INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SODIUM NITROPRUSSIDE [Na₂(Fe(CN)₅NO)] (SNP) is used clinically as a vasodilator to reduce markedly elevated blood pressure and reduce left ventricular impedance in the acute management of hypertension. The potential adverse effects of SNP on cells and tissues have recently been identified and include SNP-induced cell death in murine neuroblastoma cells (47), the brain (37), vascular smooth muscle cells (32, 44), and certain tumor cells (42). However, this cytotoxicity is not demonstrable in all cells and tissues. SNP does not induce cell death in murine macrophages (8) or thymocytes (1). In contrast, a cell-protective function has been attributed to SNP in PC12 cells, sympathetic neuronal cells (13), and endothelial cells (10) and in ovarian follicles (7). If the potential cytotoxicity of SNP is dependent on cell type, then it becomes imperative to determine whether SNP induces cardiomyocyte cell death, because the loss of cardiomyocytes is part of the underlying pathogenesis leading to irreversible failure of cardiac function (23). We therefore sought to determine the potential toxicity of SNP in cardiomyocytes and the type of cell death that would be involved. Cell death can be divided mainly into two kinds: apoptosis, which involves the activation of a genetically programmed set of enzymes that leads to the cell participating in its own destruction, or necrosis, a more passive process of cell destruction that occurs after severe damage to the cell membrane (23, 43, 46).

The underlying mechanisms accounting for SNP-induced cell toxicity are not completely understood but include several mechanisms related to free radical-mediated cell injury. First, metabolites of SNP may undergo redox cycling with oxygen to form superoxide anion (O₂·) that in turn produces H₂O₂ and O₂ (36). Second, the iron moiety of SNP appears to mediate the prooxidative properties of SNP (37). Third, the metabolism of SNP releases cyanide, which complexes with the iron of mitochondrial cytochromes to inhibit ATP production that is critically important for cell metabolism (25). Fourth, metabolism of SNP releases nitric oxide (NO) that in turn can regulate myocardial oxygen metabolism (33). NO also has the potential for cytotoxicity through the formation of peroxynitrite, although the extent to which this is relevant in vivo is still open to debate (2, 28). We sought to determine the extent to which the putative action of SNP may be mediated through oxidative damage to the cardiomyocyte.

We further sought to explore the potential synergy between SNP and reactive oxygen species. Reactive oxygen species have been implicated in myocardial damage in clinically important conditions such as reperfusion injury and heart failure (for reviews, see Refs. 3, 14). Hydrogen peroxide (H₂O₂), one of the reactive oxygen species released during reperfusion of the ischemic myocardium (38), induces contractile dysfunction, oxidative damage to membranes, and cell death (4, 5, 9, 21, 38, 40). In addition to the acute cardiotoxicity of reactive oxygen species, the concept has recently developed that heart failure may develop because of limited antioxidant reserve of the myocardium (38). Reduction in antioxidant reserve may also be an important determinant of prognosis in patients with heart failure (22). Thus myocardial damage may occur acutely due to high bursts of free radical generation during reperfusion, after ischemia, as well as chronically because of low grade oxidative stress. High concentrations of O₂· produce H₂O₂ that, in combination with NO, released by SNP, produce peroxynitrite anion (ONOO) and hydroxyl radicals that are especially damaging to tissues (2). The potential adverse consequence of an interaction between SNP and low concentrations of H₂O₂ is a novel hypothesis. Although most studies focused on factors other than those that mitigate acute H₂O₂-induced cardiotoxicity, we explored the potential of SNP as a factor that might act to accentuate or augment the effect of low levels of H₂O₂ to induce cardiomyocyte injury and cell death.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell Cultures

Cell Viability

MTT assay. The MTT assay, an index of cell viability and cell growth, is based on the ability of viable cells to reduce MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] from a yellow water-soluble dye to a dark blue insoluble formazan product (26). Cardiomyocytes were seeded in multiwell microtiter plates (Falcon 3072, Becton-Dickinson, Lincoln Park, NJ) at 20,000 cells per well and incubated at 37°C for 72 h. Cardiomyocytes were then treated with SNP, H₂O₂, or other agents and returned to the incubator for 24 h. SNP and H₂O₂ were added directly to the media, and potential inhibitors were added 15 min before either agent. MTT dye was added to each well for the last 4 h of treatment. The reaction was stopped with the addition of solubilization reagent (Promega, Madison, WI), and the optical density was determined at 570 nm on a multiwell plate reader (Bio-Rad model 3550, Bio-Rad, Mississauga, Canada). Background absorbance of medium in the absence of cells was subtracted. All samples were assayed in duplicate, and the mean for each experiment was calculated. The relationship of the absorbance of MTT to cell number was verified in separate experiments in which 1,200 to 37,000 cardiomyocytes were added to different wells of a microtiter plate. There was a significant (P < 0.01) linear relationship between cell number and absorbance with a correlation coefficient of 0.98.

Trypan blue exclusion. Cardiomyocytes were grown on coverslips in dishes for 72 h in culture before treatment with SNP. After the addition of SNP to the media, dishes were returned to the incubator for 24 h. The coverslips were removed, stained with 0.4% trypan blue, and then examined microscopically in a hemocytometer to determine the total number of cells and the number of dead cells, i.e., those retaining trypan blue. Four representative high-power fields (hpf) of each slide, ~300 cells, were assessed, and the mean was calculated per coverslip.

Microscopy

H₂O₂ Assay

7

Nitrite Assay

Flow Cytometry (FACS Analysis)

To examine nuclear characteristics based on PI staining after cardiomyocytes were exposed to drugs, the reaction was stopped by the removal of media from the cells followed by brief exposure to trypsin (0.01% in DMS8) to suspend the adherent culture. Trypsinization was halted by dilution with 818A media, which contains 6% fetal calf serum. The suspended cardiomyocytes were gently spun down and washed with PBS. Cells were then permeabilized with 70% ethanol for 30 min at room temperature. Cells were then gently spun down, and the ethanol was removed. The resulting permeabilized cells were then stained with PI staining mix (0.1% Triton X-100, 0.1 mM EDTA, 0.5 mg/ml RNAse A, and 50 µg/ml PI) to visualize the nuclei.

Materials

Data Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SNP treatment of cardiomyocytes produced a dose-dependent reduction in cell survival. Cardiomyocytes treated with SNP for 24 h displayed a significant (P < 0.001) dose-dependent increase in the proportion of dead cells determined by the trypan blue assay (Fig. 1). The proportion of dead to live cells per hpf increased from 0.25 ± 0.06 (means ± SE) in control to 0.96 ± 0.33 for 0.1 mM SNP to 1.5 ± 0.2 for 0.5 mM SNP and to 4.9 ± 0.7 for 1.0 mM SNP. Each concentration of SNP was significantly (P < 0.01) greater than control. The percentage of dead cells per hpf increased significantly (P < 0.01) from control of 20.3 ± 2.2 to 48.9 ± 10.0%, 57.3 ± 6.8, and 81.8 ± 11% for, respectively, 0.1, 0.5, and 1.0 mM SNP. Evaluation of cell survival by the MTT assay demonstrated a significant (P < 0.0001) dose-dependent reduction in MTT absorbance in cardiomyocytes treated with SNP for 24 h (Fig. 2). On the basis of the ability of viable cells to reduce MTT and the direct relationship between cardiomyocyte cell number and MTT absorbance, SNP induced a significant loss of cell viability (see inset, Fig. 2). The EC₅₀ was 0.53 mM (confidence intervals 0.45-0.59 mM).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Cardiomyocyte viability after treatment with sodium nitroprusside (SNP) as assayed by trypan blue exclusion. Cells were treated with 0.1 mM (n = 4), 0.5 mM (n = 4), or 1.0 (n = 4) SNP or control (n = 14) for 24 h. The proportion of dead to alive cells per high power field (hpf) was calculated. Four representative hpfs are counted with a total of ~300 cells, and the mean is calculated for each coverslip. The percentage of dead cells of the entire population per hpf was calculated and is shown in the inset. The results are presented as the means ± SE, and the data with SNP are compared with control (**P < 0.01).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Cardiomyocyte viability after treatment with SNP as assayed by the MTT assay. Cardiomyocytes were seeded in multiwell microtiter at 20,000 cells per well. After 72 h, cells were treated with SNP for 24 h. MTT dye was added to each well for the last 4 h of treatment. The reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for 0.01 mM (n = 7), 0.1 mM (n = 30), 0.5 mM (n = 40), or 1.0 mM (n = 17) SNP and control (n = 40) for 24 h. Hypothesis testing compared SNP and control cells (*P < 0.05; **P < 0.01). The inset shows loss of cell viability as % control.

Because the iron moiety of SNP mediates some of the cytotoxicity of SNP (37), cardiomyocytes were treated with the iron chelator deferoxamine, and the impact of SNP-mediated cell death was ascertained. A SNP concentration of 0.5 mM was chosen because it is close to the EC₅₀ and it significantly reduces cell viability. Deferoxamine, 0.001 µM, which had no cytotoxicity, did not alter SNP-induced loss of cell viability (Fig. 3). A higher concentration of deferoxamine, 0.01 µM, which slightly reduced cardiomyocyte viability alone, in combination with SNP did not prevent SNP-induced cell death.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   The effect of deferoxamine (Def) on cardiomyocyte viability after treatment with SNP as assayed by the MTT assay. Cardiomyocytes were seeded in multiwell microtiter at 20,000 cells per well. After 72 h, cells were treated with SNP with or without Def for 24 h. MTT dye was added to each well for the last 4 h of treatment, the reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for 0.001 µM (n = 4) or 0.01 µM Def (n = 11), 0.05 mM SNP (n = 40), 0.5 mM SNP + 0.001 µM (n = 4) or 0.01 µM Def (n = 7), and control (n = 40) for 24 h. The inset shows the loss of cell viability as a percentage of control. Hypothesis testing compared each data set and control (*P < 0.05 or **P < 0.01). There were no differences between SNP + Desf and SNP alone.

Because metabolites of SNP may undergo redox cycling with oxygen to produce H₂O₂ (36), we sought to determine whether catalase that degrades H₂O₂ would antagonize the action of SNP (Fig. 4). Cardiomyocytes were pretreated with catalase beginning 2 h before SNP. Catalase, 1.0 mg/ml, completely blocked the reduction of cell viability induced by 0.1 mM SNP and significantly (P < 0.05) blunted the effect of 0.5 mM SNP. We then sought confirmatory evidence that SNP induced H₂O₂ production in these cells. Measurement of H₂O₂ release into the media revealed that SNP produced a significant (P < 0.05) increase in H₂O₂ after 6 or 24 h of SNP treatment that was, respectively, 3.5 ± 0.2- and 4.3 ± 0.7-fold greater than control (Fig. 5). In contrast, in the absence of cells, there was no significant change in H₂O₂ production, as H₂O₂ was 1.0 ± 0.1- and 1.0 ± 0.2-fold times control for, respectively, 0.5 mM SNP for 24 h (n = 4) and 0.5 mM SNP plus 10⁷ M H₂O₂ (n = 3).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   The effect of catalase (Cat) on cardiomyocyte viability after treatment with SNP as assayed by the MTT assay. Cardiomyocytes were seeded in multiwell microtiter at 20,000 cells per well. After 72 h, cells were treated with SNP with or without Cat for 24 h. MTT dye was added to each well for the last 4 h of treatment, the reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for 0.1 mM SNP (n = 30), SNP + 0.1 mg/ml Cat (n = 6), SNP + 1.0 mg/ml Cat (n = 7), 0.5 mM SNP (n = 40), SNP + 0.1 mg/ml Cat (n = 5), SNP + 1.0 mg/ml Cat (n = 10), and control (n = 40) for 24 h. The inset shows the loss of cell viability as a percentage of control. Hypothesis testing compared each data set to control (*P < 0.05 or **P < 0.01) or SNP (P < 0.05).

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Ability of SNP to generate H2O2. Cardiomyocytes were grown in petri dishes for 72 h, after which they were treated with 0.5 mM SNP for 6 or 24 h. The medium was removed and assayed at 590 nm using a quantitative assay kit for H2O2. A standard curve for absorbance of H2O2 permitted calculation of H2O2 in the medium. The effect of endogenous iron was eliminated by subtracting a blank with reagents, and, to eliminate the effect of SNP itself, media plus SNP without cells was also measured and subtracted from the samples of cells treated with SNP. Top, data after 6 and 24 h of SNP treatment; bottom, data compared with control cells. The data are presented as the means ± SE (*P < 0.05; **P < 0.01).

Because H₂O₂ production is thought to be mediated by NO release from SNP (33) and NO release indicates SNP metabolism (29), we sought to obtain an indication of the amount of NO released during SNP treatment. NO_x concentration in the media was determined after treatment of cardiomyocytes with SNP for 24 h. There was a significant (P < 0.05) increase in NO_x generation concentration with SNP treatment (Fig. 6). SNP, 0.5 mM, produced a 4.6 ± 0.5-fold increase in NO_x concentration compared with control. Cyanide can also be released by SNP metabolism. Although it is difficult to know the extent of SNP metabolism in cell culture and cyanide liberation, considering that 1 mM SNP produced an increase in NO_x in the 1 µM range, we examined the effect of cyanide on cell viability of cyanide ~1 µM. We did not find any increase in cell death assessed by the MTT assay after exposure of cardiomyocytes to cyanide (KCN) of 1 × 10⁶ M for 24 h [2.5 ± 3.1% (n = 8)], whereas 10⁵ M produced a small 13.3 ± 4.4% (n = 13) reduction in cell viability.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Nitric oxide (NO) release in SNP-treated cardiomyocytes. Cardiomyocytes that had been in petri dishes for 72 h after culture were treated with SNP for 24 h, after which the medium was removed. Medium, 100 µl, was added to equal volumes of the Greiss reagent, and the absorbance was measured at 570 nm. SNP, 0.1 (n = 4), 0.5 (n = 7), and 1.0 mM (n = 4), and control (n = 7) are shown. The inset shows the change relative to control. The data are presented as the means ± SE (**P < 0.01 from control).

To further examine the relationship between SNP and H₂O₂, it was first necessary to determine the concentration-effect relationship between chronic treatment with H₂O₂ and cell viability (Fig. 7). Cardiomyocytes were treated with a range of H₂O₂ concentrations for 24 h. H₂O₂ produced a significant (P < 0.001) dose-dependent reduction in MTT absorbance and reduction in cell viability (Fig. 5) The EC₅₀, the H₂O₂ concentration at which there was a 50% increase in cell death, was 4.2 × 10⁴ M (confidence interval 1.2-8.4 × 10⁴ M). The nature of the concentration-effect relationship was such that there was a small nonsignificant reduction in cell survival, <10%, at H₂O₂ concentrations of 10⁶ M or less. To examine for a potential interaction of SNP with a low concentration of H₂O₂ that induced minimal cell death, 10⁶ M H₂O₂ was selected (Fig. 8). H₂O₂, 10⁶ M, increased cell death 9.9 ± 7.3% compared with control; 0.1 mM SNP increased cell death 20.2 ± 5.9%; and the combination of both agents increased cell death 33.9 ± 11.1%, similar to the expected arithmetic sum of these concentrations of H₂O₂ and SNP. Higher concentrations of 0.5 mM SNP increased cell death by 56.1 ± 4.9%. The combination of this concentration of SNP plus 10⁶ M H₂O₂ increased cell death by 59.1 ± 6.5%, which is similar to the arithmetic sum of SNP plus H₂O₂. To examine the impact of catalase on the combination of SNP, the cardiomyocytes were pretreated with catalase beginning 2 h before SNP plus H₂O₂ treatment (Fig. 9). Catalase antagonized the reduction in cell viability produced by the combination of H₂O₂ and 0.1 mM SNP and significantly (P < 0.05) blunted the effect of the combination of H₂O₂ and 0.5 mM SNP.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   The effect of H2O2 on cardiomyocyte viability as assayed by the MTT assay. Cardiomyocytes were seeded in multiwell microtitre at 20,000 cells per well. After 72 h, cells were treated with H2O2 for 24 h. MTT dye was added to each well for the last 4 h of treatment, the reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for 108 M (n = 12), 106 M (n = 30), 105 M (n = 15), 104 M (n = 8), and 103 M (n = 4) H2O2, and control (n = 40) for 24 h. The inset shows the loss of cell viability as a percentage of control. Hypothesis testing compared each data set and control (**P < 0.05).

View larger version (26K): [in this window] [in a new window]   Fig. 8.   The effect of the combination of H2O2 and SNP on cardiomyocyte viability as assayed by the MTT assay. Cardiomyocytes were seeded in multiwell microtiter at 20,000 cells per well. After 72 h, cells were treated with SNP for 24 h. MTT dye was added to each well for the last 4 h of treatment, the reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for control (n = 40), 106 M H2O2 (n = 30) (shown on both sides of the graph), 0.1 mM SNP (n = 4), SNP plus H2O2 (n = 13) (shown on the left side of the graph), 0.5 SNP (n = 40), and 0.5 mM SNP plus H2O2 (n = 23) (shown on the right side of the graph) for 24 h. The inset shows the loss of cell viability as a percentage of control. Hypothesis testing compared data set and control (*P < 0.05; **P < 0.01).

View larger version (28K): [in this window] [in a new window]   Fig. 9.   The effect of catalase on the combination of H2O2 and SNP on cardiomyocyte viability. Cardiomyocytes were seeded in multiwell microtiter at 20,000 cells per well. After 72 h, cells were treated with SNP plus H2O2 for 24 h with and without pretreatment with catalase. MTT dye was added to each well for the last 4 h of treatment, the reaction was stopped, and the absorbance (optical density) was determined at 570 nm. Background absorbance of medium in the absence of cells was subtracted. The results are presented as the means ± SE for 106 M H2O2 (n = 30), 0.01 mM SNP (n = 4), SNP plus H2O2 (n = 13), 0.5 SNP (n = 40), and 0.5 mM SNP plus H2O2 (n = 23), and control (n = 40) for 24 h. The inset shows the loss of cell viability as a percentage of control. Hypothesis testing compared data set and control (*P < 0.05; **P < 0.01) or compared with SNP (P <0.05).

Microscopic examination of cardiomyocytes was next undertaken to explore the structural nature of SNP-induced cardiotoxicity and the potential interaction with H₂O₂. Cardiomyocytes were examined after staining with acridine orange, which preferentially stains nuclei and lysosomal structures. SNP, 0.5 mM, produced a reduction in cell size accompanied by nuclear fragmentation and membrane bleb formation (Fig. 10). Low concentrations of H₂O₂ produced little morphological alteration of cell structure, with the most notable being reduction in cell size and slight nucleolar fragmentation. H₂O₂ plus SNP produced a dramatic alteration in cell structure with membrane bleb formation, shrunken cells, and nuclear fragmentation. There was more intense cytosolic staining that reflects, in part, lysosomal changes that occur with reduced pH in these structures.

View larger version (89K): [in this window] [in a new window]   Fig. 10.   The cell morphology of cardiomyocytes after SNP and H2O2. Cardiomyocytes were grown in culture on coverslips for 72 h and then treated with 0.5 mM SNP, 108 M H2O2, and the combination. Cells were stained with acridine orange and examined with a fluorescent microscope at ×1,600. Representative photomicrographs are shown of control cardiomyocytes (A), SNP (B), 108 M H2O2 (C), and the combination (D).

To determine the nature of the cardiomyocyte death, cells were treated with SNP and subjected to FACS analysis using two different approaches. The first approach examined for the presence of apoptotic cells focused on the nuclear changes of apoptosis. Cells were first permeabilized, allowing PI to enter the cell and stain the nucleus. The intensity of PI staining is indicative of the amount of DNA. Cells undergoing apoptotic cell death are characterized by low DNA content or very low PI nuclear staining (30). A representative experiment shows the cell population distribution according to the degree of PI staining (FL3) (Fig. 11A). Of the 10,000 cells analyzed, only a very small proportion showed PI (staining) fluorescence around 10. In contrast, the proportion of the population with this degree of low staining (hypoploidy) was increased in SNP-treated cells. Indeed, 0.5 mM SNP produced a significant (P < 0.001) and 2.1 ± 0.2-fold increase in apoptosis compared with control (Fig. 11B).

View larger version (13K): [in this window] [in a new window]   Fig. 11.   A: DNA content frequency histograms of cardiomyocytes treated with SNP determined by FACS analysis. Cardiomyocytes were treated with SNP 0.5 mM for 24 h. Approximately 10,000 cells permeabilized with ethanol and stained with propidium iodide (PI) were aspirated into the flow cytometer and examined for fluorescence. Representative studies are shown for control (top) and SNP-treated (bottom) cardiomyocytes. Apoptotic cells (Apo) are identified by low PI staining (FL3). B: apoptotic population defined by low DNA content (PI staining) is shown in control and SNP-treated cardiomyocytes. The proportion of the population with apoptotic nuclei per 100 cells is shown at top and the ratio of apoptotic nuclei in SNP-treated cells compared with control is shown at bottom. Data analysis was performed comparing SNP with control (*P < 0.05; **P < 0.01).

With a low H₂O₂ concentration, 10⁸ M, that did not induce cell death, cardiomyocytes were treated with the combination of H₂O₂ and 0.5 mM SNP and were examined by FACS (Fig. 12). H₂O₂ plus SNP increased the proportion of the cell population with low PI staining indicative of apoptosis. SNP, 0.5 mM, produced a 2.1 ± 0.2-fold increase in the apoptotic population, whereas the combination of SNP plus 10⁸ M H₂O₂ produced a 3.0 ± 0.5-fold increase in the population of cells with apoptotic nuclei compared with control cells untreated with either SNP or H₂O₂. To further examine the relationship between SNP and H₂O₂, a second FACS technique, dual staining with FDA and PI, was used. The FDA dye readily enters the cell in the uncharged form. FDA is a substrate for esterases, an enzyme ubiquitous in most cells, which converts FDA into a charged, fluorescent form. If the cell is alive with an intact membrane, FDA will remain trapped within the cell. Necrotic cells have damaged plasma membranes, and FDA readily leaks out of the cell, whereas apoptotic cells have intact plasma membranes. In the late stages of apoptosis, there is loss of membrane integrity so the necrotic population is a mixture of necrotic and apoptotic cells. PI is a viability exclusion dye because PI does not enter viable cells but enters dead cells, where it primarily stains nuclei. Hence, early necrotic cells will exhibit poor membrane integrity and death (low FDA and high PI), whereas early apoptotic cells exhibit good membrane integrity with little death (low FDA and low PI). In the later stages of cell death, cellular debris is present and is not recognized by either stain. A representative dot plot exhibits three distinct populations: alive, apoptotic, and necrotic cells defined by flow cytometry analysis of cardiomyocytes dually stained with FDA and PI (Fig. 13). Low concentrations of H₂O₂, 10⁸ M, did not change the proportion of apoptotic or necrotic cells, whereas 0.5 M SNP increased the population of apoptotic and necrotic cells. The combination of H₂O₂ plus SNP produced a marked increase in the cell population of apoptotic cells. SNP produced a significant (P < 0.05) and 4.1 ± 0.8-fold increase in the apoptotic population and a significant (P < 0.001) and 3.5 ± 0.6 increase in the necrotic population (Fig. 14). The combination of H₂O₂ plus SNP produced a 5.7 ± 1.5-fold increase in the apoptotic population compared with control. In contrast to apoptosis, there was no increase in the necrotic population with the combination of H₂O₂ plus SNP compared with SNP alone, i.e., respectively 3.6 ± 0.5- versus 3.5 ± 0.6-fold greater than control.

View larger version (16K): [in this window] [in a new window]   Fig. 12.   The effect of SNP on nuclear evidence of apoptosis. Top: proportion of cells with apoptotic nuclei per 100 cells; bottom: apoptotic nuclei relative to control. Cardiomyocytes were treated with 0.5 mM SNP (n = 5), 108 M H2O2 (n = 5), and the combination (n = 6), and control cells (n = 6), that had been permeabilized with ethanol, stained with PI, were aspirated into the flow cytometer. The proportion of the 10,000 cells with low DNA content (PI staining) defining the apoptotic population was calculated. The data for control and SNP are repeated from Fig. 9 for ease of comparison. The data are presented as the means ± SE with data analysis compared with control (*P < 0.05; **P < 0.01) or SNP (P < 0.05).

View larger version (39K): [in this window] [in a new window]   Fig. 13.   Cell population characteristics were assessed using flow cytometry. Cardiomyocytes were dually stained with fluorescein diacetate (FDA) (y-axis) and propidium iodide (x-axis). The defined apoptotic (A) population [low fluorescein diacetate (FDA), low PI] was observed in cardiomyocytes (circled population) and a necrotic (N) population defined as low FDA with higher PI. Four representative studies of 10,000 cells each showing control, 0.5 mM SNP, 108 M H2O2, and 108 M H2O2 plus SNP.

View larger version (22K): [in this window] [in a new window]   Fig. 14.   The proportion of apoptotic and necrotic cells as defined by dual staining with FDA and PI. Cardiomyocytes treated with 0.5 mM SNP (n = 5), 108 M H2O2 (n = 5), the combination (n = 6), and control cells (n = 6) were stained with FDA and PI and aspirated into the flow cytometer. The proportion of the population of 10,000 cells that were classified as apoptotic are shown at top and as necrotic at bottom, with the ratio of treatment to control displayed at right. The data are presented as the means ± SE with data analysis compared with control (**P < 0.01) or SNP (P < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This study, to our knowledge, is the first detailed report of the ability of nitroprusside to induce cell death in cardiomyocytes. We demonstrated the cardiotoxicity of SNP by finding that SNP damaged cellular functions that are integral markers of cell viability, specifically loss of plasma membrane integrity, as assessed by the inability to exclude trypan blue or PI from entering the cell and the failure to keep a fluorescent probe from leaving the cell. These data were further supported by SNP-induced damage to mitochondrial function, as assessed by the lack of mitochondrial dehydrogenase to reduce MTT (18, 26). The adverse impact of SNP on cardiomyocyte cell viability is consistent with the findings in murine neuroblastoma N1E-115 cells (47), the brain (37), and vascular smooth muscle cells (32, 44). In other cell types, SNP does not induce cell death (1, 7, 8, 10, 13). Although the toxicity of SNP may be specific for only certain cell types, it is noteworthy that our finding that 1.0 mM SNP produced an 87% increase in cell death in cardiomyocytes is similar to the 95% reported by Yamada et al. (47) in neuroblastoma cells with the same SNP concentration.

Although several factors have been implicated to explain SNP-induced cell death, the precise mechanism has been uncertain (2, 27, 36, 37). Our data provide a compelling argument for the involvement of free radicals in SNP-induced cardiotoxicity. We and others have documented SNP-induced H₂O₂ production in cell culture (47) as SNP metabolism yields metabolites that undergo redox cycling with oxygen to produce O₂· (36). The O₂· formed during this cycle readily produces H₂O₂ (36). Treatment with SNP required 6 h before loss of cell viability was recognized (47), consistent with the time lag needed to produce a metabolite such as H₂O₂. We also demonstrated NO production from SNP in cardiomyocytes. Although NO can be released spontaneously from SNP (19), in the absence of cells, SNP releases trivial amounts of NO (19, 20). NO is released from SNP following metabolism involving one electron reduction (36). NO induces mitochondria H₂O₂ production through an effect on cytochrome oxidase explained by Ponderoso et al. (33) and demonstrated to occur in the isolated heart (33). The potential of SNP toxicity directly from NO is possible, but other NO donors such S-nitrosoglutathione and S-nitroso-N-acetyl-penicillamine afforded protection against reactive oxygen species-induced toxicity in some tissues and situations (44). The role of H₂O₂ in SNP-mediated cardiac toxicity is supported by our finding that catalase completely prevented SNP-induced cell death at low SNP concentrations and significantly blunted it at high SNP concentrations. Although we cannot exclude the possibility that the preparation of catalase may have small amounts of superoxide dismutase, the dominant effect should be that of catalase. Our findings with catalase are supported by the data that catalase has the capacity to totally inactivate H₂O₂ (5) and exogenous catalase that can enter or become associated with myocytes inhibits H₂O₂-induced cell death in this cardiomyocyte type (17).

In contrast to the protective effects of catalase, we found no evidence to support the contention that the iron moiety of SNP mediates cell toxicity in cardiomyocytes compared with the findings in other cell types (27, 36). The iron chelator deferoxamine did not alter SNP-induced cardiomyocyte cell death. Although there are data that deferoxamine can blunt H₂O₂-induced cardiomyocyte cell damage, as reflected in loss of lactate dehydrogenase activity, deferoxamine does not significantly eliminate H₂O₂-mediated cardiomyocyte injury (5). An alternative possibility is that any potential beneficial effects of deferoxamine on SNP-mediated cardiotoxicity is offset by the deleterious effects of deferioxamine reacting with hydroxyl radicals to produce a nitroxide free radical (31). Another potential mechanism for SNP-induced cardiotoxicity is the release of cyanide from the SNP molecule (25). However, cyanide itself does not readily induce cell death, as we found in these cardiomyocytes and others have found in different cells culture (29, 47). Further cyanide antidotes and trapping agents do not ameliorate but rather increase SNP-induced cytotoxicity and lipid peroxidation in isolated hepatocytes (29). Although a role for cyanide in SNP-induced cardiotoxicity has not been excluded, the ability of catalase to effectively reverse SNP-induced cell death at some SNP concentrations strongly implicates a major role for H₂O₂ in the pathogenesis of SNP-induced cardiomyocyte cell death. Furthermore, cyanide does not induce the release of H₂O₂ (12).

SNP metabolism occurs rapidly. Within minutes of its intravenous administration in animals and humans, SNP is metabolized (39), making calculation of serum concentration and volume of distribution difficult. SNP metabolism also occurs in cell culture (29, 47) as contact with the sulfhydryl groups in the cell wall immediately initiates SNP degradation (39). Although the rates of SNP metabolism in cell culture appears to be slower than those in humans (29), the rapid metabolism presents problems in extrapolating the data from cell culture to intact animals and humans. High concentrations of SNP were used in the present study. However, the SNP concentrations were the same or even less than other studies of the cytotoxicity of SNP in other cell types in culture (29, 47). Clinically, SNP is administered up to 800 µg/min so patients may receive >1 g in 24 h (39). The volume of distribution of SNP is not known, but it is likely large enough to considerably reduce the effective SNP concentration. The sensitivity of the human myocardium to SNP and its metabolites are additional factors that would influence the potential for SNP-induced cardiotoxicity in humans.

Apoptotic cell death was demonstrated in cardiomyocytes after SNP treatment. Cardiomyocyte cell structure was altered by SNP treatment with cardiomyocytes displaying the characteristic features of apoptosis, namely, reduction in cell size, membrane bleb formation, and nuclear disintegration (46). The DNA staining characteristics of cardiomyocyte nuclei evaluated by flow cytometry demonstrated SNP-induced alterations of nuclear structure characteristic of apoptosis (30). Evaluation of membrane permeability by dual staining with FDA and PI (16) also established that SNP treatment produced a significant increase in apoptosis. The degree of apoptotic cell death was between twofold when assessed by nuclear PI staining and fourfold when assessed by dual staining with PI and FDA. The differences in the magnitude of apoptosis reflect the different methodologies as the former technique examines the nuclear properties of permeabilized cells and demonstrates hypoploidy or low DNA content due to DNA fragmentation, a late event in the process of apoptosis. The dual-staining approach with FDA and PI evaluates membrane integrity and esterase activity of intact cardiomyocytes, an early event in apoptosis. SNP was also found to damage cellular membrane integrity in a manner consistent with necrotic cell death; however, in the late stage of apoptosis there is also a loss of membrane integrity as the cell dies (43). Our demonstration of SNP-induced apoptosis in cardiomyocytes is consistent with the findings in other cell types (32, 42, 44, 47). In addition, we were able to determine the magnitude of the effect of SNP on cell viability and apoptosis and demonstrated that SNP-induced membrane alterations of the type associated with necrotic cell death was the predominant change produced by SNP.

A major contribution of the present study is the demonstration that low concentrations of H₂O₂ at levels that do not alter cardiomyocyte viability produce a significant enhancement of SNP-induced apoptosis. The concentration-effect relationship between H₂O₂ and cardiomyocyte viability was established and was consistent with the findings of Byler et al. (5), who used the same cell type and duration of H₂O₂ treatment but only a single high H₂O₂ concentration. We chose low concentrations of H₂O₂ that produced minimal or no significant damage to cardiomyocytes; however, when combined with SNP, there was a significant increase in apoptotic cell death. Interestingly, the amount of necrotic cell death was not altered by the combination of SNP and H₂O₂. As the later mode of cell death predominated in SNP-induced cardiotoxicity, the potential synergy between SNP and H₂O₂ on apoptotic cell death could not be identified using a total assessment of cellular death as measured by MTT reduction. Recognizing the need for caution in using a single-parameter measurement as an indicator of apoptosis (16), we used two different flow cytometry approaches. FDA/PI staining of intact cells and PI staining of nuclei of permeabilized cells provided concordant data of a synergistic effect of SNP and H₂O₂. Cellular structural alterations reflective of apoptosis (46), as determined by fluorescent microscopy, were also accentuated when H₂O₂ was combined with SNP. Taken together, these data indicate that the combination of SNP plus H₂O₂ is especially toxic to cardiomyocytes, and the effect is essentially restricted to apoptotic cell death, especially in the early phase of apoptosis. The adverse effect of the combination of nitroprusside and H₂O₂ has been recognized by H₂O₂ enhancement of the tumoricidal activity of NO in Fu5 cells (20), cytotoxicity in V79 Chinese hamster lung fibroblasts (45), and DNA fragmentation and cell lysis in murine lymphoma cells (15). In contrast to these studies that used lethal concentrations of H₂O₂, our data is novel in the use of low, nontoxic concentrations of H₂O₂ as well as primary cells in culture rather than tumor cells or a cell line.

Extrapolation of the data must take into account the embryonic and avian nature of the cardiomyocytes examined. However, this model has several advantages, especially the ability to study the morphological aspects of apoptosis in the entire cell. Second, these cells are ideally suited for the study of chronic exposure to agents because they are stable for days in culture and our studies used prolonged, 24-h exposure to SNP and H₂O₂. Another positive feature of using this cell type is that elements of the apoptotic pathway such as the bcl-2 gene are abundant in the chick heart and show a high sequence homology to human bcl-2 (11). When the caveats associated with extrapolation from cell culture data to patients are recognized, the data suggest the need for caution in the use of SNP in situations associated with increases in myocardial H₂O₂, such as myocardial ischemia and reperfusion or heart failure (3, 14, 38). The clinical implications of the use SNP in situations with enhanced free radical generation may warrant investigation from the perspective of enhanced apoptotic cell death.

In summary, our results present novel findings for cardiomyocyte injury induced by SNP. We established that the mode of SNP-induced cardiomyocyte cell death includes apoptosis. Because of the nature of the nitroprusside molecule, we focused on free radical generation as a potential cause of SNP-induced cardiotoxicity and implicated H₂O₂ by demonstrating SNP-induced H₂O₂ generation and reversal of SNP-induced loss of cell viability by catalase. The toxicity of H₂O₂ to cardiac cell is consistent with many studies (3, 4, 5). We extended those results by demonstrating that under certain circumstances even low concentrations of H₂O₂ have the potential for toxicity.