In vivo administration of calpeptin attenuates calpain activation and cardiomyocyte loss in pressure-overloaded feline myocardium

Abstract

Calpain activation is linked to the cleavage of several cytoskeletal proteins and could be an important contributor to the loss of cardiomyocytes and contractile dysfunction during cardiac pressure overload (PO). Using a feline right ventricular (RV) PO model, we analyzed calpain activation during the early compensatory period of cardiac hypertrophy. Calpain enrichment and its increased activity with a reduced calpastatin level were observed in 24- to 48-h-PO myocardium, and these changes returned to basal level by 1 wk of PO. Histochemical studies in 24-h-PO myocardium revealed the presence of TdT-mediated dUTP nick-end label (TUNEL)-positive cardiomyocytes, which exhibited enrichment of calpain and gelsolin. Biochemical studies showed an increase in histone H2B phosphorylation and cytoskeletal binding and cleavage of gelsolin, which indicate programmed cardiomyocyte cell death. To test whether calpain inhibition could prevent these changes, we administered calpeptin (0.6 mg/kg iv) by bolus injections twice, 15 min before and 6 h after induction of 24-h PO. Calpeptin blocked the following PO-induced changes: calpain enrichment and activation, decreased calpastatin level, caspase-3 activation, enrichment and cleavage of gelsolin, TUNEL staining, and histone H2B phosphorylation. Although similar administration of a caspase inhibitor, N-benzoylcarbonyl-Val-Ala-Asp-fluoromethylketone (Z-VD-fmk), blocked caspase-3 activation, it did not alleviate other aforementioned changes. These results indicate that biochemical markers of cardiomyocyte cell death, such as sarcomeric disarray, gelsolin cleavage, and TUNEL-positive nuclei, are mediated, at least in part, by calpain and that calpeptin may serve as a potential therapeutic agent to prevent cardiomyocyte loss and preserve myocardial structure and function during cardiac hypertrophy.

cardiac hypertrophy is initiated as a compensatory mechanism to normalize the increase in wall stress caused by conditions such as hypertension and volume overload (17). The increase in overall cardiac mass, however, is associated with adverse myocardial remodeling, which, if unregulated, results in dilated cardiomyopathy and heart failure. A number of molecular pathways, including hyperactivation of calpain, a cysteine protease activated upon elevated intracellular Ca²⁺ levels (19, 52), have been proposed to cause cardiac adverse remodeling. The calpain family of proteases consists of 15 members; the μ- and m-calpains (calpain I and calpain II, respectively) are the ubiquitous and relatively well-studied isoforms (23). Limited proteolysis of specific substrate proteins by calpain in distinct subcellular locations leads to their altered function in cells, and this makes calpain an important regulator of cellular signaling mechanisms. In the heart, increased calpain activation has been observed in pressure-overloaded (PO) myocardium, and attenuation of calpain activation was found to improve contractile function (26, 27). Similarly, calpain activation and its substrate cleavage have been reported during mechanical unloading (45) and during ischemic preconditioning and myocardial infarction (28, 54), and some of these events are often accompanied by programmed cell death (48, 59). However, whether cell death is due to necrosis, apoptosis, or oncosis is still under debate (2, 32).

A common feature during programmed cell death is the loss of cytoskeletal integrity and cleavage of several cytoskeletal proteins that are mediated by the activation of cytoplasmic proteases (7). In this context, gelsolin, an actin-binding protein, is known to regulate actin organization by severing actin filaments, capping filament ends, and nucleating actin assembly (35, 53). Whereas intact gelsolin plays a cell-protective role (31), the NH₂-terminal gelsolin fragment generated by caspase-3 is known to cause morphological changes reminiscent of apoptosis (7, 22). In addition to cleavage by caspase-3, gelsolin cleavage was demonstrated during calpain activation (66). Since elevated calpain activation has been reported in hypertrophic myocardium (26, 27) and cardiomyocytes (41), we hypothesized that activation of this protease might contribute to myocardial cell death and that the deleterious effects of calpain activation in PO myocardium could be prevented by administration of calpain inhibitors in vivo. Using a feline right ventricular (RV) PO (RVPO) model, we show that calpain and caspase-3 are transiently activated in the early phases of cardiac hypertrophy, which are accompanied by TdT-mediated dUTP nick-end label (TUNEL)-positive cardiomyocytes, indicating programmed myocardial cell death. Our studies also show that these changes are accompanied by actin-cytoskeletal binding, cleavage of gelsolin, and histone H2B phosphorylation. Furthermore, inhibition of calpain, but not caspases, prevented cardiomyocyte death and its associated changes, suggesting that calpain is a potential therapeutic target for preservation of cytoskeletal structure and prevention of myocardial cell loss in hypertrophying myocardium.

MATERIALS AND METHODS

Chemicals.

The protease inhibitors 4-(2-aminoethyl) benzene sulfonyl fluoride, aprotinin, leupeptin, pepstatin A, p-aminobenzamidine, and E-64 [trans-epoxysuccinyl-l-leucylamido-(4-guanidino)butane] and the phosphatase inhibitors p-aminobenzamidine, β-glycerophosphate, okadaic acid, and sodium orthovanadate were obtained from Sigma Chemical (St. Louis, MO) or Calbiochem (San Diego, CA). Tris·HCl, EGTA, and Triton X-100 were obtained from Sigma Chemical. The caspase-3 inhibitor N-benzoylcarbonyl-Val-Ala-Asp-fluoromethylketone (Z-VD-fmk) and calpeptin were obtained from EpiCept (San Diego, CA) and Calbiochem (San Diego, CA), respectively. The calpain activity assay kit was obtained from BioVision Research Products (Mountain View, CA).

Antibodies.

Antibodies against the following proteins were obtained from the indicated vendors: gelsolin for Western blotting from BD Biosciences (San Jose, CA) and for confocal immunostaining from Sigma Chemical and Santa Cruz Biotechnology (Santa Cruz, CA); active caspase-3 for Western blotting from Sigma Chemical and for confocal immunostaining from Promega (Madison, WI); cathepsin B from Calbiochem; cathepsin D from Santa Cruz Biotechnology; α-actinin, α-sarcomeric actin, and actin from Sigma Chemical; phosphorylated histone H2B from Upstate Biotechnologies (Charlottesville, VA); fodrin from Cell Signaling Technology (Danvers, MA); calpastatin from Santa Cruz Biotechnology; horseradish peroxidase-labeled secondary antibodies from Vector Laboratories (Burlingame, CA); fluorescent-labeled secondary antibodies from Jackson ImmunoResearch Laboratories (West Grove, PA) and Molecular Probes (Carlsbad, CA); and GAPDH from Research Diagnostics (Concord, MA). The generation of antibody against calpain has been described previously (4).

Animal model.

RVPO was accomplished by partial occlusion of the pulmonary artery of adult male cats (2.8–3.5 kg body wt), as we described previously (18, 49). In this RVPO model, the systemic arterial pressure remains the same while the pulmonary arterial pressure at least doubles (57, 63). This allows the left ventricle (LV) to serve as same-animal internal control for RVPO. Surgical operations were performed using full surgical anesthesia with ketamine HCl (50 mg/kg im), meperidine (2.2 mg/kg im), and acepromazine maleate (0.25 mg/kg im). PO was achieved via placement of an external 3.2-mm-ID band, and the animals were allowed to recover after anesthesia. These cats were later anesthetized and killed at the specified times, and RV and LV tissue were processed separately as described below.

Sham operation of control cats consisted of thoracotomy and pericardiotomy without arterial occlusion.

The care of the animals and all experiments were conducted in accordance with the US National Institutes of Health guidelines for the care and use of laboratory animals and approved by the Institutional Animal Care and Use Committee of the Medical University of South Carolina.

In vivo administration of drugs.

Calpain and caspase inhibitor studies were performed in 24-h-PO cats. Calpeptin (25 mg) was dissolved in 1 ml of DMSO and further diluted in physiological saline (250 μg/ml). Z-VD-fmk (67) was dissolved in 0.05 M Tris·HCl (pH 8.5, 10 mg/ml). The pH of the drug solutions was adjusted to 7.2 before they were administered. Each drug was given by bolus intravenous injections twice, 15 min before and 6 h after induction of PO. The initial and final doses of calpeptin were 0.6 mg/kg. The first dose of Z-VD-fmk was 20 mg/kg, and the subsequent dose was 10 mg/kg.

Preparation of tissue lysates.

Triton X-100-soluble and -insoluble fractions were prepared from ventricular tissue samples, as we described previously (33). Briefly, a 100-mg ventricular tissue sample was homogenized in 2 ml of ice-cold Triton X-100 extraction buffer [final concentration 100 mM Tris·HCl (pH 7.4), 10 mM EGTA, 2% Triton X-100, 0.5 mM phenylmethysulfonyl fluoride, 10 μg/ml leupeptin, 10 μg/ml aprotinin, 2 μg/ml pepstatin, 2 μM E-64, 200 μg/ml p-aminobenzamidine, 1 μM okadaic acid, 10 mM β-glycerophosphate, and 1 mM sodium orthovanadate]. After initial homogenization and centrifugation at 14,000 g for 10 min, the supernatant was preserved for a subsequent high-speed spin. The pellet (insoluble material) was reextracted with extraction buffer for removal of any remaining detergent-soluble proteins, pelleted, resuspended in 0.5 ml of 1× SDS sample buffer, and boiled to obtain a cytoskeletal fraction. The supernatant from the previous centrifugation step was spun again at a higher speed (100,000 g) for 2.5 h at 4°C. The pellet was solubilized in 1 ml of 1× SDS sample buffer to obtain a membrane skeletal fraction, whereas the supernatant from the high-speed spin was mixed with an equal volume of 2× SDS sample buffer to obtain the soluble fraction.

Western blotting.

Proteins in SDS sample buffer were resolved by SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membranes were blocked for 1 h using 1% BSA and 5% milk in Tris-buffered saline + Tween 20 [TTBS: 10 mM Tris, 0.1 M NaCl, and 0.1% Tween 20 (pH 7.4)]. Blots were incubated with primary antibodies in TBST buffer overnight at 4°C with gentle agitation. Blots were washed five times, each for 5 min, with TBST buffer and then incubated with horseradish peroxidase-conjugated secondary antibody in TBST buffer for 1 h at room temperature. After five washes, each for 5 min, with TBST, the proteins were detected by enhanced chemiluminescence.

Fixation of ventricular tissue samples.

Perfusion fixation of the heart was performed as described previously (51) with minor modifications. Briefly, animals were euthanized, and the carotid artery was cannulated and the descending aorta was clamped. The heart was perfused for 2 min with buffered saline [10 mM sodium-potassium phosphate (pH 7.4)] under physiological pressure and then fixed with freshly prepared 1% formaldehyde for 20–30 min. Samples were excised, cryoprotected with 20% sucrose overnight at 4°C, and snap frozen in methylbutane at −130°C. Sections (15 μm thick) were prepared using a cryomicrotome (Leica) and used for immunostaining. For flash-frozen tissue samples, fresh tissue was embedded in optimal cutting temperature compound, flash frozen in liquid nitrogen, cryosectioned, and then fixed with 4% paraformaldehyde for 15 min at room temperature.

Confocal imaging.

Perfused fixed and fresh-frozen tissue sections were used for immunostaining. Apoptotic cell death was detected by TUNEL according to the manufacturer's protocol (Chemicon, Temecula, CA). For immunostaining, sections were blocked with 10% normal donkey serum for 1 h at room temperature and incubated overnight at 4°C with the following primary antibodies: mouse anti-gelsolin (1:100 dilution), mouse anti-α-actinin (1:200 dilution), mouse anti-α-sarcomeric actin (1:200 dilution), rabbit anti-active caspase-3 (1:200 dilution), or rabbit anti-calpain (1:100 dilution) antibody. The sections were washed with PBS and incubated for 2 h at room temperature in the appropriate secondary antibody [Cy5- or FITC-labeled anti-mouse IgG, Cy3-labeled anti-rabbit IgG, or Cy5-labeled anti-goat IgG (Jackson ImmunoResearch Laboratories)]. Nuclei were stained with TO-PRO-3 or 4′,6-diaminino-2-phenylindole (Molecular Probes). The fluorescence staining was visualized with a laser scanning confocal microscope (Fluoview, Olympus Optical, Tokyo, Japan) or a fluorescence microscope (model IX71, Olympus Optical) that was customized for a UV light source.

Measurement of calpain activity.

Calpain activity was measured using a kit from BioVision. Tissue lysates were prepared from LV and RV of sham and RVPO cats by Dounce homogenization in extraction buffer provided with the kit. The homogenates were centrifuged at 10,000 g for 5 min at 4°C, and the supernatants containing 200 μg of protein were used for calpain assay, as described by the manufacturer. The activity of calpain was measured in triplicates as the rate of cleavage of a calpain substrate, Ac-leucine-leucine-tyrosine-7-amino-trifluoromethyl coumarin (Ac-LLY-AFC). Free AFC emits a yellow-green fluorescence, which was detected using a fluorescence microplate reader (Molecular Devices; 400-nm excitation filter, 505-nm emission filter). Fluorescence per milligram protein of each sample was calculated, and relative fluorescence was compared with sham LV.

Statistical analysis.

Values are means ± SE. Statistical analysis was performed by two-tailed Student's t-test. P < 0.05 was considered to be statistically significant.

RESULTS

Feline RVPO model.

Hemodynamic data for our experimental feline model are shown in Table 1. In this feline RVPO model, external placement of a 3.2-mm-ID band on the pulmonary artery induced moderate pressure on the RV. Our earlier studies (58) indicate that the RV wall stress (load per unit ventricular mass) in these PO cats increased in the first few days of PO and then was normalized by the compensatory growth by the end of the 1st wk of PO. In all cases, the systemic pressure of cats in this model remained unaltered (63), and the LV served as a same-animal normally loaded control, in addition to sham control ventricles. We used this model system to study calpain activation and its importance during PO hypertrophy.

Table 1.

Hemodynamic parameters measured in RVPO cats

	n	RV/BW, g/kg	RVSP, mmHg	RVEDP, mmHg
Sham	5	0.68±0.05	26.96±3.99	3.30±2.45
24 h RVPO	6	0.70±0.05	50.62±14.28*	6.82±2.17*
48 h RVPO	7	0.71±0.06	50.49±13.13*	6.27±2.25*
1 wk RVPO	7	0.85±0.08*	49.33±11.35*	6.33±1.84*

Values are means ± SD; n, number of cats. RVPO, right ventricular (RV) pressure overload; BW, body weight; RVSP, RV systolic pressure; RVEDP, RV end-diastolic pressure.

^*P < 0.05 vs. respective sham value.

Calpain activation in PO myocardium.

To demonstrate calpain activation in the early period of PO feline myocardium, immunohistochemical staining was performed using calpain antibody (4) in PO RVs for 24 h, 48 h, and 1 wk and compared with normally loaded same-animal control LVs and sham control LV and RV. Whereas ventricular tissue (LV and RV) samples from sham control and 1-wk-RVPO cats did not stain for calpain, 24-h-PO and 48-h-PO RV samples, but not their respective control LVs, exhibited calpain staining. Although the antibody does not distinguish between active and inactive forms of calpain, enrichment of calpain in 24-h-PO to 48-h-PO myocardium is evident. Although detection by this antibody has been used to determine calpain activation in other studies (5, 43), we measured calpain activity directly in ventricular tissue extracts. Previous studies have suggested that in vitro measurement of calpain activity in the presence of Ca²⁺ exhibits predominantly the total activity, rather than the physiological activity (20). However, other independent studies (14) have used an in vitro assay to show increased calpain activity in ischemic heart. Therefore, we performed a similar analysis that demonstrated a twofold increase in 24-h-PO and 48-h-PO RV samples compared with normally loaded same-animal control LV or normally loaded sham control RV or LV (Fig. 1B). However, this increased calpain activity returned to control levels by 1 wk of PO. Importantly, this time course of activation is similar to calpain enrichment in PO RV samples (Fig. 1A). Next, to analyze whether the calpain-associated changes occurred in cardiomyocytes, we immunostained ventricular tissue samples from 24-h-PO cats with calpain and α-actinin antibodies (Fig. 1C). Cells with enriched calpain staining were observed in PO RV, but not in normally loaded LV, and they costained with α-actinin, suggesting that the changes occur in PO cardiomyocytes.

Fig. 1.

Calpain activation in cardiomyocytes of pressure-overloaded (PO) myocardium. A: perfusion-fixed ventricular tissue samples from sham control cats and from cats subjected to right ventricular pressure-overload (RVPO) for 24 h, 48 h, and 1 wk were processed for histochemical staining. Sections were processed for immunostaining using anti-calpain (red) antibody and nuclear staining (4′,6-diaminino-2-phenylindole, blue). LV, left ventricle; RV, right ventricle. Scale bar, 10 μm. B: cytoplasmic extracts prepared from sham control, 24-h-RVPO, 48-h-RVPO, and 1-wk-RVPO cat ventricles (LV and RV) were used for measurement of calpain activity. For each sample, the fluorescence emitted upon cleavage of the calpain substrate was measured and normalized to protein concentration and is shown as relative fluorescence. *P < 0.05 vs. LV of RVPO and LV and RV of sham control. C: perfusion-fixed ventricular (LV and RV) sections from 24-h-RVPO cats were immunostained using anti-α-actinin (green) and anti-calpain (red) antibodies. Scale bar, 10 μm. D: ventricular tissue (LV and RV) samples from sham control, 24-h-RVPO, 48-h-RVPO, and 1-wk-RVPO cats were subfractioned into Triton X-100-soluble (Sol) and -insoluble fractions. Samples were used for detection of calpastatin by Western blot analysis with anti-calpastatin antibody. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of all cats were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions.

Finally, we measured the level of calpastatin, an endogenous inhibitor of calpain, in control and RVPO feline ventricles. Calpastatin is an endogenous inhibitor and a potential substrate of calpain, and a decreased level of calpastatin via calpain-mediated degradation is known to substantially promote calpain activity (20, 56). For measurement of calpastatin level, tissue samples were processed to obtain Triton X-100-soluble and -insoluble low- and high-spin fractions, as we described previously (33). Whereas extraction of tissue or cell samples with Triton X-100 buffer releases cytoplasmic and other detergent-soluble proteins, our previous studies (33, 36) in PO myocardium demonstrate that signaling proteins are often recruited to the detergent-insoluble actin-rich (cytoskeletal) protein complex. Therefore, sham control and RVPO feline tissue samples were used to prepare Triton X-100-soluble and -insoluble fractions. Previous studies (21, 33) demonstrate that the Triton X-100-insoluble low-spin fraction, which is rich in actin, primarily represents the cytoskeletal fraction, and the high-speed centrifugation of the first supernatant produces another insoluble fraction that contains a smaller amount of actin, representing the membrane skeletal fraction. During SDS-PAGE, protein loading for soluble and insoluble fractions was monitored by immunoblotting with anti-GAPDH and anti-actin antibodies, respectively. As shown in Fig. 1D, calpastatin was present predominantly in the detergent-soluble fraction, and, in the case of the sham control cat, RV and LV samples exhibited equal levels of this endogenous inhibitor of calpain. However, calpastatin was substantially decreased in 24-h-PO to 48-h-PO, but not 1-wk-PO, RV compared with normally loaded control LV. This decrease in calpastatin level was not due to its recruitment to the detergent-insoluble compartments, since this protein was not present in the Triton X-100-insoluble fractions of normally loaded (LV or RV) or PO (RV) myocardium. Therefore, the calpain enrichment and activation in PO cardiomyocytes are accompanied by a decrease in calpastatin level.

Cardiomyocyte death in PO myocardium.

Earlier reports in the heart (55) and other tissues (42) implicate calpain involvement in cell death. Since we observed calpain enrichment/activation in 24-h-PO myocardium, we first analyzed whether the time course of calpain enrichment coincides with programmed cell death. Ventricular samples were subjected to 0 h, 24 h, 48 h, and 1 wk of PO and then stained with TUNEL, an indicator of DNA fragmentation and programmed cell death (Fig. 2A). Whereas sham controls (LV or RV) did not exhibit TUNEL staining, 24-h-PO and 48-h-PO, but not 1-wk-PO, cardiomyocytes exhibited TUNEL-reactive cells compared with normally loaded same-animal control LVs, and this time course of TUNEL reactivity was similar to our earlier findings with calpain enrichment/activation. Since 24-h-PO RV samples consistently showed higher levels of TUNEL-positive cells, we quantitated TUNEL-reactive nuclei using three 24-h-RVPO and three sham-operated cats. By counting ∼1 × 10⁵ nuclei, we estimated the number of TUNEL-positive nuclei in the RV to be 0.8% (Fig. 2B), whereas the same-animal LV and sham ventricles exhibited negligible TUNEL reactivity. To provide supportive evidence of programmed cell death in the PO RV, we performed Western blot analysis to detect Ser¹⁴-phosphorylated histone H2B, which has been reported to be a marker for programmed cell death (12). Phosphorylated histone H2B, which was absent in all fractions of normally loaded control ventricles (same-animal LV and sham animal LV and RV), was readily seen in the detergent-insoluble high-spin fraction and, to a lesser extent, in the detergent-insoluble low-spin RVPO fraction at 24 h (Fig. 2C), when a substantial level of TUNEL-positive nuclei was observed. This phosphorylation pattern was also present in 48-h-RVPO fractions, although to a lesser extent, and was not observed in 1-wk-RVPO samples. Finally, to establish that TUNEL-positive cells represent cardiomyocytes, we simultaneously performed TUNEL and α-actinin staining (Fig. 2D). Our studies clearly demonstrate that the TUNEL-positive cells in RVPO samples are indeed cardiomyocytes. Furthermore, α-actinin staining indicated sarcomeric disruption of cardiomyocytes, shown by the absence of Z lines in the RV (Fig. 2D). This observation was similar to the findings with calpain-enriched cardiomyocytes (Fig. 1B). Together, these data suggest that PO triggers transient programmed cell death during the 1st wk, primarily within 24 h, which accounts for ∼1% of myocardial cells.

Fig. 2.

Programmed cell death of cardiomyocytes in PO myocardium. A: perfusion-fixed ventricular (LV and RV) samples from sham control cats and from cats subjected to RVPO for 24 h, 48 h, and 1 wk were processed for TUNEL (green) and nuclear (blue) staining. B: percentage of TUNEL-positive cells was quantitated in both LV and RV of sham control and 24-h RVPO samples. Approximately 10⁵ nuclei were counted in ≥3 independent cats for each group. *P < 0.05 vs. LV of RVPO and LV and RV of sham control. C: ventricular tissue (LV and RV) samples prepared from sham control, 24-h-RVPO, 48-h-RVPO, and 1-wk-RVPO cats were subfractioned into Triton X-100-soluble and -insoluble fractions and used for detection of phosphorylated (phospho) histone-H2B by Western blot analysis with specific antibody. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of all cats were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. D: perfusion-fixed ventricular (LV and RV) samples from 24-h-RVPO cats were processed for immunostaining with α-actinin antibody (red) and TUNEL (green) and nuclear (blue) staining. Scale bar, 10 μm.

Gelsolin-associated changes in calpain- and TUNEL-positive 24-h-PO cardiomyocytes.

Cleavage of several cytoskeletal proteins by cytoplasmic proteases has been shown to disrupt actin cytoskeleton, causing programmed cell death, and gelsolin, a protease cleaved by caspase-3 (30) and calpain (66), has been shown to be one such major player (30). Therefore, we analyzed whether cardiomyocytes with disrupted sarcomeric structure in 24-h-PO myocardium stain positive for gelsolin. Our confocal immunostaining studies for α-actinin and gelsolin clearly demonstrate gelsolin enrichment in PO cardiomyocytes showing disrupted sarcomeric structure (Fig. 3A). We also performed biochemical analyses to demonstrate actin-cytoskeletal binding and cleavage of gelsolin. Sham control and 24-h-RVPO ventricular tissue samples were processed to yield Triton X-100-soluble and -insoluble low- and high-spin fractions and used for Western blot analyses with gelsolin antibody (Fig. 3B). In sham control ventricles, the full-length gelsolin was found in equal quantities in LV and RV and was predominantly present in the detergent-soluble fraction. In 24-h-PO RV, gelsolin levels in the detergent-insoluble low-spin (cytoskeletal) fraction were substantially increased compared with the normally loaded control LV fraction. On the basis of the ratio of protein loading, this recruited gelsolin level was ∼30% of the total gelsolin level. More importantly, part of the cytoskeleton-bound gelsolin in PO RV was cleaved, yielding a 65-kDa fragment. Caspase and calpain have been shown to cleave gelsolin, generating fragments of similar size. This cleavage product was observed in the PO RV, but not in the same-animal LV control or in ventricles from sham controls. Furthermore, this change in gelsolin was present in 24-h-PO to 48-h-PO RV, but not in sham control and 1-wk-PO cats (data not shown).

Fig. 3.

Enrichment of gelsolin in calpain and TUNEL-positive cardiomyocytes of PO myocardium. A: perfusion-fixed ventricular (LV and RV) samples from 24-h-RVPO cats were processed for immunostaining with anti-α-actinin (red) and anti-gelsolin (blue) antibodies. Scale bar, 10 μm. B: ventricular tissue (LV and RV) samples prepared from sham control and 24-h-RVPO cats were subfractioned into Triton X-100-soluble (Sol) and -insoluble fractions and used for detection of gelsolin by Western blot analysis with anti-gelsolin antibody that was raised against the middle region (amino acids 592–768) of gelsolin. Positions of full-length (80 kDa) and cleaved (65 kDa) gelsolin are indicated by closed and open arrows, respectively. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of all cats were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. C: perfusion-fixed ventricular (LV and RV) samples from 24-h-RVPO cats were processed with anti-gelsolin antibody (blue) and for TUNEL (green) and nuclear (red) staining. Scale bar, 10 μm. D: perfusion-fixed ventricular (LV and RV) samples from 24-h-RVPO cats were processed for immunostaining with anti-calpain (red) and anti-gelsolin (blue) antibodies and for TUNEL (green) and nuclear (purple) staining. Scale bar, 10 μm.

We next analyzed whether gelsolin-enriched cardiomyocytes undergo programmed cell death in PO myocardium. Ventricular tissue samples from a 24-h-RVPO cat were processed for TUNEL and gelsolin, as well as nuclear, staining. In PO RV, but not in normally loaded LV, cardiomyocytes that costain for gelsolin and TUNEL were readily observed (Fig. 3C). The gelsolin- and TUNEL-reactive cardiomyocytes with disrupted morphology could also be seen in the phase-contrast images of PO RV sections. Finally, we also analyzed whether calpain-enriched cardiomyocytes in 24-h-PO myocardium stain positive for gelsolin and TUNEL. Again, in 24-h-PO RV, but not in normally loaded LV, cardiomyocytes showed positive staining for calpain, gelsolin, and TUNEL and their colocalization (Fig. 3D).

Attenuation of cardiomyocyte death in PO myocardium by calpeptin administration.

To study whether activated calpain in PO myocardium leads to sarcomeric disruption and cardiomyocyte death, we used calpeptin, a pharmacological inhibitor of calpain (43). Calpeptin (0.6 mg/kg iv) was administered 15 min before PO and again 6 h after the surgery. Hemodynamic data of calpeptin-treated cats are shown in Table 2. These 24-h-RVPO cats showed an increase in RV pressure similar to non-calpeptin-treated 24-h-RVPO (Table 1), and their systemic pressure was unaltered (data not shown). Ventricular tissue samples from untreated or calpeptin-treated were then processed for immunohistochemical and biochemical studies. In PO RV, but not in normally loaded control LV (data not shown), cardiomyocytes stained positive for calpain, gelsolin, and TUNEL (Fig. 4A), as observed in our previous experiments (Fig. 3D). However, in calpeptin-treated cats, PO-induced calpain and gelsolin staining and TUNEL reactivity were completely abolished (Fig. 4A), indicating that calpain is a major mediator of these changes. About 0.8% of cardiomyocytes stained positive for gelsolin, TUNEL, or calpain, and the level of costaining (gelsolin + TUNEL, gelsolin + calpain, and gelsolin + TUNEL + calpain) was similar (∼0.8%). Upon calpeptin administration, individual staining of these markers, as well as their colocalization, was almost abolished (Fig. 4B).

Table 2.

Hemodynamic parameters measured in Z-VD-fmk-treated 24-h-RVPO cats

RV/BW, g /kg	RVSP, mmHg	RVEDP, mmHg
0.71±0.07	44.93±10.75*	6.93±1.79*

Values are means ± SE (n = 4). Z-VD-fmk, N-benzoylcarbonyl-Val-Ala-Asp-fluoromethylketone.

^*P < 0.05 vs. respective sham value in Table 1.

Fig. 4.

Effect of in vivo administration of calpeptin on calpain-, TUNEL-, and gelsolin-associated changes in PO myocardium. Perfusion-fixed RV samples from 24-h-RVPO cats that had been treated with vehicle (24 h PO-RV) or calpeptin (24 h PO-RV + calpeptin) were processed for histochemical staining. A: sections were processed for immunostaining with anti-calpain (red) and anti-gelsolin (blue) antibodies and for TUNEL (green) and nuclear (purple) staining. Scale bar, 10 μm. B: quantitation of gelsolin (Gel)-, calpain (Calp)-, and TUNEL-positive cells and their colocalization were accomplished using ≥3 independent 24-h-RVPO cats in each of the drug-treated and untreated groups, as described in Fig. 2 legend. *P < 0.05 vs. PO-RV without drug. C: ventricular tissue (LV and RV) samples from 24-h-RVPO cats, which were treated with vehicle or calpeptin in vivo, were subfractioned into Triton X-100-soluble and -insoluble fractions. Samples were used for detection of phosphorylated histone H2B and gelsolin by Western blot analysis with respective antibodies. Positions of full-length (80 kDa) and cleaved (65 kDa) gelsolin are indicated by closed and open arrows, respectively. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of all cats were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. D: ventricular tissue (LV and RV) samples from a sham control cat and 24-h-RVPO cats that were treated in vivo with calpeptin were subfractioned into Triton X-100-soluble and -insoluble fractions. Samples were used for Western blot analysis with anti-calpastatin antibody. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of all cats were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. For 24-h-RVPO and 24-h-RVPO + calpeptin, normalization of protein concentration in each pair of LV and RV is shown in C.

To further explore the effect of calpeptin treatment, we performed biochemical studies using Triton X-100-soluble and -insoluble fractions. As shown in our previous experiments, the untreated 24-h-PO cats exhibited phosphorylated histone H2B and cytoskeletal assembly and cleavage of gelsolin (Fig. 4C). However, calpeptin treatment blocked these changes, and this effect is similar to the loss of TUNEL staining and gelsolin enrichment in PO feline cardiomyocytes (Fig. 4A). Finally, we also analyzed whether calpeptin treatment blocks the loss of calpastatin during PO. As observed in our earlier experiments (Fig. 1C), the calpastatin level was similar in LV and RV samples of sham control cats (Fig. 4D). However, calpastatin level was substantially reduced in 24-h-PO RV compared with the normally loaded same-animal control LV. Importantly, calpeptin treatment during PO nearly restored calpastatin to the level observed in normally loaded control LV. Together, our studies demonstrate that calpeptin blocks calpain activation, gelsolin cleavage, and programmed cell death in PO myocardium.

Caspase-associated changes in PO myocardium.

Although the present data indicate the involvement of calpain in TUNEL staining and cleavage of cytoskeletal proteins in PO myocardium, it is possible that other proteases, such as caspase-3, also contribute to these processes. Ample evidence exists for caspase-mediated apoptosis (10, 39, 46, 65), and gelsolin has been shown to be a potential substrate for caspase-3 where the generated product, NH₂-terminal gelsolin, would severe actin polymers in a Ca²⁺-independent, unregulated manner (3, 30). Therefore, we performed immunostaining using an antibody against active caspase-3 in ventricular tissue samples from a 24-h-RVPO cat (Fig. 5A). Our studies clearly demonstrate active caspase-3-positive cells in PO RV, but not in normally loaded control LV (Fig. 5A). Furthermore, staining for gelsolin and TUNEL, along with active caspase-3, demonstrates their colocalization only in PO RV. By coimmunostaining with α-actinin antibody (data not shown), we also confirmed that most of the active caspase-3-positive cells were indeed cardiomyocytes.

Fig. 5.

Effect of in vivo administration of N-benzoylcarbonyl-Val-Ala-Asp-fluoromethylketone (Z-VD-fmk) on caspase-3-, calpain-, TUNEL-, and gelsolin-associated changes in PO myocardium. A: perfusion-fixed ventricular (LV and RV) samples from a 24-h-RVPO cat were processed for histochemical staining. Sections were processed for immunostaining with anti-active caspase-3 (red) and anti-gelsolin (blue) antibodies and for TUNEL (green) and nuclear (purple) staining. Scale bar, 10 μm. B: perfusion-fixed RV samples from 24-h-RVPO cats, which were adminstered with either vehicle (24 h PO-RV) or Z-VD-fmk (24 h PO-RV + Z-VD-fmk), were processed for histochemical staining. Sections were processed for immunostaining with anti-active caspase-3 (red) and anti-gelsolin (blue) antibodies and for TUNEL (green) and nuclear (purple) staining. Scale bar, 10 μm. C: quantitation of caspase-3 (Casp 3)-, gelsolin (Gel)-, and TUNEL-positive cells and their colocalization were accomplished using ≥3 independent 24-h-RVPO cats in each of the drug-treated and untreated groups as described in Fig. 2 legend. *P < 0.05 vs. PO-RV without drug treatment. D: perfusion-fixed RV samples from 24-h-RVPO cats treated with vehicle (24 h PO-RV) or Z-VD-fmk (24 h PO-RV + Z-VD-fmk) were processed for histochemical staining. Sections were processed for immunostaining with anti-calpain (red) and for nuclear (blue) staining. Scale bar, 10 μm. E: cytoplasmic extracts from ventricular (LV and RV) samples were prepared from 24-h-RVPO cats treated with vehicle, Z-VD-fmk, or calpeptin and used for measurement of calpain activity. For each sample, fluorescence emitted upon cleavage of the calpain substrate was measured, normalized to protein concentration, and shown as relative fluorescence. *P < 0.05 vs. LV of the RVPO cat. #P < 0.05 vs. RV of the 24-h-RVPO cat without drug treatment. F: ventricular (LV and RV) samples from 24-h-RVPO cats treated with vehicle or Z-VD-fmk in vivo were subfractioned into Triton X-100-soluble and -insoluble fractions, and samples were used for detection of phosphorylated histone-H2B, active caspases-3, gelsolin, and calpastatin by Western blot analysis with respective antibodies. Positions of full-length (80 kDa) and cleaved (65 kDa) gelsolin are indicated by closed and open arrows, respectively. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. G: ventricular (LV and RV) samples from 24-h-RVPO cats treated with vehicle or calpeptin in vivo were subfractioned into Triton X-100-soluble and -insoluble fractions, and samples were used for detection of caspase-3 activation by Western blot analysis with anti-active caspase-3 antibody. Protein concentrations in each pair of LV and RV samples of Triton X-100-lysed subfractions of 24-h-RVPO cats treated with calpeptin were adjusted by Western blot analysis with anti-GAPDH antibody for Triton X-100-soluble fraction and with anti-actin antibody for Triton X-100-insoluble fractions. For 24-h-RVPO cats treated with vehicle alone, protein normalization in each pair of LV and RV is shown in F.

Next, to demonstrate whether inhibition of caspase-3 activation would block gelsolin cleavage and reduce cell death, we used a specific caspase inhibitor, Z-VD-fmk, which has been shown to block caspases in vivo (67). Z-VD-fmk (20 mg/kg iv) was administered to 24-h-RVPO cats by a bolus injection 15 min before the initiation of PO again 6 h after surgery. Hemodynamic data of Z-VD-fmk-treated cats are shown in Table 3. These 24-h-RVPO cats exhibited increased RV pressure similar to the non-Z-VD-fmk-treated 24-h-RVPO cats (Table 1), while their systemic pressure remained unaltered (data not shown). Untreated and Z-VD-fmk-treated ventricular tissue samples from 24-h-RVPO cats were then used for immunostaining and Western blot analysis. Similar to the previous observation (Fig. 5A), PO RV in the absence of drug treatment showed costaining of gelsolin, TUNEL, and active caspase-3 (Fig. 5B). However, in Z-VD-fmk-treated cats, PO-induced activation of caspase-3 in RV was almost completely abolished, although gelsolin enrichment and TUNEL staining were still present. These observations were confirmed in two additional cats. To determine whether Z-VD-fmk had a partial effect on gelsolin and/or TUNEL staining, we counted 10⁵ nuclei and determined the number of cells that stained positive for gelsolin, TUNEL, and caspase-3 (Fig. 5C). In the untreated PO RV, the TUNEL-positive nuclei accounted for 0.9% of all nuclei. Furthermore, gelsolin and active caspase-3 costaining was observed in ∼0.6% of the counted nuclei. Such analysis in the Z-VD-fmk-treated PO RV revealed that the level of gelsolin- and TUNEL-positive cells was unaffected by the drug treatment. However, the level of active caspase-3 and its costaining with gelsolin and TUNEL were significantly reduced (Fig. 5C). Next, we analyzed whether Z-VD-fmk, which blocks load-induced caspase-3 activation, affects calpain enrichment. As shown in our earlier experiment (Figs. 1A and 3D), calpain enrichment was observed in 24-h-PO myocardium (Fig. 5D), and this trend was found to be unaffected by Z-VD-fmk. To further study the effect of calpeptin and Z-VD-fmk on calpain, we directly measured the activity of calpain in 24-h-PO tissue samples that were treated with these drugs and compared these drug-treated samples with samples from non-drug-treated PO cats (Fig. 5E). Again, calpain activity was enhanced in 24-h-PO RV compared with the normally loaded same-animal counterpart, and this observation is similar to our earlier findings (Fig. 1B). Z-VD-fmk did not affect this PO-induced activation of calpain, although calpeptin blocked this activation. These data suggest that Z-VD-fmk specifically blocked caspases, and not calpains, and that the loss of caspase-3 activation in PO myocardium was not sufficient to block the associated myocardial cell loss.

Table 3.

Hemodynamic parameters measured in calpeptin-treated 24-h-RVPO cats

RV/BW, g /kg	RVSP, mmHg	RVEDP, mmHg
0.71±0.06	50.90±3.57*	6.47±2.30*

Values are means ± SE (n = 4).

^*P < 0.05 vs. respective sham value in Table 1.

To further confirm that Z-VD-fmk-induced inhibition of caspases was not sufficient to block programmed cell death, we performed biochemical analyses in 24-h-RVPO feline samples with and without Z-VD-fmk. Z-VD-fmk did not affect PO-induced histone H2B phosphorylation, although caspase-3 activation was blocked (Fig. 5F). Similarly, PO-induced cytoskeletal recruitment and cleavage of gelsolin (Fig. 5F) and fodrin (data not shown) were unaffected by Z-VD-fmk, supporting our histochemical studies. Importantly, under these conditions, Z-VD-fmk did not prevent the PO-induced decline in calpastatin, supporting our observation that Z-VD-fmk does not affect load-induced calpain activation. All these data strongly suggest that Z-VD-fmk affected neither calpain activation nor cardiomyocyte death in PO myocardium. Finally, to explore whether the caspase-3 activation in PO myocardium occurs downstream of calpain activation, we analyzed the level of active caspase-3 in 24-h-RVPO cats with and without calpeptin. Whereas active caspase-3 was observed in the absence of drug treatment, it was absent in calpeptin-treated cats (Fig. 5G). These data indicate that, in contrast to Z-VD-fmk, which affects only caspases, calpeptin blocks caspase-3, in addition to its inhibitory effect on calpain.

DISCUSSION

The calpain system, consisting of large and small subunits of calpain and the inhibitory subunit calpastatin, regulates numerous cellular processes, including cell migration, cytoskeletal rearrangement, autophagy, and apoptosis. The dysregulation of the calpain system, especially hyperactivation of calpain, is linked to multiple pathological conditions, such as muscular disorders (42) and cardiac dysfunction (55). Thus targeting calpain for amelioration of such disorders is underway. Our histochemical studies in 24-h-PO to 48-h-PO myocardium demonstrate enrichment of calpain in TUNEL-positive cardiomyocytes. To demonstrate that this enrichment is reflective of calpain activation, we measured its activity directly in the cytosolic extracts of control and PO feline ventricles. Although Greyson et al. (26) suggest in vitro measurements of calpain activity in tissue extracts for determination of the total activity, consisting of stimulated and latent activities, other independent studies in ischemic reperfused hearts utilized this method to demonstrate calpain activation (14). Our analyses show increased calpain activity and decreased calpastatin level in 24-h-PO to 48-h-PO myocardium that returns to basal level by 1 wk of PO, suggesting that PO causes a transient activation of calpain. In support of our findings, Greyson et al. demonstrated calpain activation in an acute PO model, and blockade of this activation was shown to improve ventricular function. Our present study suggests that a potential mechanism for the improved ventricular function upon calpain inhibition could be blockade of the load-induced programmed cell death in cardiomyocytes. Previous studies demonstrate that the loss of contractile mass and function during hypertensive cardiomyopathy correlates to the loss of cardiomyocytes (32).

Cellular activation of calpain results in the cleavage of several of its substrate proteins, and the generated protein fragments often have modified functions. Multiple proteins have been identified as calpain substrates, and cellular localization of these substrates and activated calpain dictates a specific pattern of protein degradation under a given condition. Although degradation of many such proteins may contribute to cardiomyocyte death, we primarily focused on gelsolin-associated changes because of its cytoskeletal localization, cleavage, and role in actin dynamics. In this context, although intact gelsolin displays antiapoptotic activity (34), the cleaved NH₂-terminal gelsolin fragment, which is catalytically active, severs actin and causes cell death (3, 8, 30). We primarily analyzed gelsolin, since it is highly expressed in the heart (24), and it is cleaved by caspases and calpain. Our histochemical studies clearly demonstrate that ∼70% of gelsolin-enriched cells stain positive for TUNEL (Fig. 4B), and most, if not all, of these cells represent cardiomyocytes (Fig. 1A). Furthermore, our biochemical studies demonstrate recruitment to the actin-rich cytoskeletal fraction and cleavage of gelsolin in 24-h-PO myocardium. Importantly, both of these processes could be prevented by the administration of calpain, but not caspase, inhibitor in vivo. Previous studies (13) demonstrate that cytoplasmic actin binds and inhibits the activity and nuclear translocation of DNase I and that the NH₂-terminal fragment of gelsolin disrupts the actin-DNase I interaction, resulting in the activation and nuclear localization of DNase I for programmed cell death. Although we show a strong correlation between gelsolin-associated changes and programmed cell death in PO myocardium, further studies are needed to establish whether gelsolin cleavage contributes to PO-induced cardiomyocyte death.

Several previous studies using animal models of hypertrophy report cardiomyocyte apoptosis and its importance during heart failure. A marked increase in apoptosis in the first 7 days after PO with a peak at day 4 has been reported in rats (61). Similarly, in response to chronic PO in rats, a significant loss of cardiomyocytes via apoptosis was shown to contribute to the transition from LV hypertrophy to LV dysfunction (16). In humans, cardiomyocyte apoptosis has been shown to increase in hypertensive patients with LV hypertrophy, and that level was even higher during hypertension and chronic heart failure (25). Our studies showing 1% TUNEL-positive cardiomyocytes in 24-h-PO to 48-h-PO myocardium could be an important contributor to the development of contractile dysfunction for the following reasons: 1) previous studies in transgenic mice have demonstrated that even a low level of myocyte apoptosis (0.02%) is sufficient to cause lethal dilated cardiomyopathy (65), and 2) in addition to the input by the loss of cardiomyocytes, the viable myocytes that have not completed their cell death process could also contribute to the contractile dysfunction (15, 39). Therefore, our present study, which shows the activation of calpain, cleavage of cytoskeletal proteins, and the associated myocardial cell loss in PO myocardium, indicates a maladaptive process, where controlling such cell death could benefit patients with hypertrophic cardiomyopathy. In support of this view, our ongoing independent work indicates that calpeptin administration in PO rat heart improves ventricular function (37), suggesting that calpain could be a potential therapeutic target to preserve ventricular function in PO myocardium. Similarly, studies from other laboratories have demonstrated an increase in calpain activation in PO myocardium, and attenuation of such activation was found to improve contractile function (26, 27). Two potential mechanisms may contribute to calpain activation in PO myocardium. 1) PO-induced activation of angiotensin II type 1 receptor can increase endoplasmic reticulum (ER) stress (40). In this context, ER stress-induced apoptosis was shown to require calpain for the caspase-12 activation (60), and the ER stress-induced caspase-12 activation was shown to function upstream of caspase-3 while mediating apoptotic cell death (1). 2) Activation of matrix metalloproteinases in PO myocardium may result in the generation of cleaved collagen products that interact with α₂-integrin, resulting in calpain activation (9). Therefore, one or both of these pathways, which are activated during PO, may contribute to calpain activation and the associated programmed cell death.

Our studies showing the importance of calpain in mediating cardiomyocyte death in PO myocardium are based on in vivo treatment of calpeptin, a peptide aldehyde and a potent inhibitor of calpains (5). In rats with acute intracranial hypertension, calpeptin at 3.5 mg·rat⁻¹·day⁻¹ for 1 wk results in improved ventricular function (27). In this context, in the present study, administration of calpeptin at a far lower dose (0.6 mg/kg, twice per day) abolished calpain enrichment in cardiomyocytes and restored the calpastatin level. Furthermore, gelsolin-associated changes and cleavage of other potential calpain substrates, including its own endogenous inhibitor calpastatin, were absent when calpeptin was administered at this concentration during PO. Although calpeptin is known to specifically block calpain, a low-level inhibition has been reported on the cathepsin family of proteases (6, 50). However, our Western blot analysis in 24-h-PO myocardium (data not shown) revealed that no active (cleaved) forms of cathepsin B and cathepsin D isoforms are involved in cardiomyocyte apoptosis (47), suggesting that the calpeptin effect is not mediated by the loss of cathepsin proteases.

Finally, since a large body of evidence supports caspase-3-mediated apoptosis, we tested whether the caspase-3 system is active and amenable to inhibition by its inhibitors in PO hypertrophy. We used a specific pharmacological agent, Z-VD-fmk, which is known to block caspases in vivo and in vitro (67). Our studies demonstrate that Z-VD-fmk completely blocks caspase-3 activation in PO myocardium but does not affect the level of TUNEL-positive cardiomyocytes or gelsolin-associated changes. Furthermore, the PO-induced drop in calpastatin level was not restored by Z-VD-fmk treatment, suggesting that calpain activation under these conditions was unaffected. All these studies show that a caspase-independent mechanism involving calpain may contribute to the programmed cell death in PO myocardium. In support of our observation, several studies demonstrate the presence of a calpain-mediated mechanism independent of caspase-3 for apoptotic cell death (11, 64). Whereas Z-VD-fmk treatment, which affected caspase-3, did not block the PO-induced calpain activation, calpeptin treatment blocked calpain and caspase-3 (Fig. 5). In this context, previous studies demonstrate that calpain can function as an upstream activator of caspase-3 in ischemic glial cells (38) and that it does not have a significant effect directly on caspase-3 activity (44). These data suggest the upstream role of calpain on caspase-3 activation in PO myocardium, where blockade of calpain activity could result in the loss of caspase-3 function. In support of this idea, a recent study demonstrates that a calpain-dependent calpastatin cleavage mechanism regulates caspase-3 activation during apoptosis of Jurkat T cells stimulated by Entamoeba histolytica, and under these conditions, calpeptin, but not Z-VAD-fmk, abolished calpastatin degradation and caspase-3 activation (29a). Additional studies performed in a calpain-null background are needed to establish the importance of calpain activation and confirm whether calpeptin-mediated protection of cardiomyocyte loss in PO myocardium is primarily mediated by calpain activation.

Although several markers of apoptosis, including TUNEL-positive cardiomyocytes, caspase-3 activation, histone H2B phosphorylation, and calpain enrichment/activation, were observed in PO myocardium, cardiomyocyte death under these conditions might occur through necrosis, since the pathway leading to either type of cell death is quite interchangeable (29). In this context, calpain-mediated release of DNase II from the lysosome of ischemic neuronal cells was shown to contribute to necrotic cell death (62). Whereas calpain and cathepsin under these conditions degrade cytoplasmic proteins, DNase II is expected to participate in the DNA degradation. Therefore, it is possible that although an apoptosis cascade is activated, cardiomyocyte death may occur via necrosis in the early period of PO myocardium, where calpain might be the key mediator.

Overall, our data indicate that PO activates the calpain system in the myocardium, and calpeptin administration blocks the activation of calpain and caspase-3, cleavage of their substrates, and cardiomyocyte programmed cell death. On the basis of our earlier studies (57), the RV wall stress in cats is expected to increase in the first few days of PO. Therefore, our expectation is that the increased wall stress triggers calpain activation and programmed cell death to a degree shown to be sufficient to cause cardiomyopathy (65). Since induction of severe PO for a longer duration (≥3 wk) often results in a second wave of wall stress and an increase of clinical signs of heart failure (57), we expect calpain activation and programmed cell death to remerge in failing heart. We therefore expect that the administration of calpain inhibitor at an appropriate dose and duration in patients with hypertrophic cardiomyopathy will have beneficial effects.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Program Project Grant HL-48788 and by a Merit Review Award from the Research Service of the Department of Veterans Affairs.