Skip Navigation

European Journal of Heart Failure 2008 10(7):635-645; doi:10.1016/j.ejheart.2008.04.014
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Gary-Bobo, G.
Right arrow Articles by Mericskay, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gary-Bobo, G.
Right arrow Articles by Mericskay, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2008 European Society of Cardiology

Mosaic inactivation of the serum response factor gene in the myocardium induces focal lesions and heart failure

Guillaume Gary-Boboa,b, Ara Parlakiana,b, Brigitte Escoubetc,d,e, Claudio Areias Francoa,b, Sophie Clémentf, Patrick Brunevalg, David Tuilh, Dominique Daegelenh, Denise Paulina,b, Zhenlin Lia,b,* and Mathias Mericskaya,b

a UPMC Univ Paris 06, UMR7079, physiology and Physiopathology Paris, France
b CNRS, UMR 7079 Paris, France
c Denis Diderot University Paris 7, CEFI IFR02 Faculté de Médecine Xavier-Bichat Paris, France
d Assistance Publique-Hôpitaux de Paris, Bichat Hospital Paris, France
e INSERM, U722 Paris, France
f University of Geneva, Department of Clinical Pathology Geneva, Switzerland
g Assistance Publique-Hôpitaux de Paris, George Pompidou European Hospital Paris, France
h René Descartes University, Cochin Institute, INSERM U567 and CNRS UMR 8104, Genetics and development department Paris, France

* Corresponding author. UPMC Univ Paris 06, CNRS UMR7079, physiology and Physiopathology, BP256, 7 Quai St-Bernard, 75005 Paris, France. Tel.: +33 1 44 27 21 36; fax: 33 1 44 27 21 35. E-mail address: zhenli{at}ccr.jussieu.fr (Z. Li).


   Abstract
Top
 Abstract
1. Introduction
 2. Material and methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 References
 
Background and aims: Regional alterations in ventricular mechanical functions are a primary determinant for the risk of myocardial injuries in various cardiomyopathies. The serum response factor (SRF) is a transcription factor regulating contractile and cytoskeletal genes and may play an important role in the remodelling of myocardium at the cellular level.

Methods: Using Desmin-Cre transgenic mice, we generated a model of mosaic inactivation of a floxed-Srf allele in the heart to analyze the consequence of regional alterations of SRF-mediated functions in the myocardium.

Results: Two types of cardiomyocytes co-existed in the Desmin-Cre:Sf/Sf mice. Cardiomyocytes lacking SRF became thin and elongated while cardiomyocytes containing SRF became hypertrophic. Several physiological contractile genes were down-regulated while skeletal {alpha}-actin was induced in SRF positive area only. Mutants developed heart failure associated with the presence of focal lesions in the myocardium, and died before month 11.

Conclusions: Juxtaposition of functional SRF wild-type and failing SRF mutant cardiomyocytes generates deleterious heterogeneity in the myocardium. Our results show that SRF contributes to the capacity of cardiomyocytes to remodel their shape and contractile functions in response to their local environment; suggesting that it may play a role in pathologies involving regional alterations of ventricular mechanics in the heart.

Key Words: Myocardial injury • Regional ventricular dysfunction • Mosaic Cre recombination • Actin • Eccentric and concentric hypertrophy

Received January 11, 2008; Revised April 3, 2008; Accepted April 24, 2008


   1. Introduction
Top
 Abstract
 1. Introduction
2. Material and methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 References
 
Regional mechanical dysfunctions that are observed after myocardial ischaemia and other cardiac pathologies, arise in part from local alterations in stress and strain generated at the interface between normal and injured muscle [1]. Mechanical stress-strain sensors embedded in the cardiac cytoskeleton are able to transduce biomechanical stimuli to transcriptional pathways leading to changes in gene expression and modification of the cell phenotype [2]. These sensors, located at the level of the Z disks, the M-band, the intercalated disks or the costameres, contain transduction molecules that appear to shuttle from the cytoskeleton to the nucleus. Recently, several studies point toward the serum response factor (SRF), a transcription factor of the MADS (MCM1, ARG80, Deficiens, SRF) box family, as an effector of signal transduction pathways linked to cytoskeleleton. Cardiac-specific overexpression of SRF in transgenic mice leads to hypertrophic remodelling of the heart and cardiomyopathy [3].

SRF is a major regulator of actin genes as well as other muscle genes encoding cytoskeletal proteins and proteins involved in the regulation of contraction like Serca2a and muscle creatine kinase (MCK) [4,5]. SRF activity is intimately linked to actin filament dynamics through the interaction of the SRF cofactor MRTFA/B (Myocardin-related transcription factors A and B) with G-actin. When signals, mainly those acting through RhoA, stimulate the incorporation of G-actin into the filamentous F-actin, MRTFA/B is released from G-actin and can interact with SRF in the nucleus where it activates a specific subset of SRF target genes including β-actin [6]. STARS (Striated Activator of RhoA Signalling), an actin-binding protein is increased during cardiac hypertrophy and increases SRF activity [7]. Conversely, SRF can be repressed by Murf2 (Muscle specific ring of finger protein), a component of the ubiquitin-proteasome system of protein degradation that is associated to the M-band titin kinase-like domain and is released in response to arrest of contraction [8]. Murf1 partially blunts cardiac hypertrophy in mice and interacts with SRF [9]. Interestingly, a truncated form of SRF, that may be formed by caspase-3 cleavage, has been detected in human failing hearts, suggesting that SRF is the target of multiple degradation pathways in the context of heart failure [10].

Several groups have used Cre-LoxP technology to inactivate the Srf gene in the mouse embryo and shown that SRF is crucial for the maturation of the cardiac chambers in the embryo, particularly for the normal growth of the compact layer and the trabeculae [11-13]. We used a tamoxifen-inducible Cre recombinase to inactivate floxed-Srf allele in the adult heart [5]. Inactivation of SRF throughout the myocardium of adult mice resulted in the down-regulation of critical target genes including cardiac {alpha}-actin, MCK and Serca2a, and ultimately fatal dilated cardiomyopathy within two months after tamoxifen injection. These results showed that SRF plays an important role in the maintenance of cardiomyocyte functions in the adult heart.

Our aim was to create a new model to generate regional alteration of contractility in the myocardial wall. Based on the central role of SRF in the regulation of contractile proteins, we crossed our floxed-Srf mice with a new Desmin-Cre transgenic line that expresses the Cre recombinase in a mosaic pattern in the heart. Our results show that the simple juxtaposition of functional wild-type and failing mutant cardiomyocytes led to the formation of severe focal lesions in the myocardial wall in absence of any other injuries.


   2. Material and methods
Top
 Abstract
 1. Introduction
 2. Material and methods
3. Results
 4. Discussion
 Appendix A. Supplementary data
 References
 
2.1. Transgenic mice
Desmin-Cre transgenic mice bear the desmin 4-kb promoter [14] linked to a Cre cDNA. The linearized Desmin-Cre fragment was injected into (C57BL/6XDBA2) eggs and the Desmin-Cre F0 founders were bred with C57BL/6 mice. Srf-flex2neo (Sf/Sf) and ROSA26R mice and PCR genotyping have been described previously [11,15]. Desmin-Cre males were bred with Rosa26R (C57BL/6 background) females to generate animals for X-gal staining. Desmin-Cre males were crossed with Sf/Sf females (C57BL/6 background) to generate first Desmin-Cre:Sf/+ and then Desmin-Cre:Sf/Sf mutants. Sf/Sf mice from the same litter were used as controls. A total of 66 controls and 68 mutants have been used for this study. The investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1985).

2.2. X-Gal staining
Tissues were fixed in 1% formaldehyde and 0.2% glutaraldehyde for 30 min at 4 °C, and incubated overnight at 37 °C in the X-Gal staining solution as described previously [10].

2.3. Histology and immunohistochemistry
Fibrosis was detected by Sirius red (0.1%) staining. Immunohistochemistry using anti-SRF (1:200, sc-335, Santa Cruz) was performed as previously described [5]. Immunofluorescence staining were done on frozen sections with the following primary antibodies: anti-vinculin, (1:400, hVIN-1, Sigma), anti-dystrophin (1:50, Dy8/6C5, Novocastra); anti-skeletal {alpha}-actin (1:20, [16]); anti-complement 9 (1:50, 10A6, Novocastra); anti-SRF (1:40, Santa Cruz); anti-phospho-Histone H3 (1:50, Cell Signalling), followed by Cy3 or Alexa488-coupled secondary antibody. Apoptosis detection was performed by TUNEL assay (Roche). TRITC-phalloidin (0.33 {varepsilon}M, Sigma) was used at dilution 1:2000.

2.4. Isolation of adult mouse cardiomyocytes
Cardiomyocytes were isolated as described in online protocol (http://www.signaling-gateway.org/data/ProtocolLinks.html). Briefly, hearts were harvested from adult mice and perfused through the aorta for 10 min with 0.25 mg/ml liberase blendzyme 1 (Roche) and 0.14 mg/ml trypsin (Invitrogen) in presence of 10 mM 2,3-butanedione monoxime. Rod-shaped cardiomyocytes were allowed to attach on coverslips coated with 10 µg/ml Laminin (BD Biosciences) in MEM/FCS 10% for 90 min and fixed in 4% formaldehyde.

2.5. Morphological measurements
Measurements of the fibrosis area and number of lesions were performed by colour selection on photographs of sirius red staining with constant parameters for all samples using ImageJ software (http://rsb.info.nih.gov/ij/). Quantification of SRF positive and SRF negative nuclei was performed manually on 4 fields (total>1000 nuclei) for each animal. Cross-sectional area of the cardiomyocytes was measured on TRITC-phalloidin labelled sections in the left ventricle (LV). Cell area, maximal length and width of isolated cardiomyocytes were measured in 30 cardiomyocytes for each group. Quantification of TUNEL or phospho-Histone H3 labelling was done on entire transversal sections at the median level of the heart. The total number of nuclei per section was estimated from the calculated average numerical density from 4 independent 10X fields and multiplied by the total area of the section. For all quantifications, 3 sections per heart were counted at each stage for each animal (n=3 to 6 for each group).

2.6. Echocardiography
Echocardiography was performed with a Toshiba Powervision 6000, SSA 370A device equipped with an 8- to 14-MHz linear transducer under isoflurane anesthesia with spontaneous ventilation, as previously described [5].

2.7. Relative quantification of gene expression by real-time RT-PCR
PCR analysis of reversed transcribed RNA was performed with SYBR green (ABGene) on an iQTM Real-time PCR detection system (Bio-Rad) as previously described [5]. Primer sequences are available on request. Target: GAPDH expression ratio was expressed as a fold over mean control value and data are given±SEM.

2.8. Statistical analysis
We performed ANOVA and Fisher post hoc tests for repeated measures and used Student's unpaired t tests for comparisons with control mice at specific times. The data shown are means ± SEM. P-values of p<0.001 (***), p<0.01 (**) and p<0.05 (*) were considered statistically significant.


   3. Results
Top
 Abstract
 1. Introduction
 2. Material and methods
 3. Results
4. Discussion
 Appendix A. Supplementary data
 References
 
3.1. Mosaic pattern of Desmin-Cre transgene expression
Desmin is an intermediate filament protein which is expressed in all muscle cells including the heart [17]. We have shown previously that the 4 kb regulatory region upstream of the desmin gene targets transgene expression in all muscle cell types [14]. Desmin-Cre lines were generated and bred with Rosa26R mice bearing a Cre-inducible lacZ reporter gene [15]. Three of the five transgenic lines expressed Cre in similar patterns, as detected by whole-mount X-Gal staining of tissues. The staining was mostly restricted to the trabeculae of the heart in the embryo (Supplementary Fig. S1, A-B). Whole-mount staining of the young adult heart and histological sections (1.5 month-old) revealed a mosaic pattern of lacZ expression with the presence of positive cardiomyocytes, roughly 50% of the cells, close to negative cardiomyocytes in the myocardium (Supplementary Fig. S1, C-D). Mosaic Desmin-Cre expression was also observed in skeletal and smooth muscles (Supplementary Fig. S1, E-I).

3.2. Mosaic pattern of SRF deletion in the heart of Desmin-Cre:Sf/Sf mutants
Desmin-Cre:Sf/Sf mutant mice were generated in the predicted Mendel ratio at birth. PCR analysis of genomic DNA showed that the deletion level was low at birth but increased with age (Fig. 1A). Real-time RT-PCR analysis showed a progressive decrease of SRF transcripts to 61±5% at 35 days in the mutant heart (Fig. 1B). Immunostaining on heart sections showed that the number of SRF positive (SRFPos) nuclei in the mutant myocardium gradually decreased from 20 days onwards (Fig. 1C-D, I). The numbers of SRFPos nuclei in the skeletal muscles of the mutants were unchanged (Fig. 1E-F, I), although the amount of wild-type SRF mRNA decreased to the same extent as in the heart (Supplementary Fig. S2). We did not detect evidence of skeletal muscle dystrophy in the mutant mice. The number of coronary SRFPos smooth muscle cells (SMCs) was not affected (Fig. 1G-I). Hence, the Desmin-Cre:Sf/Sf mutants gradually lose their SRF, mainly in the heart, beginning 15 to 20 days after birth, when the proliferation of cardiomyocytes decreases and hypertrophic growth begins.


Figure 01
View larger version (84K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Analysis of SRF expression in Desmin-Cre:Sf/Sf mice. (A) PCR of genomic DNA from the hearts of newborn (nb) and 45 day-old mice (ad). Floxed-Srf allele (490 bp), wild-type Srf allele (448 bp) and mutated allele lacking exon 2 (320 bp) are detected. (B) Real-time RT-Q-PCR analysis of wild-type Srf gene expression in the heart of controls (Co) and Desmin-Cre:Sf/Sf mutants (Mu) at 15, 25 and 35 days. (C-H) Immunoperoxidase staining of nuclei (brown) with anti-SRF antibody in control (C, E, G) and mutant (D, F, H) at day 30. (C, D) Heart. (E, F) Skeletal muscle. (G, H) Coronaries. The SRF negative nuclei were counter-stained in blue by methyl green. (I) Quantification of SRF positive nuclei as a percentage of age-matched control. N=3 for each group.

 
3.3. Focal lesions, hypertrophy, and heart failure in the myocardium of Desmin-Cre:Sf/Sf mutants
About half of the Desmin-Cre:Sf/Sf mutant mice died before they were 100 day-old and all of them were dead by 330 days (Supplementary Fig. S3). Echocardiography demonstrated alteration of LV systolic function at 40 and 88 days as shown by the reduction of the ejection fraction and circumferential fibres contraction velocity (Table 1). LV end-diastolic diameter was increased but there was no LA remodelling (Table 1). Macroscopic observation of the mutant hearts revealed focal lesions in the walls of the ventricles (compare Fig. 2A-B, arrowheads). The extent of the focal lesions varied, and could be transmural in the most severe cases (compare Fig. 2C-D). There was a mild diffuse interstitial fibrosis throughout the myocardium (Fig. 2D-E) and the lesions were strongly fibrotic (Fig. 2D, F). Quantitative analysis of the Sirius red stained sections revealed the presence of 4 to 6 lesions in the LV versus 0.33 to 1 lesion in the RV showing that the LV was the most damage-prone region (Supplementary Table 1). The fibrosis index reached 10% of the total area at 30 days (Fig. 2G). Mutant mice developed cardiac hypertrophy, starting 20-30 days after birth (Fig. 2H). Thus the loss of SRF in 50% of the cardiomyocytes leads to slow progressive heart failure with focal lesions, extensive fibrosis and LV remodelling.


View this table:
[in this window]
[in a new window]

  		Table 1 Echocardiography in Desmin-Cre:Sf/Sf mutant mice

 


Figure 02
View larger version (56K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Morphological modifications in Desmin-Cre:Sf/Sf mutant hearts. (A, B) Macroscopic view of 30 day-old control and mutant hearts. Note the presence of red scars at the LV surface (arrowheads) of the mutant heart. Bar = 1 mm. (C, D) Sirius red staining of 30 day-old control and mutant hearts. Arrowhead: lesion regions with intensive fibrosis. Bar = 1 mm. LV: left ventricle; RV: right ventricle. (E, F) Higher magnification of boxed area in D showing interstitial fibrosis in the ventricular septum (E) and a fibrotic transmural lesion (F). (G) Quantification of fibrosis area in the heart. N=4 in each group. (H) Heart weight/body weight ratio. N=6 in each group.

 
3.4. Increased cell death and proliferation in the SRF depleted myocardium
In the heart of the Desmin-Cre:Sf/Sf mice, we found a higher number of TUNEL-positive apoptotic cells that were localized in the lesions as well as scattered throughout the myocardium where they were co-labelled by phalloidin-TRITC (Fig. 3, A-C) and we found a high level of necrosis in the focal lesions (Fig. 3D). There were more proliferative cells in the young Desmin-Cre:Sf/Sf mutants than in the controls (not shown, quantification in Fig. 3C). Proliferation decreased with age, down to that of controls at 90 days.


Figure 03
View larger version (81K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Cell proliferation and cell death analysis in mutant hearts. (A) TUNEL (green), DAPI (blue), and phalloidin-TRITC (red) staining of mutant hearts. Inset highlights the characteristic shape of the apoptotic nucleus as labelled with DAPI. (B) TUNEL positive cells (arrows) at the lesion site. Bars = 20 µm. (C) Quantification of TUNEL and anti-phospho-Histone H3 labelling of mitotic chromosomes from controls and mutants at 15, 30, and 90 days. Data are expressed as the number of positive nuclei for 105 cells. The number of mice analyzed is shown in the bars. (D) Anti-complement-9 (red) labelling the terminal C5b-9 membrane attack complex and DAPI (blue) staining in mutant heart. Bar = 100 µm.

 
3.5. Reduced F-actin and dystrophin content in injured cardiomyocytes
We analyzed the changes in the cytoarchitecture associated with SRF depletion in cardiomyocytes. Sections were stained for F-actin with TRITC-phalloidin and for vinculin-h1, which connects actin to integrin at the intercalated disks (Fig. 4A). The focal lesions were filled with atrophic cardiomyocytes with weak F-actin and remnants of intercalated disks (Fig. 4A, arrowhead). There were large isolated cardiomyocytes with strong F-actin (Fig. 4A, arrows). The atrophic cardiomyocytes were SRFNeg (Fig. 4B, arrowheads), whereas the large cardiomyocytes were SRFPos (Fig. 4B, arrows). Immunostaining for dystrophin, a linker of actin filaments to the dystroglycan complexes, homogeneously labelled the sarcolemma membranes in control sections (Fig. 4C). In contrast, the dystrophin labelling was reduced and discontinuous in small clusters of cardiomyocytes as early as day 15 (Fig. 4D). Skeletal {alpha}-actin was strongly induced in many fibres in the heart of Desmin-Cre:Sf/Sf (Fig. 4F) compared to the scattered pattern in the control hearts (Fig. 4E). Thin atrophic cardiomyocytes were negative for skeletal {alpha}-actin and SRF (Fig. 4F, H, arrowheads). We also observed an up-regulation of laminin and vimentin expression as soon as day 15 in the mutant hearts (Supplementary Fig. S4). These results show that alterations of the extracellular matrix (ECM) composition and activation of the cardiac fibroblasts may also contribute to the phenotype of Desmin-Cre:Sf/Sf mice.


Figure 04
View larger version (114K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Cytoskeletal alterations in the mutant heart. (A-F) Frozen sections were labelled with phalloidin-TRITC (red) for F-actin, DAPI (blue) for nuclei and various antibodies (green). (A) Focal lesion at day 30, filled with small cardiomyocytes that stain weakly for F-actin and contain remnants of vinculin-positive intercalated disks (arrowheads). Note the presence of isolated rounded-up hypertrophic cardiomyocytes in the lesion (arrows). (B) SRF staining shows that atrophic cardiomyocytes in the lesions are SRFNeg (arrowheads) whereas large cardiomyocytes are SRFPos (arrows). (C, D) Anti-dystrophin staining at day 15, clusters of 5 to 10 atrophic cardiomyocytes show a reduced labelling in the mutant at day 15 (arrowheads in D). Insets: dystrophin and dapi staining only. Bar = 10 {varepsilon}m in A to D. (E-H) Immunostaining with polyclonal rabbit antibodies on serial hearts paraffin sections from 30 day-old control (E, G) and mutant (F, H). (E, F) Skeletal {alpha}-actin. (G-H) SRF. Note the presence of thin elongated cardiomyocytes in the mutant heart that are negative both for SRF and skeletal {alpha}-actin (arrowheads). Bar = 50 {varepsilon}m. Co: control; Mu: mutant.

 
3.6. Hypertrophic remodelling of the normal myocardium in mutants
Most cardiomyocytes in the remote myocardium of Desmin-Cre:Sf/Sf mice were hypertrophic (Fig. 5, compare C-D to A-B) with an intense F-actin labelling. Cross-sectional area (CSA) of cardiomyocytes showed a significant increase in the mutants versus the controls as early as 20 days and this difference was maintained over time (Fig. 5E-F). To better assess the type of hypertrophic remodelling associated with the presence or absence of SRF, we performed a double immunostaining for F-actin and SRF on cardiomyocytes isolated from 30 day-old Sf/Sf control mice (Fig. 5G) and Desmin-Cre:Sf/Sf mice (Fig. 5H-I). Morphometric analysis showed that the two populations of SRFPos and SRFNeg cardiomyocytes from Desmin-Cre:Sf/Sf mice differed clearly from cardiomyocytes from control Sf/Sf mice (Fig. 5J). SRFPos cardiomyocytes have increased both cell width and length while SRFNeg cardiomyocytes have increased cell length only and reduced cell width. These results show that SRFNeg cardiomyocytes developed eccentric hypertrophy whereas SRFPos cardiomyocytes developed concentric hypertrophy.


Figure 05
View larger version (64K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Cardiomyocytes remodelling in mutant hearts. (A-D) Phalloidin (red) and vinculin (green) staining of remote myocardium in control (Co) and mutant (Mu) hearts. (A, C) longitudinal sections. (B, D) transversal sections. Arrows: intercalated disks. (E) Cross sectional area (CSA) of cardiomyocytes in transversal sections of controls (white) and mutants (black). (F) Frequency plot of CSA at day 30. N=4 in each group. Small degenerating cardiomyocytes at the sites of injury were not included in this analysis because of the difficulty to delineate cell contour. (G-I) Phalloidin (red) and SRF (green) staining of isolated cardiomyocytes from 30 day-old control Sf/Sf (G) and Des-Cre:Sf/Sf mice (H, SRFPos; I, SRFNeg cells). Bar = 20 {varepsilon}M. Cardiomyocytes were isolated from 2 controls and 3 mutants. (J) Quantification of area, length and width of isolated cardiomyocytes. N=30 cardiomyocytes in each group.

 
3.7. Altered post-natal gene expression in the mutant heart
We assessed the consequence of the loss of SRF on cardiac gene expression by real-time RT-PCR. Cardiac {alpha}-actin started to decline at 25 days (Fig. 6A), slightly later than the decrease in wild-type SRF mRNA at 15 days (see Fig. 1B). In contrast, skeletal {alpha}-actin, a marker of cardiac hypertrophy and an SRF target, started to increase at 25 days (Fig. 6B). The myosin heavy chain (MHC) isoforms underwent a similar switch, with a loss of the post-natal {alpha}-MHC isoform and an increase in the embryonic β-MHC isoform (Fig. 6, C-D). The expression of the Serca2 gene, whose promoter harbours a CArG box, was decreased at 35 days (Fig. 6E). Dystrophin mRNA expression was not affected (Fig. 6F). IGF1, which is involved in cardiomyocyte growth and survival, was decreased by 40% at days 25 and 35 (Fig. 6G) while GATA4 was transiently up-regulated at 25 days (Fig. 6H). Altogether, these results show that there was a switch toward a pathological gene program in the hearts of the Desmin-Cre:Sf/Sf mutant mice between day 15 and 35.


Figure 06
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Time course analysis of gene expression. Real-time RT-PCR of control (Co) and mutant (Mu) heart transcripts at 15, 25, 35 days postnatal. Data were normalized on GAPDH and expressed as a percentage of mean control value±SEM at each age. N=3 in each group.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Material and methods
 3. Results
 4. Discussion
Appendix A. Supplementary data
 References
 
Our principal finding is that the Desmin-Cre:Sf/Sf mutant mice develop focal myocardial lesions during the post-natal period. These lesions appeared around the third week of life and were accompanied by extensive fibrosis, cardiac hypertrophy and ultimately heart failure. The formation of the myocardial lesions was correlated with the mosaic inactivation of SRF in the postnatal cardiomyocytes by the Desmin-Cre transgene. We found that SRF plays an important role for maintenance of cardiomyocyte functions in the post-natal heart.

The inactivation of floxed-Srf alleles in the heart by {alpha}- or β-MHC-Cre transgenes results in embryonic lethality [11-13]. In contrast, none of the Desmin-Cre:Sf/Sf mice died in utero. This could be related to a low-level of the Desmin-Cre transgene expression in the embryonic heart (Supplementary Fig. S1) that is not sufficient to trigger SRF inactivation in a significant number of cardiomyocytes at these early stages, allowing the mutants to develop and reach birth. These contrasting results highlight the fact that very different outcomes can be obtained using Cre-Lox technology for the same gene in a tissue, merely by changing the percentage of Cre-expressing cells. Similarly, different phenotypic outcomes were reported for cardiac-specific deletion of floxed-Gata4 alleles when using different Cre transgenic mice [18-19].

Post-natal cardiomyocytes proliferate up to day 5 to 10 in the mouse and then display intense hypertrophic growth [20]. Cardiac lesions appeared only 15 days after birth in Desmin-Cre:Sf/Sf mice. Perhaps the increased cell proliferation that we observed at early ages (Fig. 3C) is initially sufficient to compensate for the loss of SRF deficient cardiomyocytes. However, proliferation ultimately decreased, while cell death increased in Desmin-Cre:Sf/Sf mutant hearts. This might be related to the downregulation of the prosurvival factor IGF1 at day 25 and later. IGF1 is a direct target of SRF in skeletal muscle [21].

Our previous work using {alpha}-MHC-MerCreMer:Sf/Sf inducible mice showed that deletion of SRF in 90 to 100% of the adult cardiomyocytes led to a 80% down-regulation of essential target genes expression such as cardiac {alpha}-actin and MCK, and 50% for Serca2a [5]. The expressions of the same contractile genes are decreased in Desmin-Cre:Sf/Sf mutants, although only to 40-50%. This smaller absolute loss is in good agreement with the mixture of SRFNeg and SRFPos cardiomyocytes in the mutant heart. On the other hand, skeletal {alpha}-actin, a marker of cardiac hypertrophy and a well defined SRF target, is induced in the Desmin-Cre:Sf/Sf mutant heart, while it is down-regulated in the {alpha}-MHC-MerCreMer:Sf/Sf mutant [5]. We suggest that this up-regulation reflects the activation of the gene in some SRFPos cardiomyocytes that become hypertrophic in Desmin-Cre:Sf/Sf mutants and our results using anti-skeletal {alpha}-actin antibody (Fig. 4) are in agreement with this hypothesis. Mosaic SRF inactivation generated two phenotypically different populations of cardiomyocytes. SRFNeg cardiomyocytes were elongated and thinner than control cardiomyocytes, which is a characteristic of eccentric hypertrophy, whereas SRFPos cells developed concentric hypertrophy (Fig. 5). Strikingly, our analysis on isolated cardiomyocytes clearly shows that the ability of cardiomyocytes to develop concentric hypertrophy is strictly dependent on SRF expression. In Desmin-Cre:Sf/Sf mutant hearts, the clusters with a majority of SRFNeg cardiomyocytes probably form hypo-contractile regions that are bordered by more contractile regions containing SRFPos cells. The differences in biomechanical properties of hypo-contractile and contractile regions appear to be sufficient to trigger a hypertrophic response in the remote SRFPos myocardium. While this response may initially help to preserve cardiac output, it probably increases the stress on neighbouring SRFNeg cardiomyocytes. Focal lesions were not observed in our previous model in which the Srf gene was inactivated throughout the whole myocardium of adult mice using a tamoxifen-inducible {alpha}-MHC-MerCreMer transgene [5]. These mutant mice displayed an early decrease of contractility followed by a fatal dilated cardiomyopathy. One of the main differences between the {alpha}-MHC-MerCreMer:Sf/Sf model and the Desmin-Cre:Sf/Sf is that in the former, almost all the cardiomyocytes simultaneously start to down-regulate SRF target genes, including cardiac {alpha}-actin and {alpha}-myosin heavy chain. Therefore, the stress induced by the volume overload is distributed over a homogeneously weak myocardium. Another difference between the two models could be age, adult heart at baseline in the {alpha}-MHC-MerCreMer:Sf/Sf model [5] versus postnatal "growing" heart in the Desmin-Cre:Sf/Sf model. However, we found that triggering SRF inactivation in 10 day-old {alpha}-MHC-MerCreMer:Sf/Sf neonates led to early dilated cardiomyopathy and death within 1 month in absence of detectable focal lesions (D. Tuil, unpublished data), suggesting that it is the heterogeneous (mosaic) versus homogeneous inactivation that explains the difference.

In addition to the lower actin content, we looked for alterations in the molecular linkers of actin cytoskeleton to the sarcolemma, such as vinculin and dystrophin that were reported as putative SRF targets [22,23]. We could not find a down-regulation of these genes at the mRNA level in both the Desmin-Cre:Sf/Sf heart and the adult inducible model, suggesting that other transcriptional regulators can sustain the expression of these genes in the adult heart (Fig. 6 and data not shown). However, alterations of dystrophin pattern were readily observed at the sarcolemma of lesional cardiomyocytes of the Desmin-Cre:Sf/Sf mutant hearts (see Fig. 4D). This defect in sarcolemma integrity in the absence of SRF may be secondary to the actin cytoskeleton defect or due to the loss of other unidentified SRF targets. It is interesting to note that mosaic inactivation of dystrophin in the heart also leads to cardiomyopathy in humans [24]. Indeed, human females carryings a mutated dystrophin allele have mosaic dystrophin expression in the heart because each cardiomyocyte can express only one of the two dystrophin alleles, either the wild-type or the mutant allele, due to the silencing of one X-chromosome. A cohort study (N=197) showed that 44% of these female patients experienced clinically relevant cardiomyopathy and less than 10% had normal cardiac status while they rarely develop skeletal muscle dystrophy [24]. This study and our observations in Desmin-Cre:Sf/Sf mice show the importance of a homogeneous transmission of force across the myocardium.

In conclusion, we have generated, through genetic engineering, a new model of cardiac pathology in which focal lesions form in the myocardium in response to SRF inactivation in sub-regions of the myocardium. The use of the Desmin-Cre transgenics with floxed-Srf and other floxed-alleles will prove useful for research on the molecular, cellular and biomechanical consequences of abnormal wall motion and could be a tool for testing different therapeutic approaches. Drugs like statins that repress RhoA activity may have an impact on SRF activity in the heart, which could be beneficial in the context of cardiac hypertrophy [25]. Because of the central role of SRF downstream of biomechanical activated pathways and upstream of cytoskeletal and contractile genes expression, it will be relevant to assess the status of SRF expression or post-transcriptional modifications in lesions, border zone and remote myocardium after coronary infarct or other pathologies involving regional alterations of ventricular mechanics. These studies would help to determine the benefit of therapeutic strategies aimed at modulating SRF activity in the heart.


   Appendix A. Supplementary data
Top
 Abstract
 1. Introduction
 2. Material and methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
References
 
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ejheart.2008.04.014.


   Acknowledgement
 
We acknowledge financial support from the Agence National de la Recherche, the Association Française contre les Myopathies and Fondation de France. G.G.B holds a fellowship from the GRRC and the Chancellerie des Universités de Paris and C.A.F a fellowship from the Fondation pour la Recherche Médicale (FRM).


   References
Top
 Abstract
 1. Introduction
 2. Material and methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 References
 

  1. Mazhari R., McCulloch A.D. Integrative models for understanding the structural basis of regional mechanical dysfunction in ischemic myocardium. Ann Biomed Eng (2000) 28:979–990.[CrossRef][Web of Science][Medline]
  2. Hoshijima M. Mechanical stress-strain sensors embedded in cardiac cytoskeleton: Z disk, titin, and associated structures. Am J Physiol Heart Circ Physiol (2006) 290:H1313–H1325.[Abstract/Free Full Text]
  3. Zhang X., Azhar G., Chai J., et al. Cardiomyopathy in transgenic mice with cardiac-specific overexpression of serum response factor. Am J Physiol Heart Circ Physiol (2001) 280:H1782–H1792.[Abstract/Free Full Text]
  4. Miano J.M., Long X., Fujiwara K. Serum response factor: master regulator of the actin cytoskeleton and contractile apparatus. Am J Physiol Cell Physiol (2007) 292:C70–C81.[Abstract/Free Full Text]
  5. Parlakian A., Charvet C., Escoubet B., et al. Temporally controlled onset of dilated cardiomyopathy through disruption of the SRF gene in adult heart. Circulation (2005) 112:2930–2939.[Abstract/Free Full Text]
  6. Miralles F., Posern G., Zaromytidou A.I., Treisman R. Actin dynamics control SRF activity by regulation of its coactivator MAL. Cell (2003) 113:329–342.[CrossRef][Web of Science][Medline]
  7. Kuwahara K., Teg Pipes G.C., McAnally J., et al. Modulation of adverse cardiac remodeling by STARS, a mediator of MEF2 signaling and SRF activity. J Clin Invest (2007) 117:1324–1334.[CrossRef][Web of Science][Medline]
  8. Lange S., Xiang F., Yakovenko A., et al. The kinase domain of titin controls muscle gene expression and protein turnover. Science (2005) 308:1599–1603.[Abstract/Free Full Text]
  9. Willis M.S., Ike C., Li L., et al. Muscle ring finger 1, but not muscle ring finger 2, regulates cardiac hypertrophy in vivo. Circ Res (2007) 100:456–459.[Abstract/Free Full Text]
  10. Chang J., Wei L., Otani T., et al. Inhibitory cardiac transcription factor, SRF-N, is generated by caspase 3 cleavage in human heart failure and attenuated by ventricular unloading. Circulation (2003) 108:407–413.[Abstract/Free Full Text]
  11. Parlakian A., Tuil D., Hamard G., et al. Targeted inactivation of serum response factor in the developing heart results in myocardial defects and embryonic lethality. Mol Cell Biol (2004) 24:5281–5289.[Abstract/Free Full Text]
  12. Miano J.M., Ramanan N., Georger M.A., et al. Restricted inactivation of serum response factor to the cardiovascular system. Proc Natl Acad Sci U S A (2004) 101:17132–17137.[Abstract/Free Full Text]
  13. Niu Z., Yu W., Zhang S.X., et al. Conditional mutagenesis of the murine serum response factor gene blocks cardiogenesis and the transcription of downstream gene targets. J Biol Chem (2005) 280:32531–32538.[Abstract/Free Full Text]
  14. Mericskay M., Parlakian A., Porteu A., et al. An overlapping CArG/octamer element is required for regulation of desmin gene transcription in arterial smooth muscle cells. Dev Biol (2000) 226:192–208.[CrossRef][Web of Science][Medline]
  15. Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet (1999) 21:70–71.[CrossRef][Web of Science][Medline]
  16. Clement S., Chaponnier C., Gabbiani G. A subpopulation of cardiomyocytes expressing alpha-skeletal actin is identified by a specific polyclonal antibody. Circ Res (1999) 85:e51–e58.[Web of Science][Medline]
  17. Lazarides E., Hubbard B.D. Immunological characterization of the subunit of the 100 A filaments from muscle cells. Proc Natl Acad Sci U S A (1976) 73:4344–4348.[Abstract/Free Full Text]
  18. Zeisberg E.M., Ma Q., Juraszek A.L., et al. Morphogenesis of the right ventricle requires myocardial expression of Gata4. J Clin Invest (2005) 115:1522–1531.[CrossRef][Web of Science][Medline]
  19. Oka T., Maillet M., Watt A.J., et al. Cardiac-specific deletion of Gata4 reveals its requirement for hypertrophy, compensation, and myocyte viability. Circ Res (2006) 98:837–845.[Abstract/Free Full Text]
  20. Leu M., Ehler E., Perriard J.C. Characterisation of postnatal growth of the murine heart. Anat Embryol (Berl) (2001) 204:217–224.[CrossRef][Medline]
  21. Charvet C., Houbron C., Parlakian A., et al. New role for serum response factor in postnatal skeletal muscle growth and regeneration via the interleukin 4 and insulin-like growth factor 1 pathways. Mol Cell Biol (2006) 26:6664–6674.[Abstract/Free Full Text]
  22. Gilgenkrantz H., Hugnot J.P., Lambert M., et al. Positive and negative regulatory DNA elements including a CCArGG box are involved in the cell type-specific expression of the human muscle dystrophin gene. J Biol Chem (1992) 267:10823–10830.[Abstract/Free Full Text]
  23. Gineitis D., Treisman R. Differential usage of signal transduction pathways defines two types of serum response factor target gene. J Biol Chem (2001) 276:24531–24539.[Abstract/Free Full Text]
  24. Politano L., Nigro V., Nigro G., et al. Development of cardiomyopathy in female carriers of Duchenne and Becker muscular dystrophies. Jama (1996) 275:1335–1338.[Abstract/Free Full Text]
  25. Saka M., Obata K., Ichihara S., et al. Attenuation of ventricular hypertrophy and fibrosis in rats by pitavastatin: potential role of the RhoA-extracellular signal-regulated kinase-serum response factor signalling pathway. Clin Exp Pharmacol Physiol (2006) 12:1164–1171.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Gary-Bobo, G.
Right arrow Articles by Mericskay, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gary-Bobo, G.
Right arrow Articles by Mericskay, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?