| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Cardiovascular Division, University of Minnesota Medical School; 2Minneapolis Heart Institute Foundation at Abbott Northwestern Hospital; and 3Department of Biochemistry, University of Minnesota, Minneapolis, Minnesota
Submitted 25 April 2005 ; accepted in final form 9 January 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-phenyl-tert-butylnitrone (PBN) (0.14 mol/l) in blood collected from the aorta and coronary sinus (CS). To improve signal detection, the free radical adducts were deoxygenated over a nitrogen stream for 15 min and extracted with toluene. The hyperfine splitting constants of the radicals were N = 13.7 G and H = 1.0 G, consistent with an alkoxyl or carbon-centered radical. Resting aortic and CS PBN adduct concentrations were 6.7 and 6.3 x 108 arbitrary units (P = not significant). Both aortic and CS adduct concentrations increased during exercise, but there was no significant difference between the aortic and CS concentrations. Thus, in contrast to skeletal muscle, submaximal treadmill exercise did not result in detectable free radical production by the heart.
spin trap; coronary blood flow; myocardial oxygen consumption; electron paramagnetic resonance
Because of the exceedingly short half-life, in vivo detection of ROS has generally relied on the indirect detection of their activity using measurements of lipid peroxidation or chemiluminescence of isolated tissue specimens. Electron paramagnetic resonance (EPR) spectroscopy can provide highly sensitive measurements of ROS when used in conjunction with spin traps that act to stabilize the free radical and permit ex vivo measurements. This technique has recently been exploited in humans to produce the first measurements of free radical outflow from an isolated muscle bed in resting and exercising skeletal muscle (3, 4)
Although no previous studies have examined the effect of exercise on ROS production in the normal heart, several investigations have used EPR techniques in humans (15, 28) and animals (5, 6, 11) to demonstrate a robust free radical signal in coronary sinus blood during reperfusion after periods of myocardial ischemia. However, no significant EPR signal was detected in coronary sinus blood before the onset of reperfusion. This may suggest that the basal release of ROS is lower in the heart compared with skeletal muscle or that the myocardium is able to quench free radicals more effectively due to greater endogenous antioxidant activity.
The presence of molecular oxygen in the spin-trap adduct sample can interfere with EPR signal analysis. Because oxygen is a paramagnetic substance, it may contribute to line broadening and blurring of the spectra due, in part, to electron spin exchange between the colliding paramagnetic species (27). This observation is supported by Ashton et al. (2), who found that deoxygenation of their spin adduct samples by vacuum degassing resulted in the appearance of a significant EPR signal in mixed venous blood during rest and exercise in humans using the spin trap -phenyl-tert-butylnitrone (PBN). This additional treatment has not generally been used in EPR samples obtained from coronary venous blood, possibly contributing to the failure to detect a free radical signal in the heart during normal conditions.
Because increased oxidative stress appears to be a common link in the pathogenesis of many cardiac disorders, such as hypertension, congestive heart failure, and endothelial dysfunction, the ability to measure cardiac free radical production in vivo could be a potentially important tool to monitor disease activity and the response to therapy. In this study we used the spin trap PBN in conjunction with oxygen degassing to demonstrate for the first time a significant EPR signal in the coronary sinus blood of a normal awake dog during rest and exercise.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental protocol.
Studies were performed 2–3 wk after surgery. On the day of the study, pressures and CBF were recorded continuously with the dog standing on the treadmill. Aortic and coronary venous blood (6 ml each) were withdrawn for determination of resting myocardial oxygen consumption (M
O2) and PBN adduct concentration. Exercise was then begun at 3.2 km/h at 0% grade. Three minutes later, the treadmill speed and grade were increased to 6.4 km/h at 10% grade. Aortic and coronary venous blood specimens were withdrawn for blood gas analysis and PBN adduct concentration after 6 min of total exercise. MO2 was calculated by using CBF averaged over the last 30 s of rest and at the end of the second exercise stage.
Preparation of spin trap. The spin trap PBN (Sigma) was dissolved in normal saline (0.14 mol/l) in the dark, and 1.5 ml of the solution were placed in 6-ml glass serum separator tubes (shielded from light) and mixed with 4.5 ml of blood from the aorta or coronary sinus. The tubes were gently inverted 10 times and centrifuged for 10 min at 3,500 rpm. The separated plasma was mixed with an equal volume of spectroscopic grade toluene (Sigma), vortexed for 10 s, and recentrifuged for 10 min, and the organic layer containing the spin adduct was separated and stored at –80°C until EPR analysis.
Validation of spin-trap measurements of free radicals. To validate that the EPR signal was proportional to the spin-trap adduct concentration, a calibration curve was constructed after exposure of the spin trap to increasing concentrations of free radicals. Canine arterial blood (3.5 ml) was placed into 6-ml serum separator tubes (shielded from light) containing 1.5 ml of the spin trap PBN. A free radical generating system was created using the reaction of H2O2 (500 µmol/l) and ammonium iron (II)-sulfate (100 µmol/l) (11). Saline (1 ml; control) or reaction mixture (1 ml) obtained after 1 and 2 min of mixing was placed into the tubes containing blood and PBN. The samples were immediately centrifuged, extracted with toluene, and analyzed with EPR spectroscopy after oxygen degassing in an identical manner as performed in the study protocol.
EPR spectroscopy. EPR spectra were acquired with a Bruker EleXsys E500 spectrometer (Bruker Instruments, Billerica, MA), using the Bruker SHQ cavity with quartz Dewar (Wilmad). During the experiment, samples were maintained at 25°C in nitrogen atmosphere using a temperature controller, regulating gas flow through the Dewar with the sample. Spin-adduct EPR signal was observed only in deoxygenated samples. To increase EPR sensitivity and deoxygenate samples effectively, 10 Teflon tubes (22 gauge, 0.7 mm ID, and 70 mm long) were loaded with 200 µl spin adduct, sealed on both ends with Crytoseal, and placed into a quartz tube (Wilmad). The multibore sample was placed into a cavity dewar and extended through the entire 40-mm active height of the SHQ cavity. Samples were deoxygenated in the dewar for 15 min before spectra acquisition. Spectra were obtained with 70 G scan width, using 100 kHz field modulation with peak-to-peak amplitude of 0.5 G to determine hyperfine splittings of spin adducts and 5 G to quantify spin adduct concentration. The microwave frequency was 9.378 GHz, microwave power was 20 mW, magnetic field was center 3,345 G, receiver gain was 90 dB, time constant was 20.48 ms, and the number of scans were from 30 to 90. Because the PBN adduct concentration is proportional to the EPR signal, the quantity of radicals in each sample was obtained from the double integration of the EPR signal from a computer program modeled on the known triplet of doublets of the nitroxide radical. Toluene was scanned in each study to ensure it was devoid of any EPR signal.
MO2.
Aortic and coronary venous PO2, PCO2, and pH were determined with a blood gas analyzer (model 113, Instrumentation Laboratory). Blood oxygen content was calculated as (0.0136 x hemoglobin x %oxygen saturation) + (PO2 x 0.0031). MO2 in the LAD distribution was calculated as the arteriovenous difference of oxygen content multiplied by CBF.
Data analysis. Data are presented as means ± SE. Heart rate, pressures, and coronary velocity were measured from the strip chart recordings. CBF was calculated from the Doppler frequency shift. Myocardial free radical production was calculated as the arteriovenous spin adduct concentration [expressed in arbitarary units (au)] multiplied by CBF. Hemodynamic data were compared between the rest and exercise groups by ANOVA for repeated measures; a value of P < 0.05 was considered significant. PBN adduct concentrations were compared between aortic and coronary sinus concentrations during rest and exercise. When a significant result was found, individual comparisons were performed with the Wilcoxon signed-rank test. A finding of P < 0.05 was required for statistical significance.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
O2.
Resting MO2 was 4.4 ± 1.7 ml O2/min and increased to 8.5 ± 1.7 ml O2/min during exercise (P < 0.05). Oxygen extraction increased during exercise, and the coronary sinus PO2 decreased from 27 ± 5 to 19 ± 3 mmHg (P < 0.05). EPR signal in response to free radical generator. Figure 1 demonstrates the increase in EPR signal in response to an increase in free radical concentration generated by the reaction of H2O2 and ammonium iron (II)-sulfate in arterial blood containing the spin trap PBN compared with control.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Measurement of ROS in skeletal muscle. Mitochondria represent the single largest source of ROS production in the body. The mitochondrial electron transport chain is the major cellular source of O2–, largely originating from complex I and III (19) as a result of incomplete reduction of molecular oxygen. During normal mitochondrial respiration, it has been estimated that up to 2–5% of the electron flux may contribute to the formation of short-lived ROS, such as O2– and H2O2 (7, 12), although other studies (26) suggest much less.
Because exercise is associated with increases in oxygen flux through the mitochondrial electron transport chain, free radical production might be expected to increase. This hypothesis is indirectly supported by numerous studies demonstrating an increase in lipid peroxidation by-products (malondialdehyde) in mixed venous blood from skeletal muscle (8) or by an increase in exhaled pentane levels during exercise (10). However, these findings cannot provide information on the source or type(s) of radicals produced. Direct measurements of ROS production with EPR spectroscopy in exercising humans were first reported by Ashton et al. (2), using the spin trap PBN in conjunction with vacuum degassing of the adducts. They observed that the EPR signal in mixed venous blood after exhaustive exercise increased in proportion to O2max with hyperfine splitting constants (N = 1.37 mT, and H = 0.194 mT), consistent with a secondary lipid-derived carbon or oxygen-centered radical. Bailey et al. (3, 4) provided the first evidence of free radical outflow from an exercising muscle group in humans by using ex vivo spin trapping with PBN during single-leg exercise after cannulation of the femoral artery and vein. They demonstrated EPR spectra consistent with several ROS that are the by-product of free radical-mediated damage to membrane phospholipids that increased in proportion to leg blood flow and O2 during exercise.
Previous measurements of in vivo ROS production in hearts. Zweier et al. (29) designed a loop-gap resonator to measure free radical spectra in conjunction with EPR spectroscopy in the isolated beating rat heart using perfusate containing the radical 2,2,6,6-tetramethylpiperdine-1-oxyl (TEMPO). Although the method was capable of detecting radical concentrations as low as 0.4 µM, no EPR signal was detected from the heart before infusion of the radical. In humans with acute myocardial infarction, Grech et al. (15) demonstrated a free radical signal in coronary sinus blood using the spin trap PBN in conjunction with EPR spectroscopy after reperfusion with angioplasty. Similar findings were demonstrated by Tortolani et al. (28) using ex vivo spin trapping with PBN in patients after reperfusion from cardioplegic arrest at the time of coronary bypass surgery. In both studies, no PBN adduct was detected before reperfusion. Interestingly, the hyperfine splitting constants measured during reperfusion were consistent with a secondary carbon or oxygen-centered radical originating from primary radical attack on membrane phospholipids, similar to that observed in exercising skeletal muscle (2–4).
Bolli et al. (6) administered the spin trap PBN by the intracoronary route in regionally ischemic open-chest dogs. No EPR signal was detected in the coronary venous blood before ischemia, but a significant EPR signal was observed during reperfusion after 15 min of coronary occlusion. In a follow-up study (5), they demonstrated that the EPR signal was markedly reduced by administration of SOD and catalase, consistent with the original free radical species being derived from O2– and H2O2. In anesthetized dogs, Duilio et al. (11) reported that the neutrophil NADPH oxidase inhibitor VF244 decreased the PBN adduct signal during reperfusion after a prolonged coronary occlusion (90 min), implying a significant contribution from neutrophil-derived O2– generation during reperfusion.
These studies demonstrated that a robust EPR signal can be detected in coronary sinus blood during reperfusion after brief or prolonged ischemia in both anesthetized and awake subjects. However, no detectable EPR signal has previously been reported in coronary venous blood from normal awake animals during rest or exercise. The failure to detect an EPR signal could be the result of a lesser generation of free radicals in the heart during normal conditions or insufficiently sensitive EPR signal measurement. To improve signal detection, we employed the deoxygenation strategy reported by Ashton et al. (2) and Bailey et al. (3, 4), who vacuum degassed their samples before analysis for ROS in exercising humans. Because oxygen has paramagnetic qualities, it can interact with nitoxide spin traps, such as PBN, and result in broadening and blurring of the EPR signal (27). Thus, before the removal of oxygen from our samples, only a small signal was detected (Fig. 2A), which was similar to previous results in the coronary circulation of the dog before an ischemic insult, and likely explains why a measurable EPR signal was not detected in those studies.
The hyperfine splitting constants of nitrogen in our spectra (N = 1.37 mT) are in excellent agreement with those found by Bailey et al. (1.38 mT) in the femoral vein and by Ashton et al. (1.37 mT) in the antecubital vein during exercise (2–4) and in coronary venous blood after reperfusion of the ischemic heart (5, 6) using the same spin-trap and extraction solvent. However, the 
-hydrogen coupling constant in these studies was much more variable, being between 0.17 and 0.279 mT. Our value of H = 0.10 mT is lower than previously reported values for PBN in toluene. This may be related to the decreased signal resolution in our study due to the much smaller generation of free radicals in the normal heart compared with those produced by ischemia/reperfusion or could be the result of oxygen degassing over nitrogen instead of vacuum degassing. Our findings and those from previous studies (2–4) suggest that several radical species are present in the spectra. However, based on the hyperfine splitting constants, they are likely to be predominantly a carbon- and/or oxygen-centered alkoxyl radical that results from primary free radical attack on membrane phospholipids.
Source of ROS. The report that free radical production is increased during exercise in an isolated muscle group (3, 4) that is proportional to oxygen consumption suggests that the primary source of free radicals may originate from mitochondrial respiration. However, the ability of O2– derived from the mitochondria to reach the vascular space and be captured by the spin trap would be limited by its dismutation to H2O2 by MnSOD and CuZnSOD in the matrix and intermembrane space or by glutathione, which is present in large quantities in the mitochondria (23). Furthermore, unlike H2O2, the O2– anion diffuses poorly across cell membranes because of its charged structure, further limiting its ability to enter the vascular space. Although previous studies (7, 12) suggest that significant (2–5%) ROS production can occur in respiring mitochondria, studies (26) of isolated heart mitochondria respiring on palmitoyl carnitine reported that only 0.15% of the electron flux gave rise to H2O2. In electrically stimulated isolated skeletal myotubules, hydroxyl radical and nitric oxide (NO) production increased with stimulation frequency, but the increase in O2– production as assessed by cytochrome-c reduction was independent of the pacing rate (22). This suggests that mitochondria were not the primary source of the increase in O2– and that the extracellular source of O2– was derived from other sources such as plasma membrane oxidoreductases.
Our observation that no free radical production could be detected across the coronary circulation during exercise must be considered in light of previous reports in skeletal muscle. To study free radical production by skeletal muscle, Bailey et al. (4) used ex vivo spin trapping with PBN on blood from the femoral artery and vein of healthy men during incremental single leg/knee extensor exercise. Exercise caused an increase in the net outflow of PBN spin adducts, but the investigators noted an "uncoupling" between O2 consumption and radical outflow, with a more marked increase in radical outflow between the low to moderate exercise intensity than during the transition from moderate to heavy exercise; greatest radical outflow was recorded between 25 and 75% of the maximal work rate. Because the O2max for single leg/knee extensor exercise is 
50 ml·min–1·100 g–1 (24), the work rates corresponding to the greatest increase in PBN adduct (from 25% to 70% of maximal work) correspond to a muscle O2 consumption of between 12.5 and 35 ml·min–1·100 g–1. This rate of O2 uptake is within the range of MO2 observed during exercise in our study (24.3 ± 5.0 ml·min–1·100 g–1) where no radical production was detected. Although it is possible that free radical production by the myocardium might be detected during heavier levels of exercise, the present data indicate that submaximal exercise, which results in levels of O2 uptake that are associated with production of free radicals by skeletal muscle, do not result in detectable free radical production by the heart.
In skeletal muscle the increase in oxygen flux and the resultant increase in ROS production during exercise may overwhelm the endogenous antioxidant systems, resulting in a net outflow of ROS (3, 4). This is supported by measurements of antioxidant and lipid peroxidation levels in skeletal and cardiac muscle in rats after exercise (21), where the by-products of lipid peroxidation were much higher in skeletal muscle (23.9 vs. 1.7 µmol/g tissue), whereas levels of tissue SOD were significantly greater in the heart.
Several sources of O2– and H2O2 could contribute to ROS production in the coronary vessels. In addition to endothelial mitochondria and cytochrome P-450 enzymes, the activities of endothelial NO synthase (eNOS), NAD(P)H oxidase (9), and xanthine oxidase (XO) are all augmented by shear stress. In cultured bovine aortic endothelial cells, acute application of shear stress increased O2– production measured by EPR spectroscopy and O2– generation was inhibited by the XO inhibitor oxypurinol or the NAD(P)H oxidase inhibitor apocynin (20). However, chronic increases in endothelial shear stress that are associated with exercise training may reduce ROS production. In humans with coronary artery disease, those previously randomized to an exercise program before bypass surgery had a decrease in ROS production and NAD(P)H gene expression in arterial bypass grafts (1). In a separate group of patients, exercise training improved endothelial function and flow-mediated vasodilation (16) and is consistent with the observation that exercise increases NO production and eNOS gene expression (25) and decreases the NO-mediated destruction by O2– due to increases in vascular antioxidants, such as extracellular SOD (13).
Because resting blood flow is chronically higher in the myocardium compared with skeletal muscle, the relatively greater increase in blood flow that occurs during exercise in the peripheral compared with the coronary circulation may explain, in part, the greater EPR signal observed in skeletal versus cardiac muscle.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-phenyl N-tert-butyl nitrone. J Clin Invest 82: 476–485, 1988.[Web of Science][Medline]O2max in the exercise-trained human quadriceps. J Appl Physiol 86: 1048–1053, 1999.This article has been cited by other articles:
|
|
 |
|
  S. Lavi, J. P. McConnell, C. S. Rihal, A. Prasad, V. Mathew, L. O. Lerman, and A. Lerman Local Production of Lipoprotein-Associated Phospholipase A2 and Lysophosphatidylcholine in the Coronary Circulation: Association With Early Coronary Atherosclerosis and Endothelial Dysfunction in Humans Circulation, May 29, 2007; 115(21): 2715 - 2721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-i. Saitoh, C. Zhang, J. D. Tune, B. Potter, T. Kiyooka, P. A. Rogers, J. D. Knudson, G. M. Dick, A. Swafford, and W. M. Chilian Hydrogen Peroxide: A Feed-Forward Dilator That Couples Myocardial Metabolism to Coronary Blood Flow Arterioscler Thromb Vasc Biol, December 1, 2006; 26(12): 2614 - 2621. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
