Originally thought of as simply damaging or toxic “accidents” of in vivo chemistry, free radicals are becoming increasingly recognized as redox signaling molecules implicit in cellular homeostasis. Indeed, at the vascular level, it is plausible that oxidative stress plays a regulatory role in normal vascular function. Using electron paramagnetic resonance (EPR) spectroscopy, we sought to document the ability of an oral antioxidant cocktail (vitamins C, E, and α-lipoic acid) to reduce circulating free radicals, and we employed Doppler ultrasound to examine the consequence of an antioxidant-mediated reduction in oxidative stress on exercise-induced vasodilation. A total of 25 young (18–31 yr) healthy male subjects partook in these studies. EPR spectroscopy revealed a reduction in circulating free radicals following antioxidant administration at rest (∼98%) and as a consequence of exercise (∼85%). Plasma total antioxidant capacity and vitamin C both increased following the ingestion of the antioxidant cocktail, whereas vitamin E levels were not influenced by the ingestion of the antioxidants. Brachial artery vasodilation during submaximal forearm handgrip exercise was greater with the placebo (7.4 ± 1.8%) than with the antioxidant cocktail (2.3 ± 0.7%). These data document the efficacy of an oral antioxidant cocktail in reducing free radicals and suggest that, in a healthy state, the aggressive disruption of the delicate balance between pro- and antioxidant forces can negatively impact vascular function. These findings implicate an exercise-induced reliance upon pro-oxidant-stimulated vasodilation, thereby revealing an important and positive vascular role for free radicals.
- oxidative stress
- electron paramagnetic resonance spectroscopy
oxidative stress is ultimately determined by the balance between pro- and antioxidant forces. In low concentrations, free radicals are thought to act as mediators and modulators of cell signaling and contribute to other key functions such as regulating the activity of transcription factors and gene expression (18, 24). In contrast, high levels of free radicals in the vasculature have been associated with hypertension, atherosclerosis, diabetes, heart failure, sepsis, as well as the aging process (33). Since the initial observation that superoxide (O2−) and other free radicals inactivate nitric oxide (NO) (35), it has become increasingly apparent that this may contribute to the origin and progression of vascular dysfunction in many of these pathologies (17, 28). Such conclusions are supported by the improvement in endothelial function afforded by the infusion of high levels of exogenous antioxidants in these conditions (16, 23). However, such studies have generally relied on indirect markers of oxidative stress with somewhat limited sensitivity and specificity (16). With the wide ranging impact of oxidative stress in disease and disease progression, there is a clear need to better understand the role of free radicals and antioxidants in terms of healthy vascular function.
In humans, the high reactivity and low steady-state concentration of free radicals makes direct molecular detection a formidable analytical challenge (11). Consequently, investigators have typically relied on exercise-induced changes in biological footprints [e.g., lipid hydroperoxides (LH)] formed as a consequence of the interaction of free radicals with cellular components containing lipids and proteins. Electron paramagnetic resonance (EPR) spectroscopy is the only current technique capable of directly detecting free radicals (29), yet its application to the physiological environment has to date been limited due to the complicated nature of biological materials and environments (20). With this approach we have previously demonstrated the feasibility of applying EPR spectroscopy in humans and directly documented an exercise intensity-dependent free radical outflow from the skeletal muscle bed and gained insight into the regulation of this exercise-induced oxidative stress (2–4). In terms of the link between vascular function and free radicals, there are two equally plausible, but starkly contrasting, theories: As already noted free radicals have been proposed to limit vasodilation and blood flow by reducing NO bioavailability (16), but it is also possible that they promote vasodilation via their direct vasoactive properties (22). Somewhat surprisingly, the vascular consequences of an acutely ingested antioxidant cocktail, with a documented efficacy in terms of reducing free radical load, during exercise is unknown.
Therefore, with the mindset that free radicals are not simply damaging or toxic accidents of in vivo chemistry, we sought to determine the efficacy of an orally administered antioxidant cocktail (vitamin C, vitamin E, and α-lipoic acid) on free radical concentration and evaluate the antioxidants effects on exercise-induced brachial artery vasodilation in young healthy human subjects. Specifically, we tested the following two hypotheses: 1) this antioxidant cocktail will significantly attenuate the circulating free radical signal both at rest and as a consequence of exercise, as measured by spin trapping and ERP spectroscopy, and 2) in these normal healthy subjects a large antioxidant-induced reduction in free radicals will actually attenuate exercise-induced brachial artery vasodilation.
MATERIALS AND METHODS
Subjects and general procedures.
Twenty five young (25 ± 2 yr) healthy men participated in the current study. All subjects were normally active, nonsmokers, normotensive (<140/90 mmHg), and free of overt cardiovascular disease. Informed consent was obtained according to the University of California San Diego Human Subjects Protection Program requirements. Twelve subjects partook in the antioxidant cocktail efficacy, cycle exercise, and EPR spectroscopy studies, whereas 13 subjects took part in the antioxidant cocktail, handgrip exercise, and Doppler studies. Subjects always reported to the laboratory in the fasted state, which was maintained as a thermoneutral environment.
Venous blood samples from the antecubital fossa were evaluated at rest and within 5 s following the end of a graded maximal cycle exercise test (10–12 min duration). The exercise test was repeated on two separate days with and without antioxidant supplementation in a balanced design. Spin trapping and EPR spectroscopy were performed on these blood samples to establish the efficacy of the antioxidant cocktail in reducing free radical load.
Forearm handgrip exercise.
Initially, a single maximal voluntary handgrip contraction was determined for reference and then following sufficient rest, handgrip exercise was performed at three low-intensity workloads (3, 6, and 9 kg at 0.5 Hz) using the same hydraulic handgrip dynamometer (Rolyan Ability One, Germantown, WI). Subjects exercised for 3 min to ensure the attainment of steady-state hemodynamics before Doppler ultrasound measurements were made. The exercise test was repeated on two separate days with and without antioxidant supplementation in a balanced design. All data collection was performed following 30 min of rest in the supine position and with the arm at heart level.
All subjects received either an antioxidant cocktail or placebo supplementation in a randomized, double-blind balanced design. The formulation and timing of this antioxidant cocktail was the result of our pilot work employing vascular sampling and EPR spectral analysis to document its efficacy in reducing free radical concentration but restrained by the intent to not vastly exceed the common “over-the-counter” dosage for each individual antioxidant. Consequently, supplements were taken in two doses separated by 30 min, with the first dose ingested 2 h before the graded handgrip exercise protocol. The first dose consisted of 300 mg of α-lipoic acid, 500 mg vitamin C, and 200 IU vitamin E, whereas the second included 300 mg α-lipoic acid, 500 mg vitamin C, and 400 IU vitamin E (water dispersible). Placebo microcrystalline cellulose capsules were of similar taste, color, and appearance and were likewise consumed in two similarly timed doses.
EPR and spin-trapping.
EPR was performed on venous blood samples, as described previously. Briefly, 4.5 ml of venous blood were collected into a vacutainer that contained 1.5 ml of the spin trap α-phenyl-tert-butylnitrone (PBN) (0.140 mol/l). After centrifugation, the PBN adduct was extracted from the serum supernatant with toluene, and the adduct (200 μl) was pipetted into a precision-bore quartz EPR sample tube (Wilmad) that had been flushed with compressed N2. EPR was performed at 21°C with the use of an EMX X-band spectrometer (Bruker, MA) by using commercially available software (version 2.11, Bruker Win EPR System), with data processing blinded to experimental condition (achieved by coded numbering of samples).
Plasma antioxidant levels.
Total antioxidant capacity (TAC) was measured via enhanced chemiluminescence using a modification of the technique by Whitehead et al. (36). For vitamin C (ascorbic acid) measurements, plasma was stabilized and deproteinated by adding 900 μl of 5% metaphosphoric acid (Sigma Chemical, Dorset, UK) to 100 μl EDTA plasma. Vitamin C was subsequently assayed by fluorimetry based on the condensation of dehydroascorbic acid with 1,2-phenylenediamine (34). The plasma concentration of vitamin E (α-tocopherol) was determined by using an HPLC method (8, 30). Exercise samples were corrected for plasma volume shifts (with a general hemoconcentration noted) according to established techniques (12).
Ultrasound Doppler measurements and calculations.
The ultrasound system (Logiq 7, GE Medical Systems, Milwaukee, WI) was equipped with a linear array transducer operating at an imaging frequency of 10 MHz. Vessel diameter was determined at a perpendicular angle along the central axis of the scanned area, where the best spatial resolution was achieved. Anatomic landmarks and printed ultrasound images were utilized to ensure a similar site of measurement between scans and across experimental days. The brachial artery was insonated approximately midway between the antecubital and axillary regions, medial to the biceps brachii muscle. For each 20-s ultrasound Doppler segment, mean velocity (Vmean) was averaged across the first and last 10 s of the recorded clip. Diameter measurements were made at a perpendicular angle along the central axis of the scanned area, where the best spatial resolution was reproducibly achievable, during diastole, as described previously (13, 39). With the use of artery diameter and Vmean, blood flow in the brachial artery was calculated as:
Blood viscosity was not measured, so shear rate was calculated using the equation (7):
In this study, all ultrasound Doppler measurements and analyses were performed by a single sonographer who demonstrated equal or better ability in terms of reproducibility to previously published manual measurements of vessel diameter (38). Specifically, repeated analysis of a single brachial artery image for diameter across time, coefficient of variance equaled 9% (10 rereads); repeated analysis of a single subject's brachial artery diameter across time (i.e., probe replacement and verification of site with landmarks, etc.), coefficient of variance equaled 19%. All Doppler measurements and analyses were additionally performed in a blinded fashion such that the numerically coded identification of subjects and treatments was not available to the investigators until the completion of the study.
Statistics were performed with the use of commercially available software (SPSS). Data were analyzed using parametric statistics, following mathematical confirmation of normal distribution using a Shapiro-Wilk test. Repeated measure analysis of variance (RM ANOVA), analysis of variance (ANOVA), and Student's t-tests were used to identify significant changes in measured variables between trials, with the Tukey test used for post hoc analysis when a significant main effect was found. The sample size for each group (n = 12 and n = 13) was initially found to be sufficient via preliminary power analyses and confirmed by post hoc power analyses of the major comparisons. All group data are expressed as means ± SE. Significance was established at P < 0.05.
All of the 20 healthy young subjects completed the studies, none were taking any medication, and their general characteristics are presented in Table 1.
Antioxidant cocktail efficacy.
A clear and characteristic “triplet of doublets” EPR signal proportional to free radical concentration was detected at rest [6,264 ± 2,069 arbitrary units (au)] in the PBN spin-trapped venous blood of all 12 subjects, and this signal was largely ablated in resting venous samples after ingestion of the antioxidant cocktail (Fig. 1). Maximal cycling exercise increased the concentration of PBN adduct, resulting in a significantly greater signal magnitude (8,996 ± 2,323 au) (Fig. 1). The dominant signal exhibited hyperfine coupling constants of aN = 13.7 G and abH=1.9 G, consistent with published values for an O2-centered alkoxyl species (PBN-LO·) using similar extraction solvents (1). Consumption of the antioxidant cocktail before exercise markedly attenuated the free radical signal following exercise (1,309 ± 200 au) (Fig. 1). Plasma total antioxidant capacity (TAC) and vitamin C (ascorbic acid) both increased with acute exercise and following the ingestion of the antioxidant cocktail. Vitamin E (α-tocopherol) revealed only a small, but significant, increase from rest to exercise with no change as a result of ingesting the antioxidant cocktail (Table 2).
Exercise-induced brachial artery vasodilation, blood flow, and shear rate.
All 13 subjects were successfully imaged with the ultrasound system at rest and during the three levels of handgrip exercise (Fig. 2). The ingestion of the oral antioxidant did not significantly alter either resting brachial artery diameter (placebo: 0.489 ± 0.014; antioxidant: 0.495 ± 0.013 cm) or blood flow (placebo: 80 ± 34; antioxidant: 108 ± 30 ml/min) (Figs. 3 and 4). The exercise-induced brachial artery vasodilation increased with increasing handgrip workload (Fig. 3). Antioxidant administration markedly attenuated the exercise-induced brachial artery vasodilation when compared with the placebo trial (Fig. 3). Calculated shear rate also increased with increasing handgrip workload (Fig. 4). However, although shear rates were not significantly different between the placebo and antioxidant trials, there was a clear tendency for shear rate to be increased in the antioxidant trials (P = 0.1) (Fig. 3).
The recent recognition that free radicals are redox signaling molecules implicit in the regulation of cellular O2 homeostasis and not merely damaging accidents of chemistry justifies continued interest into their mechanisms of generation and actions in vivo. In this context, the current investigation has provided two unique and important findings. First, we have directly documented the efficacy of an oral antioxidant cocktail (vitamin C, vitamin E, and α-lipoic acid) in terms of reducing blood-borne free radicals, both at rest and when oxidative stress is elevated due to acute exercise. Second, from this platform of knowledge we have demonstrated that this antioxidant-induced reduction in free radicals negatively impacts exercise-induced brachial artery vasodilation, supporting the concept that oxidative stress actually plays an important role in the normal healthy vascular response to exercise.
Pro- and antioxidant forces.
The delicate balance between pro- and antioxidant forces and the subsequent positive versus negative effects of free radicals is likely a crucial aspect of life (14). Low concentrations of free radicals appear to have both important mediating and modulating roles in cell signaling (14, 18, 24); however, without restraint higher levels of free radicals such as O2− can cause a wide spectrum of cellular damage, including lipid peroxidation, alteration of intracellular redox state, inactivation of enzymes, and damage to DNA. Consequently, there are numerous endogenous antioxidants that act as a defense system against oxidative stress. These antioxidants are generally divided into the nonenzymatic antioxidants (e.g., antioxidant vitamins) and the enzymatic antioxidants (e.g., SOD). Under normal circumstances the endogenous array of antioxidants combine to minimize, but, as evidenced in the current study by EPR spectroscopy, there is some normal background level of in vivo oxidative stress which is magnified by exercise (Fig. 1). However, the acute ingestion of an antioxidant cocktail effectively ablated the blood-borne alkoxyl radicals at rest and severely attenuated them following exercise (Fig. 1). It should be noted that in light of the instability of O2-centered alkoxyl species (15), any primary intramuscular alkoxyl radicals formed during the exercise challenge would likely have reacted with other molecules in vivo before exposure to the spin trap. However, these EPR data certainly are indicative of the more “classical,” although indirect, markers of peroxidation and provide clear evidence for peroxidative stress and its reduction with the antioxidant cocktail. The direct assessment of increased plasma total antioxidant capacity and vitiamin C further validate this oral approach of tipping the balance toward pro-oxidant forces (the lack of a treatment effect with vitamin E was likely due to limited absorption time provided) (Table 2). With such an aggressive and clearly documented reduction in free radical concentration, the stage is set to determine the role of oxidative stress in the vascular response to exercise.
Exercise, oxidative stress, and antioxidants.
Within all mammalian tissue oxidative phophorylation and the formation of ATP is accompanied by the univalent reduction of O2 to produce O2−. Early in vitro studies estimated that mitochondrial electron “leak” accounts for 2% of the total oxygen consumed during state 4 (basal) respiration (9), although more recent evidence suggests that this is probably an overestimation, may actually be in the 0.15–0.8% range (19, 27). Although the magnitude of this “electron leakage” is still under debate, most agree that is significant and is thought to occur at the NADH dehydrogenase (32) and the ubiquinone cytochrome-bc segment of complex III (25) in the mitochondrial electron transport chain. As O2 flux through the skeletal muscle mitochondria increases considerably during exercise, muscle itself has been commonly implicated as the main site of free radical production during elevated muscle metabolism (e.g., exercise). However, free radicals are also generated by various oxidases that are not directly linked to the muscle mitochondria [e.g., xanthine oxidase, NAD(P)H oxidases] (37) but are tightly coupled with exercise and the concomitant increase blood flow-induced shear stress. Additionally, recent findings have been suggesting that in human coronary resistance vessels, flow-induced dilation requires the production of H2O2, which is formed from O2− generated by mitochondrial respiration. Thus nonmuscle mitochondrial respiration and oxidative stress appears necessary for flow-mediated vasodilation to occur (22). The degree to which each of these potential sources of free radicals contribute to the elevated oxidative stress associated with exercise is not yet well understood.
What is well accepted is that the rapid changes in redox state of skeletal muscle associated with exercise offers a unique scenario that affords the study of the mechanisms and impact of elevated free radical generation. Whereas some direct measurements in animals and the use of putative manifestations of oxidative stress have established much about the species, source, and compartmentalization of free radicals during muscle contraction (10, 21, 26), direct evidence of increased rates of free radical production during exercise in humans is still scarce (3, 14). In terms of the link between oxidative stress and vascular function, it is well accepted that free radicals likely limit vasodilation and blood flow by reducing NO bioavailability (16); however, it is also plausible that some free radicals and their end products increase vasodilation via their direct vasoactive properties (22). Clearly, the current data support the latter concept, because the large antioxidant-induced reduction in free radical concentration (Fig. 1) resulted in a diminished vasodilatory response to exercise in the brachial artery (Fig. 3).
Despite the successful amelioration of the age-related endothelial dysfunction with high plasma levels of ascorbic acid attained by infusion, to date orally administered vitamin C appears to have little impact (16). This observation may be a consequence of the method employed to assess vascular function [flow-mediated vasodilation (FMD) (16)] and the potentially limited perturbation of the balance between pro- and antioxidant forces in the conduit vessel of interest in the FMD model. In contrast, the present study implemented an acute exercise model, which provides the opportunity to study conduit arterial function in a state of elevated oxidative stress suggested to be the consequence not only of elevated skeletal muscle metabolic activity, but also by a prolonged elevation in vascular shear stress and free radical outflow into the vascular plasma (4, 10, 21). To our knowledge the current data are the first to directly reveal that “unbalancing” the equation through a documented reduction in free radicals (and an increase in plasma antioxidants) in normal healthy subjects during exercise negatively impacts exercise-induced vasodilation (Table 2 and Fig. 3). These data reveal a potentially important role for oxidative stress in provoking an appropriate vasodilation during exercise.
Mechanism for the attenuated exercise-induced vasodilation.
Certainly, a powerful NO-mediated vasodilatory force is shear stress, which if changed by either blood viscosity, blood velocity, or vessel diameter, could impact vasodilation. However, during the handgrip exercise performed in this study, our surrogate indicator of shear stress (shear rate) (7) did not change significantly with and without antioxidant use, and thus shear rate could not explain the difference in exercise-induced brachial artery vasodilation (Fig. 4). Indeed, although the antioxidant administration did not significantly alter shear rate, there was a clear tendency for shear rate to be elevated at each level of work examined (explained by a tendency for greater blood flow and less vasodilation during exercise in the antioxidant trials) (Figs. 3 and 4). These observations add even greater credence to the concept that the change in shear rate was a consequence rather than a cause of the attenuated exercise-induced vasodilation in the antioxidant trials. Though we readily accept that shear forces are but one of the multiple and complex possible stimuli for vasodilation in a conduit vessel during exercise (see Experimental considerations), shear is readily quantified with the current experimental design and appears not to be the mechanism responsible for attenuated vasodilation following antioxidant administration.
We speculate that this attenuated exercise-induced vasodilation is likely the direct result of disturbing the natural balance between pro- and antioxidant forces. As already indicated, oxidative stress is typically regarded as an unwanted byproduct of cellular oxidation and is seen as a negative risk factor for cellular and vascular health (5). However, the downstream consequences of free radicals such as H2O2 and ONOO−, perhaps released in response to increased shear stress, may act as potent vasodilators (6), and as such possess the capacity to alter vascular responsiveness (22). Thus it is likely that the large and clearly documented reduction in free radical concentration following antioxidant administration (Fig. 1) may have removed oxidative species, which possess some beneficial vasoactive properties. This possible vasodilatory role of free radicals may explain the severely attenuated exercise-induced brachial artery following antioxidant ingestion.
The exercise model employed in these studies of arterial vasodilation has not been proven to differentiate the effects of endothelial function from metabolic vasodilation. However, even in FMD studies, a well-accepted technique to evaluate endotheliium-dependent vasodilation, there is still a potential for retrograde conducted vasodilation, myogenic responses, and changes in sympathetic activity, all of which may modify conduit artery vasodilation (31). Importantly, the exercise-induced experimental model of vasodilation emulates the physiological conditions present during activities of daily living. Also, unlike the “single shot” FMD assessment of vascular function, this graded exercise approach evaluates multiple levels of stimuli that, based on our current and previous results (13), appears to be highly linear (Fig. 3), increasing confidence in the data and interpretations of change.
An additional experimental consideration is that due to technical difficulties, the repeat measurements of free radicals during the handgrip exercise were not acquired, limiting our direct knowledge of redox status during this exact protocol with and without the antioxidant cocktail. However, our validation of the effectiveness of the antioxidant cocktail during a much more metabolically demanding scenario (cycle exercise), demonstrates that at rest and immediately following high-intensity exercise, it was highly effective at ablating or greatly attenuating (Fig. 1) the measured blood-borne alkoxyl and alkyl radicals.
In conclusion, this study demonstrates that the brachial artery of healthy subjects vasodilates less following the ingestion of an antioxidant cocktail during submaximal exercise, despite similar or slightly elevated shear rates. This suggests that an antioxidant-induced reduction in oxidative stress, as supported by blood EPR measurements, negatively impacts the natural balance between pro- and antioxidant forces that exists in normal healthy subjects. This reveals a reliance on free radical-mediated vasodilation, most likely induced by the increased oxidative stress of exercise per se (increased metabolic rate) or the shear-induced release of free radicals by the vascular endothelium.
This research was supported by National Heart, Lung, and Blood Institute Grant HL-17731, Tobacco Related Disease Research Program Grant 15RT-0100, and The Stein Institute for Aging Research.
The authors thank the subjects for their time in volunteering for this study.
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.
- Copyright © 2007 by the American Physiological Society