The Effect of Consuming Instant Black Tea on Postprandial Plasma ...

Original Research

The Effect of Consuming Instant Black Tea on Postprandial Plasma Glucose and Insulin Concentrations in Healthy Humans Judith A. Bryans, PhD, Patricia A. Judd, PhD, and Peter R. Ellis, PhD King’s College London, School of Biomedical and Health Sciences, Nutrition Health and Food Research Centre, Biopolymers Group, London (J.A.B., P.R.E.), University of Central Lancashire, Lancashire School of Health and Postgraduate Medicine, Preston (P.A.J.), UNITED KINGDOM Key words: tea, polyphenols, postprandial glycemia, plasma insulin concentrations Objective: To determine the effects of black tea on postprandial plasma glucose and insulin concentrations in healthy humans in response to an oral glucose load. Methods: A four-way randomised, crossover trial was designed in which 16 healthy fasted subjects would consume 75g of glucose in either 250ml of water (control), 250ml of water plus 0.052g of caffeine (positive control) or 250 ml of water plus 1.0g or 3.0g of instant black tea. Blood samples were collected at fasting and at 30min intervals for 150min from commencement of drink ingestion. Glucose and insulin concentrations were measured using standard methodology. The tea was chemically characterised using colorimetric and HPLC methods. Results: Chemical analysis showed that the tea was rich in polyphenolic compounds (total, 350mg/g). Results from only 3 treatment arms are reported because the 3.0g tea drink caused gastrointestinal symptoms. Plasma glucose concentrations ⬍60min in response to the drinks were similar, but were significantly reduced at 120min (P⬍0.01), following ingestion of the 1.0g tea drink, relative to the control and caffeine drinks. Tea consumption resulted in elevated insulin concentrations compared with the control and caffeine drinks at 90min (P⬍0.01) and compared with caffeine drink alone at 150min (P⬍0.01). Conclusions: The 1.0g tea drink reduced the late phase plasma glucose response in healthy humans with a corresponding increase in insulin. This may indicate that the attenuation in postprandial glycemia was achieved as a result of an elevated insulin response following stimulation of pancreatic ␤-cells. This effect may be attributable to the presence of phenolic compounds in the tea

INTRODUCTION

(EGCG). In addition, oolong and black teas contain theaflavins and black tea also contains thearubigins [1]. Many of these flavonoid polyphenols may offer benefits in the prevention of chronic diseases such as cardiovascular disease [2– 4] and certain cancers [5–7], and may also be useful in improving glucose tolerance. Flavonoids, such as those found in tea, have been shown to

Tea beverage, made using the leaves of the plant Camellia sinensis (L.) O. Kuntze, is consumed in large quantities worldwide and is a rich source of flavonoid polyphenols, particularly epicatechin (EC), epicatechin gallate (ECG), epigallocatechin (EGC) and epigallocatechin gallate

Address reprint requests to: Dr Peter R. Ellis, King’s College London, School of Biomedical and Health Sciences, Nutritional Sciences Division, Biopolymers Group, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NN, UNITED KINGDOM. E-mail: [email protected] Contributors: J.A.B. was involved in the study design, subject recruitment, collation and analysis of data and writing of the manuscript. P.A.J. and P.R.E. were involved in the design of the experiment, the writing of the manuscript and provided advice. None of the authors had a conflict of interest. Abbreviations: AUC ⫽ area under the curve, CGA ⫽ chlorogenic acids, EC ⫽ epicatechin, ECG ⫽ epicatechin gallate, EDTA ⫽ ethylendiaminetetraacetic acid, EGC ⫽ epigallocatechin, EGCG ⫽ epigalocatechin gallate, EIA ⫽ enteroinsular axis, GIP ⫽ glucose-dependent insulinotropic-polypeptide, GLP-1 ⫽ glucagon-like peptide-1, SGLT-1 ⫽ sodium glucose co-transporter-1.

Journal of the American College of Nutrition, Vo, 26, No. 5, 471–477 (2007) Published by the American College of Nutrition 471

Tea Reduces Postprandial Glycemia in Humans inhibit luminal and brush border carbohydrate-digesting enzymes in vitro and in vivo [8,9]. Inhibition of these enzymes would result in delayed carbohydrate hydrolysis and consequently a delay in the absorption of the monosaccharide products of carbohydrate digestion (i.e. mainly glucose). In addition, tea polyphenols are believed to directly inhibit glucose absorption by competitively binding with the sodium-glucose co-transporter-1 (SGLT-1) [10,11]. This active carrier system is the major route of glucose transport from the small intestine and is energised by an ATP-dependent Na⫹ pump. Welsch and colleagues [12] reported that tannic acid and chlorogenic acids (CGA) reduced glucose uptake into the brush border membrane vesicles of rats. They explained these observations by suggesting that phenolic compounds could dissipate the Na⫹ electrochemical gradient required to fuel SGLT-1 active transport. More recently, Johnston and co-workers [13] demonstrated that coffee, a rich source of CGA, attenuated plasma concentrations of glucose-dependent insulinotropic-polypeptide (GIP), while enhancing concentrations of postprandial glucagon-like polypeptide-1 (GLP-1) in healthy humans. According to these authors their data were consistent with the mechanism involving CGA-mediated dissipation of the Na⫹ electrochemical gradient. Regardless of the mechanism of action, inhibition of SGLT-1 activity and thus glucose transport by polyphenols would likely lead to a lowering of postprandial glycemia. Flavonoids are also believed to exert effects on insulin secretion. For example, in studies using in vitro techniques, it has been known for some time that EC increases insulin secretion in isolated pancreatic islets in a time, temperature, and dose dependent manner [14,15]. More recently, EGCG has been shown to enhance insulin activity in fat epididymal cells [16]. If the effect of isolated tea polyphenols is retained in humans consuming a normal tea drink, tea consumption may be of particular benefit to individuals with or at risk of diabetes. Indeed, it has recently been reported that oolong tea can ameliorate long term glycemic control in people with type 2 diabetes [17]. However, currently there is no information in the literature on the effects of drinking black tea on postprandial blood glucose concentrations in humans. The objective of this study, therefore, was to determine the effects of a nutritionally-relevant amount of black tea, with a known polyphenol content, on postprandial glycemia and insulinemia in response to a glucose challenge in humans. Since tea also contains caffeine, a methylxanthine known to alter blood glucose and insulin responses in humans [18 –20], the study was designed to account for caffeine as a confounding variable.

METHODS Subjects Sixteen healthy subjects (12 female and 4 male) with a mean age of 35.5 ⫾ 1.5 years and a mean BMI of 23.8 ⫾ 0.7

472

kg/m2 were recruited from the staff and student populations at King’s College London to participate in this study. Subjects were excluded if they had any of the following: diabetes mellitus, a history of iron deficiency anaemia, gastrointestinal pathologies, or were taking medication which would interfere with normal gastrointestinal function. It was further required that female subjects should not be pregnant or lactating. The Research Ethics Committee at King’s College London approved the study (record number: 00-01/02) and all subjects gave written consent after a full explanation of the study procedures.

Composition of the Experimental Drinks and Study Design The experimental drinks used for testing were as follows: (1) 75g glucose and 250ml of water (the control); (2) 75g glucose, 0.052g caffeine and 250ml of water (positive control); (3) 75g glucose, 1.0g instant black tea and 250ml of water; and (4) 75g glucose, 3.0g instant black tea and 250ml of water. Chemical analysis of the instant black tea allowed the positive control to be formulated as it provided a value for the caffeine content (see section on chemical analysis). A four-way randomised crossover design was used, so that each subject received each of the four drinks on a different day in randomised order. Subjects were asked to maintain their usual dietary intake and not to change their physical activity patterns during the study. Unlike other studies, the subjects were not required to exclude tea from their diets, since many of the flavonoids found in tea are present in a wide variety of other foods and beverages, including fruits, vegetables, wine, beer and chocolate, and their avoidance would have unduly restricted their diet.

Preparation of the Experimental Drinks and Subject Feeding On four separate days subjects attended the metabolic unit within the Department of Nutrition and Dietetics at King’s College London following an overnight fast (12h). On arrival at the unit, each subject provided a baseline blood sample and consumed either a control, caffeine or tea drinks, as described in Composition of the experimental drinks and study design. The drinks were prepared fresh immediately prior to feeding on each study day and were each served at the same temperature (58°C ⫾ 1°C). The control drink was prepared by dissolving 75g of glucose in 150ml of boiling tap water, adding 100ml of cold tap water (4°C) and mixing thoroughly. The caffeine drink (positive control) was prepared by dissolving 0.052g of caffeine (pulverised anhydrous caplets) and 75g of glucose in 150ml of boiling tap water, adding 100ml of cold tap water and mixing thoroughly. The two tea drinks were prepared by dissolving either 1.0g or 3.0g of instant black tea (obtained from commercial sources; Brooke Bond Foods, Crawley, UK) and 75g of glucose in 150ml of boiling tap water, adding 100ml of

VO, 26, NO. 5

Tea Reduces Postprandial Glycemia in Humans cold tap water and mixing thoroughly. The cold water, which had been stored in the fridge at 4°C (⫾ 1°C), was added to the drinks to enable the subjects to consume the drinks within an allocated time period of 15min. Once prepared, the drinks were given immediately to the subjects. Each drink had a total weight of ⬃325g and ingestion of the drinks was supervised and timed to ensure compliance.

Blood Sampling Venous blood samples were taken by repeated vene-puncture in a forearm vein and collected in vacutainers containing ethylendiaminetetraacetic acid (EDTA) at fasting (0 min) and at 30 min intervals from the commencement of drink ingestion for a total period of 3h. The venous blood samples were centrifuged for 10 min at 3000 rpm, (relative centrifugal force (RCF) ⫽ 1811) using a Beckman Coulter benchtop centrifuge. The plasma was aspirated and stored in sterile cryovials at ⫺80°C until required for glucose and insulin analysis.

Biochemical Assays Venous plasma glucose concentrations were analysed using an enzymic colorimetric method using the Cobas Integra 400 autoanalyser with a Glucose HK liquid cassette (Roche Diagnostic Products Limited, Welwyn, Hertforshire, UK). The samples were run as a continuous batch. The intra-assay coefficient of variation (CV) was 0.64%. Venous plasma insulin concentrations were determined by radioimmunoassay (RIA) using a kit purchased from DiaSorin (DiaSorin, Charles House, Toutley Road, Wokingham, Berkshire, UK). The intra-assay and inter-assay CVs were 4.5% and 10.5% respectively.

Chemical Analysis of the Instant Black Tea The total polyphenol concentration of the tea was determined using the Folin-Ciocalteu method. Briefly, an aliquot of an appropriately diluted sample (i.e. 1ml) and gallic acid standards, covering the range of 0 –50␮g/ml, were added to disposable test tubes. A reagent blank using de-ionised water was also prepared. Diluted Folin-Ciocalteu reagent was added to each of the tubes and 3 to 8 min later sodium carbonate solution (7.5%, w/v, 4ml) was added. The tubes were immediately capped and mixed. After incubation for 60min at room temperature (⬃ 21°C), the absorbance of the standards and the samples were measured at 765nm relative to de-ionised water in a 1cm cell. The total phenolic content of the samples is expressed as percentage in weight for weight of gallic acid equivalents. Samples for the measurement of total phenolic concentration were prepared in triplicate. The caffeine, flavan-3-ol and theaflavin contents of the tea was quantified by HPLC analysis with Diode Array Detection (HPLC-DAD) using a Dionex summit system equipped with a model P580 pump, a GINA 50 auto-sampler, an UVD340S diode array detector and a Chromeleon data system. Peak

JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION

detection was carried out at 274nm. Catechin analysis was performed on a Luna phenyl-hexyl column (5mm, 250⬘4.60mm, Phenomenex, UK) with a catechin elution programme. Visualisation by HPLC was carried out by the general characterisation method. Theaflavin analysis was performed on Hypersil C18 column (3mm, 100⬘.4.60mm, Phenomenex, UK) by theaflavin elution. Gallic acid analysis was carried out on a Customsil ODS1 column (300mm ⫻ 4.6mm i.d., ex. Phenomenex, UK). All of the mobile phases were composed of solvent A (2%, v/v, acetic acid in acetonitrile) and solvent B (2%, v/v, acetic acid in high grade pure water). General characterisation: initial, 95% B; gradient to 69% B in 50min; isocratic at 69% B for 5min; at 55 min, back to initial conditions 95% B and isocratic for 10min; and flow rate, 1.0ml/min. Prior to analysis, samples were prepared at a suitable concentration (either 10 or 20 mg/ml) in a stabiliser solution (acetonitrile, ascorbic acid and EDTA).

Statistical Analysis Statistical analyses were carried out using SPSS version 10.0 (SPSS Inc, Chicago, Illinois, USA) and Prism 3.0 (GraphPad Software Inc, San Diego CA, USA). Incremental changes in plasma glucose and insulin concentrations were calculated relative to fasting at all postprandial times. Incremental areas under the curve (AUC) for glucose and insulin values were calculated at 0 –150min, using the trapezoid rule. Values below fasting were treated as zero. Differences between the effects of the experimental drinks on plasma glucose and insulin concentrations and AUC values were analysed by repeated measures analysis of variance (ANOVA). Significant differences between the drinks were accepted at P⬍0.05. Contrast analysis of the incremental glucose and insulin concentrations was carried out to examine differences at specific time points and adjusted with a Bonferroni correction to compensate for multiple comparisons [21].

RESULTS This study was originally designed as a four-way randomised crossover, which included the strongest 3.0g tea drink. However, some of the subjects allocated the 3.0g drink first in the randomised sequence experienced vomiting and palpitations soon after ingestion, so that it was necessary to terminate this treatment. The results presented in this section are for the control, caffeine and 1.0g tea drinks. All of the results given in this section are means ⫾ the standard error of the mean.

Chemical Analysis of the Tea Product The polyphenol and caffeine contents of the instant tea black tea are given in Table 1. Colorimetric analysis revealed the total phenolic content of the instant tea to be high, comprising more than one third of the dry weight of the product.

473

Tea Reduces Postprandial Glycemia in Humans Table 1. Total Polyphenol, Catechin, Theaflavin and Caffeine Content of the Instant Dried Black Tea (Dry Weight Basis) Compound

Amount (mg/g)

Total polyphenols Flavan-3-ols Theaflavins Caffeine

350.0 39.0 21.0 52.0

Values for phenolic concentrations are means of triplicates; pooled coefficient of variation for all phenolic components 1.2%.

HPLC analysis showed the presence of the flavan-3-ols and theaflavins. The same analysis showed that there was a significant amount of caffeine (about 5% dry weight) in the tea product, which was the amount used for formulating the caffeinated drink (i.e, the positive control).

Study Beverages All the subjects consumed the study beverages within the allocated time of 15 min. Mean drinking times for each of the study beverages was 9.0 ⫾ 0.2 min for the control, 9.0 ⫾ 0.2 min for the caffeine and 10.0 ⫾ 0.2 min for the black tea drinks. No significant differences were found between the drinking times of the beverages using ANOVA (P ⫽ 0.211).

Plasma Glucose The fasting plasma glucose concentrations of the subjects (n ⫽ 16) were within the normal range (WHO, 1999) with a pooled mean of all fasting values of 4.4 ⫾ 0.2 mmol/l. Mean fasting glucose concentrations did not differ significantly at the beginning of each study day using ANOVA (P ⫽ 0.093). Fig. 1a-c shows the mean increments in plasma glucose concentrations relative to baseline following ingestion of the test drinks. ANOVA revealed a significant treatment-time interaction effect (P ⫽ 0.003). Contrast testing of glucose concentrations at individual time points with a Bonferroni correction showed a

significant reduction in glucose concentration after tea ingestion at 120min compared with the control (P⬍0.01) and the caffeine (P⬍0.01) drinks (Fig. 1a and 1b, respectively). No statistically significant differences were found between the tea drink and the control or caffeine drinks (ANOVA; P ⫽ 0.484) when the AUC (0 –150min) glucose values were analysed (control ⫽ 207 ⫾ 23 mmol/l.min, caffeine ⫽ 195 ⫾ 21 mmol/ l.min, tea ⫽ 176 ⫾ 21 mmol/l.min).

Plasma Insulin The pooled mean fasting plasma insulin concentrations of the subjects (n ⫽ 16) were 103 ⫾ 0.4 pmol/l. Mean fasting concentrations did not differ significantly at the beginning of each study day (ANOVA, P ⫽ 0.907). Mean increments in plasma insulin concentration relative to baseline following ingestion of each of the study drinks are presented in Fig. 2a– c. Statistical analysis using ANOVA found a significant treatment-time interaction effect after tea compared with the control and caffeine drinks (P ⫽ 0.01) (Fig. 2a and 2b respectively). Contrast testing with Bonferroni correction of insulin concentrations at individual time points showed a significant increase in insulin concentration after tea at 90min compared with the control (P⬍0.01) and caffeine (P⬍0.01) drinks, and at 150min compared with the caffeine drink only (P⬍0.05). After caffeine ingestion, significant increases in insulin concentration were observed at 30min (P⬍0.01) and 120min (P⬍0.05) compared with the tea drink. No statistically significant differences were found between the tea drink and the control or caffeine drinks (ANOVA; P ⫽ 0.625) when the AUC (0 –150min) insulin values were analysed (control ⫽ 41910 ⫾ 2628 pmol/l.min, caffeine ⫽ 42843 ⫾ 1946 pmol/l.min, tea ⫽ 44578 ⫾ 2312 pmol/l.min).

DISCUSSION It is well known that the leaves of the tea plant contain large quantities of phenolic compounds. Chemical analysis of the

Fig. 1. Incremental changes in postprandial plasma glucose concentrations (mmol/L) relative to fasting in healthy subjects following consumption of the control drink (䡲), the caffeinated drink (positive control) (Œ) and tea drink (●) (n ⫽ 16). Values are means ⫾ SEM. A statistically significant reduction in glucose concentration was found after tea vs the control and after tea vs caffeine at 120min. Mean is highlighted with ** and denotes significance at P⬍0.01 (ANOVA, contrast testing and Bonferroni correction), n ⫽ 16.

474

VO, 26, NO. 5

Tea Reduces Postprandial Glycemia in Humans

Fig. 2. Incremental changes in postprandial plasma insulin concentrations (pmol/L) relative to baseline in healthy subjects following the control drink (䡲), the caffeinated drink (positive control) (Œ) and tea drink (●) (n ⫽ 16). Values are means ⫾ SEM. A significant increase in insulin concentration was found after tea vs the control at 90min and after tea vs caffeine at 90mins and 150min. A significant increase in insulin concentration after caffeine vs tea was found at 30min and 120min. Means highlighted with * denote significant difference at P⬍0.05 and with ** denote statistical difference at P⬍0.01 (ANOVA, contrast testing and Bonferroni correction), n ⫽ 16.

instant black tea tested in the current human study confirmed that this product was rich in these compounds, with a total phenolic concentration of 350mg/g in the tea on a dry weight basis. This is similar to published values obtained for black tea leaf products [22]. Further analysis of the instant tea using HPLC showed the presence of the flavonoid compounds, flavan-3-ols and theaflavins, concentrations of which were within the range of values previously published for black tea leaf [23–25]. Previous in vitro studies have shown that polyphenolic compounds, including those found in tea, can inhibit intestinal glucose transport [10,11,13] and enhance insulin secretion of pancreatic ␤-cells [14,15], suggesting that tea has the potential to influence postprandial glycemia. The objective of our human study therefore, was to determine whether consuming black tea, in amounts that could be easily drunk at any one time, affects postprandial plasma concentrations of glucose and insulin in response to an oral glucose load. When comparing the tea drink with the control, there was no difference in the early plasma glucose responses, as seen by the similar mean incremental glucose values for both drinks at 30 and 60min. The major effect was at 120min, where there was a reduction in plasma glucose of 1.3mmol/l following the tea drink, relative to the control. This effect is both statistically and biologically significant, since a mean change of 1.3mmol/l represents 40% of the postprandial peak. The lack of any apparent effect on glucose within the first hour would suggest that, in our study at least, inhibition of glucose absorption in the small intestine did not occur. As well as its effects on glucose, the consumption of tea produced a significant increase in plasma insulin at 90min compared with the control and caffeine drinks, and at 150min compared with the caffeine drink only. Whilst it is not possible to discount impaired hepatic clearance as an explanation for these findings, previously published in vitro data exists that indicates that polyphenols found in tea enhance insulin secretion [14,15,26]. We suggest therefore that the elevated insulin concentrations seen in our study are more likely to be related to


enhanced insulin secretion rather than impaired hepatic clearance. However, it is not known whether any insulinotropic effect of tea in humans is due to direct action of its polyphenolic compounds on the pancreatic ␤-cell or through stimulation of the incretin hormones of the enteroinsular axis (EIA). Certainly, substances which stimulate GIP and GLP-1 release will enhance insulin secretion, but it is not known if tea polyphenols exert any effect on these hormones. In a recent study in healthy humans, Johnston and co-workers [13] reported that polyphenols found in coffee attenuated GIP and enhanced GLP-1 concentrations in the peripheral blood, potentially delaying glucose absorption. It is possible that the polyphenols present in tea also exert effects on GIP and GLP-1. However, it is important to consider that the polyphenols present in tea and coffee differ in molecular weight and structure and it cannot be assumed that they exert the same biological effects. The late effect of the tea drink on plasma insulin concentrations in this study may reflect the time taken for the tea polyphenols or their metabolites to be absorbed. Thus, for example, Leenen and colleagues [27] reported the appearance of catechins in human plasma 30min after the ingestion of a single dose of green or black tea with peaks occurring after 60min. This matches the late plasma insulin increase of the tea drink observed in the current study, again suggesting that the glucose-lowering effect is not related to a decrease in glucose absorption, but is more likely caused by stimulation of the pancreatic ␤-cells and/or hormones of the EIA. In addition to its phenolic constituents, tea also contains methylxanthines. The major methylxanthine in tea is caffeine (about 5% in the instant tea product), which is a potent adenosine receptor antagonist. Adenosine is an endogenous nucleoside believed to influence insulin binding and signalling as well as translocation of glucose transporters [20]. It is known that adenosine receptor antagonists can inhibit muscle glucose uptake even in the presence of insulin [28]. Consequently, caffeine is likely to influence carbohydrate metabolism via its ability to act as an adenosine receptor antagonist. However, the nature of the effects of caffeine on glucose tolerance in humans

475

Tea Reduces Postprandial Glycemia in Humans remains controversial, with different research groups reporting conflicting results [18,19]. In our study, the plasma glucose concentrations of our human volunteers were not altered following ingestion of the caffeinated drink when compared with the control of drinks. However, the plasma insulin concentrations were significantly elevated compared with the tea drink (but not the control) at 30 and 120min, although this response was reversed at 90 and 150min. These results are perhaps more consistent with the findings of Graham and colleagues [19], who reported that caffeine ingestion caused elevated insulin concentrations in human subjects following an oral glucose tolerance test without a corresponding lowering of glucose concentrations. Interestingly, although the caffeine drink in our study had a caffeine content identical to that of the 1.0g tea drink, the elevation in insulin observed at 30min, following caffeine ingestion, was not mirrored by an increase in insulin at the same time point after tea. This may reflect differences in bioavailability between different sources of caffeine. Caffeine is known to form insoluble complexes with theaflavins [29] and may therefore be less bioavailable in tea than when present in other sources. Tea is one of the most commonly consumed beverages worldwide [30]. The average intake of tea in the UK is approximately 3 cups a day [23]. This study has demonstrated that instant black tea, consumed at an amount equivalent to ⬃1.5 cups per day, can reduce the late phase of postprandial glycemia. Corresponding increases in plasma insulin concentrations, suggest that bioactive compounds in tea, most probably the phenols, may elicit an insulin-stimulating effect by direct action on pancreatic B-cell functions. However, a possible role of these compounds in influencing the insulinotropic effect of gut hormones, such as GIP and GLP-1, cannot be excluded. It is important to note also that the physiological effects seen in this study were relatively small and were achieved under test conditions. Under normal tea drinking conditions before or after food, the presence of other phenolic compounds could potentially alter, or even enhance, the effects seen in our study. It is certainly an area of research that warrants further investigation.

ACKNOWLEDGMENTS We would like to thank the subjects who participated in the study, John R Lewis (Unilever Research, Colworth House) for providing the tea and the polyphenols analysis, Dave Lincoln and Maryjo Searle, for their technical support and Peter Milligan for assistance with statistics. The tea and tea analysis were obtained as a gift from Unilever, UK. The present study was supported by a grant from the International Foundation for the Promotion of Nutrition Research and Nutrition Education, ISFE, Basel, Switzerland.

476

REFERENCES 1. Harbowy ME, Balentine DA: Tea chemistry. Crit Rev Plant Sci 16:415–480, 1997. 2. Uchida S, Ozaki M, Akashi T, Yamashita K, Niwa M, Taniyama K: Effects of (⫺)-epigallocatechin-3-O-gallate (green tea tannin) on the life span of stroke-prone spontaneously hypertensive rats. Clin Exp Pharm Phys 22:S302–S303, 1995. 3. Tijburg LB, Mattern T, Folts JD, Weisgerber UM, Katan MB: Tea flavonoids and cardiovascular disease: a review. Crit Rev Food Sci Nutr 37:771–785, 1997. 4. Hodgson JM, Puddey IB, Burke V, Watts GF, Beilin LJ: Regular ingestion of black tea improves brachial artery vasodilator function. Clin Sci 102:195–201, 2002. 5. Mitscher LA, Jung M, Shankel D, Dou JH, Steele L, Pillai SP: Chemoprotection: a review of the potential therapeutic antioxidant properties of green tea (Camellia sinensis) and certain of its constituents. Med Res Rev 17:327–365, 1997. 6. Chung FL: The prevention of lung cancer induced by a tobaccospecific carcinogen in rodents by green and black tea. Proc Soc Exp Biol Med 220:244–248, 1999. 7. Sakamoto K: Synergistic effects of thearubigin and genistein on human prostate tumor cell (PC-3) growth via cell cycle arrest. Can Lett 15:103–109, 2000. 8. Kashket S, Paolino V: Inhibition of salivary amylase by watersoluble extracts of tea. Arch Oral Biol 33:845–846, 1988. 9. Zhang J, Kashket S: Inhibition of salivary amylase by black and green teas and their effects on the intraoral hydrolysis of starch. Caries Res 32:233–238, 1998. 10. Shimizu M, Koybayashi Y, Suzuki M, Satsu H, Miyamoto Y: Regulation of intestinal glucose transport by tea catechins. Biofactors 13:61–65, 2000. 11. Kobayashi Y, Suzuki M, Satsu H, Arai S, Hara Y, Suzuki K, Miyamoto Y, Shimizu M: Green tea polyphenols inhibit the sodium glucose-dependent transporter of intestinal epithelial cells by a competitive mechanism. J Agric Food Chem 48:5618–5623, 2000. 12. Welsch CA. Lachance PA, Wasserman BP: Dietary phenolic compounds: inhibition of Na⫹-dependent D-glucose uptake in rat intestinal brush border membrane vesicles. J. Nutr 11:1698–1704, 1989. 13. Johnston KL, Clifford MN, Morgan LM: Coffee acutely modifies gastrointestinal hormone secretion and glucose tolerance in humans: glycaemic effects of chlorogenic acid and caffeine. Am J Clin Nutr 78:728–733, 2003. 14. Hii CS, Howell SL: The effects of epicatechin on rat islets of Langerhans. Diabetes 33:291–196, 1984. 15. Hii CS, Howell SL: Effects of flavonoids on insulin secretion and 45Ca2⫹ handling in rat islets of Langerhans. J Endocrinol 107:1–8, 1985. 16. Anderson RA, Polansky MM: Tea enhances insulin activity J Agric Food Chem 50:7182–7186, 2002. 17. Hosoda K, Wang MF, Liao ML, Chuang CK, Iha M, Clevidence B, Yammamot S: Antihyperglycaemic effect of oolong tea in type 2 diabetes. Diabetes Care 26:1714–1718, 2003. 18. Pizziol A, Tikhonoff V, Paleari CD, Russo E, Mazza A, Ginocchio G, Onesto C, Pavan L, Casiglia E, Pessina AC: Effects of caffeine

VO, 26, NO. 5

Tea Reduces Postprandial Glycemia in Humans

19.

20.

21.

22. 23.

24. 25.

on glucose tolerance: a placebo-controlled study. Eur J Clin Nutr 52:846–849, 1998. Graham TE, Sathasivam P, Rowland M, Marko N, Greer F, Battram D: Caffeine ingestion elevates plasma insulin response in humans during an oral glucose tolerance test. Can J Physiol Pharmacol 79:559–565, 2001. Greer F, Hudson R, Ross R, Graham T: Caffeine ingestion decreases glucose disposal during a hyperinsulinemic-euglycaemic clamp in sedentary humans. Diabetes 50:2349–2354, 2001. Bland M: Significance tests. In “An Introduction to Medical Statistics,” 2nd ed. Oxford UK: Oxford University Press, pp133–147, 1996. Balentine DA, Wiseman SA, Bouwens LCM: The chemistry of tea flavonoids. Crit Rev Clin Food Sci Nutr 37:693–704, 1997. Marks V, Walker R, Mazwell S, Forman D, Marcott: “Tea and Health - A Report on the Influence of Tea Drinking on the Nation’s Health.” London: Tea Council, 1996. USDA Database for the flavonoid content of selected foods: http:// www.nal.usda.gov/fnic/foodcomp/Data/Flav/flav.pdf 2003. Mulder TP, Rietvold AG, van Amelsvoot JM: Consumption of both black and green tea results in an increase in the excretion of


26. 27.

28.

29.

30.

hippuric acid into urine. Am J Clin Nutr 81(Suppl):256S–260S, 2005. Chakravarthy BK, Gupta, S, Gode KD: Antidiabetic effect of (⫺)-epicatechin. Lancet 2:272–273, 1982. Leenen R, Roodenburg, AJ, Tijburg LB, Wiseman SA: A single dose of tea with or without milk increases plasma antioxidant activity in humans. Eur J Clin Nutr 54:87–92, 2000. Vergauwen L, Hespel P, Richter EA: Adenosine receptors mediate synergistic stimulation of glucose uptake and transport by insulin and by contractions in rat skeletal muscle. J Clin Invest 93:974– 981, 1994. Charlton AJ, Baxter NJ, Khan ML, Davis AL, Jones DP, Lewis JR, Davis AP, Haslam E, Willianson MP: The self-association of the black tea polyphenol theaflavin and its complexation with caffeine. J Chem Soc Perkin Trans 2:317–322, 2000. Weisburger J: Tea and health: a historical perspective. Cancer Lett 114:315–317, 1997.

Received November 11, 2005; Revision accepted August 28, 2006.

477