JOURNAL OF APPLIED TOXICOLOGY. VOL. 12(3), 191-192 (1992)

Tannin Content of Tea and Coffee H. Savolainen Institute of Occupational Health Sciences, University of Lausanne, Rue Bugnon 19, CH-1005 Lausanne, Switzerland

Key words: tannin; tannic acid; tea; green coffee; protein precipitation.

The tannin content of over-the-counter Indian tea, of green coffee beans and of the roasted coffee beans prepared from the same green beans was determined with a radial diffusion - protein precipitation technique and with a spectrophotometric method. The green beans contained 6.6 f 0.6 mg g-' weight tannic acid equivalents as found by protein precipitation ( n = 5, f SD) or 6.8 2 2.3 mg g-' by spectrophotometry. The same figures for roasted beans were 18 2 1.7 and 17 rt 2.7 mg g-', respectively. Tea contained 37 2 2.6 mg g-' weight tannic acid equivalents as analysed by spectrophotometry and 24 2 2.8 mg g-' by the protein precipitation technique. The latter finding may show that the biological reactivity of tannins is variable, although no major changes in the tannin-precipitated albumin occurred as shown by electrophoretic analysis. Both methods provide an easy analysis of the reportedly carcinogenic plant tannins.

INTRODUCTION

was roasted by the supplier immediately on purchase. Two grams of powered tea, green coffee or roasted coffee were extracted with 10 ml of a methanol-water mixture ( l : l , v/v). The extracts were divided into two aliquots and analysed for tannins using a spectrophotometric method"' and using the bovine serum albumin precipitation technique on agarose plate^.^ Tannic acid was used as a standard, and the results were expressed as tannic acid equivalents. Tannic acid was obtained from Aldrich, and it was used with special care because it is a carcinogen." The detection limit for the spectrophotometric method was 0.1 pg ml-', while at least 50 pg of tannin per sample were required in the precipitation method. After having read the agarose plates on a black background under a daylight lamp, the precipitates were cut out of the gel with a disposable knife and the agarose was homogenized in 2 ml of 0.1 mM phosphate buffer containing 0.2% (w/v) sodium dodecyl sulphate (SDS). This preparation was incubated overnight at 4"C,its protein concentration was determined with the biuret method and they were analysed by electrophoresis on 12% polyacrylamide gel plates using the abovementioned phosphate buffer. The gels were stained with Coomassie blue and, after destaining, were dried and photographed. All data were statistically evaluated with variance analysis.

Tannins are a complex family of plant polyphenols' with considerable biological activity. They may activate platelets,* they are implicated in the epithelial injury of bronchi by cotton dust' and the class includes notorious carcinogens4 that may play an important role in the woodworkers' nasal sinus ~ a n c e r . ~ Because of their ubiquitous presence in foodstuffs, animals have developed defense mechanisms to neutralize their toxicity. Salivary proline-rich proteins seem to interact with tannins to render them inactive,6 and this biological activity of tannins towards protein has been exploited in an elegant quantitative technique in which a methanol extract containing tannins IS allowed to diffuse out of a central well into agarose gel containing albumin. The radial diameter of the albumin precipitate depends on the initial tannin ~ o n t e n t In .~ the original work the reading of the results was directly visual, whereas in later techniques the precipitates have been detected by analysis of alkaline phosphatase activity after having allowed the enzyme to react with tannins in the geL8 Tea and green coffee dust are known lung irritants in occupational exposure and give rise to rhinitis or bronchial hyperreactivity .9 If the tannins are implicated, their mechanism may be comparable to the mechanism of cotton dust-induced bronchial d i ~ e a s e , ~ and therefore the tannins could have importance when RESULTS AND DISCUSSION workplaces are contaminated with dust of plant origin. In this paper the tannin content of two products Tea contained more tannic acid equivalents than green taken daily was measured using two different analytical or roasted coffee (Table 1). However, spectrophototechniques. metric analysis gave higher tea tannin contents than that achieved by the albumin precipitation (37 2 2.6 vs. 24 5 2.8 mg g-' weight ( n = 5 , k SD, P < 0.01), MATERIALS AND METHODS respectively). For both types of coffee, the results by the two methods did not deviate statistically (Table 1). The roasted coffee seemed to contain more tannins Indian tea and green coffee samples were purchased. than the green beans on a weight basis. This probably The green coffee beans were divided into two lots. does not represent a true increase in their concentration The first portion was retained as such, and the other but a loss of ethereal oils and humidity during roasting. 0260-437)(/92/03019 1-02$05 .OO @ 1992 by John Wiley & Sons. Ltd.

Received 25 September 1991 Accepted I October 1991

192

H. SAVOLAINEN

Table 1. Tea and coffee tannin content" Method

Tea

Green coffee

Roasted coffee

Spectrophotometric Precipitation

37 t 2.6b 24 t 2.8

6.8 t 2.3 6.6 -+ 0.6

17 t- 2.7 18 t- 1.7

"The figures give the mean of five analyses ( m g tannic acid equivalents g ' weight) ? SD. I, Differs from all other values at P < 0.01.

The mean dissolved albumin precipitate protein content cut from the agarose gels was 20 ? 6 mg mlF' (n=20, ? SD). This rather large variation was caused by the fact that the precipitated albumin area on the gel varied, and the agarose samples were taken by cutting at the precipitate border. The albumin band prevailed in all electrophoretograms, and its migration was grossly similar to that extracted from the blank agarose gel area without visible precipitates (Fig. 1). There might have been a minor shift towards slower mobility of precipitated albumin by green coffee tannins as compared to that by roasted coffee (Fig. 1). Be it as it may, the electrophoretic analysis showed that reaction products were rather similar in all cases, and thus it may be accepted that bovine serum albumin can be used as the indicator protein in precipitation analysis. There is, however, an important discrepancy in the tannin content of the Indian tea by the two methods. A speculative explanation could be that about onethird of the tea tannins do not have a similar affinity for albumin as, for instance, the coffee bean tannins. As a practical consequence, the tea tannin content could be underestimated by relying on the precipitation method only. Tea tannins are also implicated in other than only gastrointestinal malignancies." This would show that their inactivation by salivary protein could be reversible or not quantitative. Another confounding factor in these type of epidemiological comparisons is that the tea-consuming populations are often exposed to other plant tannin sources. For example, cotton fibres before

Figure 1. Electrophoresis of molecular weight standard mixture (S) of albumin precipitated by tannic acid (1) or by tannins from roasted coffee (2), from green coffee (3) and from tea (4) as compared to albumin from background (5). (396 2 145 pg protein were applied on the gels.) The albumin band is located just above the ovalbumin standard (third from origin at the top of the gel, molecular weight 49000). The minor bands near the origin of the electrophoresis (top) represent impurities in the commercial bovine serum albumin; they are probably not reaction products as they are found also i n the background albumin pattern (5),which is comparable t o other patterns. Thus, the precipitation of albumin by tannins may not profoundly change its electrophoretic behaviour.

spinning contained 1.6 t 0.13 mg g-I weight tannic acid equivalents in our tests (unpublished observations). The salivary proline-rich protein seems to contain tens of polypeptides,I3 and individual variability is by no means excluded. Perhaps the next logical step would be the use of this salivary protein mixture, or one of its speclfic components, in the precipitation analysis of tannins. Still, in its present form precipitation analysis could be a valuable tool in laboratories with limited resources, as costly instruments or maintenance are not needed. Acknowledgements The author thanks Mrs Nicole Chavannes for hcr excellent technical assist itncc.

REFERENCES

1. M. A. Rousselle, M. H. Elissalde and L. H. Domelsmith, Effect of alkali treatment on physiological activity of cotton condensed tannin. Br. J. lnd. Med. 47, 698-703 (1990). 2. S. L. Hempel, C. L. Wheatley and M. S. Rohrbach, Protein phosphorylation during tannin-mediated activation of human platelets. J. Lab. Clin. Med. 111, 299-306 (1988). 3. G. H. Ayars, L. C. Altman, C. E. O'Neill, B. T. Butcher and E. Y. Chi, Cotton dust-mediated lung epithelial injury. J. Clin. Invest. 78, 1579-1588 (1986). 4. K. S. Kirby, Induction of tumours by tannin extracts. Br. J. Cancer 14, 147-150 (1960). 5. P. Catilina, Woodworking: occupational hazards and prevention. Arch. Ma/. Prof. 42, 253-285 (1981). 6. H. Meshansho, L. G . Butler and D. M. Carlson, Dietary tannins and salivary proline-rich proteins: interactions, induction, and defense mechanisms. Ann. Rev. Nutr. 7, 423- 440 ( 1987). 7. A. E. Hagerman, Radial diffusion method for determining tannin i n plant extracts. J. Chem. Ecol. 13, 437-449 (1987). 8. Y. Ittah, Titration of tannin via alkaline phosphatase activity. Anal. Biochem. 192, 277-280 (1991).

9. E. Zuskin, B. Kanceljak, T. J. Witek and E. H. Schachter, Acute effects of herbal tea dust extracts on lung function. Chest 96, 1327-1331 (1989). 10. D.-W. Lau, S.-F. Luk and H.-L. Huang, Spectrophotometric determination of tannins in tea and beer samples with iron (Ill) and 1,lO-phenanthroline as reagents. Analysf 114, 631-633 (1989). 11. B. Korpassy, and M. Mosonyi, The carcinogenic activity of tannic acid: liver tumours induced in rats by prolonged subcutaneous administration of tannic acid solutions. Br. J. Cancer 4, 41 1-420 (1960). 12. F. J. Tewes, L. C. Koo, T. J. Meigsen and R. Rylander, Lung cancer risk and mutagenicity of tea. Environ. Res. 52, 23-33 (1990). 13. D. L. Kauffman, A. Bennick, M. Blum and P. J. Keller, Basic proline-rich proteins from human parotid saliva: relationship of the covalent structures of ten proteins from a single individual. Biochemistry 30, 3351-3356 (19911.

Tannin content of tea and coffee.

The tannin content of over-the-counter Indian tea, of green coffee beans and of the roasted coffee beans prepared from the same green beans was determ...
308KB Sizes 0 Downloads 0 Views