Functional consequences of caspase activation in cardiac myocytes

In Vitro Caspase-Substrate Assay.

Ten micrograms of purified cardiac myofibrillar proteins from the thick filament myosin heavy chain, myosin light chain 1/2 (from rabbit and bovine muscle, respectively; Sigma), from the thin filament cardiac α-actin in monomer form (bovine cardiac muscle; Cytoskeleton, Denver), tropomyosin (bovine muscle; Sigma), troponins T, I, and C (human cardiac muscle; Advanced Immunological, Long Beach, CA), and from the cytoskeletal structure, α-actinin (rabbit skeletal muscle; Cytoskeleton) were resuspended in caspase assay buffer (50 mM Hepes/2 mM EDTA/0.15 CHAPS/10% sucrose/5 mM DTT and protease inhibitors, pH 7.4). The following proteases were used to prepare the myofilaments: 1 mg/ml aprotinin, 1 mg/ml leupeptin, and 0.1 mM PMSF.

These purified proteins then were incubated in the presence or absence of recombinant human caspase-3 [protein/caspase-3 = 500:1 (wt/wt) for 4 h at room temperature]. In a different set of experiments, troponins, reconstituted in a complexed form (combining recombinant mouse cardiac TnT, TnI, and human TnC in a ratio of 1:1:1.5) (13), were exposed to the buffer assay described above in the presence or absence of caspase-3. Cardiac myofilaments, isolated from male Sprague–Dawley rat left ventricles, also were exposed to the buffer assay described above in the presence or absence of caspase-3. In the experimental set-up, inhibitors of caspase-3, DEVD-CHO (10 μM) or z-VAD-fmk [z-VAD (Z-Val-Ala-Asp(Ome)-FMK), 10 μM] also were added. Poly(ADP-ribose) polymerase was used as an internal control of the activity of caspase-3. Four hours after the addition of caspase-3, the cleavage reaction was stopped by the addition of Laemmli buffer followed by sonication and heating at 95°C for 7 min. The myofilaments were spun down, and then the cleavage reaction was stopped as mentioned above.

SDS/PAGE and Western Blot Analysis.

Samples were separated on a 12% SDS-polyacrylamide gel. Gels were subjected to Coomassie blue or silver staining or transferred to nitrocellulose. Western blot analysis was achieved by using the following antibodies: anti-cTnT that recognized the C-terminal region (clone JLT-12, dilution 1:200; Sigma); two antiactin directed against C-terminal (dilution 1:200; Sigma) and N-terminal of actin, respectively (clone C4, dilution 1:1,000; Chemicon); anti-TnI (clones 10F4 and 6F9; Advanced Immunological); anti-TnC (clones 1A2, 7B9, 12G3; Advanced Immunological); and anti-α-actinin sarcomeric (clone EA-53, dilution 1:800; Sigma). The percentage of myofibrillar protein cleavage was determined by densitometry, using nih image.

Cell Isolation, Culture, and Treatment.

Adult rat ventricular myocytes (ARVM) were isolated by enzymatic dissociation from hearts of adult male Sprague–Dawley rats (240–280 g) as described. The cells were resuspended in DMEM, supplemented with albumin (2 mg/ml), l-carnitine (2 mM), creatine (5 mM), taurine (5 mM), and 0.1% penicillin–streptomycin, and plated on laminin (1 μg/cm²)-coated dishes. ARVM were treated for 6 h with (−) norepinephrine (NE; 10 μM) in the presence of an α₁-adrenergic receptor blocker, prazosin (0.1 μM), and ascorbate acid (0.1 mM) (14). In specific experiments, ARVM were pretreated for 4 h with z-VAD-fmk (100 μM). Myofilaments then were isolated as mentioned above, and protein analysis was performed.

Skinned Fibers, Myofibrillar ATPase, and Forces Measurements.

Force/Ca²⁺ relationships and myofibrillar ATPase activity were measured simultaneously in skinned fibers by using a system described before (15, 16). In these experiments, fibers obtained from trabeculae of mice hearts were skinned overnight in a relaxing buffer (150 mM KCl/20 mM Mops/10 mM EGTA/1 mM free Mg₂/5 mM ATP/12 mM creatine phosphate/10 units/ml creatine kinase/0.5 mM DTT and protease inhibitors) and 1% Triton X-100. The fibers were attached to a displacement generator at one end and a force transducer at the other end. The resting sarcomere length was adjusted to 2.3 μm on the basis of a laser diffraction pattern. The fiber width, height, and length were measured, and the cross-sectional area was estimated based on an elliptical model. Tension was expressed as force/cross-sectional area. To determine the relationship of free Ca²⁺ concentration to tension, the fibers were bathed sequentially in solutions (with pCa ranging from pCa 8 to 4.5, at 20°C), the corresponding force was measured, and the data were stored with an analog-to-digital converter to a computer. Fiber ATPase activity was measured as follows: ATP regeneration from ADP was coupled to the breakdown of phosphoenolpyruvate to pyruvate, and ATP was catalyzed by pyruvate kinase, which was linked to the synthesis of lactate catalyzed by lactate dehydrogenase. The breakdown of NADH, which is proportional to the amount of ATP consumed, was measured by using a UV light that was projected through the quartz window of the bath and recorded at 340 nm. The ratio of light intensity at 340 nm to a signal reference at 410 nm was obtained by the mean of an analog divider. After each recording, the UV absorbance of NADH was calibrated by injections of ADP in the bathing solution. Data were analyzed with labview (National Instruments, Austin, TX). Data were linearized by using the Hill transformation, and the force/pCa was fitted to the Hill equation with nonlinear regression analysis to derive the pCa50 and Hill coefficient (sigmaplot, Chicago). The protocol consisted of two series of measurements: a first measurement at t = 0 min under control conditions, followed by a second measurement after exposure of the fibers to caspase-3 [ratio of protein/caspase = 1:50 (wt/wt)] after 45 min. In sham experiments, the fibers were bathed in the caspase buffer in the absence of caspase-3. The protocol was repeated in the presence of the inhibitor of caspase-3, DEVD-CHO (10 μM). The data were calculated as percentages compared with the maximum control values obtained at t = 0 min. Cardiac myofibrillar proteins isolated from these same fibers also were tested for breakdown of troponins.

Statistical Analysis.

Data are presented as mean ± SE and were analyzed with a one-way ANOVA, with statistical differences identified at P < 0.05.

Caspase-3 Activation and Myofilament Cleavage.

To determine whether caspase-3 cleaves the myofilaments, we treated myofibrillar proteins individually and/or in complexed form with highly purified human recombinant caspase-3 for 4 h and analyzed the cleavage products by SDS/PAGE and Western blot analysis. Fig. 1 demonstrates that caspase-3 induced a robust cleavage of cTnT into fragments of ≈25 kDa only when cTnT was in complexed form with other troponins (% of cleavage = 37 ± 4%, P < 0.01, as compared with control; Fig. 1b) or in the myofilament (% of cleavage = 57 ± 5%, P < 0.01, as compared with control; Fig. 1c) but not individually (Fig. 1a). Of note, myosin, myosin light chains, tropomyosin, TnI, and TnC were not cleaved by caspase-3.

We then examined whether caspase-3 targets actin. Using silver staining and Western blot analyses with two different types of antiactin antibodies, directed against the either the N- or C-terminal region, we found that caspase-3 cleaved cardiac actin in its monomeric form in two major fragments (≈30 and ≈15 kDa) and two minor fragments (both ≈20 kDa) (Fig. 2 a and b). Caspase-3 also cleaved cardiac actin when polymerized in the lattice of the myofilaments (Fig. 2c). The generation of the ≈20-kDa fragments was more evident in the myofilament.

In addition, Coomassie blue staining demonstrated that caspase-3 cleaved the cytoskeletal element α-actinin individually in fragments ≈45 kDa. This was blocked by the addition of caspase inhibitor z-VAD (Fig. 3a). Furthermore, caspase-3 also cleaved α-actinin in the lattice of the myofilaments. A fragment of ≈25 kDa was detected by using Western blot analysis (Fig. 3b). The addition of capase-3 inhibitors also blocked the cleavage of cTnT, cardiac α-actin (data not shown).

Effect of β-Adrenergic Stimulation on Myofilament Breakdown.

We have shown that β-adrenergic receptor stimulation induces apoptosis in adult rat ventricular myocytes, and this was associated with the early activation of the mitochondrial pathway and activation of caspase-3. Therefore, ARVM were exposed to β-adrenergic stimulation for 6 h and myofilaments then were isolated during the early phase of apoptosis when no major DNA breakdown (terminal deoxynucleotidyltransferase-mediated UTP end labeling-positive cells) is found. β-Adrenergic receptor stimulation induced similar cleavage of cTnT (≈25 kDa) in ARVM (Fig. 4). The antibodies against the N-terminal of actin (residues 23–24) and C-terminal (last 11 residues) recognized, in addition to the intact actin, a similar, major fragment of ≈20 kDa. Cleavages of cTnT and actin were reduced after pretreatment with VAD-fmk (Fig. 4), demonstrating that β-adrenergic receptor-stimulated myofilament cleavage occurred mainly via caspase activation.

Cleavage Site on TnT.

To determine the major site of caspase-3 cleavage of cTnT, caspase-3-generated cTnT fragments from the troponin complex were isolated from a polyvinylidene membrane, and partial amino acid sequences were obtained by Edman degradation and HPLC. The fragment that migrated near 25 kDa displayed an N sequence of DIHRKXVEKD, indicating that it did, in fact, result from cTnT cleavage between Asp-96 and Asp-97 as shown in Fig. 5. This finding points to the cleavage site for caspase-3 as being DFDD. The consensus caspase-3 cleavage site is usually DEVD. The two aspartic acid residues are absolutely required, but the glutamic acid and valine residues may vary. Comparison between cardiac TnT from mice, rat, and human revealed that this site is highly conserved.

Because we had no access to antibodies directed against the N-term of cTnT, we treated a troponin complex (cMyc attached to cTnT) with caspase-3, and, after treatment, we isolated the troponin complex and probed with antibodies directed against cMyc. We could not find the cMyc fragment (N-terminal fragment), suggesting that only the C terminus remains in the complex or in the lattice of the myofilament. This N-terminal region has been shown to be important for the cardiac function. Indeed, the truncation of the N-terminal region of cTnT impairs the force/Ca²⁺ relationship and decreases myofibrillar ATPase activity in skinned fiber.

Caspase Activation and Contractile Activation.

To examine whether the activation of caspase-3 and the concomitant breakdown of cardiac TnT, α-actin, and α-actinin result in functional effects, we measured the effect of caspase-3 treatment on force/Ca²⁺ relationship and simultaneous myofibrillar ATPase activity in skinned fibers isolated from the mice hearts. Fig. 6 illustrates the force/Ca²⁺ relationship and simultaneous myofibrillar ATPase before (t = 0) and after 45 min of caspase-3 treatment. After 45 min of caspase-3 treatment, there was a significant decrease in maximal Ca²⁺-activated force and in Ca²⁺ sensitivity (Fig. 6c) as compared with sham fibers that were treated with buffer solution (Fig. 6a). There was a parallel decrease in maximal Ca²⁺-activated myofibrillar ATPase activity (Fig. 6d) as compared with sham fiber (Fig. 6b). The addition of caspase-3 inhibitor DEVD-CHO prevented the functional changes mediated by caspase-3 (Fig. 6 e and f) but DEVD-CHO had no independent effects on myofilament activation (data not shown).

To ascertain that these fibers, which were used for the mechanical studies, indeed had breakdown of cardiac TnT, actin, and α-actinin, we analyzed these fibers by Western blotting. As in the myofilaments, we found one detectable cleavage product for cTnT (at 25 kDa), two detectable products for actin (20 and 15 kDa), and one cleavage product for α-actinin (at 25 kDa) (data not shown).