Skip Navigation

Cardiovascular Research 1999 41(3):575-585; doi:10.1016/S0008-6363(98)00264-8
© 1999 by European Society of Cardiology
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 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
Google Scholar
Right arrow Articles by Peterson, D. J.
Right arrow Articles by Dixon, I. M.C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Peterson, D. J.
Right arrow Articles by Dixon, I. M.C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1999, European Society of Cardiology

Expression of Gi-2{alpha} and Gs{alpha} in myofibroblasts localized to the infarct scar in heart failure due to myocardial infarction

David J. Peterson, Haisong Ju, Jianming Hao, Marcello Panagia, Donald C. Chapman and Ian M.C. Dixon*

Molecular Cardiology Laboratory, Institute of Cardiovascular Sciences, St. Boniface General Hospital Research Centre, Faculty of Medicine, University of Manitoba, 351 Tache Avenue, Winnipeg, Manitoba, Canada R2H 2A6

* Corresponding author. Tel. +1-204-235-3419; Fax: +1-204-233-6723; E-mail: iand@sbrc.umanitoba.ca

Received 17 November 1997; accepted 3 August 1998


   Abstract
Top
 Abstract
1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 
Objective: Patients surviving large transmural myocardial infarction (MI) are at risk for congestive heart failure with attendant alteration of ventricular geometry and scar remodeling. Altered Gi-2{alpha} and Gs{alpha} protein expression may be involved in cardiac remodeling associated with heart failure, however their expression in scar tissue remains unclear. Methods: MI was produced in Sprague–Dawley rats by ligation of the left coronary artery. Gi-2{alpha} and Gs{alpha} protein concentration, localization and mRNA abundance were noted in surviving left ventricle remote to the infarct, in border and in scar tissues from 8 week post-MI hearts with moderate heart failure. Results: We observed a 4.5- and 5.0-fold increase in immunoreactive Gi-2{alpha} protein concentration occurs in the border and scar regions vs. control values, respectively, in 8-week post-MI rat hearts. Similarly, immunoreactive Gs{alpha} protein concentration was increased 3.4- and 8.2-fold, respectively, in these tissues vs. controls. Double-fluorescence labeling and phenotyping studies revealed that both Gi-2{alpha} and Gs{alpha} proteins were localized to myofibroblasts in the infarct scar and to viable myocytes bordering the scar. Northern analysis revealed that the Gi-2{alpha}/GAPDH ratio was increased in both viable and scar regions (1.24- and 1.85-fold respectively) from experimental hearts when compared to sham-operated control values when compared to noninfarcted left ventricle, the value of this ratio in scar tissue was elevated ~1.5 fold. The Gs{alpha}/GAPDH ratio was significantly increased (1.28-fold) only in the scar region vs. control. Conclusion: Our results indicate a marked increase in the expression of Gi-2{alpha} and Gs{alpha} from myofibroblasts of the infarct scar as well as remnant myocytes bordering the scar in 8-week post-MI rat hearts. We suggest that these changes may be associated with ongoing remodeling in the infarct scar in chronic post-MI phase of this experimental model.

KEYWORDS Myofibroblasts; Gene expression; G-proteins; Congestive heart failure; Myocardial infarction; Scar; Remodeling; Immunofluorescence


   1 Introduction
Top
 Abstract
 1 Introduction
2 Methods
 3 Results
 4 Discussion
 References
 
Abnormal function of the myocardial β-adrenoceptor G-protein coupled pathway has been implicated in end-stage heart failure [1]. In addition to the downregulation of the β1-adrenoceptor itself in cardiac cells, altered expression of Gi-2{alpha} and Gs{alpha} are believed to contribute to cardiac dysfunction [2]. However, previous data regarding the expression of Gi-2{alpha} and Gs{alpha} in ischemic myocardial disease is inconclusive. Several studies demonstrate increased immunoreactive Gi-2{alpha} proteins in the failing human heart [3–5]. While this trend is matched by findings indicating elevated steady-state cardiac Gi-2{alpha} mRNAs in human end-stage heart failure [6], another study indicates no significant change in this parameter in an experimental model of heart failure [7]. Furthermore, in human hearts with ischemic heart disease, it was observed that Gi-2{alpha} protein concentration was similar to that detected in normal hearts [8, 9]. Thus the available data do not provide clear general trend for Gi-2{alpha} expression in failing myocardium. In this regard, the type and stage of heart disease may be a determining factor for relative expression of trimeric G-proteins. For example, an increase in the level of Gi-2{alpha} in dilated human cardiomyopathy but not human ischemic heart disease was demonstrated within the same study [8]. Healthy heart tissue is a composite of myocytes and nonmyocytes i.e., fibroblasts, while the healing infarct is marked by the presence of contractile myofibroblasts [10]. While cardiac nonmyocytes (including myofibroblasts and fibroblasts) are known to express vimentin, cardiac myocytes do not express this intermediate filament, and these two broad categories of cells may be distinguished on this basis [11]. We have recently noted upregulation of the cardiac Gq{alpha}/phospholipase C (PLC)-β pathway in the 8-week infarct scar (healed or chronic phase), as well as tissue bordering the scar and remote viable (remnant) myocardium from post-myocardial infarction (MI) rats [12]. A role for this pathway in the ongoing evolution of scar structure in the chronic phase of heart failure after MI was suggested [12]. As cardiac myofibroblasts are abundant at the site of cardiac tissue repair and are known to mediate the synthesis of fibrillar collagens, they are major players in formation of infarct scar structure in post-MI heart [13]. The healed infarct scar is specifically organized to confer anisotropic mechanical properties and to deform similarly to noninfarcted myocardium and thus may help preserve ventricular function [14].

Myocardial infarction is characterized by early infarct expansion in which the infarct site thins and elongates, and is followed by discrete scar formation via the wound healing response [15, 16]. The effect of scar formation is contingent upon the size of the initial MI. Decreased net contractile force associated with relatively small acute MI may be compensated by the viable myocardium and is characterized by maintenance of normal cardiac geometry [17]. On the other hand, large MI is attended by progressive ventricular chamber dilatation and sphericulization with interstitial fibrosis, leading to loss of normal cardiac function [17]. Thus scar tissue formation and maturation are relevant components of overall ventricular remodeling and an understanding of the processes that underlie infarct healing in post-MI hearts is warranted [16, 18]. Specific information about the expression of Gi-2{alpha} or Gs{alpha} (iso)subunits from heterotrimeric G-proteins in scar from failing hearts after MI is lacking.

In the current study, expression patterns of Gi-2{alpha} and Gs{alpha} were determined in myofibroblasts localized to the 8 week infarct scar, as well as in surviving myocytes in the border and noninfarcted myocardium in post-MI heart and respective control rat heart. Altered expression of Gi-2{alpha} and Gs{alpha} in these tissues was noted at the levels of both protein and mRNAs. Thus the relationship between the cellularity of the infarct scar and relative G-protein expression is explored.


   2 Methods
Top
 Abstract
 1 Introduction
 2 Methods
3 Results
 4 Discussion
 References
 
2.1 Experimental model, hemodynamic measurements, infarct size and collagen concentration
All experimental protocols for animal studies were approved by the Animal Care Committee of the University of Manitoba, Canada, following guidelines established by the Medical Research Council of Canada. MI was produced in male Sprague–Dawley rats (weighing 200–250 g) by surgical occlusion of the left coronary artery, as described previously, with minor modifications [19]. Left ventricular (LV) function and blood pressure of control and MI animals were measured 8 weeks following induction of MI, as described previously [19]. Infarct size was determined as described previously [12]and only animals with large infarcts (≥40% of the left ventricular free wall) were used in this study. Collagen content by way of hydroxyproline assay was measured according to the method of Chiariello et al. [20]. The border regions in experimental hearts were removed as a 2 mm area surrounding the transmural infarct scar, and these tissues were differentiated by a specific procedure using 2,3,5-triphenyltetrazolium chloride (TTC) stain. In addition, the separation of these tissues was confirmed by routine trichrome staining for total cardiac extracellular matrix and cardiac myocytes. Regions of homogenous matrix in the absence of myocytes was taken to be the infarct scar.

2.2 Immunofluorescence
A total of eighteen experimental rats were used in this assay including eight sham-operated animals and ten in the post-MI group. After anesthesia with ketamine–xylazine, animals were sacrificed by decapitation. Hearts were rapidly excised and immersed immediately in phosphate buffered saline (PBS) solution, pH 7.4. The viable left ventricle remote to the infarct and scar were immersed in OCT compound (Miles, Elkhart, IN, USA) and stored at –80°C. Serial cryostat sections, 7 µm thick, of the ventricular tissues were mounted on gelatin coated slides, fixed in 4% paraformaldehyde for 15 s and then washed 6x30 s in cold PBS. Slides were then submersed in 0.1% Triton X-100 (Sigma-Aldrich)–PBS solution for 15 s and washed again. A minimum of six sections from each ventricle of each group was processed and representative sections were chosen. Immunohistochemical staining was performed by the indirect immunofluorescence technique described in detail previously [21]. Rabbit polyclonal anti-Gs{alpha} subunit and anti-Gi-2{alpha} subunit (Calbiochem-Novabiochem, San Diego, CA, USA) at 0.4 mg/ml were diluted 1:500 with 1% BSA in PBS and applied as the primary antibodies. After overnight incubation at 4°C, the sections were washed (3x5 min) in PBS and incubated first with biotinylated anti-rabbit IgG secondary antibody (1:20 with 1% BSA in PBS) and subsequently incubated with FITC labeled streptavidin (1:20 with 1% BSA in PBS, Amersham Life Sciences, Canada) for 90 min each with washes between Ab incubations. Serial sections of sham-operated control ventricular tissue, viable (remnant), or infarct scar and remnant border myocardium were double-stained to identify vimentin-positive and myosin-positive cells in order to characterize the putative distribution of cardiac fibroblasts and cardiac myocytes, respectively, in these tissues. In order to double-immunostain the tissue parallel immunohistochemical experiments were carried out with anti-Gi-2{alpha} or anti-Gs{alpha} and anti-vimentin or anti-myosin. Monoclonal anti-vimentin clone #V9 (1:100 with 1% BSA in PBS, Sigma-Aldrich, Oakville, ON, Canada) or monoclonal anti-myosin MF-20 (1:100 with 1% BSA in PBS, Developmental Studies Hybridoma Bank, U of Iowa, Iowa City, IA, USA) were combined with either anti-Gi-2{alpha} or anti-Gs{alpha} primary antibodies. As above, a secondary biotinylated anti-rabbit IgG antibody was added to facilitate the adherence of streptavidin–FITC to the G-protein primaries. To distinguish vimentin or myosin staining vs. Gi-2{alpha} or Gs{alpha} staining, an anti-mouse linked Texas Red conjugate (1:20 with 1% BSA in PBS, Amersham Life Sciences) was added with streptavidin–FITC. Thus, conjugation of vimentin or myosin occurred with Texas Red while Gi-2{alpha} or Gs{alpha} primaries were conjugated with FITC. All washing and incubations herein were timed as described above. After washing (3x5 min) with cold PBS, all sections were immersed for 30 s in 10 µg/ml of Hoechst Dye 33342 in order to stain cellular nuclei, and then subjected to an additional wash (3x5 min) in cold PBS. Finally, the slides were mounted and coverslipped. Assessment of cell phenotype in cardiac tissue samples was carried out using three additional primary antibodies including {alpha}-smooth muscle actin ({alpha}SMA) (Sigma, mouse monoclonal at 1:400 dilution), smooth muscle myosin (SMM) (Sigma, human monoclonal at 1:250 dilution), factor VIII (von Willebrand factor) (Sigma, rabbit at 1:250 dilution). The presence of SMM and factor VIII was taken as a marker of smooth muscle cells and endothelial cells in tissue preparations, respectively. Cells that expressed {alpha}SMA but not SMM were designated myofibroblasts. The tissue sections were examined under a Nikon Labophot microscope equipped with epifluorescence optics and appropriate filters. The results were recorded by photography on Kodak TMAX 400 black and white film and Fujichrome Provia 400 color reversal film.

2.3 Western blot
Immunoreactive Gi-2{alpha} and Gs{alpha} proteins were detected using Western blot analysis. Cardiac tissues from sham-operated left ventricle, viable left ventricle, border area and scar were homogenized in 100 mM Tris (pH 7.4) containing 1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM phenymethylsulfonyl fluoride (PMSF), 4 µM leupeptin, 1 µM pepstatin A, and 0.3 µM aprotinin. Samples were sonicated for 3x5 s. Crude membrane and cytosolic fraction was isolated according to the method of Gettys et al. [22]. Briefly, samples were centrifuged for 3000xg at 4°C for 10 min to remove unbroken cells and nuclei. The supernatant was further subjected to centrifugation for 48 000xg for 20 min at 4°C. The subsequent crude membrane pellet was resuspended in the homogenizing buffer. Total protein concentration of membrane fractions was measured using the BCA method [23]. Prestained high molecular mass marker (Bio-Rad, Hercules, CA, USA) and 20 µg proteins from samples were separated on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Separated proteins were transferred on to 0.45 µM polyvinylidene difluoride (PVDF) membrane. PVDF membrane was blocked overnight at 4°C in Tris-buffered saline with 0.1% Tween-20 (TBS-T) containing 5% skim milk and probed with primary antibodies for Gi-2{alpha} or Gs{alpha} (Calbiochem-Novabiochem). Primary antibodies were diluted 1:1000 in TBS-T. Horseradish peroxidase (HRP)-labeled anti-rabbit IgG was diluted in 1:10 000 in TBS-T and used as secondary antibody. Gi2-{alpha} and Gs{alpha} were visualized by enhanced chemiluminescence (ECL) according to the manufacturers’ instruction (Amersham Life Science). Autoradiographs from Western blot analysis were quantified using a CCD camera imaging densitometer (Bio-Rad).

2.4 RNA extraction and Northern blot analysis
Total RNA was isolated from sham-operated, viable left ventricle, border and scar 8 weeks after operation by the method of Chomczynski and Sacchi [24]described previously [25]. A total of 12 animals were included in this assay. Recovered RNA was dissolved in diethyl pyrocarbonate (DEPC)-treated water and the concentration of nucleic acid was calculated from the absorbance at 260 nm prior to size fractionation. Twenty micrograms of total RNA was electrophoresed in a 1.2% agarose–formaldehyde gel and the fractionated RNA was transferred to a 0.45-µm positive charge-modified nylon membrane (Nytran Plus, Schleicher & Schuell). The RNA was covalently crosslinked to the membrane using UV radiation (UV Stratalinker 2400, Stratagene). Blots were prehybridized at 42°C for 16 h. Each membrane was hybridized with cDNA probes labeled (32P) using a random primer labeling kit (specific activity >109 cpm per µg DNA) at 42°C for 16–20 h (Gibco BRL). After washing, the membranes were exposed to X-ray film (Kodak X-OMAT) at –80°C with intensifying screens. The cDNA fragments for rat Gs{alpha} and Gi-2{alpha} were a generous gift from Randall R. Reed from the Howard Hughes Medical Institute Research Laboratories, New York and the cDNA fragment for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was obtained from the American Type Culture Collection (Rockville, MD, USA). Results of autoradiographs from Northern blot analysis were quantified by densitometry (Bio-Rad). The signals of specific mRNAs were normalized to those of GAPDH to normalize for differences in loading and/or transfer of mRNA.

2.5 Statistical analysis
All values are expressed as mean±S.E.M. One way analysis of variance (ANOVA) followed by Bonferroni’s test was used for comparing the differences among multiple groups (SigmaStat). Significant differences among groups were defined by a probability of less than 0.05.


   3 Results
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
4 Discussion
 References
 
3.1 General observations: left ventricular cardiac hypertrophy, fibrosis and heart failure
Experimental animals in this study were characterized by the presence of large MI, 42±3% of the total LV circumference (determined by planimetric techniques), with a total scar weight of 0.27±0.02 g, which was comparable to values reported earlier [21]. Hearts from experimental animals were characterized by significant cardiac hypertrophy which was reflected by increase in the mass of viable LV weight (0.89±0.02 g and 0.99±0.02 g, control and experimental groups, respectively; no scar) and also by the increased ratio of LV weight/body weight in experimental animals at 8 weeks (1.74±0.03 mg/g and 1.99±0.03 mg/g control and experimental groups, respectively). The incidence and magnitude of left ventricular hypertrophy noted in this study was comparable to our previous finding [26], as also was the averaged transmural scar weight (one measure of the extent of myocardial infarction) from experimental animals. Animals were assessed for left ventricular function at 8 weeks post-MI and these results revealed a significant increase in LV end-diastolic pressure (3.66±0.62 mm Hg and 12.46±1.2 mm Hg for control and experimental groups, respectively), decreased rates of contraction (+dP/dtmax: 5631±273 mm Hg/s and 4425±215 mm Hg/s for control and experimental groups, respectively) and relaxation (– dP/dtmax: 5416±215 mm Hg/s and 3921±234 mm Hg/s control and experimental groups respectively). Lung congestion in the experimental group was noted by the significantly elevated wet/dry weight ratio (3.49±0.21 vs. 4.92±0.22 control and experimental group values, respectively) [12]. Collagen concentration in the surviving myocardium (47.0±3.2 µg/mg dry wt.) and scar tissues (110.4±12.4 µg/mg dry wt.) was significantly higher than that of the sham control value (22.4±2.4 µg/mg dry wt.), and these results were similar to our previous finding [21, 27].

3.2 Localization of cardiac Gi2-{alpha} and Gs{alpha}
Gi-2{alpha} and Gs{alpha} protein distribution in 8-week experimental and age-matched control tissues was localized using immunofluorescence techniques. In the representative photograph (Fig. 1 top row), the staining pattern of immunoreactive Gs{alpha} is marked by bright clusters in the scar tissue as well as relatively bright staining in and among the cardiac myocytes near the border. Strong staining for Gi-2{alpha} (Fig. 1 bottom row) protein (relative to the sham section) was localized in scar tissue and in the myocytes that border the scar tissue (demarcated ‘border’), while slightly less staining was visualized in viable myocardium of these post-MI hearts. Slightly less immunoreactive Gs{alpha} protein was present in the sham-operated myocardium when compared to viable samples while the differences between sham sections and border or infarct scar are marked.


Figure 1
View larger version (85K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 (Top) Immunohistochemical stained sections showing Gs{alpha} in sham hearts, viable, border and scar tissue from post-myocardial infarction (MI, 8 weeks) respectively. Immunoreactive Gs{alpha} protein appears as brightly stained material. Magnification, x400. (Bottom) Immunohistochemical stained sections of Gi-2{alpha} with sham, viable, border and scar tissue from 8 week post-MI. Magnification x400.

 
Infarct scar and remnant myocardial tissue bordering the scar were double-stained to mark and localize immunoreactive Gi-2{alpha} and vimentin proteins in these tissues using polyclonal anti-Gi-2{alpha} (Fig. 2A) and monoclonal anti-vimentin antibodies (Fig. 2B). The identical section was also stained for cellular nuclei using Hoechst nuclear stain (Fig. 2C). Relatively bright staining of immunoreactive Gi-2{alpha} was observed in vimentin-positive cells in the nuclei-rich 8-week infarct scar. In a parallel series of experiments, double immunostaining of a different section of infarct scar and remnant myocardial tissues for Gi-2{alpha} proteins (Fig. 2D) and myosin (using monoclonal anti-myosin antibody — designated MF-20, Fig. 2E) revealed the predominance of cardiac myocytes in the tissues bordering the infarct scar. Surviving myocytes in the border region were marked by the presence of comparatively few nuclei (Fig. 2F) and by relatively bright anti-Gi- 2{alpha} staining. In both series, it is apparent that vimentin-rich cells are present in relatively large numbers in the infarct (Fig. 2B) and that the scar is not populated with myocytes (Fig. 2E). Vimentin-positive staining in the border area was present in relatively modest levels between the myocytes. In a different series of experiments, similar results were obtained in sections double immunostained with anti-Gs{alpha} and vimentin or myosin (data not shown). Finally, vimentin, myosin and nuclear staining was noted among frozen sections of sham, viable as well as scar and border tissues for the purpose of comparison (Fig. 3). From left to right, in samples of sham-operated and viable myocardium, vimentin staining (top row, apparent as brightly staining network) was prominent in the intramyocytic spaces. Cardiac myocytes in these sections (Fig. 3 middle row) stained positively for MF-20 (myosin) and these tissues appear as homogeneously brightly stained sections. The infarct scar proper stained negatively for myosin.


Figure 2
View larger version (93K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Infarct scar and adjacent remnant myocardium bordering the scar tissue, double-stained for either Gi-2{alpha} and vimentin (panels A, B and C) or Gi-2{alpha} and myosin (MF-20, panels D, E, and F) in 8-week post-MI rat heart. Panels A, B and C are a representative series of photographs of one common field, and panels D, E and F represent a different common field. S, infarct scar area; B, tissue bordering the scar. Panels A and D represent the pattern of cellular staining for immunoreactive Gi-2{alpha}, (bright staining pattern) while panels B and E depict the staining pattern of immunoreactive vimentin and myosin (bright staining pattern), respectively. Large arrow in panel B indicates a vimentin-positive cell, while the arrow in panel E indicates a cardiac myocyte. Panels C and F illustrate the specific staining pattern imparted by Hoechst nuclear stain (bright staining pattern). Magnification x400.

 

Figure 3
View larger version (81K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Representative frozen cardiac tissue sections stained for immunoreactive vimentin, myosin (MF-20), and cell nucleus (Hoechst) staining patterns in sham, viable as well as border and scar tissue from animals at 8 weeks after myocardial infarction. M, depicts the myocytes occupying the border region; S, depicts the scar region of the field. The sections shown within the tissue regions are not derived from serial preparation. Magnification, x400.

 
As immunoreactive vimentin stains not only fibroblasts, but also myofibroblasts, endothelial cells and smooth muscle cells, we carried out a series of immunohistochemical experiments to determine the phenotype distribution of these cells in the infarct scar. Representative scar sections were stained with primary antibodies of {alpha}SMA, factor VIII (von Willebrand factor), and SMM and the results are shown in Fig. 4. In 8-week infarct scar tissue, significant {alpha}SMA staining (Fig. 4A) was noted in parallel to relatively low expression of SMM (Fig. 4C) and we took these combined findings to reflect the appearance of myofibroblasts. In remnant cardiac tissue remote to the scar, {alpha}SMA staining was localized to the perivascular space of vessels (data not shown). Staining of 8 week infarct scar sections with primary antibody against factor VIII was carried out to mark endothelial cell distribution. This tissue was characterized by a negative staining pattern for this antibody (Fig. 4B).


Figure 4
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Characterization of myofibroblasts in the infarct scar at 8 weeks after myocardial infarction. (A) Frozen sections of infarct scar were stained with immunoreactive {alpha} smooth muscle actin ({alpha}SMA); (B) factor VIII (von Willebrand factor); (C) smooth muscle myosin (SMM). Conditions for fluorescence were optimized in each field. The relatively low level of specific staining for smooth muscle myosin and negative staining pattern for factor VIII indicate minimal numbers of smooth muscle cells and endothelial cells in the 8-week infarct scar. Magnification, x400.

 
3.3 Changes in cardiac Gi-2{alpha} and Gs{alpha} protein abundance in hearts with MI
Quantitative assessment of cardiac membrane Gs{alpha} protein expression in control and left ventricular tissues of 8 week post-MI rats was carried out using Western blot techniques. Fig. 5A provides a representative autoradiograph illustrating the presence of the characteristic 52 kDa and 45 kDa bands for Gs{alpha} protein. The results indicate that Gs{alpha} was increased significantly by approximately 3.4- and 8.2-fold in border and scar tissue, respectively, compared to band intensity from control animals. Also, there was no significant alteration in the Gs{alpha} band intensity of samples from viable left ventricle compared to control. (Fig. 5B). Fig. 6A provides a representative autoradiograph illustrating the presence of the characteristic 40 kDa band for Gi-2{alpha}. The data indicates that Gi-2{alpha} increased 4.5- and 5.0-fold in border and scar tissue respectively, a significant increase when compared to control and viable tissue. We also observed a significant increase in the Gi-2{alpha} band intensity in samples from post-MI viable left ventricle compared to sham operated control (Fig. 6B).


Figure 5
View larger version (41K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 (A) Representative Western blot indicating specific 52 kDa and 45 kDa bands for Gs{alpha} in samples from sham, viable, border and infarct scar tissue from 8-week experimental animals. Lanes 1 and 2 are sham, lanes 3 and 4 are viable LV, lanes 5 and 6 represent border tissue, and lanes 7 and 8 are infarct scar. (B) Quantified data for Gs{alpha} protein concentration in sham, viable, border and infarct scar tissue, respectively. The control group is sham-operated, age-matched to the 8-week post-MI experimental group. The data depicted is the mean+S.E.M. of six experiments. *P<0.05, +P<0.05 and #P<0.05 vs. sham, viable and border sample values, respectively.

 

Figure 6
View larger version (48K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 (A) Representative Western blot indicating the specific 40 kDa band for Gi-2{alpha} in samples from sham, viable, border and scar tissue from 8-week experimental animals. Lanes 1 and 2 are sham, lanes 3 and 4 are viable LV, lanes 5 and 6 represent border tissue, and lanes 7 and 8 are infarct scar samples. (B) Quantified data for Gi-2{alpha} protein concentration in sham, viable, border and infarct scar tissue respectively. The control group is sham-operated, age matched to the 8-week post-MI experimental group. The data depicted is the mean±S.E.M. of six experiments. *P<0.05 and +P<0.05 vs. sham and viable sample values, respectively.

 
3.4 Alteration of steady-state mRNA abundance of cardiac Gs{alpha} and Gi-2{alpha}
We addressed mRNA abundance of the cardiac Gs{alpha} and Gi-2{alpha} genes in tissues taken from various left ventricular regions of rats 8 weeks post-MI. Fig. 7A shows a representative Northern blot with autoradiographic bands for Gs{alpha}, Gi-2{alpha} and GAPDH mRNAs from left ventricular samples of sham, viable, and scar tissues respectively. Gs{alpha} is identified by a 1.8 kb transcript, Gi-2{alpha} by a 1.7 kb transcript and GAPDH by a 1.4 kb transcript, in our blots (Fig. 7A). Estimation of Gs{alpha} mRNA abundance was calculated by the ratio of Gs{alpha} to GAPDH signal; this ratio was significantly increased only in the scar tissue (1.28-fold) region of the left ventricle compared to control (Fig. 7B). Gi-2{alpha} mRNA calculated by the ratio of Gi-2{alpha} to GAPDH signal was significantly increased in both the viable and scar tissues (1.24- and 1.85-fold respectively, Fig. 7C). Furthermore, mRNA signal ratios from scar tissue samples were significantly increased (1.5-fold) in Gi-2{alpha} mRNA values compared to those obtained from post-MI viable LV (Fig. 7C).


Figure 7
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 (A) Representative autoradiograph from Northern blot analysis showing a 1.8 kb band of Gs{alpha} and a 1.7 kb band of Gi-2{alpha} in sham (lanes 1 and 2), viable (lanes 3 and 4), and scar tissues (lanes 5 and 6) from 8 week post-MI rat hearts. Hybridization of fractionated total RNA with cDNA probes for Gs{alpha} and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as well as for Gi-2{alpha} and GAPDH indicates relative steady-state mRNA levels for each gene tested. (B) Quantified data of Gs{alpha}/GAPDH in sham, viable, and scar tissue, respectively. The data depicted is the mean±S.E.M. of six experiments. *P<0.05 and +P<0.05 vs. sham and viable sample values, respectively. (C) Quantified data of Gi-2{alpha}/GAPDH in sham, viable, and scar tissue respectively. The data depicted is the mean±S.E.M. of six experiments. *P<0.05 and +P<0.05 vs. sham and viable sample values, respectively.

 

   4 Discussion
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
References
 
The experimental animals in the 8-week post-MI group employed for this study were in moderate heart failure attended by decreased LV peak systolic pressure and elevated LV end-diastolic pressure, decreased ±dP/dtmax, and by the presence of pulmonary congestion [19]. These hemodynamic abnormalities were not associated with overt dyspnea, cyanosis, or marked lethargy in the experimental group. The incidence of moderate heart failure was verified in the current study to provide a basis for objective comparison of cardiac dysfunction with changes in ventricular expression and function of target genes and gene products, respectively.

Activation of the sympathetic and renin–angiotensin systems as well as the enhanced release of various growth factors are associated with the development of heart failure [28, 29]. Specifically, angiotensin II and norepinephrine are trophic hormones that may stimulate the development of cardiac fibrosis and myocyte hypertrophy [30]. It has been shown that angiotensin II concentration is increased in infarct scar [31]and that AT1 receptor expression is elevated in acute infarcted myocardium (7 days) and in 8-week infarct scars [32]. In mammalian cells, some of the specific actions of angiotensin II are known to be mediated by the activation of Gi{alpha} proteins [33]. Furthermore, it is pointed out that neurohormonal stimulation of cardiac cells may be linked to the altered expression of cardiac G-proteins themselves. In this regard, the role of norepinephrine was examined in vitro using cultured neonatal rat cardiac myocytes, and it was revealed that exposure of these cells to this hormone resulted in increased expression of Gi{alpha} [34]. We suggest that changes in the receptor density of multiple neurohormonal factors may lead to increased downstream Gi-2{alpha} and Gs{alpha} expression, and further experiments are necessary to test this hypothesis. Facilitation among myocardial angiotensin II and sympathetic hormone signaling systems is known to exist beyond the post-receptor level. In the human heart, low concentrations of angiotensin II may facilitate norepinephrine release thereby potentiating the effect of this catecholamine [35]. The expression and function of Gi-2{alpha} and Gs{alpha} may play a role in mediating the development of heart failure [30], and it is pointed out that heart failure is associated with altered expression of Gi-2{alpha} and/or Gs{alpha} [36, 37]. The current results extend this finding as it supports the hypothesis that cardiac nonmyocytes, and particularly myofibroblasts per se, may also be involved in altered expression of cardiac Gs{alpha} and Gi-{alpha}2 in heart failure due to myocardial infarction.

Fibrillar collagens are major cardiac extracellular matrix proteins which are synthesized and secreted solely by cardiac fibroblast and myofibroblast cells, and altered deposition of the cardiac matrix has been suggested to contribute to the pathogenesis of heart failure [26, 38]. Biochemical assays for cardiac hydroxyproline concentration and collagen crosslinking provide evidence for progressive fibrillar collagen deposition and altered post-translational modification of fibrillar collagens, respectively, beyond 6 weeks in both remnant heart and infarct scar from post-MI rats [39, 40]. Recently, the role of fibroblasts in the infarct scar from chronic post-MI animals was investigated and it was suggested that chronic wound healing of the scar itself may contribute to ongoing remodeling of scar morphology [12]. As vimentin is expressed by cardiac nonmyocytes but not by myocytes, we used immunoflourescent detection of this protein as a first step to highlight the distribution of these broad classes of cells in infarct scars in experimental animals. Double immunostaining of scar tissue (with vimentin and Gs{alpha} or Gi-2{alpha} antibodies) indicated that immunoreactive Gs{alpha} or Gi-2{alpha} co-localized with the vimentin-positive cells. Phenotyping of the vimentin-positive cells of the infarct scar revealed significant positive staining for {alpha}SMA but not for SMM and that endothelial cells were not present in significant numbers in 8 week post-MI infarct scar. Myofibroblasts are larger cells than usual interstitial fibroblasts, have indented nuclei reminiscent of any other contractile cell, and are characterized by their expression of {alpha}SMA [41]. Our findings indicate that the major cell type populating the 8 week infarct scar was myofibroblasts, which is consistent with previous results [10, 13]. Examination of the infarct scar for immunoreactive myosin revealed a negative staining pattern (Figs. 2 and 3Go) in this tissue. While this result suggests that cardiac myocytes may be absent in the infarct scar, it does not rule out the possible presence of myocytes in the infarct zone that do not react with MF-20 anti-myosin. Anti-myosin (MF-20) positive myocytes located in the border tissue stained brightly after exposure to Gi-2{alpha} or Gs{alpha} primary antibodies, and this area was characterized by comparatively modest vimentin-positive or {alpha}SMA staining. On the other hand, we suggest that myofibroblasts are a major cellular source for expression of Gi-2{alpha} or Gs{alpha} proteins at the site of the infarct scar. Phenotypically transformed fibroblasts i.e., myofibroblasts are believed to play a role in the pathogenesis of heart failure [26, 38]. Fibroblasts, unlike myocytes, retain phenotypic plasticity, may re-enter the cell cycle, and participate in autocrine/paracrine hormonal signaling [42–44]. Cardiac myofibroblasts are the predominant cell type in post-MI scar tissue in both human and rat [10, 45]and have been suggested to mediate wound contraction during later stages of healing [46].

Although the precise functional significance of increased Gi-2{alpha} or Gs{alpha} protein expression in myofibroblasts from the infarct scar is not yet clear, there is mounting evidence to indicate the involvement of these G-proteins in fibroblast proliferation, an event associated with enhanced matrix deposition. Specifically, it was shown that both Gi-2{alpha} and Gs{alpha} play a role in the proliferation of 3T3 fibroblasts and GH3 cells (mutated cells constitutively activating Gs{alpha}) respectively [47, 48]. Crouch and coworkers found that Gi may regulate mitosis but not DNA synthesis in growth factor-activated 3T3 fibroblasts and that Gi may stimulate this event by translocating to the nucleus of mitotic cells [47]. Insulin growth factor-I stimulation of NFB4 cells which chronically overexpress Gi-2{alpha} protein, led to the induced activation of ERK-1 and -2 (extracellular signal-regulated kinases) which are involved in cellular proliferation and differentiation [49]. Furthermore, we previously observed that Gq{alpha} and PLC-β proteins are activated in cells localized in the 8-week infarct scar [12]. The present study revealed that Gi-2{alpha} and Gs{alpha} are also activated in myofibroblasts present in this tissue. Taken together, these data suggest that myofibroblasts in the 8-week infarct scar are not quiescent, and that the period of infarct healing extends beyond the classic 3-week duration [50]in this experimental model. Although we also noted increased expression of Gi-2{alpha} and Gs{alpha} proteins in cardiac myocytes remote to the site of infarction in the border and viable regions, the significance of these alterations is unknown. Nevertheless, elevated Gi-{alpha}2 in left ventricular myocardium from failing hearts has been noted in clinical and experimental studies [4, 6, 36], which was associated with disrupted Ca2+ homeostasis and loss of normal cardiac contractile function. The concomitant increase in Gs{alpha} expression observed in border myocytes may represent a reactive attempt by the myocyte cell to potentiate adenylyl cyclase function and thereby balance the influence of increased Gi-{alpha}2 expression, with respect to beat-to-beat myocyte Ca2+ handling in these cells.

In conclusion, the present study has demonstrated that Gi-2{alpha} and Gs{alpha} proteins are highly expressed in cardiac myofibroblasts localized to the infarct scar as well as in myocytes bordering the scar in heart failure subsequent to MI. As myofibroblasts and fibroblasts in the infarct scar are responsible for the synthesis and deposition of collagen type I and III, these cells are crucial for the chronic remodeling of the scar. Therefore, it is suggested that Gi-2{alpha} and Gs{alpha} may play an important role in the evolution of scar remodeling in post-MI rat heart and that these events may be linked to the development of heart failure. Further investigations to elucidate the implications of increased G-protein expression in the infarct scar of post-MI hearts are warranted.

Time for primary review 27 days.


   Acknowledgements
 
This study was supported by funding from the Medical Research Council of Canada. HJ carried out these studies as a recipient of a Manitoba Health Research Council Studentship. We thank Tracy Scammell-La Fleur and Shufang Zhao for their excellent technical assistance and their role in the completion of this work. IMCD is a scholar of the Medical Research Council of Canada/PMAC health program with funding provided by Astra Pharma, Canada.


   References
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 

  1. Bristow M.R., Ginsburg R., Minobe W., et al. Decreased catecholamine sensitivity and beta-adrenergic-receptor density in failing human hearts. New Engl J Med (1982) 307:205–211.[Abstract]
  2. Kaura D., Takeda N., Sethi R., Wang X., Nagano M., Dhalla N.S. Beta-adrenoceptor mediated signal transduction in congestive heart failure in cardiomyopathic (UM-X7.1) hamsters. Mol Cell Biochem (1996) 157:191–196.[ISI][Medline]
  3. Bristow M.R., Anderson F.L., Port J.D., et al. Differences in beta-adrenergic neuroeffector mechanisms in ischemic versus idiopathic dilated cardiomyopathy. Circulation (1991) 84:1024–1039.[Abstract/Free Full Text]
  4. Bohm M., Eschenhagen T., Gierschik P., et al. Radioimmunochemical quantification of Gi alpha in right and left ventricles from patients with ischemic and dilated cardiomyopathy and predominant left ventricular failure. J Mol Cell Cardiol (1994) 26:133–149.[CrossRef][ISI][Medline]
  5. Neumann J., Schmitz W., Scholz H., Meyerinck L.V., Doring V., Kalmar P. Increase in myocardial Gi-proteins in heart failure. Lancet (1988) 2:936–937.[ISI][Medline]
  6. Eschenhagen T., Mende U., Nose M., et al. Increased messenger RNA level of the inhibitory G protein alpha subunit Gi alpha-2 in human end-stage heart failure. Circ Res (1992) 70:688–696.[Abstract/Free Full Text]
  7. Yamamoto J., Ohyanagi M., Morita M., Iwasaki T. Beta-adrenoceptor-G protein-adenylate cyclase complex in rat hearts with ischemic heart failure produced by coronary artery ligation. J Mol Cell Cardiol (1994) 26:617–626.[CrossRef][ISI][Medline]
  8. Bohm M., Gierschik P., Jakobs K.H., et al. Increase of Gi alpha in human hearts with dilated but not ischemic cardiomyopathy. Circulation (1990) 82:1249–1265.[Abstract/Free Full Text]
  9. Bohm M., Kirchmayr R., Erdmann E. Myocardial Gi alpha-protein levels in patients with hypertensive cardiac hypertrophy, ischemic heart disease and cardiogenic shock. Cardiovasc Res (1995) 30:611–618.[Abstract/Free Full Text]
  10. Cleutjens J.P.M., Verluyten M.J.A., Smits J.F.M., Daeman M.J.A.P. Collagen remodeling after myocardial infarction in the rat heart. Am J Pathol (1995) 147:325–338.[Abstract]
  11. Padua R.R., Kardami E. Increased basic fibroblast growth factor (bFGF) accumulation and distinct patterns of localization in isoproterenol-induced cardiomyocyte injury. Growth Factors (1993) 8:291–306.[ISI][Medline]
  12. Ju H., Zhao S., Tappia P., Panagia V., Dixon I.M.C. Expression of Gq{alpha} and PLC-β in scar and border tissue in heart failure due to myocardial infarction. Circulation (1998) 97:892–899.[Abstract/Free Full Text]
  13. Vracko R., Thorning D. Contractile cells in the rat myocardial scar tissue. Lab Invest (1991) 65:214–227.[ISI][Medline]
  14. Holmes J.W., Nunez J.A., Covell J.W. Functional implications of myocardial scar structure. Am J Physiol (1997) 272:H2123–H2130.[ISI][Medline]
  15. Pfeffer M.A., Braunwald E. Ventricular remodeling after myocardial infarction. Experimental observations and clinical implications. Circulation (1990) 81:1161–1172.[Abstract/Free Full Text]
  16. Weber K.T., Sun Y., Katwa L.C. Wound healing following myocardial infarction. Clin Cardiol (1996) 19:53–62.
  17. Pfeffer J.M., Fischer T.A., Pfeffer M.A. Angiotensin-converting enzyme inhibition and ventricular remodeling after myocardial infarction. Annu Rev Physiol (1995) 57:805–826.[CrossRef][ISI][Medline]
  18. Cohn J.N., Bristow M.R., Chien K.R., et al. Report of the National Heart, Lung, and Blood Institute special emphasis panel on heart failure research. Circulation (1997) 95:766–770.[Free Full Text]
  19. Dixon I.M.C., Lee S.L., Dhalla N.S. Nitrendipine binding in congestive heart failure due to myocardial infarction. Circ Res (1990) 66:782–788.[Abstract/Free Full Text]
  20. Ju H., Zhao S., Davinder S.J., Dixon I.M.C. Effect of AT1 receptor blockade on cardiac collagen remodeling after myocardial infarction. Cardiovasc Res (1997) 35:223–232.[Abstract/Free Full Text]
  21. Chiariello M., Ambrosio G., Cappelli Bigazzi M., Perrone Filardi P., Brigante F., Sifola C. A biochemical method for the quantification of myocardial scarring after experimental coronary occlusion. J Mol Cell Cardiol (1986) 18:283–290.[CrossRef][ISI][Medline]
  22. Gettys T.W., Sheriff Carter K., Moomaw J., Taylor I.L., Raymond J.R. Characterization and use of crude alpha-subunit preparations for quantitative immunoblotting of G-proteins. Anal Biochem (1994) 220:82–91.[CrossRef][ISI][Medline]
  23. Smith P.K., Krohn R.I., Hermanson G.T., et al. Measurement of protein using bicinchoninic acid. Anal Biochem (1985) 150:76–85.[CrossRef][ISI][Medline]
  24. Chomczynski P., Sacchi N. Single-step method of TNA isolation by acid quanidinium thiocyanate–phenol–chloroform extraction. Anal Biochem (1987) 162:156–159.[ISI][Medline]
  25. Ju H., Scammell-La Fleur T., Dixon I.M.C. Altered mRNA abundance of calcium transport genes in cardiac myocytes induced by angiotensin II. J Mol Cell Cardiol (1996) 28:1119–1128.[CrossRef][ISI][Medline]
  26. Dixon I.M.C., Ju H., Jassal S.J., Peterson D.J. Effect of ramipril and losartan on collagen expression in right and left heart after myocardial infarction. Mol Cell Biochem (1996) 165:31–45.[ISI][Medline]
  27. Pelouch V., Dixon I.M.C., Golfman L., Beamish R.E., Dhalla N.S. Role of extracellular matrix proteins in heart function. Mol Cell Biochem (1993) 129:101–102.[CrossRef][ISI][Medline]
  28. Colucci W.S. Braunwald E. Pathophysiology of heart failure. In: Braunwald E, editor. Heart disease. Philadelphia:W.B. Sanders, 1997:394–420.
  29. Eichhorn E.J., Bristow M.R. Medical therapy can improve the biological properties of the chronically failing heart. A new era in the treatment of heart failure. Circulation (1996) 94:2285–2296.[Abstract/Free Full Text]
  30. Brodde O.E., Michel M.C., Zerkowski H.R. Signal transduction mechanisms controlling cardiac contractility and their alterations in chronic heart failure. Cardiovasc Res (1995) 30:570–584.[Free Full Text]
  31. Yamagishi H., Kim S., Nishikimi T., Takeuchi K., Takeda T. Contribution of cardiac renin–angiotensin system to ventricular remodeling in myocardial-infarction rats. J Mol Cell Cardiol (1993) 23:1369–1380.
  32. Lefroy D.C., Wharton J., Crake T., et al. Regional changes in angiotensin II receptor density after experimental myocardial infarction. J Mol Cell Cardiol (1996) 28:429–440.[CrossRef][ISI][Medline]
  33. Hayashida W., Horiuchi M., Dzau V.J. Intracellular third loop domain of angiotensin II type 2 receptor. Role in mediating signal transduction and cellular function. J Biol Chem (1996) 271(36):21985–21992.[Abstract/Free Full Text]
  34. Reithmann C., Gierchik P., Muller U., Werdan K., Jakobs K.H. Pseudomonas exotoxin A prevents β-adrenoceptor-induced up-regulation of Gi protein {alpha}-subunits and adenylyl cyclase desensitization in rat heart muscle cells. Mol Pharmocol (1990) 37:631–638.[Abstract]
  35. Rump L.C., Schwertfeger E., Schaible U., Fraedich G., Schollmeyer P. β2-Adrenergic receptor and angiotensin II receptor modulation of sympathetic neurotransmission in human atria. Circ Res (1994) 74:434–440.[Abstract/Free Full Text]
  36. Shi B., Heavner J.E., McMahon K.K., Spallholz J.E. Dynamic changes in Gi{alpha}-2 levels in rat hearts associated with impaired heart function after myocardial infarction. Am J Physiol (1995) 269:H1073–H1079.[ISI][Medline]
  37. Nash J.A., Hammond H.K., Saffitz J.E. Subcellular compartmentalization of Gs{alpha} in cardiac myocytes and its redistribution in heart failure. Am J Physiol (1996) 271:H2209–2217.[ISI][Medline]
  38. Schieffer B., Wirger A., Meybrunn M., et al. Comparative effects of chronic angiotensin-converting enzyme inhibitors and angiotensin II type 1 receptor blockade on cardiac remodeling after myocardial infarction in the rat. Circulation (1994) 89:2273–2282.[Abstract/Free Full Text]
  39. Jugdutt B.I., Amy R.W.M. Healing after myocardial infarction in the dog: changes in infarct hydroxyproline and topography. J Am Coll Cardiol (1986) 7:91–102.[Abstract]
  40. McCormick R.J., Musch T.I., Bergman B.C., Thomas D.P. Regional differences in LV collagen accumulation and mature cross-linking after myocardial infarction in rats. Am J Physiol (1994) 266:H354–H359.[ISI][Medline]
  41. Gabbiani G., Hirschel B.J., Ryan G.B., Statkov P.R., Majno G. Granulation tissue as a contractile organ. A study of structure and function. J Exp Med (1972) 135:719–734.[Abstract]
  42. Sappino A.P., Schurch W., Gabbiani G. Differentiation repertoire of fibroblastic cells: expression of cytosketetal proteins as marker of phenotypic modulations. Lab Invest (1990) 63:144–161.[ISI][Medline]
  43. Weber K.T. Cardiac interstitium in health and disease: the fibrillar collagen network. J Am Coll Cardiol (1989) 13:1637–1652.[Abstract]
  44. Eghbali M., Tomek R., Sukhatme V.P., Woods C., Bhambi B. Differential effects of transforming growth factor-beta 1 and phorbol myristate acetate on cardiac fibroblasts. Regulation of fibrillar collagen mRNAs and expression of early transcription factors. Cir Res (1991) 69:483–490.[Abstract/Free Full Text]
  45. Willems I.E.M.G., Havenith M.G., DeMey J.G.R., Daemen M.J.A.P. The {alpha}-smooth muscle actin-positive cells in healing human myocardial scars. Am J Pathol (1994) 145:868–875.[Abstract]
  46. Majno G. The story of the myofibroblasts. Am J Surg Pathol (1979) 3:535–542.[ISI][Medline]
  47. Crouch M.F., Simson L. The G-protein G(i) regulates mitosis but not DNA synthesis in growth factor-activated fibroblasts: a role for the nuclear translocation of G(i). FASEB J (1997) 11(2):189–198.[Abstract]
  48. Ham J., Ivan M., Wynford-Thomas D., Scanlon M.F. GH3 cells expressing constitutively active Gs alpha (Q227L) show enhanced hormone secretion and proliferation. Mol Cell Endocrinol (1997) 127(1):41–47.[CrossRef][ISI][Medline]
  49. Sarbassov D.D., Jones L.G., Peterson C.A. Extracellular signal-regulated kinase-1 and -2 respond differently to mitogenic and differentiative signaling pathways in myoblasts. Mol Endocrinol (1997) 11:2038–2047.[Abstract/Free Full Text]
  50. Fishbein M.C., Maclean D., Maroko P.R. Experimental myocardial infarction in the rat: qualitative quantitative changes during pathologic evolution. Am J Pathol (1978) 90:57–70.[Abstract]

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 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
Google Scholar
Right arrow Articles by Peterson, D. J.
Right arrow Articles by Dixon, I. M.C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Peterson, D. J.
Right arrow Articles by Dixon, I. M.C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?