]i through a capacitative Ca2 + entry system (CCE)
[37]. CCE is now well accepted as the most important Ca2 +
influx pathway in non-excitable cells and appeared to be
physiologically important for these cells [4,33]. The path-
way has been found in glial cells [11,46,49] where it is
considered as a prolonged communication between neuron
and glia following the neurotransmitter-initiated Ca2 + tran-
sient peak. However, the endothelin-mediated CCE of
astrocytes has not been investigated.
In the present study, we used pharmacological analysis to
show that ET-1 stimulated biphasic (transient and sustained)
Ca2 + signaling through activation of the BQ123-sensitive
ETA receptors in IA-1g1 astrocytes. We also present evi-
dence that IA-1g1 expressed both ETA and ETB subtypes of
endothelin receptor subtypes, with ETA being the major
form. In these cells, ET-1 stimulated Ca2 + transient spike
and then activated the sustained increase of [Ca2 +]i by
inducing Ca2 + entry through VSCC and CCE of these cells.
Pharmacological analysis demonstrated that CCE is essen-
tial in maintaining Ca2 + homeostasis and is the major
mechanism of Ca2 + entry in IA-1g1 astrocytes. Moreover,
CCE was negatively regulated by protein kinase C (PKC)
and positively regulated by cAMP-dependent protein kinase
(PKA) in IA-1g1 astrocytes.
2. Materials and methods
2.1. Materials
8-Bromo cyclic AMP, forskolin, Fura-2/AM, H89, mico-
nazole, and nifedipine phorbol-12,13-didecanoate (PMA)
were purchased from Sigma (St. Louis, MO, USA).
BQ123, BQ788, endotheline-1, GF109203X, SKF96365,
thapsigargin (TG) and U73122 were from Calbiochem
(San Diego, CA, USA). Foetal bovine serum, F-10 nutrient
mixture and gentamicin were purchased from Gibco BRL
(Gaithersburg, MD, USA). Culture flasks and dishes were
obtained from Corning Laboratory Sciences (Corning, NY,
USA) and cuvettes (3 cm3) were from Sarstedt, Aktienge-
sellschaft (Nûmbrecht, Germany).
2.2. Cell culture and immunocytochemical analysis
IA-1g1 cell line, selected and cloned from primary
astrocyte cultures derived from neonate rat brain, was
cultured in F10 media supplemented with 10% FBS and
50 Ag of gentamicin/ml (culture medium) in a humidified
atmosphere of 95% air and 5% CO2 at 37 jC. The cells
were stock cultured in T-75 flasks as monolayer in the
culture medium.
To observe cellular morphology and analyze the expres-
sion of astrocyte marker proteins, IA-1g1 cells were sub-
cultured at a density of 2  10 4 cells/cm 2 on glass
coverslips and further cultured in culture medium for 3
days. Cells were stained with first antibodies against stanti-
GFAP or anti-A2B5. After washes with phosphate- buf-
fered saline containing 0.5% Tween 20, the coverslips were
treated with a 1:250 dilution of a secondary antibody, rabbit
anti-mouse IgG conjugated with horseradish peroxidase.
Color determination was conducted by the 3,3V-diamino-
benzidine method with nuclei stained with hematoxylin.
The negative controls for each first antibody were con-
ducted by using secondary antibody only. The photomicro-
graphs were taken with Nikon Eclipse microscope (Nikon,
Japan).
2.3. Western blot analysis of GFAP
Aliquots of protein (50 Ag) from IA-1g1 and C6
glioma cells were loaded on each lane of 12.5% sodium
Fig. 1. IA-1g1 is a type-1 astrocyte cell lines. Cells were subcultured on
glass coverslip, and cultured for 3 days, and then stained with (A) none,
first antibodies (B) anti-GFAP, and (C) anti-A2B5. After washing, the
coverslips were treated with 1:250 dilution of a secondary antibody, rabbit
anti-IgG conjugated with horseradish peroxidase. Color determination was
conducted by the 3,3V-diaminobenzidine method with nuclei counterstained
with hematoxylin. (D) Western blot analysis of the expression of the GFAP
in IA-1g1 and C6 glioma cells as indicated. Aliquots of protein (50 Ag)
from cell lysates were separated by gel electrophoresis, transblotted to
nitrocellulose membrane. Detection of GFAP (50 kDa) was performed by
reacting with anti-GFAP antibody and visualized by ECL method as
indicated.
 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207 199

(Hercules, California, USA). For detection of GFAP, the

sheet was reacted with 1:250 dilution of anti-GFAP anti-
body and a secondary antibody, rabbit anti-mouse IgG
conjugated with horseradish peroxidase. The blot was then
reacted with ECL immunodetection reagents and visual-
ized by autoradiography using Fuji medical X-ray film.

Fig. 2. ET-1 stimulated a PLC-dependent increase in [Ca2 +]i in IA-1g1

astrocytes. (A) Cells were loaded with Fura-2/AM for 30 min, washed, and
incubated with loading buffer containing Ca2 + (upper trace, n = 12) or
Ca2 +-free loading buffer (lower trace, n = 8). (B) Dose – response curve of
ET-1-stimulated net increases in [Ca2 +]i in loading buffer containing 2 mM
Ca2 + with n = 4. Fura-2 loaded IA-1g1 astrocytes were pretreated with (C)
the phospholipase C inhibitor, U73122 and (D) the endoplasmic reticulum
Fig. 3. ET-1 stimulated increases in [Ca2 +]i through predominantly ETA
Ca2 + pump inhibitor, thapsigargin (TG) for 3 min (n = 3). The addition of
receptors in IA-1g1 astrocytes. (A) IA-1g1 cells were pretreated with 5 AM
100 nM ET-1 indicated by an arrow, the fluorescence of Fura-2 and Fura-
BQ123 and (B) 5 AM BQ788. (C) Dose – response curves (1 nM – 1 AM) of
2 – Ca2 + were recorded, graphs were drawn by SigmaPlot and values are
BQ123 (.) and BQ788 (o) inhibition of ET-1-stimulated net increases in
means F S.D.
[Ca2 +]i in loading buffer containing 2 mM Ca2 + (n = 3). Data values are
means F S.D. from three determinations. (D) RT-PCR analysis of the
expression of endothelin receptor subtypes in total RNA isolated from C6
dodecyl sulfate-polyacrylamide gel for electrophoresis glioma, RBA-2 type-2 astrocytes and IA-1g1 astrocytes. Lane M: DNA size
separation. After separation, the proteins were transferred marker; +: ETA or ETB primers, and : negative control for ETA or ETB
to nitrocellulose sheets using a Semi-Dry Transfer Cell primers as indicated.
200 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207

Fig. 4. Ca2 + induced capacitative Ca2 + entry in IA-1g1 astrocytes. (A) Cells were loaded with Fura-2/AM for 30 min, washed and incubated in Ca2 +-free
loading buffer. Cells were then treated with 100 nM ET-1 and 2 mM Ca2 + was reintroduced into the media. Some cells were also treated with (B) 1 AM TG, (C)
5 AM U73122 (D) 10 AM miconazole and (E) 5 AM SKF96365. Values are means F S.D. from three determinations.
 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207
2.4. Measurement of [Ca2+]i
The increases in [Ca2 +]i were measured as described
earlier [43] according to the original method of Grynkie-
wicz et al. [15]. IA-1g1 cells cultured in the culture
medium for 48 h were washed, harvested and resus-
pended in culture medium at a density of 1  107 cell/
ml and incubated with Fura-2/AM (5 AM/ml) for 30 min
at 37 jC. The cell suspension was then rinsed twice with
serum free F10 culture medium to remove the excess
Fura-2/AM, and resuspended in culture medium at a
density of 2  106 cells/ml. It was then incubated for 30
min at 37 jC to allow the entrapped ester to hydrolyze
completely. The cell suspension (0.5 ml) was then
washed, resuspended in 2.5 ml loading buffer (150 mM
NaCl; 5 mM KCl; 1 mM MgCl2; 5 mM glucose; 10 mM
HEPES, pH 7.4; either with or without 2.2 mM CaCl2)
and then transferred to a 3-cm3 cuvette positioned in the
thermostat-regulated (37 jC) sample chamber of a dual-
excitation beam spectrofluorometer (SPEX, Model
CM1T111). The cell suspension was continually stirred
with a circular stir bar driven by a motor placed beneath
the sample chamber.
The membrane-permeant Fura-2/AM was hydrolyzed by
nonspecific cytoplasmic esterases to Fura-2. Fura-2 binds
cytosolic free Ca2 + to become AM – Ca2 +, which can then
be used as a fluorescent indicator of calcium concentration
inside the cells. The Fura-2-loaded cells were sequentially
illuminated with light at 340 and 380 nm, the excitation
wavelengths for Fura-2 –Ca2 + and Fura-2, respectively, and
Fura-2 fluorescence emission was measured at 505 nm
every second. The excitation and emission band passes
were both 1 nm, and the agonist, endothelin-1, was then
added as indicated.
To calculate [Ca2 +]i the basal, peak and sustained levels
were recorded. The net increases in [Ca2 +]i were calculated
by subtracting the basal level from the peak levels of
[Ca2 +]i. Means and S.D. from at least three experiments
on separate batches of cells were statistically analyzed by
unpaired Student’s t-test and graphs were drawn by Sigma
Plot with S.D. bars shown at every 30 s.
2.5. Reverse transcription-polymerase chain reaction
The reverse transcription-polymerase chain reaction
(RT-PCR) was performed using a GeneAmp PCR System
9600 programmable thermal controller, (Perkin Elmer,
USA). Total RNAs (5 Ag) isolated from IA-1g1 cells were
reverse-transcribed by SuperScriptk II according to the
manufacturer’s instructions (Life Technologies, USA). The
cDNA was then amplified by PCR in a reaction mixture
(100 Al) containing 50 ng of cDNA, 10 Al of 10  PCR
buffer, each deoxynucleotide 5-triphosphate at 0.25 mM,
each primer at 0.2 M, and 2 U of Taq DNA polymerase
(Life technologies, USA). The synthesized primer sets
were based on the published sequences: for ETA receptor
201
[25] sense 5V-TCGTCATGGTACCCTTCG-3Vand antisense
5V-GCTTCTGCACAGGGTTAG-3V and for ETB [7] sense
5V-TACAAGACAGCCAAAGAC-3V and anti-sense 5V-
TCCACGGTGAGGACAATG-3V. The PCR program had
the following cycling protocol: 95 jC for 45 s, 58 jC for
60 s, and 72 jC for 90 s (30 cycles); 72 jC for 5 min (1
cycle). The RT-PCR products were analyzed on a 1.2%
agarose gel stained with ethidium bromide. The negative
controls were performed as described above (i.e., in the
presence of RNA isolated from IA-1g1 and either the ETA
or the ETB primer set) but in the absence of Taq DNA
polymerase.
3. Results
3.1. IA-1g1 is a type-1 astrocyte cell line
IA-1g1 cells were first characterized by immunocyto-
chemical analysis of astrocyte markers. As shown in Fig. 1,
these cells were positive for GFAP (Fig. 1B) whereas they
were negative for A2B5 (Fig. 1C). Morphological exami-
nation indicated that these cells were processless flat cells.
A comparison of the astrocyte-specific protein expression
in C6 glioma cells (C6) and IA-1g1 cells is indicated in
Fig. 1D, which shows that antibody against GFAP revealed
a single band at 50 kDa. Taken together, these results
indicate that IA-1g1 exhibit the characteristics of type-1
astrocytes.
3.2. Endothelin-1 induced Ca2+ signaling of IA-1g1 type-1
astrocytes
As shown in Fig. 2A, ET-1 stimulated increases in
[Ca2 +]i both in the Ca2 +-added buffer system (top trace)
and in the Ca2 +-free buffer system (lower trace). In the
Ca2 +-added buffer system, ET-1 (100 nM) initially stimu-
lated a rapid (within 10 s) increase in [Ca2 +]i from a basal
Table 1
Net increases in Ca2 +-induced CCE in IA-1g1 astrocytes in the absence of
ET-1
Treatment Ca2 +-induced net [Ca2 +]i
[Ca2 +]i (nM) %
None 92.7 F 6.2 100 F 6.7
Thapsigargin 55.8 F 6.7 60.0 F 7.2
U73122 27.3 F 2.1 29.5 F 2.2
Miconazole 46.3 F 5.9 50.0 F 6.4
SKF96365 55.7 F 4.3 60.0 F 4.6
Fura-2 preloaded cells were suspended in Ca2 +-free loading buffer and
measured intracellular Ca2 + concentration.
Some cells were treated with thapsigargin (1 AM), U73122 (5 AM)
miconazole (10 AM) or SKF96365 (5 AM) for 3 min and 2 mM Ca2 + was
introduced into the media.
All values are net increases in [Ca2 +]i with mean F S.D. from three
determinations.
202 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207

level of 83.6 F 23 nM to a peak level 352 F 134 nM (a net
increase of 269 F 36 nM), after which the peak slowly
declined to a sustained level that was higher than the initial
basal level (Fig. 2A, top trace). At 150 s, the mean
amplitude of [Ca2 +]i was 163 F 38 nM and this high level
was maintained. In the Ca2 +-free buffer system, the basal
[Ca2 +]i level, mean peak amplitude and the mean amplitude
at 150 s were 55.2 F 18, 128 F 41 and 70 F 34 nM,
respectively (n = 8) (Fig. 2A, lower trace). The ET-1-stimu-
lated net increase in [Ca2 +]i is 72.8 F 27 nM in the Ca2 +-
free system. Fig. 1B shows the effect of ET-1 concentration
(1 –1000 nM) on the net increases in [Ca2 +]i in the Ca2 +-
added buffer system (n = 5). ET-1 at 1 nM elicited a minimal
net increase in [Ca2 +]i of 38 F 10 and the EC50 value was
around 14 nM (n = 5).
To verify that ET-1 stimulated the Ca2 + transient peak
through the phospholipase C (PLC)-dependent Ca2 + release
from intracellular stores, Fura-2/AM-preloaded cells were
treated with the PLC inhibitor U73123 to inhibit the
production of IP3. In addition, other fura-2 preloaded cells
were pretreated with the endoplasmic reticulum (ER) Ca2 +
pump inhibitor, thapsigargin (TG) to deplete the intracellu-
lar Ca2 + stores in ER. Pretreatment of the cells with U73123
for 3 min completely blocked the ET-1-stimulated Ca2 +
response (Fig. 2C) while the addition of TG induced a sharp
rise followed by a steady fall in [Ca2 +]i levels, and almost
completely blocked the ET-1-stimulated increases in [Ca2 +]i
(Fig. 2D). Taken together, these results reconfirm the
already well-accepted hypothesis that the action of ET-1 is
mediated through the PLC/IP3-dependent Ca2 + release.

Fig. 5. ET-1 induced capacitative Ca2 + entry in IA-1g1 astrocytes. (A) Cells were loaded with Fura-2/AM for 30 min, washed and incubated in Ca2 +-
free loading buffer. Cells were then treated with 100 nM ET-1 and 2 mM Ca2 + was reintroduced into the media. Some cells were also treated with (B)
20 AM nifedipine, (C) 10 AM miconazole and (D) 5 AM SKF96365 for 3 min as indicated by the bold arrow. Values are means F S.D. from three
determinations.
 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207
Thus, the sustained increase in [Ca2 +]i shown in the upper
trace in Fig. 2A may be due to a subsequent stimulation of
Ca2 + entry.
3.3. IA-1g1 astrocytes possess both ETA and ETB receptors,
but ET-1 stimulated Ca2+ signaling occurs through ETA
receptors
To characterize which of the endothelin receptors medi-
ated the ET-1 action, increases in [Ca2 +]i were measured in
the presence of the respective ETA and ETB receptor
antagonists, BQ123 (5 AM) and BQ788 (5 AM). Both
antagonists significantly inhibited the ET-1 stimulated
increases in [Ca2 +]i but inhibition by BQ788 was much
less effective (Fig. 3A and B). The dose – response curves
(Fig. 3C) show that BQ123 inhibited the ET-1-increased
[Ca2 +]i dose-dependently with an EI50 around 5 nM. There-
fore, IA-1g1 cells appear to possess both ETA and ETB
receptors, but ETA is evidently the major form of ET
receptor subtype in these cells.
To confirm that the IA-1g1 cells possess both ETA and
ETB receptors, reverse transcription-polymerase chain reac-
tion (RT-PCR) experiments were performed. A dense band
with the predicted size of 634 bp was observed using the
ETA receptor primer set and a much lighter band migrating
at the predicted size of 565 bp was observed with the ETB
primer set (Fig. 3D). This confirms that IA-1g1 cells
express mRNA for both ETA and ETB receptors. A
comparison analysis of the expression of ETA and ETB
mRNA in C6 glioma cells and RBA-2 type-2 astrocytes
[43] is also shown in Fig. 3D, which shows that C6
expressed ETA but not ETB mRNA, and RBA-2 expressed
none. These experiments were performed twice with iden-
tical results.
3.4. Ca2+-induced Ca2+ entry through the capacitative
Ca2+entry mechanism (CCE)
To discover whether Ca2 + enters IA-1g1 cells through a
capacitative mechanism, increases in [Ca2 +]i were initially
measured in a Ca2 +-free system and then 2 mM Ca2 + was
reintroduced into the assay buffer. As shown in Fig. 4A,
simply reintroducing 2 mM Ca2 + in the buffer induced a
sustained increase in [Ca2 +]i. Pretreatment with the ER
Ca2 + pump inhibitor TG (1 AM) of itself rapidly induced
an increase in [Ca2 +]i and significantly inhibited the Ca2 +-
induced [Ca2 +]i (Fig. 4B), while the addition of PLC
inhibitor U73122 (5 AM) slightly raised the levels of
[Ca2 +]i but greatly reduced the Ca2 +-induced [Ca2 +]i
(Fig. 4C). Pretreatment of cells with the CCE inhibitor
miconazole (10 AM, 3 min) (Fig. 4D) and SKF96365 (5
AM, 3 min) (Fig. 4E) also significantly reduced the
sustained Ca2 +-induced increase in [Ca2 +]i. In addition,
miconazole per se raised the level of [Ca2 +]i in the Ca2 +-
free media (Fig. 4D). The net increases in Ca2 +-induced
[Ca2 +]i are summarized in Table 1. Taken together, we
203
conclude from these results that the sustained Ca2 +-
induced increase in [Ca2 +]i of IA-1g1 astrocytes was
mediated via CCE.
3.5. ET-1 induced Ca2+ release activated capacitative Ca2+
entry (CCE)
The involvement of CCE in ET-1-stimulated Ca2 +
signaling was then characterized in these cells. As shown
in Fig. 5A, in an initially Ca2 +-free buffer system, ET-1
stimulated a small transient increase in [Ca2 +]i and the
subsequent reintroduction of 2 mM Ca2 + in the assay
buffer induced a second rapid and sustained rise in
[Ca2 +]i. Pretreatment with the voltage sensitive Ca2 +
channel (VSCC) inhibitor nifedipine [36] did not affect
the second Ca2 +-induced sustained increase in [Ca2 +]i
(Fig. 5B). This suggests that the VSCC may not play a
significant role in the ET-1-stimulated Ca2 + entry. The
CCE inhibitor miconazole blocked both the ET-1-induced
Ca2 + transient peak and the Ca2 +-induced increase in
[Ca2 +]i. Another CCE inhibitor, SKF96365, significantly
inhibited both the ET-1-induced Ca2 + transient and the
Ca2 +-induced sustained increase in [Ca2 +]i (Fig. 5D). The
inhibition of the Ca2 +-induced sustained increase in
[Ca2 +]i by miconalzole and SKF96365 suggests that
following the ET-1 stimulation, the second Ca2 +-induced
[Ca2 +]i was mediated through CCE in these cells. The net
increases in ET-1- and Ca2 +-induced [Ca2 +]i are summar-
ized in Table 2.
3.6. Protein kinase C (PKC) negatively regulated the ET-1-
mediated CCE of IA-1g1 astrocytes
To examine whether protein kinases regulate the ET-1-
associated CCE, cells were pretreated with the PKC acti-
vator PMA (400 nM, 3 min) in an initially Ca2 +-free
system. In the absence of PMA, the initial ET-1-induced
and the subsequent Ca2 +-induced net increases in [Ca2 +]i
were 72.9 F 12 and 242.1 F 39 nM, respectively while in
PMA-pretreated cells the corresponding increases were
27.4 F 5.2 and 174.0 F 30.6 nM (Fig. 6A). By contrast,
Table 2
Net increases in ET-1- and Ca2 +-induced [Ca2 +]i in IA-1g1 astrocytes
Treatment ET-1-induced net [Ca2 +]i Ca2 +-induced net [Ca2 +]i
2+
[Ca ]i (nM) % [Ca2 +]i (nM) %
None 90.5 F 10 100 F 12 247 F 47 100 F 19
Nifedipine 118 F 9.9 130 F 11 235 F 30 95.1 F 12
Miconazole 1.61 F 0.2 1.8 F 0.2 20.2 F 12 8.18 F 5
SKF96365 66.7 F 5.2 73.7 F 6 136 F 10 55.0 F 4
Fura-2 preloaded cells were suspended in Ca2 +-free loading buffer and
measured intracellular Ca2 + concentration.
Some cells were pretreated with nifedipine (20 AM), miconazole (10 AM)
or SKF96365 (5 AM) for 3 min. Cells were treated with ET-1 (10 nM) and 2
mM Ca2 + was introduced into the media.
All values are net increases in [Ca2 +]i with mean F S.D. from three
determinations.
204 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207

Fig. 6. Protein kinase C negatively regulated ET-1-stimulated Ca2 + signaling in IA-1g1 astrocytes. Cells were preloaded with Fura-2/AM for 30 min, washed,
reincubated in Ca2 +-free loading buffer and treated with (A) 0 and 400 nM PMA and (B) 0 and 1 AM GF109203X for 3 min. ET-1 was then added and then 2
mM Ca2 + was reintroduced into the media as indicated by the arrows. Values are means F S.D. from three determinations.

when cells were pretreated with the PKC inhibitor
GF109203X (1 AM, 3 min), the ET-1 and subsequent
Ca2 +-induced increases in [Ca2 +]i were 125.5 F 8.2 and
339.1 F 30.2 nM (Fig. 6B). Thus, we concluded that PKC
negatively regulated both the ET-1-stimulated Ca2 + release
and CCE in these cells.
3.7. Cyclic AMP-dependent protein kinase (PKA) positively
regulated the ET-1-mediated CCE of IA-1g1 astrocytes
To elucidate whether PKA also regulates the ET-1-
associated CCE, cells were pretreated with the cyclic
AMP analogue 8-bromo cAMP (8-Br-cAMP, 1 mM) and
dibutyrate cAMP (1 mM), the PKA inhibitor H89 (5 and
10 AM) and adenylyl cyclase activator forskolin (2 and 10
AM) in an initial Ca2 +-free system. As shown in Fig. 7A,
8-Br-cAMP did not affect the initial mean peak amplitude
of ET-1-stimulated Ca2 + release as compared with the
controls (123.8 F 21 and 118.8.2 F 10 nM), whereas 8-
Br-cAMP enhanced the subsequent Ca2 +-induced increases
in [Ca2 +]i. The mean peak amplitude of the ET-I-mediated
Ca2 +-induced increases in [Ca2 +]i of control and 8-Br-
cAMP-treated cells were 274 F 23 and 325 F 20 nM,
respectively. To confirm the involvement of PKA, cells
were pretreated with PKA inhibitor H89 (5 and 10 AM).
As shown in Fig. 7B, the initial peak levels of ET-1-
stimulated Ca2 + releases of control- and H89 (5 and 10
AM)-treated cells were 140 F 19, 146 F 32 and 148 F 10
 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207 205

Fig. 7. Protein kinase A positively regulated ET-1-mediated CCE in IA-1g1 astrocytes. Cells were preloaded with Fura-2/AM for 30 min, washed, reincubated
in Ca2 +-free loading buffer and treated with (A) 8-Br-cAMP (1 mM), (B) 0, 5 and 10 AM H89, (C) 5 and 10 AM forskolin, and (D) forskolin (10 AM), H89
(5 AM) and forskolin (10 AM) plus H89 (5 AM). ET-1 was then added and then 2 mM Ca2 + was reintroduced into the media as indicated by the arrows. Values are
means F S.D. from three determinations.

nM, respectively. And the subsequent mean peak ampli-
tude of Ca2 +-induced increases in [Ca2 +]i were 348.1 F
18, 214.2 F 16 and 163.2 F 15 nM, respectively. There-
fore, H89 had no effect on the initial ET-1-stimulated Ca2 +
release whereas H89 decreased the Ca2 +-induced increases
in [Ca2 +]i dose-dependently. Thus, PKA positively regu-
lated the ET-1-stimulated CCE in these cells. To elucidate
further the involvement of PKA, cells were treated with
forskolin. As shown in Fig. 7C, forskolin decreased the
ET-1-stimulated CCE in these cells dose-dependently.
Furthermore, H89 (5 AM) plus forskolin (10 AM)
inhibited CCE additively. Therefore, we concluded that
PKA positively regulated CCE and the forskolin-decreased
CCE is probably mediated through a mechanism other than
PKA.
4. Discussion
The present study characterized the ET-1-stimulated
Ca2 + signaling in a permanent clonal rat brain-derived
type-1 astrocyte cell line, IA-1g1. The results indicated that
ET-1 stimulated a biphasic, i.e. transient and sustained,
increase in [Ca2 +]i in these cells. The ET-1-stimulated
transient increase in [Ca2 +]i was primarily associated with
activation of the PLC-dependent Ca2 + release through ETA
receptors. The subsequent sustained increase in [Ca2 +]i was
due to Ca2 + entry. Although Ca2 + could also enter through
the voltage-sensitive Ca2 + channels, CCE was essential in
replenishing the Ca2 + stores and appeared to be the major
mechanism of Ca2 + entry in these cells. CCE was nega-
tively regulated by PKC and positively regulated by PKA.
Blomstrand et al. [5] showed that in astrocytes, the ET-1-
induced Ca2 + signaling was fully prevented only by treat-
ment of cells with both ETA plus ETB antagonists, indicat-
ing that astrocytes possess both receptor subtypes. In the
present study, RT-PCR revealed that IA-1g1 astrocytes
possess both ETA and ETB mRNA (Fig. 3D). However,
treatment with BQ123 alone was sufficient to block the
response (Fig. 3A and C), indicating that ETA is the major
functional receptor subtype in these cells. This result is
similar to an early finding in human astrocytoma U138MG
cells that ET-1 binds to ETA receptors and leads to long
lasting responses [50]. Nevertheless, in human U373MG
astrocytoma cells, on the other hand, ET-1 was shown to
bind to ETB receptors and then induced rapid internaliza-
tion, suggesting that ETB may play an important role in
ligand clearance through receptor internalization [51]. In the
present study, a pilot study found that ET-3 (100 nM)
stimulated increases in [Ca2 +]i in these cells; the net
206 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207
increase in [Ca2 +]i (71.9 F 20 nM, n = 3) was far less than
for ET-1 (339 F 41 nM, n = 5). The result is also similar to
findings in C6 glioma cells that BQ123 inhibited both ET-1-
and ET-3-stimulated cellular proliferation [41] and Ca2 +
signaling [40] concluded that both responses were mediated
through ETA receptors. Nevertheless, as shown in Fig. 3D,
C6 glioma cells appeared to express mRNA for ETA only.
Taken together while ETA is the major form of ET receptor
subtype, ETB receptor in IA-1g1 cells may still be functional
in IA-1g1 astrocytes.
This is the first report to show that ET-1 induced CCE in
astrocytes. This finding agrees with the common observa-
tion that stores depletion activated CCE in non-excitable
cells [19,28,32,33,35,56]. In the present study, when media
Ca2 + were removed, the net increase in ET-1-stimulated
transient increase in [Ca2 +]i was 73% lower than when the
medium contained 2 mM Ca2 + (Fig. 2A). The result
suggests that the ET-1-stimulated IP3-mediated Ca2 + release
is tightly coupled with CCE and that CCE is engaged in
both phases of Ca2 + rise. In addition, in steady state,
U73122 inhibited 70% of the Ca2 +-induced CCE (Fig.
4C), reconfirming that the IP3-mediated Ca2 + release was
also tightly linked to CCE even in the absence of an agonist.
Therefore, CCE plays a pivotal role in maintaining Ca2 +
homeostasis in IA-1g1 astrocytes.
It is interesting to note that miconazole and TG both
raised [Ca2 +]i (Figs. 4B and D, and 5C). A recent study
found that the miconazole-sensitive intracellular Ca2 + store
overlapped with the TG sensitive store [21]. Although both
miconazole and TG inhibited the Ca2 +-induced increases in
[Ca2 +]i, nevertheless, in cells pretreated with TG, the Ca2 +-
induced increase in [Ca2 +]i declined rapidly whereas, in the
miconazole-treated cells, it was sustained. The reason for
this difference is not yet known. However, TG specifically
inhibited ER Ca2 + pump [45] and is considered to be the
most reliable experimental procedure to deplete ER Ca2 +
stores [9]. So in the TG-treated cells, decline in [Ca2 +]i may
be due to Ca2 + sequestration into other Ca2 + store organ-
elles such as mitochondria [11]. The decline may also be
due to an effect of TG on protein – protein interaction
between ER Ca2 + pump and CCE channel [4]. In contrast,
miconazole may exert complex multiple effects on Ca2 +
signaling and inducing CCE on its own [21]. Our present
results are consistent with previous findings that the agonist-
mediated CCE is essential in refilling Ca2 + stores and
maintaining Ca2 + homeostasis of cells.
Although the involvement of CCE in gating Ca2 + entry
and prolonged Ca2 + signaling has recently attracted much
attention, the putative ion channels have not yet been
purified and cloned, and the mechanisms that regulate and
control CCE remain elusive [9]. Recently, a SNAP-25-
dependent regulatory mechanism [54] and a secretion-like
coupling model [34] have also been identified. In addition,
PKC inhibition of CCE has been reported in human neu-
trophils [29] and in Xenopus oocytes [35]. In the present
study, we found that activation of PKC inhibited both ET-1-
mediated Ca2 + release and CCE in IA-1g1 astrocytes (Fig.
6A). The inhibition of the Ca2 + release may be due to
desensitization of PLC-h by PKC [13,38].
The present study also demonstrated that PKA positively
regulated the ET-1-mediated CCE in IA-1g1 astrocytes. This
result correlated well with early findings in C6 glioma cells
[3] and in cerebellar astrocytes [49] which indicated that
Ca2 +-sensitive adenylyl cyclase (AC) might colocalize with
the CCE channels [8,12]. Lastly, although PKA positively
regulated CCE, forskolin caused a significant inhibition in
the ET-1-mediated CCE, and an additive effect of H89 and
forskolin was observed (Fig. 7D). Therefore, forskolin may
mediate the effect through a mechanism other than PKA,
such as glucose transport [48], that may be important in
regulating signal transduction pathways of astrocytes.
In summary, the present study demonstrates that ETA
receptors are the major ET receptor subtype of a rat brain-
derived clonal type-1 astrocyte cell line. ET-1 stimulated a
biphasic increase in [Ca2 +]i, which was the result of a
transient Ca2 + release from intracellular stores followed
by sustained Ca2 + entry via CCE. In addition, PKA pos-
itively regulated, whereas PKC negatively regulated, the
ET-1-stimulated CCE in these cells.
Acknowledgements
The study was supported by grant no. 90-B-FA22-1-4-02
from the Ministry of Education, Taipei and CMYM9005
from Chi-Mei Medical Center, Tainan, Taiwan, ROC.
References
[1] Arai H, Hori S, Aramori I, Ohkubo H, Nakanishi S. Nature 1990;348:
730 – 2.
[2] Baba A. Life Sci 1998;62:1711 – 5.
[3] Barańsk J, Chaban V, Czarny M, Sabala P. Cell Calcium 1995;17:
207 – 15.
[4] Berridge MJ. Biochem J 1995;312:1 – 11.
[5] Blomstrand F, Giaume C, Hansson E, Ronnback L. Am J Physiol
1999;277:C616 – 27.
[6] Cazaubon S, Chaverot N, Romero IA, Girault IA, Adamson P, Stros-
berg AD, et al. J Neurosci 1997;17:6203 – 12.
[7] Cheng HF, Su YM, Yeh JR, Chang KJ. Mol Pharmacol 1993;44:
533 – 8.
[8] Chiono M, Mahey R, Tate G, Cooper DMF. J Biol Chem 1995;
270:1149 – 55.
[9] Clapham DE. Cell 1995;80:259 – 68.
[10] Couraud P, Durieu-Trautmann O, Le Nguyenm P, Marin P, Gilbert F,
Strosberg AD. Eur J Pharmacol 1991;206:191 – 8.
[11] Deitmer JW, Verkhratsky AJ, Lohr C. Cell Calcium 1998;24:405 – 16.
[12] Fagan KA, Mahey R, Cooper DMF. J Biol Chem 1996;271:
12438 – 44.
[13] Fisher SK. Eur J Pharmacol 1995;288:231 – 50.
[14] Goto K, Kasuya Y, Mastsuki N, Takuwa Y, Kurihara H, Ishikawa T,
et al. Proc Natl Acad Sci U S A 1989;86:3918 – 35.
[15] Grynkiewicz G, Poenie M, Tsien RY. J Biol Chem 1985;260:
3440 – 50.
[16] Hama H, Sakurai T, Kasuya Y, Fujiki M, Masaki T, Goto K. Biochim
Biophys Res Commun 1992;186:355 – 62.
 Y.J. Ju et al. / Cellular Signalling 15 (2003) 197–207
[17] Hama H, Yamada G, Suzuki N, Masaki T, Goto K. J Neurosci Res
1997;47:590 – 602.
[18] He J-Q, Pi YQ, Walker JW, Kamp TJ. J Physiol 2000;524:807 – 20.
[19] Hoth M, Penner R. J Physiol 1993;465:359 – 86.
[20] Inoue A, Yanagisawa M, Kimura S, Kasuya S, Miyauchi T, Goto K,
et al. Proc Natl Acad Sci U S A 1989;86:2863 – 7.
[21] Jan C-R, Tseng C-J. Life Sci 1999;65:2513 – 22.
[22] Kasuya Y, Ishikawa T, Yanagisawa M, Kimura S, Goto K, Masaki T.
Am J Physiol 1989;257:H1818 – 35.
[23] Kasuya Y, Takuwa Y, Yanagisawa M, Kimura S, Gota K, Masaki T.
Biochem Biophys Res Commun 1989;161:1049 – 55.
[24] Koyama Y, Takwmura M, Fujiki K, Ishikawa N, Shigenaga Y,
Baba A. Glia 1999;26:268 – 71.
[25] Lin HY, Kaji EH, Winkei GK, Ives HE, Lodish HF. Proc Natl Acad
Sci U S A 1991;88:3185 – 9.
[26] Marsden PA, Danthuluri NR, Brenner BM, Ballermann BJ, Brock TA.
Biochem Biophys Res Commun 1989;158:86 – 93.
[27] MacCumber M, Ross CA, Snyder H. Proc Natl Acad Sci U S A
1990;87:1359 – 2363.
[28] McDonald TV, Premack BA, Gardner P. J Biol Chem 1993;268:
3889 – 96.
[29] Montero M, Garcia-Sancho J, Alvarez J. J Biol Chem 1993;268:
13055 – 61.
[30] Morga E, Faber C, Heuschling P. J Neurochem 2000;74:603 – 12.
[31] Nakagomi S, Kiryu-Seo S, Kiyama H. Neuroscience 2000;101:
441 – 9.
[32] Parekh AB, Teriau H, Stûhmer W. Nature 1993;364:814 – 8.
[33] Parekh AB, Penner R. Physiol Rev 1997;77:901 – 30.
[34] Patterson RL, Van Rossum DB, Gill DL. Cell 1999;98:487 – 99.
[35] Petersen CCH, Berridge MJ. J Biol Chem 1994;269:32246 – 53.
[36] Platano D, Pollo A, Carbone E, Aicardi G. FEBS Lett 1996;39:
189 – 94.
[37] Putney Jr JW. Cell Calcium 1986;7:1 – 12.
[38] Rebecchi MJ, Pentyala SN. Physiol Rev 2000;80:1291 – 335.
207
[39] Sakurai T, Yanagisawa M, Takuwa Y, Miyazaki H, Kimura S, Goto K,
et al. Nature 1990;348:732 – 5.
[40] Šedo A, Malik R, Vlasicova K, Rovero P. Neuropeptides 1999;
33:13 – 7.
[41] Šedo A, Provero P, Revoltslla RP, Di Bartolo V, Beffy P, Mizrahi J.
J Biol Regul Homeost Agent 1993;7:95 – 8.
[42] Shibaguchi H, Himeno A, Shigematsu K, Kataoka Y, Niwa M. Glia
2000;31:91 – 4.
[43] Sun SH, Lin LB, Hung AC, Kuo J-S. J Neurochem 1999;73:334 – 43.
[44] Teixeira A, Chaverot N, Chroder C, Strosberg AD, Couraud P-O,
Cazaubon S. J Neurochem 1999;72:120 – 8.
[45] Thastrup O, Cullen PJ, Drobal B, Hanley MR, Dawson AP. Proc Natl
Acad Sci U S A 1990;87:2466 – 70.
[46] Toescu EC, Moller T, Kettenmann H, Verkhratsky A. Neuroscience
1998;86:925 – 35.
[47] Uesugi M, Kasuya Y, Hama H, Yamamoto M, Hayashi K, Masaki T,
et al. Brain Res 1996;728:255 – 9.
[48] Williams M, Sharp P. Regulation of jejunal glucose transporter ex-
pression by forskolin. Biochim Biophys Acta 2002;1559:179 – 85.
[49] Wu M-L, Chen W-H, Liu I-H, Tseng C-D, Wang S-M. J Neurochem
1999;73:1318 – 28.
[50] Wu-Wong J, Chiou W, Magnuson SR, Bianchi BR, Lin CW. Biochim
Biophys Acta 1996;1311:155 – 63.
[51] Wu-Wong Ng, Chiou W, Magnuson SR, Opgenorth TJ. J Pharmacol
Exp Ther 1995;274:499 – 507.
[52] Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M,
Mitsui Y, et al. Nature (Lond.) 1988;332:411 – 5.
[53] Yanagisawa M, Inoue A, Ishikawa T, Kasuya Y, Kimura S, Kumagaye
S-I, et al. Proc Natl Acad Sci U S A 1988;85:6964 – 7.
[54] Yao Y, Ferrer-Montiel AV, Montel M, Tsien R. Cell 1999;98:475 – 85.
[55] Zhang M, Olsson Y. J Neurol Sci 1995;134:26 – 32.
[56] Zhang WW, Basonic T, Hoog A, Jiang MH, Ma KC, Nie XJ. J Neurol
Sci 1994;122:90 – 6.