点击显示 收起
1 From the Center for Human Nutrition (DJO, SD, SMG, and IJ) and the Division of Clinical Biochemistry and Human Metabolism, Department of Pathology and Internal Medicine (SD and IJ), University of Texas Southwestern Medical Center, Dallas.
2 Supported in part by grants from Welch Foods Inc (Concord, MA) and the National Institutes of Health (K24 AT00596). Welch Foods Inc supplied the Concord grape juice. 3 Address reprint requests to I Jialal, Laboratory for Atherosclerosis and Metabolic Research, University of California, Davis, Medical Center, 4635 Second Avenue, Room 3000, Sacramento, CA 95817. E-mail: ishwarlal.jialal{at}ucdmc.ucdavis.edu.
ABSTRACT
Background: Concord grape juice (CGJ) is a rich source of flavonoids, which have greater antioxidant efficacy in vitro than does -tocopherol; however, the efficacies of flavonoids and -tocopherol in vivo have not been compared.
Objective: We compared the in vivo antioxidant efficacy of CGJ with that of -tocopherol in healthy adults.
Design: Subjects were randomly assigned to receive either 400 IU RRR--tocopherol/d (n = 17) or 10 mL CGJ · kg-1 · d-1 (n = 15) for 2 wk. Serum oxygen radical absorbance capacity, plasma protein carbonyls, urinary F2-isoprostanes, and resistance of LDL to ex vivo oxidation were measured before and after supplementation as markers of antioxidant status and oxidative stress.
Results: After supplementation, plasma -tocopherol increased 92% in subjects who received -tocopherol (P < 0.001); plasma total and conjugated phenols increased 17% (P < 0.01) and 22% (P < 0.001), respectively, in subjects who received CGJ. There was a significant change in plasma triacylglycerols in both groups, but the concentrations were within the normal range. CGJ supplementation was associated with significantly higher triacylglycerols than was -tocopherol supplementation. Both supplementation regimens significantly increased serum oxygen radical absorbance capacity (P < 0.001) and LDL lag time (P < 0.001) and significantly decreased the LDL oxidation rate (P < 0.01), with no significant difference in effectiveness. Protein carbonyl concentrations in native plasma decreased 20% after CGJ supplementation, which was a significantly different response than that after -tocopherol supplementation (P < 0.05).
Conclusions: In healthy adults, 10 mL CGJ · kg-1 · d-1 increased serum antioxidant capacity and protected LDL against oxidation to an extent similar to that obtained with 400 IU -tocopherol/d but decreased native plasma protein oxidation significantly more than did -tocopherol. CGJ flavonoids are potent antioxidants that may protect against oxidative stress and reduce the risk of free radical damage and chronic diseases.
Key Words: Antioxidant • -tocopherol • grape juice • flavonoids • LDL oxidation • oxygen radical absorbance capacity • protein carbonyls • urinary F2-isoprostanes
INTRODUCTION
Flavonoids constitute a large group of polyphenolic compounds that are found in fruit, vegetables, chocolate, coffee, tea, wine, and grape juice (1,2). Consumption of foods rich in flavonoids is associated with a reduced risk of various chronic diseases (3–7). The protective benefits of dietary flavonoids may be due in part to their antioxidant properties and ability to reduce oxidative stress (2). In vitro studies showed significant antioxidant activity for specific dietary flavonoids and some of the major metabolites and conjugated derivatives that occur in the circulation after consumption of dietary flavonoids (2,8–12). In some experiments the antioxidant capacity of flavonoids was shown to exceed that of Trolox (Sigma Chemical Co, St Louis) or -tocopherol (2). Because of the diverse chemical structures of flavonoids and their metabolites, they can have hydrophilic or relatively lipophilic properties and may interact with plasma proteins as well as the polar surface region of phospholipid bilayers in lipoproteins and cell membranes (2,13,14). Because of the nature of these interactions, flavonoids may have the ability to protect against free radical attack in both aqueous and lipid environments, thus providing an effective antioxidant defense in biological systems.
Grape juice is a rich source of the antioxidant flavonoids catechin, epicatechin, quercetin, and anthocyanins (2). In vitro studies showed that grape juice has significant antioxidant activity and can inhibit oxidation of LDL (15–17). Human studies showed promising results but were limited by either the short duration of supplementation, the confounding effects of medications or other antioxidants, or the measurement of only a few indexes of antioxidant status (18–21). To date, no study has investigated the effects of chronic grape juice consumption on antioxidant status and markers of oxidative damage to lipids and proteins. Because evidence from in vitro experiments (2,9,10) suggested that grape juice flavonoids provide more potent and diverse in vivo antioxidant protection than does the lipophilic antioxidant -tocopherol, we were interested in comparing their antioxidant efficacies. In previous research, we found that supplementation with 400 IU -tocopherol/d is sufficient to decrease LDL oxidation (22–24) and urinary F2-isoprostanes (24), which are valid markers of lipid peroxidation (25). Thus, it was the aim of the present study to characterize the antioxidant efficacy of grape juice by comparing it with that of -tocopherol, which is a potent antioxidant that has been shown to have significant effects in humans after supplementation. In the present study, we supplemented healthy volunteers with either 10 mL 100% Concord grape juice (CGJ) · kg-1 · d-1 or 400 IU -tocopherol/d for 2 wk and measured the effects on serum oxygen radical absorbance capacity (ORAC), plasma protein carbonyls, urinary F2-isoprostanes, and ex vivo LDL oxidation.
SUBJECTS AND METHODS
Study design
The study design was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center at Dallas. After being recruited from the campus of the University of Texas Southwestern Medical Center, healthy subjects provided informed consent to participate in the study. Subjects who met the study criteria were randomly assigned to receive either 400 IU RRR--tocopherol (natural)/d or 10 mL 100% CGJ · kg-1 · d-1 (Welch Foods Inc, Concord, MA) by using an adaptive randomization program that matched subjects on the basis of age, sex, body mass index, and LDL cholesterol to provide a balanced distribution. To standardize their diets, subjects began a flavonoid-restricted diet 2 wk before initiation of the study. After stabilization with the diet, subjects were supplemented with -tocopherol or CGJ for 2 wk while continuing to follow the flavonoid-restricted diet. A fasting blood sample was obtained during the screening process (baseline), after 2 wk of the flavonoid-restricted diet (before supplementation), and after 2 wk of dietary restrictions plus supplementation (after supplementation). Early morning urine samples were obtained after 2 wk of the flavonoid-restricted diet and after 2 wk of supplementation. Subjects completed dietary questionnaires during all 3 phases of the study. A 7-d diet record was obtained from subjects during the supplementation phase of the study. Subjects were given a known quantity of -tocopherol supplements or CGJ. The remaining -tocopherol pills and CGJ were collected at the end of the study to determine the subjects’ compliance with the regimen. The indexes measured were plasma concentrations of phenols, -tocopherol, ascorbic acid, lipids, and protein carbonyls; serum ORAC; oxidative resistance of LDL; and urinary F2-isoprostanes.
Subjects
Thirty-six healthy adults participated in this study. Subjects were excluded if they smoked; took antioxidant or fish oil supplements; consumed > 3 servings of fruit and vegetables/d; consumed red wine or purple grape juice regularly during the 3 mo before initiation of the study; had a history of diabetes, hypertension, heart disease, or endocrine disorders; had an abnormal blood chemistry profile, fasting LDL-cholesterol concentration > 3.37 mmol/L (> 130 mg/dL), or fasting triacylglycerol concentration > 3.39 mmol/L (> 300 mg/dL) at the time of the screening; or were taking hypolipidemic medications or receiving anticoagulant or thyroxin therapy.
Diet
To minimize the potential confounding effects of consuming fluctuating amounts of foods and beverages that are high in dietary flavonoids, all subjects followed a flavonoid-restricted diet during the entire study, a protocol similar to that of other studies (20,26,27). The subjects received instructions on the diet from a registered dietitian and were given printed guidelines on foods that were to be avoided. To standardize the flavonoid content of the diet, the dietary restrictions included avoidance of alcohol, tea, grape products, fruit juice, citrus, berries, onions, apples, and broccoli. These specific foods were categorized as containing medium (0.040–0.099 mg/g) or high (0.100 mg/g) amounts of flavonoids (28) and are commonly consumed in quantities that might cause fluctuations in plasma flavonoid concentrations. Coffee and chocolate were not eliminated from the diet, but subjects were advised to maintain a consistent and moderate diet throughout the study. Subjects were instructed to continue to eat their normal diet with the exception of the previously mentioned food restrictions. To monitor compliance with the dietary restrictions, a dietary food-frequency questionnaire was designed to include commonly eaten foods that are considered to be high in flavonoids (28). The flavonoid content of each subject’s diet was calculated from the completed questionnaires by using values for the total phenolic content of foods and beverages obtained from published literature (15,29–32). The diet records were used for additional monitoring of dietary compliance. A post hoc analysis of the diet records was performed to evaluate whether flavonoid intake or consumption of specific flavonoid-rich foods varied between groups.
Supplementation
Subjects assigned to receive -tocopherol were instructed to take one capsule of 400 IU RRR--tocopherol each day with a meal or snack. The capsules were given in an amber-colored container, and the subjects were instructed to store the sealed container away from direct sunlight. This dose of -tocopherol was chosen because it was shown to be the minimum dose that significantly decreases LDL oxidative susceptibility (22–24). Compliance with -tocopherol supplementation was determined by pill count and plasma -tocopherol concentration.
The subjects who were assigned to receive 10 mL CGJ · kg-1 · d-1 (Welch’s 100% with no added vitamin C) received a supply of juice and a container that was marked to indicate the volume of grape juice that the subject was to drink each day. Subjects were instructed to pour their daily ration of grape juice into the container each morning and to store the grape juice in the refrigerator. Because grape juice has an osmolality of 1000, it was anticipated that diarrhea might occur if grape juice was consumed in large quantities in a short period of time without accompanying food. Therefore, subjects were advised to drink CGJ throughout the day with meals and snacks to avoid possible diarrhea. The flavonoid content of the CGJ was determined to be 560 mg phenolic equivalents/L by using the method of Serifini et al (33). Previous studies showed that 7.0–7.7 mL CGJ · kg-1 · d-1 had significant antioxidant activity for some variables (20,21); therefore, we selected the dose of 10 mL CGJ · kg-1 · d-1 to provide a slightly higher dose of CGJ that was still reasonable for the participants to drink over the course of a day. Compliance with CGJ supplementation was determined by the volume of CGJ returned at the completion of the study and by plasma total and conjugated phenol concentrations.
Blood and urine collection and storage
All blood samples were collected by venipuncture after the subjects had fasted 10–12 h. During the screening process, blood samples were collected for determination of serum glucose, total protein, and liver and renal function profiles. Blood was collected in tubes containing 1 mg EDTA/mL for determination of complete blood cell count, plasma ascorbic acid, and plasma lipids. Serum and plasma were separated from blood cells by low-speed centrifugation (800 x g for 15 min) at 4 °C. Before supplementation and after 2 wk of supplementation, blood samples were collected and dispensed immediately into evacuated serum separator tubes and tubes containing 1 mg EDTA/mL that were stored on ice and then centrifuged. Freshly separated plasma was used for determination of the lipoprotein profile and plasma ascorbic acid concentrations. The remaining aliquots of plasma were purged with nitrogen and stored at -70 °C for measurement of -tocopherol, phenols, total protein, and protein carbonyls at the conclusion of the study. Additional aliquots of plasma were purged with nitrogen and stored at -20 °C until isolation of LDL and oxidation studies could be performed at the completion of the study. Serum samples were frozen at -70 °C, and the ORAC assay was performed at the completion of the study. All samples were numbered consecutively so that sample identity and treatment assignment were not apparent. Furthermore, the laboratory staff were blinded to the treatment assignments.
After the subjects had fasted 10–12 h, urine samples were collected in the early morning both before and after the 2-wk supplementation. The urine was refrigerated and centrifuged, and the supernatant fluid was dispensed, flushed with nitrogen, and stored at -70 °C so that urinary creatinine and F2-isoprostanes could be measured at the conclusion of the study.
Analytic procedures
Plasma lipid and lipoprotein concentrations were assayed enzymatically by using the automated Paramax system (Dade, Deerfield, IL) from Parkland Hospital’s (Dallas) Division of Clinical Biochemistry. The laboratory participated in the Alert Proficiency program and has a CV of < 3% for total cholesterol. Plasma total protein; serum glucose, creatinine, and alanine transaminase; complete blood cell count with differential; and urinary creatinine were determined with standard clinical chemistry laboratory techniques. Plasma -tocopherol was measured by reversed-phase HPLC as described previously (34). Plasma ascorbic acid was measured spectrophotometrically (35).
Plasma total and conjugated phenols were measured by using the method described by Serifini et al (33). Briefly, for total phenols, 1 mL of a 1.0-mol HCl/L solution was added to 500 µL plasma to hydrolyze conjugated phenols. Samples were mixed by vortex and incubated at 37 °C for 30 min. Complexed phenols were extracted with 1.0 mL of a 2.0-mol NaOH/L solution in 75% (by vol) methanol. Proteins were precipitated by using 0.75 mol metaphosphoric acid/L and were reextracted with acetone:water (1:1, by vol). The phenolic content was measured by using the Folin-Ciocalteau method with phenol as the standard. Free phenols were measured by using this same method after precipitation of protein with 0.75 mol metaphosphoric acid/L. The concentration of conjugated phenols was assessed by subtracting the concentration of free phenols from that of total phenols. Serifini et al (33) validated this assay in plasma by using a recovery test in which increasing amounts of quercetin were added to plasma; they found that 92 ± 7% of the quercetin was recovered (n = 12). Additionally, this assay was used to show that acute consumption of 113 mL alcohol-free red wine significantly increased plasma total phenol concentrations in healthy subjects (33).
The antioxidant potential of serum was determined by the ORAC assay on the basis of the procedure described by Cao and Prior (36). The procedure uses ß-phycoerythrin as an indicator protein and 2,2'-azobis (2-amidinopropane) dihydrochloride (AAPH) as a peroxyl radical generator. Under appropriate conditions, the loss of phycoerythrin fluorescence in the presence of reactive species is an index of oxidative damage. The inhibition by an antioxidant, which is reflected in protection against the loss of phycoerythrin fluorescence in the ORAC assay, is a measure of its antioxidant capacity.
Plasma protein carbonyls were measured by enzyme-linked immunosorbent assay before and after a 4-h incubation at 37 °C with 100 mmol/L of the aqueous free radical initiator AAPH (37,38). Oxidation was arrested by the addition of 40 µmol butylated hydroxytoluene/L, which was followed by refrigeration (37).
Frozen samples of urine that was originally collected early in the morning were used for measurement of urinary F2-isoprostanes by column extraction followed by an enzyme-linked immunosorbent assay according to a modified version of the method described by Morrow et al (39) and Wang et al (40). Briefly, 1 mL urine was acidified, loaded onto a C18 column, and eluted with ethyl acetate containing 1% (by vol) methanol. Sodium sulfate was then added to the eluate, which was loaded onto a silica column, eluted with ethyl acetate:methanol (1:1, by vol), and dried with nitrogen; the volume of the dried sample was brought to 1 mL with enzyme immunoassay buffer. A 1:8 dilution was made with the enzyme immunoassay buffer, and an enzyme-linked immunosorbent assay was performed with the use of reagents from Cayman Laboratories (Ann Arbor, MI). The results obtained were standardized to milligrams urinary creatinine (24). We previously validated this method against a method in which gas chromatography–mass spectrometry was used (41).
LDL oxidation was measured by monitoring the formation of conjugated dienes after in vitro Cu2+-catalyzed oxidation. Briefly, LDL was isolated with rapid two-step ultracentrifugation by using a modified version of the method described by Kleinveld et al (42). The samples were used for oxidation experiments within 24 h of isolation. Salts and EDTA were removed by passage of the LDL samples through a 10-mL Sephadex G-10 column (Amersham Pharmacia Biotech, Inc, Piscataway, NJ). The protein concentration of freshly isolated LDL was determined by using the method of Lowry et al (43). Samples were diluted to 100 µg LDL protein/mL with phosphate-buffered saline (pH 7.4) and incubated at 37 °C with 2.5 µmol Cu2+/L for 5 h. Conjugated diene formation was measured by continuous spectrophotometric monitoring at 234 nm (44). Measurements were recorded every 10 min and were used to assess LDL oxidation. Conjugated dienes were determined by using the molar extinction coefficient 2.95 x 104 mol · L-1 · cm-1 and were expressed relative to the amount of LDL protein. The LDL oxidation lag time (amount of time before the onset of conjugated diene propagation) and oxidative rate (rate of conjugated diene production) were calculated and used to characterize the oxidative resistance of LDL in each sample. To minimize variability, each subject’s LDL samples from before and after the supplementation were isolated and oxidized during the same laboratory procedure.
Statistical analysis
Results are expressed as means ± SDs or, when appropriate, as medians with 10th and 90th percentiles. Descriptive statistics were computed for baseline data. Data at baseline and before supplementation were analyzed with 2-sample t tests or Wilcoxon rank-sum tests to assess differences between groups before supplementation. Data before and after supplementation were analyzed by using two-way repeated-measures analysis of variance (ANOVA) models. The main effects assessed were group (-tocopherol and CGJ) and time (before or after supplementation) and the interaction between group and time. Rank or log transformations were performed on some variables before analysis to meet parametric analysis assumptions. When significant interactions were observed, pairwise comparisons were made within groups by using the least-square means estimates from the ANOVA models. These statistical analyses were performed by using SAS version 8.2 (SAS Institute Inc, Cary, NC). Student’s t test was performed on dietary data by using the EXCEL 7.0 statistical analysis program (Microsoft, Seattle). Significance was set at P = 0.05 for ANOVA and at P = 0.025 for pairwise comparisons to control for type I error.
RESULTS
Thirty-two of the original 36 participants successfully completed the study (-tocopherol group, n = 17 of 18; CGJ group, n = 15 of 18). One participant who was assigned to receive -tocopherol dropped out of the study because of time constraints, and 3 subjects who were assigned to receive CGJ dropped out of the study or were excluded from the study because they were not able to comply with supplementation. Overall, the experimental diet and supplements were well tolerated, and no side effects were reported except for 2 subjects in the CGJ group: one subject reported diarrhea and the other subject reported increased bruising during the study. These 2 subjects consumed > 750 mL CGJ/d, and the subject who experienced diarrhea frequently consumed large amounts of CGJ in a short period of time.
The characteristics of the subjects at baseline are shown in Table 1. There were no significant differences between the groups with regard to age, body mass index, or lipid profile.
View this table:
TABLE 1 . Baseline descriptive characteristics of subjects assigned to receive either 400 IU -tocopherol/d or 10 mL Concord grape juice·kg-1·d-11
Dietary phenolic consumption during the study is shown in Table 2. There was no significant difference in the total phenolic content of the diets consumed by the subjects in the -tocopherol and CGJ groups while the subjects followed the flavonoid-restricted diet at baseline and before supplementation. Only the subjects who received CGJ significantly increased their consumption of dietary phenols during the supplementation phase (P < 0.001). The results of this dietary analysis were based on the phenolic content of foods and beverages of fruit, vegetable, or tea origin and did not take into account foods or beverages containing coffee or chocolate, which also contribute flavonoids to the diet. A post hoc analysis was performed on the diet records that the participants kept during the supplementation phase of the study to determine whether the groups consumed significantly different amounts of coffee or chocolate-containing foods and beverages during this phase of the study. The results of this analysis did not show any significant group differences in mean daily servings of coffee (-tocopherol group: 0.4 ± 0.6; CGJ group: 0.3 ± 0.6; P = 0.30), chocolate beverages (-tocopherol group: 0.09 ± 0.2; CGJ group: 0.04 ± 0.1; P = 0.18), chocolate bakery products (-tocopherol group: 0.3 ± 0.7; CGJ group: 0.3 ± 0.4; P = 0.49), chocolate dairy desserts (-tocopherol group: 0.1 ± 0.2; CGJ group: 0.2 ± 0.3; P = 0.30), chocolate sauces (-tocopherol group: 0.04 ± 0.09; CGJ group: 0.02 ± 0.05; P = 0.19), or chocolate candies (-tocopherol group: 0.1 ± 0.1; CGJ group: 0.2 ± 0.4; P = 0.12).
View this table:
TABLE 2 . Dietary phenolic intake of subjects assigned to receive either 400 IU -tocopherol/d or 10 mL Concord grape juice·kg-1·d-11
In response to CGJ supplementation, mean plasma concentrations of total and conjugated phenols increased significantly (17% and 22%, respectively; Table 3). Mean plasma -tocopherol concentrations increased significantly (92%) after -tocopherol supplementation but did not change significantly after CGJ supplementation (Table 4). Mean plasma ascorbic acid concentrations increased significantly (29%) after -tocopherol supplementation but were not significantly different after CGJ supplementation (Table 4).
View this table:
TABLE 3 . Plasma phenol concentrations in subjects receiving 10 mL Concord grape juice·kg-1·d-11
View this table:
TABLE 4 . Plasma -tocopherol and ascorbic acid concentrations in subjects assigned to receive either 400 IU -tocopherol/d or 10 mL Concord grape juice·kg-1·d-11
Plasma lipid concentrations before supplementation were not significantly different between the -tocopherol and CGJ groups, but after supplementation both groups experienced minor increases in plasma lipids (Table 5). Although the results of ANOVA showed an increase in plasma total, LDL-, and VLDL-cholesterol concentrations in both groups (time interaction: P < 0.05), there was no significant group x time interaction. A significant group x time interaction was detected only for the plasma triacyglycerol data. Plasma triacylglycerol concentrations were significantly greater after both supplementation regimens (-tocopherol: P = 0.027; CGJ: P < 0.001). Furthermore, CGJ supplementation increased plasma triacylglycerol concentrations significantly more than did -tocopherol supplementation (P = 0.017).
View this table:
TABLE 5 . Plasma lipid profiles of subjects assigned to receive either 400 IU -tocopherol/d or 10 mL Concord grape juice·kg-1·d-11
The antioxidant properties of -tocopherol and CGJ were compared by assessing the effects on serum ORAC, plasma protein carbonyls, urinary F2-isoprostanes, and LDL oxidative resistance (Table 6). There were no significant differences between the groups before supplementation. Supplementation with either -tocopherol or CGJ significantly increased plasma ORAC values, and there was no apparent difference in response between the 2 groups. There were no significant changes in plasma concentrations of protein carbonyls either in the native state or in AAPH-oxidized plasma after -tocopherol supplementation. However, there was a 20% decrease in the concentration of protein carbonyls in native plasma in the subjects who received CGJ, resulting in a significant group x time interaction. Although the main effect of treatment on the change in concentration over time was not significant (P = 0.098), there was a significant difference between the groups in their responses to supplementation (P = 0.048). CGJ supplementation did not protect against AAPH-induced plasma protein oxidation. Also, neither -tocopherol nor CGJ supplementation altered endogenous urinary F2-isoprostane concentrations. However, both -tocopherol and CGJ supplementation significantly increased LDL lag time and significantly decreased the rate of LDL oxidation, as measured by the formation of conjugated dienes during ex vivo LDL oxidation. There were no significant differences in LDL lag time or oxidation rate between the 2 groups.
View this table:
TABLE 6 . Antioxidant and oxidative indexes of subjects assigned to receive either 400 IU -tocopherol/d or 10 mL Concord grape juice·kg-1·d-11
DISCUSSION
Grape juice flavonoids are potent antioxidants in vitro (2,10–12,15,16), but the efficacy of CGJ flavonoids in humans is unclear. To provide protection against oxidative stress in vivo, CGJ flavonoids must be absorbed and retained in the body in a form that still has antioxidant properties. Studies showed that when dietary flavonoids from food sources are absorbed from the gut, the circulating species are almost entirely conjugated (1) and that many of these conjugated metabolites have antioxidant properties in vitro (8,9). We observed significant increases in plasma concentrations of total and conjugated phenols after subjects consumed 10 mL CGJ · kg-1 · d-1 for 2 wk, which suggests that flavonoids were absorbed, conjugated, and retained in circulation even after a 10–12-h fast. This suggestion is in accord with the results of other investigators, who found that plasma flavonoids remain in the circulation for 8–20 h after cessation of supplementation and that some species have a half-life of up to 25 h (1,8,45,46).
CGJ supplementation provided significant antioxidant protection to serum, plasma proteins, and LDL as evidenced by changes in serum ORAC activity, concentrations of protein carbonyls in native plasma, and LDL oxidative resistance. These results are in agreement with those of other studies, which showed the in vivo antioxidant activity of red grape juice concentrate or purple grape juice (19–21). However, our study is the first to show that CGJ has an apparent antioxidant effect on endogenous protein oxidation. This is significant because other investigators did not detect changes in plasma protein carbonyl concentrations after healthy volunteers were supplemented with 400 mg ascorbic acid/d, 400 IU -tocopherol/d, or 600 mg lipoic acid/d for > 5 wk (24,47). Yet in our study, CGJ supplementation decreased plasma protein carbonyl concentrations 20%, which was a significantly different effect from that of supplementation with 400 IU -tocopherol/d. In vitro studies of catechins and anthocyanins showed that these phenols increase antioxidant capacity, bind to LDL, and protect lipids and proteins from oxidation (15–17,48,49). We therefore believe that the results that we observed after CGJ supplementation were the direct result of the plasma phenols derived from the CGJ flavonoids and were not the result of any other dietary or plasma antioxidant because subjects followed a strict diet and did not take supplements or medications, and their plasma ascorbic acid and -tocopherol concentrations did not change. However, the antioxidant effects of CGJ were not as potent and universal as anticipated because supplementation did not protect plasma proteins against AAPH oxidation or reduce the concentrations of endogenous urinary F2-isoprostanes. Lipoic acid is a potent antioxidant that has been shown to decrease urinary F2-isoprostane concentrations and protect plasma proteins against in vitro oxidation after 8 wk of supplementation (24); therefore, it is possible that 2 wk of CGJ supplementation was not sufficient to fully saturate plasma proteins or lipids to protect them against in vitro oxidative damage. Although other studies showed that 2 wk of supplementation with 450 mg red wine phenols/d, 100 mg -tocopherol/d, or 600 mg -tocopherol/d was sufficient to decrease urinary or plasma concentrations of F2-isoprostanes in subjects with elevated concentrations due to oxidative stress (50–52), a recent 2-wk supplementation study showed that 1200 mg -tocopherol/d had no effect on urinary F2-isoprostane concentrations in subjects with normal concentrations (53). In our study, subjects were provided comparable doses of CGJ phenols (432 ± 92 mg phenolic equivalents/d) or -tocopherol, as cited in these other studies, but in vitro protein carbonyl formation and urinary F2-isoprostane concentrations were not altered by either supplementation regimen. Thus, individuals experiencing chronic oxidative stress apparently respond more rapidly and dramatically to antioxidant supplementation than do healthy, normolipidemic individuals who are not undergoing any abnormal oxidative stress.
In vitro studies suggest that the flavonoids that are predominant in grape juice are more potent antioxidants than are the aqueous or lipophilic forms of -tocopherol (2). Because CGJ flavonoids have hydrophilic and lipophilic properties (2,13,14), we anticipated that CGJ would provide protection against lipid and protein oxidation that was significantly better than that provided by -tocopherol. Although CGJ consumption was uniquely associated with lower plasma protein carbonyl concentrations, no other significant differences were found between the 2 supplementation groups. The discordance between the results of in vitro studies and the results of our in vivo experiment is probably due to differences in the absorption and retention of dietary flavonoids (54,55) and -tocopherol, because supplementation with CGJ increased conjugated phenol concentrations only 22%, whereas -tocopherol supplementation increased plasma -tocopherol concentrations 92%. Additionally, circulating flavonoids and metabolites probably had not reached a steady state during the 2-wk study (54), and studies of longer duration would probably show more significant effects. Furthermore, the stringent dietary protocol used in our study may have limited the effects of CGJ by also restricting ascorbic acid and other phytochemicals that interact with CGJ flavonoids and potentiate their antioxidant properties. However, it is noteworthy that both supplements retained their antioxidant properties within the strict dietary protocol and that neither CGJ nor -tocopherol acted as prooxidants when taken with the flavonoid-restricted diets. These results provide evidence against the potential prooxidant effect of antioxidants such as -tocopherol that was suggested by Stocker (56). Additionally, the modest increase in plasma triacylglycerol concentration after subjects consumed CGJ may have limited its antioxidant potency, but this should not have influenced the results of the comparison with -tocopherol. This modest change in plasma triacylglycerol concentration was probably a reflection of biological variation and the transitory response to increased energy (carbohydrates) intake from CGJ and is probably not significant in healthy populations but may be deleterious in diabetic patients or patients with hypertriglyceridemia.
In summary, we showed that supplementing healthy volunteers for 2 wk with 10 mL 100% CGJ · kg-1 · d-1 significantly increased serum ORAC and LDL resistance to oxidative modification and that this dose of CGJ had antioxidant potency comparable to that of 400 IU -tocopherol/d. Of the 2 antioxidant supplements, only CGJ decreased endogenous protein oxidation. However, the 2-wk duration of the study and the flavonoid restrictions of the dietary protocol may have limited the full antioxidant effect of CGJ on lipid and protein oxidation. Thus, future studies should investigate the long-term antioxidant effects of CGJ consumption in combination with a well-defined but less restrictive diet. The phenolic and flavonoid composition of a beverage has been shown to correlate with antioxidant activity (57,58); yet, the phenolic and anthocyanin content of CGJ varies from 1800 to 2500 parts gallic acid per million and from 350 to 500 parts malvidin per million, respectively (Welch Foods Inc). Therefore, future research should attempt to fully characterize the flavonoid composition of the CGJ and the background diet. This approach would provide optimal dietary conditions so that the in vivo antioxidant properties, metabolism, and resulting plasma, lipoprotein, cellular, and urinary concentrations of flavonoids and metabolites can be evaluated in a systematic and physiologically relevant manner. The antioxidant capacity of CGJ appears to be significant, and because it is a highly palatable beverage that can be consumed by both adults and children, it has the potential to help reduce the risks of chronic diseases associated with free radical damage.
ACKNOWLEDGMENTS
We gratefully acknowledge Gary Beecher of the US Department of Agriculture for helpful conversation and information on the flavonoid content of commonly eaten foods; Beverley Adams-Huet for assistance with the statistical analysis of the data; Karine Marangon, Shaina Hirany, Alicia Summers, and Runna Alaw for assistance in conducting the study; and Ronald Tankersley for assistance with manuscript preparation.
REFERENCES