Mutagenicity and antimutagenicity studies of tannic acid and its related compounds

Food and Chemical Toxicology 38 (2000) 1±5

www.elsevier.com/locate/foodchemtox

Research Section

Mutagenicity and Antimutagenicity Studies of Tannic Acid and its Related Compounds SSU-CHING CHEN1 and KING-THOM CHUNG2* Department of Medical Technology, Fooyin Institute of Technology, Kaosiung, Taiwan 830, ROC and 2Department of Microbiology and Molecular Cell Sciences, The University of Memphis, Memphis, TN 38152, USA 1

(Accepted 25 May 1999) AbstractÐTannic acid and its hydrolysed products such as ellagic acid, gallic acid and propyl gallate were tested for mutagenicities using Ames Salmonella tester strains TA98 and TA100. Also, the antimutagenic activities of these compounds against a number of direct mutagens including 2-nitro¯uorene (2-NF), 4,4'dinitro-2-biphenylamine, 1-nitropyrene, 1,3-dinitropyrene, 2-nitro-p-phenylenediamine, 3-nitro-o-phenylenediamine, 4-nitro-o-phenylenediamine were tested. None of these tannic acid compounds was mutagenic. They also failed to show antimutagenic activity towards the tested direct mutagens. However, tannic acid at non-growth inhibitory concentrations reduced the revertant numbers of TA98 in the presence of S9 mix when benzidine, 3,3'-4,4'-tetraminobiphenyl, 4-aminobiphenyl, and N,N-N',N'-tetramethylbenzidine were

#

used as the mutagens. These results suggest that tannic acid, but not its hydrolytic products, a€ects the metabolic activation of these mutagens.

Keywords:

2000 Elsevier Science Ltd. All rights reserved

mutagenicity; antimutagenicity; Ames test; tannic acid; mutagens.

Abbreviations: DMSO = dimethyl sulfoxide; MNNG = N-methyl-N-nitro-N'-nitrosoguanidine; 2-NF = 2-nitro¯uorene.

INTRODUCTION

Vegetable tannins are water-soluble phenolic compounds having a molecular weight between 500 and 3000 Daltons. Vegetable tannins can be classi®ed into hydrolysable and condensed tannins. Hydrolysable tannins contain either gallotannins or ellagitannins. Gallotannins yield glucose and gallic acid on hydrolysis by acids, bases or certain enzymes. Ellagitannins contain one or more hydroxydiphenoyl residues which are linked to glucose as a diester in addition to gallic acid. On hydrolysis, the hydroxydiphenoyl residue undergoes lactonization to produce ellagic acid. Condensed tannins are the polymerized products of ¯avan-3-ols and ¯avan-3,4diols or a mixture of the two. The polymers, referred to as ``¯avolans'' are popularly called condensed tannins (Chung et al., 1998). Tannins are present in almost every food plant, particularly bananas, grapes, raisins, sorghum, spinach, red wine, persimmons, co€ee, chocolate and tea (Chung *Corresponding author. Tel: (901) 678-4458; Fax: (901) 678-4457; e-mail: [email protected]. 0278-6915/00/$ - see front matter PII S0278-6915(99)00114-3

et al., 1998; Jones 1992). It is estimated that people in the United States ingest each day 1 g of tannic acid, the most simple form of hydrolysable tannins (Sanyal et al., 1997). Most of the dietary tannins are water hydrolysable tannins which are either gallotannins or ellagitannin (Wu-Yuan et al., 1988). Previously, we discovered that tannic acid had antimicrobial properties, which are associated with the ester linkage between gallic acid and other sugar or alcohol groups (Chung et al., 1993, 1995). Tannic acid is considered to be a generally regarded as safe (GRAS) food additive, which may be present from 10 to 400 ppm, depending on the type of food to which it is added. Propyl gallate is also a certi®ed food additive and is often added as an antioxidant (Sherwin, 1990). There have been numerous reports that some tannin components such as gallic acid or ellagic acid can inhibit the mutagenicity of certain mutagens (Huang et al., 1983, 1985; Mitscher et al., 1992; Shelef and Chin 1980; Smart et al., 1986; Steele et al., 1985; Stich 1991; Teel 1986; Terwel and Van der Hoeven 1985; Wood et al., 1982). However, in evaluating the anti-

# 2000 Elsevier Science Ltd. All rights reserved. Printed in Great Britain

2

Ssu-Ching Chen and King-Thom Chung

mutagenic activity of many plant phenolics, inhibitory e€ect on the growth of test organisms have been reported (De Flora et al., 1992; Mendelsoh 1992; Stich et al., 1982). As tannins are endogenous food components, their potential toxic e€ects are crucial for evaluating their impact for human health. The present paper reports the testing of the mutagenicities of tannic acid and its hydrolysed product using the Ames Salmonella/microsome mutagenicity assay at non-growth inhibitory concentrations. The antimutagenic activities of these compounds against a number of known compounds were also tested.

purchased from Aldrich Chemical Company, Inc. (Milwaukee, WI, USA). All solutions were freshly prepared by dissolving the chemicals in DMSO or in phosphate bu€er and were kept in the dark. Bacterial cultures Salmonella typhimurium TA98 and TA100 were kindly provided by Dr B. N. Ames, Department of Biochemistry, University of California, Berkeley, California. Fresh overnight tester strain cultures, to which DMSO was added as a cryoprotective agent, were stored at ÿ808C. Tester strains were checked routinely to con®rm genetic features using the procedure outlined by Maron and Ames (1983).

MATERIALS AND METHODS

Metabolic activation system

Chemicals Tannic acid (CAS No. 1401-55-4), gallic acid (149-91-7), ellagic acid (476-66-4), propyl gallate (121-79-9), 2-nitro-p-phenylenediamine (5307-14-2), 3-nitro-o-phenylenediamine (3694-52-8), 4-nitro-ophenylenediamine (99-56-9), benzidine (92-87-5), 4aminobiphenyl (92-67-1), 3,3'-4,4'-tetraminobiphenyl (91-95-2) and dimethyl sulfoxide (DMSO) (67-68-5) were purchased from Sigma (St Louis, MO, USA). 1-Nitropyrene (5522-43-0), 2-nitro¯uorene (2-NF) (607-57-8), 4,4'-dinitro-2-biphenylamine (51787-75-8) 1,3-dinitropyrene (75321-20-9), N,N-N',N'-tetramethylbenzidine (366-29-0) and N-methyl-N-nitroN'-nitrosoguanidine (MNNG) (70-25-7) were

S9 mix (Aroclor 1254-induced, Sprague±Dawley male rat liver in 0.154 M KCl solution) was purchased from Molecular Toxicology, Inc. (Annapolis, MD, USA) and stored at ÿ708C. Before use, the S9 mix was ®ltered through a 0.45 mm Nalgene disposable ®lter (Nalge Co., Rochester, NY, USA). Mutagenicity test Mutagenicity tests were performed using standard preincubation procedures and in the presence or absence of the liver S9 mix (Maron and Ames, 1983).

Table 1. Mutagenicity of tannic acid and related compounds to Salmonella typhimurium TA98 and TA100 Revertant colonies per plate (mean

2 SD)

TA98 Compound

Dose (per plate)

DMSO 2-Nitro¯uorene MNNG 2-Amino¯uorene Tannic acid (mg)

Propyl gallate (mg)

Ellagic acid (mg)

Gallic acid (mg)

a

Values are means of revertants/plate Not tested.

b

100 ml 4 mg 1 mg 10 mg 375 750 1500 3000 125 250 500 1000 125 250 500 1000 375 750 1500 3000

ÿS9

23 349 2 20 20

± ±

23 22 21 24 12 2 2 13 2 3 15 2 5 15 2 4 16 2 4 15 2 1 17 2 5 15 2 4 18 2 4 11 2 3 12 2 1 16 2 5

12 10 11 16

a

TA100 +S9 20

24

ÿS9 105

b

± ±

2 30 29 2 1 19 2 7 16 2 9 16 2 6 20 2 3 17 2 6 12 2 5 12 2 1 17 2 5 18 2 2 17 2 3 13 2 4 26 2 2 16 2 7 19 2 4 15 2 9

2112

4625

2 SD of quadruplicate runs, each with three plates/dose.

2 28

103

21

±

2 142 ±

26 2 13 29 2 17 85 2 3 89 2 8 96 2 8 96 2 23 96 2 8 87 2 9 124 2 28 103 2 33 91 2 5 126 2 8 94 2 6 86 2 12

106 107 82 85

+S9

2 135 107 2 11 98 2 11 128 2 20 120 2 12 76 2 3 71 2 1 75 2 6 121 2 14 82 2 2 116 2 4 114 2 19 98 2 3 78 2 7 103 2 7 104 2 14 80 2 2

2110

Mutagenicity and antimutagenicity studies of tannic acid

3

Table 2. E€ect of tannic acid and related compounds on the mutagenicity of nitro group-containing compounds in Salmonella typhimurium TA98 Number of His+ revertants induced by mutagens/plate* Compound Tannic acid Propyl gallate

Ellagic acid

Gallic acid

Dose (mmol)

2-Nitro¯uorene (1 mg)

2 16 2 10 199 2 10 192 2 31 223 2 21 242 2 42 246 2 21 251 2 36 242 2 42 246 2 35 245 2 4

0 0.2

265 263

0 0.1 0.2 0 0.1 0.2 0 0.1 0.2

2-Nitro-p-phenylenediamine (30 mg)

3-Nitro-o-phenylenediamine (300 mg)

2 27 2 23 436 2 38 406 2 16 383 2 33 551 2 49 504 2 79 540 2 23 551 2 49 539 2 10 470 2 30

4-Nitro-o-phenylenediamine (10 mg)

2 13 24 166 2 24 155 2 7 167 2 30 179 2 12 156 2 4 171 2 2 179 2 12 169 2 17 160 2 17

525 540

2 24 2 54 256 2 37 253 2 23 210 2 31 281 2 33 261 2 49 267 2 38 263 2 38 285 2 26 300 2 37

184 165

265 252

Number of His+ revertants induced by mutagens/platea 4,4'-Dinitro-2-biphenylamine Compound

Dose (mmol)

Tannic acid

Propyl gallate

Ellagic acid

Gallic acid

a

0 0.1 0.2

0 0.1 0.2 0 0.1 0.2

Values are means of revertants/plate Not tested.

b

1,3-Dinitropyrene

(10 mg)

(0.33 mg)

(1 mg)

(0.0067 mg)

(0.01 mg)

2 2 459 2 38 477 2 79 400 2 40 459 2 38 495 2 63 422 2 13 459 2 38 549 2 85 438 2 3

2 57 2 73 2 21 465 2 57 448 2 16 426 2 37 465 2 57 477 2 81 418 2 17 465 2 57 403 2 18 402 2 14

2 207 1628 2 202

24 ± 110 2 14 86 2 11 ± 81 2 9 86 2 11 ± 80 2 28 86 2 11 ± 82 2 4

2 10 ± 689 2 129

476 34 ±b 448 6

0 0.1 0.2

1-Nitropyrene

465 471 465

1171

2 SD of quadruplicate runs, each with three plates/dose.

Concurrently, 2-amino¯uorene, 2-nitro¯uorene and MNNG were included in all assays. Except under special conditions, 0.5 ml of S9 mix were used. The same amount of DMSO (25 ml) was delivered to each plate. All operations were conducted under yellow light to avoid photo-oxidation of the compounds. Antimutagenic test The antimutagenic tests were performed similar to the standard plate incorporation and preincubation procedures of Maron and Ames (1983), except the antimutagens were also added and preincubated for 30 min before plating. RESULTS

No tannic acid or its related compounds was mutagenic towards Salmonella tester strains TA98 and TA100 in the absence or presence of S9 mix,

±

± ± ± ± ± ± ± ± ±

93

615

± ± ± ± ± ± ± ± ±

and the results are shown in Table 1. The antimutagenicity testing of these compounds against nitrogroup containing mutagens including 4,4'-dinitro-2biphenylamine, 1-nitropyrene, 1,3-dinitropyrene, 2nitro¯uorene, 2-nitro-p-phenylenediamine, 3-nitro-ophenylenediamine and 4-nitro-o-phenylenediamine to TA98 in the plate incorporation assay in the absence of metabolic activation showed that they did not a€ect the metabolic activity of these mutagens (Table 2). However, when these compounds were tested against those mutagens that required metabolic activation, namely benzidine, 4-aminobiphenyl, 3,3'-4,4'-tetraaminobiphenyl, and N,NN',N'-tetramethylbenzidine, di€erent results were obtained. The number of histidine revertants was signi®cantly lower in the presence of tannic acid at its non-growth inhibitory concentrations (<0.2 mmol/plate) than without the tannic acid (Table 3). Tannic acid at higher concentration was inhibitory to the growth of the tester strain. Propyl

4

Ssu-Ching Chen and King-Thom Chung

gallate, ellagic acid and gallic acid showed no antimutagenic activities against all mutagens tested. DISCUSSION

Tannic acid is present in almost every food plant (Chung et al., 1998). There were reports indicating that some tannin components were carcinogenic (Oterdoom, 1985). For example, in the Caribbean, the heavy use of tannin-rich herb teas is thought to contribute to high levels of oesophageal cancer (Kapadia et al., 1983). However, Onodera et al. (1994) showed that tannic acid did not induce any tumours in F344 rats. Nagahushan et al. (1991) also reported that tannic acid was not mutagenic to Salmonella typhimurium tester strains TA98, TA100, TA 1535 and TA1538 with or without metabolic activation. In the present studies, tannic acid and its hydrolytic products are not mutagenic in the Ames Salmonella/microsome mutagenicity assay system. There were also reports to indicate that some tannin components were anticarcinogenic (Carr, 1985; Teel, 1986). Antimutagenic properties of di€erent tannin preparations had also been reported (De Flora et al., 1992; Huang et al., 1983, 1985; Mitscher et al., 1992; Shelef and Chin, 1980). Our results indicated that tannic acid and its hydrolytic products were not antimutagenic against these tested direct mutagens such as 2-nitro¯uorene, 2nitro-p-phenylenediamine, 3-nitro-o-phenylenediamine, 4-nitro-o-phenylenediamine, 4,4'-dinitro-2biphenylamine, 1-nitropyrene and 1,3-dinitropyrene. On the other hand, tannic acid at concentrations sublethal to Salmonella tester strains, but not its hydrolytic products such as gallic acid and ellagic acid, could decrease the frequency of histidine reversion caused by benzidine, 4-aminobiphenyl,

3,3'-4,4'-tetraaminobiphenyl and N,N-N',N'-tetramethylbenzidine requiring the metabolic activation. It is possible that tannic acid is inhibitory to the metabolic activation of these mutagens. This possibility needs further study. Investigating the antimutagenic potentials of tannic acid is handicapped by their toxicity to the Salmonella tester strains, especially when the concentration is high. In this paper, we tested the antimutagenic activity of tannic acid at concentrations that were non-growth inhibitory to the tester strains. Gallic acid, ellagic acid and propyl gallate were not antimutagenic at these low concentrations. The possibility should not be excluded that antimutagenic activity could be found at higher concentrations. Gichner et al. (1986, 1987) reported that tannic acid and gallic acid inhibited the mutagenicity of direct mutagen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) in Arabidopsis thaliana. Much higher concentrations were used than in the present study. Kado et al. (1983) applied a liquid incubation procedure for mutagenic testing of human urine samples. A concentrated bacterial cell suspensions (approx. 109) was used, prolonging the incubation at 378C for 90 min, which then proceeded according to the standard Ames tested protocol (Maron and Ames, 1983). They found that this liquid incubation procedure was more sensitive than the standard plate incorporation test for detecting mutagens (Kado et al., 1983). We employed the standard mutagenicity procedure and found that tannic acid and its related compounds were not antimutagenic against a number of direct mutagens. Whether the hydrolysed products of tannic acid would be antimutagenic towards direct mutagens if a more sensitive method to be employed is still in question.

Table 3. E€ect of tannic acid and related compounds on the mutagenicity of some aromatic amines requiring S9 in Salmonella typhimurium TA98 Number of His+ revertants induced by mutagens/plate* Compound Tannic acid

Propyl gallate

Ellagic acid

Gallic acid

Dose (mmol) 0 0.1 0.2 0 0.1 0.2 0 0.1 0.2 0 0.1 0.2

2

Benzidine (300 mg)

4-Aminobiphenyl (10 mg)

26 2 1** 2 2** 115 2 23 102 2 10 116 2 31 86 2 8 106 2 26 95 2 9 115 2 23 109 2 20 100 2 8 94 52 39

a

b b

a a a

a

a

a

a a

a

29 2 14 2 6** 134 2 19 109 2 7 149 2 23 134 2 19 135 2 35 137 2 16 97 2 15 89 2 21 86 2 10 93 66 42

c

c

d

c

c

c c c c c c c

3,3'-4,4'-Tetraaminobiphenyl (30 mg)

2 23 2 15** 2 7** 221 2 34 228 2 30 237 2 50 275 2 45 273 2 30 274 2 11 221 2 34 211 2 12 275 2 64 270 77 20

e

f

e e e e e e e e e

f

N,N-N',N'-Tetramethylbenzidine (10 mg)

28 2 9* 2 4* 113 2 27 95 2 12 100 2 2 113 2 27 106 2 8 108 2 18 113 2 27 114 2 24 115 2 21 88 57 37

g

h h g g

g

g

g

g g g g

*The data are the mean SD from two independent runs, each with three plate/dose. **Percentage of inhibition>30% (=100%-[(revertants/plate with compound + inhibitorÿspontaneous revertants)/(revertants/plate with compound aloneÿspontaneous revertants)] 2 100. Means with di€erent supercripts within each vertical column group were signi®cant di€erent from the control.

Mutagenicity and antimutagenicity studies of tannic acid

Tannins are a diverse group of compounds. Only a few related compounds were tested in the present study. There are other structurally complex forms of tannins such as catechin, epicatechin, epigallocatechin, epicatechin-3-gallate, epigallocatechin-3-ogallate, agrimonin, geranin, davurciin T1, leucocyanidin, leucodelphinidin, guibourtacacidin, (+)-leucorobinetinidin, (ÿ)-melacacidin and teracacidin, which are present in food and medicinal plants (Chung et al., 1998; Okuda et al., 1992). Whether they are mutagenic, antimutagenic or e€ective in exerting other genotoxic potentials a€ecting the human health still awaits further study. REFERENCES

Carr C. J. (1985) Food and drug interactions. In Xenobiotics in Foods and Feeds, ed. J. W. Finley and D. E. Schwass, pp. 221±228. ACS Symposium 234, American Chemical Society, Washington, DC. Chung K-T., Stevens S. E., Jr, Lin W. F. and Wei C. I. (1993) Growth inhibition of selected food-born bacteria by tannic acid, propyl gallate and related compounds. Letters in Applied Microbiology 17, 29±32. Chung K-T., Wong T. Y., Wei C. I., Huang Y. W. and Lin Y. (1998) Tannins and human health: a review. CRC Critical Reviews in Food and Nutrition 38, 421± 464. Chung K-T., Zhao G., Stevens S. E., Jr and Simco B. A. (1995) Growth inhibition of selected ®sh pathogens by tannic acid and related compounds. Journal of Aquatic Animal Health 7, 46±49. De Flora S., Camoirano A., D'Agostini F. and Balansky R. (1992) Modulation of the mutagenic response in prokaryotes. Mutation Research 267, 183±192. Gichner T., Pospisil F., Volkeova V., Volke J. and Veleminsky J. (1986) Gallic and tannic acids inhibit the mutagenicity of a direct acting mutagen N-methyl-N'nitro-N-nitrosoguanidine, but not of promutagen dimethylnitrosamine in Arabidopsis thaliana. Biologia Plantarum (Praha) 28, 386±390. Gichner T., Pospisil F., Veleminsky J., Volkeova V. and Volke J. (1987) Two types of antimutagenic e€ects of gallic and tannic acids towards N-nitroso-compoundinduced mutagenicity in the Ames Salmonella assay. Folia Microbiologia 32, 55±62. Huang M. T., Chang R. L., Wood A. W., Newmark H. L., Sayer J. M., Yagi H., Jerina D. M. and Conney A. H. (1985) Inhibition of the mutagenicity of bay region diol epoxides of polycyclic aromatic hydrocarbons by tannic acid, hydroxylated anthraquinones and butylated cinnamic acid derivatives. Carcinogenesis 6, 237±242. Huang M. T., Wood A. W., Newmark H. L., Sayer J. M., Jagi H., Jerina D. M. and Conney A. H. (1983) Inhibition of the mutagenicity of bay region diol epoxides of polycyclic aromatic hydrocarbons by phenolic plant ¯avonoids. Carcinogenesis 4, 1631±1637. Jones J. M. (1992) Naturally occurring food toxicants. In Food Safety, pp. 69±105. Egan Press, St. Paul, Minnesota. Kado N. Y., Langley D. and Eisenstadt E. (1983) A simple modi®cation of Salmonella liquid-incubation assay. Increased sensitivity for detecting mutagens in human urine. Mutation Research 121, 25±32. Kapadia G. J., Rao G. S. and Morton J. F. (1983) Herbal tea consumption and esophageal cancer. In Carcinogens

5

and Mutagens in the Environment, ed. H. F. Stich, Vol. 3, pp. 3±12. CRC Press, Boca Raton, FL. Maron. D. M. and Ames B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutation Research 113, 173±215. Mendelsoh M. L. (1992) Antimutagenic e€ects in humans. Mutation Research 267, 257±264. Mitscher L. A., Telikepalli H., Wang P.B-B., Kuo S., Shankel D. M. and Stewart G. (1992) Antimutagenicity of secondary metabolites from higher plants. Mutation Research 267, 229±241. Nagabhushan M., Sarode A. V., Nair J., Amonkar A. J., D'Souza A. V. and Bhide S. V. (1991) Mutagenicity and carcinogenicity of tea, Camellia sinensis. Indian Journal of Experimental Biology 29, 401±406. Okuda T., Yoshida T. and Hatano T. (1992) Pharmacological active tannins isolated from medicinal plants. Basic Life Science 59, 539±569. Onodera H., Kitaura K., Mitsumori K., Yoshida J., Yasuhara K., Shimo T., Takahashi M. and Hayashi T. (1994) Study on the carcinogenicity of tannic acid in F344 rats. Food and Chemical Toxicology 32, 1101± 1106. Oterdoom H. J. (1985) Tannin, sorghum and oesophageal cancer. Lancet ii, 330. Sanyal R., Darroudi F., Parzefall W., Nagao M. and Knasmuller S. (1997) Inhibition of the genotoxic e€ects of heterocyclic amines in human derived hepatoma cells by dietary bioantimutagens. Mutagenesis 12, 297±303. Shelef L. A. and Chin B. (1980) E€ect of phenolic antioxidants on the mutagenicity of a¯atoxin B1. Applied and Environmental Microbiology 40, 1031±1043. Sherwin E. R. (1990) Antioxidants. In Food Additives, ed. A. I. Branen, P. M. Davidson and S. Salminen, pp. 139±193. Marcel Dekker, New York. Smart R. C., Huang M-T., Chang R. L., Sayer J. M., Jerina D. M. and Conney A. H. (1986) Disposition of the naturally occurring antimutagenic plant phenols, ellagic acid, and its synthetic derivatives, 3-o-decylellagic acid and 3,3'-di-o-methylgallic acid in mice. Carcinogenesis 7, 1663±1667. Steele C. M., Lalies M. and Ioannides C. (1985) Inhibition of the mutagenicity of aromatic amines by the plant ¯avonoid (+)-catechin. Cancer Research 45, 3573±3577. Stich H. F. (1991) The bene®cial and hazardous e€ects of simple phenolic compounds. Mutation Research 259, 307±324. Stich H. F., Rosin M. P. and Bryson L. (1982) Inhibition of mutagenicity of a model nitrosation reaction by naturally occurring phenolics, co€ee and tea. Mutation Research 95, 119±128. Teel R. W. (1986) Ellagic acid binding to DNA as possible mechanism for its antimutagenic and anticarcinogenic action. Cancer Letter 45, 1±8. Terwel L. and Van der Hoeven J. C. M. (1985) Antimutagenic activity of some naturally occurring compounds towards cigarette-smoke condensate and benzo(a)pyrene in the Salmonella/microsome assay. Mutation Research 152, 1±4. Wood A. W., Huang M. T., Chang R. L., Newmark H. L., Lehr R. E., Yagi H., Sayer J. M., Jerina D. M. and Conney A. H. (1982) Inhibition of the mutagenicity of bay-region diol epoxides of polycyclic aromatic hydrocarbons by naturally occurring plant phenols: exceptional activity of ellagic acid. Proceedings of the National Academy of Sciences of the U.S.A. 79, 5513± 5517. Wu-Yuan C. D., Chen C. Y. and Wu R. T. (1988) Gallotannins inhibit growth, water-insoluble glucan synthesis and aggregation of mutans streptococci. Journal of Dental Research 67, 51±55.