THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINEVolume 13, Number 10, 2007, pp. 1119–1124© Mary Ann Liebert, Inc.DOI: 10.1089/acm.2007.7033

Selective Microbiologic Effects of Tea Extract on CertainAntibiotics Against Escherichia coli In Vitro

TIRANG R. NEYESTANI, Ph.D., NILOUFAR KHALAJI, M.Sc., and A’AZAM GHARAVI, Bs.C.

ABSTRACT

Objectives: This study evaluated the microbiologic effects of black tea, compared to green tea, alone and inconjunction with selected antibiotics against Escherichia coli, the common cause of intestinal and urinary tractinfections.

Design: This study was an in vitro evaluation of antibacterial effects of tea extracts.Methods: Black and green tea extracts were analyzed by using high-performance liquid chromatography to

compare their major polyphenol profiles. Different concentrations of the extracts or gallic acid (GA), the phe-nolic compound found with high concentration in the black tea extract, were employed for bacterial sensitiv-ity tests, using pour plate and disc diffusion methods. The latter was used to evaluate the interactions betweenthe extracts and certain anti–E. coli antibiotics.

Results: GA in black tea extract and epigallocatechin and epigallocatechin gallate in green tea extract arepresent in the highest concentrations, respectively. At concentrations of 25 mg/mL, both black and green teasafter 5 and 7 hours completely inhibited E. coli growth. GA at concentrations of 5, 10, and 25 �g/mL after 7,5 and 3 hrs, respectively, inhibited bacterial growth. Both black and green tea extracts had either synergisticor antagonistic effects at different concentrations on selected antibiotics, while GA showed a synergistic effectwith all the antibiotics tested in a dose-dependent manner. The effect was more prominent with amikacin andsulfamethoxazole.

Conclusions: The microbiologic effects of both black tea and green tea extracts on certain antibiotics againstE. coli may vary, depending on the type of the tea extract (i.e., black vs. green), the amount of the extract, andthe antibiotic being used.

1119

INTRODUCTION

Different strains of Escherichia coli are common causesof intestinal infections in infants and create urinary in-

fections and gastroenteritis in adults.1–3 There is a popularbelief in Iran that drinking black tea extract, and sometimeseven chewing its dried leaves, can alleviative effects in bothintestinal and urinary infections. Though it may not be ac-tually effective against all types of infectious agents, tea ex-tract demonstrated antidiarrheal activity in animal models4

and inhibited net fluid and electrolyte losses in diarrheacaused by enterotoxigenic E. coli.5 The microbiologic ef-

fects of some herbs, including tea leaves, are attributed tothe polyphenolic compounds,6,7 the potent antimicrobial8

and antioxidant9 agents. Black tea, a major source of dietaryphenolics,10 including flavins and thearubigins,11 has beenshown to have antimicrobial properties both in vivo and invitro.12 As herbal medicines are usually considered as anadjunct therapy, the interaction between the herb and med-ications requires investigation. The synergy between tea andthe antibiotics may, however, be considered (if not ne-glected) as a preassumption. Though this synergistic effect,for some antibiotics, has already been demonstrated,13 thismay not be a general rule.

Laboratory of Nutrition Research, National Nutrition and Food Technology Research Institute, Shaheed Beheshti Medical Univer-sity, Tehran, Iran.

In this study, we first examined the antibacterial proper-ties of both black and green tea extracts against E. coli. Chro-matographic analyses showed that gallic acid (GA), had thehighest amount of the phenolic compounds in the black teainfusion, so it was also evaluated for its antimicrobial ef-fects. Further, the concomitant effects of the tea extract, aswell as of GA, and certain antibiotics against E. coli wereinvestigated in vitro.

MATERIALS AND METHODS

Materials

The bacterial strain (ATCC 25920) was purchased fromthe Iranian Organization of Scientific and Industrial Re-searches (Tehran). All culture media were obtained fromMerck (Darmstadt, Germany). Polyphenol standards, in-cluding Polyphenon 60, were from Sigma Aldrich (St. Louis,

MO). The chromatographic solvents were all of high-per-formance liquid chromatography (HPLC)-grade (Romil,Cambridge, UK). Antibiotic discs, including gentamicin,sulfometoxazole, norfloxacin, and amikacin, were pur-chased from Padtan Teb (Tehran, Iran).

Preparation of tea extracts

Dried tea leaves (100 g, green or black) were repeatedlysoaked in 70% ethanol (2 L) and incubated at 60°C for 48hours, then the extract was filtered, using Whatman® paper(W&R Balston, Ltd., Kent, UK). The extract was concen-trated to a dried fine powder and then reconstituted in di-methyl sulfoxide (1000 mg/mL).

Total antioxidant capacity (TAC) assay

TAC was determined by using 2,2�-azinobis (3-ethyl-benzothiazoline-6-sulfonate).14 Results were compared withthe antioxidant activity of bovine serum albumin (BSA) as

NEYESTANI ET AL.1120

Name��280.0[1]

GA

EG

C

CA

F

EG

CG

A

0

0

Minutes

Volt

s

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0

250

500

750

1 G

A2

GC

3 C

. EG

C

4 C

AF

5 E

GC

G

6 ec

7 G

CG

8 E

GC

B

0 10

0

Minutes

mA

U

20

20 30 40 50 60 70

40

FIG. 1. High-performance liquid chromatography analysis of (A) black tea extract and (B) green tea extract (Polyphenon 60). GA,gallic acid; EGC, epigallocatechin gallate; CAF, caffeine; EGCG, epigallocatechin gallate.

the standard. Green and black tea extracts, at the concen-tration of 1 mg/mL, were tested for TAC and the resultswere expressed as the mmol/L of BSA equivalent.

Chromatographic analysis of black tea polyphenols

Green and black tea extracts were analyzed by using anHPLC system equipped with an ultrasonic degassing system(Young Lin, Seoul, South Korea) and the photodiode arraydetector 2800, with Chromgate software (Knauer, Berlin,Germany).

Typically, a 2.5-g tea bag (Noushine Chaai, Lahijan, Iran)was infused in deionized hot water for 15 minutes. Aftercooling, the extract was filtered (0.45-�m cartridge) and in-jected into the column. Polyphenon 60, a standard green teaextract, was also analyzed for polyphenolic compounds.Chromatographic conditions were as follows: column: C18Nova-Pak, 4 � 150 mm, 4 �m (Waters, Milford, MA); flowrate: 1.5 mL/min; pressure: 2700 psi; temperature: 30°C;mobile phase: (water:methanol: ortho-phosphoric acid,90:10:0.1, v:v:v); � max: 210 nm; run time: 25 minutes.

Bacterial sensitivity tests

Pour plate. Different concentrations of black or green teaextracts (6.25, 12.5, 25, 50, and 100 mg/mL) in brain heartbroth (final volume: 10 mL) were prepared. For the bacter-

ial suspension, 1 mL equivalent to 0.5 MacFarland wasadded to the suspensions and incubated at 37°C. After 1, 2,3, 5, 7, and 24 hours, 1 mL of the suspension was trans-ferred to a sterile petri dish containing liquid nutrient agarat 45°C and dispersed evenly. All plates were incubated at37°C and assessed after 24 hours. To evaluate the antibac-terial effects of GA, the same procedure was used with theconcentrations 2.5, 5, 10, 25, 50, 100, and 1000 �g/mL.

Disc diffusion. For black or green tea extract as well asGA (10, 50, and 100 mg/mL), 25 �L was inoculated to theblank or antibiotic discs. The discs were dried by incubat-ing at 37°C. The antibiotics used were amikacin (30�g/disc), sulfometoxazole (10 �g/disc), norfloxacin (10�g/disc), and gentamicin (10 �g/disc). The bacterial sensi-tivity disc diffusion test was repeated 8–11 times on differ-ent days. The mean diameter of growth inhibition was cal-culated and used for further statistical analyses.

Statistical analyses

Normality of data distribution was evaluated by usingKolmogrov-Smirnov. Comparison of means was performedwith the Student’s t-test or, when the distribution was notnormal, Mann-Whitney U-Wilcoxon. The predeterminedupper limit of significance throughout this investigation wasp � 0.05. All statistical analyses were performed by usingthe Windows/SPSS ver. 11.5 package.

RESULTS

Chromatographic analysis of green and black tea polyphenols

GA and caffeine were the most abundant constituents ofthe black tea, while epigallocatechin (EGC) and epigallo-catechin gallate (EGCG) were the prominent polyphenols ingreen tea (Fig. 1), which was consistent with the previousreports.15 Table 1 shows the retention times of caffeine andthe major polyphenols in tea.

TEA EXTRACT AND E. COLI 1121

TABLE 1. RETENTION TIMES OF CAFFEINE AND THREE

MAJOR POLYPHENOLIC COMPOUNDS OF TEA EXTRACT

WITH THE CONCENTRATIONS IN THE BLACK TEA

EXTRACT (MEAN � STANDARD DEVIATION)

Retention time ConcentrationConstituent (minutes) (mg/dL)

GA 1.5 3.5 � 0.8EGC 7.3 0.43 � 0.03Caffeine 11.5 30.9 � 0.4EGCG 18.58 0.25 � 0.01

GA, gallic acid; EGC, epigallocatechin; EGCG, epigallocatechingallate.

TABLE 2. EFFECT OF BLACK TEA EXTRACT (BTE) AND GREEN TEA EXTRACT (GTE) ON

THE GROWTH OF ESCHERICHIQA COLI AT DIFFERENT TIMES AND CONCENTRATIONS

Concentration (mg/mL)

6.25 12.5 25.0 50.0 100.0

Incubation time (hours) BTE GTE BTE GTE BTE GTE BTE GTE BTE GTE

1 �105 �105 �105 �103 �103 2 � 103 103 8 � 103 — —2 �105 �105 �105 �103 �103 5 � 102 7 � 102 50 — —3 �105 �105 �105 �103 �103 2 � 102 102 — — —5 �105 �105 �105 �103 �103 — — — — —7 �105 �105 �105 �103 — — — — — —24 �105 �105 �105 �103 — — — — — —

Total antioxidant capacity

The antioxidant capacities of green and black tea extractsat the concentration of 1 mg/mL were 2.68 � 0.11 and 1.22 �0.10 mM of BSA equivalent, respectively (p � 0.001).

The antibacterial effect of black and green teas

Both black and green teas, at the concentration of 25 mg/mL after 5 and 7 hours, respectively, completely inhibited E.coli growth (Table 2). GA at the concentrations as low as 5,10, and 25 �g/mL after 7, 5, and 3 hours, respectively, in-hibited bacterial growth. At higher concentrations, no growthwas observed, even after 1 hour incubation (Table 3).

Evaluation of tea extract interactions with some antibiotics

Adding 1.25 mg of black tea extract to the standard an-tibiotic discs suppressed the antibacterial effects of almostall of them, as judged by the bacterial growth inhibition, andthis suppression, with the exception of amikacin, was sta-tistically significant (p � 0.01). When doubling the amountof black tea extract, this inhibitory effect was reduced withnorfloxacin and sulfamethoxazole, though the diameters ofbacterial growth inhibition were still significantly lower thanthose of the standard discs (p � 0.02). With gentamicin, thediameter of bacterial growth inhibition was almost restoredto the baseline, while for amikacin the diameter was evenmore than that of the standard disc (p 0.014).

Green tea extract enhanced the antibacterial effects ofgentamicin and amikacin (p � 0.001) and this synergisticeffect was strengthened by the incremental amounts of greentea extract added to the standard discs (p � 0.001). How-ever, green tea extract, at the amount of 1.25 mg, showedan inhibitory effect on norfloxacin and sulfamethoxazole(p � 0.001). Increasing the amount of green tea extract to2.5 mg restored the growth inhibition diameter for sul-famethoxazole and enhanced the antibacterial effect of nor-floxacin, compared to the baseline (p � 0.001). GA showeda synergistic effect, with all the antibiotics tested in a dose-dependent manner (p � 0.001); this effect was more promi-nent with amikacin and sulfamethoxazole (Fig. 2).

NEYESTANI ET AL.1122

TABLE 3. EFFECT OF GALLIC ACID ON THE GROWTH OF ESCHERICHIA COLI AT DIFFERENT TIMES AND CONCENTRATIONS

Concentration (�g/mL)

Incubation time (hours) 2.5 5.0 10.0 25.0 50.0 100.0 1000.0

1 �104 8 � 103 3 � 103 5 � 102 — — —2 �104 103 5 � 102 30 — — —3 �104 2 � 102 80 — — — —5 �104 102 — — — — —7 �104 — — — — — —24 �104 — — — — — —

FIG. 2. The effect of adding different doses of (A) black tea ex-tract, (B) green tea extract, and (C) gallic acid on selected antibi-otic discs against Escherichia coli in a disc diffusion test.

C

B

A

DISCUSSION

Antimicrobial effects of tea are well known.16,17 It is com-monly thought that herbal extracts would be more effectiveif used in conjunction with conventional treatments.18 Forthis reason, some investigators have evaluated drug-herb in-teractions. Synergistic effects of tea with some antibioticshave been demonstrated.13,19 In a study, synergy betweengreen tea extract and levofloxacin against E. coli was ob-served.13 Our data, however, suggest selective, dose-depen-dent microbiologic effects of both green and black teas. Onemechanism for bacterial growth-inhibitory effects of plantpolyphenols (tannins) may be auto-oxidation and hydrogenperoxide production.20 However, in certain circumstances,some bacterial genes may be induced (such as OxyR in E.coli) that, by strengthening bacterial antioxidant defensemechanisms, may overcome tannin inhibitory effects.20

These bacterial defense mechanisms may be elicited in a teaextract in a dose-dependent manner. Another possible ex-planation for these observations may be the presence ofsome interference between tea constituents that, at certainconcentrations, may result in lowered antibacterial activity.This speculation is supported by the strong, dose-dependentsynergism between purified GA and all the antibiotics tested.

The most potent antimicrobial polyphenols of green tea,EGC and EGCG, were found in minute amounts in the blacktea. It is likely that microbiologic effects of black tea, alongwith its antioxidant capacity, are influenced by the processof fermentation. It has been suggested that even the manu-facturing season may affect the antimicrobial activity oftea.21 However, there were limitations in this study; this in-vestigation was in vitro, yet both antibiotics and polyphe-nolic compounds undergo metabolic processes in the bodyand there is less information on the interaction of the relatedmetabolites. Moreover, tissue distribution of ingestedpolyphenols is poorly understood. Some of the polyphenolsmay confer their beneficial effects to specific parts of thebody, in which they are concentrated.22 This explanation isfurther complicated by the immune system interactions withboth polyphenolic compounds23 and the invasive pathogen.Nevertheless, the in vitro bacterial sensitivity test is still aroutine procedure for clinical purposes, based on whichproper antibiotics are selected to treat the infection.24–26

CONCLUSIONS

As in this study, each 100 g of dried black tea leaves gave11 g of dried extract, so approximately 1.25 and 2.5 mg ofextract would be equal to about 11.4 and 22.7 g of blacktea, respectively. As each tea bag contains about 2.5 g, theabove numbers can be translated to around four to five andnine cups of tea a day, respectively. In the comprehensivefood consumption survey during 1999–2001, per capita con-sumption of dried tea was estimated at about 3.8 g/day (�2

cups a day).27 While it is less likely that the daily con-sumption of two to three cups of tea in patients underchemotherapy with the selected antibiotics is problematic,having proper amounts of tea in patients with E. coli ente-rocolitis or urinary tract infection that are under conven-tional antibiotic therapy may be proven to be helpful. Mean-while, the evaluation of antibacterial effects of GA, as anutritional supplement, can be the subject of future studies.Finally, from a herbal medicine point of view, the concen-tration of GA in the different tea varieties may be consid-ered as a measure of its possible curative effects. Our find-ings warrant further both in vitro and in vivo studies.

ACKNOWLEDGMENTS

This work was funded by the National Nutrition and FoodTechnology Research Institute (Tehran, Iran). All the labora-tory bench works have been done at the Laboratory of Nutri-tion Research, National Nutrition and Food Technology Re-search Institute.

REFERENCES

1. Villaseca JM, Hernandez U, Sainz-Espunes TR, Rosario C,Eslava C. Enteroaggregative Escherichia coli, an emergentpathogen with different virulence properties. Rev LatinoamMicrobiol 2005;47:140–159.

2. Garofalo CK, Hooton TM, Martin SM, Stamm WE, PalermoJJ, Gordon JI, Hultgren SJ. Escherichia coli from urine of fe-male patients with urinary tract infections is competent for in-tracellular bacterial community formation. Infect Immun2007;75:52–60.

3. Stapleton AE. Urinary tract infection in women: New patho-genic considerations. Curr Infect Dis Rep 2006;8:465–472.

4. Besra SE, Gomes A, Ganguly DK, Vedasiromoni JR. An-tidiarrhoeal activity of hot water extract of black tea (Camel-lia sinensis). Phytother Res 2003;17:380–384.

5. Bruins MJ, Cermak R, Kiers JL, van der Meulen J, vanAmelsvoort JM, van Klinken BJ. In vivo and in vitro effectsof tea extracts on enterotoxigenic Escherichia coli–induced in-testinal fluid loss in animal models. J Pediatr GastroenterolNutr 2006;43:459–469.

6. Si W, Gong J, Tsao R, Kalab M, Yang R, Yin Y. Bioassay-guided purification and identification of antimicrobial com-ponents in Chinese green tea extract. J Chromatogr A2006;1125:204–210.

7. Heredia N, Escobar M, Rodriguez-Padilla C, Garcia S. Ex-tracts of Haematoxylon brasiletto inhibit growth, verotoxinproduction, and adhesion of enterohemorrhagic Escherichiacoli O157:H7 to HeLa cells. J Food Prot 2005;68:1346–1351.

8. Taguri T, Tanaka T, Kouno I. Antimicrobial activity of 10 dif-ferent plant polyphenols against bacteria causing food-bornedisease. Biol Pharm Bull 2004;27:1965–1969.

9. Sang S, Hou Z, Lambert JD, Yang CS. Redox properties oftea polyphenols and related biological activities. Antioxid Re-dox Signal 2005;7:1704–1714.

TEA EXTRACT AND E. COLI 1123

10. Rechner AR, Wagner E, Van Buren L, Van De Put F, Wise-man S, Rice-Evans CA. Black tea represents a major sourceof dietary phenolics among regular tea drinkers. Free RadicRes 2002;36:1127–1135.

11. Luczaj W, Skrzydlewska E. Antioxidative properties of blacktea. Prev Med 2005;40:910–918.

12. Bandyopadhyay D, Chatterjee TK, Dasgupta A, Lourduraja J,Dastidar SG. In vitro and in vivo antimicrobial action of tea:The commonest beverage of Asia. Biol Pharm Bull2005;28:2125–2127.

13. Isogai E, Isogai H, Hirose K, Hayashi K, Hayashi S, OgumaK. In vivo synergy between green tea extract and levofloxacinagainst enterhemorrhagic Escherichia coli 0157 infection.Curr Microbiol 2001;42:248–251.

14. Rice-Evans, Miller NJ. Total antioxidant status in plasma andbody fluids. Meth Enzymol 1994;234:279–293.

15. Zuo Y, Chen H, Deng Y. Simultaneous determination of cat-echins, caffeine, and gallic acids in green, Oolong, black, andpu-erch teas using HPLC with photodiodie array detector.Ta-lanta 2002;57:307–316.

16. Bandyopadhyay D, Chatterjee TK, Dasgupta A, Lourduraja J,Dastidar SG. In vitro and in vivo antimicrobial action of tea:The commonest beverage of Asia. Biol Pharm Bull 2005;28:2125–2127.

17. Hamilton-Miller JMT. Antibacterial properties of tea (Camelliasinensis L.). Antimicrob Agents Chemother 1995;39:2375–2377.

18. Sadeghian S, Neyestani TR, Shirazi MH, Ranjbarian P. Bac-teriostatic effect of dill, fennel, caraway, and cinnamon ex-tracts against Helicobacter pylori. J Nutr Enviro Med 2005;15:47–55.

19. Tiwari RP, Bharti SK, Kaur HD, Dikshit RP, Hoondal GS.Synergistic antimicrobial activity of tea and antibiotics. IndianJ Med Res 2005;122:80–84.

20. Smith AH, Imlay JA, Mackie RI. Increasing the oxidativestress response allows Escherichia coli to overcome inhibitoryeffects of condensed tannins. Appl Environ Microbiol 2003;69:3406–3411.

21. Chou CL, Lin LL, Chung KT. Antimicorbial activity of tea asaffected by the degree of fermentation and manufacturing sea-son. Int J Food Microbiol 1999;48:125–130.

22. Corder R, Mullen W, Khan NQ, Marks SC, Wood EG, Car-rier MJ, Crozier A. Oenology: Red wine procyanidins and vas-cular health. Nature 2006;444:566.

23. Hofmann T, Liegibel U, Winterhalter P, Bub A, Rechkem-mer G, Pool-Zobel BL. Intervention with polyphenol-rich fruit juices results in an elevation of glutathione S-transferase P1 (hGSTP1) protein expression in humanleucocytes of healthy volunteers. Mol Nutr Food Res2006;50:1191–1200.

24. Metin DY, Hilmioglu-Polat S, Hakim F, Inci R, Tumbay E.Evaluation of the microdilution, Etest, and disk diffusionmethods for antifungal susceptibility testing of clinicalstrains of Trichosporon spp. J Chemother 2005;17:404–408.

25. Krobot K, Yin D, Zhang Q, Sen S, Altendorf-Hofmann A,Scheele J, Sendt W. Effect of inappropriate initial empiricantibiotic therapy on outcome of patients with community-acquired intra-abdominal infections requiring surgery. EurJ Clin Microbiol Infect Dis 2004;23:682–687.

26. Miller D, Alfonso EC. Comparative in vitro activity of lev-ofloxacin, ofloxacin, and ciprofloxacin against ocular strep-tococcal isolates. Cornea 2004;23:289–293.

27. Kalantari N, Ghafarpour M. National Comprehensive Studyon Household Food Consumption Pattern and NutritionalStatus, IR Iran, 2001–2003 [national report]. Tehran, Iran:Shaheed Beheshti Medical University, National Nutritionand Food Technology Research Institute, Department of Nu-trition Research, 2005.

Address reprint requests to:Tirang R. Neyestani, Ph.D.

Laboratory of Nutrition ResearchNational Nutrition and Food Technology Research

InstituteP.O. Box 19395-4741Tehran, 1981619573

Iran

E-mail: tneyestani@nnftri.ac.ir

NEYESTANI ET AL.1124