Bioscience, Biotechnology, and Biochemistry

ISSN: 0916-8451 (Print) 1347-6947 (Online) Journal homepage: https://www.tandfonline.com/loi/tbbb20

Estrogenic and Anti-Estrogenic Activities of the

Thai Traditional Herb, Butea superba Roxb.

Wichai CHERDSHEWASART, Tawatchai MAHAPANICHKUL & Chuenchit

BOONCHIRD

To cite this article: Wichai CHERDSHEWASART, Tawatchai MAHAPANICHKUL & Chuenchit

BOONCHIRD (2010) Estrogenic and Anti-Estrogenic Activities of the Thai Traditional Herb, Butea
superba Roxb., Bioscience, Biotechnology, and Biochemistry, 74:11, 2176-2182, DOI: 10.1271/
bbb.100159

To link to this article: https://doi.org/10.1271/bbb.100159

Published online: 22 May 2014.

Submit your article to this journal

Article views: 1085

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tbbb20
Biosci. Biotechnol. Biochem., 74 (11), 2176–2182, 2010

Estrogenic and Anti-Estrogenic Activities of the Thai Traditional Herb,

Butea superba Roxb.
Wichai C HERDSHEWASART,1 Tawatchai M AHAPANICHKUL,2;3 and Chuenchit B OONCHIRD2; y
1
Department of Biology, Faculty of Science, Chulalongkorn University,
Phyathai Road, Patumwan, Bangkok 10330, Thailand
2
Department of Biotechnology, Faculty of Science, Mahidol University,
Rama VI Road, Ratchathewi, Bangkok 10400, Thailand
3
Center of Excellence on Agricultural Biotechnology: (AG-BIO/PERDO-CHE), Bangkok 10400, Thailand

Received March 8, 2010; Accepted August 4, 2010; Online Publication, November 7, 2010
[doi:10.1271/bbb.100159]

This study evaluated the estrogenic and antiestro- menopausal symptoms.7) The tuberous plant materials
genic activities of native and in vitro hepatic metabolized from P. mirifica have exhibited stronger estrogenic
tuberous extracts of wild Butea superba collected from activity than that of Kudzu in ovariectomized rats that
23 out of the 76 provinces in Thailand by yeast estrogen was evaluated by both the vaginal cornification assay8)
screening (YES). The YES screen used consisted of the and uterotrophic assays.9)
human estrogen receptors hER and hER and the Butea superba Roxb., family Leguminosae, with the
human transcriptional intermediary factor 2 or human domestic Thai name of ‘‘red kwao krua,’’ is commonly
steroid receptor coactivator 1, respectively, together found in deciduous Thai forests and is an indigenous
with the -galactosidase expression cassette as the Thai herb popularly used for promoting male potency7)
reporter. The relative potency, effectiveness and relative as well as constituting a minor component of traditional
inductive efficiency were evaluated by determining the medicines made with P. mirifica for treating meno-
-galactosidase activity (EC50 ) of each tuberous extract pausal symptoms. It has exhibited androgen disruption
in relation to that induced by 17-estradiol. Six pure in male rats.10) The efficacy of B. superba powder has
compounds isolated from B. superba were tested in been demonstrated in a human clinical trial resulting in
parallel and exhibited a maximum relative potency the effective treatment of erectile dysfunction in Thai
compared to 17-estradiol of 15.5% and 5.27% in the males.11) B. superba products have recently been be-
respective hER and hER assays. Eighteen and coming popular as traditional medicines, dietary supple-
seventeen plant extracts were respectively found to ments and for topical applications to promote male
interact with the hER and hER receptors in the YES vigor. The traditional used of this plant ingredient
assays with higher relative potency and relative induc- needed to be scientifically evaluated to aid in an
tive efficiency with hER than with hER. The selected understanding of the bioactivity of the plant recipes.
plant extracts tested exhibited antiestrogenic activity. We therefore set up a YES-based study to evaluate the
Coincubation with the rat liver S9 mixture also elevated estrogenic activity of B. superba tuberous samples
the estrogenic potency of these plant extracts. collected from available sources in Thailand. The results
could make it possible not only to discover why the
Key words: phytoestrogen; Butea superba; yeast estro- plant is used in the recipes for treating menopausal
gen screening; estrogen receptor; anti- symptoms but also to establish the first evaluated record
estrogenic of estrogenic activity in this plant species.

There are a few medicinal plant species that exhibit a Materials and Methods
significant level of estrogenic activity including red
clover (Trifolium pretense L.), Erythrina variegata L., Plant materials. The tuberous roots of B. superba were collected
Kudzu (Pueraria lobata) and ‘‘white kwao krua’’ from 23 out of the 76 provinces of Thailand during the summer
(March-April) of 2000. The plant samples were identified by
(Pueraria mirifica Airy shaw et Suvatabhantu). Red
Cherdshewasart with reference to the herbarium voucher no. BCU
clover, a common perennial herb in Europe, has been 11046.12) The tuberous roots were cleaned, sliced, dried in a hot-air
shown to exhibit strong estrogenic activity in an oven at 70
C, subsequently ground into powder and filtered through a
evaluation by a yeast estrogen screening (YES) assay.1,2) 100-mesh-size sieve. The sieved powder was extracted with absolute
Erythrina variegata L., a traditional herb in Southeast ethanol at a ratio of 1:10 (w/v) for 3 d at room temperature, and then
Asia, India and China has been shown to have estrogenic the sediment was discarded. The ethanolic supernatant was passed
activity by expressing an anti-oesteoporotic effect in through Whatman no. 3 filter paper, and the filtrate was evaporated in a
rotary evaporator to leave a solid ethanolic tuberous extract. Each
ovariectomized rats.3) Kudzu, the common tuberous vine crude plant extract was kept in a tightly-capped tube, in a light-
of China, Korea and Japan contains a high amount of protected container and stored at 70
C until needed. A stock solution
isoflavonoids4–6) whilst ‘‘white kwao krua,’’ the common of each extract was freshly prepared in DMSO at concentrations
tuberous vine of Thailand, Myanmar and Laos is ranging from 1  103 to 1  103 mg/ml.
traditionally consumed for rejuvenating and for treating

y
To whom correspondence should be addressed. Fax: +66-2-3547160; E-mail: [email protected]
 Estrogenic and Anti-Estrogenic Activities of Butea superba Roxb. 2177

Fig. 1. The Chemical Structure of the Six Pure Compounds Isolated from B. superba (Pure Compounds 1–5 are from Ngamrojanavanich et al.,
2007).

Estrogenic activity testing of the six chemicals isolated from hERLBD and pGAD424-hSRC1 coactivator. Plasmids pGBT9-
B. superba tubers. Five previously isolated chemicals13) plus hexaco- hER
LBD, pGBT9-hERLBD and pGAD424-hSRC1 were con-
sanoic acid 2,3-dihydroxy-propyl ester (Fig. 1), which had all been structed by using genes hER
/ LBD and hSRC1 kindly provided
isolated from tubers of B. superba collected from Lampang Province, by Dr. A. Ohta of the Department of Biotechnology, the University of
Thailand, were used as reference chemicals in the evaluating the Tokyo, Japan.15) Plasmid pGAD424-hTIF2 was presented by Dr. Y.
estrogenic activity of the plant extracts. These were presented by Dr. Masamune of the Department of Cellular and Molecular Biology,
N. Ngamrojanavanich. Faculty of Pharmaceutical Science, Kanazawa University.16)

Construction of yeast estrogen screening strains ER
+ hTIF2 and Yeast culture conditions. S. cerevisiae host strain Y190 was
ER + hSRC1. The YES assay was based on a yeast two-hybrid maintained on yeast peptone dextrose agar (YPD: 1% (w/v) yeast
system and constructed by inserting the human estrogen receptor (hER) extract, 2% (w/v) peptone, 2% (w/v) glucose and 2% (w/v) agar). The
and coactivator into the yeast cells which allowed specific binding of yeast cells were incubated at 30
C for two days and then preserved at
suitable ligands with ER prior to their interaction with the yeast 4
C for routine use. Recombinant S. cerevisiae yeasts were usually
transcription machinery. The estrogenic activity was quantitatively selected and grown on a synthetic complete medium (SC) lacking
evaluated by the level of expression of the reporter gene encoding the tryptophan and leucine (SC agar: 0.67% (w/v) yeast nitrogen base w/o
-galactosidase enzyme.14) Since hER exists in the two subtypes of amino acid, 2% (w/v) glucose, 2% (w/v) dropout mix (lacking
hER
and hER that are distributed within specific tissues, only human tryptophan and leucine) and 1.5% (w/v) agar). Long term preservation
transcriptional intermediary factor 2 (hTIF2) and human steroid was achieved by growing them in SC broth lacking tryptophan and
receptor coactivator 1 (hSRC1) which have exhibited the greatest leucine at 30
C to the mid to late logarithmic phase of growth (24 h),
respective effectiveness to hER
and hER in the presence of 17- when an equal volume of 30% (v/v) glycerol was added and mixed and
estradiol14) were chosen for evaluating the estrogenic activity of the the yeast samples preserved in 1-ml aliquots at 80
C. In the case of
B. superba tuberous ethanolic extracts and their six isolated chemicals. the -galactosidase assay, the recombinant S. cerevisiae yeast samples
Yeast Saccharomyces cerevisiae strain Y190 (MATa, ura3-52, his3- were grown in a synthetic dextrose minimal medium (SD) broth
D200, ade2-101, trp1-901, leu2-3,112, gal4Dgal80D, URA3::GAL- supplemented with 0.02 mg/ml of adenine at 30
C while vigorously
LacZ, cyhr2, LYS2::GAL-HIS3; Clontech
, USA) was used as the shaking overnight for 14–16 h.
host for constructing the two yeast strains, YES-hER
+ hTIF2 and
YES-hER + hSRC1, respectively harboring the plasmid pGBT9- Recombinant yeast assay. The assays were performed by incubating
hER
LBD and pGAD424-hTIF2 coactivator, and plasmid pGBT9- 50 ml of an overnight yeast culture and 2.5 ml of one of (i) an ethanolic
2178 W. C HERDSHEWASART et al.
3 3
tuberous extract (10 –10 mg/ml final concentration), or (ii) an needed. The final respective concentrations of S9-treated B. superba
individual isolated chemicals from B. superba at 106 –1 mg/ml final and puerarin in the assay were 1000 and 42 mg/ml.
concentration, or (iii) 17-estradiol, E2 (102 –1010 M final concen-
tration, as a positive control) or (iv) DMSO (the vehicle) in each tube Evaluation of the antiestrogenic activity. The antiestrogenic activity
containing 200 ml of a fresh selective medium. After incubating at 30
C of B. superba was evaluated by examining the tuberous ethanolic
while shaking for 4 h, a 150 ml aliquot of each cultured cell suspension extracts that showed no detectable estrogenic activity were examined
was allocated to each well of a 96-well microtiter plate for measurement for anti-estrogenic activity based on their ability to inhibit the -
of the optical density (OD) at 660 nm, and 100 ml was used to harvest the galactosidase induction by 17-estradiol in the YES-hER
+ hTIF2
cells by centrifugation at 12;000 gmax for 5 min. The resulting cell pellet and YES-hER + hSRC1 assay systems. These two samples were
was then resuspended in 200 ml of a Z-buffer (0.1 M sodium phosphate at accordingly selected based on their likely sensitivity and ease of
pH 7.0, 10 mM KCl, 1 mM MgSO4 and 3.5 mM -mercaptoethanol) detection of anti-estrogenic activity, since they would either have had
containing 1 mg/ml of zymolyase 20T, and the mixture incubated at too low an estrogenic activity level, relative to any anti-estrogenic
37
C for 15 min. The resulting cell lysate was incubated with 40 ml of a activity, to mask overcoming the exogenous 17-estradiol activity.
substrate (4 mg/ml of o-nitrophenyl--D-galactoside (ONPG) in a 0.1 M
sodium phosphate buffer at pH 7.0) at 30
C for 30 min, the reaction Statistical analyses. The -galactosidase unit activity presented in
being terminated when the yellow color of o-nitrophenol (ONP) had this study is presented as the mean  standard deviation (SD) of three
developed by adding of 100 ml of 1 M Na2 CO3 . To remove all the cell independent experiments. An unpaired Student’s t-test and Duncan’s
debris, the reaction tube was clarified by centrifugation at 12;000 gmax multiple-range test were used for analyzing of the test results with
for 5 min, then 150 ml of the clarified supernatant was transferred into p < 0:05 being considered significant, by using SPSS Version 11
each well of a 96-well microplate and the absorbance levels at 420 nm statistical software program.
and 550 nm were measured by a microplate reader.14) The experiment
was done in triplicate for each sample. Results
-Galactosidase assay. The -galactosidase activity is defined in ER
- and ER-estrogenic activities
terms of Miller units and calculated by the following equation: The ER
- and ER-estrogenic activities, as respec-
Miller unit ¼ 1000½OD420  1:75ðOD550 Þ=½OD660 ðtÞðvÞ tively determined with the YES-hER
+ hTIF2 and
YES-hER + hSRC1 based assays of the six pure
where OD420 is the absorbance of yellow ONP, OD550 is light
compounds isolated from B. superba are summarized
scattering from the cell debris at the end of reaction, OD660 is the cell
density at the start of the assay, t is the time of reaction (min) and v is
in Table 1. The relative ER
-estrogenic potency of these
the volume of culture used in the assay (ml). six isolated B. superba chemicals formed three broad
groupings with, in order of potency, group 1 (7-hydroxy-
Calculation of the estrogenic EC50 values for each extract. The 6,40 -dimethoxyisoflavone) > group 2 (medicarpin >
yeast cells were incubated with none (control) and with 103 to formononetin) > group 3 (prunetin > hexacosanoic acid
103 mg/ml of the B. superba tuberous ethanolic extract before the 2,3-dihydroxy-propyl ester > 7,40 -dimethoxyisoflavone).
-galactosidase activity was subsequently analyzed. The relative However, although the ER-estrogenic activity showed
potency, efficiency and relative inductive efficiency were evaluated
the same three groupings, with group 1 also showing the
and then compared with the values obtained from yeast cells incubated
with 17-estradiol at a dose of 102 to 1010 M as a standardized highest potency, that of groups 2 and 3 was reversed
reference for the estrogenic activity amongst the B. superba plant such that the actual order in terms of relative potency
extracts. was; 7-hydroxy-6,40 -dimethoxyisoflavone > 7,40 -dime-
The data obtained from the -galactosidase assays were each thoxyisoflavone > hexacosanoic acid 2,3-dihydroxy-
fitted with a four-parameter logistic dose-response model. Calculations propyl ester > medicarpin > formononetin > prunetin.
were performed with Sigmaplot software for Windows, version 9.0 To evaluate the ER
- and ER-estrogenic activities of
(Microsoft Inc., USA), using the following function;
the B. superba tuberous ethanolic extracts, the relative
½A  D potency, efficiency and relative inductive efficiency of
Y¼ þD
1 þ ½C=XB the plant extracts derived from the YES-hER
+ hTIF2
where Y is the response value (-galactosidase activity), X is the and YES-hER + hSRC1 YES assays were compared
sample concentration in the test, A is the maximum response of (Tables 2 and 3). The results for the 23 geographical
-galactosidase activity (ligand efficiency), B is the relative slope of samples of B. superba are ranked in order of the highest
the middle region of the curve as estimated from a linear/log relative potency observed for ER
(Table 2) and ER
regression of the linear part of the dose–response curve, C is the
sample concentration that resulted in 50% efficiency, and D is the
(Table 3). The mean value for the relative potency of
detection limit. The EC50 value is the value of C in the equation and each plant extract by the YES-hER + hSRC1 assay
represents the ligand potency. was found to be approximately six-fold greater than that
by the YES-hER
+ hTIF2 assay, this difference being
Preparation and use of the S9 mixture. The in vitro metabolic statistically significant (p < 0:05). Presentation in terms
activation of each sample resulting from the addition of a freshly of the efficiency and relative inductive potency showed a
prepared rat liver S9 mixture, was tested against the tuberous ethanolic similar pattern for both. Of some note is the significant
extracts from two plant samples that exhibited no detectable estrogenic
variation in both ER
and ER activities between the
activity in either of the two YES assays. Thus the plant samples were
selected on the basis of their likely sensitivity and ease of detection, different samples. Whether this was attributable to
since they had a low endogenous estrogenic activity level which would different geographical locations (cultivars and or culti-
otherwise potentially saturate the assay, masking the newly formed vation conditions) or occurred with any given cultivation
activity by the in vitro hepatic metabolic activation. remains to be ascertained.
A 10-ml aliquot of a plant extract dissolved in DMSO was incubated
with a 990-ml aliquot of freshly prepared S9 mixture as described17) Anti-estrogenic activity
(containing 0.5 mg of the rat liver S9 fraction (Wako Pure Chemical
Industries, Ltd., Japan), 0.16 M MgCl2 , 0.1 M NADP, 0.1 M G-6-P, 0.5 M
The plant extracts that expressed very little or no
sodium phosphate buffer at pH 7.4, and 1 M KCl) at 37
C for 4 h. The detectable estrogenic activity in the YES-hER
+
negative control consisted of the S9 fraction heat-inactivated at 95
C hTIF2 assay (collected from Chaiyaphum Province) and
for 5 min. Each S9-treated plant extract was stored at 80
C until YES-hER assay (collected from Saraburi Province)
 Estrogenic and Anti-Estrogenic Activities of Butea superba Roxb. 2179
Table 1. Apparent in Vitro ER
and ER Activating Estrogenic Activities of B. superba Derived Pure Compounds Evaluated by the YES-
hER
+ hTIF2 and YES-hER + hSRC1 Assays

YES-hER
+ hTIF2 YES-hER + hSRC1

Pure compound Relative Relative
EC50 (mg/ml) EC50 (mg/ml)
potency (%) potency (%)
Prunetin 63:5  0:1d 0.02 349:1  27:9c 0.03
Medicarpin 1:13  0:36b 1.27 142:6  31:6b 0.06
Formononetin 1:58  0:32c 0.91 316:0  29:1c 0.03
7-Hydroxy-6,40 -dimethoxyisoflavone 0:09  0:35a 15.47 1:73  0:18a 5.27
7,40 -Dimethoxyisoflavone 220:2  0:2f  0.01 22:6  4:4a 0.40
Hexacosanoic acid 2,3-dihydroxy-propyl ester 162:3  0:04e 0.01 27:3  3:4a 0.33
17-Estradiol (E2) 0:01  0:43a 100 0:09  0:24a 100
Relative potency (EC50 of E2/EC50 of plant chemical)  100.
Expressed as mean  SD, n ¼ 3. means not sharing a common superscript letter in the same column are significantly different (p < 0:05), as determined by Duncan’s
multiple range test.

p < 0:05 significantly different from the negative control (DMSO) by Student’s t-test.

were selected to screen for the potential anti-estrogenic extracts as well as those in six pure chemicals that had
activity. These two plant extracts clearly demonstrated previously been isolated from B. superba tubers.
an anti-estrogenic effect in both these assays, showing The YES-hER
+ hTIF2 based assay revealed highly
clear inhibition of the 17-estradiol effect. The apparent potent ER
-estrogenic activity with 7-hydroxy-6-40 -
anti-estrogenic activity of both plant extracts was dimethoxyisoflavone at a level broadly comparable to
slightly higher in the ER (YES-hER + hSRC1) assay that of the 17-estradiol positive control. Moderately
than in the ER
assay (YES-hER
+ hTIF2) assays good ER
-estrogenic activity, and thus assumed binding
(Fig. 2). Regardless of this, these results are consistent and or activation, was also apparent with medicarpin and
with the notion that at least these two, if not all the formononetin. However, the other three compounds each
unmetabolized B. superba tuberous ethanolic extracts showed a much weaker ER
-estrogenic activity, the
harbored anti-estrogenic activity. weakest ER
-estrogenic activity being observed with
7,40 -dimethoxyisoflavone. This highlights the diverse
In vitro metabolic activation variation in estrogenic activity amongst the compounds.
One of the plant samples that exerted very little With respect to the ER-estrogenic activity, 7-
detectable estrogenic activity in the YES-hER
+ hydroxy-6-40 -dimethoxyisoflavone also exhibited by
hTIF2 assay (collected from Chaiyaphum Province) far the strongest activity of the six compounds, and
and the one plant showing negative in the YES-hER + thus could bind strongly and activate both ER
and
hSRC1 assay (collected from Saraburi Province), were ER. However, its ER-estrogenic activity was less than
tested for potential in vitro hepatic metabolism by that of the 17-estradiol positive control. Moreover,
coincubating with a fresh S9 mixture prior to assaying medicarpin and formononetin, which had moderate
for the estrogenic activity. Both the ER
and ER ER
-estrogenic activity, showed only a weak level of
estrogenic activities of these two tuberous ethanolic ER-estrogenic activity, whilst 7,40 -dimethoxyisofla-
plant extracts were significantly, but differently increased vone and hexacosanoic acid 2,3-dihydroxy-propyl ester,
by the S9 mixture treatment, with an enhanced ER
- but which both showed weak ER
-estrogenic activity,
not ER-estrogenic activity for the sample from the revealed a moderate ER-estrogenic activity. This
Chaiyapum Province and, conversely, enhanced ER- means that the B. superba tuberous ethanolic extracts
but not ER
-estrogenic activity for the sample from the would have been likely to initiate binding to both the
Saraburi Province (Fig. 3). These results are consistent human ER
and ER receptors with different affinity
with the notion that hepatic metabolism, in this case depending upon their relative chemical constituents.
attained in vitro by incubating with the S9 mixture, Importantly, variation between the B. superba sam-
could elevate the net estrogenic activity (i.e., that ples in their estrogenic activity was apparent in this
observed above any level of inhibitory anti-estrogenic study. The majority of the plant extracts expressed a
activity also present) of the B. superba tuberous etha- detectable, but low level of both ER
- and ER-
nolic extracts and furthermore, that such enhanced ER
- estrogenic activity. However, the plant estrogenic
or ER-estrogenic activity may be different in different activity was classified as of low efficiency in all 23
plant extracts, suggesting differing chemical constitu- samples, this was especially true for the ER
-estrogenic
ents. However, the bioactive compounds so metabolized activity which was some four- to six-fold lower than the
and the new bioactive compounds so formed remain to corresponding ER-estrogenic activity, depending on
be elucidated. which parameter was used for calculating the estrogenic
activity. The clear suggestion is that the non-metabo-
Discussion lized phytochemicals in the B. superba tuberous etha-
nolic extracts could bind to and activate ER more
We tested in this study, a modified yeast two-hybrid strongly than ER
, at least with these two different YES
system based upon the human estrogen receptors (ER
constructions. However, it should be noted with these
and ER, separately) coupled to the -galactosidase YES assays that 17-estradiol bound or activated ER

reporter gene to separately monitor the ER
- and ER- more strongly than ER with the same YES construc-
estrogenic activities in the B. superba tuberous ethanolic tions. This correlates with a previous finding in a study
2180 W. C HERDSHEWASART et al.
Table 2. ER
-Estrogenic Activity of B. superba Tuberous Ethanolic Extracts Evaluated by the YES-hER
+ hTIF2 Assay

Relative Efficiency
EC50 RIE
No. Province potency (-galactosidase
(mg/ml) (%)
(%) unit/mg plant extract)
1 Sisaket 83:25  3:03b 0.017 211:11  19:43fg 5.55
2 Ratchaburi 88:55  18:05bc 0.016 244:09  9:40g 6.42
3 Khon Kaen 124:66  4:15cd 0.012 454:76  14:28i 11.97
4 Nakorn Ratchasima 150:00  1:27d 0.010 379:19  9:48h 9.98
5 Lopburi 153:83  15:97d 0.009 123:55  2:95de 3.25
6 Loei 223:02  61:35e 0.006 70:61  1:37abcd 1.86
7 Chiang Rai 232:09  5:48e 0.006 45:99  4:12abc 1.21
8 Nong Bua Lam Phu 292:11  6:24f 0.005 128:74  7:94de 3.39
9 Lampang 387:69  34:18g 0.004 101:33  7:62cd 2.67
10 Chachaengsao 399:93  4:63g 0.004 87:55  5:98bcd 2.30
11 Saraburi 417:85  7:72g 0.003 22:96  13:32ab 0.60
12 Nakhon Sawan 463:07  35:57h 0.003 177:10  14:32ef 4.66
13 Phrachin Buri 508:83  56:60i 0.003 53:34  6:33abc 1.46
14 Mae Hong Sorn 510:05  4:71i 0.003 32:49  1:21abc 0.85
15 Tak 522:82  39:23i 0.003 98:45  8:42cd 2.59
16 Uttraradith 620:78  7:70j 0.002 6:11  2:99a 0.16
17 Phetchabun 666:92  7:84k 0.002 11:94  3:16a 0.31
18 Chonburi 823:83  6:81l 0.002 35:87  5:54abc 0.94
19 Chaiyaphum >1000m <0:001 NA NA
20 Chantaburi >1000m <0:001 NA NA
21 Chiang Mai >1000m <0:001 NA NA
22 Kanchanaburi >1000m <0:001 NA NA
23 Phitsanulok >1000m <0:001 NA NA
17-Estradiol (E2) 0:01  0:00a 100 3;800:40  179:32j 100
NA, not applicable
Relative potency; (EC50 of E2 /EC50 of plant extract)  100.
The plateau of the curve per mg plant extract designated as the ligand efficiency.
Relative inductive efficiency (RIE): (Efficiency of plant extract/Efficiency of E2 )  100.
Expressed as mean  SD, n ¼ 3. means not sharing a common superscript letter in the same column are significantly different (p < 0:05) as determined by Duncan’s
multiple range test.

Table 3. ER-Estrogenic Activity of B. superba Tuberous Ethanolic Extracts Evaluated by the YES-hER + hSRC1 Assay

Relative Efficiency
EC50 RIE
No. Province potencya (-galactosidase
(mg/ml) (%)
(%) unit/mg plant extract)
1 Khon Kaen 82:99  3:14b 0.11 225:55  15:81g 47.74
2 Nakhon Ratchasima 83:42  4:54b 0.11 261:29  17:89h 55.30
3 Lopburi 99:96  0:12b 0.09 150:30  2:72f 31.81
4 Chiang Rai 99:96  0:12b 0.09 58:85  7:47d 12.46
5 Tak 199:76  5:73c 0.05 103:42  8:21e 21.89
6 Nong Bua Lam Phu 354:23  19:50d 0.03 59:46  5:70d 12.58
7 Lampang 371:75  45:03d 0.02 68:20  10:45d 14.43
8 Phrachin Buri 399:73  0:17e 0.02 21:43  8:90abc 4.54
9 Loei 400:11  0:21e 0.02 13:84  7:02ab 2.93
10 Chantaburi 413:50  2:26e 0.02 9:13  0:85ab 1.93
11 Sisaket 413:50  13:38e 0.02 35:65  1:45c 7.55
12 Nakhon Sawan 538:59  13:84f 0.02 25:31  2:52bc 5.36
13 Uttraradith 599:59  0:15g 0.02 28:45  2:45bc 6.02
14 Phetchabun 599:70  0:42g 0.02 21:18  4:98abc 4.48
15 Ratchaburi 619:02  5:15g 0.01 108:17  5:35e 22.89
16 Kanchanaburi 694:31  17:01h 0.01 9:91  1:52ab 2.10
17 Chaiyaphum 733:08  8:78i 0.01 15:31  0:95abc 3.24
18 Saraburi 0a 0 0a 0
19 Chonburi >1000j <0:001 NA NA
20 Chachaengsao >1000j <0:001 NA NA
21 Mae Hong Son >1000j <0:001 NA NA
22 Chiang Mai >1000j <0:001 NA NA
23 Phitsanulok >1000j <0:001 NA NA
17-Estradiol (E2) 0:09  0:002a 100 472:48  46:09i 100
NA, not applicable.
Relative potency; (EC50 of E2 /EC50 of plant extract)  100.
The plateau of the curve per mg plant extract designated as the ligand efficiency.
Relative inductive efficiency (RIE); (Efficiency of plant extract/Efficiency of E2 )  100.
Expressed as mean  SD, n ¼ 3. means not sharing a common superscript letter in the same column are significantly different (p < 0:05) as determined by Duncan’s
multiple range tests.
 Estrogenic and Anti-Estrogenic Activities of Butea superba Roxb. 2181

on soy isoflavonoids that those phytoestrogens could of the plant samples with the six chemicals isolated in
bind to or activate ER more strongly than ER
whilst this study.
estrogen bound to ER
more strongly than to ER.16) The tuberous ethanolic extracts from B. superba that
The results of this present study also showed variation in did not exhibit estrogenic activity in the YES-hER
+
the estrogenic activity of the plant samples collected hTIF2 and YES-hER + hSRC1 based assays, did
from different provinces in Thailand although there was exhibit anti-estrogenic activity in the incubation assay
not a sufficient number of samples from within each of with 17-estrodiol. However, this result was obtained
the 23 localities to distinguish intra-regional variation with non-metabolized samples whereas the oral admin-
from inter-regional variation. Nevertheless, the results istration of herbs like B. superba would be likely to
were not directly related to the region of the existent result in hepatic metabolism. Such hepatic metabolism
plant samples. This implies the likely combined influ- may enhance or reduce any anti-estrogenic activity, an
ence of the environment, plant age/development stage aspect which was not assayed here. The potential
and plant genetics on such bioactive difference. There importance of this point is illustrated by the estrogenic
was no assay on the concentration of the six isolated activity, whereby the addition of an in vitro hepatic
compounds in all analyzed samples, except for the metabolizing S9 system clearly increased the net
one collected from Lampang Province. It is correspond- observed estrogenic activity of the two B. superba
ingly not possible to correlate the estrogenic activity samples tested and differentially increased the ER
-
but not ER-estrogenic activity in one plant extract, but
visa versa in the other sample.
Various methods have been used to evaluate the
estrogenic activity of phytoestrogens or plant materials.
The uterotrophic assay with ovariectomized rats is a
classical method and has been used to establish the
estrogenic activity in phytoestrogen-rich plants.18) The
vaginal cornification assay is a simpler method that is
performed with ovariectomized rats, and has also
enabled the successive differential evaluation of estro-
genic activity for the phytoestrogen-rich Pueraria mirifica
and P. lobata populations.8,19) The MCF-7 screening test
(E-assay), based on the estrogen-dependent in vitro
proliferation of the human mammary adenocarcinoma
cell line in the presence of a tested phytoestrogen, has
been able to demonstrate the biphasic estrogenic
response of the phytoestrogen-rich B. superba and
Fig. 2. The Anti-Estrogenic Activity of B. superba Tuberous Etha- P. mirifica extracts12,20) and has also revealed that this
nolic Extracts at a Final Concentration of 1000 mg/ml as Determined was dependent on where the plant samples were
by the YES-hER
+ hTIF2 and YES-hER + hSRC1 Assays.
A plant sample at a 103 mg/ml, final concentration of was mixed
collected. Our previous analysis using an MCF-7
with 17-estradiol at a 109 M final concentration and determined proliferation assay of the same plant samples of
against the same concentration of 17-estradiol (as a control). Data B. superba collected from Lampang Province as those
are expressed as the mean  SD, n ¼ 3.  p < 0:05; significantly used in this study, found no proliferative effect but
different from the 17-estradiol treatment by the Student’s t-test. instead an antiproliferative effect in the tested dose

Fig. 3. Estrogenic Activity of the B. superba Tuberous Ethanolic Extracts after Treating with the Rat Liver S9 Fraction Evaluated by the YES-
hER
+ hTIF2 and YES-hER + hSRC1 Assays.
Abbreviation: S9-inactive is the heat-inactivated S9 fraction (negative control): S9-active is the active S9 fraction. Each value is the mean of
three independent experiments. The final concentration of each S9-treated B. superba tuberous ethanolic extracts at 1000 mg/ml and of puerarin
(control) at 42 mg/ml was tested. Data are expressed as the mean  SD, n ¼ 3.  p < 0:05; significantly different from the S9-inactive treatment
by Student’s t-test.
2182 W. C HERDSHEWASART et al.
12)
range of 10–1000 mg/ml. In contrast to the results Dr. Yukito Masamune and Prof. Dr. Akinori Ohta for
with the MCF-7 assay, most of the B. superba samples generosly providing the YES plasmids and to Dr. Robert
in this YES-based study, including those from Lampang Butcher, PCU, Faculty of Science, Chulalongkorn
Province, exhibited estrogenic activity in the yeast two- University, for English proofreading.
hybrid system suggesting this hybrid system to be more
sensitive than the standard MCF-7 cell proliferation test References
in determining the potential net estrogenic activity,
especially of plant materials. However, these two test 1) Beck V, Rohr U, and Jungbauer A, J. Steroid Biochem. Mol.
systems involve no hepatic metabolic activation and are Biol., 94, 499–518 (2005).
theoretically unlikely to represent the real situation 2) Beck V, Unterrieder E, Krenn L, Kubelka W, and Jungbauer A,
following an oral administration, indeed, the estrogenic J. Steroid Biochem. Mol. Biol., 84, 259–268 (2003).
3) Zhang Y, Li X-L, Lai W-P, Chen B, Chow H-K, Wu C-F, Wang
activity of P. mirifica experimentally assayed with
N-L, Yao X-S, and Wong M-S, J. Ethnopharmacol., 109, 165–
HepG2 cells has suggested the requirement for hepatic 169 (2007).
activation to demonstrate maximal estrogenic activity.21) 4) Kaufman PB, Duke JA, Brielmann H, and Hoyt JE, J. Altern.
In contrast, the establishment of the YES test system in Complement. Med., 3, 7–12 (1997).
the presence of an S9 mixture comprising rat liver 5) Kirakosyan A, Kaufman PB, Warber S, Bolling S, Chang S, and
enzymes that was used in this study supports our Duke JA, Plant Sci., 164, 883–888 (2003).
previous finding that in vitro metabolic activation could 6) Cherdshewasart W, Subtang S, and Dahlan W, J. Pharm.
Biomed. Anal., 43, 428–434 (2007).
enhance the estrogenic activity of the plant extracts.22) 7) Suntara A, ‘‘The Remedy Pamphlet of Kwao Krua Tuber of
Even though the in vitro YES assay is a rapid Luang Anusarnsuntarakromkarnpiset, Chiang Mai,’’ Upatipong-
method and could be quantitatively applied, it may not sa Press, Chiang Mai (1931).
have the high level of sensitivity for detecting estro- 8) Malaivijitnond S, Chansri K, Kijkuokul P, Urasopon N, and
genic activity that the in vivo vaginal cornification Cherdshewasart W, J. Ethnopharmacol., 107, 354–360 (2006).
assay offers. Moreover, it may not be possible to 9) Kim HY, Hong JH, Kim DS, Kang KJ, Han SB, Lee EJ, Chung
HW, Song KH, Sho KA, Kwack SJ, Kim SS, Park KL, Kim
reliably apply the results from the same plant materials
MC, Kim CM, and Song IS, Food Technol. Biotechnol., 12, 29–
in tests with animals to predict what would be the case 35 (2003).
with human consumption. Moreover, the successful 10) Cherdshewasart W, Bhuntaku P, Panriansaen R, and Malaivi-
detection of estrogenic activity in tuberous ethanolic jitnond S, Maturitas, 60, 131–137 (2008).
extracts from diverse B. superba samples provides 11) Cherdshewasart W and Nimsakul N, Asian J. Androl., 5, 243–
additional support for the efficacy of the in vitro yeast 245 (2003).
two-hybrid system for detecting estrogenic, and perhaps 12) Cherdshewasart W, Cheewasopit W, and Picha P, J. Ethno-
pharmacol., 93, 255–260 (2004).
anti-estrogenic, activity. This system has previously
13) Ngamrojanavanich N, Loontaisong A, Pengpreecha S, Cherd-
been used to study estrogens in such other plants as shewasart W, Pornpakakul S, Pudhom K, Roengsamran S, and
Moghania philippinensi,23) the traditional Chinese herb, Petsom A, J. Ethnopharmacol., 109, 354–358 (2007).
Polygonum cuspidatum,24) several Mediterranean plants25) 14) Nishikawa J, Saito K, Goto J, Dakeyama F, Matsuo M, and
and Millettia pachycarpa.26) We demonstrated in this Nishihara T, Toxicol. Appl. Pharmacol., 154, 76–83 (1999).
study that most of the Thai B. superba populations 15) Lee HS, Miyauchi K, Nagata Y, Fukuda R, Sasagawa S, Endoh
showed detectable estrogenic activity and also poten- H, Kato S, Horiuchi H, Takagi M, and Ohta A, J. Biochem., 131,
399–405 (2002).
tially anti-estrogenic activity. The anti-estrogenic activ- 16) Morito K, Hirose T, Kinjo J, Hirakawa T, Okawa M, Nohara T,
ity of the B. superba tuber extract is in accordance with Ogawa S, Inone S, and Muramatsu M, Biol. Pharmacol. Bull.,
the study of B. superba samples collected from Lampang 24, 351–356 (2001).
Province by a hormonal assay and histological assay of 17) Takatori S, Kitagawa Y, Oda H, Miwa G, Nishikawa J-I,
the uterus in ovariectomized rats.27) This might be one Nishihara T, Nakasawa H, and Hori S, J. Health Sci., 49, 91–98
of the reasons why the plant is used in the recipes for (2003).
18) Benson GK, Cowie AT, and Hosking ZD, J. Endocrinol., 21,
treating menopausal symptoms. The YES test systems
401–409 (1961).
could provide useful information in screening potential 19) Cherdshewasart W, Kitsamai Y, and Malaivijitnond S,
B. superba material for further developing phytoestrogen- J. Reprod. Dev., 53, 385–393 (2007).
rich products, especially in conjugation with the S9 20) Cherdshewasart W, Traisup V, and Picha P, J. Reprod. Dev., 54,
in vitro hepatic metabolic activation. 63–67 (2008).
21) Lee YS, Park JS, Cho SD, Son JK, Cherdshewasart W, and
Acknowledgments Kang KS, J. Vet. Sci., 3, 273–277 (2002).
22) Cherdshewasart W and Sriwatcharakul S, Maturitas, 59, 128–
136 (2008).
This study was supported by the Higher Education 23) Ahn EM, Nakamura N, Akao T, Nishihara T, and Hattori M,
Development Project, Subproject: Graduate and Re- Biol. Pharm. Bull., 27, 548–553 (2004).
search in Agricultural Biotechnology Subproject, under 24) Zhang C, Zhang X, Zhang Y, Xu Q, Xiao H, and Liang X,
the Commission on Higher Education, by the Ministry J. Ethnopharmacol., 105, 223–228 (2006).
of Education. Wichai Cherdshewasart was supported by 25) Garritano S, Pinto B, Giachi I, Pistelli L, and Reali D,
The National Research University Project of the CHE Phytomedicine, 12, 143–147 (2005).
26) Ito C, Itoigawa M, Kumagaya M, Okamoto Y, Ueda K,
and Ratchadaphiseksomphot Endowment Fund (FW Nishihara T, Kojima N, and Furukawa H, J. Nat. Prod., 69,
011A) and by the Thai Government Stimulus Package 138–141 (2006).
2 (TKK2555), under the Project for Establishment of a 27) Malaivijitnond M, Ketsuwan A, Watanabe G, Taya K, and
Comprehensive Center for Innovative Food, Health Cherdshewasart W, J. Ethnopharmacol., 121, 123–129 (2009).
Products and Agriculture. We would like to thank Prof.