INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

NITRIC OXIDE (NO) can affect a number of processes associated with cardiac blood flow and metabolism. These properties of NO have made it a central issue in many models of cardiac metabolic regulation. NO is a potent vasodilator in the coronary vascular bed, and modification of the activity of NO synthesis has effects on vascular tone regulation (16). These results suggest that NO plays a key role in the regulation of coronary blood flow (6, 18). In isolated mitochondria, NO has been shown to inhibit ATP production (4), presumably through the inhibition of cytochrome oxidase (5). Thus both of these important processes in cardiac energy metabolism seem to have a common potential regulatory agent in NO. Because NO is believed to be primarily produced from the vascular endothelial cells, most models suggest that NO might provide a signaling pathway between the blood vessels and cardiac myocytes. However, the high concentration of myoglobin in heart cells (1) would predictably buffer the short-lived NO molecule making it unlikely that a NO concentration ([NO]) signal could rapidly traverse the cytosol to the mitochondria. The putative myoglobin buffer would blunt any transient No signaling from the vasculature.

Recently, it has been shown that a NO synthase (NOS) is present in rat liver mitochondria (14). Monoclonal antibody studies have localized NOS to the mitochondria of rat hearts (3, 10) as well as skeletal muscle (11). Mitochondrial NOS might provide a local source of NO that could be an effective modulator of oxidative phosphorylation even with a cytosol rich in myoglobin. The sensitivity of NOS to Ca²⁺ (7) [a putative regulator of oxidative phosphorylation (15)] as well as other metabolites including NADPH (8) suggests that NOS could play an important role in the regulatory network controlling aerobic ATP production in the heart.

The purpose of this study was to determine whether NOS was present in porcine heart mitochondria and whether it could be induced to generate a sufficient amount of NO to modify oxidative phosphorylation. This was accomplished by evaluating several substrates and inhibitors of NOS on aerobic respiration in isolated porcine heart mitochondria. In addition, direct assays of NO production as well as NOS protein and activity were performed on this preparation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cardiac mitochondria preparation. Porcine heart mitochondria were isolated as previously reported (15). Isolated mitochondria had a respiratory quotient (state 3 respiration/state 4 respiration) of >10 using 5 mM glutamate and 5 mM malate as substrates (15). Experiments were run in experimental buffer solution containing (in mM) 125 KCl, 15 NaCl, 20 HEPES, 1 EDTA, 1 EGTA, and 5 MgCl₂ and 800 µM of total (535 nM free) Ca²⁺ at pH 7. All experiments were run at 37°C. The concentration of mitochondria used in these studies was 1 nmol of cytochrome a (cyt a) per milliliter of buffer unless otherwise specified.

Cyt a and protein content. Mitochondria concentration was determined from the spectrophotometric determination of cyt a content as previously described (2). Protein concentration was determined by biuret assay (9) with the modifications introduced by Yonetani (17), and bovine serum albumin was used as a standard.

Monitoring of chamber O₂ and NO content. O₂ consumption was monitored using a YSI Clark electrode (Yellow Springs, OH) that was calibrated to 100% air saturated at 37°C as previously described (12). NO was simultaneously measured using a NO-selective electrode (WPI; Sarasota, FL) inserted into the water-jacketed chamber. The NO-selective electrode was calibrated using injections of nitrite into 0.1 M KI and 0.1 M H₂SO₄.

Determination of L-citrulline. Porcine heart mitochondria were frozen and thawed three times using liquid nitrogen. An aliquot of these samples (containing 1-2 mg of protein/ml) was incubated (and constantly stirred) with 2 ml of solution containing 1 mM MgCl₂, 0.2 mM CaCl₂, 30 µM L-arginine, 0.1 mM NADPH, and 0.1 M triethanolamine (pH 7.4) supplemented with 10 µM tetrahydrobiopterin. When N-monomethyl-L-arginine (L-NMMA) was used in the incubation instead of L-arginine, the mitochondrial protein concentration was 3-5 mg/ml. L-Citrulline produced during the NOS-catalyzed reaction was measured using a colorimetric assay essentially as described by Prescott and Jones (13).

SDS-PAGE analysis. Samples from porcine heart mitochondria were separated by SDS-PAGE using three 10% polyacrylamide gels. Samples were heated for 5 min at 95°C in a sample buffer that contained 10 mM Tris · HCl (pH 8.0), 1 mM EDTA, 2.5% (wt/vol) SDS, and 2.5% mercaptoethanol. The molecular weight markers (Amersham Life Science) were subjected to the same treatment as the samples before electrophoresis. After the samples were cooled, 0.01% (wt/vol) bromophenol blue was added, and 10 µg of protein were loaded per lane. Gels were electrophoresed using a Bio-Rad (Hercules, CA) apparatus set at 200 V and 15 mA for 1 h at 4°C. For Western blot analysis, the proteins were blotted by electrodiffusion for 3 h at 40 V on nitrocellulose membranes (0.45-µm pore size Trans-Blot membranes; Bio-Rad). Membranes were blocked with 5% bovine serum albumin and 10% normal goat serum in Tris-buffered saline (TBS; 150 mM NaCl and 10 mM Tris · HCl; pH 7.6) with 0.05% Tween 20 (TBST) for 1 h. The membranes were thoroughly washed with TBST and incubated separately with mouse monoclonal antibody to macrophage NOS (macNOS), neuronal NOS (nNOS), and endothelial NOS (eNOS) (1/2,000 dilution in TBST) for 1 h. The membranes were extensively washed with TBST and subsequently incubated with goat antibodies conjugated with horseradish peroxidase to mouse IgG (1/30,000, Upstate Biotech; Lake Placid, NY) for 1 h. After the membranes were washed with TBST, the immunocomplexes were developed using an enhanced horseradish peroxidase/luminol chemiluminescence reaction, which was detected with photographic film (Hyperfilm ECL; Amersham) and recorded after 10 s to 3 min of exposure. A lysate of mouse macrophage RAW 264.7 cell line, rat pituitary lysate, and human endothelial cells were used as positive controls for inducible NOS (iNOS), brain NOS (bNOS), and eNOS, respectively.

Chemical sources. NADPH, ADP, ATP, NO, L-citrulline, L-arginine, 2,3-butanedione monoxine, antipyrine, and L-NMMA were purchased from Sigma (St. Louis, MO). The antibodies to iNOS, nNOS, and eNOS were purchased from Transduction Laboratories (Lexington, KY). The antibodies to iNOS (or macNOS) were obtained using a 21-kDa protein fragment corresponding to amino acids 961-1,144 of mouse macNOS as immunogen. All other reagents were of analytical grade.

Statistical analysis. Data are expressed as means ± SE. Statistical analysis was performed by one-way ANOVA.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Kinetics of NO inhibition of mitochondrial respiration. To confirm that NO inhibits cardiac mitochondrial O₂ consumption, the effects of NO on cardiac mitochondria were first determined. NO was directly injected into the chamber with mitochondria while O₂ consumption and NO content were simultaneously monitored. An example from these studies is shown in Fig. 1A, where ADP (2 mM final concentration) and NO (1-4 µM final concentration) were simultaneously injected into the chamber. With the injection of NO, the NO tension rapidly rose in the chamber and then quickly fell as it was metabolized. The NO was associated with inhibition of the ADP-dependent increase in respiration as previously described (4). As the NO was metabolized, the respiratory rate recovered in a dose-dependent manner. These data could be used to calculate several factors involving NO kinetics in isolated heart mitochondria. First, the rate of NO metabolism in the presence of the mitochondria was determined and found to be linear with the [NO]. This suggests that the rate of metabolism is [NO] dependent. Because no saturation of this effect was observed, we assumed that this reaction was a pseudo-first-order reaction with regard to [NO] (in µM) and mitochondria concentration, resulting in
where NO is the rate of NO metabolism and k is the rate constant, which was determined to be 0.91 × 10³ l · min¹ · nmol of cyt a¹ by performing a linear regression of the data in Fig. 1B. This mitochondrial NO metabolic rate constant can be extrapolated to the intact heart using cyt a conversion (37 nmol cyt a/g wet weight) (12) to 3.4 × 10² l · min¹ · g wet weight¹.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   A: mitochondrial respiratory response (change in O2 over time, dO2/dt) to nitric oxide (NO). Because NO gas decay was rapid, NO-saturated water was injected into the mitochondrial suspension [containing 1 ml of 1 nmol cytochrome a (cyt a) per milliliter] simultaneously with 2 µmol ADP. B: concentration-dependent rate of NO decay in the presence of mitochondria (1 nmol cyt a per milliliter). Various volumes of saturated NO solution were injected into mitochondria and the initial decay was monitored with a NO-selective electrode. Nitrite injected into 0.1 M H2SO4 and 0.1 M KI was used to calibrate the electrode. [O2], O2 concentration; [NO], NO concentration.

Mitochondrial NO generation. Establishing the kinetics for NO inhibition in this preparation, we set out to determine whether mitochondrial NOS could generate sufficient NO to inhibit mitochondrial respiration and therefore regulate ATP production. To accomplish this, we used previously established procedures to generate NO from endogenous NOS using L-arginine. An example of the basic assay procedure is shown in Fig. 2A, where the effects of L-arginine could be assessed for both state 4 (no ADP) and state 3 (saturating ADP and P_i) respiration. Ca²⁺ was added (535 nM free Ca²⁺) to ensure activation of mitochondrial ATP production (15) as well as any Ca²⁺-sensitive NOS (7). There were no significant decreases in state 3 (2 mM ADP) or state 4 respiration with the addition of 3-300 µM L-arginine (Fig. 2B). In addition, no increase in NO generation was observed using the NO-sensitive electrode at any dose of L-arginine (not shown). These data are not consistent with the significant generation of NO by mitochondrial NOS under these circumstances.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   A: mitochondrial respiratory response to L-arginine (L-Arg) injection. L-Arg (24 µmol) was injected into 3 ml of mitochondrial suspension (1 nmol cyt a per milliliter) in the presence of glutamate and malate. B: mitochondrial respiratory response to the NO synthase (NOS) substrate L-Arg during states 3 () and 4 (). Because state 3 respiratory rates were rapid, studies were conducted at a mitochondrial concentration of 0.5 nmol cyt a per milliliter. mO2, mitochondrial O2 consumption.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Mitochondrial respiratory response to the NOS inhibitor N-monomethyl-L-arginine (L-NMMA). States 3 () and 4 () were run with the same conditions as described in Fig. 2B with L-NMMA replacing L-Arg.

NOS activity and quantitation. Another method of monitoring NOS activity is to track the production of L-citrulline from L-arginine, which is cosynthesized with the production of NO by NOS (13). L-Citrulline was produced at a rate of 0.22 ± 0.03 nmol · mg mitochondrial protein¹ · min¹ in heart mitochondria. This activity is 14-fold smaller than that found in rat liver mitochondria (8). The pig heart mitochondria NOS activity was inhibited by L-NMMA and was negligible in the absence of L-arginine.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Equal protein amounts of porcine heart and rat liver mitochondria were loaded onto SDS-PAGE gels and electroblotted onto nitrocellulose membranes. Western blots were performed using monoclonal antibodies to inducible (iNOS), brain (bNOS) or neuronal (nNOS), and endothelial (eNOS) NOS. Positive controls were run using those provided by Transduction Laboratories.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The concentration of NOS in porcine heart mitochondria is ~10 times lower than in the rat liver based on Western blot assays as well as homogenate activity. NO production by NOS was not detected in mitochondria suspensions based on the lack of L-arginine-dependent or L-NMMA-sensitive NO production or any effect of these agents on oxidative phosphorylation. The discrepancy between the intact mitochondrial data and homogenate data is likely due to the rapid mitochondrial metabolism of NO. The NO metabolism was ~1-4 nmol · mg mitochondrial protein¹ · min¹ in the mitochondria, whereas the maximum rate of NO production in the homogenates was 0.22 nmol · mg mitochondrial protein¹ · min¹ based on L-citrulline production. This comparison suggests that any production of NO by mitochondrial NOS was immediately metabolized and resulted in undetectable levels of NO production. The fact that oxidative phosphorylation was not inhibited by NOS activity suggests that significant levels of NO were not being generated even in the mitochondrial matrix.

These data suggest that the small amount of NOS activity detected in heart mitochondria is unlikely to play a significant role in the regulation of oxidative phosphorylation due to the rapid metabolism of NO. It is possible that the NO metabolic rate in these isolated mitochondria is much higher than that occurring in the intact cell. However, no data on the in vivo rate of mitochondrial NO production or metabolism are currently available. All of these studies were conducted at extramitochondrial Ca²⁺ concentrations near the optimal (~500 nM free Ca²⁺) for stimulating oxidative phosphorylation (15). Studies were also conducted using Ca²⁺ concentrations two- to threefold higher to test for a Ca²⁺-sensitive NOS; however, no evidence of NO production was detected even at these higher Ca²⁺ levels.

The small amount of NOS detected in this mitochondrial preparation could be intimately associated with the mitochondria. The precise location within the mitochondrion is unknown, but with no NOS gene in the mitochondrial genome, its source must ultimately be the cytosol and a nuclear gene. A small cytosolic contamination causing the detection of NOS in this preparation cannot be ruled out. Enzymatic assays demonstrate that the preparation is highly purified based on the enrichment of the mitochondrial markers cyt a and citrate synthase. The levels of a cytosolic marker, phosphoglycerate kinase, in the preparation were <0.5% of the total tissue homogenate (not shown). Electron microscopy evaluation of the preparation also suggested very high purity based on visual inspection. However, these studies cannot rule out a small cytosolic contamination causing the residual NOS activity detected in this preparation.

In summary, these data suggest that the local production of NO by mitochondrial NOS is not significant even under optimal conditions and does not contribute to the regulation of oxidative phosphorylation in heart cells. If NO does play a role in the regulation of mitochondrial function, the NO must be reaching the mitochondria through the myoglobin-rich cytosol. This extramitochondrial NO production could be in the cytosol itself or from other cells (i.e., vascular endothelial cells) in the tissue. It is important to note that not all mitochondria in the myocytes are surrounded by myoglobin and that the subplasmalemma mitochondria could be almost immediately exposed to extracellular NO levels without myoglobin buffering. These mitochondria might be the most susceptible population to transient NO regulation of oxidative phosphorylation by vascular production.