点击显示 收起
1 From the Departments of Nutrition, Internal Medicine, and Food Science, University of California, Davis, and Analytic and Applied Sciences, MARS Inc, Hackettstown, NJ.
2 Supported in part by CNRU grant DK-35747. The aortic endothelial cells used were a gift from ME O'Donnell, University of California, Davis. 3 Address reprint requests to DD Schramm, Department of Nutrition, University of California, One Shields Avenue, Davis, CA 95616-8669. E-mail: derek{at}ipworld.com.
ABSTRACT
Background: Polyphenolic phytochemicals inhibit vascular and inflammatory processes that contribute to disease. These effects are hypothesized to result from polyphenol-mediated alterations in cellular eicosanoid synthesis.
Objective: The objective was to determine and compare the ability of cocoa procyanidins to alter eicosanoid synthesis in human subjects and cultured human aortic endothelial cells.
Design: After an overnight fast, 10 healthy subjects (4 men and 6 women) consumed 37 g low-procyanidin (0.09 mg/g) and high-procyanidin (4.0 mg/g) chocolate; the treatments were separated by 1 wk. The investigation had a randomized, blinded, crossover design. Plasma samples were collected before treatment and 2 and 6 h after treatment. Eicosanoids were quantitated by enzyme immunoassay. Endothelial cells were treated in vitro with procyanidins to determine whether the effects of procyanidin in vivo were associated with procyanidin-induced alterations in endothelial cell eicosanoid synthesis.
Results: Relative to the effects of the low-procyanidin chocolate, high-procyanidin chocolate induced increases in plasma prostacyclin (32%; P < 0.05) and decreases in plasma leukotrienes (29%; P < 0.04). After the in vitro procyanidin treatments, aortic endothelial cells synthesized twice as much 6-keto-prostaglandin F1 (P < 0.01) and 16% less leukotriene (P < 0.05) as did control cells. The in vitro and in vivo effects of procyanidins on plasma leukotriene-prostacyclin ratios in culture medium were also comparable: decreases of 58% and 52%, respectively.
Conclusion: Data from this short-term investigation support the concept that certain food-derived flavonoids can favorably alter eicosanoid synthesis in humans, providing a plausible hypothesis for a mechanism by which they can decrease platelet activation in humans.
Key Words: Chocolate • eicosanoids • flavonoids • endothelial cells • phytochemicals • leukotrienes
INTRODUCTION
A therapeutic value of agents that either increase prostacyclin synthesis or decrease leukotriene synthesis has been postulated for 2 decades (1, 2). In both cases, the resulting effects were suggested to be of value in the treatment or prevention of vascular and respiratory pathologies. In this regard, recent clinical investigations focused on the value of prostacyclin analogues and leukotriene antagonists. For example, cysteinyl leukotriene receptor antagonists (eg, pranlukast) and 5-lipoxygenase inhibitors (eg, zileuton) were reported to provide asthma control in some patients (3). Likewise, prostacyclin analogues (eg, iloprost) were reported to be effective in the treatment of many cardiovascular conditions, including pulmonary embolism and ischemia-induced neuronal necrosis (4–6). Complementing the use of eicosanoid-modulating pharmacologic agents in the treatment of vascular and inflammatory pathologies, there has been increasing interest in the concept that certain dietary factors may be of value in the treatment or prevention of these pathologies by altering eicosanoid synthesis and through other mechanisms (7–9). One group of dietary compounds that has received considerable interest in this regard is the polyphenolic flavonoids.
Numerous investigators have reported data from epidemiologic studies, acute and chronic feeding studies, and in vitro studies that support the hypothesis that polyphenolics can reduce the risk of cardiovascular disease (10–15). However, mechanisms that underlie the reported effects are a subject of considerable debate. Recent postulated mechanisms of action for these compounds include interference with oxidation, decreased platelet activation, and increased eicosanoid synthesis (7, 9). Presumably, the mechanisms are dependent on many factors, including the amount of polyphenolic compounds consumed, genetics of the individual, and numerous environmental factors (eg, diet and lifestyle).
In the present study, we focused on the hypothesis that select dietary polyphenolics can influence eicosanoid synthesis in vivo. This hypothesis is supported by the observation that, in vitro, the activity of eicosanoid synthesizing enzymes (eg, prostaglandin-endoperoxide synthase; EC 1.14.99.1) is altered by select polyphenolics (16, 17) and the observation that polyphenolics alter cellular eicosanoid synthesis (18–20). Ten subjects consumed high- and low-polyphenolic chocolate in a randomized double-blind study. Chocolate was used because it has been shown to be a dense dietary source of biologically active procyanidins (21–24; MM Bearden, DA Pearson, D Rein, et al, unpublished observations, 2000) and because of our ability to alter its polyphenolic content without inducing additional alterations in its composition (25). In this investigation, we determined and compared the in vivo and in vitro effects of chocolate procyanidins on eicosanoid synthesis.
SUBJECTS AND METHODS
Experimental design
Eleven subjects were recruited in the spring and summer of 1999 from the University of California at Davis to participate in a randomized, double-blind, crossover trial that involved the consumption of low-procyanidin and high-procyanidin chocolate. The subjects had a mean (±SD) age of 39 ± 5 y (range: 20–55 y), body weight of 68 ± 7 kg, body height of 1.67 ± 0.3 m, and body mass index (in kg/m2 ) of 24 ± 3. Ten subjects (4 men and 6 women) completed both chocolate treatments. One subject dropped out after reporting an upset stomach. The 2 d of treatment were separated by 1 wk. The subjects were free from known disease and were nonsmokers. The unit of randomization was the individual. Treatments were assigned numerical designations by an investigator who was independent of the executor of assignment. Chocolate treatments did not differ in size, color, or shape, and only minor differences in taste were noted. On the day before each of the 2 experimental days, the subjects were instructed to refrain from consuming foods other than meat, egg noodles, and water after 1830 and were required to fast after 2200. After the overnight fast, experimentation was conducted at baseline as follows. Eighteen milliliters of blood was taken by venipuncture to establish baseline values. The subjects consumed 45 g of a bagel and 37 g chocolate within 15 min after the first blood sample was taken. A second blood sample was taken 2 h after baseline and then the subjects consumed 95 g of a plain bagel during the next hour. A final blood sample was taken 6 h after baseline. Venous blood samples were drawn into evacuated tubes containing EDTA or heparin. Plasma was separated by low-speed centrifugation (1500 x g) at 4°C for 10 min, portioned, and stored at -80°C until analyzed. All subjects gave informed consent before entering the study and the procedures were in accordance with the guidelines of the Human Subjects Review Committee of the University of California, Davis.
Low- and high-procyanidin foods
The bagels fed in this study contained nondetectable amounts of procyanidins. The low- and high-procyanidin chocolates were produced, analyzed, and supplied by Mars, Inc (Hackettstown, NJ). The high-procyanidin chocolate bar was a standard commercial product (DOVE Dark Chocolate); the low-procyanidin chocolate bar was identical in appearance but contained only trace amounts of the procyanidins. The compositions of the low-procyanidin and high-procyanidin bars, respectively, were as follows: fat, 327 and 330 g/kg; carbohydrate, 493 and 511 g/kg; caffeine, 0.5 and 0.6 g/kg; theobromine, 4.1 and 4.7 g/kg; total procyanidins, 0.09 and 4.0 g/kg; and epicatechin procyanidin, 0.05 and 1.1 g/kg. The procyanidin content was determined according to the method of Adamson et al (25).
Assessment of plasma epicatechin
Sample extraction
Chemicals were purchased from Sigma Chemical Co (St Louis) unless otherwise stated. The plasma concentration of epicatechin was determined as described by Richelle et al (26). In brief, 200 µL heparin-containing human plasma was mixed by vortex with 20 µL of a mixture of ascorbic acid and EDTA (1 mol ascorbic acid/L + 2.5 mmol EDTA/L) and 20 µL of a ß-glucuronidase solution (83 nkat/L, or 5000 U/mL). After 45 min of incubation (37°C), 0.5 mL acetonitrile was added and the samples were mixed by vortex and centrifuged. The supernatant fraction was combined with 50 mg alumina and 1 mL tris buffer (pH 7.6), mixed by vortex, and centrifuged. After the supernate was discarded, the alumina was washed with 1 mL tris buffer (pH 7.6) followed by 1 mL methanol. Residual methanol was evaporated under nitrogen and epicatechin was extracted from the alumina with 250 µL of 250 mmol perchloric acid/L. The sample was then centrifuged and the supernate was filtered through a 0.2-µm polyethersulfone membrane (Nalgene, Rochester, NY). All centrifugations were done at 10000 x g for 5 min at 4°C unless otherwise noted. Fifty microliters of the final filtrate was used for chromatographic analysis of epicatechin.
Chromatographic conditions
Samples were injected onto an 1100 series system with electrochemical detection (ESA Coulochem II; Hewlett-Packard, Chelmsford, MA), a guard cell setting of 550 mV, a conditioning cell setting of 100 mV, and an analytic cell setting of 400 mV. Chromatography was carried out by using an Alltima C18 reversed-phase column (5 µm, 150 x 4.6 mm; Alltech, Deerfield, IL) with a 0.5-µm frit precolumn filter (Upchurch, Oak Harbor, WA). A gradient elution with a flow rate of 1 mL/min was performed by using 2 solvents: solvent A [methanol: 50 mmol sodium acetate/L (40:60), pH 5.0] and solvent B [methanol: 100 mmol sodium acetate/L in water (7:93), pH 5.2]. The gradient was 20% solvent A at t = 0 min, was increased to 40% solvent A at 1 min, and then was increased to 80% solvent A at 3.5 min. Solvent A was maintained at 80% until t = 20 min and then decreased linearly to 20% by t = 30 min.
Measurement of eicosanoids
Immunoassay procedures were conducted in culture medium and plasma as reported previously (18). Eicosanoids were measured as their parent compounds or as their stable metabolites (eg, 6-keto-prostaglandin F1 for prostacyclin). Samples were analyzed for each subject and time point in triplicate. The enzyme immunoassays were purchased from Cayman Chemical Company (Ann Arbor, MI) and Oxford Biomedical Research (Oxford, MI).
Cell culture and treatment methods
Cell culture materials were purchased from Sigma Scientific (St Louis). Cells were cultured in minimum essential medium (MEM) as described previously (19–21). Endothelial cells (passage 10) were seeded onto 6-well plates with MEM containing 2 mmol glutamine/L, 10% fetal bovine serum, 100 U penicillin/L, 0.1 mg streptomycin/L, and 0.25 µg amphotericin/L. Endothelial cell integrity was monitored by trypan blue exclusion.
Confluent cells were treated in 1 mL MEM containing treatment compounds or vehicle (5% ethanol). The procyanidin extract was purified from Cocoapro cocoa (Mars, Inc). This extract was characterized and quantified according to Adamson et al (25) and contained 49% total procyanidins by weight with the following contributions from the various oligomeric classes: 43.3% monomer, 15.7% dimer, 13.5% trimer, 10.1% tetramer, 6.6% pentamer, 4.4% hexamer, 2.2% heptamer, 2.0% octamer, 1.2% nonamer, and 1.0% decamer. Treatment was applied at a concentration of 2 mg/L. Confluent endothelial cell layers were treated for 3 d with fresh medium placed on the cells at 0800, 1300, and 1800. Before use, MEM was incubated for 12 h in 5% CO2 to allow proper gas equilibration. MEM from the incubations at 1300 and 1800 on day 3 was collected and stored at -80°C for analysis by immunoassay. Experiments were conducted in triplicate.
Statistical analysis
Wilcoxon's signed-rank test was used to determine changes from baseline and Wilcoxon's rank-sum test was used to determine differences between men and women. Cell culture data were compared by using Student's t test. P values <0.05 were considered statistically significant. Unless otherwise stated, results are expressed as means ± SDs. The statistical analyses were conducted by using STATVIEW (SAS Institute, Inc, Cary, NC).
RESULTS
Plasma concentrations of the monomeric procyanidin epicatechin at baseline (0 h) and 2 and 6 h for subjects in the low- and high-procyanidin-chocolate groups are shown in Figure 1. All 10 subjects absorbed the procyanidin monomer epicatechin. At 2 h, plasma epicatechin concentrations in subjects who consumed the high-procyanidin chocolate were 19.6 times those of subjects who consumed the low-procyanidin chocolate (21.2 ± 8.3 compared with 1.1 ± 1.3 nmol/L). After treatment with the high-procyanidin chocolate, plasma concentrations of epicatechin correlated inversely with the weight of the subjects (r = -0.715, P < 0.02). There were no significant differences or trends in the absorption or clearance of the procyanidins between men and women.
FIGURE 1.. Changes in plasma epicatechin monomer concentrations after subjects ingested 37 g low-procyanidin chocolate, which provided 3.3 mg total procyanidin, and after ingestion of 37 g high-procyanidin chocolate, which provided 148 mg total procyanidin. n = 10.
Relative to the low-procyanidin chocolate, the high-procyanidin chocolate induced significant changes in plasma leukotriene and prostacyclin concentrations at 2 h but not at 6 h. Two hours after the low- and high-procyanidin chocolates were consumed, plasma leukotriene concentrations were 9.18 ± 0.75 and 6.53 ± 0.83 nmol/L (P < 0.03) and plasma prostacyclin (as 6-keto-PGF1) concentrations were 397 ± 40 and 554 ± 37 pmol/L (P < 0.05), respectively. Procyanidin consumption induced a significant decrease in the leukotriene-prostacyclin ratio (P < 0.01). Plasma leukotriene, prostacyclin, and the leukotriene-prostacyclin ratios of individual subjects in both chocolate-treatment groups are depicted in Figure 2. No sex differences were noted. Regression analysis showed that the relation between the plasma leukotriene-prostacyclin ratio and the plasma epicatechin concentration was asymptotic (y = -6.62 +1.8x - 0.129x2; P < 0.05).
FIGURE 2.. Comparison of the effects of low-procyanidin and high-procyanidin chocolate on plasma leukotriene (LTC4 + LTD4 + LTE4) concentrations, prostacyclin I2 (PGI2) concentrations, and the leukotriene-prostacyclin ratio 2 h after ingestion. n = 10.
In an effort to determine whether the identified in vivo effects of procyanidins could have resulted from the effects of procyanidin on endothelial cells, the influence of purified procyanidins on endothelial cells was studied in vitro. As depicted in Table 1, the cocoa-derived procyanidin extract induced significant decreases in concentrations of cysteinyl leukotrienes C4 + D4 + E4 and significant increases in prostacyclin in MEM. Values for the concentrations of 6 other eicosanoids are also reported.
View this table:
TABLE 1.. Autocrine-product profile of medium from cells treated with the procyanidin extract of cocoa (2 mg/L)1
DISCUSSION
Data from this short-term investigation support the concept that certain food-derived flavonoids can favorably alter eicosanoid synthesis in humans, providing a plausible hypothesis for a mechanism by which they can decrease platelet activation in humans. The consumption of high-procyanidin chocolate resulted in 1) an increase in plasma prostacyclin concentrations, 2) a decrease in plasma leukotriene concentrations, and 3) a decrease in the plasma leukotriene-prostacyclin ratio, a measure of the proinflammatory-antiinflammatory eicosanoid balance. The results also support the idea that procyanidins, in part, mediate their in vivo effects on eicosanoid concentrations by modulating endothelial cell eicosanoid synthesis. Other potential eicosanoid-synthesizing targets include platelets and white blood cells. Plasma epicatechin concentrations increased rapidly after consumption of the procyanidins; however, concentrations also rapidly returned to baseline. The transient nature of the effects observed in this investigation illustrates the critical nature of sample timing to investigations involving polyphenolics, and supports the concept that the consumption of polyphenolics many times each day might be necessary for optimal vascular protection.
The ability of dietary polyphenolics to modulate eicosanoid synthesis in vivo might be due to their ability to induce specific effects on the activity of enzymes that synthesize or degrade eicosanoids. Prostaglandin-endoperoxide synthase activity, for example, is modulated by numerous specific polyphenolics. This is an important observation relative to our data because the amount of prostaglandin H2, which is synthesized by prostaglandin-endoperoxide synthase, has been shown to be rate-limiting to prostacyclin synthesis (27). Polyphenolic effects on the activity of both cellular (28) and purified (16) prostaglandin-endoperoxide synthase have been observed. When present at physiologically relevant concentrations, polyphenolics stimulated the activity of purified prostaglandin-endoperoxide synthase; the level of stimulation increased with increasing polyphenolic hydroxylation (16).
Polyphenolics are also known to modulate the activity of numerous leukotriene-synthesizing enzymes. For example, the activities of 5-lipoxygenase and 12-lipoxygenase can be inhibited by many specific polyphenolics in vitro (17, 18). 5-Lipoxygenase adds oxygen to the C-5 of arachidonic acid, producing leukotriene A4, the precursor to leukotriene C4, leukotriene D4, and leukotriene E4. Thus, we hypothesized that the effects of procyanidins on leukotriene synthesis observed in this investigation resulted from decreased 5-lipoxygenase activity. Finally, because of the rapid and transient nature of the observed procyanidin effects, we can eliminate the possibility that procyanidins mediated their effects at the transcription and translation levels.
In conclusion, the results of this study support the hypothesis that modulation of eicosanoid synthesis is one mechanism by which certain food-derived polyphenolics, at normal intakes, confer cardioprotective effects. For example, prostacyclin inhibits platelet aggregation through its ability to elevate platelet cyclic AMP concentrations, inhibiting agonist-induced increases in glycoprotein IIb-IIIa expression and phosphorylation (29–31). Thus, the results provide an explanation for the ability of polyphenolics to decrease platelet aggregation in vivo and ex vivo (13, 22, 32). Other possible benefits of the food-derived procyanidins include effects on the activity of nitric-oxide synthase and lipid peroxidation. In addition, it is important to note that although the use of different study designs makes comparisons of the results of different studies difficult, cocoa procyanidins induced comparable changes in leukotriene-prostacyclin ratios (decreases of 52% and 58%, respectively) in plasma and culture medium. Therefore, we conclude that the endothelial cell culture system used can provide data that mimic the in vivo effects of procyanidins. Future studies are needed to determine whether the acute effects of procyanidins in humans are maintained with chronic procyanidin feeding, as was reported recently in rats (33), and to determine the extent of other potential mechanisms (7) of procyanidin action in vivo.
ACKNOWLEDGMENTS
We thank Alina Wettstein for her assistance with this investigation, Gerald Robinson for his inspiration, and Mary E Meyers and Eric J Whitacre for their expertise in the preparation of the chocolate products used in this study.
REFERENCES