Antibacterial Activity of Tea and Coffee: Their Extracts and Preparations

Abstract

Tea, Camellia sinensis (L) O. Kuntze (Theaceae) and coffee, Coffea arabica Linn. (Rubiaceae) possess antimicrobial activity in addition to various biological properties. In this study six strains of human pathogenic bacteria were assessed for their sensitivity to aqueous extracts of tea and coffee along with their preparations i.e. supplemented with milk and sugar. Antibacterial potential of the extracts has been compared with some of the commonly employed antibiotics. Bactericidal activity of the selected extracts was assessed by viable cell count method. Both tea and coffee inhibited bacteria to a variable extent and retained their antibacterial activity even after addition of milk and sugar. Equal effectiveness of tea/coffee extracts and their preparations justifies their potential as antibacterial agents.

Full text

Letters to the Editor / International Journal of Antimicrobial Agents 31 (2008) 380–399 393
In vitro antibacterial activity of three plants belonging to
the family Umbelliferae
Sir,
Plant-based medicine has served mankind for centuries
and an impressive number of modern drugs have been iso-
lated from natural sources [1]. The last 20–25 years have
witnessed renewed interest in folkloric remedies to fight
microbial diseases owing to the emergence of multidrug-
resistant microorganisms [2–4].
The aim of the present study was to expand the antibac-
terial spectrum from natural resources and to validate
traditional uses of three medicinal plants, namely Anethum
graveolens, Foeniculum vulgare and Trachyspermum ammi
(family Umbelliferae). These plants are used against a variety
of gastrointestinal disorders and also as spices and condi-
ments. Scant literature is available on the antimicrobial effect
of their extracts. To our knowledge, the present study is the
first systematic positive report on the efficacy of their aqueous
extracts.
Reference bacterial strains were obtained from Microbial
Type Culture Collection (IMTECH, Chandigarh, India). A
4-h activated bacterial suspension was used as the inocu-
lum. Seeds of different plants were obtained from the local
market and their aqueous extracts were prepared accord-
ing to Arora and Kaur [4]. Organic (hexane, ethyl acetate,
acetone and ethanol) extracts were prepared by soaking a
weighed amount of sample in a known volume of solvent
for 24 h to achieve the desired concentration (20%). After
filtration through Whatman filter paper no. 1, the super-
natant was used for further testing. The antibacterial activity
of extracts was assessed by the Kirby–Bauer agar diffu-
sion assay (ADA) [5] and compared with eight commonly
employed antibiotics. The minimum inhibitory concentration
(MIC) was determined by agar dilution method and the bac-
tericidal activity was determined by viable cell count (VCC)
studies [6].
The different plant extracts showed inhibitory activity
against all the tested bacteria, except Klebsiella pneumo-
niae 1 and 2 and Pseudomonas aeruginosa 1. Hot water
extracts gave inhibitory zones ranging from 12 mm to 25 mm,
whilst organic extracts gave zone sizes of 9–30mm; only the
data pertaining to hot water and acetone extracts have been
presented (Table 1). Staphylococcus aureus was the most sen-
sitive organism, closely followed by Enterococcus faecalis,
whilst P. aeruginosa 2 was the most sensitive among the
Gram-negative bacteria. Plant extracts were better or equally
effective against some of the bacterial strains compared with
antibiotics (Table 1).
Acetone extracts were more effective compared
with aqueous extracts, with MICs of 0.5–1.5% and
2–8%, respectively. During VCC studies, different plant
extracts caused 100% killing of the cells within 10–14 h
(Fig. 1).
The results of the present study are encouraging as all
the tested plants showed antibacterial potential, although the
method of extracting the plant material, its form (whether
crushed or finely powdered) and the choice of solvent affected
their antibacterial activity. The powdered seed extracts
showed no or very little antibacterial activity compared with
crushed seed extracts, which could be attributed to the heat
generated during grinding.
Table 1
Antibacterial activity of aqueous and acetone extracts of various plants compared with standard antibiotics
BacteriaaZone of inhibition (mm)
ExtractsbAntibiotics (␮g/disk)c
P1 P2 P3 A 10 Cfx 5 C 30 Co 25 G 10 I 10 Pt 10 Tb 10
Aq Ac Aq Ac Aq Ac
EF 24 26 22 29 22 28 21 – 23 – 11 23 16 09
SA 25 30 24 28 23 30 30 21 22 21 17 39 33 18
EC 12 24.5 15 22 17 23 20 20 19 21 18 28 20 16
KN1 – – – – –13 –22232414211817
KN2 –11 – – –12 –22242213 – – –
PA 1 – 12 – 9 – 15 – – 16 – 17 21 17 19
PA2242424252425 – – – –13261818
ST – 18 12 20 – 18 10 17 19 16 15 21 17 13
STM11215 –13 –142623322220302620
STM 2 20 26 14 26 12 24 – 20 23 18 21 24 17 17
SF 14 25 18 26 15 24 23 23 25 18 27 32 23 17
Mean 11.91 19.23 11.73 18.00 10.18 20.55 11.82 15.27 20.55 12.82 16.00 24.09 18.64 14.91
S.D. 10.492 8.948 10.061 10.826 10.381 6.362 12.287 9.951 7.942 10.400 3.098 9.741 8.028 5.804
–, no zone of inhibition; S.D., standard deviation.
aEF, Enterococcus faecalis; SA, Staphylococcus aureus; EC, Escherichia coli; KN, Klebsiella pneumoniae;PA,Pseudomonas aeruginosa;ST,Salmonella
typhi; STM, Salmonella typhimurium;SF,Shigella flexneri.
bP1, Anethum graveolens; P2, Foeniculum vulgare; P3, Trachyspermum ammi; Aq, aqueous extract; Ac, acetone extract.
cA 10, ampicillin; Cfx 5, cefixime; C 30, chloramphenicol; Co 25, co-trimoxazole; G 10, gentamicin; I 10, imipenem; Pt 10, piperacillin/tazobactam; Tb 10,
tobramycin.

394 Letters to the Editor / International Journal of Antimicrobial Agents 31 (2008) 380–399
Fig. 1. Viable cell count (VCC) of various bacteria with aqueous extracts of (a) Anethum graveolens, (b) Foeniculum vulgare and (c) Trachyspermum ammi.EF,
Enterococcus faecalis; SA, Staphylococcus aureus; EC, Escherichia coli;PA,Pseudomonas aeruginosa;ST,Salmonella typhi; STM, Salmonella typhimurium;
SF, Shigella flexneri.
The higher sensitivity of S. aureus and E. faecalis could
be attributed to physiological differences among Gram-
positive and Gram-negative bacteria. In an earlier study [2],
aqueous and hexane extracts of F. vulgare and T. ammi
did not show any antibacterial activity, whilst alcoholic
extracts gave lesser activity compared with our study against
Escherichia coli,P. aeruginosa,Salmonella typhimurium and
S. aureus. In another study, methanolic extracts of A. grave-
olens did not show any antibacterial activity [7], whilst its
ethanolic extract in the present study showed reasonable
antibacterial potential. Such variations could be attributed
to environmental conditions, strain differences and protocols
used.
Acetone extracts of all the plants exhibited greater activity
than the corresponding aqueous extracts, although the dif-
ference observed was not statistically significant except for
T. ammi. Stronger extraction capacity and thus better effi-
cacy of acetone was further supported by MIC studies. VCC
studies further supported the data obtained by ADA
and MIC studies. Staphylococcus aureus was completely
killed after 8 h of incubation; however, Salmonella typhi,
being the least sensitive, took the longest time for 100%
killing. Plant extracts showed statistically significant activ-
ity in comparison with antibiotics (P< 0.05) against E.
faecalis and P. aeruginosa 2, which, however, were resis-
tant to some of the commercially available antibiotics
(Table 1).
Most of the microorganisms are developing resistance to
commonly employed antibiotics and are a common cause of
nosocomial infections. Microbial pathogens such as E. coli
and Shigella flexneri are associated with gastrointestinal dis-
orders. Thus, the antibacterial activities of medicinal plants
reported in the present study are noteworthy considering their
importance in nosocomial infections and treatment of gastric
ailments.
In conclusion, all three extracts possessed equally good
inhibitory activity against all the tested bacteria. Aqueous as
well as acetone extracts showed almost comparable antibac-
terial activity, supporting the traditional use of these plants.
Thus, the present study reflects a hope for the development
of novel chemotherapeutic agents.
Funding: No funding sources.
Competing interests: None declared.
Ethical approval: Not required.
References
[1] Cowan MM. Plant products as antimicrobial agents. Clin Microbiol Rev
1999;12:564–82.
[2] Ahmad I, Mehmood J, Mohammad F. Screening of some Indian
medicinal plants for their antimicrobial properties. J Ethnopharmacol
1998;62:183–93.
[3] Arora DS, Kaur J. Antimicrobial activity of spices. Int J Antimicrob
Agents 1999;12:257–62.
[4] Arora DS, Kaur GJ. Antibacterial activity of some Indian medicinal
plants. J Nat Med 2007;61:313–7.
[5] Bauer AW, Kirby WMM, Sherris JC, Turck M. Antibiotic suscepti-
bility testing by a standardized single disk method. Am J Clin Pathol
1966;43:493–6.
[6] Toda M, Okubo S, Hiyoshi R, Shimamura T. The bactericidal activity of
tea and coffee. Lett Appl Microbiol 1989;8:123–5.
[7] Bazzaz BSF,Haririzadeh G. Screening of Iranian plants for antimicrobial
activity. Pharm Biol 2003;41:573–83.

Letters to the Editor / International Journal of Antimicrobial Agents 31 (2008) 380–399 395
Gurinder Jeet Kaur
Daljit Singh Arora ∗
Microbial Technology Laboratory, Department of
Microbiology, Guru Nanak Dev University, Amritsar
143005, Punjab, India
∗Corresponding author. Tel.: +91 183 2258802–09x3316;
fax: +91 183 2258819–20.
E-mail address: daljit 02@yahoo.co.in (D.S. Arora)
doi: 10.1016/j.ijantimicag.2007.11.007
Post-antibiotic effects of linezolid and other agents
against Mycobacterium tuberculosis
Sir,
The persistent suppression of bacterial growth after
limited exposure to an antimicrobial is known as the post-
antibiotic effect (PAE). Recognition of PAE is important in
helping to design the appropriate drug dosing interval. The
oxazolidinone linezolid has been reported to possess good in
vitro antituberculosis activity [1]. Combination of linezolid
and fluoroquinolones has been shown to yield synergistic
activity against Mycobacterium tuberculosis [2]. It appears
that the optimal design of new dosing regimens would benefit
from further pharmacokinetic and pharmacodynamic stud-
ies.
In vitro PAEs of rifampicin, isoniazid, amikacin, strep-
tomycin, ethambutol, pyrazinamide, ofloxacin, moxifloxacin
and rifapentine, alone or in combination, had been previ-
ously studied [3,4]. In this study, we examined the PAEs of
linezolid, gatifloxacin, capreomycin and rifampicin as single
agents and in combination against M. tuberculosis. Line-
zolid and gatifloxacin are potential new antimycobacterial
agents. Rifampicin and capreomycin were also included as
they represent the key members of first- and second-line anti-
tuberculosis drugs, respectively.
The standard strain of M. tuberculosis H37Rv (ATCC
27294) was used. The minimal inhibitory concentra-
tions (MICs) of linezolid, gatifloxacin, capreomycin and
rifampicin were 0.5, 0.25, 2.5 and 0.5 mg/L, respectively,
as determined by the broth macrodilution method [5]. The
PAEs of the four agents were assessed individually using
drug concentrations falling into the likely therapeutic ranges
in humans (linezolid 10 mg/L; gatifloxacin 3 mg/L; capre-
omycin 40 mg/L; and rifampicin 8–16 mg/L). Gatifloxacin
and linezolid drug powders were gifts from Bristol-Myers
Squibb Pharmaceuticals (Princeton, NJ) and Pfizer Corp.
(Kalamazoo, MI), respectively. Rifampicin and capreomycin
drug powders were purchased from Sigma Co. (St Louis,
MO). Stock solutions were stored at −70 ◦C. For each
experiment, aliquots of the stock solutions were diluted in
Middlebrook 7H9 broth supplemented with 2% glycerol
and 10% oleic acid–dextrose–catalase (Difco Laboratories,
Sparks, MD).
The PAEs of the four drugs alone and in combination
were determined. Briefly, a homogeneous suspension of bac-
terial cells at 1 McFarland standard was obtained from a
2-week-old solid culture and stored at −70 ◦C in 1.0 mL
aliquots. A single vial of cells was inoculated into 10 mL
of BACTEC 12B medium (Becton Dickinson, Sparks, MD)
supplemented with 2.5% PANTA reconstituting fluid (PRF)
(polyoxyethylene stearate-containing fluid without antibi-
otics). This suspension was incubated for 15 days at 37 ◦C
to obtain mycobacteria in late logarithmic phase. The seed
vial was sonicated for 3 min in a Branson ultrasonic water-
bath. Nine millilitres of the various drugs with concentrations
equivalent to 12–32 times their MICs, either alone or in com-
bination, and a drug-free control were inoculated with 1.0 mL
of the prepared seed (final inoculum 1–3 ×106colony-
forming units/mL). After 2 h of incubation, drug removal was
accomplished by a 1:1000 dilution into fresh pre-warmed
BACTEC 12B medium supplemented with 2.5% PRF. Con-
trols containing similarly diluted drugs and serial dilutions of
previously unexposed organisms were included. All experi-
ments were carried out thrice in duplicate on different days.
Inoculated BACTEC vials were incubated and read daily on
a BACTEC 460 instrument until the BACTEC growth index
(GI) reached 999. The viable numbers of organisms, imme-
diately before and after drug exposure, were determined by
plating appropriate dilutions onto Middlebrook 7H10 agar
slopes (Difco Laboratories) in screw-cap flasks and incu-
bating for 3–4 weeks. The PAEs were calculated using the
formula PAE = TE−TC, where TEand TCrepresent, respec-
tively, the time for the exposed and control cultures to reach
the cumulative GI values of 100.
The PAE results are shown in Table 1. Rifampicin at
16 mg/L was found to have the longest median PAE of 37.3 h.
When its concentration was lowered to 8 mg/L the median
PAE was also shortened to 23.3 h, probably reflecting con-
centration dependence as shown previously [3]. Gatifloxacin
at 3 mg/L showed a limited median PAE of 8.8 h, similar to
those of other fluoroquinolones that we have investigated pre-
viously [3,4]. Linezolid only showed a minimal PAE, with a
median duration of only 4 h. Both linezolid and gatifloxacin
had a significantly shorter duration of PAE compared with
capreomycin and rifampicin (at 8 mg/L and 16 mg/L) when
tested singly (P< 0.05). Linezolid conferred no additive or
synergistic PAE when added to rifampicin, gatifloxacin or
capreomycin (P> 0.05).
When used in the treatment of multidrug-resistant tubercu-
losis (MDR-TB), linezolid at one-half the daily dose (600 mg
once daily instead of 600 mg twice daily) could reduce
adverse reactions, especially haematological ones. Our pre-
liminary findings regarding the PAE of linezolid, coupled
with its known half-life of only 5–7 h [6], would suggest
a possible need for twice-daily dosing in the treatment of
MDR-TB. This dosing requirement might prove to be a dis-
advantage in the long-term use of the drug owing to its
attendant toxicity and cost. Further investigation in this direc-
tion appears warranted.