The Effect of Brassica Vegetable Metabolism in Humans

Consumption

on

Caffeine

R.E. McDanell, L.A. Henderson, K. Russell & A.E.M. McLean Toxicology Laboratory, Department of Clinical Pharmacology, University College London, The Rayne Institute, 5 University Street, London WCIE 6JJ, UK Ten healthy volunteers were used in two studies investigating the effect of short-term Brassica consumption on caffeine metabolism. In the first study volunteers were given three Brassica-containing meals, the last one 3 h prior to caffeine administration. In the second study volunteers were given two Brassica-containing meals and then fasted

overnight before caffeine administration. In both studies the mean plasma half-life of caffeine was reduced by approximately 20% following a Brassica diet, suggesting that Brassica vegetables stimulate caffeine metabolism. When caffeine was given 3 h after the last meal, plasma caffeine concentrations over 6 h, were increased by up to 27% on the Brassica diet compared to controls. This may be due to a transient increased permeability of the intestine to caffeine, immediately following Brassica consumption. This effect was not seen in the second study where there was a 12-h period between the last meal and caffeine administration. There was large interindividual variation in the effect of the Brassica diet on caffeine metabolism. Introduction

Cytochrome P450 isoenzymes are involved in the oxidative metabolism of a wide range of both endogenous and foreign compounds. To date, 154 P450 sequences have been identifies, along with numerous metabolizing functions, both activating and inactivating. This enzyme system is inducible by both synthetic and naturally occurring compounds. Among the inducing compounds are several dietary components, including Brassica vegetables which have been shown to induce MFO activity in the liver and small intestine of experimental animals,2-4 and man, 5,6 and also to protect against chemicallyinduced tumours in experimental animals.’-9 Modification of the P450-enzyme profile by Brassica vegetables may account for this protective effect.10 Previous studies from this laboratory have looked at dietary effects on enzyme induction in experimental animal. Currently, this work has been extended to examine the effect of Brassica vegetables on human drug metabolism. Caffeine was selected as a model substrate as it is a widely consumed stimulant drug&dquo; and is selectively metabolized in man by

cytochrome P450IA2, an isoenzyme of considerable toxicological significance 12 and one of two polycyclic aromatic hydrocarbon (PAH) inducible forms of P450 found in mammalian species This choice of substrate is appropriate as the mechanism of induction by Brassica vegetables is believed to be via PAH-inducible P450 isoenzymes.1o In previous human studies investigating the induction of MFO enzymes, 5,6 volunteers were given a Brassica diet twice daily for 7 d. However, with human dietary habits, consumption of Brassica vegetables is usually more intermittant. In the present study, we therefore chose to examine the effect of short-term consumption (two or three meals) of Brassica vegetables, particularly as in earlier work in the rat2 we have shown significant induction of intestinal MFO activity within 6 h of the consumption of a single Brassica meal, returning to control levels at 24 h. In the present study we have therefore looked at caffeine metabolism in human volunteers 3 h and 15 h after a Brassica meal.

Correspondence: RE. McDanell

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015



168

Methods Volunteers Normal volunteers between 20 and 57 years of age took part in these studies. The subjects were all non-smokers and were healthy as determined by medical history, biochemical and haematological measurements. Each volunteer acted as his/her own control. Diets

Control

diet

The

control

diet

of two control meals to which 200 g of lightly steamed sprouts had been added. The following day, after an overnight fast, volunteers received 30 mg caffeine (1 g Nescafe Gold Blend in 100 ml water) and fasting continued for 2.5 h after caffeine administration. Blood samples were taken as described for the control diet. In two of the volunteers, a second control study was carried out several weeks after the Brassica diet, to give an indication of intraindividual variation.

excluded

caffeine-containing foods and beverages (tea, coffee, cocoa, caffeine-containing soft drinks and

Extraction

chocolate and also excluded known enzyme-

feine from plasma

inducing dietary components (alcohol, fried, grilled, roasted or toasted foods and Brassica vegetables).

phase octadecylsilane bonded silica columns

Brassica diet This was essentially the same as the control diet, but with Brassica vegetables added as detailed below (see ’Studies’ section).

Studies

Study I Four

male volunteers were given control diet for 24 h. At 8.00 h the following day they received breakfast (control diet) and then 3 h later were given 60 mg caffeine (2 g Nescafe Gold Blend in 200 ml water). Blood samples were taken from a heparinized indwelling catheter

immediately prior

to caffeine administra-

tion and at

10, 20, 30, 40, 50, 60, 70, 80, 90, 120 150, 180, 240, 300 and 360 min after caffeine

administration. Immediately after sampling, blood was centrifuged at 2000 rpm for 10 min and plasma stored at - 20 C. Not less than 1 week later, volunteers were given the Brassica diet for one day. This consisted of three meals of control diet to which had been added 200 g of lightly steamed cabbage. At 8.00 h on the following day volunteers received breakfast (control diet with 200 g cabbage) and then 3 h later were given 60 mg of caffeine (2 g Nescafe Gold Blend in 200 ml of water). Blood samples were taken as for the control diet.

Study

2

Caffeine analysis of samples Rapid extraction of cafwas

acheived using

reverse-

(Bond Elut C18, Jones Chromatography, LLandbradach, UK). The columns were conditioned prior to use by drawing through 2 x 1 ml methanol followed by 2 x 1 ml water, under vacuum (Vac Elut apparatus, Jones Chromatography, Landbradach, UK). After releasing the vacuum, the columns were loaded with 200 gl of the plasma sample or standard, followed by 100 til of the internal standard ((3-hydraxyethyltheophylline, Sigma, Poole, Dorset). The vacuum left to stand for

released and the columns further 2 min before washing, by drawing through 2 x 1 ml of water. The sample was then eluted with 400 gl acetone which was evaporated to dryness under nitrogen at 5 5 ° C. The eluent was then reconstituted in 200 ui of mobile phase. was

a

Chromatography Using the method of Hartley 13 separation of caffeine and its metabolites was carried out on a reverse phase analytical column (Waters Associates Nova Pak C 18 stainless steel column 3.9 cm x 15 cm, Millipore). The column was protected by a Waters Associate Guard pre-pack module containing a Bonda Pak C 18 insert. The mobile phase consisted of 1 % glacial acetic acid: methanol (88:12) which had been filtered and degassed before use. Flow rate was 1.5 ml min-’. UV detection was at 273 nm (Applied Biosystems 757 detector).

Three male and three female volun-

teers received the control diet for 2 d. On the

Standard.

third day, after an overnight fast, volunteers received 30 mg caffeine (1 g Nescafe Gold Blend in 100 ml water) and fasting continued for 2.5 h after caffeine administration. Blood samples were taken from an indwelling cannula at 10, 20,

cals, Poole, Dorset) were made up at concentrations of 0.2, 0.5, 1.0, 1.5 and 2.0 yg ml - in 3% BSA (Sigma). The stock solutions were prepared

25, 30, 35, 40, 45, 50, 60, 70, 80, 90, 120, 150, 180, 240, 300, 360 and 420 min after caffeine administration. Blood samples were centrifuged at 2000 rpm for 10 min and plasma stored at 20 C. Not less than 1 week later, volunteers received the Brassica diet for 1 d. This consisted -

Caffeine standards (Sigma Chemi-

according to the method of Hartley.’ The internal standard (3-hydroxyethyltheophylline was used at a concentration of 2 yg ml -’ I in disodium hydrogen phosphate buffer pH 4. Statistical and pharmacokinetic analysis Levels of significance were determined using a Student’s t-test. Area under the plasma/

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015

169

concentration curve (AUC) was calculated using the trapezoidal method and the elimination halflife (tll2) was obtained from the least square best fit for curve stripping. Mean plasma clearance was determined as MCR=DOSE/AUC.

theAUC, Cmax, T... and MCR values determined for the control diet and the Brassica diet

(Table 4). In the volunteers who received two control

diets, several weeks apart, the intra-individual variation of the plasma half-life of caffeine was less then 4% (data not shown).

Results

Study

1 In volunteers consuming Brassica diets, mean plasma caffeine levels were increased at all time points between 10 min and 4 h post-caffeine administration. This increase was between 11 % and 27% when compared to the same volunteers on a control diet (Figure 1). The greatest differences were seen in the early absorption phase. Because of wide interindividual variation these differences did not reach statistical significance. The mean AUC for the Brassica group was correspondingly higher than for the controls. (Table 1). In three out of four volunteers the plasma half-life of caffeine was decreased between 22% and 26% in Brassica-fed volunteers

compared

to

Discussion Caffeine is selectively metabolized in man by hepatic cytochrome P450IA2,12 while Brassica

Study 1: 60 mg of caffeine was administered as single cup of coffee (200 ml) approximately 3 h after their last meal. The subjects were non-fasting. The data given below is the mean data for the group (s.d.). Table 1 a

controls (Table 2).

Study 2 In volunteers

consuming

plasma levels of caffeine

Brassica

diets,

mean

increased at all time points between 10 min and 4 h postcaffeine administration (Figure 2). However, these mean values were heavily weighted by a single volunteer (DB) as in five of the six volunteers plasma caffeine levels were similar after both the Brassica and control diet. In the sixth volunteer (DB) plasma levels of caffeine after Brassica diet were twice those seen after a control diet (Figure 3). In five volunteers there was a decrease in the plasma half-life of caffeine of between 7% and 52% when on the Brassica diet (Table 3). There not a significant difference between was were

Figure 1 Study 1: Mean plasma caffeine levels (gg ml-’) ± s.d. vs time (min). n=4 (A-A control, .-. Brassica).

Table 2 a

1: 60 mg of caffeine was administered as cup of coffee. Half-lives of caffeine are in

Study

single

minutes.

Study 2: Mean plasma caffeine levels (jig s.d. vs time (min). n=6 (A-A control, Brassica).

Figure

2

ml-’)

±

-

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015

170

Study 2 30 mg of caffeine was administered as single cup of coffee (100 ml) approximately 12 h after their last meal. The subjects were fasting. The data below is the mean data for the whole group (s.d.). Table 4 a

Figure 3 Study 2: Plasma caffeine levels (wg ml -’ ) ± s.d. vs time (min) for subject D.B. (A-A control, ·-· Brassica).

Table 3 a

Study 2: 30 mg of caffeine was administered as

single cup of coffee (100 ml). Half-lives of

caffeine are

in minutes.

vegetables have recently been shown to induce this isoenzyme in rat liver.10 Caffeine metabolism might therefore be stimulated by consumption of these vegetables. In both the current

studies, the

mean

plasma

half-life of caffeine was reduced by approximately 20% in volunteers fed the Brassica diet, indicating a stimulatory effect of Brassica vegetables on caffeine metabolism.. A large interindividual variation in caffeine metabolism was found, while the intra-individual variation of plasma half-life (measured in two volunteers) was less than 4%. The mean plasma half-life of caffeine was reduced by the same amount in Brassica-fed volunteers, with or without an overnight fast before caffeine administration. This suggests that the stimulation of caffeine metabolism can occur rapidly after only two Brassica meals, and is maintained for up to 15 h. This is in agreement with our findings in the rat where significant induction of intestinal MFO activity is seen as early as 4 h after eating a single cabbage meal, returning to control levels at 24 h.2 A rapid, dietary-induced increase in drug metabolizing

activity is likely

to be of consequence to the metabolism and toxicity of xenobiotic compounds consumed as part of that diet. Pantuck and his co-workers have shown a similar inducing effect in man5 where a 13% decrease in the plasma half-life of antipyrene and an 11% increase in mean metabolic clearance was seen after 7 d on Brassica diet. A stimulatory effect of Brassica vegetables on antipyrine metabolism is somewhat surprising as it is poorly metabolized by PAH-inducible P450 isoenzymes. In the same studys the plasma half-life of phenacetin was not altered by the Brassica diet, even though phenacetin is a selective substrate for cytochrome P450IA2 in man. There was, however, a 34-67% decrease in the mean plasma concentration of phenacetin in Brassica-fed volunteers compared to controls, and the authors suggest that Brassica diets may enhance the metabolism of phenacetin in the gastrointestinal tract and/or during first pass through the liver. This is in agreement with previous findings in this and other laboratories where Brassica vegetables significantly induced both intestinal and hepatic cytochrome P450. It is interesting to note here that in a recent study,1O Brassica vegetables induced both cytochrome P450IA1 and cytochrome P4501A2 in the liver, but only cytochrome P450IA1 in the colon. P450 profiles were not measured in the small intestine which is usually the site of highest MFO enzyme activity in the gastrointestinal tract.

In the present study using caffeine, although the mean plasma half-life was decreased by 20% in Brassica volunteers, there was no corresponding decrease in plasma concentrations as previously described for phenacetin. In contrast, in non-fasted volunteers, where there was only 3 h between the Brassica diet and caffeine administration, plasma levels of caffeine over 6 h were appreciably higher than those seen after the control diet. These elevated plasma levels of caffeine were not seen in volunteers fasted overnight after a Brassica meal. It is possible that

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015

171

Brassica vegetables may cause a transient increase in the permeability of the intestine to caffeine, an effect which disappears after an overnight fast. A similar effect on intestinal permeability has been demonstrated with other natural components of the diet such as saponins, 14 which were shown to rapidly increase the permeability of the small intestine in vitro, inhibiting active nutrient transport and aiding the uptake of materials to which the gut would normally be impermeable. Dietary effects on intestinal permeability and cytochrome P450 activity, both have obvious toxicological significance. The human diet contains a heterogeneous mixture of nutrient and non-nutrient compounds, many of which are metabolized via cytochrome P450. Brassica, and similarly inducing diets may alter both the absorption and metabolism of these components and ultimately affect their toxicity. Evidence from animal experiments suggest that the consumption of Brassica vegetables protects against chemically-induced tumours in several organs. 7-9 Similarly, the consumption of these vegetables has been correlated with a decreased risk for cancer of the colon, rectum and lung in man. IS,16 Indolic compounds in these vegetables are thought to modify the carcinogenic effect through their induction of cytochrome P450IA in the colon and cytochrome P450IA1 and IA2

in the liver. 10,17 However, cytochrome P450IA2 has also been implicated in carcinogen activa-

tion, notably N-oxidation of heterocyclic amines carcinogenic/mutagenic forms. 22 It seems likely that dietary inducers are in-

to their

volved in both the activation and inactivation of many xenobiotic compounds to which we are exposed, and that while the dietary induction of P450 has been recognized in man and experimental animals for some time, we still have only a limited understanding of its specificity and toxicological significance. There may be a fine balance between activating and inactivating pathways of metabolism so that any alteration in enzyme activity may upset this balance, with unpredictable consequences. Consequently, more research is needed in this field to provide a better understanding of the dietary modification of toxic and carcinogenic processes.

Acknowledgements The authors gratefully acknowledge the contribution to this study of the following; Mr Barry Graham for help with statistical analysis, all our volunteers, and our colleagues at The AFRC Institute of Food Research, Norwich for provision and analysis of Brassica samples. This project was supported by the Ministry of Agriculture, Fisheries and Food.

References 1 Nebert DW, Nelson DR, Coon MJ et al. The P450 superfamily: update on new sequences, gene mapping and recommended nomenclature. DNA Cell Biology 1991; 10: 1-14.

2 McDanell RE, McLean AEM, Hanley AB, Heaney RK & Fenwick GR. The effect of feeding Brassica vegetables and intact glucosinolates on mixed-function-oxidase activity in the liver and intestines of rats. Food and Chemical Toxicology 1989; 27: 289-93. 3 Pantuck EJ, Hsiao KO, Loub WD et al. Stimulatory effect of vegetables on intestinal drug metabolism in the rat. Journal of Pharmacology and Experimental Therapeutics 1976; 198: 278-83. 4 Wattenberg LW. Studies of polycyclic hydrocarbon hydroxylases of the intestine possibly related to cancer. Cancer (New York) 1971; 28: 99. WA al. Stimulatory 5 Pantuck EJ, Pantuck CB, Garland et effect of brussel sprouts and cabbage on human drug metabolism. Clinical and Pharmacological Therapeutics 1979; 25: 88-95. 6 Pantuck EJ, Pantuck CB, Anderson et KE al. Effect of brussel sprouts and cabbage on drug conjugation. Clinical and Pharmacological Therapeutics 1984; 35: 161-9. 7 Stoewsand GS, Babish JB & Wimberley HC. Inhibition of hepatic toxicities from polybrominated biphenyls and aflatoxin B, in rats fed cauliflower. Journal of Environmental and Pathological Toxicology 1978; 2: 399-406. 8 Boyd JN, Babish JB & Stoewsand. Modification by beet and cabbage diets of aflatoxin B -induced rat plasma 1

alfa-foetoprotein elevation, hepatic tumorogenesis and mutagenicity of urine. Food and Chemical Toxicology

1982; 20: 47-52. Wattenberg LW, Schafer HW & Davis DW. Inhibition of mammary tumor formation by brocolli and cabbage. Proceedings of the American Association of Cancer Research 1989; 30: 181. 10 Vang O, Jensen MB & Autrup H. Induction of cytP450IA1 in rat colon and liver by indole-3-carbinol and 5,6benzoflavone. Carcinogenesis 1990; 11: 1259-63. 11 Benowitz NL. Clinical pharmacology of caffeine. Annual Reviews in Medicine 1990; 41: 277-88. 12 Butler MA, Iwasaki M, Guengerich FP & Kadlubar FF. Human cytochrome P450 PA (P450IA2) the phenacetin-odeethylase, is primarily responsible for the hepatic 3demethylation of caffeine and N-oxidation of carcinogenic arylamines. Proceedings of the National Academy of Science USA 1989; : 7696-700. 86 13 Hartley R, Smith IJ & Cookman JR. Improved high performance liquid chromatographic method for the si9

multaneous

determination

of caffeine

and

its

N-

demethylated metabolites in plasma using solid phase extraction. Journal ofChromatography 1985; 342: 105-17. Johnson IT, Gee JM, Price K, Curl C & Fenwick GR. Influence of saponins on gut permeability and active nutrient transport in vitro Journal of Nutrition 1986; 116: 2270-7. 15 Graham S, Dayal H, Swanson M, Mittelman A & Wilkinson G. Diet in the epidemiology of cancer of the colon and rectum. Journal of the National Cancer Institute 1978; 61: 709-14. 16 Marchand LL, Yoshizawa CN, Kolonel LN, Hankin JH & Goodman MT. Vegetable consumption and lung cancer 14

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015

172

risk. A Journal 64. 17

population-based case-control study in Hawaii. of the National Cancer Institute 1989; 81: 1158-

McDanell

Fenwick GR. Chemical and biological properties of indole glucosinolates. (glucobrassicins): a review. Food and Chemical Toxicology 1988; 26: 59-70.

RE, McLean AEM, Hanley AB, Heaney RK &

Downloaded from het.sagepub.com at UNIV CALIFORNIA SAN DIEGO on March 12, 2015

The effect of brassica vegetable consumption on caffeine metabolism in humans.

Ten healthy volunteers were used in two studies investigating the effect of short-term Brassica consumption on caffeine metabolism. In the first study...
432KB Sizes 0 Downloads 0 Views