INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

MECHANISMS RESPONSIBLE for the death of heart muscle cells in chronic cardiovascular diseases have been of increasing interest in recent years. In some diseases, there is necrosis of cardiac muscle cells that is characterized by cell swelling, disruption of cell membranes, and cell lysis with the induction of an inflammatory response in the surrounding tissue. In other important cardiovascular diseases, it has become apparent that apoptosis of cardiomyocytes is also responsible for cardiac muscle cell loss. Apoptosis is a morphologically and biochemically distinct form of programmed cell death that was first identified in 1972 (31). Apoptosis differs from necrosis in that there is condensation of nuclear chromatin and cell shrinkage with preservation of intracellular organelles (52). Subsequently, there is blebbing of nuclear and cytoplasmic membranes and fragmentation of the dying cells into membrane-bound "apoptotic" bodies that undergo phagocytosis. The sequence of events during apoptosis is rapid (hours), requires energy, and eventuates in activation of endonucleases that degrade chromosomal DNA into oligosomal fragments that are multiples of 180- to 200-base pair length (which form DNA "ladders" after electrophoresis). The apoptotic program, which can proceed even in anucleate cytoplasts (26), involves multiple gene products that either promote or hinder the death pathway. In addition to its known role in embryogenesis, apoptosis has been demonstrated in a number of pathophysiological processes. Apoptosis of cardiac muscle cells has been reported in acute myocardial infarction (10, 29), cardiac allograft rejection (60), right ventricular dysplasia (27, 28), dilated cardiomyopathy and heart failure (44, 53), pacing-induced congestive heart failure (38), ischemia-reperfusion injury (22), and papillary muscles subjected to passive stretch (11).

The five-electron oxidation of L-arginine to citrulline and biologically active nitric oxide (NO) is important to a large variety of physiological and pathological processes. NO synthesis is accomplished by the three isoforms of NO synthase (NOS), neuronal NOS (nNOS, NOS I), originally identified in brain, inducible NOS (iNOS, NOS II), originally identified in macrophages, and endothelial NOS (eNOS, NOS III), originally identified in endothelial cells and subsequently in cardiac myocytes (reviewed in Ref. 43). Constitutive NOS (nNOS and eNOS) require calcium and calmodulin as cofactors and generate small amounts of NO. Small amounts of NO released by endothelial cells in response to hormones or sheer stress interact with soluble guanylyl cyclases to increase the formation of cGMP in target cells such as platelets and endothelial and vascular smooth muscle cells, promoting inhibition of platelet adhesion and aggregation, inhibition of leukocyte adhesion and migration, and vasodilation, respectively. NO produced by nNOS acts as a neurotransmitter in the brain, cells of the nonadrenergic nervous system, and skeletal muscle. The iNOS expressed in macrophages, endothelial cells, vascular smooth muscle cells, and cardiac myocytes in response to cytokines or bacterial endotoxin is activated in a calcium-independent manner and generates large (up to micromolar) amounts for protracted periods. NO synthesized by iNOS in activated macrophages is cytotoxic and participates in their antiviral and antimicrobial actions. NO synthesized by iNOS in vascular smooth muscle cells and cardiac myocytes in response to endotoxin has been implicated in the pathogenesis of hypotension in association with Gram-negative bacterial infections. NO produced by iNOS in cardiomyocytes has also been demonstrated to influence myocardial contractile responses to -adrenergic agonists and heart rate. In 1995, our group reported (47) that NO produced by iNOS in activated J744 macrophages was capable of killing adjacent isolated adult rat cardiac myocytes and that NO produced by iNOS in cytokine-treated, isolated adult rat cardiomyocytes was capable of causing death of the cardiomyocytes. Furthermore, the death of heart muscle cells could be blocked by inhibitors of NOS or by the administration of transforming growth factor (TGF)-, which potently suppressed iNOS expression. However, in those experiments, the mechanism of cell death (necrosis vs. apoptosis) was not determined.

Albina and co-workers (2, 13) first reported that NO produced by iNOS in cytokine-treated macrophages was capable of triggering apoptosis of both macrophages and certain tumor cells cocultured with the macrophages. Subsequently, this work has been confirmed and NO has been found to also promote apoptosis in chrondrocytes (7), pancreatic islet cells (3, 30), macrophage-like cells (42, 50, 54), certain tumors (13), and vascular smooth muscle cells (46). In contrast, NO has also been reported to inhibit apoptosis in other cells including eosinophils (4), murine and human B lymphocytes (19, 40), vascular endothelial cells (16), and cardiac myocytes that have been subjected to passive stretch (11). The mechanisms responsible for these marked differences in the response to NO have not been clearly defined.

In several of the cardiac diseases in which apoptosis of cardiac myocytes has been reported, other investigators have documented iNOS expression using PCR, Western blotting, or immunostaining with specific anti-iNOS antibodies and/or measurements of iNOS enzyme activity. These diseases include experimental myocardial infarction (59), dilated cardiomyopathy (14), end-stage heart failure (23), and rat and human cardiac allograft rejection (8, 37, 60, 61, 68). In the study of rat cardiac allograft rejection, increases in iNOS mRNA, protein, and enzyme activity in the allografts paralleled in time and extent apoptosis both of macrophages infiltrating the graft and of cardiac myocytes. These data are compatible with but do not prove the hypothesis that NO can trigger apoptosis of cardiac myocytes in these settings. Accordingly, the present studies were designed to investigate whether NO released from an NO-donor drug is capable of triggering apoptosis of adult rat cardiac myocytes in vitro.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Isolation and purification of adult rat ventricular myocytes. Cardiac myocytes were isolated from the ventricles of adult male Wistar-Furth rats (250-300 g) using a method described previously (47). Rats were anesthetized and heparinized (1,500 U/kg), after which their hearts were rapidly excised and placed in cold, nominally calcium-free buffer. In aseptic fashion, perfusate was introduced into the cross-clamped aortic root at a constant rate (3-4 ml/min) using a Langendorff-type apparatus. Thirty milliliters of an oxygenated, calcium-free perfusate were infused for the initial 5 min [in mM: 111 NaCl, 5.4 KCl, 1.5 NaH₂PO₄, 1.6 MgCl₂, 4.2 NaHCO₃, 20 HEPES, 5.4 glucose, 4.1 L-glutamine, and 10 taurine (Sigma, St. Louis, MO), dissolved in MEM with MEM amino acid and vitamin supplement (GIBCO, Grand Island, NY)]. After this initial perfusion was completed, collagenase B (final concn 0.05%; Boehringer-Mannheim, Indianapolis, IN) was added, and recirculating perfusion continued at 35°C for 15 min. The ventricles were minced and then placed into 10 ml of a digestion solution consisting of the base perfusion solution described above, to which 10 mg of collagenase B, CaCl₂ (final concn 0.6 mM), and bovine serum albumin (5 g/l) were added. The tissue was placed in a 33°C shaker, and every 15 min, the myocyte-rich solution was decanted and new digestion solution was added. This process was repeated four or five times. The decanted fractions were filtered through an autoclaved nylon mesh, pooled, and then subjected to Percoll density-gradient centrifugation (Sigma). Myocytes sedimenting at the 1.082/1.062 g/ml interface were collected, resuspended in DMEM with 10% fetal bovine serum (Gemini Bioproducts, Calabasas, CA), 1% penicillin-streptomycin, and 1% L-glutamine, and plated on plastic ware precoated with laminin (Sigma). For studies of cell death, myocytes were plated into 96-well ELISA plates (Corning). Myocytes designated for in situ nick labeling were plated onto chamber slides, and those designated for DNA extraction to assess ladders were plated onto 100-mm dishes. The medium was changed 2 h after plating to remove cellular debris and contaminating cells. In addition to light microscopic characterization of the characteristic rod-shaped or rhomboidal appearance and striations of cardiac muscle cells, the purity of representative cell preparations was confirmed by plating cells onto chamber slides, followed by immunostaining with either anti-tropomyosin or anti-sarcomeric myosin. These procedures indicated that >99% of all adherent cells were myocytes.

Quantification of cell death. Cardiac myocyte death was determined using a fluorometric method with calcein AM, a membrane-permeant dye that is not fluorescent until the acetoxymethyl group is cleaved by cellular esterases in living cells (1, 39, 66). Cells were plated in 96-well plates at a density of ~50,000 cells/well and were left untreated (these cells served as living cells for control) or treated with ethanol (40%; these cells served as dead cells for control) or the indicated concentrations of S-nitroso-N-acetylpenicillamine (SNAP, Sigma, St. Louis, MO). After 4 h of treatment (ethanol was added to wells 20 min before termination of experiments), calcein AM (2 µM final concn; Molecular Probes, Eugene, OR) was added to all wells. Excitation and emission spectra were 485 and 530 nm, respectively. Data are expressed as the means of 14 observations normalized to the mean fluorescent intensity of the living cells. To visualize myocytes directly, parallel experiments were performed with cells plated on chamber slides with the addition of both calcein AM and ethidium homodimer (4 µM, Molecular Probes; this was for nuclear identification); fluorescent images were obtained using an Olympus System Microscope (model FLA-BX) with simultaneous excitation filters set at 485 and 530 nm to simultaneously visualize ethidium-stained nuclear material and calcein-stained cytoplasm.

Preparation of reduced hemoglobin. Reduced hemoglobin was prepared by dissolving bovine hemoglobin (Sigma) in double-distilled water to a concentration of 5 mM, after which a 10-fold molar excess of sodium dithionite (Fisher Scientific) was added for 10 min at room temperature with gentle vortexing. The slurry was placed in a dialysis membrane (molecular mass cutoff 13 kDa) for dialysis at 4°C. Samples were aliquoted and frozen at 80°C until time of use.

Identification and quantification of apoptotic cells. In situ nick-end labeling was performed to identify broken strands of DNA (18). This procedure takes advantage of the ability of the Klenow fragment of DNA polymerase I to use the complementary DNA strand as a template to add nucleotides to the 3' end of a nicked strand of DNA (21). In this procedure, chamber slides were treated with SNAP (1 mM, Sigma), N-acetylpenicillamine (NAP, 1 mM), or SNAP (1 mM) + hemoglobin (10 mM) for the indicated durations. After this treatment, residual medium was gently aspirated and the slide was allowed to air dry for 1 h. After 30-min to 1-h fixation at 40°C in 2.5% DMSO in 10% phosphate-buffered Formalin, slides were placed overnight at 40°C in phosphate-buffered saline (pH 7.40).

Confirmation of apoptosis by DNA ladders. DNA fragmentation into characteristic 180- to 200-bp oligonucleosomes as a function of duration of exposure to SNAP was detected by 3' end-labeling DNA fragments with terminal deoxynucleotidyl transferase followed by separation of DNA fragments by agarose gel electrophoresis and detection by autoradiography. This method has been previously described in detail (64). Cellular DNA was prepared by scraping cells into a buffer consisting of 0.1 M NaCl, 0.01 M EDTA (pH 8.0), 0.05 M Tris · HCl (pH 8.0), and 0.2 M sucrose, after which they were placed in microcentrifuge tubes containing 1:10 (vol/vol) proteinase K (20 mg/ml, Boehringer-Mannheim) for 60 min at 50°C to facilitate membrane and protein disruption. After addition of 35 µl of 8 M potassium acetate, samples were placed on ice for 60 min and centrifuged at 5,000 g for 10 min to pellet cellular debris. DNA was extracted from supernatants using two extractions, first with an equal volume of phenol-chloroform-isoamyl alcohol (25:24:1) and then with chloroform-isoamyl alcohol (24:1). DNA was then precipitated from the upper aqueous phase using 0.1 vol of 3 M sodium acetate with 2.5 vol of ice-cold ethanol and left at 20°C for 60 min before centrifugation. Pellets were resuspended in 50 µl of TE buffer, followed by a 60-min incubation with DNase-free RNase (Boehringer-Mannheim) at 37°C. Samples were reextracted, and DNA was reprecipitated as described above. After a 10% ethanol wash and vacuum drying, DNA pellets were resuspended in 25 µl of sterile water, and DNA concentrations were quantified by measuring absorbance at 260 nm.

Cytokine and N^G-monomethyl-L-arginine exposure. Twenty-four hours after primary culture of the myocytes (at the time of medium change), cytokines were added to the medium to induce iNOS as we showed previously (47); these included recombinant human tumor necrosis factor (TNF)- (25 ng/ml; Genzyme, Cambridge, MA), interleukin (IL)-1 (5 ng/ml; Genzyme), and recombinant murine interferon- (100 U/ml; Genzyme). In certain instances where indicated, at the time of cytokine addition, N^G-monomethyl-L-arginine (L-NMMA, 2 mM; Calbiochem, La Jolla, CA) was added to the cell culture medium.

Detection of p53. Cardiac myocytes were lysed in a lysis buffer containing 150 mM NaCl, 1.0% NP-40, 0.1% SDS, 1 mM EDTA, and 50 mM Tris (pH 7.7), supplemented with protease inhibitors (10 µg/ml of antipain and leupeptin and 1 mM phenylmethylsulfonyl fluoride), and centrifuged at 10,000 rpm for 20 min at 4°C. The protein extracts (20 µg/lane) were electrophoresed on a 15% SDS-polyacrylamide gel, transferred to a nitrocellulose filter, and then immunoblotted with a 1:1,000 dilution of a murine anti-p53 IgG (clone PAb 122, Pharmingen, San Diego, CA). Horseradish peroxidase-conjugated sheep anti-murine IgG was used as a secondary antibody. Bands were detected with the enhanced chemiluminescence method (Amersham). For immunohistochemistry, myocytes were prepared on chamber slides and Formalin fixed as described in Identification and quantification of apoptotic cells. Slides were initially blocked using 4% horse serum with 0.1% Triton X-100 in PBS, after which the anti-p53 IgG (Pab 122) was applied in PBS at a 1:250 dilution overnight at room temperature. An ExtrAvidin-2A alkaline phosphatase staining kit (Sigma), which incorporates a biotinylated goat anti-mouse secondary antibody, was used with procedures described by the manufacturer. Slides were developed using Fast Red TR/naphthol AS-MX alkaline phosphatase substrate tablets (Sigma) according to the manufacturer's instructions.

Statistics. Data were evaluated using ANOVA for unpaired variables to compare three or more groups (cardiac myocyte death). For quantitative apoptosis experiments in which apoptotic nuclei were counted, differences between the presence or absence of apoptosis were discriminated using the ² statistic. Data are expressed as means ± SE, with P < 0.05 considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Figure 1 shows the effects of NO on cardiac myocyte cell death. Purified adult rat cardiac myocytes were exposed to increasing doses of the NO donor SNAP. Cell viability was determined at 4 h by the ability of viable cells to cleave the AM group from calcein AM to generate the fluorochrome calcein (1, 39, 66). These data (Fig. 1) demonstrate that SNAP kills myocytes in a dose-dependent fashion. In these experiments, ethanol was added to a separate group of myocytes to serve as a positive control for cell death (Fig. 1). In parallel experiments, fluorescence was visually confirmed using myocytes plated on laminin-coated chamber slides, which were either left untreated or treated with SNAP (1 mM) for 4 h. The myocytes were treated with both calcein AM (to visualize calcein in viable cells) and ethidium homodimer (for nuclear localization) and viewed using a simultaneous double-fluorescence technique with excitation wavelengths of 485 and 530 nm. With this technique, viable myocytes demonstrated intense green cytoplasmic fluorescence, which was not observed in dead cells (Fig. 2).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Dose-response effect of a nitric oxide (NO) donor on cardiac myocyte death. Adult rat ventricular myocytes were purified as described in MATERIALS AND METHODS, plated in 96-well plates at a density of ~50,000 cells/well, and either left untreated (0 mM SNAP) or incubated with indicated concentrations of NO donor S-nitroso-N-acetylpenicillamine (SNAP) or 40% ethanol (EtOH) to serve as a positive control for cell death. Conversion of calcein AM (2 µM) to the fluorochrome calcein, which occurs in living cells, was measured using excitation and emission spectra of 485 and 530 nm, respectively. Data are normalized to mean fluorescent intensity of living cells (controls). Means ± SE are shown.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Direct visualization of live and dead myocytes using calcein AM and ethidium homodimer for nuclear colocalization. Cells were prepared as described in Fig. 1, except that they were plated on chamber slides. Fluorescent images were obtained using an Olympus System Microscope (model FLA-BX) with simultaneous excitation filters set at 485 and 530 nm. A: control (living) myocyte demonstrates green cytoplasmic fluorescence. B: dead (SNAP treated) myocyte lacking calcein fluorescence but with nuclear (ethidium) staining. C: untreated myocytes; both a living and a dead myocyte can be observed in same field.

Because NO gas (2) or chemical NO donors (41) have been reported to induce apoptosis in cells of the macrophage lineage, we investigated whether SNAP could cause cardiac myocyte death by apoptosis. To detect apoptosis of cardiac myocytes, two complementary procedures were used. In the first procedure, apoptosis of individual cardiac myocytes was identified in situ (on chamber slides) using a nick-end labeling procedure (18). In this technique, digoxigenin-labeled dUTP is incorporated at the 3' ends of DNA fragments using the Klenow fragment of DNA polymerase I (21). Sites of dUTP accumulation are then identified using an alkaline phosphatase-conjugated secondary antibody. Nuclei containing fragmented DNA stained darkly and were easily identified (Fig. 3). In contrast to nonapoptotic myocytes, whose nuclei do not stain by this procedure, apoptotic cells exhibited a characteristic shrunken appearance accompanied by condensation of nuclear material and membrane blebbing (63). To determine whether NO may induce apoptosis in adult rat cardiac myocytes, the NO donor SNAP was added to cardiac myocytes for 4 h and the incorporation of fluorescent dUTP end label was examined. Apoptotic nuclei were clearly identified at each of the two different magnifications (Fig. 4, C and D). In contrast, when similar procedures were used, but with the addition of the control compound (NAP), only rare nuclei were identified as being apoptotic (Fig. 4, A and B). To confirm that the observed increase in SNAP-induced apoptosis was mediated by release of NO from the SNAP, reduced hemoglobin was added to SNAP-treated myocytes in parallel experiments to scavenge free NO. Under these conditions, the presence of hemoglobin reduced the numbers of apoptotic cells in SNAP-treated preparations (Fig. 4, E and F). The in situ technique was also used to quantify the numbers of apoptotic nuclei observed after either 4 or 8 h of SNAP treatment, after which cells containing apoptotic nuclei on the chamber slides were counted and expressed as a percentage of the total cells (Fig. 4G). These data demonstrate that SNAP treatment induces apoptosis of cardiac myocytes and that this can be blocked by simultaneous addition of a compound that scavenges free NO.

View larger version (52K): [in this window] [in a new window]   Fig. 3.   Gross morphological appearance of control and apoptotic myocytes stained using in situ nick-end labeling. Cardiac myocytes were purified as described in MATERIALS AND METHODS. Digoxigenin 11-dUTP was incorporated into DNA fragments using Klenow fragment of DNA polymerase I, developed with an alkaline phosphatase-conjugated anti-digoxigenin antibody, and visualized with NBT-BCIP. Representative nonapoptotic cardiac myocytes (left) appear rectangular or rhomboid, with a striated cytoplasmic appearance and faint or absent nuclear staining. Apoptotic cardiac myocytes (right) appear rounded, with darkly staining nuclei and plasmalemmal blebbing.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Effect of an NO donor or an NO donor with an NO scavenger on cardiac myocyte apoptosis. Apoptotic nuclei were detected using fluorescein-dUTP incorporated into DNA at sites of strand breakage using terminal deoxynucleotidyl transferase. Myocytes were plated on chamber slides and treated with N-acetylpenicillamine (NAP, 1 mM; A and B) as a control, NO donor SNAP (1 mM; C and D), or SNAP + reduced Hb (1 and 10 mM, respectively; E and F) for 4 h. Magnification: ×100 (A, C, E) and ×200 (B, D, F). G: quantitative apoptosis, determined by counting apoptotic myocytes and expressing them as percentage of total cells counted. Conditions and time points are as indicated. Means ± SE are shown. ** P < 0.01 vs. controls, !! P < 0.01 vs. SNAP.

The second procedure to identify apoptosis of cardiac myocytes was performed to ascribe the observed nuclear staining by the in situ technique to internucleosomal DNA cleavage rather than nonspecific DNA strand breakage, as has been reported in other cell types by authentic NO (45) or SNAP (51). With agarose gel electrophoresis (64), DNA extracted from control or SNAP-treated myocytes demonstrated internucleosomal DNA cleavage into integer multiples of 180-200 bp, forming DNA ladders that are the hallmark of apoptosis (Fig. 5). Although some apoptosis was observed in the 4-h controls, the intensity of the DNA ladders was markedly increased when purified cardiac myocytes were exposed to SNAP for 4 h (Fig. 5A). This increase in DNA ladders was even more apparent by 16 h, although basal levels of apoptosis were also increased at the 16-h time point. To confirm that oligosomal DNA fragmentation into ladders was the result of NO liberated by SNAP, hemoglobin was added simultaneously with the addition of SNAP in parallel experiments. Because hemoglobin reduces the appearance of ladders after treatment with SNAP, this experiment supports the histological finding that hemoglobin reduces the ability of SNAP to trigger apoptosis of cardiac myocytes (Fig. 5B).

View larger version (87K): [in this window] [in a new window]   Fig. 5.   Identification of oligonucleosomal DNA fragmentation in cardiac myocytes in response to an NO donor in presence or absence of an NO scavenger. Ventricular myocytes were prepared as described in MATERIALS AND METHODS and either left untreated (control), treated with SNAP (1 mM), or treated with a combination of SNAP (1 mM) + reduced Hb (10 mM) for indicated durations. Cellular DNA was prepared, and equal amounts (2 µg) were incubated with terminal deoxynucleotidyl transferase (A) or Klenow fragment of DNA polymerase I (B) in presence of [32P]dATP or [32P]dCTP. After removal of unincorporated radionucleotides, DNA was subjected to low-voltage agarose gel electrophoresis and autoradiography. A: DNA ladder formation in SNAP (1 mM)-treated or untreated (control) myocytes at indicated duration. B: effect of Hb (10 mM) added to SNAP (1 mM) on appearance of DNA ladders in cardiac myocytes (8-h exposure).

In previous work (47), we have shown that cytokines induce iNOS in cardiac myocytes and that this promotes NO-dependent killing of cardiac myocytes. However, in those particular studies, apoptosis was not examined. To extend this initial observation, and to show that NO-dependent killing of cytokine-stimulated cardiac myocytes is not restricted to application of an NO donor, experiments were performed to detect NO-dependent oligonucleosomal DNA fragmentation in cytokine-stimulated cardiac myocytes. For these experiments, which took place over 24 h, there is a clearly detectable increase in oligonucleosomal DNA fragmentation after treatment with IL-1, TNF-, and interferon- (Fig. 6). Application of SNAP to cardiac myocytes not treated with cytokines, shown here for comparison, causes a marked increase in apoptosis over baseline untreated conditions. When the NOS inhibitor L-NMMA is given to cytokine-stimulated cardiac myocytes, there is a substantial reduction in oligonucleosomal DNA fragmentation.

View larger version (47K): [in this window] [in a new window]   Fig. 6.   Effect of cytokines on NO-dependent induction of apoptosis. Cardiac myocytes were treated for 24 h with interleukin-1, tumor necrosis factor-, and interferon- (cytokines) with or without NO synthase inhibitor NG-monomethyl-L-arginine (L-NMMA; 2 mM). Oligonucleosomal DNA fragmentation was detected as described in Fig. 5. A lane is shown for comparison in which myocytes were not treated with cytokines but to which SNAP (1 mM) was applied.

Because the tumor suppressor gene p53 has a recognized role to promote apoptosis in response to DNA damage (49), and because NO is able to damage DNA chemically (45, 51, 58, 67), we hypothesized that p53 expression might be increased in cardiac myocytes after exposure to an NO donor. Accordingly, cardiac myocytes were divided into three groups after purification: one group was treated with NAP (1 mM for 4 h; Fig. 7, A and B), one group was exposed to SNAP (1 mM for 4 h; Fig. 7, C and D), and the final group was exposed to the identical dose and duration of SNAP in the presence of hemoglobin (10 mM; Fig. 7, E and F). Both Western blots (Fig. 7G) and immunohistochemistry performed on these myocytes demonstrated a distinct increase in immunoreactive p53 protein in response to SNAP exposure; the expression of p53 was reduced when the cell cultures were coincubated with both SNAP and the NO scavenger hemoglobin.

View larger version (103K): [in this window] [in a new window]   Fig. 7.   Immunohistochemical detection of p53 in cardiac myocytes in response to an NO donor. Myocytes were plated on chamber slides and treated with NAP (1 mM; A and B) as a control, NO donor SNAP (1 mM; C and D), or SNAP + reduced Hb (1 and 10 mM, respectively; E and F) for 4 h. p53 Expression was detected using a murine anti-p53 primary antibody and an alkaline phosphatase-conjugated secondary antibody. Sites of increased in p53 expression can be observed as dark red-brown staining. Magnification: ×100 (A, C, E), ×250 (B, D, F). G: Western blot showing an increase in immunoreactive p53 after exposure of cardiac myocytes to the NO donor SNAP (1 mM) for 4 h. Molecular mass markers on left.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present experiments indicate that exposure of purified adult rat ventricular myocytes to either an NO donor or cytokines that have been previously shown to cause iNOS induction (47) can induce the apoptosis of cardiac myocytes in an NO-dependent manner. Detection of apoptosis was performed by gel electrophoresis, searching for oligonucleosomal DNA fragmentation, and a nick-end labeling procedure to identify apoptosis in individual cardiomyocytes. Both NO-donor treatment and cytokine exposure increased apoptosis above the low level observed in control cells. The ability of cytokines that induce iNOS to cause apoptosis was blocked by an inhibitor of NO synthesis; the ability of a chemical NO donor (SNAP) to induce apoptosis was dose dependent, was not produced by the chemically similar NAP (which does not release NO), and was reduced when hemoglobin was added to the incubation media to scavenge NO released by the NO-donor drug. These data demonstrate directly that NO is capable of triggering apoptotic cell death in adult rat ventricular myocytes.

These results are consistent with previous reports that NO gas or NO released from an NO-donor drug is capable of inducing apoptosis in macrophages (2, 41, 42, 50, 54), chrondrocytes (7), certain tumor cell lines (13), vascular smooth muscle cells (46), and pancreatic islet cells (3, 30). They are also consistent with the report of Iwashina et al. (25), who, using a liposomal method of transfection, transfected vascular smooth muscle cells with iNOS. Overproduction of NO in the cells induced their death by apoptosis, as shown by characteristic morphological changes and by the development of internucleosomal DNA fragmentation. This proapoptotic effect of iNOS was associated with p53 accumulation and was inhibited by inhibiting nitric oxide synthesis with L-NMMA. In a recent paper (Ref. 24, published during the review of this work), neonatal rat cardiac myocytes were shown to undergo apoptosis in response to cytokine induction of NO synthesis or when exposed to an NO donor (S-nitrosoglutathione). The cytokine cocktail for these studies that was used to induce NO, consisting of IL-1, TNF-, and interferon-, was identical to that which we used in the current work and that which we previously reported (47). However, Ing et al. (24) showed that the effects of this cytokine cocktail on NO production or apoptosis were caused by IL-1 rather than by TNF- or interferon-. These data, taken together with our own data showing L-NMMA to inhibit the proapototic effects of these cytokines, suggest that these cytokines do not exert NO-independent actions to promote apoptosis under the experimental conditions that were used. Although p53 was not studied in the work of Ing et al. (24), in neonatal rat ventricular myocytes, the rate and extent of apoptosis appeared to be governed by alterations in the cellular balance between Bak and Bcl-x(L).

The propensity of NO to promote apoptosis in cardiac myocytes in vitro is diametrically opposite to the effect of NO in other cell types and under different conditions. NO has been reported to inhibit apoptosis in eosinophils (4), murine (19) and human (40) lymphocytes, certain tumor cell lines (13), human endothelial cells (16), and rat cardiac muscle cells subjected to mechanical stress (11). The mechanisms responsible for these different responses to NO have not been defined. However, one factor may be the dose of NO interacting with the target cells. In the settings in which apoptosis has been triggered by NO, the doses of NO or the donor drug were high or there were cells in the vicinity expressing iNOS, which has the capacity to synthesize large amounts of NO for prolonged periods of time. The doses used in the current experiments are likely to be physiologically relevant in the heart, because direct measurements in the beating heart have identified similar peak (albeit pulsatile) levels in the vicinity of cardiac myocytes (48).

The fact that NO is capable of triggering apoptosis in cardiomyocytes and in other cells is consistent with the effect of NO and other such triggers to induce DNA scission and damage. NO has been documented to induce DNA scission by processes involving deamination of nucleotides (45, 51, 58, 67), with cleavage occurring especially at cytosine-rich regions (67). Burney and co-workers (9) recently performed a mechanistic analysis of NO-induced toxicity in two cell lines. They confirmed that NO can induce a wide range of damage to cells, including inhibition of DNA synthesis, production of DNA strand breaks, and apoptosis. Because DNA damage is capable of causing cell cycle arrest, further investigation by this group revealed that NO also induced arrest of the cell cycle in both cell lines investigated. However, the precise mechanisms associated with NO-induced cellular damage are incompletely understood, because NO can interact with a large array of cellular proteins (55, 56) and can also interact with molecular oxygen and/or superoxide ions and be converted to a variety of active molecular species including N₂O₃ and peroxynitrite (ONOO) (12). Other influences on the cellular fates of NO are its rates of production and diffusion, the distance between source and target, the concentrations in the environment of potential reactants, the levels of such enzymes as superoxide dismutase and catalase, and the concentrations of cellular antioxidants such as glutathione.

The tumor suppressor gene p53 is a transcription factor that can effect cycle arrest or apoptosis (36, 49). Because levels of p53 are increased in response to DNA damage in several cell lines, in the current experiments, levels of p53 expression were evaluated in the myocytes incubated with the NO donor. Incubation of the isolated purified cardiomyocytes with SNAP, but not with NAP or SNAP plus reduced hemoglobin, was associated with increased p53 protein and increased immunostaining of p53 in the cardiac muscle cells. The association of p53 with apoptosis has been observed in macrophages (41), vascular smooth muscle cells (25), and cardiac myocytes from dogs with heart failure (35, 53). Although expression of p53 may be a component of the apoptotic gene program triggered by NO, and in some cells involved in the transcription of the proapoptotic gene Bax and suppression of the antiapoptotic gene Bcl-2, it may not be a required component of the apoptotic response. Kitsis and co-workers (6) demonstrated that apoptosis in response to myocardial infarction occurred as readily in mice lacking p53 as in wild-type littermate controls. Similarly, immunostaining for Fas antigen revealed that Fas was present in adult rat cardiomyocytes but did not appear to increase in response to the NO donor (data not shown). In cardiac myocytes in which apoptosis was triggered by hypoxia (62) or by mechanical stretch (11), expression of Fas appeared to be correlated with apoptosis.

The control incubations in the present study indicated that isolated adult rat cardiac myocytes exhibit a basal low level of apoptosis, even in the absence of an NO donor. Although not a focus of the present study, this raises the possibility that cardiac myocytes, like neurons in primary culture, require continual trophic support of growth or other factors (e.g., nerve growth factor for neurons) to prevent apoptosis.

Although the present experiments indicate that NO is capable of triggering apoptosis of cardiac myocytes, it must be kept in perspective that NO may kill heart muscle and other cells by a variety of mechanisms. These include auto-ADP-ribosylation of glycolytic enzymes (15), inhibition of ribonucleotide reductase (32), activation of poly(ADP-ribose) synthetase (69), inhibition of mitochondrial respiration and depletion of cellular energy stores (20, 57), formation of iron-nitrosyl complexes (17, 33, 34), inhibition of tricarboxylic acid cycle enzymes such as aconitase (65), and formation of toxic oxidants such as peroxynitrite (5, 12).

The contextual framework in which the current experiments can be viewed is that of myocardial disease processes, in which both iNOS induction and p53 expression have an impugned role. Apoptosis of cardiac myocytes and the expression of high-output iNOS have been observed in several cardiac diseases including myocardial infarction (44, 53, 59), reperfusion injury (22), dilated cardiomyopathy (14), heart failure (23, 44, 53), and cardiac allograft rejection (60, 61). Because p53 expression is increased in heart failure (35) and its expression increases in vascular smooth muscle cells made to overexpress iNOS and undergo apoptosis (25), these observations may be relevant to many cardiovascular diseases. The present study with an NO donor suggests that triggering of apoptosis by NO derived from iNOS may contribute to heart muscle cell loss and declines in ventricular function seen in several cardiovascular diseases. These considerations raise the possibility that myocyte loss caused by apoptosis might be reduced by drugs that inhibit NO synthesis by iNOS in myocytes or infiltrating macrophages, by gene therapies that alter the apoptotic gene program, or by drugs that inhibit proteases involved in the final stages of apoptotic cell destruction.