Content uploaded by Wandee Gritsanapan
Author content
All content in this area was uploaded by Wandee Gritsanapan
Content may be subject to copyright.
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Original Paper
Med Princ Pract 2010;19:281–286
DOI: 10.1159/000312714
Anti-Acne-Inducing Bacterial Activity of
Mangosteen Fruit Rind Extracts
Werayut Pothitirat a, c Mullika Traidej Chomnawang b Wandee Gritsanapan a
Departments of
a Pharmacognosy and b Microbiology, Faculty of Pharmacy, Mahidol University, and
c Faculty of
Pharmacy, Siam University, Bangkok, Thailand
acnes and S. epidermidis . Using HPLC, the dichloromethane
extract yielded the highest content (46.21% w/w) of ␣ -man-
gostin followed by the ethanol extract (18.03% w/w), the
hexane extract (17.21% w/w) and the water extract (0.54%
w/w). Conclusions: Dichloromethane extract exhibited the
strongest anti-acne-inducing bacterial effect and this ex-
tract yielded the highest amount of ␣ -mangostin.
Copyr ight © 2010 S. Karge r AG, Base l
Introduction
Acne vulgaris is a cutaneous pleomorphic disorder of
the pilosebaceous unit involving abnormalities in sebum
production and is characterized by both inflammatory
and noninflammatory lesions. It has painful social and
psychological effects on patients. The affected areas con-
tain the largest oil glands, including the face, back and
trunk. It is characterized by comedones, papules, pus-
tules, cysts, nodules and often scars
[1] . The pathogenesis
of acne is multifactoral, namely hormonal imbalance,
bacterial infection, stress, food or cosmetic application.
Propionibacterium acnes and Staphylococcus epidermi-
dis , which are bacteria isolated from the skin surface, are
often involved in the development of acne
[2] . P. acn es are
obligate anaerobic bacteria and act as an immunostimu-
lator that produces a variety of biologically active mole-
Key Words
Acne vulgaris ⴢ Antiacne ⴢ Mangosteen ⴢ Mangostin ⴢ
Propionibacterium acnes ⴢ Staphylococcus epidermidis
Abstract
Objective: The aims of this study were to determine the
most effective solvent extract of mangosteen, anti-acne-
inducing bacterial activity and the amount of ␣ -mangostin,
a major active component in each mangosteen fruit rind
extract, using high-performance liquid chromatography
(HPLC). Materials and Methods: The fruit rinds of mango-
steen were extracted with hexane, dichloromethane, etha-
nol and water. The extracts were tested for antibacterial ac-
tivity against bacteria that induce acne, including Propioni-
bacterium acnes and Staphylococcus epidermidis . Thin-layer
chromatographic autobiography against these bacteria was
also performed for each ex tract, while the ␣ -mangostin con-
tent was analyzed using a validated HPLC method. Results:
The dichloromethane extract exhibited the strongest anti-
bacterial effect with minimum inhibitory concentration val-
ues for both bacterial species at 3.91 g/ml, while the mini-
mum bactericidal concentration values against P. acn es
and S. epidermidis were 3.91 and 15.63 g/ml, respectively.
Thin-layer chromatographic autobiography indicated that
␣ -mangostin was present in all extracts, except the water
extract, and is a major active component against both P.
Recei ved: March 5, 2009
Revised: August 25, 2009
Dr. Wandee Gr itsanapan
Depa rtment of Pharmacognosy, Facu lty of Pharmac y, Mahidol University
447 Sr i-Ayudhay a Road, Ratchathewi, Ba ngkok 10400 (Thailand)
Tel. +66 26 4 486 7789, ext 1500/5530, Fa x +66 2644 8701
E-Mail pywgs
@ mahidol.ac.th
© 2010 S. Karger AG, Basel
1011–7571/10 /0194– 0281$26 .00/0
Accessible online at :
www.karger.com/mpp
Pothitirat /Chomnawang /Gritsanapan
Med Princ Pract 2010;19:281–286
282
cules and enzymes, such as lipases, proteases, hyaluroni-
dase and chemotactic factors, which influence the devel-
opment of inflammatory acne
[3] . S. epidermidis , an
aerobic organism, is usually involved in superficial infec-
tions within the sebaceous unit
[4] . The common therapy
for acne vulgaris includes oral and topical treatment us-
ing comedolytics and antibiotics
[5] . However, these ther-
apies can produce a number of potential side effects, in-
cluding the development of resistance to frequently used
antibiotics. Thus, the development of new antimicrobial
agents for resistant organisms is becoming critically im-
portant. Plants produce many secondary metabolites
with pharmacological activities that can be sources of
pharmacologically active agents against pathogenic mi-
cro-organisms.
The fruit of Garcinia mangostana Linn. or mango-
steen, of the family Guttiferae, is known as ‘the queen of
fruit’ due to its delicious taste and pleasant aroma. The
origin of this plant is in Southeast Asia
[6, 7] and it is
mainly distributed in Thailand, India, Sri Lanka, Myan-
mar, Indonesia, Malaysia, Philippines, China and other
tropical countries. The fruit rind of G. mangostana has
been used in Asian traditional medicines for the treat-
ment of skin infections, wounds, diarrhea, dysentery,
suppuration, leucorrhea, chronic ulcer and gonorrhea. It
contains high amounts of xanthones, such as ␣ -mangos-
tin, which is a major component and other bioactive sub-
stances including tannins, f lavonoids and polyphenolics
compounds
[8] . ␣ -Mangostin is soluble in alcohol, ether,
acetone, chloroform and ethyl acetate, while flavonoids
and polyphenolic compounds are soluble in water and
other polar solvents
[9, 10] . Recent reports have shown
that extracts from G. mangostana fruit rind have several
medicinal properties, such as antioxidant
[11–14] , anti-
inflammatory
[15] a nd also inhibition of HIV [16] . In ad-
dition, it has been known to promote high antimicrobial
activity against bacteria frequently involved in acne in-
flammation, P. ac nes and S. epidermidis [17] . This extract
is popularly used as a raw material in herbal cosmetics
and herbal drug preparations to prevent or treat acne.
Therefore, it is necessary to study the appropriate extrac-
tion solvent promoting high antiacne activity.
Materials and Methods
This study was to determine and compare anti-acne-inducing
bacterial activity by broth microdilution method and the content
of a major active component, ␣ -mangostin, in mangosteen fruit
rind extracts prepared using different solvents including hexane,
dichloromethane, ethanol and water. Thin-layer chromatograph-
ic (TLC) autobiography for antibacterial activity against P. acn es
and S. epidermidis o f e ac h e xt r ac t w as a ls o p er fo r me d. Th e e x tr ac t
with high anti-acne activity should be recommended as the ap-
propriate extract for further development of antiacne prepara-
tion.
Preparation of G. mangostana Fruit Rind Extract Plant
Material
The ripe fruits of G. mangostana were collected from Lang
Suan District, Chumphon Province, in the South of Thailand
in June 2006. The samples were identified by Dr. Wandee Grit-
sanapan at the Faculty of Pharmacy, Mahidol University, Bang-
kok, Thai land. The voucher specimen (WGM080 6) was deposited
at the Department of Pharmacognosy, Faculty of Pharmacy,
Mahidol University, Bangkok, Thailand.
The fresh fruits were cleaned and the edible aril parts were
removed. The fruit rinds were cut into small pieces and dried in
a hot air oven at 50
° C for 72 h. The dried samples were ground
into powder and passed through a sieve (20 mesh). The powdered
sample was kept i n an air-tight contai ner and protected from light
until used.
E x t r a c t i o n o f G. mangostana Fruit Rind Extract
Powd ered rinds (10 g) were extrac ted separate ly with 1,0 00 m l
of hexane, dichloromethane and 95% ethanol using a Soxhlet ap-
paratus. Each extract was fi ltered through a Whatma n No. 1 filter
paper. The filtrate was concentrated under reduced pressured at
50
° C using a rotary vacuum evaporator. To prepare a water ex-
tract, the powdered G. mangostana fruit rind (10 g) was boiled
with 200 ml of water for 1 h and filtered through a Whatman No.
1 filter paper. This method was repeated 5 times for exhaustive
extraction and the filtrates were combined and evaporated on a
boiling water bath. Each extraction was done in triplicate and
yields of the dried extracts were recorded.
Isolation of
␣
-Mangostin
␣ -Mangostin was separated from the dichloromethane ex-
tract of G. mangostana fruit rind by column chromatography
eluted with hexane, hexane/ethyl acetate and then ethyl acetate
with increasing polarity. After purification by recrystallization,
␣ -mangostin was identified by comparison of
1 H-NMR spectra
with reference data of ␣ -mangostin [18] . The purity of isolated ␣ -
mangostin was determined by high-performance liquid chroma-
tography (HPLC), which wa s used as a standa rd compound in this
study.
Anti-Acne-Inducing Bacterial Susceptibility Testing
The test organisms used in this study were P. acnes (ATCC
6919) and S. epidermidis (ATCC 14990), which were obtained
from the American Type Culture Collection, USA. Brain-heart
infusion and tryptic soy broth were purchased from DIFCO (De-
troit, Mich., USA).
P. a cne s was incubated in brain-heart in fusion for 72 h at 37
° C
under anaerobic conditions, while S. epidermidis was incubated
in tryptic soy broth for 24 h at 37
° C, and they were adjusted to
approximately 10
8 CFU/ml.
The minimal inhibitory concentration (MIC) values were de-
termined by 2-fold serial microdilution assay
[17] . The extracts
were incorporated into media to obtain a concentration of serial
dilution from 500 to 0.24 g/ml. A sample of 10 l standardized
Anti-Acne-Inducing Bacterial Activity of
Mangosteen
Med Princ Pract 2010;19:281–286
283
suspension of each tested organism was transferred to each well.
The broth cultures of S. epidermidis and P. acn es were incubated
for 24 and 72 h, respectively. The MIC defined as the lowest con-
centration of t he compound that can inhibit the micro-organisms
were determined . The minima l bactericidal concentr ation (MBC)
values of the extracts were recorded as the lowest concentration
that showed no visible growth after subculture of each clear well
onto a new plate containing suitable media.
TLC Analysis
The TLC chromatograms of each extract and of a reference
␣ -mangostin were performed on a precoated aluminium plate of
silica gel 60F254 (Merck KGaA, Darmstadt, Germany) using di-
chloromethane:methanol (96:
4) as a mobile phase. After remov-
ing the plate from the chamber, the plate was dried using an air
dryer and sprayed with 10% sulfuric acid in ethanol, followed by
heating at 110
° C for 10 min. The plate was examined under ultra-
violet light (366 nm). The R
f value of the main component was
determined by comparing it with the R
f value of the reference
standard. Videodensitometry of the chromatoplate was carried
out using CAMAG Reprostar 3 with cabinet cover and mounted
digital camera. The plates were run in duplicate; one set was used
as the reference chromatogram and the other for bioautography.
Bioautography
The developed TLC plates were carefully overlaid with nutri-
ent agar containing an aliquot of an overnight culture. The cul-
tures were incubated at 37
° C for 72 h under anaerobic condition
for P. acne s and 24 h for S. epidermidis . The plates were sprayed
with 1% of 2,3,5-triphenyl-tetrazolium chloride solution. The ar-
eas of inhibition were indicated by clear zones on the chromato-
gram and were compared with the R
f values of the related spots
on the reference TLC plate.
Determination of
␣
-Mangostin Content in Each Extract by
HPLC Method
Standard solutions of ␣ -mangostin (purity = 99.63%) were
prepared by diluting the stock solution (1,000 g/ml) with meth-
anol t o give the concentrat ion range o f 10–200 g/ml. For sample
preparations, 10 mg of each dried extract was dissolved in metha-
nol and the volume was adjusted to 10 ml in a volumetric flask
(concentration = 1 mg/ml). Aliquot of this solution (1.5 ml) was
Tab le 1. Yields of crude extracts, contents of ␣-mangostin in G. mangostana fruit rind extracts prepared using
different solvents and their MIC and MBC values against P. a cne and S. epidermidis
Solvent Yields of crude extracts
% dry weight
␣-mangostin content
% w/w of extract
Susceptibility of bacteria to various extracts, g/ml
P. acnes S. epidermidis
MIC MBC MIC MBC
HEX 0.9780.03 (7.54) 17.2180.50 7.81 7.81 15.63 31.25
DCM 8.0180.21 (1.39) 46.2182.46 3.91 3.91 3.91 15.63
EtOH 20.2080.36 (0.34) 18.0380.71 7.81 15.63 7.81 62.50
H2O27.5080.43 (0.02) 0.5480.01 500 >500 500 >500
HEX = Hexane extract; DCM = dichloromethane extract; EtOH = ethanol; H2O = water extracts. Figures in
parentheses indicate price in USD per gram of each extract.
abc
Fig. 1. Bioautograms against S. epidermi-
dis ( a ) and P. acn es ( c ) and TLC chromato-
grams (
b ) of various extracts of G. man-
gostana fruit rind using silica gel 60F254
as stationary phase and dichloromethane:
methanol (96:
4) as a mobile phase: T1 =
hexane extract; T2 = dichloromethane ex-
tract; T3 = ␣ -mangostin reference stan-
dard; T4 = ethanol extract; T5 = water ex-
tract. Arrows indicate clear zone.
Color vers ion available onlin e
Pothitirat /Chomnawang /Gritsanapan
Med Princ Pract 2010;19:281–286
284
diluted with methanol to make a final concentration of 150 g/
ml. The sample solution was filtered through a 0.45- m mem-
brane filter before injection.
A validated HPLC method [19] was performed on a Shimadzu
SCL-10A HPLC system, equipped with a UV-vis detector SPD-
10A. The separation was carried out using a Hypersil 쏐 BDS C18
column (250 ! 4.6 mm, 5 m size) with a C18 guard column. The
mobile phase consisted of 0.1% v/v orthophosphoric acid (A) and
acetonitrile (B). The elution was performed with gradient solvent
systems at a f low rate of 1 ml/min and monitoring at 320 nm. The
gradient program was as follows: 70% B for 0–15 min, 70–75% B
for 3 min, 75–80% B for 1 min, constant at 80% B for 6 min, 80–
70% B for 1 min and 11 min for postrunning for reconditioning.
The quantitative determination was analyzed with a CLASS VP
software program using the external calibration method.
0
05
a10 15 20 25 30 35
0.025
0.050
mAU
0
05
b10 15 20 25 30 35
0.025
0.050
mAU
0
05
c10 15 20 25 30 35
0.050
mAU
0
05d10 15 20 25 30 35
0.025
0.050
mAU
0
05
e10 15 20
Time (min)
25 30 35
0.025
0.050
mAU
Fig. 2. HPLC chromatograms of the fruit
rind extracts of G. mangostana : ␣ -man-
gostin standard (
a ), hexane extract ( b ), di-
chloromethane extract (
c ), ethanol extract
(
d ) and water extract ( e ).
Color vers ion available onlin e
Anti-Acne-Inducing Bacterial Activity of
Mangosteen
Med Princ Pract 2010;19:281–286
285
R e s u l t s
The TLC autobiographs of all the extracts of G. man-
gostana fruit rinds except that of water ( fig. 1 ) promoted
main inhibition zones at the same R
f value (R
f = 0.4). The
activity of ␣ -mangostin against P. a cnes and S. epidermi-
dis had the same MIC but different MBC values against
P. ac nes at 1.95 g/ml and S. epidermidis at 3.91 g/ml.
The dichloromethane extract of G. mangostana fruit rind
showed the strongest antibacterial activities against P.
acnes and S. epidermidis at the same MIC values of 3.91
g/ml but MBC values at 3.91 and 15.63 g/ml against P.
acnes and S. epidermidis , respectively. The hexane and
ethanol extracts promoted moderate activity with the
same MIC at 7.81 g/ml and MBC at 7.81 and 15.63 g/
ml, respectively, against P. ac nes . For S. epidermidis , the
hexane extract promoted the MIC at 15.63 g/ml and
MBC at 31.25 g/ml, while the ethanol extract gave the
MIC and MBC values at 7.81 and 62.50 g/ml, respec-
tively. On the other hand, the water extract showed the
lowest activity (MIC = 500 g/ml, MBC 1 500 g/ml for
both of the tested bacteria; table 1 ).
The content of ␣ -mangostin in each extract was in the
following order: dichloromethane extract (46.21% w/w),
ethanol extract (18.03% w/w), hexane extract (17.21%
w/w) and water (0.54% w/w), respectively. The purity of
isolated ␣ -mangostin extract was 99.6%.
D i s c u s s i o n
The results showed that hexane, dichloromethane
and ethanol extracts of mangosteen fruit rind exhibited
higher activity against P. a cnes than S. epidermidis ,
thereby implying that the polarity of extractant had in-
fluence upon antibacterial activity. Using HPLC analy-
sis, the highest content of ␣ -mangostin, the main xan-
thone found in G. mangostana fruit rinds, was found in
the dichloromethane extract followed by ethanol, hex-
ane and water extracts, respectively. These results show
that because of the moderate polarity of the chemical
structure of ␣ -mangostin, it is more soluble in a moder-
ate polarity solvent (dichloromethane) than a nonpolar
solvent (hexane) or polar solvents (ethanol or water)
[9] .
HPLC and TLC chromatograms of all extracts showed
different patterns, but ␣ -mangostin was found to be a
major component in all extracts, supporting the previ-
ous report
[20, 21] ( fig. 1 , 2 ). This compound also pro-
motes the strong antibacterial activities against P. acne s
and S. epidermidis . The results indicate that the antiacne
activity of the extracts might depend on the ␣ -mangos-
tin content.
Based on the current findings, dichloromethane
seemed to be the appropriate solvent to extract G. man-
gostana fruit rind for antiacne purposes because it pro-
moted the highest ␣ -mangostin content and anti-acne-
inducing bacterial activity. However, dichloromethane
could be toxic to humans
[22] . On the other hand, etha-
nol was found to be a less toxic
[23] and cheaper solvent
that also provided a high antiacne property. Thus, etha-
nol extract of G. mangostana fruit rind might be an al-
ternative choice for this purpose in pharmaceutical pro-
duction.
C o n c l u s i o n s
Dichloromethane provided the extract of G. man-
gostana fruit rind with high content of ␣ -mangostin and
high anti-acne-inducing bacterial activity. Considering
various factors, i.e. safety, cost and antiacne activity, eth-
anol extract might be used as an alternative natural source
in antiacne preparations instead of antibiotic drugs.
However, toxicology and clinical trials in animal models
have to be studied before use in humans.
References 1 Kumar A , Baboota S, Agar wal SP, Ali J, Ahu-
ja A: Treatment of acne with special empha-
sis on herbal remedies. Expert Rev Dermatol
2008;
3: 111–122.
2 Hamnerius N: Acne – aetiology and patho-
genesis. Treat Acne 1996;
32: 29–38.
3 Jeremy AH T, Holland DB, Robert SG , Thom-
son KF, Cunliffe WJ: Inflammatory events
are involved i n acne lesion initiat ion. J Invest
Dermatol 2003;
121: 20–27.
4 Burkhart CG, Burkhart CN, Lehmann PF:
Acne: a review of immunologic and micro-
biologic factors. J Postgrad Med 1999;
75:
328–331.
5 Clark C: Acne – general practice manage-
ment. Practitioner 1993;
237: 160–16 4.
6 Martin FW: Durian and mangosteen; in
Na g y S , Sh aw PE (e d s): Tr op ic a l a n d S ub t ro p-
ical Fruits. New York, AVI Publishing Inc,
1980, pp 401–414.
7 Kanch anapoom K, Kanc hanapoom M: Ma n-
gosteen; in Shaw PE, Chan Jr HT, Nagy S
(eds): Tropical and Subtropical Fruits. Au-
burndale, Agscience Inc, 1998, pp 191–215.
Pothitirat /Chomnawang /Gritsanapan
Med Princ Pract 2010;19:281–286
286
8 Pedraza-Chaverri J, Cárdenas-Rodríguez N,
Orozco-Ibarra M, Pérez-Rojas JM: Medici-
nal proper ties of mangost een (G arc ini a man -
gostana) . Food Chem Toxicol 2008;
46: 3227–
3239.
9 Budavari S: The Merck Index, ed 11. White-
house Station, Merck & Co, Inc, 1989.
10 Marston A, Hostettmann K: Separation and
quantification of flavonoids; in Andersen
OM, Markham KR (eds): Flavonoids: Chem-
istry, Biochemistry and Applications. Boca
Raton, CRC Press, 2006, pp 1–32.
11 Weechar angsan W, Opanasopit P, Sukma M,
Ngawhiru npat T, Sotanaphun U, Siripong P:
Antioxidative and neuroprotective activities
of extracts from t he fruit hull of mangosteen
( Garcinia mangostana Linn.). Med Princ
Pract 20 06;
15: 281–287.
12 Guzmán-Beltrán S, Orozco-Ibarra M,
González-Cuahutencos O, Victoria-Mares
S, Mercha nd-Reyes G, Medina- Campos ON,
Pedraza-Chaverri J: Neuroprotective effect
and reactive oxygen species scavenging ca-
pacity of ma ngosteen peric arp ext ract in cul-
tured neurons. Curr Top Nutr Res 2008;
6:
149 –158.
13 Márquez-Valadez B, Lugo-Huitrón R,
Valdivia-Cerda V, Miranda-Ramírez LR,
Pér ez-D e La C ruz V, Gonz ále z-C uahu tenc os
O, R ivero -Cru z I, Mata R, S anta maría R , Pe-
draza-Chaverrí J: The natural xanthone ␣ -
mangostin reduces oxidative damage in rat
brain tissue. Nutr Neurosci 2009;
12: 35–42.
14 Pedraza-Chaverrí J, Reyes-Fermín LM, No-
lasco-Amaya EG, Ibarra MO, Medina-Cam-
pos ON, González-Cuahutencos O, Rivero-
Cruz I, Mata R: ROS scavenging capacity
and neuroprotective effect of ␣ -mangostin
against 3-nitropropionic acid in cerebel lar
granule neurons. Exp Toxicol Pathol 2009;
61: 491–501.
15 Lin CN, Chung MI, Liou SJ, Lee TH, Wang
JP: Synthesis and anti-inf lammatory effects
of xanthone derivatives. J Pharm Pharmacol
1996;
48: 532–538.
16 Chen SX, Wan M, Loh BN: Active constitu-
ents against HIV-1 protease from Garcinia
mangostana . Planta Med 1996;
62: 381–382.
17 Chom nawang MT, Surassmo S, Nukoolkarn
VS, Gritsanapan W: Antimicrobia l effects of
Thai med icinal plant s against acne-i nducing
bacteria. J Ethnopharmacol 2005;
101: 330 –
333.
18 Iinuma M, Tosa H, Tanaka T, Asai F, Ko-
baya shi Y, Shi man o R, M iya uchi K: A nti bac-
terial activity of xanthones from guttifer-
aeous plants against methicillin-resistant
Staphylococcus aureus. J Pharm Pharmacol
1996;
48: 861–865.
19 Pot hitir at W, Gritsanapan W: HPLC qua nti-
tative analysis method for the determination
of ␣ -mangostin in mangosteen fruit rind ex-
tract. Thai J Agric Sci 2009;
42: 7–12.
20 Pothitirat W, Gritsanapan W: Quantitative
analysis of total mangostins in Garcinia
mangostana fru it rind. J Health Res 2008 ;
22:
161–166 .
21 Pothitirat W, Chomnawang MT, Supabphol
R, Gritsanapan W: Comparison of bioactive
compounds content, free radical scavenging
and anti-acne inducing bacteria activities of
extracts from the mangosteen fruit rind at
two stages of maturity. Fitoterapia 2009;
80:
442–447.
22 Skrabalak SD, Babish GJ: Safety standards
for occupational exposure to dichlorometh-
ane. Regul Toxicol Pharm 1983;
3: 139–143.
23 Greim H, Reuter U: Classification of carci-
nogenic chemicals in the work area by the
German MAK Commission: current exam-
ples for the new categories. Toxicology 2001;
166: 11–23.
