Antimicrobial Activities of Tea Catechins and Theaflavins and Tea Extracts Against Bacillus Cereus

354

1 Journal of Food Protection, Vol. 69, No. 2, 2006, Pages 354-361 71 Antimicrobial Activities of Tea Catechins and Theaflavins and Tea Extracts against Bacillus cereus

MENDEL FRIEDMAN,I PHILIP R. HENIKA, CAROL E. LEVIN, ROBERT E. MANDRELL, AND NOBUYUKI KOZUKUE2

Western Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture, 800 Buchanan Street, Albany, California 94710, USA; and Department of Food Service Industry, Uiduck University, Gyongbuk, 780-713, Korea MS 05-266: Received 31 May 2005/Accepted 8 September 2005

ABSTRACT

We evaluated the antimicrobial activities of seven green tea catechins and four black tea theaflavins, generally referred to as flavonoids, as well as the aqueous extracts (infusions) of 36 commercial black, green, oolong. white, and herbal teas against Bacillus cereus (strain RM3I90) incubated at 21°C for 3, 15, 30, and 60 mm. The results obtained demonstrate that (I) (- )-gallocatechin-3-gallate, (- )-epigallocatechin-3-gallate. (- )-catechin-3-gallate, (- )-epicatechin-3-gallate, theaflavin-3, 3-digallate, theaflavin-3-gallate, and theaflavin-3-gallate showed antimicrobial activities at nanomolar levels; (ii) most compounds were more active than were medicinal antibiotics, such as tetracycline or vancomycin, at comparable concentrations; (iii) the bactericidal activities of the teas could be accounted for by the levels of catechins and theaflavins as determined by high-pressure liquid chromatography; (iv) freshly prepared tea infusions were more active than day-old teas; and (v) tea catechins without gallate side chains, gallic acid and the alkaloids caffeine and theobromine also present in teas, and herbal (chamomile and peppermint) teas that contain no flavonoids are all inactive. These studies extend our knowledge about the antimicrobial effects of food ingredients.

The gram-positive, facultative aerobe Bacillus cereus gallate, (- )-gallocatechin, ( - )-gal locatechin-3-gallate, gallic acid, is a foodborne bacterium that has, on occasion, been shown chloramphenicol, gentamycin sulfate, tetracycline, theobromine, to contaminate baked goods, carrots, meat, milk products, and vancomycin were obtained from Sigma (St. Louis, Mo.). Caf- rice, sauces, soups, sprouts, and zucchini (10, 13, 15, 18, feine, clindamycin, and rifampicin were obtained from Fluka 20). In previous studies, we determined the relative activ- (Steinheim, Switzerland), and theaflavin, theaflavin-3-gallate, theaflavin-3-gallate, and theaflavin-3, 3-digallate were obtained ities of plant essential oils and phenolic compounds against from Wako Chemical Company (Richmond, Va.). Thirty-six teas pathogenic bacteria (4, 6-8). These studies have provided in tea bags were obtained from the Stash Tea Company (Portland, insight into the structural features that govern bactericidal Oreg.). activities as well as the candidates for use in formulations designed to reduce pathogens in foods. The present study HPLC analysis of teas. Flavonoids were extracted from tea was designed to extend our knowledge of the antimicrobial leaves by stirring them for 5 min in boiled water previously ad- effectiveness of various flavonoids, individually and as justed to p1-1 7 to approximate the pH of the PBS buffer used for the bactericidal assays. The HPLC technique used to determine mixtures found in teas. To extend our knowledge about the the content of the flavonoids in teas has been described in detail structure-antimicrobial activity relationships of different elsewhere (9). Briefly, HPLC was carried out in Korea on a Hi- classes of plant compounds, the specific objectives of this tachi instrument equipped with an Inertsil column, an Autosam- study were (i) to compare the bactericidal activities against pler, and a Shimadzu UV-VIS detector. The mobile-phase gradient B. cereus of seven tea catechins and four theaflavins to the consisted of a mixture of acetonitrile and 20 mM KH 2PO4. The activities of clinical antibiotics shown to be active against II flavonoids were determined in a single run. The flavonoid con- Bacilli (16) and (ii) to relate the antimicrobial effects of 36 tent of the pH 7 water extracts did not differ from the levels in aqueous tea extracts to their content of these flavonoids as extracts obtained with unadjusted water. The same tea bags were determined by high-pressure liquid chromatography extracted in California for the bactericidal assays described below. (HPLC). Source of bacteria. B. cereus/thuringiensis (strain RM3 190) MATERIALS AND METHODS was isolated in our laboratory by Dr. Michael Cooley from soil and was identified by two bacterial identification systems (Biolog. Test compounds. (+ )-Catechin, (± )-catechin, (- )-catechin, Hayward, Calif., and Microbial ID, Newark, Del.). (-)-catechin-3 -gal late, (+ )-epicatechin, (- )-epicatcchin, (- )-epicatechin-3-gallate, (- )-epigallocatechin, (- )-epigallocatechin-3- Preparation of test substances for bactericidal assays. For steeping of tea leaves, a stainless steel screen from a "Beehouse" Author for correspondence. Tel: 510-559-5615; Fax: 510-559-5777; teapot. obtained at a local Peets Coffee and Tea store, was placed E-mail: [email protected]. with 0.9 g of tea leaves into a 250-ml beaker with 150 ml of hot Presented at the XI International Congress of Bacteriology and Applied 50 mM phosphate-buffered saline (PBS, pH 7.0) that had been Microbiology. IUMS-2005 Meeting, San Francisco, California, 23 to 28 brought to a boil and then allowed to sit for 2 mm. The teas were July 2005. Abstract B-I 160. steeped with gentle stirring for 5 mm. A steeped tea aliquot (200 J. Food Prot., Vol. 69, No. 2 ANTIMICROBIAL ACTIVITIES OF TEA FLAVONOIDS AND TEAS AGAINST B. CEREUS 355

FIGURE 1. Structures of the tea catechins 2kOH .. OH H and theaflavins evaluated in this study. H 0 6LJL..i;fI 6 HOH I 4 0H OH OH HO" OH OH °H H (.)_catechèn (4 ).epitechjn (-)-epigallocatechin

H OH HO HqH - OH ^OH HO5 OH OH OH OH OH O((OH HO H OH OH OH (-)-catechin-3-gallate tH ()-epicatechin3gallate OH (-)-gallocatechin-3-gallate

Hç(JOH OH

OH 0

HO H (-)-epigallocatechin-3-gallate

theaflavin-3-gallate theaflavin-3 .gatate theaflavin-3,3-digallate

jil) was then added to the PBS (1,800 III). This sample (200 p.l) brewed and day-old tea extracts (infusions): 12 wells for PBS- was again diluted with PBS (1,800 11) for a final 100-fold dilu- negative controls and 60 wells for five different teas, each with tion. The latter was used as the first test dose. The second dose the two dilution series as described above for the preparation of was diluted by one third, i.e.. tea (333 p.l) plus PBS (667 .LI). The the tea infusions. The contents of each of the 72 wells were then pH of these samples was 7.0 at 21°C. Steeped teas were left stand- tested after a 60-min incubation at 21°C as described below. ing on the bench top for 24 h at 21°C and tested as "day-old" teas. Antimicrobial assays. The bactericidal assay described pre- For catechins, the test solution was prepared by dissolving viously (6, 8) was modified for this study. B. cereus organisms the test substance (10 mg) in PBS (9.9 ml) in a 50-ml test tube were stored on streaked plates subcultured for 16 to 18 h at 37°C to give a 0.1% solution. The tube was warmed in a microwave with Luria-Bertani agar (LB) plates (Difco, Becton Dickinson, oven for 10 s and vortexed to clarity. Theaflavins were available Sparks, Md.). Overnight LB broth cultures were prepared by har- only in 1-mg quantities in glass vials. PBS (I ml) was added to vesting a few isolated colonies from the plate with a sterile loop these vials. These vials were then warmed in a microwave oven and suspending them into 5 ml of LB broth in a 15-ml sterile for 5 s and vortexed to give clear solutions. These test solutions plastic tube. The capped tubes were incubated with shaking (200 were then diluted 100-fold with PBS before being added to mi- rpm) at 37°C for 18 h. crotiter plates. For the assay, bacterial suspensions (ca. 100 to 200 CFU per The pure compounds in PBS (500 Id each) were added to lane) were placed on the square plates with grids used for count- five sterile tubes (1.9 ml). Starting with the 1/100 test solution (I ing. Briefly, each sample (1 ml) of an 18-h LB broth culture of MI), we performed serial dilutions, using 500 til for each transfer, B. cereus was added to a 1.9-ml microfuge tube, and the bacteria to the sterile tubes for a total of five dilutions. In a typical ex- were pelleted by centrifugation in a microfuge (15,800 X g) for periment, microtiter plates (96 wells; Nalge, Nunc, Rochester, 1 mm. After removal of the supernatant, sterile PBS (1 ml) was N.Y.) were prepared with PBS-negative controls (100 i.l each in added to the pellet. The pellet was then resuspended by gentle six wells) as well as three test substances with five dilutions (100 aspiration in and out of a transfer pipette. The optical density at sl each dilution per well). The contents of these 24 wells were 620 nm of the sample was adjusted with PBS to ca. 1.0. The then tested at three time intervals. suspension (40 il) was added to the PBS (960 p,l). The latter The tea samples. in 72 wells, were prepared for the freshly (160 1.d) was then added to PBS (5 ml), vortexed, and poured into 356 FRIEDMAN El AL. Food Prot., Vol. 69, No. 2

TABLE 1. Antimicrobial activities of tea flavonoids and clinical antibiotics against Bacillus cereus incubated at 21°C for 3, 15, and 30 min"

BA50 (nmol/well) Relative ratios of 30-mm Compound 3 mm 15 mm 30 mm act iv it ies"

(-)-Gal locatechin-3 -gallate (458) 24 ± l8d 0.73 ± 0.37" 0.43 ± 0.14 2,307 Theaflavin-3,3-digallate (869) ND 1.6 ± 0.21 0.54 ± ND" 1,837 (- )-Epigallocatech in- 3-gallate (458) 37 ± 201 0.99 ± 0.389 0.68 ± 031d 1,459 Theaflavin-3-gal late (717) ND 2.4 ± 0.54 1.8 ± 0.08 551 Rifampicin (823) 1.6 ± 11b 2.4 ± 0.79 2.1 ± 0.461 472 Vancornycin HCI (1486) 2.8 ± 0.52 2.6 ± Q4i 2.3 ± 0.48 431 Theaflavin-3-gallate (717) ND 3.6 ± 0.49 2.8 ± 0.11 354 (- )-Catechin-3-gallate (442) 52 ± 15 7.8 ± 3.7 5.5 ± 1.6 180 Tetracycline HCI (481) 32 ± l3 17 ± 6" 12 ± 6.8" 83 Clindamycin (461) 16 ± 3.9 26 ± 13 19 ± 6.1" 52 (-)-Epicatechin-3-gallate (442) 234 ± 9° 25 ± 3.2° 22 ± 141 45 Gentamycin sulfate (391) 36 ± 8 87 ± 60" 75 ± 56 13 97d 103k Chloramphenicol (323) 152 ± 201 ± 180 5.5 Theaflavin (564) ND 833 ± ND" 283 ± 100 3.5 (-)-Epigallocatechin (306) >2,146q 1,167 ± 424 800 ± 141 1.2 (-)-Gallocatechin (306) >2,146q 1,393 ± 883 992 ± 452 1.0 a Listed values are for n = 2, ±SD, except as indicated below; ND, not determined. The 60-min values (not shown) were similar to the 30-min values. The following compounds with the indicated BA 50 values (nnmol/well) were considered inactive: caffeine (>3,454), (+ )-catechin (>2,310), (±)-catechin (>2,310), (-)-catechin, (>2,310), (+)-epicatechin (>2,310), (-)-epicatechin (>2,310), gallic acid monohydrate (>3,564), and theobromine (>3,772). b Ratios based on (-)-gallocatechin BA 50 (nmol per well) = 1. Values in parentheses are molecular weights. It = 8. "n = 1. I n = 15. 5 n = 9. n = 5. n = 6. i n = 10. k = 7. n = 12. "n = 11. "n = 3. °n = 4. = 13. Inactive.

a sterile plastic petri dish. We drew bacterial suspensions (50 p.l) substance. The lower the BA 50 or the higher the 1/BA 50 value, with a multichannel Eppendorf pipette using six channels and add- the greater the activity. ed them to six microtiter plate wells. This procedure was repeated until all the prepared wells were inoculated. RESULTS AND DISCUSSION Immediately after inoculation or at the end of a specified Teas can be classified into three major categories: un- incubation time, aliquots (10 u.1) from each of the six wells were drawn with an Eppendorf multichannel pipette for the spotting of fermented green teas containing catechins, fully fermented six 10-hi drops at the top of a square petri plate made with LB black teas containing theaflavins, and semifermented teas agar. The plates were tilted before spotting to avoid coalescence containing both catechins and theaflavins (19). An exami- of drops and tapped gently to facilitate movement of the liquid to nation of the structures of the catechins evaluated in this the bottom. The inoculated plates were tested three times at 3 mm study (Fig. I) shows that (depending on the configuration (it takes 2 to 3 min to spot four plates). IS mm, and 30 mm. or of the 3, 4-dihydroxyphenyl and phenolic OH groups at :11 3, 30. and 60 mm. After spotting, the plates were left to dry the 2- and 3-positions of the C-ring) catechins can exist as r 10 min in a biological safety hood at 21°C without agitation. two isomers: (-)-catechin and (+)-epicatechin. They can The plates were incubated for 8 to 9 It at 37°C and then refrig- be further modified to form (-)-epigallocatechin, (-)-cat- rated at 4°C overnight. The plates with B. cereus colonies were echin-3-gallaic, (- )-epicatechin-3-gallate, (- )-gallocate- then removed from the refrigerator the next day and incubated for chin-3-gallate, and (- )-epigallocatechin-3-gallate. Thea- further I to 2 h at 37°C before counting. The bacterial load of cach microtiter well was ca. 1,500 to 3,000 cells per 150 p.]. flavins are formed when catechins are oxidatively coupled. I \periments were performed in duplicate. We did not observe Antimicrobial effects of green tea catechins. Because porulation. the test compounds differ in molecular weights. we calcu- Bactericidal activities (BA values), defined as the percent- -50 lated the activities of all pure compounds in terms of nano- ige of test compound that kills 50% of the bacteria, were calcu- moles per well that kill 50% of the bacteria (BA lated as follows. The numbers of CFU from each dilution were 50, nmol matched with the average control value to determine the percent- per well) (Table 1). To summarize, Table I shows that (i) :tte of bacteria killed per well. The CFU values from all experi- the antimicrobial activities of the active catechins increased nients were transferred to a Microsoft Excel 8.0 Spreadsheet. Each with time of incubation from approximately 3 to 30 mm; of the dose-response profiles (% test compound versus % BA50) the results after 60 min were similar to the results after 30 sos examined graphically, and the BA50 values were estimated mm; (ii) the 30-mm nmol/well BA50 values for the six ac- by linear regression. We also converted the BA 50 values to nano- tive catechins ranged as follows: (-)-gallocatechin-3-gal- moles per mole per well using the molecular weight of each test late, 0.43 (highest activity) (- )-epigallocatechin-3-gallate, J. Food Prot., Vol. 69, No. 2 ANTIMICROBIAL ACTIVITIES OF TEA FLAVONOIDS AND TEAS AGAINST B. CEREUS 357

TABLE 2. Antimicrobial activities (BA 50 and 1/BA 50 values) of freshly brewed and day-old teas against Bacillus cereus incubated for 60 min at 21°C

Fresh tea 24-h-old tea Ratios of activities Tea names BA 50 (A)" 11BA50 BA,, (B)" I/B A50 (A/B)

Darjeeling Green Organic 0.018 ± 0.003" i5.6 ± 9.26 0.042 ± 0.006" 23.8 ± 3.4 2.3 Darjeeling Summer 0.02 ± 0.01" 50 ± 25 0.11 ± 0.02 9.1 ± 1.65 5.5 Exotica Silver Jasmine 0.023 ± oo, 3.5 ± 0 0.15 ± 0.03 6.7 ± 1.33 6.5 Daijeeling Black 0.024 ± 0.01" 1.7 ± 17.36 0.08 ± 0.009 12.5 ± 1.41 3.3 Premium Green 0.025 ± 0•02b 40 ± 32 0.2 ± 0.02" 5 ± 0.5 8.0 Exotica Champagne Oulong 0.027 ± 0.006" 37 ± 8.23 0.3 ± 0.01 3.3 ± 0.11 11.1 Exotica Dragonwell Green 0.027 ± 0008b 37 ± 10.97 0.32 ± 0.06 3.1 ± 0.59 11.9 Kopili Assam Black 0.028 ± 0.006" 5.7 ± 7.65 0.11 ± 0.006 9.1 ± 0.5 3.9 Darjeeling Black Organic 0.029 ± 0•008h 4.5 ± 9.51 0.065 ± 0.001 15.4 ± 0.24 2.2 Exotica Golden Darjeeling 0.032 ± 001b 31.3 ^ 9.77 0.18 ± 0.003 5.6 ± 0.09 5.6 Earl Grey Black & Green Organic 0.034 ± 0.02e 29.4 ± 17.3 0.17 ± 0.04 5.9 ± 1.38 5.0 Fusion Green and White 0.038 + 0.006 26.3 ± 4.16 0.19 ± 0.02 5.3 ± 0.55 5.0 Green Organic 0.039 ± 0.01 25.6 ± 6.57 0.12 ± 0.02 8.3 ± 1.39 3.1 English Breakfast 0.040 ± 0.02r 25 ± 12.5 0.37 ± 009" 2.7 ± 0.66 9.3 Exotica China White 0.043 ± 00211 23.3 ± 10.82 0.13 ± 0.02 7.7 ± 1.18 3.0 Sushi Bar Green 0.045 ± 0007b 22.2 ± 3.46 0.14 ± 0.04 7.1 ± 2.04 3.1 Darjeeling Spring 0.046 ± 0.02 21.7 ± 9.45 0.11 ± 0.02 9.1 + 1.65 2.4 Jasmine Blossom Green 0.049 ± 0.008 20.4 ± 333 0.14 ± 0.02 7.1 ± 1.02 2.9 China Oolong 0.052 ± 0.02 19.2 ± 7.4 0.19 ± 0.04 5.3 ± 1.11 3.7 Earl Grey Black 0.058 ± 0.01 17.2 ± 2.97 0.18 ± 0.04 5.6 ± 1.23 3.1 Exotica Ceylon Estate Earl Grey 0.058 ± 0.01 17.2 ± 2.97 0.48 ± 0.007" 2.1 ± 0.03 8.3 Exotica Assam Breakfast 0.063 ± 0.008 15.9 ± 2.02 0.1 ± 0.008 10 ± 0.8 1.6 Breakfast Blend Organic 0.065 ± 0.01 15.4 ± 2.37 0.12 ± 0.02 8.3 ± 1.39 1.8 Orange Spice Black 0.065 ± 0.03 15.4 ± 7.1 0.19 ± 0.02 5.3 ± 0.55 2.9 Nigiri Black 0.068 ± 0.02 14.7 ± 4.33 0.17 ± 0.02 5.9 ± 0.69 2.5 Lemon Spice Green 0.073 ± 0.02 13.7 ± 3.75 0.16 ± 0.02 6.3 ± 0.78 2.2 0. 1 9b Premium Green Decaffeinated 0.099 ± 0.03 10.1 ± 3.06 0.44 ± 2.3 ± 0.98 4.4 Exotica Osmanthus 0.11 ± 0.01 9.1 ± 0.83 0.62 ± 0.06" 1.6 ± 0.16 5.6 Exotica Reserve Blend 0.15 ± 0.0 7.7 ± 1.18 0.22 ± 0.03 4.5 ± 0.62 1.7 Fusion Red and White 0.15 ± 0.04 6.7 ± 1.78 0.3 ± 0.11 3.3 ± 1.22 2.0 Kukicha 0.58 ± 0.13" 1.7 ± 0.39 0.66 ± 0004 1.5 ± 0.01 1.1 "Percentage (ml/100 ml) of tea in PBS buffer that killed 50% of the bacteria; n = 4. -_SD (no superscript), unless indicated otherwise. The following teas exhibited no antimicrobial activity: Chamomile. Moroccan-Mint Green, Peppermint, Pu-arh Oolong, and Wild Raspberry Herbal. b0 = 2. = 3.

0.68; (- )-catechin-3 -gal late, 5.5; (-)-catechin-3-gallate, Antimicrobial effects of black tea theaflavins. Table 22; (-)-epigallocatechin, 800; and (-)-gallocatechin, 992; 1 shows that the theaflavins also showed strong antibacte- (iii) there was thus a 2,307-fold difference in activities for rial effects against B. cereus. Thus, the antibacterial activ- the most active compared to the least active catechin; (iv) ities for the four theaflavins in terms of nanomolar BA50 the highest activity resides in the 3-gallate esters of both values were as follows: theaflavin-3, 3-digallate, 0.54 catechins and cpicatechins: and (v) the epimers, (-)-gal- (highest activity); theaflavin-3 -gallate, 1.8; theaflavin-3- locatechin-3-gallate and (- )-epigallocatechin-3-gallate. gaUate, 2.8; and theaflavin, 283. These data demonstrate the have identical activities, whereas (-)-catechin-3-gallate following: (i) the three theaflavin gallates showed signifi- was four times more active than the epimeric (-)-epicate- cantly greater antimicrobial activities than the theaflavin chin-3-gallate. that lacked the gallate side chain, (ii) the difference in ac- The following catechins lacking a gallate side chain tivities among the four theaflavins was 520-fold, and (iii) were inactive under the test conditions of the assay: the the activities of the three theaflavin gallates were of the (H-)-catechin, the isomeric (-)-catechin, the mixture of the same order as those cited above for the catechin-3-gallates. two, the (±)-catechin, the (+)-epicatechin, and the isomeric Because this study was directed toward possible appli- (-)-epicatechin. Gallic acid and the alkaloids caffeine and cations of tea compounds to foods, we carried out time theobromine also present in teas were also inactive. Cate- studies designed to discover short-term effects. The time chin structure appears to strongly influence antimicrobial study in Table 1 shows that the medicinal antibiotics acted effects. more rapidly than the flavonoids. 358 FRIEDMAN ET AL. J. Food Prot.. Vol. 69, No. 2

TABLE 3. HPLC analysis of (-)-epigallocarechin (EGC), catechin (C), epicatechin (EC), (-)-epigallocatechin-3-gallate (EGCG). (-)-gallocatechin-3-gallate (GCG), (-)-epicatechin-3-gallare (ECG), (-)-catechin-3-gallate (CG), theaflavin (TF), theaflavin-3-gallate (TF3G), theaflavin-3 -ga/late (TF3 G), and theaflavin-3,3 -digallate (TF33 G) in 36 teas evaluated in this study

Catechins, CATS (mg/g tea)

Stash tea EGC C EC EGCG GCG

Darjeeling Green Organic 8.1 ± 0.06 5.1 ± 0.29 1.9 ± 0.05 53.6 ± 1.41 3.9 ± 0.03 Darjeeling Summer 5.1 ± 0.18 4.9 ± 0.01 2.2 ± 0.02 26.8 ± 0.01 3 ± 0.04 Exotica Silver Jasmine 6.1 ± 0.17 7.9 ± 0.36 2.9 ± 0.09 36.3 ± 0.2 2.7 ± 0.1 Darjeeling Black 4.2 ± 0.15 0.8 ± 0.01 1.4 ± 0.03 27 ± 0.56 4.2 ± 0.03 Premium Green 13.6 ± 0.61 2 ± 0.09 0.9 ± 0.05 38.3 ± 0.42 5.1 ± 0.09 Exotica Champagne Oolong 2.9 ± 0.47 0.8 ± 0.01 1.1 ± 0.17 37.5 ± 0.05 4.4 ± 0.11 Exotica Dragonwell Green 0.7 ± 0.01 7.9 ± 0.06 1.1 ± 0.01 43.6 ± 0.52 1.7 ± 0 Koplii Assami Black 2.3 ± 0.57 6.7 ± 0.05 4.2 ± 0.03 13.3 ± 0.31 6.7 ± 0.13 Darjeeling Black Organic 3.1 ± 0.02 5.3 ± 0.05 4.1 ± 0.2 23.5 ± 0.16 6.1 ± 0.26 Exotica Golden Darjeeling 5.2 ± 0.02 1.7 ± 0.06 2.3 ± 0.02 28.2 ± 0.13 3.2 ± 0.06 Earl Grey Black & Green Organic 2.8 ± 0.11 4.3 ± 0.08 2.3 ± 0.03 9 ± 0.68 1.8 ± 0.02 Fusion Green and White 5.8 ± 0.25 6.2 ± 0.09 2.5 ± 0.02 33.5 ± 0.99 2.7 ± 0.11 Green Organic 13.9 ± 0.09 1.7 ± 0.01 1.7 ± 0.02 41.8 ± 0.17 2.4 ± 0.2 English Breakfast 3.4 ± 0.82 3 ± 0.16 2.3 ± 0.1 6.6 ± 0.18 2.4 ± 0.01 Exotica China White 2.1 ± 0.11 4.2 ± 0.05 1.3 ± 0.04 30.4 ± 0.8 2.9 ± 0.06 Sushi Bar Green 13.8 ± 0.42 5.8 ± 0.15 2.6 ± 0.03 41.2 ± 0.39 5.1 ± 0.1 Darjeeling Spring 4.1 ± 0.3 1.2 ± 0.07 2 ± 0.01 28.4 ± 0.56 4.4 ± 0.18 Jasmine Blossom Green 6.1 ± 0.1 6.2 ± 0.02 0.9 ± 0.01 29.4 ± 0.22 5 ± 0.08 China Oolong 5.7 ± 0.24 5.7 ± 0.15 2.5 ± 0.01 23.2 ± 0.04 3.5 ± 0.16 Earl Grey Black 6.4 ± 0.01 5 ± 0.07 2.1 ± 0.02 5.2 ± 0.13 2.3 ± 0.02 Exotica Ceylon Estate Earl Grey 10.3 ± 0.2 2.7 ± 0 4.3 ± 0.05 23 ± 0.03 1.1 ± 0.1 Exotica Assaic Breakfast 2.3 ± 0.09 7.1 ± 0.43 2.6 ± 0.1 9.5 ± 0.05 1.6 ± 0.01 Breakfast Blend Organic 2.5 ± 0.09 7.4 ± 0.16 2.5 ± 0.06 12.7 ± 0.73 2.1 ± 0.1! Orange Spice Black 1.9 ± 0.05 3 ± 0.06 1.7 ± 0.13 6.1 ± 0.22 1.5 ± 0.1 Nigiri Black 2.3 ± 0.03 8.1 ± 0.5 2.4 ± 0.03 6.1 ± 0.06 1.9 ± 0.03 Lemon Spice Green 3.4 ± 0.01 2.8 ± 0.19 1.4 ± 0.01 12.9 ± 0.02 2.9 ± 0.01 Premium Green Decaffeinated 4.8 ± 0.09 4.4 ± 0.05 0.8 ± 0.02 13.5 ± 0.21 2.7 ± 0.01 Exotica Osminthus 2.2 ± 0.05 3.7 ± 0.63 0.9 ± 0.03 9.9 ± 0.02 3.7 ± 0.16 Exotica Reserve Blend 3 ± 0.03 4.6 ± 0.06 2.3 ± 0.04 16.9 ± 0.07 4.1 ± 0.04 Fusion Red and White 1.6 ± 0.04 2.3 ± 0.1 1.1 ± 0.01 11.4 ± 0.11 4.8 ± 0.22 Kukicha Li ± 0.04 3.5 ± 0.07 0.5 ± 0.01 7.4 ± 0.07 6.3 ± 0.18 Chamomile ND ND 0.2 ± 0.01 2.1 ± 0.01 0.9 ± 0.09 Moroccan Mint Green 1 ± 0.04 33.2 ± 0.29 1.3 ± 0.05 6.7 ± 0.3 3.2 ± 0.23 Peppermint 0.2 ± 0.03 1 ± 0.02 ND ND ND Pu-arh Oolong Trace 1.5 ± 0.17 0.4 ± 0.01 2.1 ± 0.02 1.7 ± 0.04 Wild Raspberry Herbal Trace 1.4 ± 0.36 0.2 ± 0.01 0.2 ± 0.02 0.6 ± 0.04 "Listed values are averages ± SD (n = 3); ND, not detected.

Antimicrobial effects of medical antibiotics. Because Antimicrobial effects of teas. Table 2 lists the bacte- the catechin-3-gailates and theaflavins were active against ricidal effects of 36 black, green, oolong, white, and herbal B. cereus at nanomolar levels, we compared the sensitivity tea infusions in terms of the percentage of fresh and day- of this bacterium to antibiotics that have been shown to be old teas in PBS buffer that killed 50% of the bacteria inactive against Bacilli, including B. anthracis, a pathogen cubated at 21°C for 60 mm (BA 50 values). Table 3 lists the that is genetically related to B. cereus (11, 16, 22). levels of catechins and theaflavins of the same teas as de- Table 1 shows that (i) unlike the time-dependence data termined by HPLC. for catechins and theaflavins against B. cereus, the BA50 The data demonstrate that (i) the BA 50 values for the values for the antibiotics were identical after incubation for teas ranged from 0.018% (0.018 ml of tea in 100 ml of approximately 3, 15, or 30 mm; (ii) the 30-nun nmol/well PBS buffer killed 50% of the bacteria) for Darjeeling Green BA50 values for the antibiotics ranged as follows: rifam- Organic to 058% for Kukicha, about a 30-fold ratio from p1cm, 2.1 (highest activity); vancomycin, 2.3; tetracycline, the least to the most active tea, and (ii) freshly prepared 2; clindamycin, 19; and chloramphenicol, 180; and (iii) teas were generally more active than were teas that had the most active catechin-3-gallates and theaflavin gallates been left standing for 24 h. These observations demonstrate are more effective antimicrobial agents against B. cereus the strong antimicrobial activities of a widely consumed than sonic of the antibiotics. beverage against a foodhorne pathogen.

J. Food Prot., Vol. 69, No. 2 ANTIMICROBIAL ACTIVITIES OF TEA FLAVONOIDS AND TEAS AGAINST B. CEREUS 359

TABLE 3. Extended

Catechins. CATS (mg/g tea) - Theaflavins TFS (mg/g tea)

Sum of Sum ECG CG CATS TF TF3G TF3G TF33G of TFS

27.1 ± 0.49 0.3 ± 0.0 100.0 ND ND ND ND 0.0 20.9 ± 0.1 0.1 ± 0.01 62.9 0.8 ± 0.01 0.6 ± 0.03 0.3 ± 0.01 6.36 ± 0.02 8.1 26.9 ± 0.12 0.3 ± 0.02 83.1 ND ND ND ND 0.0 20.2 ± 0.35 0.4 ± 0.02 58.1 0.7 ± 0.01 0.4 ± 0.02 0.2 ± 0.01 0.46 ± 0.02 1.7 15.9 ± 0.14 0.1 ± 0.01 75.9 ND ND ND ND 0.0 18.2 ± 0.32 0.5 ± 0.01 65.3 ND ND ND ND 0.0 19.2 ± 0.88 0.2 ± 0.01 74.5 ND ND ND ND 0.0 24.4 ± 0.41 1.3 ± 0.05 58.9 0.8 ± 0.02 1.8 ± 0.02 0.8 ± 0.01 3.48 ± 0.07 6.9 26.8 ± 0.45 0.6 ± 0.01 69.6 0.9 ± 0.03 0.8 ± 0.01 0.2 ± 0.01 1.13 ± 0.02 3.0 24.2 ± 0.22 0.7 ± 0.03 65.6 ND ND ND ND 0.0 11 ± 0.28 0.8 ± 0.01 31.9 1.2 ± 0.01 1.4 ± 0.0 0.5 ± 0.01 1.18 ± 0.02 4.3 16.1 ± 0.15 0.2 ± 0.01 67.0 ND ND ND ND 0.0 18.6 ± 0.17 0.1 ± 0 80.2 ND ND ND ND 0.0 10.1 ± 0.17 0.7 ± 0.03 28.5 1.2 ± 0.01 1.6 ± 0.06 0.6 ± 0.02 1.99 ± 0.03 5.4 19.2 ± 0.51 0.4 ± 0.02 60.4 ND ND ND ND 0.0 18.4 ± 0.12 0.2 ± 0.01 87.1 ND ND ND ND 0.0 22.9 ± 0.57 0.4 ± 0 63.3 0.7 ± 0.01 0.4 ± 0.01 0.1 ± 0 0.36 ± 0.02 1.6 15.9 ± 0.12 0.3 ± 0.01 64.2 ND ND ND ND 0.0 2.2 ± 0.35 2.3 ± 0.03 44.9 ND ND ND ND 0.0 10 ± 0.15 0.7 ± 0 31.7 I ± 0.01 1.6 ± 0.04 0.6 ± 0.02 1.83 ± 0.07 5.0 20.6 ± 0.53 0.5 ± 0.02 62.5 ND ND ND ND 0.0 16.1 ± 0.14 0.5 ± 0 39.6 1.2 ± 0.05 2.1 ± 0.04 0.9 ± 0.01 3 ± 0.04 7.3 14.5 ± 0.57 0.6 ± 0.02 42.2 1.7 ± 0.08 1.3 ± 0.05 0.6 ± 0.03 1.47 ± 0.02 5.1 10.8 ± 0.14 1 ± 0.03 26.1 1.2 ± 0.01 1.4 ± 0.04 0.4 ± 0.01 5.83 ± 0.07 8.8 11.8 ± 0.38 0.6 ± 0.03 33.2 1.2 ± 0.08 1.5 ± 0.06 0.6 ± 0.08 1.78 ± 0.03 5.0 11.7 ± 0.01 0.4 ± 0.05 35.5 0.6 ± 0.02 0.6 ± 0.02 0.8 ± 0.03 0.64 ± 0.03 2.6 6 ± 0.1 0.3 ± 0 32.4 ND ND ND ND 0.0 7.8 ± 0.04 0.6 ± 0.01 28.8 0.2 ± 0.02 0.2 ± 0.03 0.1 ± 0.01 0.49 ± 0.02 1.0 16.8 ± 0.29 0.5 ± 0.08 48.2 0.8 ± 0.02 1.1 ± 0.04 0.4 ± 0.02 1.36 ± 0.04 3.7 9.1 ± 0.08 0 ± 0 30.3 ND ND ND ND 0.0 5 ± 0.04 0.1 ± 0 23.9 ND ND ND ND 0.0 1.8 ± 0.27 0.3 ± 0.01 6.4 ND ND ND ND 0.0 7.7 ± 0.38 0.4 ± 0.01 53.4 ND ND ND ND 0.0 24.6 ± 0.37 ND 26.4 ND ND ND ND 0.0 1.3 ± 0.06 0.1 ± 0.02 7.4 ND ND ND ND 0.0 0.4 ± 0.02 0.2 ± 0.01 3.4 ND ND ND ND 0.0

The loss of activity of teas after 24 h does not appear late esters in the tested teas, other factors probably contrib- to follow any obvious trend. This loss ranged from no de- ute to the observed bactericidal effects. Such factors may crease for Kukicha tea to a 12-fold decrease for the Exotica include competitive, additive, antagonistic, and synergistic Dragonwell Green tea (Table 2, last column). However, this interactive effects among the flavonoids at the molecular tea was still highly active after 24 h, with a BA50 value of and cellular levels of the bacteria. 0.32% (0.32 ml of tea per 100 ml of PBS). Losses in ac- It is also relevant to note that (-)-epigallocatechin-3- tivity of teas could be due to oxygen-, light-, and metal gallate, which we found to strongly inhibit B. cereus, has ion-induced degradation or polymerization of active com- also been reported to strongly inhibit the anthrax lethal fac- pounds. tor produced by B. anthracis. with an IC 60 value (concen- Attempts to relate levels of a specific flavonoid in the tration of the catechin that inhibited 50% of the toxin) of large number of teas shown in Table 3 to activities of the 97 nM (3), as well as inhibiting the growth of human leu- corresponding pure flavonoid shown in Table I were not kemia cells with an IC 50 value of 100 iM (17). We surmise successful. The best correlations we could make were with that the inactivation of the bacterial toxin is a result of the the sum of the three most active catechins (R2 = 0.524) binding of phenolic OH groups of the (-)-epicatechin-3- and the sum of the 11 flavonoids (R2 = 0.583) (Fig. 2). gallate (Fig. 1) to the zinc atom associated with the metal- These observations suggest that although there appears to loproteinase of the toxin (1, 5) or that it is a result of the be an association between activity and the presence of gal- antioxidative effects of the catechin (2). 360 FRIEDMAN ET AL. J. Food Prot., Vol. 69, No. 2

120 A Sum of catechins and theaflavins 0) - - U-- Sum of EGCG, EGC, and ECG y= 1.131x+28.41 Linear (Sum of catechins and theaflavins) 100 R2 = 0.583 - - - Linear (Sum of EGCG, EGC, and ECG) co

co 80 IL -AL co 5 60 -c Dc co 40 Y= 1.110x+ 16.35 I-, i•Iu R2 = 0.521 Ar 20 E Cs

0 0 10 20 30 40 50 60 1/BA5O FIGURE 2. Relationships between the flavonoid content of fresh teas as determined by 1-IPLC and their bactericidal activities values) against B. cereus. The equations for the two straight lines were calculated with the aid of an Excel program for least-square fits represented by y = mx + b, where y represents the flavonoid Content listed in Table 3, x represents the activity shown in Table in 2, the slope, and b is the intercept.

It is not known whether the tea compounds will be as 5. Friedman, M., 0. K. Grosjean, and J. C. Zahnley. 1986. Inactivation active against other strains of B. cereus, whether theaflavins of metallo-enzymes by food constituents. Food Chem. Toxicol. 24: and tea infusions will also inhibit the anthrax toxin, or 897-902. 6. Friedman, M., P. R. Henika, C. E. Levin, and R. E. Mandrell. 2004. whether similar mechanisms govern both the antimicrobial Antibacterial activities of plant essential oils and their components and anticarcinogenic effects of tea flavonoids. against Escherichia Co/i 0157:H7 and Salmonella enterica in apple juice. J. Agric. Food Chem. 52:6042-6048. The observed bactericidal effects of tea catechin-3-gal- 7. Friedman, M., P. R. Henika, and R. F. Mandrel[. 2002. Bactericidal lates, theaflavins, and tea infusions complement and extend activities of plant essential oils and some of their isolated constitu- reported studies on the antimicrobial activities of tea com- ents against Camp ylobacier jejuni, Escherichia co/i 0157:H7, Lis- pounds against other pathogens (12, 14, 21, 23, 24). The teria monocytogenes. and Salmonella enterica. J. Food Prot. 65: cited trends in activities of different tea flavonoids also con- 1545-1560. 8. Friedman. M., P. R. Henika, and R. E. Mandrell. 2003- Antibacterial tribute to the elucidation of the structural features of these activities of phenolic bcnzaldehydes and benzoic acids against Cam- compounds that influence bactericidal effects. Such an un- pvlohacter jejun,, Escherichia coli 0157:H7, Listeria nionocytoge- derstanding may help devise food formulations that reduce nec, and Salmonella enterica. J. Food Prot. 66:1811-1821. pathogens in foods. The data offer the consumer a choice 9. Friedman, M., S.-Y. Kim, S.A. Lee, G.-P. Han, J.-S. Han, K.-R. Lee, of tea brands with very high antimicrobial properties and N. Kozukue. 2005. Distribution of catechins, theaflavins, caf- against B. cereus and possibly other Bacilli. feine. and theobromine in 77 teas consumed in the United States. J. Food Sci. 70:C550-0559. ACKNOWLEDGMENT 10. Guinebretiere, M. H., H. Girardin. C. Dargaignaratz, F Carlin, and C. Nguyen-The. 2003. Contamination flows of Bacillus cereu.s and We thank Dr. Michael Cooley for the B. cereus culture. spore-forming aerobic bacteria in a cooked, pasteurized and chilled zucchini puree processing line. mt. J. Food Microbiol. 82:223-232. REFERENCES II. Hoffmaster, A. R., J. Ravel, D. A. Rasko, G. D. Chapman, M. D. I. Renelli. R_ R. Vene. D. Bisacchi, S. Garbisa. and A. Albini. 2002. Chute, C. K. Marston, B. K. Dc, C. T. Sacchi, C. Fitzgerald, L. W. Anti-invasive effects of green tea pol yphenol epigallocatechin-3-gal- Mayer, M. C. Maiden, F G. Priest, M. Barker, L. Jiang, R. Z. Cer, late (EGCG), a natural inhibitor of metallo and serine proteases. Biol. J. Rilstone, S. N. Peterson, R. S. Weyant. D. R. Galloway, T. D. (hen,. 383:101-105. Read, T Popovic, and C. M. Fraser. 2004. Identification of anthrax 2. Cabrera, C., R. Gimenez. and M. C. Lopez. 2003. Determination of toxin genes in a Bacillus cereus associated with an illness resembling tea components with antioxidant activity. J. Agric. Food Chem. 51: inhalation anthrax. Proc. Natl. Acad. Sci. USA 101:8449-8454. 4427-4435. 12. Hu, Z. Q., W. H. Zhao, Y. Yoda, N. Asano, Y. Hara, and T. Shi- 3. DeIlAica, I., M. Donà, F Tonello, A. Pins, M. Mock, C. Monte- mamura. 2002. Additive, indifferent and antagonistic effects in com- cucco. and S. Garbisa. 2004. Potent inhibitors of anthrax lethal factor binations of epigallocatechin gallate with 12 non-13-lactam antibi- from green tea. EMBO Rep. 5:418-422. otics against methicillin-resistant Staphylococcus aureu,s. J. Anti,ni- 4. Friedman, M., R. Buick, and C. T Elliott. 2004. Antibacterial activ- crob. Chemother. 50:1051-1054. ities of naturally occurring compounds against antibiotic-resistant 13. Jaaskelainen, E. L., M. M. Haggblom, M. A. Andersson, L. Vanne. Bacillus cereus vegetative cells and spores, Escherichia co/i, and and M. S. Salkinoja-Salonen. 2003. Potential of Bacillus cereus for .uiphvlococctis oureur. J. Food Prot. 67:1774-1779. producin g an emetic toxin. cereulide, in bakery products: quaittita-

J. Food Prot., Vol. 69, No. 2 ANTIMICROBIAL ACTIVITIES OF TEA FLAVONOIDS AND TEAS AGAINST B. CEREUS 361

tive analysis by chemical and biological methods. J. Food Prot. 66: 19. Shahidi, F, and M. Naczk. 2004. Phenolic compounds of beverages, 1047-1054. p. 241-248. In Phenolics in food and neutraceuticals. CRC Press, 14. Kim, S., C. Ruengwilysup, and D. Y. Fung. 2004. Antibacterial ef - Boca Raton. Fla. fect of water-soluble tea extracts on foodborne pathogens in labo- 20. Smith, J. P., D. P. Daifas, W. El-Khoury, J. Koukoutsis, and A. El- ratory medium and in a food model. J. Food Prot. 67:2608-2612. Khoury. 2004. Shelf life and safety concerns of bakery products- 15. Little, C. L., J. Barnes, and R. T. Mitchell. 2002. Microbiological a review. Crit. Rev. Food Sd. Nutr. 44:19-55. quality of take-away cooked rice and chicken sandwiches: effective- 21. Taguri, T., T. Tanaka. and I. Kouno. 2004. Antimicrobial activity of ness of food hygiene training of the management. Commun. Dis. 10 different plant polyphenols against bacteria causing food-borne disease. Biol. Pharm. Bull. 27:1965-1969. Public Health 5:289-298. 22. Turnbull, P. C., N. M. Sirianni, C. I. LeBron, M. N. Samaan, E N. 16. Mohammed, M. J., C. K. Marston, T. Popovic, R. S. Weyant, and F Sutton, A. E. Reyes, and L. F Peruski, Jr. 2004. MICs of selected C. Tenover. 2002. Antimicrobial susceptibility testing of Bacillus an- antibiotics for Bacillus anthracis, Bacillus cereus, Bacillus thurin- thracis. comparison of results obtained by using the National Com- giensis, and Bacillus ,nycoides from a range of clinical and environ- mittee for Clinical Laboratory Standards broth microdilution refer- mental sources as determined by the Etest. J. C/in. Microbiol. 42: J. Gun. Microbiol. ence and Etest agar gradient diffusion methods. 3626-3634. 40:1902-1907. 23. Yee, Y. K., M. W. Koo. and M. L. Szeto. 2002. Chinese tea con- 17. Nam, S. H., S. P. Choi, M. Y. Kang, H. J. Koh, N. Kozukue. and sumption and lower risk of He/icobacter infection. J. Gastroenterol. M. Friedman. 2005. Bran extracts from pigmented rice seeds inhibit Hepatol. 17:552-555. tumor promotion in lymphoblastoid B cells by phorbol ester. Food 24. Zhao, W. I-I., Z. Q. Ho, V. Hara. and T. Shimamura. 2002. Inhibition Chem. Toxicol. 43:741-745. of penicillinase by epigallocatechin gallate resulting in restoration of 18. Nauta, M. J., S. Litman, G. C. Barker, and F Carlin. 2003. A retail antibacterial activity of penicillin against penicillinase-producing and consumer phase model for exposure assessment of Bacillus Ce- Staphylococcus aureus. Antimicrob. Agents Chemother. 46:2266- reus. mt. J. Food Microhiol. 83:205-218. 2268.