点击显示 收起
1 From the University of Western Australia, Department of Medicine, and the West Australian Heart Research Institute, Royal Perth Hospital, Perth, Australia.
2 Supported by research grants from the National Heart Foundation of Australia and The Tetley Group. 3 Address reprint requests to JM Hodgson, University of Western Australia, Department of Medicine, GPO Box X2213, Perth, WA 6847, Australia. E-mail: jonathan{at}cyllene.uwa.edu.au.
ABSTRACT
Background: Tea has been associated with a reduced risk of cardiovascular disease. One proposed mechanism of this risk reduction involves inhibition of lipoprotein oxidation in vivo by antioxidant polyphenolic compounds derived from tea. However, controlled interventions uniformly failed to show that ingestion of tea can inhibit LDL oxidation ex vivo. The absence of effects in previous studies may be due to the isolation of LDL particles from polyphenolic compounds that are present in the aqueous phase of serum.
Objective: The objective of this study was to examine the acute effects of ingestion of black and green tea on ex vivo Cu2+-induced lipoprotein oxidation without prior isolation of lipoproteins from serum.
Design: The acute effects of 4 hot drinks—green tea and black tea (each at a dose equivalent to 4 standard cups), water matched to the teas for caffeine content, and water—were assessed in 20 healthy men by using a Latin-square design. The lag time to lipoprotein diene formation, slope of the propagation phase of the oxidation curve, and area under the oxidation curve were calculated. Urinary concentrations of 4-O-methylgallic acid were used as a marker of uptake and metabolism of polyphenolic compounds from tea.
Results: Significant increases in urinary 4-O-methylgallic acid for black and green tea (P < 0.0001) were observed. Caffeine did not significantly influence lipoprotein oxidation. Compared with the water control, there was a greater lag time for black tea (5.4 ± 2.9 min; P = 0.05) that was of borderline significance and a similar trend for green tea (4.4 ± 2.8 min; P = 0.17). Slope and area under the oxidation curve were not altered.
Conclusion: Black tea has a mild acute effect on ex vivo lipoprotein oxidation in human serum. 2000;71:–7.
Key Words: Black tea • green tea • antioxidant • caffeine • flavonoid • polyphenolic compounds • lipoprotein • Australia
INTRODUCTION
Results of epidemiologic studies show that tea (1–3) and flavonoids derived from tea (1, 3–5) are associated with a reduced risk of cardiovascular disease. A potential mechanism for such an effect involves inhibition of lipoprotein oxidation by flavonoids derived from tea (6). The development and progression of atherosclerosis is thought to be critically dependent on lipoprotein oxidation (7).
The antioxidant activity of tea has been assessed in vitro (8–10) and ex vivo after ingestion of tea (9–13). We showed previously that extracts of black and green tea inhibit lipoprotein oxidation in human serum to a similar extent in vitro (8). In addition, significant increases in the total antioxidant capacity of serum after tea ingestion have been reported (10, 11), but this has not been a consistent finding (9, 11). Although assessment of lipoprotein oxidizability in vitro may provide a useful guide as to the potential of tea to inhibit oxidation, it does not establish any in vivo effects.
Controlled interventions of tea drinking have consistently failed to show that tea ingestion can inhibit LDL oxidation ex vivo (9, 11–13). The absence of effects on LDL oxidizability ex vivo may be related to the method used to assess LDL oxidizability. The standard method involves isolation of the LDL particles from the aqueous phase of serum. Catechins, the major class of flavonoids found in tea, were measured in LDL but only at <10% of the total serum concentration (14). Most of the catechins were associated with the water-soluble fraction and it was concluded that there was insufficient accumulation of catechins in LDL to improve resistance to LDL oxidation ex vivo (14). It was suggested that flavonoids may have action in the aqueous phase (15, 16), perhaps at the surface of lipoprotein particles. Therefore, the isolation of LDL from antioxidant flavonoids in the water-soluble fraction of serum may be inappropriate. Increases in total antioxidant capacity of serum after ingestion of tea (10, 11) may be due to the presence of phenolic antioxidants in the aqueous phase.
To assess possible in vivo effects of tea, we studied the acute effects of ingestion of black and green tea on ex vivo lipoprotein oxidation without prior isolation of lipoproteins from serum. Such effects have not been examined previously without isolation of lipoproteins from the aqueous phase of serum. The use of this method may resemble in vivo conditions more closely than previously. Total antioxidant capacity of serum was also assessed and urinary concentrations of 4-O-methylgallic acid, a major gallic acid metabolite (17, 18), were used as a marker of tea polyphenol uptake.
SUBJECTS AND METHODS
Subjects
Twenty healthy men aged 35–73 y were recruited from the general population in response to media advertisements. Potential volunteers were excluded after initial screening if they reported the use of any medication or dietary supplements; a history of major illness, including heart disease, diabetes mellitus, liver disease, and renal disease; current smoking or smoking cessation for <6 mo; a body mass index (BMI; in kg/m2) >33; an alcohol intake averaging >40 g/d; or regular tea or coffee intake averaging <1 cup/d. The Royal Perth Hospital Ethics Committee approved the study and all participants gave written, informed consent.
Experimental design
The acute effects on lipoprotein oxidation in serum of 4 hot drinks—black tea, green tea, water matched to the teas for caffeine content, and water—were assessed and compared by using a Latin-square design. The use of caffeine and water allowed determination of any effects of caffeine on lipoprotein oxidation.
A total of 4 clinic visits were conducted 1 wk apart and as close as possible to the same day of the week and at the same time of the day. Subjects consumed each of the 4 drinks over 30 min, 1 at each visit, in random order. The subjects were fasting and had avoided tea, coffee, and other caffeine-containing beverages for 12 h before each visit. The subjects were instructed to maintain their usual diets throughout the 4 wk within which the clinic visits occurred. A blood sample was taken just before the subject started drinking each drink and 60 min after the subject finished the drink. A urine sample was taken before the subject started the drink and again 90 min after he finished the drink.
Preparation of tea
Tea bags with a total of 7.6 g tea leaves were extracted, with constant stirring movement, for 4 min in 400 mL boiled water. The black tea was a blended black tea and the green tea was a Japanese-type "Sen-Cha" tea. The other 2 drinks were 400 mL water with a caffeine content matched to that of the teas and 400 mL water. The caffeine content of each tea and the caffeine drink was 180 mg. All drinks were consumed hot and were prepared and consumed without additives such as milk and sugar.
Lipoprotein oxidation in serum
We used a modification of the serum oxidation method described by Regnstrom et al (19), which was described previously in detail (20, 21). This method provides an indication of diene formation in lipoprotein fatty acids in serum exposed to Cu2+, assessed by measuring change in absorbance at 234 nm. Briefly, the fasting blood sample was drawn from the antecubital fossa vein into evacuated glass tubes and the serum was separated by using standard procedures. Serum samples were then frozen at -80°C until assayed. Serum was thawed and then diluted to a final concentration of 0.67% in phosphate-buffered saline solution (pH 7.4). Oxidation was initiated by the addition of 12 µmol Cu2+/L (final concentration) added as CuSO4•5H2O. Oxidation kinetics were determined by measuring the absorbance at 234 nm with a DU650 ultraviolet-visible spectrophotometer (Beckman Instruments Inc, Fullerton, CA). Absorbance readings were taken every 20 min over 240 min at 37°C.
The variables used to describe any differences between oxidation curves were lag time to lipoprotein diene formation (lag time), the slope of the propagation phase of the oxidation curve (slope), and the area under the oxidation curve. Lag time was measured from the plot of absorbance against time and was defined as the intercept between the baseline and the tangent of the absorbance curve during the propagation phase. The lag time provides an indication of protection of lipoproteins against oxidation. Two blinded observers measured the lag times and the results were averaged. All comparisons were made on within-run assessments to limit between-run variability. The within-run CV for the assessment of lag time when this technique was used was 8.8%. Slope and area under the curve were calculated by using PRISM software (GraphPad, San Diego).
Total antioxidant activity in serum
The total antioxidant activity of serum was determined by a spectrophotometric assay for total peroxyl radical–trapping antioxidant potential in human serum, as described in detail by Valkonen and Kuusi (22). Briefly, serum was thawed and then diluted to 1.25% in phosphate-buffered saline solution (pH 7.4). The oxidation of the free radical probe dichlorofluorescin-diacetate (DCFH-DA) (14 µmol/L) was then monitored at room temperature after addition of the free radical generator 2,2'-diazobis(2-amidinopropane)dihydrochloride (AAPH) (56 mmol/L) by measuring absorbance at 504 nm. Oxidation of DCFH-DA begins once serum antioxidants have been used. Trolox (Aldrich Chemical Co, Milwaukee) (8.4 µmol/L) was used as an internal standard. Each Trolox molecule is able to neutralize 2 molecules of peroxyl radicals. Therefore, the total antioxidant activity could be quantified by using the following equation:
RESULTS
The mean age of the men recruited to this study was 56.2 ± 1.1 y. The subjects were healthy nonsmokers with BMIs between 19.0 and 32.4 (
View this table:
TABLE 1.. Serum lipoprotein oxidation–derived variables and total antioxidant activity of serum before and 60 min after and urinary 4-O-methylgallic acid concentrations before and 90 min after ingestion of water, caffeine, black tea, and green tea1
Baseline-adjusted differences in postdrink lag time, slope, and area under the curve and total antioxidant activity of serum are shown in Table 2. A greater lag time for black tea than for water was of borderline significance. There were no other significant differences in other variables derived from curves of Cu2+-induced lipoprotein oxidation in serum for black tea or green tea. A trend toward greater total antioxidant activity for green tea than for water was also observed.
View this table:
TABLE 2.. Baseline-adjusted differences in 60-min postdrink assessments of serum lipoprotein oxidation–derived variables and total antioxidant activity of serum for black tea and green tea compared with water1
The total polyphenol intakes from the tea infusions were 1.44 and 1.37 g for black tea and green tea, respectively. The total gallic acid intakes from the tea infusions were 103 and 102 mg for black tea and green tea, respectively. Changes in 4-O-methylgallic acid concentrations in urine with ingestion of each drink are shown in Table 1. Compared with water, black tea (P < 0.0001) and green tea (P < 0.0001) were associated with significantly greater 4-O-methylgallic acid concentrations. The increases were significantly larger for black tea than for green tea (P < 0.001).
DISCUSSION
The main objective of this study was to determine whether consumption of black or green tea had acute inhibitory effects on lipoprotein oxidation ex vivo. We found a greater lag time to lipoprotein oxidation in human serum after ingestion of black tea than after ingestion of the control drinks that was of borderline significance, and a similar trend for green tea that did not reach significance. However, the observed effects on variables derived from curves of Cu2+-induced lipoprotein oxidation, and on total antioxidant activity, were generally small.
"Black" and "green" are the 2 main classes of tea. Black teas are produced by promoting enzymatic oxidation of flavonoids, whereas enzymes involved in flavonoid oxidation are inactivated during the manufacture of green tea. Flavonoids and other polyphenols with antioxidant activity constitute 40% of the weight of the black and green tea leaves (25). The measurements of total polyphenols in the tea infusions used were 1.44 g (19%) for black tea and 1.37 g (18%) for green tea. These amounts correspond to 50% recovery and are reasonable given that infusions of tea prepared with hot water will not extract all polyphenols from the tea leaf.
About 5% of the weights of black and green tea leaves are in the form of gallic acid esters or free gallic acid (25). The measured gallic acid contents of the black and green tea infusions were similar (103 and 102 mg, respectively). A major metabolite of gallic acid in humans is 4-O-methylgallic acid (17), which appears in the urine soon after ingestion of tea. Therefore, urinary concentrations of 4-O-methylgallic acid were used as a marker of uptake and metabolism of polyphenolic compounds from the tea.
The increases in urinary 4-O-methylgallic acid concentrations were significantly larger after ingestion of black tea than of green tea. Although the total concentration of gallic acid is similar for black and green tea, the form in which the gallic acid is present differs as a result of oxidative changes in black tea during production. Free gallic acid is higher in black tea—estimated at 1% of dry weight (25)—and possibly is more available for uptake and metabolism.
Ingestion of tea can make a major contribution to the intake of polyphenolic compounds (1). The high concentration of antioxidant polyphenolics in tea and the proposed link between lipoprotein oxidation and cardiovascular risk has led to investigations of possible antioxidant effects of tea on lipoprotein oxidation ex vivo.
Controlled interventions have consistently failed to show that ingestion of tea can inhibit LDL oxidation ex vivo (9, 11–13). One study did show a significant prolongation of lag time after ingestion of black tea (12), but this was an uncontrolled comparison. We suggest that the lack of effects of tea on LDL oxidation ex vivo in previous controlled interventions (9, 11–13) may relate to the method used to assess LDL oxidizability. The isolation of LDL from serum for the LDL oxidation assay, and findings showing little incorporation of tea polyphenols into LDL (14), strengthen this explanation. The assay used in the present study to assess ex vivo lipoprotein oxidation does not involve isolation of lipoproteins (19, 20). The oxidation of lipoprotein fatty acids occurs in the presence of the aqueous phase of serum. Assessment of lipoprotein oxidation in serum therefore may more closely resemble subendothelial fluid, where lipoprotein oxidation is suggested to occur (26).
Our findings are generally consistent with the theory that ingestion of tea can inhibit lipoprotein oxidation ex vivo. In addition, our results agree with the suggestion that assays that measure lipoprotein oxidation in an environment more closely resembling in vivo conditions (26) may be more likely to show significant effects. However, observed effects on variables related to serum lipoprotein oxidation were small and generally not significant.
Small, nonsignificant increases in total antioxidant activity were observed in this study 60 min after ingestion of black or green tea. These increases were 3% and 4% for black tea and green tea, respectively. Using a similar assay, Serifini et al (10) found a significant 40–50% increase in total antioxidant activity 30–50 min after ingestion of 1 cup of black or green tea. However, McAnlis et al (9) found no increase in total antioxidant activity from 30 to 180 min after ingestion of a single cup of strong tea. In addition, in studies of long-term tea consumption, increases in total antioxidant activity were small (3–10%) and generally not significant (9, 11).
The timing of blood sampling after tea ingestion is a factor that has the potential to influence the results obtained. In the present study, drinks were consumed over 30 min and blood samples were taken 60 min after the subjects finished the drink (90 min after they started). Data published before this study began suggested that antioxidant activity peaks within 90 min of ingestion of black or green tea (10). However, more recent findings suggest that the peak concentration of the main class of polyphenolic antioxidants in tea occurs 120–140 min after ingestion of green tea (27). Therefore, the assessment of lipoprotein oxidation in serum and total antioxidant activity closer to 120 min after tea ingestion could have resulted in more significant effects.
In conclusion, polyphenolic compounds from tea were absorbed and excreted in the urine within 90 min of tea ingestion. We showed that black tea can have a mild acute effect on the oxidation of lipoproteins in serum ex vivo. Tea consumption had no significant acute effect on total antioxidant activity in serum. Overall, although effects on oxidation-related variables were generally consistent in direction with the proposed antioxidant activity of tea, they were small. The final conclusion as to whether tea consumption affects lipid peroxidation in vivo will have to await the outcome of long-term studies using markers that are more representative of oxidative stress in vivo.
ACKNOWLEDGMENTS
We thank Nerissa Jordan and Penny Rogers for their help in carrying out the study.
REFERENCES