INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

AFTER MYOCARDIAL INFARCTION (MI) there is progressive myocardial remodeling characterized by left ventricular (LV) dilation, contractile dysfunction (30, 33), myocyte hypertrophy (19), and altered expression of contractile (39), calcium-handling (37), and extracellular matrix proteins (38). In mice (4, 25), there is evidence that apoptosis of cardiac myocytes occurs transiently during the first several days after MI in the ischemic area but not in myocardium remote from the area of ischemic injury. However, apoptosis of cardiac myocytes has been observed in LV myocardium from patients with chronic cardiomyopathy late after MI (28), leading to the thesis that myocyte apoptosis contributes to the chronic progression of myocardial failure (2).

Therefore, an important unresolved question is whether myocyte apoptosis occurs late (i.e., months) after MI in tissue remote from the ischemic injury. Several factors that may be present in remodeling myocardium (6, 12, 17, 23, 40) have been shown to stimulate myocyte apoptosis in vitro, including increased mechanical strain (11), neurohormones (13), reactive oxygen species (1, 35), and inflammatory cytokines (21). Because these potential stimuli for apoptosis are not restricted to the infarct, per se, it is possible that late remodeling is associated with apoptosis in myocardium remote from the area of ischemic damage.

The goal of this study was to test the hypothesis that LV remodeling late after MI is associated with apoptosis in myocardium remote from the infarct and, if so, to determine the temporal course of apoptosis and its relationship to the progression of structural and functional remodeling. Accordingly, we determined LV dilation and maximal LV contractile function using an isovolumic balloon-in-LV Langendorff technique and measured myocyte apoptosis by terminal deoxynucleotidyl transferase-mediated nick end-labeling (TUNEL) in hearts from mice 1, 4, and 6 mo after coronary artery ligation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Myocardial infarction. MI was induced in male CD-1 mice (Charles River) by ligation of the left coronary artery as previously described (15, 26). The mice were 8-10 wk old and weighed 30-32 g. The animals were killed and studied at 1, 4, and 6 mo after MI. The protocol was approved by the Institutional Animal Care and Use committee at Boston University School of Medicine.

Langendorff preparation. LV volume and function were determined using an isolated, isovolumic red cell-perfused Langendorff heart preparation as previously described (15). After 20 min of baseline perfusion at a systolic pressure of 80-100 mmHg and a pacing rate of 7 Hz (420 beats/min), the intracardiac balloon volume was set to the lowest volume at which minimal LV pressure tracings could be recorded. This volume (V₀) was defined as the "zero" volume. Systolic and diastolic pressure-volume relationships were established by increasing the LV balloon volume in 5-µl increments using an air-tight Hamilton syringe. LV functional measurements were obtained 1-2 min after each increment in volume when a new steady state was reached. The LV volume was increased up to a value (V_max) at which peak LV developed pressure (LVDP) was reached and a further increase in balloon volume led to a decrease in peak LVDP.

Infarct size. After fixation, the LV was embedded in paraffin, and sections were taken at the apex, midcavity, and base and stained with trichrome Masson. Infarct size was determined as the mean percentage of epicardial and endocardial circumference occupied by scar tissue for the three sections as previously described by Pfeffer et al. (32).

Northern hybridization. Mice were killed at 1, 4, and 6 mo after MI, the atria and right ventricle were removed, and the LV was weighed. The infarct was carefully dissected along with a 0.5- to 1-mm rim of normal-appearing tissue (peri-infarct region) to yield samples referred to as "infarct/peri-infarct" and "remote," which were weighed and rapidly frozen in liquid nitrogen. Infarct size for these hearts was expressed as the weight ratio of the infarct/peri-infarct sample to the whole LV (i.e., before removal of the infarct). Only samples from hearts with infarcts 30% were used. mRNA was extracted and subjected to Northern hybridization as previously described (8) using a full-length cDNA for rat prepro-atrial natriuretic peptide (ANP). Blots were exposed to X-ray film and quantified by laser densitometry using Molecular Analyst software (Bio-Rad, Hercules, CA). mRNA levels were normalized by reprobing with an oligonucleotide complimentary to 18S rRNA (8).

TUNEL and Hoechst 33258 staining. Sections 5 µm thick from the LV apex, midcavity, and base were stained with TUNEL using a cell death detection assay (Boehringer Mannheim, Indianapolis, IN) according to the manufacturer's instructions. Sections were incubated with the TUNEL reaction mixture containing terminal deoxynucleotidyl transferase and fluorescein-dUTP for 1 h at 37°C. Staining with Hoechst 33258 (10 µM for 30 min) was performed on the same sections after TUNEL staining for visualization of nuclei. Sections were visualized by fluorescence microscopy using a Nikon Diaphot 300 microscope, and images were acquired with a Bioquant (Memphis, TN) image acquisition and analysis system using an Optronics camera.

Caspase activity. Total myocardial caspase activity was measured using the ApoAlert CPP32/Caspase-3 assay kit from Clontech (Palo Alto, CA). Samples of ventricular myocardium remote from the infarct and sham-operated animals were removed and rapidly frozen as described for isolation of mRNA (see Northern hybridization). Samples were finely ground under liquid nitrogen using a mortar and pestle. Samples of 20-40 mg of tissue were lysed using the cell lysis buffer from the kit using 100 µl of buffer per 10 mg of tissue and were left on ice for 20 min. Samples were centrifuged at 10,000 g for 3 min, and CPP32 (cysteine protease protein-32) activity was measured in 50-µl volumes of supernatant per the manufacturer's instructions. Protein concentration in the same samples was measured using the Bradford assay, and units of caspase activity were calculated per microgram of protein.

Statistical analysis. Differences in pressure-volume relationships were tested by two-way analysis of variance (ANOVA) for repeated measures. When two-way ANOVA indicated a significant (P < 0.05) difference, data were further examined by the method of least significant differences. Group differences for a single variable were tested by Student's unpaired t-test (two-tailed). A P value of <0.05 was considered significant. All data are presented as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Myocardial infarction. A total of 140 mice were studied. Of the 122 mice that underwent coronary artery ligation, ~50% died in the first 24 h after surgery. In the surviving mice, the 6-mo survival rate as assessed by Kaplan-Meier analysis was 72 ± 7% (vs. 100% in sham-operated mice).

Heart weight. Body weight increased between 1 and 4 mo in both the MI and sham groups, reflecting normal age-related growth (Table 1). There was no difference in body weight between MI and sham groups at 1, 4, or 6 mo. Heart weight likewise increased between 1 and 4 mo but was greater in the MI group at each time point, with or without normalization to body weight. Neither ascites nor pleural effusion was found at death in any animal, and ratios of wet to dry lung and liver weights were similar in the MI and sham groups (data not shown).

LV volume and function. LV size and function were measured using the isovolumic, blood-perfused (balloon-in-LV) Langendorff preparation paced at 7 Hz (15). The LV end-diastolic pressure-volume relationship was shifted rightward (vs. sham) at 1, 4, and 6 mo after MI, and the degree of rightward shift increased progressively from 1 to 6 mo (Fig. 1).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Mean left ventricular (LV) end-diastolic pressure-volume relationships in hearts from mice 1 (A), 4 (B), or 6 mo (C) after myocardial infarction (MI) and their sham-operated controls. At 1, 4, and 6 mo the volumes of the infarcted LVs are greater than those from age-matched sham-operated controls (*P < 0.05 for all time points). Among the infarcted hearts (MI), LV volume at 6 mo was greater than that at 1 or 4 mo (P < 0.05). Values are the mean data for 8 MI and 6 sham hearts at 1 mo, 7 MI and 6 sham hearts at 4 mo, and 13 MI and 6 sham hearts at 6 mo. LVEDP, LV end-diastolic pressure.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   A: mean curves describing LV developed pressure vs. volume in mice 1, 4, and 6 mo post-MI and their sham-operated controls. The curves are shifted progressively rightward and downward, and the peak developed pressure declines progressively at 4 and 6 mo (*P < 0.05 vs. sham; **P < 0.01 vs. sham). The numbers of hearts in each group are as reported in Fig. 1. B: peak LV developed pressure at 1, 4, and 6 mo post-MI normalized to the LV volume that yielded the maximum developed pressure (referred to as Vmax). The curves were shifted rightward in infarcted hearts at 4 and 6 mo (**P = 0.003 vs. shams; ***P = 0.001 vs. shams). The numbers of hearts in each group are as reported in Fig. 1.

ANP mRNA expression. Prepro-ANP mRNA was measured by Northern hybridization as a marker of fetal gene expression (Fig. 3A). Infarcted hearts were dissected into two parts, one consisting of the infarct with the peri-infarct rim and the other consisting of the remaining noninfarcted (i.e., remote) LV. In noninfarcted LV, ANP mRNA was not increased (vs. shams) at 1 mo but was increased 7- and 4-fold at 4 and 6 mo, respectively (Fig. 3B). In the infarct/peri-infarct sample, ANP mRNA was increased 10-, 10-, and 9-fold at 1, 4, and 6 mo, respectively.

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Expression of prepro-atrial natriuretic peptide (ANP) mRNA in LV myocardium after MI. A: representative Northern hybridizations for ANP and 18S in LV from sham-operated mice and from the remote or infarcted (including peri-infarct rim) regions of infarcted hearts. B: mean levels of ANP mRNA (normalized to 18S RNA) in the infarct/peri-infarct and remote regions compared with age-matched sham-operated hearts (*P < 0.05 vs. shams; ***P < 0.001 vs. shams; n = 4 hearts for each group).

TUNEL staining, nuclear morphology, and caspase activity. Cross sections from the LV apex, midcavity, and base of each heart were stained by the in situ TUNEL technique (Fig. 4A). There were rare TUNEL-positive nuclei in sham-operated animals. In myocardium remote from the infarct, the number of TUNEL-positive cells appeared to be increased at all time points. High-power microscopy demonstrated that TUNEL-positive staining was localized primarily to nuclei of myocytes, which had evidence of chromatin condensation by Hoechst staining (Fig. 4A). This impression was confirmed by confocal microscopy in selected sections. TUNEL-positive cells appeared to be randomly distributed throughout the remote myocardium.

View larger version (91K): [in this window] [in a new window]   Fig. 4.   Apoptosis in remote myocardium late post-MI. A: representative terminal deoxynucleotidyl transferase-mediated nick end-labeling (TUNEL)- (left) and Hoechst 33258-stained sections (right) from the remote (i.e., noninfarcted) region of a mouse heart 6 mo post-MI. The arrows indicate TUNEL-positive cells and the corresponding Hoechst 33258-stained nuclei (calibration bar = 50 µm). B: mean number of TUNEL-positive nuclei per 105 myocytes in the remote myocardium at 1, 4, and 6 mo post-MI. The number of TUNEL-positive myocytes (27 ± 8/105 nuclei) in sham-operated mice was similar for all time points. The number of TUNEL-positive cells increased progressively from 1 to 6 mo (P = 0.005 by log rank test), reaching significance at 6 mo (**P = 0.002 vs. sham; n = 4-6 hearts for each group).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the mouse heart, myocyte apoptosis occurs transiently in the immediate area of ischemic damage for up to several days after MI but not in myocardium remote from the area of ischemia (4, 25). The major new finding of our study is that late (i.e., 1-6 mo) after MI, chronic LV remodeling with chamber dilation and impaired systolic pump function is temporally associated with increased myocyte apoptosis in myocardium remote from the area of initial ischemic damage.

Structural and functional remodeling. Post-MI LV remodeling consists of at least two temporal phases (32). During the early phase (first few days), there is expansion of the infarcted tissue, leading to early LV chamber dilation and increased mechanical stress on the remaining viable myocardium. During the late phase (subsequent weeks to months), there is progressive remodeling of the remote areas of the LV characterized by further chamber dilation, myocardial hypertrophy, reexpression of a fetal phenotype, and progressive deterioration in systolic pump function.

Late myocyte apoptosis. The mouse MI model has been used to characterize the mode of cell death during the first 7 days after MI (4, 24). In mice 7 days after MI, Li et al. (25) found that LV chamber dilation was associated with evidence of both necrosis and apoptosis of cardiac myocytes in the infarct/peri-infarct region, but not in the septum, which is remote from the area of ischemic damage. Furthermore, they found that, in mice that overexpress insulin-like growth factor-1, there was less apoptosis, less extensive remodeling, and better LV function. During the first 72 h after MI, Bialik et al. (4) also found evidence of myocyte apoptosis that was restricted to the area of hypoxic myocardium. Likewise, in rats apoptosis is localized to the infarct and peri-infarct regions early after MI, with little to none in areas remote from the infarct (10).

Survival. The infarcted mice had a reduced survival over 6 mo of this study. Overt heart failure and organ congestion were not found in mice that survived to 6 mo, suggesting that the more severely affected animals had died before that time. We have observed that some mice carried beyond 6 mo experience rapid deterioration and death associated with respiratory distress and increased lung water, suggesting that animals that are compensated at 6 mo may subsequently progress to overt heart failure. Thus the animals included in our analysis may have less LV dysfunction and apoptosis than those that died before euthanization.

Implications. These observations raise several important questions. First, what is the cause of increased apoptosis in remote myocardium? In vitro, several factors that may be relevant to chronic remodeling can stimulate myocyte death by apoptosis, including mechanical stress (11), angiotensin (23), inflammatory cytokines (34), norepinephrine (13), and oxidative stress (1). The progressive increase in the number of apoptotic myocytes suggests that the stimulus for apoptosis (e.g., mechanical stress) increases with time after MI. We did not determine which factor(s) might be stimulating apoptosis late post-MI. However, the estimated LV diastolic wall stress (27) was ~2-4 kdyn/cm² in shams (assuming an LVEDP of 2-5 mmHg) and increased by ~5-10-fold at 6 mo post-MI (assuming an LVEDP of 15 mmHg) (data not shown). These values are similar in magnitude to the mechanical stress that has been shown to cause apoptosis of cardiac myocytes in vitro (11). Thus increased mechanical strain may have been a stimulus for apoptosis late post-MI.