| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* The Herman B. Wells Center for Pediatric Research, Department of Pediatrics, Indiana University, School of Medicine, Indianapolis, Indiana, USA;
Section of Cardiovascular Sciences, Department of Medicine,
Center for Cardiovascular Development, Baylor College of Medicine, Houston, Texas, USA;
The Institute of Biosciences and Technology, The Texas A&M University System Health Science Center, Houston, Texas, USA
1Correspondence: The Herman B. Wells Center for Pediatric Research, Department of Pediatrics, Indiana University, School of Medicine, R4 building, Rm. 370, 1044 West Walnut St., Indianapolis, IN 46202-5225, USA. E-mail: lewei{at}iupui.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
q. Thus, these results indicate that ROCK1 contributes to the development of cardiac fibrosis and induction of fibrogenic cytokines in cardiomyocytes in response to pathological stimuli. Zhang, Y.-M., Bo, J., Taffet, G. E., Chang, J., Shi, J., Reddy, A. K., Michael, L. H., Schneider, M. D., Entman, M. L., Schwartz, R. J., Wei, L. Targeted deletion of ROCK1 protects the heart against pressure overload by inhibiting reactive fibrosis.
Key Words: Rho kinase • pressure overload • fibrosis • hypertrophy • fibrogenic cytokines
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. To maintain cardiac output against pressure overload, the heart activates an adaptive response in the form of cardiac hypertrophy, characterized by increases in myocyte size and protein content, and a change in gene expression profiles from adult to a "fetal-like" programs (2
, 3)
. Although this adaptation is initially a compensatory response, persistent stress eventually leads to the development of heart failure. Pressure overload also results in apoptosis and fibrosis, which could be part of the mechanism of cardiac decompensation. Identification of signaling pathways regulating hypertrophic growth, myocyte contractility, survival and fibrosis will provide useful information for therapy aimed at preventing or retarding the development of heart failure.
Recent studies suggest that RhoA GTPase is involved in the pathogenesis of cardiac hypertrophy (4
, 5)
. Rho kinase is a downstream mediator of RhoA, and is believed to play a critical role in mediating the effects of RhoA on stress fiber formation, smooth muscle contraction, cell adhesion, membrane ruffling and cell motility (6
, 7)
. Recent studies using Rho kinase inhibitors, Y27632 and fasudil, suggest that Rho kinase is involved in the pathogenesis of cardiac hypertrophy and remodeling. Administration of Y27632 to Dahl salt-sensitive hypertensive rats, leads to the regression of cardiac hypertrophy and decreased pathological remodeling (8
, 9)
. Administration of fasudil suppresses angiotensin (ANG) II-induced cardiomyocyte hypertrophy in rats (10)
, and left ventricular remodeling after myocardial infraction in mice (11)
. Whether Rho kinase is implicated in the development of cardiac hypertrophy induced by transverse aortic constriction, a well-established surgical model for acute and chronic pressure overload, has not been investigated. To maintain cardiac output against acute pressure overload, the heart activates an adaptive physiological response in the form of concentric cardiac hypertrophy (12
, 13)
. Although this adaptation is initially a compensatory response, persistent stress eventually leads to the development of cardiac dilation and heart failure (14)
.
The previous studies relied heavily on pharmacological inhibitors, and not on gene deletion. The chemical inhibitors of Rho kinase do not distinguish between ROCK1 (p160ROCK) and ROCK2 (ROK), the two isoforms of Rho kinase family, and could also have nonselective effects (15
16
17)
. Loss-of-function studies for Rho kinase through gene targeting approach are necessary to provide more direct and conclusive evidence for the role of Rho kinase in the pathological remodeling of myocardium, and also to dissect the individual functions of Rho kinase isoforms.
To investigate biological functions of ROCK1 and its role in the development of cardiac hypertrophy and ventricular remodeling, we generated ROCK1 knockout (ROCK1–/–) mice. The homozygous ROCK1 knockout mice are morphologically indistinguishable from wild-type (WT) littermates. These mice were subjected to pressure overload by transverse aortic constriction. We found that ROCK1 homozygous deficient mice developed cardiac myocyte hypertrophy in response to pressure overload. These mice exhibited significantly reduced interstitial fibrosis, which was associated with a reduction in pathophysiological synthesis of collagens and fibrogenic cytokines, compared to WT mice. This inhibitory effect of ROCK1 deletion on pathophysiological induction of fibrogenic cytokines was further confirmed in the myocardium of transgenic mice with cardiomyocyte-specific overexpression of -subunit of Gq, which recapitulates many cellular, molecular and functional characteristics of pressure-overload induced hypertrophy (18
, 19)
.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Generation of ROCK1 knockout mice
ROCK1+/– embryonic stem (ES) cells were generated by homologous recombination and used to create mice by using the standard methods. A murine strain 129 genomic clone (10 kb) containing exons 3–6 was isolated from a 129 mouse genomic library. A targeting vector was designed to insert the coding sequence of ß-galactosidase in frame downstream of residue 180 (in exon 5) of ROCK1 followed by PGK-Neo cassette (Fig. 1
A). The targeting vector was linearized with NotI and introduced into AB2.2 ES cells by electroporation, then placed under G418 selection. Nine independent ROCK1+/– ES cell clones were used for injection of C57BL/6 blastocysts to generate chimeric mice, which were then mated with C57BL/6 mice to produce heterozygous ROCK1+/– mice. Two ROCK1+/– ES cell clones have shown transmission through the germ line to establish heterozygous ROCK1+/– mice (Fig. 1B, C
). ROCK1+/– heterozygous mice were then intercrossed to produce homozygous ROCK1–/– mice (Fig. 1C
). The genotypes of the offspring were identified by Southern blot analysis and polymerase chain reaction (PCR) on DNA obtained from embryos or tails of adult mice as described previously (20)
. Mice used in the present study had been backcrossed to FVB for at least six generations.
|
Mouse model of cardiac hypertrophy
Pressure-overload cardiac hypertrophy was induced in 10- to 12-wk-old adult WT and ROCK1–/– mice by transverse aortic banding as described previously (21)
. Mice were anesthetized by inhalation of isoflurane (1% in 100% oxygen). The degree of pressure-overload was measured by right-to-left carotid artery flow velocity ratio after constricting the transverse aorta. Only the mice with a ratio from 5:1 to 10:1 were used. As the control, a "sham" operation without occlusion was performed on age-matched mice. Cardiac dimension and contractile performance were evaluated by noninvasive M-mode echocardiography (22)
. Surface electrocardiography (ECG) was performed using limb leads and a dedicated ECG unit.
Histology and quantitative analysis
Assessment of cardiac hypertrophy was performed as described previously (23)
. Total heart wt was indexed to tibial length. For histology, hearts were perfused with PBS followed by 10% buffered formalin and embedded in paraffin. Tissue sections were then stained with hematoxylin/eosin for initial evaluation, or picrosirius red to identify collagen fibers. Myocyte diameter was measured using transnuclear width at the midventricular concentration. The picrosirius red-stained slides were scanned by using a microscope equipped with a digital camera (Zeiss, Thornwood, NY, USA), and quantitative evaluation was performed with Zeiss IMAGE analysis software. The collagen-stained area was calculated as a percentage of the total myocardial area.
Generation of MHC-Gq/ROCK1–/– compound mice
Transgenic FVB mice overexpressing Gq in cardiomyocytes (MHC-Gq) were kindly provided by Dr. Gerald Dorn, University of Cincinnati (18
, 19)
. These mice express Gq at levels 
5-fold higher than the endogenous protein (18
, 19)
. These transgenic mice were mated with ROCK1–/– (FVB background) to generate MHC-Gq/ROCK1+/– mice, which were then bred with ROCK+/– mice to generate MHC-Gq/ROCK1–/– mice. Southern blot analysis was carried out for ROCK1 as described above and PCR analysis for MHC-Gq transgene.
Protein analysis
Protein samples were prepared as described previously (20)
. Protein expression of ROCK1 was examined using a rabbit anti-ROCK1 polyclonal antibody (Santa Cruz; directed against human ROCK1 residues 755–840, 1:200), and also by a mouse anti-ROCK1 monoclonal antibody (mAb) (BD Transduction Laboratories, San Diego, CA, USA; directed against mouse ROCK1 residues 906-1012, 1:1000). Both antibodies detect ROCK1 expression in WT mouse hearts but not in ROCK1–/– mouse hearts. Protein expression of ROCK2 was examined using a mouse anti-ROCK2 mAb (BD Transduction laboratories, 1:1000). The extent of phosphorylation of the ERM (ezrin, radixin, and moesin) family, a substrate of Rho kinase, was measured by Western blot analysis using an antiphospho-ERM antibody (Ab) (Cell Signaling, Beverly, MA, USA; 1:1000). The amount of the ERM phosphorylation was normalized to that of total ERM, which was measured by Western blot analysis using an anti-ERM Ab (Cell Signaling, 1:1000).
Gene expression analysis
Total RNA was extracted from ventricular tissues by using TRIZOL (Gibco BRL, Gaithersburg, MD, USA) and RNAeasy (Qiagen, Chatsworth, CA, USA). For expression profiling, samples were labeled with biotinylated nucleotides by reverse transcription, hybridized to Affymetrix mouse genomic 430 2.0 arrays, and stained with streptavidin-phycoerythrin. Fluorescence intensities were captured with an Affymetrix GeneArray 3000 Scanner, quantified with Affimetrix Microarray Suite 5.0, and analyzed using dChip 1.3 (Harvard University, Cambridge, MA, USA). To confirm the results of microarray hybrizidation, mRNA transcript levels of selected genes, were assessed by real-time RT-PCR (ABI Prism 7700, Perkin Elmer, Norwalk, CT, USA). To assess mRNA transcript levels by real-time RT-PCR, TaqMan primers and probes for mouse GAPDH, ROCK1, ROCK2, MHC, ßMHC, phospholamban skeletal -actin, atrial natriuretic peptide (ANP), sarcoplasmic reticulum Ca2+-ATPase (SERCA2a), smooth muscle -actin, collagen I 1 (COL11), collagen III 1 (COL31), transforming growth factor-ß1 (TGFß1), TGFß2, connective tissue growth factor (CTGF) were purchased from Applied Biosystems (Foster City, CA, USA).
Statistical analysis
Data are reported as mean ± SE. Comparisons between groups were analyzed by Student’s t test or ANOVA as appropriate, with P < 0.05 considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). As the entire kinase domain of ROCK1 is contained within residues 76–338, this targeting approach results in the deletion of 158 amino acids from the kinase domain and all amino acids from the coiled-coil and the PH domains (see Fig. 1A
). Disruption of the ROCK1 gene was confirmed by Western blot analysis with multiple antibodies (Fig. 1C
). In contrast, the abundance of ROCK2 protein concentration did not differ between ROCK1 mutants and WT mice (Fig. 1C
), indicating that the expression of ROCK2 did not increase to compensate for the loss of ROCK1. The ROCK1+/– and ROCK1–/– mice on FVB background were used in the present study. Heterozygous ROCK1+/– mice were found to be viable and fertile with no detected anatomical abnormalities. Intercrossing of heterozygous mice generated homozygous ROCK1–/– mice, which were also viable with no detected anatomical abnormalities. No increase in morbidity or mortality of ROCK1–/– mice was observed up to 24 months of age.
Despite the normal phenotype of ROCK1–/– mice, analysis of the genotype distribution in offspring from heterozygous mating revealed that the homozygous ROCK1 knockout mice were significantly underrepresented (Table 1
) from the value predicted by Mendelian inheritance law. In addition, we obtained the same low frequency for ROCK1–/– embryos ranging from E9.5 to E15.5 from heterozygous mating and no cardiovascular defect was detected with the knockout genotype. The low number of ROCK1–/– mice from heterozygous breeding is likely due to early lethality of 50–60% ROCK1–/– embryos before implantation or early postimplantation. The heterozygous ROCK1 knockout mice were also underrepresented from breeding between heterozygotes or the breeding of heterozygotes with WT mice (Table 1)
, suggesting there is 30% lethality in ROCK1+/– embryos. We were not able to detect any developmental or gross anatomic malformation in ROCK1–/– and ROCK1+/– embryos or adult mice except for the under-represented values for these mice derived from heterozygous breeding. These results suggest that ROCK1 acts on an early stage of embryonic development prior to organogenesis (before E9.5).
|
ROCK1 expression and Rho kinase activity were increased in hypertrophic hearts induced by pressure overload
To determine if ROCK1 is required for the development of cardiac hypertrophy, we subjected WT and ROCK1–/– mice to transverse aortic banding. Both ROCK1 and ROCK2 are present in WT mouse myocardium as revealed by real-time RT-PCR (Fig. 2
A) and Western blot analyses (Fig. 2B
). After 1 wk of banding (early adaptive phase), the mRNA and protein levels of ROCK1 were not significantly increased in WT hypertrophic hearts compared with the sham-operated group. ROCK1 expression was significantly increased at 3 wk after banding (the state of stable hypertrophy) (Fig. 2B
). On the other hand, phosphorylation of ERM proteins (known substrates of Rho kinase), which was not significantly increased in WT hypertrophic hearts at 1 wk after banding, was also increased at 3 wk after banding (Fig. 2B
). In addition, this increased phosphorylation of ERM proteins was not observed in ROCK1–/– hypertrophic hearts (Fig. 2B
). Both mRNA and protein levels of ROCK2 remained unchanged in WT and ROCK1–/– hearts after 3 wk of aortic banding (Fig. 2B
). These observations indicate that pressure overload induced a delayed increase in ROCK1 expression and Rho kinase activity.
|
Disruption of ROCK1 did not affect the development of cardiac hypertrophy
Concentric cardiac hypertrophy was induced in the WT and ROCK1–/– hearts by transverse aortic banding (Fig. 3
A). Both ROCK1–/– and WT mice received a comparable load, based on the right-to-left carotid artery flow velocity ratio (from 5:1 to 10:1) after constricting the transverse aorta (Table 2
). The numbers of mice that did not survive of aortic banding were very small for WT (4 out of 29) and ROCK1–/– (4 out of 32) mice, and were similar to those of sham-operated mice. There was no difference between WT and ROCK1–/– mice in the heart wt/tibial length under sham-operated condition (Fig. 3B
). The time course and extent of increase in the heart wt/tibial length in response to pressure overload were similar between the WT and ROCK1–/– mice (Fig. 3B
). Cardiac myocyte diameter, another index of hypertrophy, increased similarly in WT and ROCK1–/– mice (Fig. 3C
). These findings suggest that hypertrophic responses were not impaired in ROCK1–/– mice. In addition, the levels of ANP, BNP, -skeletal actin, and ßMHC mRNAs had significantly increased in response to pressure overload to a similar extent (except for -skeletal actin) in WT and ROCK1–/–hearts after 1 wk of banding (Fig. 3D
). However, induction of BNP, -skeletal actin, and ßMHC mRNA levels was reduced in the ROCK1–/– hearts compared with the WT hearts after 3 wk of banding, when the state of stable hypertrophy had been reached (Fig. 3D
).
|
|
Cardiac function was preserved in ROCK1–/–and WT hypertrophic hearts induced by pressure overload
There was no significant difference in echocardiographic parameters including left ventricular end-diastolic diameter (LVEDD), left ventricular end-systolic diameter (LVESD) and left ventricular percent fractional shortening (%FS) between the WT and ROCK1–/– mice under baseline and sham conditions (Table 2)
. The only difference observed between the WT and ROCK1–/– mice is the heart rate, which was significantly increased (20%) in ROCK1–/– hearts under these conditions. The reason for this increased heart rate remains to be determined as no sinus and AV nodal dysfunction and cardiac conduction defect were detected in ROCK1–/– hearts by electrocardiography and intracardiac electrophysiological analyses (data not shown). As cardiac-specific over-expression of Rho GDI, an endogenous inhibitor of Rho family proteins, led to progressive atrioventricular conduction defects (23)
, the presence of ROCK2 may compensate for the absence of ROCK1 in regulating cardiac conduction.
Under banding condition, the degree (measured by the right-to-left carotid artery flow velocity ratio) and the duration of banding (up to 3 wk) did not induce cardiac failure in WT and ROCK1–/– mice as no cardiac dilation occurred in these mice (Table 2)
. The echocardiographic parameters (LVEDD, LVESD and fibrous FS) remained similar between WT and ROCK1–/– mice at 1 and 3 wk after banding, indicating that cardiac dimension and contractile function were preserved in WT and ROCK1–/– mice.
Interstitial fibrosis was reduced in ROCK1–/– hearts
Cardiac hypertrophy induced by pressure overload is associated with increased fibrosis in the myocardium, which is characterized by over-production of ECM proteins, predominantly collagen types I and III, into the interstitial and perivascular space (24)
. Picrosirius red staining, which identifies collagen fibers, and real-time RT-PCR analysis of the mRNA levels of COL1 and COL31 revealed that cardiac fibrosis significantly increased in WT and ROCK1–/– mice after aortic banding (Fig. 4
). Compared with WT mice, this increased fibrosis was significantly less in ROCK1–/– mice at 3 wk after banding (Fig. 4A, B
). However, after 1 wk of banding, the mRNA levels of COL11 and COL31 were similarly increased in ROCK1–/– and WT mice (Fig. 4C
) and histological staining did not detect significant difference between ROCK1–/– and WT mice at this early time point (data not shown). These observations indicate a delayed antifibrotic effect of ROCK1 deficiency to pressure overload.
|
Rho and Rho kinase have been shown to be major regulators of cytoskeletal rearrangement and cell cycle progression in a variety of cells including fibroblasts (25
26
27)
. One possible explanation for the reduced fibrosis in ROCK1–/– mice could be a reduced number of cardiac myofibroblasts, which are responsible for myocardial collagen metabolism. However, immunostaining analysis with antismooth -actin Ab revealed similar myofibroblast density and quantitative RT-PCR analysis revealed similar smooth -actin mRNA levels in ROCK1–/– and WT mice after 1 and 3 wk of aortic banding (data not shown).
Gene expression analysis of ROCK1–/– and WT hearts
To investigate the molecular mechanisms underlying the antifibrosis effects of ROCK1 ablation, we have performed DNA microarray analysis of ventricular myocardium from ROCK1–/– and WT mice after 3 wk of aortic banding. Real-time quantitative RT-PCR was then performed to confirm the findings with a number of ECM proteins and molecules involved in ECM protein metabolism (Table 3
). In addition to collagen types I and III, mRNA expression levels of a number of procollagens were depressed in ROCK1–/– hearts compared with WT hearts after 3 wk of banding (Table 3)
. A variety of other genes associated with ECM including laminins, fibrillins, fibulins were likewise depressed in ROCK1–/– hearts. The expression levels of TGFß2 and CTGF, which are potent fibrogenic cytokines (28)
, were significantly lower in ROCK1–/– mice than in WT mice 3 wk after aortic banding (Table 3
, Fig. 5
A). In addition, the mRNA levels of TGFß2 and CTGF were not significantly different between ROCK1–/– and WT mice after 1 wk of banding (Fig. 5A
), consistent with the finding on the mRNA levels of COL1 and COL3 at this time point. These observations suggest that ROCK1 deficiency reduces reactive fibrosis induced by pressure overload at least in part through modulating expression of these fibrogenic cytokines.
|
|
ROCK1 deficiency reduced Gq-induced fibrogenic cytokine expressions
Cardiomyocytes have been shown to produce fibrogenic cytokines including TGFß and CTGF under hypertrophic stimuli (29
, 30)
. To further investigate the cellular mechanisms underlying the antifibrotic effect of ROCK1 ablation, we examined the consequences of ROCK1 ablation on fibrogenic cytokines in the heart of transgenic mice with cardiomyocyte-specific overexpression of Gq (MHC-Gq) (18
, 19)
. Homozygous deletion of ROCK1 was introduced into MHC-Gq transgenic mice by cross-breeding these lines to generate MHC-Gq/ROCK1–/– mice. ROCK1 ablation did not affect expression of Gq transgene and increase in heart wt/tibial length induced by Gq (30–40% increase at age of 12 wk) (data not shown). The mRNA levels of TGFß2 and CTGF were determined by real-time RT-PCR in 12 wk-old male mice, and their levels were significantly up-regulated in MHC-Gq hearts compared to WT hearts (Fig. 5B
). Similar to the finding in hypertrophic hearts induced by pressure overload, the mRNA levels of these fibrogenic factors in MHC-Gq/ROCK1–/– hearts were significantly reduced compared to those in MHC-Gq hearts (Fig. 5B
), indicating ROCK1 ablation significantly reduced the Gq-induced up-regulation of fibrogenic cytokine production in cardiomyocytes.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
50–60%) and ROCK1+/– (30%) embryos died before implantation or early postimplantation. This partially penetrant and dosage-dependent embryonic lethality at very early stages suggests that there is a threshold effect for ROCK1 during a critical window of development before organogenesis. Further studies are under way to determine the precise stages of the early lethality of ROCK1–/– embryos and to investigate whether ROCK1 deficiency alters expression of cytokines known to be important during preimplantation or early postimplantation development (31)
.
The present study demonstrates potential therapeutic importance of ROCK1 for inhibiting maladaptive development of reactive fibrosis induced by pressure overload. One of the major morphological differences between the pathological (pressure overload-induced) vs. physiological (exercise-induced) compensatory hypertrophy is that only pathological hypertrophy is accompanied by reactive fibrosis (accumulation of collagen) (14
, 32
, 33)
. Both ROCK1–/– and WT mice subjected to 3 wk of aortic banding (compensatory phase) developed cardiac hypertrophy without signs of congestive heart failure, indicating that disruption of ROCK1 does not affect the development of compensatory hypertrophy. Strikingly, the maladaptive development of reactive fibrosis was reduced in ROCK1–/– hearts during the compensatory phase (at 3 wk after banding), accompanied by reduced pathological induction of profibrotic gene expression including collagens and fibrogenic cytokines, suggesting that ROCK1 signaling is involved in fibrosis at least in part through regulation of TGFß2 and CTGF expression.
TGFß is a well established profibrogenic cytokine implicated in cardiac fibrosis (28)
. CTGF has recently received much attention as a downstream mediator of TGFß and various hypertrophic stimuli during cardiac fibrosis, promoting the expression of ECM proteins (29
, 34
35
36)
. As TGFß2 and CTGF are secreted by cardiomyocytes, cardiac fibroblasts and inflammatory cells in response of pressure overload (36
, 37)
and systemic deletion of ROCK1 was employed in the present study, it remains to be determined which cell types are involved in the reduced production of TGFß2 and CTGF in ROCK1–/– hearts. As the reduction of mRNA levels of some cardiac hypertrophic markers such as BNP, ßMHC and skeletal -actin was associated with reduced mRNA levels of TGFß2 and CTGF in ROCK1–/– hearts, ROCK1 deficiency in cardiomyocytes may be in part responsible for the reduced mRNA levels of TGFß2 and CTGF. This hypothesis is supported by our finding that ROCK1 deficiency also reduced mRNA levels of TGFß2 and CTGF in the myocardium of transgenic mice with cardiomyocyte-specific overexpression of Gq, which recapitulates many cellular, molecular and functional characteristics of pressure-overload hypertrophy (18
, 19)
. Gq is the heterotrimeric protein that couples membrane receptors for neurohumoral factors including 1-adrenergic agonists, ANG II and endothelin-1 to the cardiac hypertrophic response. Previous studies have also demonstrated that Gq signaling is necessary for pressure-overload induced cardiac hypertrophy (38
, 39)
.
The antifibrotic effect of ROCK1 deficiency was not observed at 1 wk after aortic banding, but at 3 wk. One potential factor that could contribute to this delayed antifibrotic effect is the wall stress, which is believed to be significantly different between the early adaptive and compensatory phases because the development of concentric cardiac hypertrophy normalizes wall stress (12
, 14)
. Under pressure overload condition, an increase in collagen production occurs in response to three types of stimuli: 1) biomechanical stress that directly activates fibroblasts, 2) activation of neurohumoral systems, in particular, systemic and local production of ANG II, which plays a major role in modulating the synthetic activity of cardiac fibroblasts, and 3) inflammatory cells that infiltrate the overloaded cardiac tissue and release fibrogenic cytokines (40
, 41)
. Biomechanical stress to the heart could be significantly greater at 1 wk compared to 3 wk after banding and may activate ROCK1-independent profibrotic signal pathways. The delayed antifibrotic effect of ROCK1 deficiency (at 3 wk after banding) suggests that ROCK1 may mediate a delayed response initiated by release of hormones and/or cytokines induced by pressure overload. This hypothesis is supported by our finding that ROCK1 deficiency reduced induction of fibrogenic cytokines, TGFß2 and CTGF in the myocardium of transgenic mice with cardiomyocyte-specific overexpression of Gq, which activates pathways downstream of multiple neurohumoral hypertrophic signals (18
, 19)
. In addition, there was also a delayed increase in ROCK1 expression induced by pressure overload in WT hearts (not at 1 wk, but at 3 wk of banding), suggesting this delayed up-regulation of ROCK1 expression may also contribute to reactive fibrosis induced by pressure overload.
Our observation that homozygous deletion of ROCK1 did not reduce pressure overload-induced cardiac hypertrophy suggests that this isoform of Rho kinase may not play an essential role in regulating myocyte hypertrophy. This observation is clearly contrasted to previous studies performed with Rho kinases inhibitors Y27632 and fasudil. Inactivation of ROCK1 and ROCK2 by these inhibitors has been shown to inhibit both myocyte hypertrophy and fibrosis during the left ventricular remodeling process under several pathological conditions including myocardial infraction in mice (11)
, ANG II-induced cardiac hypertrophy in rats and mice (10
, 42)
, and hypertension-induced heart failure in Dahl salt-sensitive rats (8
, 9)
. Several potential explanations could account for the absence of an antihypertrophic effect with ROCK1 deficiency observed in the present study. First, ROCK1 may have a specific role in fibrosis while ROCK2 may be involved in hypertrophy. Second, there may be a threshold effect of the total Rho kinase activity in fibrosis and hypertrophy, with fibrosis being more sensitive to the reduction in Rho kinase activity. It is worth noting that our finding that ROCK1 deficiency selectively reduced cardiac fibrosis shares some similarities with a previous study in which inhibition of p38 MAPK prevented development of cardiac fibrosis but not hypertrophy in response to pressure overload (43)
.
Recently, another group also reported the targeted disruption of ROCK1 (44)
. Their mutant mice had apparently normal development, except failure of closure of eyelid and umbilical ring in varying degree at birth. None of these abnormalities was detected in our mutant mice. The similarity between their and our mutant mice is that both adult mutant mice were fertile and apparently healthy. The differences between their and our mutant mice may arise from several factors. First, in their targeting vector, exons 3 and 4 were replaced by ß-galactosidase resulting in disruption of ROCK1 at residue 59, while in our targeting vector, ROCK1 coding sequence was disrupted at residue 180 in exon 5. In both cases (disruption at residue 180 or at residue 59 of ROCK1), disruption of ROCK1 should result in completely abolished ROCK1 kinase activity and function. However, effect on the expression of other genes close to ROCK1 locus may be different. Second, different genetic backgrounds were used: C57BL/6 was used in their study, while in our study C57BL/6–129/SvJ mixed (when ROCK1–/– mice were generated) and FVB (for the subsequent characterizations) backgrounds were used. These factors have been shown to cause phenotypic differences within similar genetic knockouts (31
, 45)
. Further study with ROCK1 deletion in various strain backgrounds should be performed to evaluate the developmental role of ROCK1. One additional limitation of the present study should be mentioned. Although ROCK1–/– mice used in the present study are apparently normal to the WT mice, the possibility of the presence of subtle changes in the structure and/or function of the heart and other organs due to the absence of ROCK1 during embryonic development can not be excluded. Future study with inducible disruption of ROCK1 in adults is needed to confirm the antifibrotic effect of ROCK1 deficiency observed in the present study.
In summary, the present study provides a direct support for a role of ROCK1 in non adaptive fibrosis in response to pressure overload and in the regulation of fibrogenic cytokine expressions in cardiomyocytes in response to hypertrophic stimuli. This study demonstrated potential therapeutic importance of ROCK1 for cardiac fibrotic diseases.
Additional note:
While this manuscript was under review, one publication has addressed the role of ROCK1 in cardiac hypertrophy and remodeling using ROCK1+/– mice (46)
. That study supports many of the findings outlined here.
|
ACKNOWLEDGMENTS |
|---|
MHC-Gq line. This work was supported by a Scientist Development Grant from American Heart Association and by NIHR01-HL72897 (to L. W.); by NIH P01-HL49953 and NIH R01-HL64356 (to R. J. S.); by NIHR01-AG18599 (to G.E.T.) and P01-HL42550 (to M.L.E). Received for publication October 8, 2005. Accepted for publication December 20, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Ohtsu, S. Higuchi, H. Shirai, K. Eguchi, H. Suzuki, A. Hinoki, E. Brailoiu, A. D. Eckhart, G. D. Frank, and S. Eguchi Central Role of Gq in the Hypertrophic Signal Transduction of Angiotensin II in Vascular Smooth Muscle Cells Endocrinology, July 1, 2008; 149(7): 3569 - 3575. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Phrommintikul, L. Tran, A. Kompa, B. Wang, A. Adrahtas, D. Cantwell, D. J. Kelly, and H. Krum Effects of a Rho kinase inhibitor on pressure overload induced cardiac hypertrophy and associated diastolic dysfunction Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1804 - H1814. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Kass, H. C. Champion, and J. A. Beavo Phosphodiesterase Type 5: Expanding Roles in Cardiovascular Regulation Circ. Res., November 26, 2007; 101(11): 1084 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Goel, Y. Zhang, L. Anderson, and R. Rahimian Gender difference in rat aorta vasodilation after acute exposure to high glucose: Involvement of protein kinase C {beta} and superoxide but not of Rho Kinase Cardiovasc Res, November 1, 2007; 76(2): 351 - 360. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L.M. Peters and M. C. Michel The RhoA/Rho kinase pathway in the myocardium Cardiovasc Res, July 1, 2007; 75(1): 3 - 4. [Full Text] [PDF]
|
|
|
|
|
|
L. Hauck, C. Harms, D. Grothe, J. An, K. Gertz, G. Kronenberg, R. Dietz, M. Endres, and R. von Harsdorf Critical Role for FoxO3a-Dependent Regulation of p21CIP1/WAF1 in Response to Statin Signaling in Cardiac Myocytes Circ. Res., January 5, 2007; 100(1): 50 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Yoneda, D. Ushakov, H. A.B. Multhaupt, and J. R. Couchman Fibronectin Matrix Assembly Requires Distinct Contributions from Rho Kinases I and -II Mol. Biol. Cell, January 1, 2007; 18(1): 66 - 75. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
