INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

CARDIAC HYPERTROPHY is a fundamental, adaptive response of the heart to an increased workload (for review, see Ref. 17). The hypertrophic response is, at least initially, considered to be a beneficial, compensatory mechanism balancing the increased workload on the myocardium imposed during congestive heart failure (CHF), thereby contributing to maintain cardiac output. Several studies have demonstrated that cardiomyocyte hypertrophy is associated with elevated levels of intracellular calcium or enhanced sensitivity to calcium, suggesting a crucial role of calcium in the signaling pathways leading to myocardial hypertrophy (for review, see Ref. 23). In addition, many studies have demonstrated that hypertrophic stimuli result in reprogramming of gene expression in the myocardium, with reexpression of a fetal phenotype (3, 19). However, the molecular mechanisms that couple increased intracellular calcium to the changes in myocardial gene expression are unknown. Recently, it has been hypothesized that the calcium/calmodulin-dependent protein phosphatase calcineurin might be a nodal point in the conversion of elevated intracellular calcium levels to the trophic signals that mediate hypertrophy of the myocardium (16). Cardiac-specific overexpression of a constitutively active form of calcineurin in transgenic mice led to a phenotype demonstrating myocardial hypertrophy and dilated cardiomyopathy with strong similarity to the molecular and pathophysiological responses of the heart in human heart failure (16). In addition, as shown in the latter report, development of cardiac hypertrophy was prevented by administration of the immunosuppressant drug cyclosporin A (CsA), an inhibitor of calcineurin phosphatase activity. Furthermore, in transgenic mouse models of hypertrophic cardiomyopathy, inhibition of calcineurin has been shown to block or attenuate the hypertrophic response (15, 26). These results indicate that activation of calcineurin may mediate cardiac hypertrophy in response to at least some pathological stimuli and that the calcineurin pathway may be a putative pharmacotherapeutical target in the treatment of dysfunctional cardiac hypertrophy. However, to what extent calcineurin is a principal mediator of the myocardial hypertrophy associated with ischemic heart failure has not been studied.

The aim of the present study was to test the hypothesis that calcineurin is a mediator of cardiac hypertrophy in ischemic heart failure, i.e., the most prevalent form of CHF in the Western hemisphere. To test this hypothesis, we first investigated to what extent intervention with CsA during ischemic heart failure in rats would lead to attenuation of gross myocardial hypertrophy and corresponding attenuation of gene expression of key markers of myocardial hypertrophy. Second, we investigated the effects of CsA treatment on left ventricular (LV) dimensions and LV functional parameters. Third, we investigated whether intervention with CsA during CHF affects myocardial apoptosis and fibrosis.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Surgical preparation. We used the left coronary artery-ligated rat model of CHF according to the method of Selye et al. (24) with minor modifications (20). Male Wistar rats (~295 g) were anesthetized with halothane and ventilated with a rodent ventilator with 1% halothane in a mixture of one-third O₂ and two-thirds N₂O. Rats were subjected to ligation of the proximal portion of the left coronary artery. Except for ligation of the coronary artery, sham-operated rats underwent the identical procedure. Body temperature was maintained at 37°C during the surgical procedure with the use of a heat blanket and a homeothermic blanket control unit (Harvard Apparatus, Edenbridge, UK). Surgical mortality was 30% in the rats with myocardial infarction (MI). The animal experiments, procedures, and housing were in accordance with institutional guidelines and national legislation conforming to The European Convention for The Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes of 18 March 1986.

Study protocol. The rats were allocated to four groups: sham-operated rats to be treated with CsA (sham-CsA rats, n = 5); sham-operated rats to be treated with vehicle (sham-vehicle rats, n = 4); CHF rats to be treated with CsA (CHF-CsA rats, n = 6); and CHF rats to be treated with vehicle (CHF-vehicle rats, n = 6). CsA (50 mg · kg¹ · day¹; Novartis, Basel, Switzerland) or vehicle (saline) was administered continuously by subcutaneously implanted mini osmotic pumps (Alzet 2ML1, Alza, Palo Alto, CA). It was previously reported (28) that CsA administered at a dose of 10 mg · kg¹ · day¹ is sufficient to completely inhibit myocardial calcineurin activity in rats. The mini osmotic pumps were implanted 24 h after induction of MI to minimize potential effects of CsA on infarct size. The rats were anesthetized subcutaneously with a mixture of fluanisone (6.25 mg/kg)-fentanyl citrate (0.2 mg/kg) (Janssen-Cilag, Buckinghamshire, UK) and midazolam (3.1 mg/kg) (Hoffmann-La Roche, Basel, Switzerland) given 10 min before implantation of the mini osmotic pumps. The treatment period was 14 days. To achieve continuous administration of CsA for 14 days, mini osmotic pumps were implanted on days 1 and 8, consecutively, after induction of MI. Eight days after MI, three CHF rats died and one CHF rat was euthanized because of poor condition (2 CHF-vehicle rats and 2 CHF-CsA rats).

Hemodynamic measurements. Fifteen days after the induction of MI, the rats were anesthetized with fluanisone-fentanyl citrate-midazolam as described and ventilated with a rodent ventilator. A 2-F micromanometer-tipped catheter (model SPR-407, Millar Instruments, Houston, TX) was inserted through the right carotid artery into the aorta to determine mean arterial blood pressure (MAP). The catheter was subsequently advanced further into the LV, and end-diastolic pressure (LVEDP) and systolic pressure (LVSP), as well as peak positive and peak negative first derivatives of the LV pressures (+dP/dt and dP/dt), were recorded using a computer data acquisition workstation (model MP 100A-CE, Biopack Systems, Santa Barbara, CA).

Echocardiographic examination. Immediately after the hemodynamic measurements, echocardiography was performed as previously described (19), using the fully digital Vingmed System FiVe (GE Vingmed Ultrasound, Horten, Norway) and a 7.5-MHz linear array transducer. Cine loops and still images were digitally stored for subsequent analysis using the EchoPac software (GE Vingmed Ultrasound). Briefly, two-dimensional long-axis and short-axis views of the LV were recorded with a typical frame rate of 196 frames/s. The short-axis dimensions were recorded at the level of the tip of the papillary muscle. After gain settings were optimized, M-mode tracings were recorded at the same level, and interventricular septum and posterior wall thicknesses at end-diastole (IVSEDT and PWEDT, respectively) and at end-systole (IVSEST and PWEST, respectively) were measured. LV internal dimensions [end-diastolic diameter (LVEDD) and end-systolic diameter (LVESD)] were recorded as the largest anteroposterior diameter outside the infarcted area. LV fractional shortening (FS) and relative wall thickness (RWT) were calculated according to the following formulas: FS (%) = [(LVEDD  LVESD)/LVEDD] × 100 and RWT = (IVSEDT + PWEDT)/LVEDD. LV systolic wall stress was calculated as follows using a spherical curve fit: LV systolic wall stress = 1.36 × (aortic systolic pressure × LVESD)/(IVSEST + PWEST). The tracings were analyzed by one observer (R. Bjørnerheim) who had no knowledge of the study group.

Blood and tissue sampling. After the echocardiographic examination, blood samples were drawn from the abdominal aorta. Blood for analysis of plasma endothelin-1 (ET-1) was collected in prechilled tubes containing EDTA (1 mg/ml blood), whereas blood for analysis of renin was collected in prechilled tubes containing EDTA (1 mg/ml blood) and the angiotensin-converting enzyme inhibitor 2,3-dimercapto-1-propanol (0.4 mol/l). The blood was centrifuged at 4°C, and the plasma was aspirated and stored at 70°C until use. The rats were subsequently euthanized by excision of the heart. The atria and the ventricles were sectioned, separated, and weighed. The infarcted area (scar tissue) was excised from the noninfarcted LV, with care taken to avoid contamination of viable myocardial tissue with necrotic tissue. The excised scar tissue was weighed to estimate infarct size. The noninfarcted LV from the CHF rats was further sectioned by separating the tissue contiguous to the infarct (viable LV free wall) and the tissue distal to the infarct (interventricular septum). The tissue samples were snap-frozen in liquid nitrogen and stored at 70°C. A piece of the apical part of the nonischemic interventricular septum was processed separately for histochemical analysis of myocardial fibrosis and apoptosis. The right tibia of each rat was isolated, and its length was measured.

Histochemistry and in situ detection of nuclear DNA fragmentation. Frozen myocardial tissue from the nonischemic interventricular septum was cut into 5-µm sections using a cryostat microtome. Sections were fixed in acetone for 10 min and stained with van Gieson to identify extracellular collagen in the myocardium. Parallel sections were fixed in 1% paraformaldehyde for 10 min for in situ terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-digoxigenin nick end-labeling (TUNEL) analysis of myocardial apoptotic cell death (ApopTag, Oncor, Gaithersburg, MD). The TUNEL analysis was performed according to the manufacturer's instructions. Briefly, residues of digoxigenin-labeled UTP were catalytically added to DNA by TdT and subsequently incubated with an anti-digoxigenin peroxidase-labeled antibody. Finally, nuclear DNA fragmentation was visualized by incubation with diaminobenzidine as the chromogen. The slides were counterstained with hematoxylin. A tissue section of female rodent mammary gland after weaning served as positive control for the TUNEL reaction.

RNA isolation. Total RNA from the noninfarcted LV free wall and the interventricular septum was prepared by ion-exchange chromatography using a commercially available RNA isolation procedure (RNeasy Maxi Kit, Qiagen). Briefly, the tissue was lysed and homogenized under denaturing conditions with the use of guanidinium isothiocyanate and -mercaptoethanol and then applied to a column with a silica gel-based membrane for binding of total RNA. High-quality RNA was then eluted in RNase-free water.

Northern blot analysis. Northern blot analysis was performed as previously described (19). The cDNA probes used were cDNA fragments of rat atrial natriuretic peptide [ANP; nucleotides (nt) 1-635, GenBank accession no. X00665], rat brain natriuretic peptide (BNP; nt 180-365, GenBank accession no. M25297), rat skeletal -actin (nt 2,581-3,037, GenBank accession no. J00692), and rat transforming growth factor-₁ (TGF-₁; nt 855-1,341, GenBank accession no. X52498). The cDNA probes were subcloned between the BamH I/EcoR I restriction sites of pBluescript SK⁺ and then radiolabeled by the random priming method in the presence of [³²P]dCTP (specific activity ~6,000 Ci/mmol) with the use of a random-priming labeling kit (Megaprime DNA labeling system, Amersham Pharmacia Biotech, Amersham, UK). Autoradiography of the gel was performed using storage phosphor plates and a scanning phosphorimager (PhosphorImager 445 SI, Molecular Dynamics, CA). Densitometric analysis of the bands was performed with the Image-Master software package (Pharmacia Biotech). To control for variations in loading and transfer efficiencies, the blots were rehybridized with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe (a fragment of rat GAPDH cDNA; nt 458-994, GenBank accession no. M17701).

Plasma and serum measurements. Plasma renin levels were quantified in a bioassay measuring renin enzymatic activity. Plasma samples were incubated at 37°C for 60 min. The rate of conversion of angiotensinogen to angiotensin I was assayed by radioimmunoassay and expressed as nanomoles of angiotensin I produced per liter of plasma per hour. Plasma ET-1 concentrations were measured by ELISA (1), and serum creatinine levels were determined by photometric analysis.

Statistical analysis. All data are presented as means ± SE. Statistical analysis was assessed by unpaired two-tailed Student's t-test. P values <0.05 were considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Effects of CsA on myocardial hypertrophy. As shown in Table 1, CsA decreased both heart weight and body weight of the sham-operated rats compared with vehicle (P < 0.05), whereas the length of the tibia was not significantly affected. Both the LV and right ventricular (RV) weight-to-tibial length ratios were significantly lower in the sham-CsA group than in the sham-vehicle group. However, CsA did not affect RWT in the sham-operated rats. In the CHF-vehicle rats, the RV weight-to-tibial length ratio as well as the atrial weight-to-tibial length ratio was substantially increased compared with that of sham-vehicle rats at day 15 after MI (P < 0.05), demonstrating compensatory myocardial hypertrophy in the CHF-vehicle rats. Intervention with CsA virtually prevented ventricular hypertrophy during CHF. As shown in Table 1, RV weight-to-tibial length ratio was substantially decreased in the CHF rats after treatment with CsA (3.9 ± 0.1 vs. 6.3 ± 0.6 mg/mm, P < 0.05). Although the RV weight-to-tibial length ratio was still significantly elevated in the CHF-CsA group compared with the sham-CsA group, the decrease of the ratio after treatment with CsA was more substantial for the CHF rats (38%) than for the sham-operated rats (27%). Also the LV weight-to-tibial length ratio was significantly decreased in the CHF rats after intervention with CsA. In addition, echocardiographic analysis of the rats 15 days after MI revealed that RWT of the LV was substantially reduced in the CHF-CsA group compared with the CHF-vehicle group (0.26 ± 0.003 vs. 0.42 ± 0.03 mg/mm, P < 0.05), clearly demonstrating the detrimental effects of inhibition of compensatory hypertrophy during postinfarction failure. There was no significant effect of CsA on the atrial weight-to-tibial length ratio in the CHF rats. Furthermore, the scar weight-to-tibial length ratio in the CHF rats was not affected after treatment with CsA, indicating no effect of CsA on myocardial infarct size when treatment was initiated 24 h after induction of MI.

Effects of CsA on myocardial fibrosis and apoptosis. Histochemical analysis revealed myocardial deposition of collagen predominantly in perivascular tissue (Fig. 1). There were no visually detectable differences in collagen density or distribution between sham-operated rats and CHF rats irrespective of treatment strategy.

View larger version (109K): [in this window] [in a new window]   Fig. 1.   Sections of nonischemic myocardial tissue 15 days after sham operation or induction of myocardial infarction (MI). A-D: sections stained with van Gieson stain, demonstrating collagen (red), cardiomyocytes (yellow), and nuclei (blue). Collagen is shown surrounding small vessels between cardiomyocytes (arrowheads) in sham-operated rats treated with vehicle (A) or cyclosporin A (CsA) (B) and in congestive heart failure (CHF) rats treated with vehicle (C) or CsA (D). Original magnification, ×200. E-F: sections demonstrating rare TdT-mediated dUTP-digoxigenin nick end-labeling (TUNEL)-positive cells (arrow) in the myocardium of CHF rats after treatment with vehicle (E) or CsA (F). Original magnification, ×400.

Effects of CsA on LV pressures and dimensions. As shown in Table 2, both MAP and LVSP were lower, but not statistically different, in the sham-CsA group compared with the sham-vehicle group. However, MAP and LVSP were markedly and significantly lower in the CHF-CsA group compared with the CHF-vehicle group (49 ± 4 and 66 ± 4 mmHg vs. 71 ± 6 and 92 ± 5 mmHg, respectively; P < 0.05), suggesting severe cardiac dysfunction in the CHF rats after intervention with CsA. There were no significant differences in LVEDP between the CHF-CsA group (9 ± 3 mmHg) and the CHF-vehicle group (8 ± 2 mmHg).

Effects of CsA on LV function. As shown in Table 2, peak +dP/dt was reduced in the sham-operated rats after treatment with CsA (P < 0.05), whereas LV FS and peak dP/dt were not significantly affected. In the CHF-vehicle rats, peak +dP/dt and LV FS were decreased compared with the sham-vehicle rats (P < 0.05). Intervention with CsA caused an additional reduction of peak +dP/dt in the CHF rats compared with the vehicle (from 4,879 ± 394 mmHg/s in the CHF-vehicle group to 2,773 ± 223 mmHg/s in the CHF-CsA group, P < 0.05). LV FS was also lower in the CHF-CsA group compared with the CHF-vehicle group, but the decline was not statistically significant (Table 2). Our data therefore indicate that CsA aggravates systolic function during CHF. In addition, peak dP/dt was reduced from 4,143 ± 351 mmHg/s in the CHF-vehicle rats to 2,529 ± 218 mmHg/s in the CHF-CsA rats (P < 0.05). Thus our data also indicate impaired diastolic function in the failing hearts after intervention with CsA.

Effects of CsA on myocardial fetal gene expression during CHF. Myocardial gene expression of ANP, BNP, skeletal -actin, and TGF-₁ were low in the LV of sham-operated rats (Fig. 2). However, substantial upregulation of these genes was seen after induction of CHF. The ANP, BNP, skeletal -actin, and TGF-₁ mRNA levels were increased 10.8-, 2.7-, 3.2-, and 4.7-fold, respectively, in the interventricular septum compared with the sham-operated rats (P < 0.05) and were increased 15.8-, 3.9-, 6.3-, and 7.9-fold, respectively, in the LV contiguous to the ischemic region (P < 0.05). Treatment of CHF rats with CsA caused a 28% reduction of the increase in mRNA expression of skeletal -actin in the interventricular septum (P < 0.05). Similar reduction was seen in the nonischemic LV bordering the ischemic area. However, the decrease observed in this region was not statistically significant (P = 0.07). Myocardial gene expression of ANP, BNP, and TGF-₁ were not significantly affected after intervention with CsA.

View larger version (71K): [in this window] [in a new window]   Fig. 2.   Northern blot analysis demonstrating myocardial mRNA expression of genes encoding a fetal phenotype in the left ventricle (LV) of sham-operated rats (Sham-veh) and in the interventricular septum (IVS) and the LV free wall bordering the ischemic region (LV border) of CHF rats treated with CsA (CHF-CsA) or vehicle (CHF-veh) for 14 days. In each lane, 15 µg of total RNA were used. A: the autoradiograph was exposed to storage phosphor plates (48 h) and analyzed by phosphorimaging. ANP, atrial natriuretic peptide; BNP, brain natriuretic peptide; TGF-1, transforming growth factor-1. B: densitometric analysis of the scanning data. The data are presented as ratios of mRNA expression levels of the fetal genes to the levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA to normalize for variations of the amount of total RNA loaded on the gel and transfer efficiencies. The data are presented as means ± SE of n = 4 rats in each group. The statistical analysis was performed by unpaired Student's t-test. *P < 0.05 vs. sham-vehicle; P < 0.05 vs. CHF-vehicle.

Effects of CsA on plasma renin activity, ET-1 levels and serum creatinine levels during CHF. Plasma renin activity was increased from 24 ± 9 nmol · l¹ · h¹ in the vehicle-treated CHF rats to 73 ± 5 nmol · l¹ · h¹ in the CsA-treated CHF rats (P < 0.05), whereas plasma ET-1 levels were increased from 2.6 ± 1.0 to 4.8 ± 0.4 pg/ml after intervention with CsA. Creatinine serum levels measured to monitor renal function were not significantly different between the CHF-CsA group (52 ± 2 µmol/l) and the CHF-vehicle group (49 ± 2 µmol/l).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates for the first time that treatment with the calcineurin inhibitor CsA prevents myocardial hypertrophy during postinfarction failure in rats. CsA strongly inhibited the hypertrophic response in the right and left ventricles. Our data therefore support the hypothesis that calcineurin-dependent pathways are involved in the myocardial hypertrophy associated with CHF. In addition, CsA also decreased the cardiac mass in the sham-operated rats, indicating an inhibitory effect of CsA on normal cardiac growth.

Recent reports (15, 26) have demonstrated that inhibition of calcineurin prevents or attenuates cardiac hypertrophy in different transgenic mice models of hypertrophic cardiomyopathy. As to the involvement of calcineurin in pressure-overload hypertrophy, recent data are conflicting. Two studies (14, 26) have demonstrated prevention or attenuation of cardiac hypertrophy in aortic-banded mice after treatment with CsA, whereas other studies (6, 13, 28) have not found evidence for inhibition of cardiac hypertrophy using the same experimental models and intervention with CsA. Postinfarction failure is also associated with increased load on the heart. However, this condition has more complex pathophysiology, with the release of several vasoactive peptides and cytokines from the granulation tissue replacing the MI. Many of these substances have been implicated in hypertrophy and remodeling of the myocardium (4, 21). Our data demonstrate that even in this setting inhibition of calcineurin strongly inhibited myocardial hypertrophy, supporting the hypothesis that calcineurin is a crucial mediator in the pathways of myocardial hypertrophy.

In addition to the inhibitory effects of CsA on compensatory myocardial hypertrophy, the present study demonstrates that CsA impairs normal cardiac growth. Inhibition of normal cardiac growth was also associated with reduction of body weight. Thus it might be argued that diminished cardiac mass after intervention with CsA could be due to toxic or catabolic actions of the drug. In terms of the very short time span of the present treatment protocol, this scenario is less likely. The rats included in the CsA intervention protocol were not overtly sick compared with the vehicle-treated rats. Rather, recent reports (8, 18, 25) have demonstrated that CsA also inhibits calcineurin-mediated growth of skeletal muscle. This growth-inhibitory effect of CsA on skeletal muscle is more likely to account for reduction of body weight. Although CsA also inhibited cardiac growth in the sham-operated rats, intervention with CsA decreased RV mass to a greater extent in the CHF rats than in the sham-operated rats. This indicates more potent actions of CsA during CHF, a condition associated with powerful hypertrophic stimulation of the heart.

Several investigators (23) have demonstrated that elevation of intracellular calcium is associated with initiation of myocardial hypertrophy. Furthermore, it has previously been shown (10) that myocardial overexpression of calmodulin causes hypertrophy. However, the hypothesis that the calcium/calmodulin-dependent serine/threonine phosphatase calcineurin may be the link between increased levels of intracellular calcium and activation of transcription factors involved in expression of the hypertrophic phenotype in failing myocardial tissue has not been addressed until recently (16). Calcineurin is activated by sustained elevations of intracellular calcium (7). In contrast, calcineurin appears to be relatively insensitive to the transient oscillations of calcium concentrations associated with the cycles of contraction and relaxation of cardiac myocytes (5). Thus cardiomyocytes are able to distinguish between the transient elevations of calcium associated with the contraction-relaxation cycles of the myocardium and the sustained elevations elicited by, for example, neurohumoral factors, growth factors, or mechanical overload. Recently, Molkentin and colleagues (16) reported convincing evidence for a crucial role of the transcription factor NF-AT3 as a mediator of cardiac hypertrophy after activation by calcineurin. Calcineurin dephosphorylates NF-AT3, enabling it to translocate to the nucleus (9) and, subsequently, to bind to the transcription factor GATA4 (16), an established regulator of cardiac gene expression (11). It is also conceivable that other signaling mechanisms independent of calcineurin may mediate hypertrophic stimuli in myocardial tissue. For instance, our data demonstrate no significant effect of CsA on atrial hypertrophy in CHF rats, supporting the existence of a calcineurin-independent pathway at least in atrial hypertrophy.

It was recently reported (26) that, in a transgenic mice model of hypertrophic cardiomyopathy due to aberrant expression of myosin light chain-2, CsA prevented both cardiac hypertrophy and the activation of the fetal genes ANP and skeletal -actin in the heart. Similarly, Mende et al. (15) reported that in transgenic mice with cardiomyocyte-specific overexpression of constitutively active G_q, the moderate inhibition of myocardial hypertrophy after treatment with CsA was parallel to a reduction of myocardial ANP mRNA levels. In another study (6), however, transverse aortic banding in mice and subsequent inhibition of calcineurin by CsA did not attenuate induction of cardiac hypertrophy or myocardial ANP mRNA expression. In the present study, we demonstrate moderate reductions of skeletal -actin mRNA levels in the LV in the CHF rats after treatment with CsA, indicating that during postinfarction failure calcineurin may activate cardiac transcription factors that oppose repression of the skeletal -actin gene. However, CsA did not affect TGF-₁ or ANP mRNA levels in the failing LV. Furthermore, CsA treatment had no significant effect on BNP mRNA expression, i.e., a gene previously reported (16) to be induced by NF-AT3/GATA4. Our data therefore suggest that the attenuation of postinfarction myocardial hypertrophy after inhibition of calcineurin is only partially associated with decreased myocardial expression of the fetal gene program. It may be difficult to find a unifying explanation for the different effects of CsA on expression of the fetal gene program. Obviously, the disparate findings may be due to different experimental models and treatment protocols. It has also been reported (22) that hemodynamic stress rather than cardiac hypertrophy correlates with induction of the fetal phenotype of the myocardium. In the present study, the inhibition of cardiac hypertrophy during postinfarction failure after treatment with CsA was not associated with alteration of LV wall stress.

As demonstrated in this study, CsA did not affect myocardial apoptosis in the viable failing myocardium. Thus the attenuation of the ventricular weight increase observed in the CHF rats after intervention with CsA did not appear to be a result of increased myocardial cell loss through apoptosis. It has previously been reported (12) that intervention with CsA may augment mitochondrial damage and myocardial fibrosis after transplantation of the heart. We did not find evidence for any effect on myocardial fibrosis after 14 days of treatment with CsA during postinfarction failure.

The dramatic effects of CsA on inhibition of ventricular hypertrophy during ischemic heart failure, as demonstrated in the present study, were associated with increased dilatation of the LV. Dilatation of the LV is an ominous finding shown to be a major predictor of mortality in patients with CHF (27). Although our study was not designed to investigate mortality of CsA intervention, the CHF-CsA rats displayed impaired myocardial function as evidenced by decreased peak +dP/dt and dP/dt as well as reduced LV FS. Furthermore, both plasma renin and ET-1 levels were increased in the CHF-CsA rats compared with the CHF-vehicle rats, providing neurohormonal cues of aggravated cardiac function. However, inhibition of calcineurin-mediated hypertrophy of the ventricles may not account entirely for the detrimental effects of CsA on myocardial function. It has recently been reported (2) that CsA may also impair systolic myocardial function ex vivo. Thus cardiotoxic effects on myocardial contractility may also contribute to the impaired cardiac function. This is supported by the findings in the present study demonstrating reduction of peak +dP/dt in the sham-operated rats after treatment with CsA. However, the decreased peak +dP/dt may be due to decreased contractile mass rather than cardiotoxic effects on myocardial contractility. Nevertheless, the profound inhibition of myocardial hypertrophy of the failing myocardium suggests that attenuation of calcineurin-mediated hypertrophy is the predominant action of CsA in the myocardium. It could also be argued that the reduction of ventricular hypertrophy observed in the CHF rats after intervention with CsA might be secondary to reduced MAP. Although we did observe a decrease of MAP in the CHF rats after treatment with CsA, there were no differences in LV wall stress, the major stimulus for cardiac hypertrophy, between the CHF-vehicle and CHF-CsA groups.

In conclusion, the present study demonstrates inhibition of normal cardiac growth as well as prevention of postinfarction myocardial hypertrophy in rats after intervention with CsA. Our data therefore support the hypothesis that calcineurin may be a nodal point in the conversion of elevated intracellular calcium levels to the trophic signals that initiate myocardial hypertrophy. Inhibition of the hypertrophic response by CsA was accompanied by progressive dilatation of the LV cavity and reduction of cardiac function. Therefore, our data indicate that inhibition of ventricular hypertrophy during the early phase after MI exerts detrimental effects on cardiac remodeling and function.