|Home | About | Journals | Submit | Contact Us | Français|
We hypothesize that persistent alterations in molecular signaling may drive recurrent pathologic remodeling even after the reduction of mechanical stress achieved via surgical ventricular reconstruction (SVR). We developed a murine model of SVR that would facilitate molecular analysis of the post-SVR myocardium and allow future exploitation of genetic models.
C57/B6 mice underwent coronary artery ligation. For SVR at 4 weeks post-MI, a purse string suture (7-0 polypropylene) achieved at least partial exclusion of the apical aneurysm. Serial echocardiography was correlated to measurements of apoptosis and to Western blot analysis of key signaling cascades.
An immediate 21.7 ± 2.6% improvement in fractional shortening (FS) was seen in the remaining myocardium after SVR. Reduction in LV volume and improved function persisted at 1 week, but recurrent dilatation at 4 weeks (LVEDV of 63.5 ± 2.5 vs. 42.1± 5.4 uL immediately after SVR, P < 0.05) was associated with a loss of functional improvement (FS 41.2 ± 2% vs. 46 ± 0.9%, P < 0.01). At 1 week after SVR, there was a transient reduction in myocardial apoptosis. A steady reduction in cardioprotective myocardial Akt activation, however, was not impacted by SVR.
This murine model recapitulates both the immediate benefits of SVR and the longer-term recurrence of dilated cardiomyopathy seen previously in some animal models and human studies. Early analysis has begun to implicate persistent signaling changes in the post-MI myocardium that may be responsible for recurrent dilatation after SVR, and that may become targets for combined surgical and molecular interventions.
Current management of congestive heart failure (CHF) primarily
in slows progression of disease. Surgical treatments have been proposed to improve the biophysics of the dilated left ventricle (LV) and to reduce the stimulus for ongoing pathologic remodeling . Early studies of surgical ventricular reconstruction (SVR), and particularly of aneurysm resection, have demonstrated both an improvement in ventricular function and CHF symptoms. [2-8]. These studies, however, have also demonstrated recurrent LV dilatation, suggesting that a reduction in mechanical stress alone does not address all of the changes in myocardial biology that drive the remodeling process [4,8,9].
Much has been learned regarding molecular signaling in the myocardium. A number of pathways have been identified that drive both ventricular hypertrophy as well as a transition to dilated cardiomyopthay . Genetic mouse models have revealed changes in even a single enzyme that can influence macroscopic ventricular structure and function [12-16]. In particular, phosphotidyl inositol-3 kinase (PI3K)/Akt signaling can protect cardiac myocytes from apoptosis [11,12], although the pathway has been associated with both pathologic hypertrophy  and adaptive, physiologic hypertrophy . Mitogen activated protein (MAP) kinases also help determine cardiac myocyte and ventricular fate; p38 MAP kinase and JNK have, in some studies, been associated with cardiac cell apoptosis and ventricular dilatation [14,15], while the MAP kinase ERK may protect these cells .
We describe a murine model of post-MI heart failure and SVR that may help elucidate the molecular mechanisms of recurrent ventricular dilatation. We hypothesize that critical changes in myocardial signaling persist despite ventricular reconstruction (Figure 1), and that hybrid molecular and surgical interventions may sustain and even enhance benefit. We measured the activation of key elements of the PI3K/Akt and MAP kinase cascades in post-MI hearts subjected to either SVR or sham re-operation.
Male C57/B6 mice (25 gm) were anesthetized with 1.5% inhaled isoflurane at 115 breaths/min. A left lateral thoracotomy incision at the fourth interspace exposed the left ventricle and atrial appendage. Although mouse coronary anatomy differs from that of the typical human heart, there is generally at least one large branch of the left main coronary on the anterior wall of the left ventricle. A 7-0 polypropylene suture was used to ligate this vessel, which, for convenience, we refer to as the left anterior descending (LAD), approximately 1/3 the distance from the base to the apex. Preliminary studies documented a reproducible infarction of 30-40% of the left ventricle as assessed by computerized image analysis (Scion Corp., Frederick, MD). A sodium hyaluronate-carboxymethyl cellulose membrane patch (Seprafilm®, Genzyme) was placed between the left ventricle and chest wall to facilitate reoperation. SVR 4 weeks post-infarction involved placement of a pursestring suture (7-0 polypropylene) around the infarcted/aneurismal region. Sham operation consisted simply of redo thoracotomy. A total of 56 mice with LAD ligation were randomly divided into: 1) sham control (20 mice); and 2) SVR (36 mice). Mortality from thoracotomy was <10%, and ~20% with SVR. Significant morbidity was limited primarily to failure to thrive after MI. Diastolic arrest was induced high potassium injection at the time of sacrifice. All procedures conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996), and were approved by the Institutional Animal Care and Use Committee of the San Francisco Veterans Affairs Medical Center.
Transthoracic echocardiography was performed on awake, minimally restrained mice using an Acuson Sequoia 512 machine and a 13-MHz probe. A two-dimensional short-axis view of the left ventricle was obtained at the level of the papillary muscles, and M-mode tracings were recorded. LV fractional shortening was calculated as (LV diastolic dimension – LV systolic dimension [17,18]. Measurements were made prior to and immediately post SVR, and at 1, 2, and 4 weeks.
Mouse hearts arrested in diastole were pressure-fixed in formalin and paraffin-embedded. Serial sections (5 μm) were stained with Gomori Trichrome to identify fibrillar, collagen-rich scar. For apoptosis, thin sections underwent TUNEL staining (Chemicon, Temecula, CA) and apoptotic indices were calculated in 3 sections/heart. Ligase staining (ApopTag. Peroxidase ISOL Kit, Chemicon) of selected adjacent sections was used to confirm the specificity of TUNEL.
Hearts were divided into infarct/borderzone (defined as the translucent infarct tissue plus the immediate ~1 mm of surrounding myocardium) and remote, uninfarcted myocardium, and were homogenized in a lysis buffer (0.13M KCl, 1mM EDTA, 1mM EGTA, 1mM Na3(VO4), 5mM NaF, 20mM HEPES and Protease inhibitor cocktail tablet (Roche Diagnostics, Mannheim, Germany). BCA Protein Assay Reagent Kit (Pierce, Rockford, IL) was used to measure protein concentration. Equal amounts of protein were separated by NuPAGE Novex Bis-Tris Gels (Invitrogen) and transferred to PVDF membranes (Invitrogen). Blots were probed with antibodies specific for phospho- and total Akt, , phospho- and total ERK, phospho- and total p38, phospho- and total p70S6 kinase, Bcl-2, Bax, or BAD (Cell Signaling, Beverly, MA) with appropriate horseradish peroxidase-conjugated antibodies as secondary antibodies (Cell Signaling). SuperSignal West Femto Maximum Sensitivity substrate (Pierce) was used for visualization. Density analysis using Scion Image (Scion Corporation, Frederick, MD), was used to determine relative protein quantitification.
Data are reported as mean ± SEM. Comparisons between groups were made using Student's t test and ANOVA. A P-value of less than 0.05 was considered to denote statistical significance.
Trichrome staining of a representative mouse heart immediately after SVR demonstrated the exclusion of the majority of post-infarct LV aneurysm, with subsequent alteration of LV chanber size and geometry (Figure 2). The inability to access the septal component of an infarct and excessive mortality with overly-aggressive purse string reductions in LV size resulted in only partial aneurysm exclusion in many animals Figure 2
Echocardiography confirmed similar left ventricular end diastolic volume (LVEDV), end systolic volume (LVESV) and fractional shortening (FS) between the groups that were subsequently randomized to either SVR or sham control at 4 weeks after LAD ligation. Immediately after SVR we observed an anticipated reduction in LV dimensions (Figure 3A) and LVEDV (42.1 ± 5.4 vs. 69.0± 5.8 μL, P < 0.01, Figure 3B). No differences were observed in LV volumes after sham operation. Serial echocardiography demonstrated that the LVEDV gradually increased after SVR (Figure 3B), and that LVEDV in the post-SVR animals was no longer statistically different from that in the sham group, or pre-SVR levels, by 4 weeks after reoperation.
Echocardiography was chosen as a primary functional analysis since it may be the only modality capable of measuring cardiac function in awake, minimally restrained mice, thereby eliminating the profound impact of general anesthesia on cardiac function in rodents. Furthermore, it allowed serial measurements in the same animals, felt to be critical for this study of surgical reconstruction. Focusing the analysis of FS at the level of the more easily identifiable mitral apparatus helped standardize the echocardiographic imaging, which was carried out in a blinded fashion. Echocardiographic evidence of improved hemodynamic function immediately after SVR was corroborated in a small number of hearts via ex vivo Langendorff preparation (data not shown).
Echocardiography revealed an improvement in ejection fraction (EF) after SVR that corresponded to the reduction in LV chamber size (74.5 ± 0.6% vs. 66.9 ± 1.4% before SVR, P < 0.01). While a reduction in volume can directly lead to an improvement in EF, even more striking was an improvement in FS (21.7 ± 2.6% improvement over pre-SVR) measured in the myocardial wall at the level of the proximal mitral apparatus, distant from the region of the infarct (Figure 4). These data suggest that an immediate reduction in LV wall stress allowed a functional improvement in more global myocyte contractility. By 4 weeks after SVR, however, serial echocardiography indicated a significant decline in both EF and FS (Figures 3B, ,4).4). Parameters of LV function in the sham controls remained relatively stable during this time period.
Previous studies have suggested that LV dilatation is mediated at least in part through the loss of cardiac myocytes to programmed cell death, or apoptosis. We therefore measured both rates of myocardial apoptosis and the protein expression levels of several key mediators of apoptosis in the post-SVR ventricle, and compared them to those in sham controls (Figure 5 A&B). The regulation of apoptosis is characterized by an intricate balance between proteins that play anti- and pro-apoptotic roles. Although levels of the anti-apoptotic protein Bcl-2 did not change significantly between weeks 1 and 4 after re-operation, either in SVR or sham control animals, a decrease in expression of pro-apoptotic Bax resulted in an increase in the Bcl-2/Bax ratio at week 1 after re-operation compared to normal hearts (Figure 5C). This increase was more pronounced in SVR vs. sham controls (P < 0.05). By week 4, this ratio in both groups returned to levels seen in normal hearts. This early anti-apoptotic signal was counterbalanced in sham controls by a drop in pro-apoptotic BAD phosphorylation (inactivation) at week 1 (Figure 5D). Interestingly, BAD phosphorylation remained at normal levels in SVR hearts at 1 week, at which time apoptotic indices were down. Early after re-operation, levels of anti-apoptotic Bcl-XL expression were decreased, but then increased very significantly in both treatment groups at week 4 (Figure 5E)..
Compared to normal myocardium, post-MI sham control hearts displayed a progressive decrease in Akt activation (as reflected in the relative phosphorylation of Akt on Western blot, Figure 6A), and this drop in cardioprotective signaling was not impacted by SVR. A small increase was observed at week 1 after re-operation in phosphorylation (i.e. activation) of mitogen activated protein (MAP) kinase p38 in sham controls, although in general, levels of p38 phosphorylation, did not vary substantially between SVR and sham controls (Figure 6B). Similarly, increases in ERK phosphorylation (i.e. activation) were seen in both sham controls and in SVR hearts at week 1 after re-operation, although there was a return to baseline ERK phosphorylation in SVR, but not sham, hearts at week 4 (Figure 6C). Interestingly, a drop in phosphorylation (i.e. activation) of JNK was observed only in sham control hearts at week 1, although phosphorylation of this so-called stress activated kinase, associated in some studies with myocardial apoptotsis, was increased above baseline in normal hearts in both treatment groups at week 4 (Figure 6D).
In our murine model, SVR improved cardiac function. These reconstructed ventricles, however, were susceptible to recurrent dilatation, and a reduction in function was observed as early as 4 weeks post reconstruction. These results are similar to the findings of Nishina et al. in a rat model of aneurysm exclusion [3,19], and are reminiscent of longer-term changes documented in clinical studies of linear or patch ventricular repair [20-22].
The complex integration of multiple signaling pathways is not yet well understood in the evolution of post-MI cardiomyopathy. However, studies have suggested a cardioprotective role for the PI3K/Akt pathway [23,24]; the drop we observed in Akt phosphorylation in the post-MI myocardium may therefore play an important role in the loss and dysfunction of myocardial tissue; this drop persisted even after an improvement in ventricular geometry achieved via SVR.
Only minor differences were observed between sham control and SVR hearts in terms of the phosphorylation of MAP kinases p38 and ERK early after re-operation. MAP kinase activation is generally upregulated after ischemic insults, although the role of this activation has been studied more extensively in the progression from pressure overload-induced hypertrophy to dilated cardiomyopathy. It is not clear what role, if any, the observed reduction in phosphorylation of JNK, generally considered, like p38, a “stress-induced” kinase, might have played in sham control hearts 1 week after re-operation, although this change was not observed in SVR hearts. JNK phosphorylation was then increased in both groups by week 4.
SVR results in a reduction in ventricular volume and a decrease in chamber radius. According to LaPlace's law, these changes decrease wall stress. A reduction in wall stress reduces myocardial oxygen demand and enhances ventricular contraction , and there is an increase in the extent and velocity of systolic fiber shortening. This phenomenon was reflected in the increases in FS and EF observed in our murine model. In addition, SVR is likely associated with acute and subacute changes in hemodynamic parameters, that, in turn, are likely to affect both ventricular remodeling as well as associated changes in myocyte signaling. Future studies may be necessary to sort out the various stimuli for changes in cardiac myocyte gene expression and kinase activation in order to optimize human translation of these interventions.
Furthermore, a reduction in mechanical stress may have also contributed to the reduced level of myocardial apoptosis seen early after SVR, and that may have been mediated by an increase in the Bcl-2/ Bax ratio and a preservation of BAD phosphorylation. As suggested by the schema in Figure 1, persistent changes in molecular signaling, however, such as the reduction of Akt phosphorylation, may drive cardiac remodeling at the cellular level, even after SVR. Subsequent recurrence of LV dilatation could, in turn, instigate a return to pathologic levels of wall stress, and to the vicious cycle of apoptotic cell loss and further progressive LV enlargement.
Current surgical treatments for ischemic cardiomyopathy may not yield optimal long-term outcomes [2,7]. This study explored changes in myocardial signaling that might induce reverse remodeling or “physiologic hypertrophy” to compliment the immediate improvement in ventricular geometry achieved by surgical reconstruction. In addition, cell-based and tissue engineering approaches may be combined with reconstruction. Numerous studies have demonstrated benefit from stem cell delivery to injured myocardium [25-27], and more recent reports have also described bio-artificial matrices that may provide mechanical support to the injured myocardium and enhance myocardial regeneration [28,29]. The direct access of the surgeon at the time of SVR may facilitate otherwise challenging, early applications of these approaches.
Like any small animal model, ours must be understood in light of its numerous limitations. We induced acute infarction in the setting of otherwise normal coronary anatomy and function; more complex human disease may be better mimicked through application of this model in genetic mouse models of diffuse coronary atherosclerosis. In addition, permanent coronary occlusion resulted in a large transmural infarction/aneurysm that exaggerates the typical infarctions of modern candidates for aneurysm resection. Future variations of this model, such as temporary ligation, may extend these observations toward a broader range of human clinical scenarios. The rapid progression of LV remodeling after MI in mice, and of recurrent dilatation after SVR, may require transition through larger animal models for accurate extrapolation to human translation.
If one key to recurrent LV dilatation, however, lies in the molecular biology of the cardiac myocyte, , then the murine model described here may be well-suited to identify critical molecular pathways in the response of the ventricle to SVR. The relative ease with which transgenic strains can be developed in mice has already yielded a wide array of available genetic models. Interestingly, the rapid progression of recurrent dilatation in mice may provide an advantage in the early discovery process. Ongoing studies have begun to examine SVR in the context of myocardial overexpression of activated Akt in transgenic mice. Data from these and other studies may in turn provide a foundation for the development of novel, hybrid surgical interventions for failing hearts, in which genetic, molecular or cell-based therapies may be applied intra-operatively based on a more detailed appreciation of the molecular and cellular parameters of ventricular recovery after reconstruction.
This work was supported by National Institutes of Health, Grant numbers:1R01 HL083118; 1K08HL079239 and American Heart Associstion, Grant number: 0465090Y to MJM.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.