www.elsevier.com/locate/cellsig

Cellular Signalling 17

Disruption of the filamentous actin cytoskeleton is necessary for the

activation of capacitative calcium entry in naive smooth muscle cells

Sara Moralesa, Pedro J. Camelloa, Juan A. Rosadoa, Gary M. Maweb, Marıa J. Pozoa,*

aDepartment of Physiology, University of Extremadura, 10071 Caceres, SpainbDepartment of Anatomy and Neurobiology, University of Vermont, Burlington, Vermont, VT05405, USA

Received 6 September 2004; accepted 11 October 2004

Available online 13 November 2004

Abstract

It has been proposed that cytoskeleton plays a key positive role in the activation of capacitative calcium entry (CCE), which supported the

secretion-like hypothesis for the mechanisms underlying this process. However, its role on CCE in native smooth muscle is unknown. Here

we demonstrate that CCE in isolated gallbladder myocytes was enhanced by cytochalasin D or latrunculin A treatments (agents that cause

actin disassembly) whereas it was reduced by jasplakinolide treatment (which causes actin polymerization), suggesting that actin

cytoskeleton acts as a barrier in CCE. In addition, we show for the first time that depletion of intracellular Ca2+ stores by thapsigargin and

cholecystokinin in BAPTA-loaded cells induced a decrease in F-actin content that was consistent with a link between CCE and actin

reorganization. In conclusion, these data suggest an active participation of actin reorganization in the implementation of CCE and support a

conformational coupling model for this process in naive smooth muscle cells.

D 2004 Elsevier Inc. All rights reserved.

Keywords: F actin content; Store depletion; Thapsigargin

1. Introduction

In smooth muscle, changes in cytosolic Ca2+ concen-

tration ([Ca2+]i) control a vast array of cellular functions

ranging from contraction or relaxation to growth and

apoptosis. This ubiquitous signal is tightly controlled by a

variety of cellular transport systems that act to increase or

remove Ca2+ from the cytosol [1]. One of the least

understood systems is activation of Ca2+ entry through

plasma membrane following depletion of intracellular Ca2+

stores, a process termed capacitative Ca2+ entry (CCE) [2].

The base of this mechanism, initially observed in non

excitable cells, is that Ca2+ concentration within the intra-

cellular Ca2+ pools (mainly endoplasmic/sarcoplasmic

reticulum, E/SR) determines the permeability of plasma

0898-6568/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.cellsig.2004.10.002

* Corresponding author. Department of Physiology, Nursing School,

Avda Universidad s/n, 10071 Caceres, Spain. Tel.: +34 927 257450; fax:

+34 927 257451.

E-mail address: mjpozo@unex.es (M.J. Pozo).

membrane to external Ca2+, so that Ca2+ release from the

stores increases Ca2+ influx, resulting in a sustained Ca2+

plateau during stimulation. Contrary to non-excitable cells,

reports of CCE in excitable tissues are relatively scarce. In

gallbladder smooth muscle (GBSM), we have recently

demonstrated that release of Ca2+ from internal stores

activates a CCE pathway, in addition to activating Ca2+

influx through L-type Ca2+ channels [3].

Despite the intense research in the field, the mechanism

that links fall of Ca2+ concentration in the stores to opening

of plasma membrane Ca2+ channels remains highly con-

troversial. One set of hypotheses postulates the release of a

diffusible messenger by the pools, while others claim a

physical interaction between the empty stores and plasma

membrane involving membrane proteins, secretory vesicles

and possibly cytoskeletal elements (reviewed in Ref. [4]).

The cytoskeleton is an integrated, dynamically arranged

network of actin fibers, microtubules and intermediate

filaments that is involved in various cellular processes

including cell motility, intracellular transport and smooth

(2005) 635–645

S. Morales et al. / Cellular Signalling 17 (2005) 635–645636

muscle contraction [5,6]. F-actin cytoskeleton rearrange-

ment provides cells a valuable tool to control a variety of

functions such as cell strength and contractility [7,8]. The

actin cytoskeleton forms a complex network, providing the

structural basis for simultaneous interactions between multi-

ple cellular structures. A number of studies in different cell

types have reported a positive role for the actin cytoskeleton

in the activation of CCE. For example, experimental

manipulation of the cortical actin cytoskeleton has been

used as evidence for a secretion-like coupling model in

cultured smooth muscle cells [9] and other cell types

[10,11]. However, in other cells, such as RBL-1 cells

[12,13] or vascular smooth muscle cells [14], CCE has been

shown to be unaffected by such cytoskeletal disruption.

In the current study, we sought to test the hypothesis that

the actin cytoskeleton plays a role in the CCE process of

GBSM cells. Our results demonstrate that disruption of actin

cytoskeleton is associated with an increase in CCE, which

supports a negative role for actin cytoskeleton acting as a

barrier in the activation of capacitative mechanisms. In

addition, we show for the first time that CCE activation by

experimental and physiological stimuli is linked to actin

depolymerization. All together, our data indicate that a

direct coupling model is the one that best fit with our

findings.

2. Methods

2.1. Tissue preparation

Gallbladders, isolated from 300- to 500-g male guinea

pigs, after deep halothane anesthesia and cervical disloca-

tion, were immediately placed in cold Krebs-Henseleit

solution (K-HS; for composition see Solutions and chem-

icals) at pH 7.35. The gallbladder was opened from the end

of the cystic duct to the base, and trimmed of any adherent

liver tissue. After the preparation was washed with the

nutrient solution to remove any biliary component, the

mucosa was carefully dissected away. All the experiments

were carried out according to the guidelines of Animal Care

and Use Committees of the University of Extremadura.

2.2. Cell isolation

GBSM cells were dissociated enzymatically using a

previously described method [15]. Briefly, after preparing

the tissue as indicated above, the gallbladder was cut into

small pieces and incubated for 35 min at 37 8C in

enzyme solution (ES, for composition see Solutions and

chemicals) supplemented with 1 mg/ml BSA, 1 mg/ml

papain and 1 mg/ml dithioerythritol (DTT). Then the tissue

was transferred to fresh ES containing 1 mg/ml BSA, 1 mg/

ml collagenase and 100 AM CaCl2, and incubated for 9 min

at 37 8C. The tissue was then washed three times using Na+-

HEPES solution (for composition see Solutions and

chemicals), and the single smooth muscle cells were isolated

by several passages of the tissue pieces through the tip of a

fire-polished glass Pasteur pipette. The resultant cell

suspension was kept in Na+- HEPES solution at 4 8C until

use, generally within 6 h. All experiments involving isolated

cells were performed at room temperature (22 8C).

2.3. F-actin content measurement

The F-actin content of resting and stimulated GBSM

cells was determined according to a previously published

procedure [16]. Briefly, samples of GBSM cell suspen-

sions (200 Al) were challeged in Na+-HEPES solution

and quickly transferred to 200 Al ice-cold 3% (w/v)

formaldehyde in phosphate-buffered saline solution (PBS,

for composition see Solution and chemicals) for 10 min.

Fixed cells were permeabilised by incubation for 10 min

with 0.025 % (v/v) Nonidet P-40 detergent dissolved in

PBS. Cells were then incubated for 30 min with fluorescein

isothiocyanate-labeled phalloidin (FITC-phalloidin; 1 AM)

in PBS solution supplemented with 0.5 % (w/v) bovine

serum albumin (BSA). After incubation, the cells were

collected by centrifugation for 2 min at 10000�g and

resuspended in PBS solution. Staining of actin filaments

was measured using a Shimadzu fluorescence spectro-

fluorimeter. Samples were excited at 496 nm and emission

was recorded at 516 nm.

For actin filament visualization, aliquots of FITC-

phalloidin-stained cells were transferred to an experimen-

tal chamber made with a glass poly-d-lysine-treated

coverslip (0.17 mm thick), and mounted on the stage

of an inverted microscope (Eclipse TE300, Nikon). F-

actin was visualised using a confocal laser-scanning

system (model MRC-1024, Bio-Rad) with excitation

wavelength of 488 nm and emission at 515 nm. The

cell F-actin content was quantified as arbitrary units of

fluorescence using the ImageJ software.

2.4. Cell loading and [Ca2+]i determination

[Ca2+]i was determined by epifluorescence microscopy

using the fluorescent ratiometric Ca2+ indicator, fura-2.

Isolated cells were loaded with 4 AM fura 2-AM at room

temperature for 25 min. An aliquot of cell suspension was

placed in an experimental chamber made with a glass poly-

d-lysine-treated coverslip (0.17 mm thick) filled with Na+-

HEPES solution, and mounted on the stage of an inverted

microscope (Diaphot T200, Nikon). After the cell sedimen-

tation, a gravity-fed system was used to perfuse the chamber

with Na+-HEPES solution in absence or presence of

experimental agents. For deesterification of the dye, z20

min were allowed to elapse before Ca2+ measurements were

started.

Cells were illuminated at 340 and 380 nm by a computer-

controlled filter wheel (Lambda-2, Sutter Instruments) at

0.3–1 cycles/s and the emitted fluorescence was selected by

S. Morales et al. / Cellular Signalling 17 (2005) 635–645 637

a 500-nm band-pass filter. The emitted fluorescence images

were captured with a cooled digital charge-coupled device

camera (model C-4880-91, Hamamatsu Photonics) and

recorded using dedicated software (Argus-HisCa, Hama-

matsu Photonics). The ratio of fluorescence at 340 nm to

fluorescence at 380 nm (F340/F380) was calculated pixel by

pixel and used to indicate the changes in [Ca2+]i. A

calibration of the ratio for [Ca2+] was not performed in

view of the many uncertainties related to the binding

properties of fura-2 with Ca2+ inside of smooth muscle cells.

For loading with dimethyl BAPTA, cells were incubated

for 30 min at 37 8C with 10 AM dimethyl BAPTA/AM.

2.5. Intracellular recording from smooth muscle

The methods to be used for intracellular electrophysio-

logical recording were similar to those previously described

[17]. The gallbladder whole mount preparation was pinned,

serosal side up, in a 3-ml tissue chamber and placed on the

stage of an inverted microscope (Diaphot T300, Nikon).

Smooth muscle bundles were visualized at �200 with

Hoffman Modulation Contrast optics (Modulation Optics,

Greenvale, NY, USA). The preparations were continuously

perfused at a rate of 10–12 ml/min with modified Krebs

solution (for composition see Solutions and drugs) aerated

with 95% O2–5% CO2. Temperature was maintained

between 36 and 37 8C at the recording site.

Glass microelectrodes were filled with 2.0 M KCl and

had resistances in the range of 50–110 MV. A negative-

capacity compensation amplifier (Axoclamp 2A; Axon

Instruments, Foster City, CA, USA) with bridge circuit

was used to record membrane potentials, and outputs were

displayed on an oscilloscope (Hitachi VC-6050). Electrical

signals were recorded using the computer program, MacLab

(CB Sciences, Milford MA, USA). Experimental com-

pounds were applied by addition to the superfusing solution.

2.6. Solutions and chemicals

The K-HS solution contained (in mM): NaCl 113, KCl

4.7, CaCl2 2.5, KH2PO4 1.2, MgSO4 1.2, NaHCO3 25 and

d-glucose 11.5. This solution had a final pH of 7.35 after

equilibration with 95% CO2–5% O2. The composition (mM)

of the modified Krebs solution used in the intracellular

recordings was: NaCl 121, KCl 5.9, CaCl2 2.5, NaH2PO4

1.2, MgCl 1.2, NaHCO3 25 and d-glucose 8. The ES

solution used to disperse cells was made up of (in mM): N-

2-hydroxyethylpiperazine-NV-2-sulphonic acid (HEPES) 10,

NaCl 55, KCl 5.6, Na-Glutamate 80, MgCl2 2, and d-

glucose 10, with pH adjusted to 7.3 with NaOH. The Na+-

HEPES solution contained (in mM): HEPES 10, NaCl 140,

KCl 4.7, CaCl2 2, MgCl2 2 and d-glucose 10, with pH

adjusted to 7.3 with NaOH. The PBS solution used in F-

actin studies contained (in mM): NaCl 137, KCl 2.7,

Na2HPO4 5.62, NaH2PO4 1.09 and KH2PO4 1.47 with pH

adjusted to 7.2.

Drug concentrations are expressed as final bath concen-

trations of active species. Drugs and chemicals were

obtained from the following sources: cholecystokinin (26–

33) (CCK-8) sulfated, 1,4-dithio-dl-threitol (DTT), nonidet

P40, paraformaldehyde, fluorescein isothiocyanate-labelled

phalloidin (FITC-phalloidin) and thapsigargin (TPS) were

from Sigma (St. Louis, MO, USA), cytochalasin D (CytD)

and latrunculina A (LatA) were from Calbiochem (La Jolla,

CA, USA), 2-aminoethoxydiphenylborane (2-APB) was

from Tocris (Bristol, UK), dimethyl bis-(o-aminophe-

noxy)-ethane-N,N,NV,NV-tetra-acetic acid acetoxymethyl

ester (dimethyl BAPTA AM), fura-2 AM and jasplakinolide

(JP) were from Molecular Probes (Molecular Probes

Europe, Leiden, Netherlands), collagenase was from Fluka

(Madrid, Spain), and papain was from Worthington Bio-

chemical (Lakewood, NJ, USA). Other chemicals used were

of analytical grade from Panreac (Barcelona, Spain).

Stock solutions of 2-APB, dimethyl BAPTA-AM, fura 2-

AM, JP, LatA and TPS were prepared in dimethylsulph-

oxide (DMSO) and stocks solutions of CytD and FITC-

phalloidin were prepared in ethanol. The solutions were

diluted such that the final concentration of DMSO or

ethanol was V0.1% v/v. These concentrations of solvents did

not themselves affect the mechanical activity of the tissue

nor interfere with fura-2 fluorescence.

2.7. Quantification and statistics

Results are expressed as meanFthe standard error of the

mean of n cells. Statistical differences between means were

determined by Student’s t-test. Differences were considered

significant when Pb0.05.

3. Results

3.1. F-actin content in GBSM was decreased by cytocha-

lasin D and latrunculin A and increased by jasplakinolide

In the current investigation, cytochalasin D (CytD),

latrunculin A (LatA) and jasplakinolide (JP) were used to

modify the F-actin content in GBSM. To inhibit actin

filament assembly we used CytD and LatA [18] whereas JP

was the tool of choice to increase actin polymerization [19].

In the current study, F-actin was quantified using the

fluorescent derivative of phalloidin, FICT-phalloidin, and

two different fluorescence methods. Fig. 1 shows the

distribution of actin in control GBSM cells (panel A) and

after exposure to 10 AM CytD, 3 AM LatA and 10 AM JP

(panels B, C and D). Isolated GBSM cells do not have

prominent actin stress fibres that are typical of cultured

smooth muscle cells, rather their cytoskeleton resem-

bles that of other fresh isolated smooth muscle cells [20].

Both CytD and LatA induced a decrease in phalloidin

fluorescence (control: 34.31F2.95 arbitrary units, a.u., n=8;

CytD: 19.03F1.43 a.u., n=9, Pb0.001 vs. control; LatA:

Fig. 1. Effects of cytochalasin, latrunculin A and jasplakinolide on GBSM cell F-actin content. Confocal fluorescence images of FITC-labelled phaloidin

showing the distribution of F-actin content in GBSM cells in control conditions (A), and after treatment with 10 AM CytD (B), 3 AM LatA (C) and 10 AM JP

(D). Cells were treated with the cytoskeleton modifying drugs for 45 min, except for LatA that was always incubated for 60 min. Images are representative of

six to nine experiments. Panel E shows summary data of fluorescence intensities from confocal images in the above-described conditions. Histograms are

meanFSEM of six to nine experiments. Similar pattern in F-actin content is shown in panel F where fluorescence was measured in a cell suspension using a

fluorescence spectrophotometer. F-actin content is expressed as percentage of control. To reduce variability due to cell isolation and to changes in the cell

density along the experiment each time two volumes of cell suspension were taken: one for control and the other for the corresponding experimental group.

Histograms are meanFSEM of four to nine experiments. *Pb0.05, **Pb0.01, ***Pb0.001 respect to control, respectively.

S. Morales et al. / Cellular Signalling 17 (2005) 635–645638

20.75F2.43 a.u., n=6, Pb0.01 vs. control; Fig. 1E), whereas

the treatment with JP caused a significant increase in

detectable fluorescence, indicating an increase in actin

polymerization (JP: 44.65F3.35 a.u., n=6, Pb0.05 vs.

control). Similar results were obtained when a suspension

of FITC-phalloidin stained GBSM cells was evaluated by

spectrofluorometry. To minimize variability in fluorescence

due to changes in the density of the cell suspension

throughout time or in different preparations, we routinely

acquired two samples at the same time: one for control and

the other for the experimental conditions. As indicated

above, CytD decreased F-actin content (in arbitrary units) by

62.3% (7.20F0.5 vs. 2.62F0.39 a.u., n=5, Pb0.001, Fig.

1F), LatA also caused a significant reduction (11.25F1.91

vs. 6.45F1.17 a.u., n=9, 41.6% of reduction, Pb0.05, Fig.

1F) whereas JP increased the F-actin content (9.28F0.99 vs.

13.22F0.8 a.u., 46.2% of increase, n=4 , Pb0.05, Fig. 1F).

Based on the consistency of the results obtained by both

methods, and to avoid cellular variability, we selected the

spectrofluorimeter technique to quantify actin content

throughout the remainder of the study.

3.2. CCE was altered by disassembly and polymerization/

reorganization of the actin cytoskeleton

To test whether the actin cytoskeleton plays a role in CCE

mechanisms in GBSM cells we activated CCE by using a

protocol previously validated in this cellular model [3]. Ca2+

stores were depleted by the SERCA pump inhibitor,

thapsigargin (TPS, 1 AM) in a Ca2+-free medium for 30

S. Morales et al. / Cellular Signalling 17 (2005) 635–645 639

min. When external Ca2+ was reintroduced, a sustained

[Ca2+]i increase (0.074F0.003 DF340/F380, n=39) was

observed (Fig. 2A), indicative of enhanced permeability to

extracellular Ca2+. To assess the effects of actin cytoskeleton

in the activation of Ca2+ entry, fura 2-loaded cells were pre-

treated with 10 AMCytD for 45 min before emptying of Ca2+

stores. Interestingly, under these conditions Ca2+ reintroduc-

tion induced a [Ca2+]i increase that reached a plateau of

0.138F0.014 DF340/F380 (n=11, Pb0.01 vs. control, Fig.

2B). A similar pattern was also observed when actin

disassembly was evoked with 3 AM LatA (0.137F0.014

Fig. 2. CCE is altered by disassembling and polymerization/reorganization of

fluorescence ratio in response to Ca2+ store depletion and Ca2+ restoration in contr

latrunculin A (C) and 10 AM jasplakinolide (D). Fura-2 loaded GBSM cells were

stores. When indicated, cells were perfused with a 2 mM Ca2+ HEPES solution r

incubated with the cytoskeleton modifying drugs at least 45 min prior to TPS trea

above described conditions. Histograms are meanFSEM of 12–15 experiments. *

DF340/F380, n=15, Pb0.001 vs. control, Fig. 2C). However,

when 10 AM JP was used, restoration of external Ca2+

induced a capacitative behaviour, although the plateau was

smaller than in control cells (0.050F0.004DF340/F380, n=12,

Pb0.001 vs. control, Fig. 2D). These data suggest that the

membrane-associated cytoskeleton acts a physical barrier

that inhibits Ca2+ entry by capacitative mechanisms.

We have previously demonstrated that in addition to

bclassicalQ capacitative channels L-type channels are acti-

vated in GBSM cells in response to depletion of Ca2+ stores

[3]. To determine whether actin cytoskeleton integrity is

the actin cytoskeleton. Representative original traces of changes in the

ol GBSM cells (A) and in the presence of 10 AM cytochalasin D (B), 3 AMtreated with 1 AM thapsigargin (TPS) in Ca2+ free solution to deplete the

esulting in a sustained [Ca2+]i increase. In panels B, C, D, cells were pre-

tment. Panel E shows summary data of DF340/F380 from experiments in the

*Pb0.01, ***Pb0.001 respect to control.

S. Morales et al. / Cellular Signalling 17 (2005) 635–645640

necessary in the overall influx or specifically to a particular

channel, CCE was activated in the presence of the L-type

Ca2+ channel inhibitor, nitrendipine (1 AM). As observed in

Fig. 3A and B, CytD-induced actin disassembly caused a

significant increase in CCE (0.055F0.004 vs. 0.108F0.006

DF340/F380, 198 % of increase, n=15 and 13, respectively;

Pb0.001).

When external Ca2+ restorationwas performed in presence

of 100 AM 2-APB, to block capacitative channels and allow

Ca2+ influx through L-type Ca2+ channels, disruption of the

actin cytoskeleton did not modify Ca2+ influx (0.044F0.003

vs. 0.052F0.003 DF340/F380 in the absence and presence of

CytD, respectively, n=14; PN0.05; Fig. 3C and D). These

results suggest a specific and negative role for the actin

cytoskeleton in the activation of mechanisms gating CCE

channels in GBSM cells. According to this hypothesis, in the

presence of JP the activation of capacitative channels should

be impaired. Consistent with this model, in the presence of JP

Ca2+ restoration-induced Ca2+ influx was almost completely

Fig. 3. Actin cytoskeleton disassembly induces activation of Ca2+ influx through

presence of the L-type Ca2+ channel blocker, nitrendipine (1 AM) and the capac

presence (B, D) of 10 AM CytD, respectively. CCE was activated as described in

treatment. Panel E shows summary data of DF340/F380 from experiments in the abo

***Pb0.001 respect to control.

blocked when the membrane potential was hyperpolarized by

the KATP channel activator pinacidil (5 AM) (81.3F7.0% of

inhibition, n=15, Pb0.001), indicating that in these con-

ditions there was not Ca2+ influx through voltage-independ-

ent Ca2+ channels.

To further investigate if actin cytoskeleton disruption was

affecting Ca2+ influx through L-type Ca2+ channels, we

studied the effects of CytD-induced actin filament dis-

assembly on KCl-induced Ca2+ signaling. CytD by itself did

not modify resting [Ca2+]i, nor had any effects on the

fluorescent probe, fura-2, as there was no changes in the

time course of DF340/F380 when fura-2-loaded cells where

monitored in the absence and presence of 10 AM CytD for

45 min (0.855F0.038 vs. 0.864F0.037, n=20). This treat-

ment did not cause any substantial alteration in cell

morphology. The presence of CytD also did not modify

Ca2+ influx through L-type channels, as similar changes in

[Ca2+]i were recorded in response to exposure of the cells to

60 mM KCl in the absence or presence of CytD (KCl:

capacitative Ca2+ channels. Representative original traces of CCE in the

itative Ca2+ channel blocker, 2-APB (100 AM) in the absence (A, C) and

the legend of Fig. 2. In panels B, D CytD was applied 45 min prior to TPS

ve-described conditions. Histograms are meanFSEM of 13–15 experiments.

S. Morales et al. / Cellular Signalling 17 (2005) 635–645 641

0.28F0.03 DF340/F380; KCl+CytD: 0.29F0.04 DF340/F380,

n=13, PN0.05, Fig. 4A and B). Consistent with these results,

JP also had no effects on KCl-induced Ca2+ entry (KCl:

0.153F0.032 DF340/F380; KCl+JP: 0.153F0.036 DF340/

F380, n=7, PN0.05, Fig. 4C).

To confirm our hypothesis that actin cytoskeleton

disruption specifically affected specifically CCE mecha-

nisms, we used 10 nM CCK to activate IP3-mediated Ca2+

release and activate CCE. As shown in Fig. 4D, we recorded

the typical pattern for SR depleting agents as the response

consisted of a transient elevation followed by a steady state

level slightly above the resting level. After CytD pre-

treatment, the peak, indicative of store depletion, was

unchanged (CCK: 0.37F0.06 DF340/F380; CCK+CytD:

0.40F0.07 DF340/F380, n=13 and 16 for control and CytD,

respectively, PN0.05, Fig. 4D and E). However, the steady

state level, which reflects Ca2+ entry activated by store

depletion, was significantly increased (CCK: 0.065F0.020

DF340/F380; CCK+CytD: 0.160F0.030 DF340/F380, n=13

and 16 for control and CytD, respectively, Pb0.05, Fig. 4D

and E). When cells were pre-treated with JP and then

challenged with 10 nM CCK there was not any significant

change in CCK-evoked Ca2+ transient (CCK: 0.246F0.044

DF340/F380; CCK+JP: 0.254F0.046 DF340/F380, n=10,

PN0.05, Fig. 4F). However, the sustained plateau indicative

Fig. 4. Effects of cytoskeleton disruption on KCl and CCK-induced Ca2+ transi

GBSM fura-2 loaded cells in response to 60 mM KCl and 10 nM CCK, in the ab

(panels C, F). These recordings are representative of 10–16 such experiments. In

drug at least 45 min prior to the stimuli.

of CCE, was significantly reduced (CCK: 0.031F0.006

DF340/F380; CCK+JP: 0.011F0.002 DF340/F380, n=10,

Pb0.01, Fig. 4F). These findings suggest that actin

disassembly does not change the size of the Ca2+ pools or

the release mechanisms of the stores but it favours CCE,

which is consistent with a specific inhibitory role of

cytoskeleton on CCE.

As the increase in CCE we found could be due to

alterations in the membrane potential and thus, to changes

in the driving force for Ca2+ influx, we performed

intracellular recordings in the whole mount gallbladder

preparation to test whether actin disruption affects the

resting membrane potentials. GBSM cells generate rhyth-

mic spontaneous action potentials with a depolarizing

spike that results from activation of L-type voltage-

dependent calcium channels [17]. As demonstrated in

Fig. 5, addition of 10 AM CytD did not cause a detectable

change in the resting membrane potential (control:

�48.0F4.3 mV; 3 min after 10 AM CytD: �47.5F4.1

mV; 10 min after 10 AM CytD: �50.3F5.1 mV; 30 min

after 10 AM CytD: �48.8F7.1 mV, n=3–7, PN0.05 vs.

control), the frequency of action potentials (control:

0.28F0.05 Hz; 3 min after 10 AM CytD: 0.31F0.05

mV; 10 min after 10 AM CytD: 0.23F0.04 mV; 30 min

after 10 AM CytD: 0.23F0.06 mV, n=3–7, PN0.05 vs.

ents. Representative original traces of changes in the fluorescence ratio in

sence (panels A, D) or presence of 10 AM CytD (panels B, E) or 10 AM JP

panels B, C, E, F cells were pre-incubated with the cytoskeleton modifying

Fig. 5. Actin cytoskeleton disassembly does not alter resting membrane potential or spontaneous action potentials. Representative transmembrane voltage

recordings (Vm) from an intact gallbladder smooth muscle cell showing the effects of 10 AM CytD. Expanded speed traces below demonstrate individual action

potentials recorded in the two experimental conditions (n=7). The resting membrane potential of the cell impaled was �42.7 mV.

S. Morales et al. / Cellular Signalling 17 (2005) 635–645642

control) or the shape of action potentials, indicating no

changes in membrane conductances that contribute to the

active or passive membrane properties of these cells.

3.3. CCE induced actin disassembly

Our results suggest that actin cytoskeleton manipulation

affects CCE, supporting a role for the actin cytoskeleton as a

Fig. 6. CCE activation causes actin cytoskeleton disassembly. (A) Changes in F-ac

stimulation of GBSM cells with 10 nM CCK and 60 mM KCl. Data are expressed a

meanFSEM of 7–11 experiments. These decreases in F-actin content are not relate

reduction when TPS and CCK were applied in 10 AM BAPTA loaded cells (B).

barrier for Ca2+ influx after store depletion. These data are the

result of using pharmacological tools to manipulate the

cytoskeleton, which might have a general effect on CCE,

although changes in the cytoskeleton should not participate in

the activation of CCE. To assess whether CCE activation is

related to changes in actin cytoskeleton, the F-actin content

was measured after inducing CCE with TPS in Ca2+-free

medium or 10 nM CCK. TPS treatment (for 30 min) caused a

tin contents after the depletion of stores with TPS in Ca2+ free medium and

s % of F-actin content in control or unstimulated conditions. Histograms are

d to TPS depletion- or CCK-induced increases in [Ca2+]i as there was also a

S. Morales et al. / Cellular Signalling 17 (2005) 635–645 643

reduction in F-actin content (7.32F0.95 vs. 4.72F0.45 a.u.,

26.2% reduction, n=12, Pb0.05, Fig. 6A). Application of 10

nM CCK for 7–8 min also decreased actin polymerization

(8.68F0.42 vs. 4.25F0.46 a.u., 51.0% reduction, n=8, P b

0.001, Fig. 6A). To our knowledge, this is the first

experimental evidence demonstrating that cell stimulation

with Ca2+ releasing-agonists induces a net and sustained actin

filament depolymerization. To investigate whether this effect

was due to Ca2+ store depletion we performed a series of

experiments using KCl depolarization. When 60 mM KCl

was applied, there was no significant change in F-actin

content (10.08F2.23 vs. 10.75F2.62 a.u., n=7, PN0.05, Fig.

6A). To rule out the possibility that increases in [Ca2+]ievoked by store depletion would be responsible for actin

reorganization, we repeated the TPS and CCK experimental

groups in BAPTA-loaded cells. BAPTA loading was able to

abolish the [Ca2+]i increases in response to both KCl and

CCK in fura-2 loaded cells (data not shown). As shown in

Fig. 6B, pre-treatment with BAPTA did not have any effect

on actin reorganization induced by both TPS and CCK (TPS:

8.79F0.672 vs. 5.91F0.052 a.u., 30.9% reduction, n=8,

Pb0.01; CCK: 8.90F0.56 vs. 3.94F0.43 a.u., 55.7 %

reduction, n=8, Pb0.001; Fig. 6B), suggesting that actin

depolymerization was mediated by store depletion (Fig. 6).

4. Discussion

The present study was conducted to elucidate the

possible role of actin cytoskeleton in CCE mechanisms in

GBSM cells. This study provides the first demonstration

that Ca2+ store depletion is linked to actin cytoskeleton

depolymerization to activate capacitative channels and

induce Ca2+ influx in naive smooth muscle cells. We

provide evidence for a novel negative regulatory role of

actin cytoskeleton on CCE.

In this cellular model, the FITC-phalloidin stained actin

cytoskeleton appears as a uniform structure in contrast with

previous reports involving cultured smooth muscle cells,

where stress fibers are evident [9,21,22]. Similar to our

results, actin cytoskeleton in isolated mesenteric artery

myocytes is more densely packed and is not organized into

thick strands [20], as in myocytes from intact preparations

[23,24]. Thus, actin stress fibers reflect an actin filament

reorganization that is associated with maintenance of cells in

culture.

In the current study, actin disassembly induced by pre-

treatment with two unrelated depolymerizing agents, CytD

and LatA, enhanced CCE, whereas actin stabilization

induced by JP caused a reduction of capacitative Ca2+

influx. These effects appear to be specific to a CCE

mechanism as [Ca2+]i mobilization from intracellular stores

or Ca2+ influx through L-type Ca2+ channles remained

insensitive to actin cytoskeleton dynamics.

A link between cytoskeletal structure and intracellular

levels of Ca2+ has been suggested in other studies, and a

number of underlying mechanisms have been proposed [25–

29]. The actin cytoskeleton may itself be part of the

intracellular Ca2+ store apparatus, and actin depolymeriza-

tion might result in cytoplasmic Ca2+ elevation [25]. This

mechanism probably is not responsible for the CytD effects

in our experiments because there was no changes in [Ca2+]iin response to CytD application in resting conditions.

Smooth muscle cells, as other excitable tissues, are

largely dependent on ionic conductances in the sarcolemma

[30], a cell region that is closely related to cytoskeleton.

This structural element could modify Ca2+ entry by

modulating Ca2+ channels or other ionic conductances in

the plasma membrane that determine membrane potential

[30]. However, results from this investigation rule out the

possibility of changes in the membrane potential and hence

driving force for Ca2+ entry to explain the increase in CCE

caused by disruption of actin cytoskeleton, as under CytD

treatment membrane membrane potential remained the

same. This was also confirmed by the lack of effects of

CytD in resting Ca2+ levels in isolated smooth muscle cells.

Provocative results have been reported regarding the role

of actin cytoskeleton on L-type Ca2+ channel activity in

smooth muscle; either inhibition or activation has been

reported for vascular cells [31,32] and no effects were found

on cells isolated from ileum [33]. In the present study, we

did not detect any change in KCl-induced [Ca2+]i elevation

in CytD or JP pre-treated cells compared to controls.

Moreover, when store-depletion induced-Ca2+ influx was

induced in the presence of 2-APB, to block capacitative

Ca2+ channels, and isolate store depletion-induced L-type

Ca2+ channel activation [3], no changes were induced by

modifications in actin cytoskeletal dynamics. In addition,

spontaneous action potentials, which involve Ca2+ entry

through L-type Ca2+ channels [17], were unaltered by CytD.

These results suggest that, at least in GBSM cells, L-type

Ca2+ channels are not regulated by actin cytoskeleton,

consistent with results reported for another gastrointestinal

smooth muscle type [33]. The discrepancy with the results

reported by Gokina and Osol [32] could be due to

interferences of mechanoreception-induced changes in the

cytoskeleton, as this study was performed in pressurized

arteries. On the other hand, the study of Nakamura et al.

[31] was performed in cultured vascular cells, and, as

discussed above, maintenance of smooth muscle cells in

culture conditions induces an obvious rearrangement in the

cytoskeleton, which could also be responsible for changes in

the regulatory functions of this element.

Several investigations have suggested the existence of a

physical relationship between the ER and the actin

cytoskeleton [26,34]. Moreover, there is evidence for

impaired function of ER enzymes and/or receptors in cells

with cytoskeletal disruption. In agreement with this, it has

been proposed that the cytoskeleton is involved in the

regulation of IP3 binding, IP3 receptor-mediated Ca2+

release and the spatial relationship between plasma mem-

brane and IP3 receptors [26,35,36]. Our results do not

S. Morales et al. / Cellular Signalling 17 (2005) 635–645644

indicate a primary role for the cytoskeleton in affecting Ca2+

release through IP3Rs, as the Ca2+ transients evoked by the

IP3 releasing agent, CCK remain unaltered after CytD or JP

pre-treatment. This might be related to the lack of changes

in cell shape induced by cytoskeletal alterations in our

cellular model. However, the significant changes in CCK-

evoked [Ca2+]i plateau induced by CytD and JP treatment,

clearly indicate a role of the cytoskeleton in the communi-

cation between depleted stores and CCE channels in the

plasma membrane.

The activation of Ca2+ entry following depletion of

intracellular Ca2+ stores is a signalling process of great

relevance both in excitable and non-excitable cells. In this

process, studies on the role of the actin cytoskeleton have

provided conflicting results in different cell types. In this

context, it is important to differentiate between the roles

suggested to the experimentally induced cortical membrane-

associated actin barrier and the normal cytoplasmic actin

network. The former has been found to act as a negative

clamp preventing Ca2+ entry in several cell types such as

cultured smooth muscle cells [9], platelets [29], corneal

endothelial cells [10], DT40 lymphocytes [37] and pancre-

atic acinar cells [11]. On the other hand, a role for the

integrity of the cytosolic actin network in CCE has been

demonstrated by using cytoskeletal disrupters in some cells,

including vascular endothelial cells [38], astrocytes [39],

platelets [29], HepG2 cells [40], pancreatic acinar cells [11],

while this experimental manoeuvre had no effect in NIH

3T3 [26], cultured smooth muscle cells [9] or RBL-1 cells

[12]. Although the reasons for this discrepancy are unclear,

the proximity of the ER to the plasma membrane, and

therefore the requirement of transport of portions of the ER

toward the cell membrane, supported by the actin cytoske-

leton, might be important.

Using different cytoskeletal modifications our results

provide new insight regarding the involvement of the actin

cytoskeleton in the activation of CCE in a naive smooth

muscle cellular model. Our results clearly indicate that the

actin cytoskeleton in GBSM cells acts as a barrier

preventing conformational coupling between the sarcoplas-

mic reticulum and plasma membrane. In these cells, store

depletion stimulated by the physiological agonist CCK or

induced by TPS resulted in a net decrease in the F-actin

content. To our knowledge, this is the first demonstration

that the disruption of actin cytoskeleton can be linked to

store depletion. This effect was found to be entirely

dependent on store depletion and not on rises in [Ca2+]isince BAPTA loading did not modify the effects of CCK or

TPS on the F-actin content. Consistent with this finding,

disruption of the GBSM cell actin filament network with

CytD or LatA was found to facilitate the coupling

mechanism increasing CCE whereas increased actin poly-

merization by JP significantly inhibited CCE in these cells.

Taken together, the findings of this investigation indicate

that the conformation coupling model is the paradigm that

best describes the activation of CCE in GBSM cells. In

contrast to the model described by Patterson and co-workers

[9] in cultured smooth muscle cells, in GBSM cells CytD-

or LatA-treatment increased CCE. In this cellular model

vesicle trafficking or any other event that requires the

support of the actin cytoskeleton does not seem to be needed

for CCE. Our results are inconsistent with the hypothesis

that Ca2+ pool depletion promotes an insertion of vesicles

containing capacitative Ca2+ entry channels [41], since such

an exocytotic mechanism usually depends on an intact

cytoskeletal network. Instead, the physiological cortical

actin network would act as a barrier, preventing coupling

between elements in the SR and plasma membrane. Thus,

CCE is facilitated by cytoskeletal disrupters and prevented

by stabilization of the cortical barrier. Although speculative,

the different sources of the smooth muscle cells and/or the

use of cultured or fresh cells might account for these

differences. In summary, our results support a coupling

mechanism involving direct, spatially restricted interactions

between SR and plasma membrane proteins in the activation

of CCE in GBSM cells. The intracellular signaling between

depleted Ca2+ stores and changes in the actin cytoskeletal

dynamics remains to be determined.

Acknowledgements

The authors wish to thank Dr Onesmo Balemba for

assistance with the intracellular electrophysiological studies

and to M.P. Delgado for her technical assistance. This work

was supported by the Spanish MCyT grant SAF-2001-0295

0295 (MJP) and NIH grant NS26995 (GMM). S. Morales is

supported by Ministry of Education Predoctoral Research

Grant.

References

[1] K.M. Sanders, J. Appl. Physiol. 91 (2001) 1438.

[2] J.W. Putney, Cell Calcium 7 (1986) 1.

[3] S. Morales, P.J. Camello, S. Alcon, G.M. Salido, G. Mawe, M.J.

Pozo, Am. J. Physiol., Gastrointest. Liver Physiol. 286 (2004)

G1090.

[4] J.W. Putney, R.R. McKay, Bioessays 21 (1999) 38.

[5] C.L. Carpenter, Crit. Care Med. 28 (2000) N94.

[6] P.A. Janmey, Physiol. Rev. 78 (1998) 763.

[7] D. Mehta, S.J. Gunst, J. Physiol. 519 (1999) 829.

[8] C. Li, Q. Xu, Cell. Signal. 12 (2000) 435.

[9] R.L. Patterson, D.B. van Rossum, D.L. Gill, Cell 98 (1999) 487.

[10] Q. Xie, Y. Zhang, C. Zhai, J.A. Bonanno, J. Biol. Chem. 277 (2002)

16559.

[11] P.C. Redondo, A.I. Lajas, G.M. Salido, A. Gonzalez, J.A. Rosado,

J.A. Pariente, Biochem. J. 370 (2003) 255.

[12] D. Bakowski, M.D. Glitsch, A.B. Parekh, J. Physiol. 532 (2001)

55.

[13] D. Bakowski, R.D. Burgoyne, A.B. Parekh, J. Physiol. 553 (2003)

387.

[14] T. Smani, S.I. Zakharov, P. Csutora, E. Leno, E.S. Trepakova, V.M.

Bolotina, Nat. Cell Biol. 6 (2004) 113.

[15] M.J. Pozo, G.J. Perez, M.T. Nelson, G.M. Mawe, Am. J. Physiol.:

Gasterointest. Liver Physiol. 282 (2002) G165.

S. Morales et al. / Cellular Signalling 17 (2005) 635–645 645

[16] J.A. Rosado, S.O. Sage, Biochem. J. 347 (2000) 183.

[17] L. Zhang, A.D. Bonev, M.T. Nelson, G.M. Mawe, Am. J. Physiol. 265

(1993) C1552.

[18] I. Spector, N.R. Shochet, D. Blasberger, Y. Kashman, Cell Motil.

Cytoskelet. 13 (1989) 127.

[19] M.R. Bubb, A.M. Senderowicz, E.A. Sausville, K.L. Duncan, E.D.

Korn, J. Biol. Chem. 269 (1994) 14869.

[20] V. Pucovsky, R.F. Moss, T.B. Bolton, J. Physiol. 552 (2003) 119.

[21] S. Saegusa, H. Tsubone, M. Kuwahara, Eur. J. Pharmacol. 413 (2001)

163.

[22] S.S. An, R.E. Laudadio, J. Lai, R.A. Rogers, J.J. Fredberg, Am. J.

Physiol., Cell Physiol. 283 (2002) C792.

[23] M.J. Cipolla, N. Lessov, E.S. Hammer, A.B. Curry, Stroke 32 (2001)

1658.

[24] S.M. Smiley-Jewell, M.U. Tran, A.J. Weir, Z.A. Johnson, L.S. Van

Winkle, C.G. Plopper, J. Appl. Physiol. 93 (2002) 1506.

[25] K. Lange, U. Brandt, FEBS Lett. 395 (1996) 137.

[26] C.M. Ribeiro, J. Reece, J.W. Putney Jr., J. Biol. Chem. 272 (1997)

26555.

[27] P.A. Janmey, J. Kas, J.V. Shah, P.G. Allen, J.X. Tang, Biol. Bull. 194

(1998) 334.

[28] Z. Wu, K. Wong, M. Glogauer, R.P. Ellen, C.A. McCulloch, Biochem.

Biophys. Res. Commun. 261 (1999) 419.

[29] J.A. Rosado, S.O. Sage, J. Physiol. 526 (2000) 221.

[30] T.B. Bolton, S.A. Prestwich, A.V. Zholos, D.V. Gordienko, Annu.

Rev. Physiol. 61 (1999) 85.

[31] M. Nakamura, M. Sunagawa, T. Kosugi, N. Sperelakis, Am. J.

Physiol., Cell Physiol. 279 (2000) C480.

[32] N.I. Gokina, G. Osol, Am. J. Physiol., Heart Circ. Physiol. 282 (2002)

H1410.

[33] K. Obara, H. Yabu, Eur. J. Pharmacol. 255 (1994) 139.

[34] O. Baumann, B. Walz, Int. Rev. Cytol. 205 (2001) 149.

[35] L.Y. Bourguignon, N. Iida, H. Jin, Cell Biol. Int. 17 (1993) 751.

[36] Y. Wang, M.P. Mattson, K. Furukawa, J. Neurochem. 82 (2002) 945.

[37] H.T. Ma, K. Venkatachalam, J.B. Parys, D.L. Gill, J. Biol. Chem. 277

(2002) 6915.

[38] J.R. Holda, L.A. Blatter, FEBS Lett. 403 (1997) 191.

[39] M. Grimaldi, A. Favit, D.L. Alkon, J. Biol. Chem. 274 (1999) 33557.

[40] J.A. Rosado, I. Rosenzweig, S. Harding, S.O. Sage, Am. J. Physiol.,

Cell Physiol. 280 (2001) C1636.

[41] B. Somasundaram, J.C. Norman, M.P. Mahaut-Smith, Biochem. J.

309 (1995) 725.