Prevention of RhoA activation and cofilin-mediated actinpolymerization mediates the antihypertrophic effect of adenosinereceptor agonists in angiotensin II- and endothelin-1-treatedcardiomyocytes

Asad Zeidan • Xiaohong Tracey Gan •

Ashley Thomas • Morris Karmazyn

Received: 17 July 2013 / Accepted: 26 September 2013 / Published online: 6 October 2013

� Springer Science+Business Media New York 2013

Abstract Adenosine receptor activation has been shown to

be associated with diminution of cardiac hypertrophy and it

has been suggested that endogenously produced adenosine

may serve to blunt pro-hypertrophic processes. In the present

study, we determined the effects of two pro-hypertrophic

stimuli, angiotensin II (Ang II, 100 nM) and endothelin-1

(ET-1, 10 nM) on Ras homolog gene family, member A

(RhoA)/Rho-associated, coiled-coil containing protein kinase

(ROCK) activation in cultured neonatal rat ventricular myo-

cytes and whether the latter serves as a target for the anti-

hypertrophic effect of adenosine receptor activation. Both

hypertrophic stimuli potently increased RhoA activity with

peak activation occurring 15–30 min following agonist

addition. These effects were associated with significantly

increased phosphorylation (inactivation) of cofilin, a down-

stream mediator of RhoA, an increase in actin polymerization,

and increased activation and nuclear import of p38 mitogen

activated protein kinase. The ability of both Ang II and ET-1 to

activate the RhoA pathway was completely prevented by the

adenosine A1 receptor agonist N6-cyclopentyladenosine, the

A2a receptor agonist 2-p-(2-carboxyethyl)-phenethylamino-

50-N-ethylcarboxamidoadenosine, the A3 receptor agonist

N6-(3-iodobenzyl)adenosine-50-methyluronamide as well as

the nonspecific adenosine analog 2-chloro adenosine. All

effects of specific receptor agonists were prevented by their

respective receptor antagonists. Moreover, all adenosine

agonists prevented either Ang II- or ET-1-induced hypertro-

phy, a property shared by the RhoA inhibitor Clostridium

botulinum C3 exoenzyme, the ROCK inhibitor Y-27632 or the

actin depolymerizing agent latrunculin B. Our study therefore

demonstrates that both Ang II and ET-1 can activate the RhoA

pathway and that prevention of the hypertrophic response to

both agonists by adenosine receptor activation is mediated by

prevention of RhoA stimulation and actin polymerization.

Keywords Adenosine receptor activation �Cardiomyocyte hypertrophy � RhoA/ROCK pathway �Cofilin phosphorylation � Actin polymerization � p38

Nuclear translocation

Introduction

Adenosine, a product of adenine nucleotide catabolism, has

been demonstrated to exert numerous effects on the car-

diovascular system, which are mediated by activation of

various receptor subtypes. The primary adenosine receptor

in the myocardium is the A1 subtype which is linked to Gi-

mediated inhibition of adenylate cyclase although both

A2a/b and A3 receptors have also been identified [1–4].

There is increasing evidence that endogenously produced

adenosine is an important negative regulator of the

hypertrophic and remodeling processes which contribute to

heart failure. Thus, from a clinical perspective, an impor-

tant observation linking adenosine to the heart failure

process was the report that plasma levels of the nucleoside

are elevated in patients with heart failure irrespective of

causative factor [5]. Moreover, the degree of elevation was

dependent on the severity of heart failure according to New

York Heart Association (NYHA) classification with the

greatest increases (more than five-fold) observed in NYHA

A. Zeidan � X. T. Gan � A. Thomas � M. Karmazyn (&)

Department of Physiology and Pharmacology, Schulich School

of Medicine and Dentistry, University of Western Ontario,

London, ON N6A 5C1, Canada

e-mail: morris.karmazyn@schulich.uwo.ca

A. Zeidan

Department of Anatomy, Cell Biology and Physiological

Sciences, American University of Beirut, Beirut, Lebanon

123

Mol Cell Biochem (2014) 385:239–248

DOI 10.1007/s11010-013-1832-2

class IV patients. The levels of adenosine were signifi-

cantly correlated with plasma norepinephrine levels [5].

These investigators also showed that the nucleoside trans-

port inhibitors dipyridamole and dilazep (which increase

adenosine levels) reduced the severity of heart failure

although the benefit reversed after drug discontinuation [6].

Experimental observations have shown a direct antihy-

pertrophic effect of adenosine receptor agonists on cardio-

myocytes, which appears to be mediated by multiple

adenosine receptor subtypes [7]. Further evidence obtained

from in vivo studies also demonstrates a salutary effect of

adenosine in reversing ventricular remodeling following

aortic coarctation in rats [8]. In addition to direct effects of

adenosine receptor activation, it has also been demonstrated

that deficiency in ecto-50-nucleotidase which catalyzes the

conversion of extracellular AMP to adenosine, thus

increasing extracellular adenosine production, increases the

degree of cardiac hypertrophy following aortic banding [9].

When taken together, the evidence supports the contention

that endogenous adenosine functions to limit the hypertro-

phic and remodeling processes which contribute to the

development of heart failure.

The RhoA/RhoA kinase (ROCK) pathway (members of

the Rho family GTPases) has, over the past few years,

emerged as a potential mediator of cardiac pathology

especially the development of cardiomyocyte hypertrophy

(reviewed in [10, 11]). There is now emerging strong

evidence implicating the RhoA/ROCK pathway as an

important target for mitigating the hypertrophic process

and reducing heart failure [12–17]. In view of the

increasing evidence implicating RhoA/ROCK in hyper-

trophy, the present study was carried out to address two

primary questions. First, we wished to determine whether

cardiomyocyte hypertrophy produced by either angiotensin

II (Ang II) or endothelin-1 (ET-1) is associated with altered

RhoA/ROCK activity culminating in cofilin-regulated actin

remodeling. Second, we determined whether the antihy-

pertrophic effects of adenosine receptor agonists are med-

iated by modulation of the RhoA/ROCK pathway.

Materials and methods

Neonatal rat cardiomyocyte culture

Experiments were carried out using cardiomyocytes isolated

from 1 to 3 day old Sprague–Dawley rats (Charles River

Canada, St. Constant, QC, Canada) as described previously

[7]. Myocytes were cultured at 37 �C in media containing

10 % fetal bovine serum for 24 h and then serum starved

for further 24 h before initiating treatments. The protocol

for myocyte isolation was approved by the University of

Western Ontario Animal Care and Use Subcommittee and

conformed to guidelines of the Canadian Council on Ani-

mal Care (Ottawa, ON, Canada).

Experimental design

Cardiomyocyte hypertrophy was induced by the addition of

either 100 nM Ang II or 10 nM ET-1 for up to 24 h,

concentrations which represented the maximum hypertro-

phic response observed as assessed by cell surface area in

the absence of any changes in myocyte morphology as

evidenced by light microscopy. To determine the effect of

adenosine receptor agonists on either Ang II- or ET-1-

induced changes, these agents were added 30 min before

administration of the hypertrophic stimuli. Additional

experiments were also performed to confirm receptor

specificity of individual agonists by assessing the effect of

their respective antagonists in which the latter was added

15 min prior to the addition of the agonist. We also

determined the effect of 2-chloro adenosine, a non-specific

analog of adenosine which is resistant to cellular reuptake

by nucleoside transport or metabolism through adenosine

deaminase. In addition, experiments were carried out to

determine whether pharmacological inhibition of the RhoA

pathway mimics the effect of adenosine receptor agonists

using an identical addition protocol. Myocytes were treated

for various durations with the hypertrophic stimuli

reflecting the parameter under study, as indicated in

‘‘Results’’ section.

Drugs and reagents

Ang II (100 nM) and ET-1 (10 nM) were obtained from

Sigma-Aldrich (Oakville, ON, Canada). To probe the

adenosine system we used the adenosine A1 receptor agonist

N6-cyclopentyladenosine (CPA, 1 lM), the A2a receptor

agonist 2-p-(2-carboxyethyl)-phenethylamino-50-N-ethylcar-

boxamidoadenosine (CGS21680, 100 nM) and the A3 recep-

tor agonist N6-(3-iodobenzyl)adenosine-50-methyluronamide

(IB-MECA, 100 nM) in the absence or presence of their

respective antagonists 8-cyclopentyl-1,3-dipropylxanthine

(DPCPX, 10 lM), 8-(3-chlorostyryl) caffeine (CSC, 10 lM),

and 3-propyl-6-ethyl-5[ethyl(thio)carbonyl]-2-phenyl-4-

propyl-3-pyridinecarboxylate (MRS 1523, 1 lM). The

concentrations of adenosine agonists were selected based

on our previous study with these agents confirming anti-

hypertrophic effects of the agonists and selective inhibition

of these effects by their respective antagonists [7]. All

adenosine receptor agonists and antagonists were pur-

chased from Sigma-Aldrich. The RhoA inhibitor Clos-

tridium botulinum C3 exoenzyme (C3, 30 ng/ml, Alexis

Biochemicals, Carlsbad, CA, USA), the ROCK inhibi-

tor (R)-(?)-trans-4-(1-aminoethyl)-N-(4-pyridyl)cyclohex-

anecarboxamide dihydrochloride monohydrate (Y-27632,

240 Mol Cell Biochem (2014) 385:239–248

123

10 lM, Sigma-Aldrich), and the actin-depolymerizing

agent latrunculin B (10 nM, Calbiochem, San Diego, CA,

USA) were studied to determine whether they can mimic

the antihypertrophic effects of adenosine receptor ago-

nists and were added in a similar manner as the latter

compounds.

Measurement of cell surface area

To determine the cell surface area we used SigmaScan

software (Systat, Inc., San Jose, CA, USA). At least 50

randomly selected cells were measured and averaged to

obtain one individual value.

Determination of RhoA activation

To assess RhoA activity we measured RhoA–GTP levels

using the RhoA G-LISA Activity Biochem Assay Kit as

per the manufacturer’s protocol (Cytoskeleton, Denver,

CO, USA). Measurements were performed using a Spec-

traMax M5 (Molecular Devices, Sunnyvale, CA, USA)

plate reader at an absorbance of 490 nm.

Preparation of subcellular fractions

Cytosolic- and nuclei-enriched fractions were prepared using

differential centrifugation as previously described [18]. In

brief, cell lysates were collected and homogenized in a cold

buffer containing 20 mM Tris–HCl, 2 mM EDTA, 137 mM

NaCl, 1 mM sodium orthovanadate, 2 mM sodium pyro-

phosphate, 10 % glycerol, 1 mM 4-(2-aminoethyl)-benzene-

sulfonyl fluoride, and 10 mg/ml leupeptin (buffer A). After

clarification of the homogenate by centrifugation at 7509g for

20 min at 4 �C, the collected lysate was further centrifuged at

10,0009g for 20 min at 4 �C and the cytosolic-enriched

fraction (supernatant) was obtained. The remaining pellet was

resuspended in a second cold buffer B (buffer A with 2 %

SDS) and kept on ice to be used as the nuclear-containing

membrane fraction. Fractions were then used for the deter-

mination of p38 levels using western blotting.

Western blotting

Western blotting was carried out to determine protein

contents of total and phosphorylated cofilin and p38

mitogen-activated protein kinase (MAPK) as well as

globular (G) and filamentous (F) actin contents, the latter

described separately in the following section. Total cellular

lysates were collected using a lysis buffer and protease

cocktail inhibitor mixture as previously described [18, 19].

Proteins were loaded equally on SDS gels after protein

quantification using BioRad Dye Reagent (BioRad,

Hercules, CA, USA) according to the manufacturer’s

instructions. The primary antibodies and respective dilu-

tions used in this study include total (1:1,000 dilution,

Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and

phosphorylated p38 (Thr180/Tyr182) forms (1:1,000 dilu-

tion, Cell Signaling, Danvers, MA, USA) as well as

phosphorylated (1:1,000 dilution, Santa Cruz) and total

cofilin-2 (1:1,000 dilution, Millipore, Billerica, MA, USA).

Goat-anti-rabbit IgG and goat-anti-mouse IgG HRP con-

jugates (BioRad) were used at 1:10,000 dilution and don-

key-anti-goat IgG HRP conjugate (Santa Cruz) was used at

1:10,000 dilution as appropriate. b-Actin (1:1,000 dilution,

Cytoskeleton) was used for cytosolic and nuclear loading

controls, respectively. Spot densitometry using FluorChem

(Alpha Innotech Corporation, Santa Clara, CA, USA)

software was performed for protein quantification.

Determination of G–F actin ratios

The ratio of G–F actin was determined as described previously

[18, 19] using a commercially available kit (Cytoskeleton). In

brief, myocytes were collected in F-actin stabilization buffer

and centrifuged to pellet F-actin while leaving G-actin in

the supernatant. The F-actin was then dissociated by 1 lM

cytochalasin D, and both fractions were subjected to western

blotting using an anti-actin antibody (1:1,000, Cytoskeleton)

as described above. Furthermore, we used fluorescence

microscopy to corroborate data obtained with western blot-

ting. For these analyses cardiomyocytes grown on glass cov-

erslips were fixed in 4 % (w/v) paraformaldehyde, and stained

with fluorescin isothiocyanate–phalloidin (FITC–phalloidin

1 lg/ml) to stain F-actin and with Texas Red-labeled DNase I

(10 lg/ml) to stain G-actin. Confocal images of F- and

G-actin were captured simultaneously with a fluorescence

Zeiss LSM 510 microscope (Carl Zeiss, Oberkochen, Ger-

many) at 490/525 nm (excitation/emission) for FITC–phal-

loidin and at 596/615 nm (excitation/emission) for Texas

Red-labeled DNase I.

Statistical analyses

Data were analyzed using a one-way ANOVA and group

differences were detected using a Student–Newman–Keuls

post hoc test. A P value of \0.05 was defined as representing

statistically significant differences between treatment groups.

Results

Adenosine receptor agonists abrogate Ang II- and

ET-1-induced RhoA activation

We first examined whether Ang II or ET-1 can activate

RhoA in cardiomyocytes. As shown in Fig. 1 both Ang II

Mol Cell Biochem (2014) 385:239–248 241

123

and ET-1 markedly increased RhoA activity by up to

approximately threefold, with peak activation evident

30 min after agonist addition and values returning to

control by 60 min. 2-Chloro adenosine, the non-metabo-

lizable adenosine analog almost completely prevented the

ability of either Ang II or ET-1 to activate RhoA

(Fig. 1). Furthermore, the effect of 2-chloro adenosine

was shared by all three adenosine receptor agonists that

were examined in the present study. The ability of

adenosine agonists to inhibit activation of RhoA was

abrogated in the presence of their respective receptor

antagonists (Fig. 2).

0

50

100

150

200

250

300

350R

ho

A a

ctiv

atio

n (

% C

on

tro

l)

ET-1 ET-1 +CPA ET-1 + CGS ET-1 + IB

*

**

*# # #

# ##

# #

# ##

## # # #

ET-1 + 2-Cl-A

0

50

100

150

200

250

300

350

Ang II Ang II + CPA Ang II + CGS Ang II + IB

Rh

oA

act

ivat

ion

(%

Co

ntr

ol)

*

* *

# # #

# ## #

#

# ##

## # # #

Ang II + 2-Cl-A

0 min 5 min 15 min 30 min 60 min

Fig. 1 Activation profile of

RhoA following addition of

either angiotensin II (Ang II, top

panel) or endothelin-1 (ET-1,

bottom panel) in the absence or

presence of the adenosine A1

receptor agonist N6-

cyclopentyladenosine (CPA,

1 lM), the A2a receptor agonist

2-p-(2-carboxyethyl)-

phenethylamino-50-N-

ethylcarboxamidoadenosine

(CGS, 100 nM), the A3 receptor

agonist N6-(3-

iodobenzyl)adenosine-50-methyluronamide (IB, 100 nM)

or the non-specific agonist

2-chloro adenosine (2-Cl-Ad,

10 lM). Values represent

means ? SE, N = 8. *P \ 0.05

from 0 min values; #P \ 0.05

from respective time value in

the absence of adenosine

agonists

Rh

oA

act

ivat

ion

(%

Co

ntr

ol)

0

50

100

150

200

250

300

350

* ** *

Rh

oA

act

ivat

ion

(%

Co

ntr

ol)

0

50

100

150

200

250

300

* * * *

Fig. 2 Prevention of specific adenosine receptor agonist-mediated

inhibition of either angiotensin II (Ang II, left panel) or endothelin-1

(ET-1, right panel) induced RhoA activation in the presence of their

respective antagonists. Abbreviations for adenosine agonists as for

Fig. 1. Adenosine receptor antagonists were the A1 antagonist

8-cyclopentyl-1,3-dipropylxanthine (DPCPX, 10 lM), the A2a

receptor antagonist 8-(3-chlorostyryl) caffeine (CSC, 10 lM), and the

A3 receptor antagonist 3-propyl-6-ethyl-5[ethyl(thio)carbonyl]-2-phe-

nyl-4-propyl-3-pyridine carboxylate (MRS 1523, 1 lM). All data

represent values obtained 15 min after addition of either Ang II or ET-1

except control group which represents 15 min time control values.

Values represent means ? SE, N = 6. *P \ 0.05 from control

242 Mol Cell Biochem (2014) 385:239–248

123

Ang II- and ET-1-induced RhoA activation is

associated with increased cofilin phosphorylation

and actin polymerization which is prevented

by adenosine receptor activation

Downstream targets for RhoA were also determined

including cofilin phosphorylation and the effects on actin

dynamics. With respect to the former, both Ang II and

ET-1 produced a significant increase in cofilin phosphor-

ylation as determined 24 h after addition of either pro-

hypertrophic agonist although these effects were prevented

by adenosine receptor agonists (Fig. 3). In contrast, aden-

osine receptor agonists exerted no effect in the presence of

their respective antagonists (Fig. 3). Moreover, both Ang II

and ET-1 produced a significant reduction in the G/F actin

ratio thus demonstrating a proportional increase in actin

polymerization (Fig. 4). As also shown in Fig. 4, all three

adenosine receptor agonists significantly prevented the

ability of Ang II and ET-1 to increase actin polymerization

to values not significantly different from control. The

ability of specific adenosine receptor agonists to prevent

changes in actin dynamics was reversed by their respective

antagonists whereas they were mimicked by 2-chloro-

adenosine (not shown).

Adenosine receptor agonists prevent Ang II- or ET-1-

induced p38 MAPK phosphorylation and translocation

Phosphorylation and translocation of MAPK p38 have been

shown to be regulated by RhoA signaling and we therefore

determined whether adenosine agonists can modulate the

effect of Ang II or ET-1 on p38. As shown in Fig. 5, both

Ang II and ET-1 significantly increased p38 phosphoryla-

tion although this was completely prevented by adenosine

receptor agonists. In addition, the administration of either

hypertrophic stimulus resulted in a robust and significant

translocation of p38 MAPK into nuclei which was pre-

vented by adenosine agonists (Fig. 6). Although not shown

in the figures, none of the adenosine receptor agonists

exerted any effect on either p38 MAPK phosphorylation or

its translocation into nuclei in the presence of their

respective receptor antagonists.

0

40

80

120

160

200

Cof

ilin

Pho

spho

ryla

tion

(% c

ontr

ol)

* * * *# # # #

0

50

100

150

200

250

300

Cof

ilin

Pho

spho

ryla

tion

(% c

ontr

ol) * * * *

# # # #

P-Cofilin

T-CofilinP-Cofilin

T-Cofilin

Fig. 3 Cofilin phosphorylation

24 h following addition of

either angiotensin II (Ang II, top

panel) or endothelin-1 (ET-1,

bottom panel) in the absence or

presence of specific adenosine

receptor agonists either alone or

in combination with their

respective antagonists.

Description of adenosine

receptor agonists and

antagonists as for Figs. 1 and 2.

Values represent means ? SE,

N = 5. *P \ 0.05 from control

(0 time); #P \ 0.05 from

respective value with either Ang

II or ET-1 alone. Representative

western blots for either

phosphorylated or total cofilin

are shown under the appropriate

bar

Mol Cell Biochem (2014) 385:239–248 243

123

Adenosine receptor agonists and inhibitors of the RhoA

pathway share a common ability to prevent Ang II-

and ET-1-induced cardiomyocyte hypertrophy

Finally, we determined the hypertrophic response to either

Ang II or ET-1 and assessed the relative abilities of

adenosine agonists and RhoA inhibitors to affect these

responses. As shown in Table 1, both Ang II and ET-1

produced a robust hypertrophic response as assessed by

either cell surface area or expression of ANP although

these effects were significantly inhibited by both adenosine

receptor agonists and RhoA inhibitors.

Ang II

Control

Ang II +CPA

Control

ET-1

ET-1+CPA

G-actin F-actin G-actin F-actin

G/F

-Act

in r

atio

G/F

-Act

in r

atio

0

1

2

3

4

Co

ntr

ol

An

g II

*

##

#

An

g II

+ C

PA

An

g II

+ C

GS

An

g II

+ IB

G-actin

F-actin

0

1

2

3

4

Co

ntr

ol

*

# #

#

ET -

1 +

CP

A

ET-

1 +

CG

S

ET -

1 +

IB

ET-

1

G-actin

F-actin

Merge Merge

Fig. 4 G- and F-actin ratios 24 h following addition of either

angiotensin II (Ang II, top left panel) or endothelin-1 (ET-1, top right

panel) in the absence or presence of specific adenosine receptor

agonists. Description of adenosine receptor agonists is as for Fig. 1.

Values represent means ? SE, N = 6. *P \ 0.05 from control (0

time); #P \ 0.05 from respective value with either Ang II or ET-1

alone. Representative western blots for G and F actin are shown under

the appropriate bar. Bottom panels show examples of confocal images

for G and F actin staining of myocytes treated with Ang II (left) or

ET-1 (right) in the absence or presence of CPA although completely

identical images were seen with all adenosine receptor agonists

0

50

100

150

200

250

300

p38

phos

phor

lyat

ion

(% C

ontr

ol)

p-p38

T-p38

*

# # # #

0

50

100

150

200

250

300

350

*

# ## #

p38

phos

phor

lyat

ion

(% C

ontr

ol)

p-p38

T-p38

Fig. 5 p38 phosphorylation 24 h following addition of either angio-

tensin II (Ang II, left panel) or endothelin-1 (ET-1, right panel) in the

absence or presence of specific adenosine receptor agonists. Descrip-

tion of adenosine receptor agonists as for Fig. 1. Values represent

means ? SE, N = 5. *P \ 0.05 from control (0 time); #P \ 0.05

from respective value with either Ang II or ET-1 alone. Represen-

tative western blots for either phosphorylated or total p38 are shown

under the appropriate bar

244 Mol Cell Biochem (2014) 385:239–248

123

Discussion

Emerging evidence suggests that adenosine exerts antihy-

pertrophic effects on the heart although the mechanistic

bases for these effects are not well understood (reviewed in

[20]). The antihypertrophic effect of adenosine as well as

evidence that plasma adenosine levels are elevated in

patients with heart failure [5] may be of substantial

importance as it suggests that the endogenous production

of the nucleoside may represent a mechanism to limit

myocardial hypertrophy and remodeling. Adenosine exerts

its effect via activation of various receptor subtypes most

of which are expressed on the cardiomyocyte [1–4]. These

receptors are markedly different in terms of cardiac cell

signaling process including the nature of second messenger

involvement and subsequent cardiac effects (reviewed in

[20]). Thus, it is quite remarkable that activation of at least

three of these receptors including the A1 receptor, the A2a

receptor, and the A3 receptor exerts virtually identical

effects in terms of limiting hypertrophic responses as pre-

viously demonstrated against the hypertrophic effect of the

a1 adrenoceptor agonist phenylephrine [7, 21], and as

shown in the present report against both Ang II and ET-1.

The specificity of adenosine receptor involvement was

reinforced by the fact that the effects of adenosine receptor

agonists were completely suppressed by their respective

antagonists (not shown). The similarity in adenosine

receptor-mediated effects on cardiomyocytes is surprising

in view of diversity of adenosine receptor-mediated cell

signaling processes, although this bears some similarity to

Nuclear fraction

Cytoplasmic fraction

T-p38 (Nuclear)

T-p38 (Cytoplasmic)

T-p3

8 co

nten

t (%

Con

trol)

0

50

100

150

200

250

*

*

#* #*#*

#

T-p3

8 co

nten

t (%

Con

trol)

0

50100

150

200

250

300

#* #* #* #

*

T-p38 (Nuclear)T-p38 (Cytoplasmic)

Fig. 6 Protein levels of total p38 in nuclear and cytosolic fraction

24 h following addition of either endothelin-1 (ET-1, left panel) or

angiotensin II (Ang II, right panel) in the absence or presence of

specific adenosine receptor agonists. Description of adenosine

receptor agonists as for Fig. 1. Values represent means ? SE,

N = 5. *P \ 0.05 from respective nuclear or cytosolic control (0

time) values; #P \ 0.05 from respective value with either Ang II or

ET-1 alone. Representative western blots are shown under the

appropriate bars

Table 1 Comparative effects of RhoA pathway inhibitors and adenosine receptor agonists on cardiomyocyte hypertrophy produced by

angiotensin II or endothelin 1 as determined by cell surface area and expression of ANP

Parameters RhoA inhibitors Adenosine agonists

Ang II alone Ang II ? C3 Ang II ? Y Ang II ? Lat B Ang II ? CPA Ang II ? CGS Ang II ? IB Ang II ? 2-Cl-Ad

CSA 129 ± 4.1* 99 ± 4.8# 99 ± 3.7# 99 ± 3.2# 101 ± 3.2# 101 ± 3.7# 98 ± 4.0# 94 ± 5.4#

ANP 178 ± 16.3* 117 ± 9.2# 122 ± 8.4# 124 ± 9.3# 129 ± 16.1# 111 ± 22.4# 89 ± 16.4# 126 ± 12.8#

ET-1 alone ET-1 ? C3 ET-1 ? Y ET-1 ? Lat B ET-1 ? CPA ET-1 ? CGS ET-1 ? IB ET-1 ? 2-Cl-Ad

CSA 133 ± 3.8* 97 ± 4.1# 98 ± 4.7# 95 ± 7.7# 103 ± 4.6# 100 ± 5.1# 101 ± 4.4# 98 ± 5.7#

ANP 187 ± 12.9* 123 ± 8.3# 126 ± 11.4# 127 ± 12.0# 114 ± 12.4# 107 ± 9.5# 115 ± 11.5# 121 ± 14.4#

All values represents mean ± SE and indicate percentage of values obtained before the addition of either angiotensin II (Ang II) or endothelin-1

(ET-1). N = 6 for all groups except Ang II or ET-1 alone where N = 10. ANP values represent changes in the ANP to GAPDH housekeeping gene

ratio

CSA cell surface area, ANP atrial natriuretic peptide, C3 RhoA inhibitor Clostridium botulinum C3 exoenzyme, Y the ROCK inhibitor Y-27632,

Lat B the actin-depolymerizing agent latrunculin B, CPA the adenosine A1 receptor agonist N6-cyclopentyladenosine, CGS the A2a receptor agonist

2-p-(2-carboxyethyl)-phenethylamino-50-N-ethylcarboxamidoadenosine (CGS21680), IB A3 receptor agonist N6-(3-iodobenzyl)adenosine-50-methyluronamide (IB-MECA)

* P \ 0.05 from either pre-Ang II or pre-ET-1 values; # P \ 0.05 from values obtained with either Ang II or ET-1 alone

Mol Cell Biochem (2014) 385:239–248 245

123

the cardioprotective effects of adenosine in that various

adenosine receptors have been shown to produce similar

salutary effects on the ischemic and reperfused myocar-

dium (reviewed in [20]).

Despite the increasing evidence that adenosine could

function as an endogenous antihypertrophic factor little is

known about the mechanisms underlying these effects. Yet,

targeting endogenous inhibitors of cardiac hypertrophy

represent an important component in the development of

new treatment strategies aimed at mitigating the hyper-

trophic process [22]. We hypothesized that the virtually

identical antihypertrophic effects of multiple adenosine

receptor activation suggest a common target for all three

adenosine receptor subtypes studied. Although the process

of cardiomyocyte hypertrophy is complex and mediated by

numerous cell signaling pathway (reviewed in [23]), RhoA,

a member of a family of small molecular weight GTPases

important in regulation of actin dynamics, has been

implicated in a variety of cardiac pathologies including

hypertrophy and heart failure and inhibition of RhoA

activity has been proposed as a potential therapeutic

approach for the treatment of cardiac disorders [10, 11].

Indeed, it has been shown that inhibition of the RhoA–

ROCK pathway reduces the hypertrophic response and the

severity of heart failure in experimental models such as

hypertension-induced heart failure [12]. In addition, reports

from a number of laboratories have shown that pharma-

cological inhibition of ROCK attenuates cardiac hypertro-

phy in mice subjected to myocardial infarction [13] as well

as rats subjected to aortic constriction [15] and reduces

diastolic dysfunction in both hypertensive [14] and aortic-

constricted rats [15]. Furthermore, ROCK deletion has

been shown to reduce heart failure progression and mor-

tality in mouse hypertrophic hearts produced by Gaq

overexpression thus implicating a role for ROCK in cardiac

decompensation [16]. Inhibition of the RhoA pathway also

likely represents the mechanism for the beneficial effects of

ginseng against leptin-induced hypertrophy [24]. Based on

this evidence, in the present study we used two hypertro-

phic stimuli to determine whether the RhoA pathway could

be a target for the antihypertrophic actions to adenosine

receptor activation. To the best of our knowledge, whether

adenosine affects RhoA activation and whether this could

be related to the antihypertrophic effect of the nucleoside

has not been previously investigated.

The ability of both ET-1 and Ang II to stimulate the

RhoA pathway has previously been demonstrated by var-

ious investigators particularly in vascular tissue where

activation of RhoA has been linked to vasopressor effects

of both peptides [25–28]. The present study provides strong

evidence that RhoA activation contributes to the hyper-

trophic response to both ET-1 and Ang II, and supports our

previous finding for a RhoA-dependent hypertrophy

induced by diverse factors including ET-1 [29] and the

16 kDa adipokine leptin [18, 19]. In the present study,

increased RhoA activity was found to be transient with

peak activation for both agonists occurring 15–30 min after

addition and returning to control thereafter.

The mechanisms which contribute to the pro-hypertro-

phic influence of RhoA activation are not well understood

although this likely reflects activation of a myriad of

transcriptional factors downstream of the RhoA pathway.

One of the downstream targets for RhoA is cofilin, an

actin-binding protein which causes depolymerisation of

actin [30]. The early increase in RhoA activity resulted in

significantly increased cofilin phosphorylation following

addition of either Ang or ET-1. Since phosphorylation of

cofilin inhibits its activity, the resultant effect was a

marked and significant decrease in the G–F actin ratio. We

have previously proposed that this p-cofilin-dependent

increased actin polymerization contributes to the hyper-

trophic effect of leptin in cultured ventricular myocytes

through a mechanism associated with a selective translo-

cation of p38 MAPK into nuclei [18, 19]. The present

report demonstrates a similar ability of both Ang II and ET-

1 to increase both p38 phosphorylation and nuclear trans-

location. The latter response is likely of critical importance

as MAPKs must translocate into nuclei to interact with

transcriptional factors and produce cellular effects [31, 32].

We believe that selective p38 translocation (no transloca-

tion was seen for ERK1/2) represents a key mechanism

which accounts for RhoA-dependent hypertrophic respon-

ses to both Ang II and ET-1. Indeed, Rho GTPases have

been shown to regulate a wide variety of intracellular

processes including activities of MAPKs as well as playing

a key role in the regulation of cellular trafficking between

cytoplasmic and nuclear compartments (reviewed in [33]).

To further reinforce the contribution of the RhoA

pathway to the pro-hypertrophic effects of both Ang II and

ET-1 as well as a potential target for the antihypertrophic

effects of adenosine receptor agonists we determined

whether the latter share similar antihypertrophic effects

with inhibitors of the RhoA pathway in myocytes treated

with either Ang II or ET-1. Indeed, the antihypertrophic

effects of adenosine receptor agonists were completely

mimicked by inhibitors of the RhoA pathway including the

RhoA inhibitors C3, the ROCK inhibitor Y-27632, as well

as latrunculin B. Although when taken together the results

implicate RhoA in hypertrophic response to both Ang II

and ET-1, the results with latrunculin B are of particular

interest as this agent produces a breakdown of F actin

thereby normalizing the G–F actin ratio. Thus, it is rea-

sonable to assume that the pro-hypertrophic effect of both

Ang II and ET-1 is critically dependent on actin poly-

merization. How actin dynamics could affect the hyper-

trophic process is not known although it has been shown

246 Mol Cell Biochem (2014) 385:239–248

123

that RhoA/ROCK-dependent cytoskeletal changes result in

activation of various transcriptional factors, likely through

mechano-dependent process [34, 35]. Moreover, there is

evidence that the state of actin dynamics is important in the

regulation of nucleocytoplasmic trafficking of a large

number of signaling molecules (reviewed in [36]).

Although, this phenomenon has not been applied to

exploring its potential relevance to the hypertrophic pro-

gram, the ability of latrunculin A to prevent p38 MAPK

import into nuclei implicates actin remodeling in this

process, downstream of RhoA activation.

Taken together, based on our results we propose the

following model for activation of the RhoA pathway by

Ang II and ET-1 and its inhibition by adenosine receptor

activation (Fig. 7). Stimulation of RhoA by these hyper-

trophic factors, acting via their specific receptors, is an

important component for the manifestation of the hyper-

trophic response. This is dependent on ROCK and LIM

kinase activation which inhibits cofilin thus increasing

actin polymerization (a decrease in the G–F actin ratio)

which is critical for activation of p38 MAPK and its

translocation into nuclei. The inhibition of RhoA activation

by adenosine receptor agonists appears to occur upstream

in the RhoA pathway cascade since receptor activation was

associated with inhibition of RhoA activity. A potential

candidate for this site is the family of guanine nucleotide

exchange factors (GEFs) which catalyze the exchange of

GDP–GTP resulting in activation of RhoA [37] and it is

therefore possible that adenosine receptor activation results

in GEF inhibition. This concept needs to be explored in

depth in future studies.

To the best of our knowledge, this is the first report

showing that adenosine receptor activation can inhibit

stimulation of RhoA activity by pro-hypertrophic factors

and, accordingly, the data suggest that this could represent

an important mechanism underlying the antihypertrophic

effect of adenosine. Although, a link between adenosine

and the RhoA pathway has not been previously established,

it has been suggested that protection of chick cardiomyo-

cytes exposed to ischemia mimetic conditions by adenosine

A3 receptor activation is mediated by the RhoA pathway

[38] and more recently RhoA has been implicated as a

cardioprotective factor in the ischemic and reperfused

mouse heart [39]. These findings underlie the concept that

that the RhoA pathway is complex and can mediate mul-

tifaceted effects, both beneficial and deleterious, under

specific experimental conditions [40]. Our findings coupled

with emerging evidence for RhoA/ROCK inhibition as an

effective strategy to inhibit heart failure provide evidence

that the antihypertrophic effect of endogenous adenosine

may occur by blunting RhoA/ROCK activation thus

keeping the hypertrophic program in check.

Acknowledgments This work was supported by a Grant from the

Heart and Stroke Foundation of Ontario. Morris Karmazyn holds a

Tier 1 Canada Research Chair in Experimental Cardiology.

References

1. Martens D, Lohse MJ, Rauch B, Schwabe U (1987) Pharmaco-

logical characterization of A1 adenosine receptors in isolated rat

ventricular myocytes. Naunyn Schmiedebergs Arch Pharmacol

336:342–348

2. Romano FD, MacDonald SG, Dobson JG (1989) Adenosine

receptor coupling to adenylate cyclase of rat ventricular myocyte

membranes. Am J Physiol 257:H1088–H1095

3. Xu D, Kong HY, Liang BT (1992) Expression and pharmaco-

logical characterization of a stimulatory subtype of adenosine

receptor in fetal chick ventricular myocytes. Circ Res 70:56–65

4. Tracey WR, Magee W, Masamune H, Oleynek JJ, Hill RJ (1998)

Selective activation of adenosine A3 receptors with N6-[3-chlo-

robenzyl]-50-N-methylcarboxamidoadenosine (CB-MECA) pro-

vides cardioprotection via KATP channel activation. Cardiovasc

Res 40:138–145

5. Funaya H, Kitakaze M, Node K, Minamino T, Komamura K,

Hori M (1997) Plasma adenosine levels increase in patients with

chronic heart failure. Circulation 95:1363–1365

6. Kitakaze M, Minamino T, Node K, Koretsune Y, Komamura K,

Funaya H, Kuzuya T, Hori M (1998) Elevation of plasma aden-

osine levels may attenuate the severity of chronic heart failure.

Cardiovasc Drugs Ther 12:307–309

7. Gan XT, Rajapurohitam V, Haist JV, Chidiac P, Cook MA,

Karmazyn M (2005) Inhibition of phenylephrine-induced car-

diomyocyte hypertrophy by activation of multiple adenosine

receptor subtypes. J Pharmacol Exp Ther 312:27–34

8. Chung ES, Perlini S, Aurigemma GP, Fenton RA, Dobson JG,

Meyer TE (1998) Effects of chronic adenosine uptake blockade

RhoA RhoA

GEFs

ROCK LIMK

G/F Actinp38

Activation of Transcription

Hypertrophy

ETA AT1 AR

ET-1 Ang II Adenosine

GTP GDP

+ +

Nucleus

?GTP

Fig. 7 Simplified scheme showing the potential site for RhoA

activation by Ang II or ET-1 and its attenuation by adenosine acting

through its receptors (AR) by modulation of one or more of the

guanine nucleotide exchange factors (GEFs). Details in ‘‘Discussion’’

section. ETA endothelin-1 ETA receptor, AT1 Ang II AT1 receptor,

LIMK LIM kinase

Mol Cell Biochem (2014) 385:239–248 247

123

on adrenergic responsiveness and left ventricular chamber func-

tion in pressure overload hypertrophy in the rat. J Hypertens

16:1813–1822

9. Xu X, Fassett J, Hu X, Zhu G, Lu Z, Li Y, Schnermann J, Bache

RJ, Chen Y (2008) Ecto-50-nucleotidase deficiency exacerbates

pressure-overload-induced left ventricular hypertrophy and dys-

function. Hypertension 51:1557–1564

10. Brown JH, Del Re DP, Sussman MA (2006) The Rac and Rho

hall of fame: a decade of hypertrophic signaling hits. Circ Res

98:730–742

11. Loirand G, Guerin P, Pacaud P (2006) Rho kinases in cardio-

vascular physiology and pathophysiology. Circ Res 98:322–334

12. Satoh S, Ueda Y, Koyanagi M, Kadokami T, Sugano M, Yos-

hikawa Y, Makino N (2003) Chronic inhibition of Rho kinase

blunts the process of left ventricular hypertrophy leading to

cardiac contractile dysfunction in hypertension-induced heart

failure. J Mol Cell Cardiol 35:59–70

13. Hattori T, Shimokawa H, Higashi M, Hiroki J, Mukai Y, Tsutsui

H, Kaibuchi K, Takeshita A (2004) Long-term inhibition of Rho-

kinase suppresses left ventricular remodeling after myocardial

infarction in mice. Circulation 109:2234–2239

14. Fukui S, Fukumoto Y, Suzuki J, Saji K, Nawata J, Tawara S,

Shinozaki T, Kagaya Y, Shimokawa H (2008) Long-term inhi-

bition of Rho-kinase ameliorates diastolic heart failure in

hypertensive rats. J Cardiovasc Pharmacol 51:317–326

15. Phrommintikul A, Tran L, Kompa A, Wang B, Adrahtas A,

Cantwell D, Kelly DJ, Krum H (2008) Effects of a Rho kinase

inhibitor on pressure overload induced cardiac hypertrophy and

associated diastolic dysfunction. Am J Physiol Heart Circ Physiol

294:H1804–H1814

16. Shi J, Zhang YW, Summers LJ, Dorn GW II, Wei L (2008)

Disruption of ROCK1 gene attenuates cardiac dilation and

improves contractile function in pathological cardiac hypertro-

phy. J Mol Cell Cardiol 44:551–560

17. Shi J, Zhang YW, Yang Y, Zhang L, Wei L (2010) ROCK1 plays

an essential role in the transition from cardiac hypertrophy to

failure in mice. J Mol Cell Cardiol 49:819–828

18. Zeidan A, Javadov S, Chakrabarti S, Karmazyn M (2008) Leptin-

induced cardiomyocyte hypertrophy involves selective caveolae

and RhoA/ROCK-dependent p38 MAPK translocation to nuclei.

Cardiovasc Res 77:64–72

19. Zeidan A, Javadov S, Karmazyn M (2006) Essential role of Rho/

ROCK-dependent processes and actin dynamics in mediating

leptin-induced hypertrophy in rat neonatal ventricular myocytes.

Cardiovasc Res 72:101–111

20. Headrick JP, Peart JN, Reichelt ME, Haseler LJ (2011) Adeno-

sine and its receptors in the heart: regulation, retaliation and

adaptation. Biochim Biophys Acta 1808:1413–1428

21. Pang T, Rajapurohitam V, Cook MA, Karmazyn M (2010) Dif-

ferential AMPK phosphorylation sites associated with phenyl-

ephrine vs. antihypertrophic effects of adenosine agonists in

neonatal rat ventricular myocytes. Am J Physiol Heart Circ

Physiol 298:H1382–H1390

22. Hardt SE, Sadoshima J (2010) Negative regulators of cardiac

hypertrophy. Cardiovasc Res 63:500–509

23. Frey N, Olson EN (2003) Cardiac hypertrophy: the good, the bad,

and the ugly. Annu Rev Physiol 65:45–79

24. Moey M, Rajapurohitam V, Zeidan A, Karmazyn M (2012)

Ginseng (Panax quinquefolius) attenuates leptin-induced cardiac

hypertrophy through inhibition of p115Rho guanine nucleotide

exchange factor-RhoA/Rho-associated, coiled-coil containing

protein kinase-dependent mitogen-activated protein kinase path-

way activation. J Pharmacol Exp Ther 339:746–756

25. Lan C, Das D, Wloskowicz A, Vollrath B (2004) Endothelin-1

modulates hemoglobin-mediated signaling in cerebrovascular

smooth muscle via RhoA/Rho kinase and protein kinase C. Am J

Physiol Heart Circ Physiol 286:H165–H173

26. Bregeon J, Loirand G, Pacaud P, Rolli-Derkinderen M (2009)

Angiotensin II induces RhoA activation through SHP2-dependent

dephosphorylation of the RhoGAP p190A in vascular smooth

muscle cells. Am J Physiol Cell Physiol 297:C1062–C1070

27. Allahdadi KJ, Hannan JL, Tostes RC, Webb RC (2010) Endo-

thelin-1 induces contraction of female rat internal pudendal and

clitoral arteries through ETA receptor and rho-kinase activation.

J Sex Med 7:2096–2103

28. Shatanawi A, Romero MJ, Iddings JA, Chandra S, Umapathy NS,

Verin AD, Caldwell RB, Caldwell RW (2011) Angiotensin II-

induced vascular endothelial dysfunction through RhoA/Rho

kinase/p38 mitogen-activated protein kinase/arginase pathway.

Am J Physiol Cell Physiol 300:C1181–C1192

29. Hunter JC, Zeidan A, Javadov S, Kilic A, Rajapurohitam V,

Karmazyn M (2009) Nitric oxide inhibits endothelin-1-induced

neonatal cardiomyocyte hypertrophy via a RhoA-ROCK-depen-

dent pathway. J Mol Cell Cardiol 47:810–818

30. Bernstein BW, Bamburg JR (2010) ADF/cofilin: a functional

node in cell biology. Trends Cell Biol 20:187–195

31. Brunet A, Roux D, Lenormand P, Dowd S, Keyse S, Pouyssegur J

(1999) Nuclear translocation of p42/p44 mitogen-activated pro-

tein kinase is required for growth factor-induced gene expression

and cell cycle entry. EMBO J 18:664–674

32. Plotnikov A, Zehorai E, Procaccia S, Seger R (2011) The MAPK

cascades: signaling components, nuclear roles and mechanisms of

nuclear translocation. Biochim Biophys Acta 1813:1619–1633

33. Hall A (2005) Rho GTPases and the control of cell behaviour.

Biochem Soc Trans 33(Pt 5):891–895

34. Sotiropoulos A, Gineitis D, Copeland J, Treisman R (1999)

Signal-regulated activation of serum response factor is mediated

by changes in actin dynamics. Cell 98:159–169

35. Chiquet M, Tunc-Civelek V, Sarasa-Renedo A (2007) Gene

regulation by mechanotransduction in fibroblasts. Appl Physiol

Nutr Metab 32:967–973

36. Aplin AE, Juliano RL (2001) Regulation of nucleocytoplasmic

trafficking by cell adhesion receptors and the cytoskeleton. J Cell

Biol 155:187–191

37. Mizuno-Yamasaki E, Rivera-Molina F, Novick P (2012) GTPase

networks in membrane traffic. Annu Rev Biochem 81:637–659

38. Mozzicato S, Joshi BV, Jacobson KA, Liang BT (2004) Role of

direct RhoA-phospholipase D1 interaction in mediating adeno-

sine-induced protection from cardiac ischemia. FASEB J

18:406–408

39. Xiang SY, Vanhoutte D, Del Re DP, Purcell NH, Ling H,

Banerjee I, Bossuyt J, Lang RA, Zheng Y, Matkovich SJ, Mi-

yamoto S, Molkentin JD, Dorn GW II, Brown JH (2011) RhoA

protects the mouse heart against ischemia/reperfusion injury.

J Clin Investig 21:3269–3276

40. Miyamoto S, Del Re DP, Xiang SY, Zhao X, Florholmen G,

Brown JH (2010) Revisited and revised: is RhoA always a villain

in cardiac pathophysiology? J Cardiovasc Transl Res 3:330–343

248 Mol Cell Biochem (2014) 385:239–248

123