Introduction
Women potentially spend the last third of their lives in postmenopause, due to their
increased life expectancy. Menopausal women suffer from a variety of symptoms, such
as hot flashes, night sweats, mood swings, insomnia, vaginal dryness, in addition
to
long-term complications such as osteoporosis [1 ], [2 ]. These symptoms arise primarily as a response to the
drastic decline in circulating endogenous estrogens [3 ].
In order to relieve menopausal symptoms, traditional HT (estrogen or estrogen plus
progestin), was designed to supplement the hormones. However, the WHI demonstrated
an increased risk of developing breast cancer associated with HT [4 ] leading women to search for natural alternatives such
as botanical supplements [5 ], [6 ].
Botanicals, which are generally perceived as safe due to their natural origin, have
a
long history of use for female complaints, particularly in traditional Chinese
medicine [7 ]. The fact that Asian women have less
frequent and severe hot flashes suggests that this effect could be associated with
their flavonoid-rich diet and that botanicals with high flavonoid content could be
effective in managing menopausal symptoms [8 ]. As a
result, many women turn to botanical dietary supplements for alleviation of
menopausal symptoms, specifically for the reduction of hot flashes [5 ], [6 ].
Botanical supplements could act through a number of different mechanisms including
estrogenic, progestogenic, and/or serotonergic pathways ([Figs. 1 ] and [2 ]) [9 ], [10 ], [11 ], [12 ], [13 ]. It is thought that botanicals with estrogenic activity might mimic the
actions of estrogens. The importance of estrogen in homeostatic regulation of many
cellular and biochemical events is well illustrated by the pathophysiological
changes that occur with estrogen deficiency [14 ], [15 ], [16 ], [17 ]. Endogenous estrogen (estradiol, E2 ) is
actively involved in the development of the mammary gland and uterus, in maintaining
pregnancy and bone density, in protection from cardiovascular diseases, and in
relieving menopausal symptoms [16 ], [17 ], [18 ], [19 ]. Estrogens mainly exert their biological effects
through binding to ERs including ERα and ERβ , followed by dimerization
of ERs and interaction with EREs at the promoter of the estrogen responsive genes,
thus activating transcription and generating estrogenic responses which are crucial
for normal physiological functions ([Fig. 1 ]) [20 ], [21 ], [22 ]. In humans around one-third of the genes that are
regulated by ERs do not contain ERE-like sequences [22 ], [23 ]. ERs can also tether to other
transcription factors such as Fos and Jun that are directly bound to DNA through
their respective responsive elements such as AP1 binding sites to regulate
transcription of the related genes [22 ], [23 ]. Estrogen also activates rapid signaling pathways
such as MAPKs and PI3K pathways, which, in turn, can modulate transcription and
proliferation [22 ], [24 ].
Studies have revealed another type of ER, namely GPER or GPR30, that is involved in
different signaling pathways [25 ], [26 ]. It is also known that mechanisms of E2
actions depend on the ligand, the cell type, and the receptor subtype [22 ], [23 ]. It is believed
that ERα induction is responsible for the proliferative effects of estrogens,
while ERβ activation balances the ERα -dependent responses [27 ], [28 ], [29 ].
Fig. 1 Classical mechanisms of the estrogenic and progestogenic
activities.
Fig. 2 Mechanism of serotonergic activity. Serotonin is released in the
synapses and binds to its receptor (5-HT7 ) in the post-synaptic
cells. A serotonin receptor, coupled with a G-protein, activates adenylate
cyclase, resulting in production of cAMP and activation of enzymatic cascades
leading to serotonergic effects.
Botanicals and specifically their phytoestrogens, such as genistein and daidzein,
preferentially bind and activate ERβ , thus may exert a safe estrogenic
activity [15 ], [30 ].
However, the use of botanicals with only plant-derived estrogens in the absence of
progestins might increase the risk of developing endometrial hyperplasia and cancer
similar to conventional estrogen-alone HT [31 ], [32 ]. It is known that women with an intact uterus who
take HT to treat estrogen-deficient menopausal symptoms must take a combination of
estrogens and progestins, and the same is likely true for phytoestrogens and
phytoprogestins. P4 , the precursor of many steroid hormones, plays a
crucial role in the normal physiology of the uterus, ovaries, mammary gland,
cardiovascular system, bone, brain, and central nervous system [33 ]. Its biological function is mainly mediated through
its binding to PRs, including PRA and PRB, followed by the receptor dimerization,
translocation to the nucleus, and interaction with PREs, thus regulating
transcription of downstream genes ([Fig. 1 ]) [34 ]. Animal models partially suggest that PRA induction
is protective in the uterus, while PRB induction might increase breast proliferation
[35 ], [36 ], [37 ]. Botanicals containing natural progestins, which can
activate progesterone-dependent pathways, in addition to estrogenic compounds are
preferred.
Estrogen withdrawal during menopause results in the decline in the release of
neurotransmitters, primarily norepinephrine and serotonin (5-HT), which will lead
to
a change in thermoregulation in the hypothalamus [38 ].
This effect ultimately results in frequent sweating and increased peripheral
circulation as heat-loss mechanisms generating hot flashes and night sweats.
Increase in the amount of serotonin and activation of certain 5-HT receptors as well
as inhibition of serotonin reuptake in synapses through the blocking of SERTs are
possible approaches in preventing hot flashes ([Fig. 2 ]). In order to avoid hormonal approaches, some women choose SSRIs to
manage menopausal discomforts, particularly vasomotor symptoms [39 ]. However, there are also a number of undesirable
outcomes such as sexual dysfunction, nausea, weight gain, and sleep disturbances
associated with these remedies [40 ], [41 ]. Therefore, some botanicals have been investigated
for their potential serotonergic effects including activation of serotonin
receptors, mainly 5-HT7 , or inhibition of serotonin reuptake through
SERTs. What follows is a review of the potential mechanisms (estrogenic,
progestogenic, serotonergic) of common botanicals for managing menopausal
symptoms.
Botanicals with Estrogenic Activity
Due to the importance of estrogens in the alleviation of menopausal symptoms,
particularly for the reduction of hot flashes, several popular botanicals have been
studied for estrogenic activity including soy, red clover, kudzu, hops, licorice,
rhubarb, yam, chasteberry, dong quai, and black cohosh ([Table 1 ]). In vitro and in vivo experiments are summarized
below describing elucidation of potential estrogenic activity of the extracts and
isolation and characterization of their active principles.
Table 1 Estrogenic potency of phytoestrogens in competitive
ER binding assaya .
Compoundb
Plant
References
IC50 (µM) reported range ER binding
ERα
ERβ
a The values are from different studies and are
included for qualitative comparison. Different methods were
employed: radiometric binding assay using purified human ER
[9 ], [50 ], [101 ], [185 ], fluorescence polarization
assay using purified human ER [124 ], [129 ], [138 ], [189 ], radiometric binding assay in cells or tissues
[113 ], [187 ], [188 ], [190 ], inhibition
ELISA using purified human ER [186 ], and dextran-coated charcoal method in cells
[89 ]. b Some
compounds might not be plant-specific. c ND: not
determined
17β -estradiol (E2 )
–
[89 ], [101 ], [185 ]
0.00 001–0.003
0.0014–0.0038
Genistein
soy, red clover, kudzu
[50 ], [101 ], [185 ]
0.59–1.145
0.025–0.09
Daidzein
soy, red clover, kudzu
[9 ], [185 ], [186 ]
0.96–17
0.1–1.20
S-equol
soy, red clover, kudzu
[50 ], [138 ], [185 ]
0.208–1.02
0.016–0.11
Kaempferol
red clover
[80 ], [186 ]
8.2
0.05–50
Puerarin
kudzu
[187 ]
0.87
NDc
Miroestrol
kudzu
[89 ]
0.0003
NDc
8-prenylnaringenin
hops
[9 ], [101 ], [188 ]
0.057–0.51
0.068–1.7
Liquiritigenin
licorice
[189 ]
2.80
0.41
Glabridin
licorice
[113 ]
5.00
NDc
Glabrene
licorice
[190 ]
1.00
NDc
Lindleyin
rhubarb
[124 ]
225–435
NDc
Trans -rhapontigenin
rhubarb
[129 ]
12
5.6
Desoxyrhapontigenin
rhubarb
[129 ]
26
28
Apigenin
chasteberry
[138 ], [139 ], [186 ]
7.88
0.08–1.00
Penduletin
chasteberry
[139 ]
NDc
0.31
Soy (Glycine max , Fabaceae) is often consumed as an alternative to HT
by menopausal women [42 ]. Genistein and daidzein ([Table 1 ], [Fig. 3 B ]) are
the most abundant estrogenic compounds in soy [43 ].
They are mainly glycosylated in the extract and are activated upon hydrolysis in
vivo, contributing to their estrogenic activity [44 ], [45 ], [46 ]. It has been shown that these flavonoids preferentially bind and activate
ERβ in cell-based assays and that daidzein had weaker estrogenic activity
compared to genistein ([Table 1 ]) [9 ], [47 ], [48 ], [49 ]. It has also been
reported that daidzein metabolism by the gut microflora results in the formation of
S-equol ([Table 1 ], [Fig. 3 B ]) which had a stronger estrogenic activity than daidzein in ER
binding and transcriptional activation assays in HEC-1 cells [50 ]. S-equol activity was more significant with ERβ , (Ki
[ERβ ] = 16 nM; β /α = 13 fold), being comparable to that of
genistein (Ki [ERβ ] = 6.7 nM; β /α = 16) [50 ], [51 ]. It has also been reported that
gut microflora variability and differences in the metabolism of soy flavonoids could
lead to individual variation in the amount of S-equol formed resulting in distinct
biological outcomes [51 ], [52 ], [53 ], [54 ].
Fig. 3 Chemical structures of phytoestrogens found in the estrogenic
botanicals.
Genistein, daidzein, and S-equol can also activate ERα -dependent responses
such as MCF-7 (ERα +) cell proliferation [55 ], [56 ], [57 ], [58 ], [59 ], [60 ], [61 ]. It has been demonstrated that activation of ERα -dependent
responses by genistein is associated with high concentrations of this isoflavone
[48 ], [62 ], [63 ], while ERβ -dependent responses are mainly
observed at low concentrations [9 ], [62 ]. Wober et al. [64 ]
observed a dose-dependent induction of alkaline phosphatase with genistein and
daidzein in the endometrial adenocarcinoma cell line, Ishikawa (ERα +), an
effect which was inhibited by the antiestrogen ICI 182,780, demonstrating an
ER-dependent estrogenic activity. Similarly, Kayisli et al. [65 ] reported a weak but dose-dependent estrogenic activity with soy
isoflavones in Ishikawa cells when they studied cell proliferation and alkaline
phosphatase activity. In the presence of E2 , these compounds had
antiestrogenic effects [65 ].
Consistent with the observed in vitro effects, subcutaneous injection of
genistein (250 mg/kg/day) in ovariectomized Sprague-Dawley rats for two weeks
significantly increased uterine weight, uterine-to-body weight ratio, femur weight,
and femur-to-body weight ratio, all of which are likely ERα -dependent effects
[10 ]. Cimafranca et al. [66 ] also showed that 2 µL/g body weight of genistein (25 mg/mL) induced
an increase in uterine weight, downregulated the progesterone receptor in uterine
epithelium, increased multioocyte follicles, and decreased thymus weight relative
to
body weight in neonatal mice after 5 days. Some of the effects including increased
multioocyte follicles and abnormal estrous cycle were also seen after 6 months.
Legette et al. [67 ] observed an increased uterine
weight and enhanced uterine proliferative index in ovariectomized Sprague-Dawley
rats receiving 200 mg/kg dietary equol, demonstrating ERα -dependent effects
in vivo .
Similarly as in the isoflavone studies, soy extract bound to ERβ at 100 µg/mL
and activated ERβ -dependent transcription in HEK-293 cells at 0.1–100 µg/mL,
while it did not have any proliferative effects in MCF-7 (ERα +) cells [68 ]. However, Vieira et al. [69 ] showed that different commercially available soy supplements
increased uterine weight in immature female rats when applied at increasing serial
doses (125–4150 mg/kg/day) for 3 days. The observed proliferative effects could be
due to the administration of high concentrations of the extract, which can activate
ERα -dependent pathways, similar to what was observed in experiments with
high concentrations of isoflavones. In their studies, the uterotrophic effects were
different when extracts from different vendors were used, demonstrating the lack of
a unified standardization system in soy extract manufacturing. Allred et al. [70 ] showed that genistein alone, mixed soy isoflavones,
Novasoy, molasses, and soy flour combined with mixed isoflavones have different
effects on estrogen-dependent MCF-7 (ERα +) tumor growth in athymic mice,
demonstrating the effect of the food matrix on the modulation of estrogenic effects
of soy and its contents. A thorough review by Hilakivi-Clarke et al. [71 ] concluded that the estrogenic effects of soy and its
isoflavones in animal models are influenced by the dose, the route of
administration, the matrix, and the age in which animals receive either whole soy
or
the isoflavones. In summary, soy contains genistein and daidzein which
preferentially bind to and activate ERβ , but at higher concentrations and
depending on the tissue, they can activate ERα -dependent responses as well
[71 ], [72 ], [73 ].
Red clover (Trifolium pratense , Fabaceae) is often used for the relief
of menopausal symptoms [42 ], and it contains the same
phytoestrogens including genistein and daidzein ([Table
1 ], [Fig. 3 B ]) as discussed above for soy.
However, unlike soy, the majority of the isoflavones in red clover are present as
the methoxy ethers, formononetin and biochanin A, which require P450-catalyzed
metabolism to generate the active phytoestrogens daidzein and genistein,
respectively ([Fig. 4 A ]) [43 ], [74 ]. In a chemical and biological
evaluation, Booth et al. [75 ] reported that a
standardized red clover extract (0.23 % daidzein, 0.41 % genistein, 0.07 %
kaempferol, 14.26 % formononetin, 14.47 % biochanin A), preferentially bound to
ERβ and induced alkaline phosphatase in Ishikawa cells
(EC50 = 2.0–2.2 µg/mL). In this study, daidzein and genistein were the
most active constituents in the alkaline phosphatase induction assay in Ishikawa
cells and in the competitive ER binding assay, with a preferential activity with
ERβ , while formononetin, biochanin A, and kaempferol did not have
estrogenic effects. Considering the abundance of formononetin and biochanin A in the
extract, relative to daidzein and genistein, P450-catalyzed metabolism obviously
plays an essential role in generating the estrogenic activity of the extract ([Fig. 4 A ]) [75 ]. Hsu et al.
[76 ] reported moderate estrogenic effects by
biochanin A in MCF-7 cells. Markiewicz et al. [77 ]
showed that compared to E2 , genistein, and daidzein, induction of
alkaline phosphatase by formononetin and biochanin A in Ishikawa cells is weak.
Similarly, in a study by Fokialakis et al. [78 ],
biochanin A weakly induced luciferase reporter activity in MCF-7:D5 L and alkaline
phosphatase activity in Ishikawa cells as well as proliferation of MCF-7 cells.
Halabalaki et al. [79 ] showed that formononetin
moderately bound to ER subtypes. Compared to genistein and daidzein, formononetin
weakly activated alkaline phosphatase in Ishikawa cells and luciferase reporter
induction in MCF-7:D5 L cells [79 ]. This might further
emphasize the role of metabolism in exerting estrogenic responses by red clover
compounds. While Booth et al. [75 ] did not observe any
estrogenic activity for kaempferol ([Fig. 3 B ]),
Zoechling et al. [80 ] reported a selective binding of
kaempferol to ERβ ([Table 1 ]).
Fig. 4 A P450-catalyzed formation of daidzein and genistein from
formononetin and biochanin A, respectively. B Metabolic formation of 8-PN
from its precursors in hops. C Metabolic formation of liquiritigenin from
isoliquiritigenin in licorice.
In another study, a weak estrogenic activity was reported for a red clover extract
through binding and activation of ERβ in transfected HEK-293 cells but
proliferative effects in MCF-7 (ERα +) cells, at concentrations > 30 µg/mL
[68 ]. Interestingly, Overk et al. [9 ] also showed that red clover extract had binding
affinity for both ER subtypes but a greater affinity for ERβ . They also
demonstrated that the extract induced ERE-luciferase and PgR mRNA
transcription in ERα + cell lines MCF-7 and Ishikawa as well as activating
estrogen responsive alkaline phosphatase in Ishikawa cells (EC50 :
2 µg/mL). Red clover phytoestrogens, as well as the extract, have a greater affinity
for ERβ, but they can bind to and activate ERα at high concentrations,
especially in cell lines such as MCF-7 and Ishikawa as well as estrogen-sensitive
tissues, which have high levels of ERα . Moreover, formation of the ERβ
selective estrogenic constituents, genistein and daidzein, depends on P450-catalyzed
metabolism which could influence the results of the cell-based assays, since some
cultured cells such as Ishikawa do not metabolize formononetin and biochanin A to
the active isoflavones [9 ].
Estrogenic activity of red clover has also been studied in animal models. In an in
vivo study with Sprague-Dawley rats, Burdette et al. [81 ] showed that a red clover extract, standardized to isoflavones,
increased the uterine weight and vaginal cell cornification, demonstrating an
estrogenic response in these tissues while they did not observe any mammary gland
ductal branching as a sign of estrogenic activity in the breast. This observation
showed that red clover can activate ERα -dependent responses in vivo,
but its effects might be tissue specific. On the other hand, in an in vivo
study by Overk et al., in which Sprague-Dawley rats were treated with lower doses
of
red clover extract, no uterotrophic effects, vaginal cell cornification, or increase
in the height of uterus luminal epithelial cells were observed [82 ]. In summary, red clover mainly contains formononetin
and biochanin A, which are converted by P450 metabolism to genistein and daidzein
and exert estrogenic effects, preferentially through ERβ . Similar to soy
flavonoids, the estrogenic activity of red clover and its flavonoids likely depend
on the administered concentration, metabolism of isoflavones, and the target tissue
[71 ], [72 ], [73 ].
Kudzu (Pueraria lobata ; Fabaceae) is one of the commonly used botanical
supplements in the United States [42 ]. Isoflavonoids
such as formononetin, biochanin A, genistein, daidzein, and puerarin ([Table 1 ], [Fig. 3 B ] and
[C ]) are among the compounds that were isolated
from this plant [83 ], [84 ]. As discussed above, genistein and daidzein have estrogenic activity
preferentially through ERβ , but activate ERα pathways as well [10 ], [82 ]. Puerarin is
metabolized to daidzein by intestinal bacteria [85 ], [86 ].
Pueraria lobata extract has been shown to preferentially bind to and activate
ERβ in HEK-293 cells transfected with ER subtypes but surprisingly had a
proliferative effect in MCF-7 (ERα +) cells [68 ].
In this study, it was not mentioned if the kudzu extract also had any effect on the
reporter gene activity in ERα transfected HEK-293 cells, and the relative ER
subtype selectivity was not clearly stated. These controversial results might also
be associated with using two different cell lines and the tissue specific behavior
of the extract. As discussed above, the estrogenic constituents of kudzu, such as
genistein and daidzein, can activate ERα -dependent estrogenic responses such
as proliferation at higher concentrations or in specific hormone responsive tissues.
A variety of studies using the yeast-based estrogenic assays reported various
potencies for kudzu extracts [35 ], [36 ], [45 ]. The observed
distinct effects could be related to the different extract preparations and
standardization between studies.
Pueraria mirifica, another species of Pueraria (commonly called Kwao
Keur) which is very popular in Southeast Asia, was shown to significantly promote
proliferation of MCF-7 (ERα +) cells at 1 µg/mL, and it was more estrogenic
than Pueraria lobata extract [87 ], [88 ]. The active compound was puerarin with a
proliferative effect at 1 µM [87 ]. It was also reported
that miroestrol ([Fig. 3 C ]), another compound in
Pueraria mirifica , induced ERE-CAT reporter gene activity as well as cell
growth in MCF-7 (ERα +) cells, indicating estrogenic activity, mainly through
ERα
[89 ]. However, using a yeast-based estrogenic assay
with ER subtypes, Boonchird et al. [90 ] observed an
8.5-fold higher relative potency for the plant extract with ERβ in comparison
to ERα . Nevertheless, it was reported that different concentrations of
various extracts of Pueraria mirifica induced vaginal epithelium
cornification, increased uterine weight and thickness, and attenuated body weight
in
ovariectomized rats [91 ], [92 ], [93 ], [94 ], [95 ], which are considered ERα
effects. Higher concentrations of the extract as well as isoflavone content and the
tissue specificity might induce both ERα - and ERβ -dependent responses
[30 ], [72 ], [73 ]. A thorough recent review by Malaivijitnond [96 ] summarized many biological studies and defined the
effects of different cultivars of Pueraria plants from various regions and
seasons as a potential source of existing discrepancies between different study
outcomes. In summary, Pueraria species contain formononetin, biochanin A,
genistein, daidzein, puerarin, and miroestrol. Miroestrol is a potent ERα
ligand while genistein and daidzein preferentially bind to and activate ERβ .
However, depending on the concentration and target tissue, ERα -dependent
responses could be observed.
Hops (Humulus lupulus ; Cannabaceae) is a popular botanical for its
sleep-inducing effects, especially in Europe [97 ], [98 ]. It is also present in some
dietary supplements for managing menopausal symptoms [42 ]. Liu et al. [99 ] reported a moderate
estrogenic activity for hops based on competitive ER binding activity, alkaline
phosphatase induction, and PgR mRNA induction in Ishikawa cells. Overk et al.
also reported estrogenic activity for a hops extract in competitive ER binding
assays, ERE-luciferase induction in MCF-7 (ERα +) cells, PgR mRNA
induction in MCF-7 and Ishikawa cells, and induction of alkaline phosphatase enzyme
in Ishikawa cells (EC50 = 1 µg/mL) [9 ]. 8-PN
has been reported as the estrogenic component of hops, equipotent for both ER
subtypes, with an activity greater than that of any of the known phytoestrogens
([Table 1 ], [Fig. 3 D ]) [9 ], [100 ], [101 ]. Bovee et al. [102 ] observed estrogenic activity of 8-PN in a
yeast-based ER-dependent reporter assay. In this study, the potencies of 8-PN for
ERα and ERβ were 100 times and 3900 times less than that of
estradiol, respectively. Milligan et al. [103 ]
reported that 8-PN induced alkaline phosphatase in Ishikawa Var I cells
(EC50 = 4.41 nM) and was active in a yeast-based estrogenic assay.
They showed that administration of 8-PN (15.9 mg/kg/day, equivalent to 100 µg/mL)
in
the drinking water for 72 h increased vaginal mitosis in ovariectomized Swiss albino
mice; however, they did not observe a significant increase in the uterine weight and
the uterine mitosis response [103 ]. In contrast, Overk
et al. [82 ] observed that 8-PN increased the uterine
weight and the height of uterus luminal epithelial cells in Sprague-Dawley rats
significantly; however, an ethanolic extract of hops standardized to its active
constituent, 8-PN, and its metabolic precursors, including isoxanthohumol,
xanthohumol, and desmethylxanthohumol ([Fig. 4 B ]), did
not induce uterotrophy, vaginal cell cornification, and changes in the height of
uterus luminal epithelial cells. Similarly, an in vivo study by Diel et al.
[104 ] in ovariectomized Wistar rats showed that
subcutaneous administration of 8-PN (10 mg/kg/day) increased the uterine weight, the
height of uterine epithelial, and the height of vaginal epithelial cells.
Additionally, ERα and clusterin genes were downregulated and complement C3
was upregulated in the uterus, indicating estrogenic activity of 8-PN in this animal
model [104 ].
Bolca et al. [105 ] showed that disposition of 8-PN in
the womenʼs breast tissue, after hop supplementation for 5 days was associated with
the dose and the metabolism of the precursor compounds. It is believed that the
formation of the estrogenic compound of hops, 8-PN, is closely related to the
metabolism of its precursor, isoxanthohumol ([Fig. 4 B ]), by intestinal microbiota and therefore, subjects with varied
microflora could experience different biological outcomes upon hops administration
[105 ], [106 ], [107 ], [108 ], [109 ]. The difference in uterotrophic effects and
vaginal histology between hops and its active compound, 8-PN, could be related to
the metabolism factor and/or to the other components of hops as an extract. Hops
might contain natural progestins which could counteract the estrogenic effects of
8-PN (discussed in the progestogenic effects section) [11 ]. Moreover, uterotrophy and vaginal cell histology are not the only
measures of estrogenicity. Hops as an estrogenic extract might have more pronounced
estrogenic effects in other target tissues, such as bone, cardiovascular, and brain
which were not evaluated in these studies. In summary, hops flavonoid, 8-PN, is the
most potent phytoestrogen known to date and is equipotent for ER subtypes. Since its
formation depends on the metabolism of its precursors in hops, the estrogenic
activity of hops extract might vary between different subjects depending on their
metabolism characteristics.
Licorice (Glycyrrhiza species, Fabaceae) is a widely used plant, mainly
as a sweetening agent in tobacco, in food and beverages, and in toothpastes. It
consists of more than 30 species from which a few have been studied for several
biological effects such as antibacterial, antiulcer, anti-inflammation, estrogenic,
and chemopreventive [110 ]. Licorice is a common
botanical in menopausal supplements in the United States, either as a single herb
or
in combination with other herbs [42 ]. The estrogenic
activities of different licorice species and extracts are not the same. For example,
Liu et al. [99 ] did not observe any estrogenic effects
with the methanolic extract of Glycyrrhiza glabra (European licorice, the
most common licorice species) when tested in the competitive ER binding assay,
alkaline phosphatase induction in Ishikawa cells, Tff1 mRNA induction in S30
cells, and PgR mRNA induction in Ishikawa cells. However, Dong et al. [111 ] showed that the boiling water extract of
Glycyrrhiza glabra stimulated MCF-7 (ERα +) cell growth at
concentrations of 0.1–10 µg/mL and enhanced ProAB/luciferase activity in the same
cell line at a range of 1–10 µg/mL, which was comparable to the estradiol effect at
10 nM. In this study, the induction of estrogen responsive genes and the activation
of rapid signaling pathways through Erk1/2 and Akt in the
proliferation of MCF-7 (ERα +) cells was observed at 10 µg/mL of the extract,
demonstrating the role of this extract in activating the nonclassical mechanism of
estrogenic activity [111 ]. Simons et al. [112 ] also observed estrogenic activity for several
fractions of an ethyl acetate extract of Glycyrrhiza glabra in the
yeast-based estrogenic assays. The activity of some fractions was abolished in the
presence of either RU58668, a selective ERα antagonist, or
(R,R)-5,11-diethyl-5,6,11,12-tetrahydro-2,8-chrysenediol (R,R -THC), a
selective ERβ antagonist, demonstrating the ER-mediated estrogenic effects.
The difference between the outcomes of these studies could be associated with using
different extracts and concentrations tested as well as the various sources of the
plant species. Simons et al. [112 ] also showed that
glabrene-rich fractions of Glycyrrhiza glabra extract were more estrogenic
with a higher potency for ERα , while glabridin ([Table
1 ], [Fig. 3 E ]) had antiestrogenic
properties. However, Tamir et al. [113 ] showed that
glabridin bound to ER in T47D cell extract (IC50 : 5 µM) stimulated
ER-dependent cell growth at concentrations lower than 10 µM and inhibited cell
growth at concentrations higher than 15 µM in an ER-independent manner. They also
observed increased activation of creatine kinase, a marker of estrogenic activity,
in female rat uterus, epiphyseal cartilage, diaphyseal bone, and cardiovascular
tissues as well as an increased uterine weight effect comparable to that of
E2 . Similarly, Somjen et al. [114 ]
showed that glabridin better than glabrene activated creatine kinase in cultured
female human bone cells as well as in female rat skeletal tissues. They also
reported the estrogenic activity of glabrene and glabridin in vascular tissues in
vitro and in vivo , with glabrene having selective estrogen receptor
modulating-like effects [115 ].
The other popular species of licorice, Glycyrrhiza uralensis (Chinese
licorice) was also reported to be estrogenic in yeast-based estrogen receptor
activity assays, but the reported activities from different studies were not the
same, indicating the lack of a unified standardized extract [116 ], [117 ]. Glycyrrhiza uralensis
extract was reported to stimulate MCF-7 (ERα +) cell growth at concentrations
of 10–100 µg/mL with the maximal growth stimulation comparable to that of estradiol
at 1 nM [118 ]. Cell cycle analysis indicated an
increased population of cells in the S phase, and Western blots showed increased
PCNA levels in response to proliferative concentrations of the extract, confirming
an enhanced cell growth. They also demonstrated reduced levels of ERα protein
as a marker of estrogenicity and a dose-dependent induction of pS2
(Tff1 ) and GREB1 mRNA [118 ]. These
data showed ERα -dependent estrogenic effects by the Glycyrrhiza
uralensis extract. In contrast, an undefined licorice extract was reported
to have no proliferative effects in MCF-7 (ERα +) cells and no uterotrophic
effects in animal models, but possessed ERβ selectivity in ERE-luciferase
induction in transfected HeLa cells [119 ]. The
contradictory results could be associated with using different extracts which
demonstrates the importance of having well-defined standardized licorice
extracts.
Studying various species of licorice cultivated in different regions of the world,
Kondo et al. [120 ] reported that Glycyrrhiza
uralensis and Glycyrrhiza glabra had the highest and the lowest
amounts of liquiritigenin ([Table 1 ], [Fig. 3 E ]), an estrogenic principle of licorice,
respectively. Liquiritigenin was reported to be a highly selective ERβ
agonist in the ER binding assay and ERβ -ERE-luciferase induction assay in
U2OS cells [121 ]. This flavonoid did not enhance
proliferation of MCF-7 (ERα +) xenograft or induction of uterine weight in
nude mice, confirming its better potency for ERβ and the corresponding
pathways [121 ]. Isoliquiritigenin ([Fig. 4 C ]), the precursor of liquiritigenin, was
reported to have estrogenic effects [122 ]. However,
the observed effects could be associated with the conversion of isoliquiritigenin
to
liquiritigenin. Therefore, Glycyrrhiza uralensis is expected to exhibit
stronger ERβ -dependent effects, since it contains the highest amount of
liquiritigenin. However, activation of ERα -dependent responses such as
increased proliferation markers could also be observed in some tissues and/or at
higher concentrations [72 ], [73 ]. In summary, the most common licorice species in dietary supplements
is Glycyrrhiza glabra which contains glabridin and glabrene in addition to
liquiritigenin, while Glycyrrhiza uralenis contains the highest amount of
liquiritigenin, an ERβ selective phytoestrogen. More in depth studies are
needed to define the estrogenic effects of licorice in vitro and in
vivo .
Rhubarb (Rheum species, Polygonaceae) is also a common herb for
menopausal symptoms [42 ]. A variety of estrogenic
activities have been reported for rhubarb extracts. For example, a Rheum
undulatum (rhizome) extract in a yeast-based assay gave an
EC50 = 80 µg/mL with a relative potency of 100 times lower than that of
estradiol [117 ]. Rheum palmatum (root) extracts
were reported to have relative potencies of 2500 and 10 000 times lower than that
of
estradiol in the yeast-based estrogenic assays [116 ], [123 ]. It was also shown that a
rhubarb extract (species not defined) induced ERE-luciferase in
ERα /ERβ transfected TSA201 cells, dose-dependently and the active
constituent was lindleyin ([Table 1 ], [Fig. 3 F ]) with a relative binding potency of 20 000
times lower than that of estradiol for ERα . The extract increased vitelogenin
(a marker of estrogenic activity) levels in the serum of Japanese Medaka [124 ]. These studies showed that different rhubarb
species might have a weak to moderate ER-dependent estrogenic potential. An extract
of Rheum rhaponticum (root), which is very popular in Germany, has also been
studied. Wober et al. [125 ] showed that the extract
activated reporter gene induction through ERβ in transfected HEC-1B
adenocarcinoma cells. Similarly, Moller et al. [126 ]
reported an ERβ activity of the extract in U2OS cells. A three-day in
vivo study on ovariectomized rats by Papke et al. [127 ] showed that the rhubarb extract did not induce uterotrophy or
markers of proliferation. Interestingly, administration of the extract in the
presence of low doses of estradiol (menopausal conditions) suppressed the
uterotrophic effects of estradiol, demonstrating an antiestrogenic effect. An
extended in vivo study for 90 days with ovariectomized rats also confirmed
that the rhubarb extract did not induce uterotrophy or markers of proliferation,
while it showed no effect on the bone mineral density [128 ]. Vollmer et al. [129 ] also reported
that different doses of the rhubarb extract did not enhance the uterine wet weight
and the proliferation marker genes in ovariectomized rats. Interestingly, when the
extract was combined with E2 , it counteracted the uterotrophic effects of
E2 , dose-dependently [129 ]. They
reported that the two compounds, trans -rhapontigenin and desoxyrhapontigenin
([Table 1 ], [Fig. 3 F ]), from the extract bound to both ER subtypes with a slight
preference for ERβ
[129 ]. Activation of ERβ with rhubarb might be
the reason that the extract does not show proliferative effects in uterine tissue,
and its antiestrogenic effects could be related to its partial agonistic effects for
ERs, which manifest as antagonistic activity when the full agonist, estradiol, is
present. In summary, rhubarb is mainly an ERβ activating plant, although its
reported active compounds, lindleyin, rhapontigenin, and desoxyrhapontigenin, are
not ERβ -selective.
Yam (Dioscorea species, Dioscoreaceae) is a common botanical for
managing menopausal symptoms [42 ]. Park et al. [130 ] showed that yam extract at a high concentration
(200 µg/mL) induced PgR and pS2 mRNA in MCF-7 cells after 24 h. These
effects were inhibited when the treatment was combined with ICI 182,780 (1 µM),
indicating an ER-dependent pathway. Similar to E2 , yam extract reduced
the levels of ERα protein and mRNA, measured by Western blot and RT-PCR. This
effect was also blocked by ICI 182,780, showing the estrogenic potential of yam
extract. On the other hand, the extract was antiproliferative in MCF-7 (ERα +)
cells when applied at 20–200 µg/mL for 72 h suggesting that the estrogenic yam
extract did not promote estrogen-dependent tumor cell growth [130 ]. However, the type of the extract and the species
of yam was not defined in this study. Our own observations with yam (Dioscorea
villosa ) showed that the methanolic extract was toxic to Ishikawa and MCF-7
(ERα +) cells at concentrations > 5 µg/mL, and therefore the estrogenic
activity could not be evaluated (unpublished data). Cheng et al. [131 ] showed that the ethyl acetate extract of yam
(Dioscorea alata ) at 10 µg/mL weakly induced the transcriptional
activation of GAL4-responsive alkaline phosphatase reporter in CHO-K1 cells with
either ER subtypes, with a slightly stronger activity with ERα . Similarly,
when yam (Dioscorea alata ) was given to menopausal women at 390 g/day as part
of their food for 30 days, a significant increase in serum concentrations of
estrone, sex hormone binding globulin (SHBG), and an increase in estradiol was
observed, showing that the yam diet enhanced the hormone levels in these subjects
[132 ]. Diosgenin ([Table
1 ], [Fig. 3 G ]) isolated from yam, was used
in pharmaceutical industry to synthesize progesterone and cortisone [133 ] and was shown to have estrogenic activity in an
animal model [134 ]; however, there are few recent
reports about its estrogenic activity. The concentration of diosgenin is relatively
low in yam species, and it will not biochemically convert to estrogens in
vivo
[131 ]. Therefore it is not clear, how yam ingestion
could lead to increased estrogen levels in menopausal women and which components
might be the active principle(s).
Chasteberry (Vitex agnus-castus , Lamiaceae) is also a popular botanical
added to botanical supplements for womenʼs health [42 ].
It was shown to have a weak binding affinity for ERs and no alkaline phosphatase
induction in Ishikawa cells; however, it induced PgR mRNA in this cell line
[99 ]. Activation of PgR while other
estrogenic markers are negative could be associated with possible progestogenic
effects of chasteberry (discussed in progestogenic effects section). Liu et al.
[135 ] reported that the methanolic extract of
Vitex agnus-castus had a weak binding affinity for ERs (IC50
ERα = 46 µg/mL, ERβ = 64 µg/mL) and upregulated ERβ mRNA in
T47D:A18 and PgR in Ishikawa cells, while inducing alkaline phosphates enzyme
in Ishikawa cells. Linoleic acid ([Fig. 3 H ]) has been
found as the “active” estrogenic component of chasteberry based on bioassay-directed
fractionation of the crude extract using the ER binding assay. In this study, while
linoleic acid induced ERβ in T47D:A18 cells and PgR in Ishikawa cells,
it did not induce alkaline phosphatase activity in Ishikawa cells. However, linoleic
acid is a fatty acid and may contribute to nonspecific binding to ERs and PRs,
generating false positive results [135 ].
Ibrahim et al. [136 ] showed that an ethanolic extract
of Vitex agnus castus increased uterine weight in Sprague-Dawley rats in
addition to an increase in the plasma levels of progesterone and estrogen and a
decrease in LH and prolactin, suggesting an estrogenic effect of chasteberry. It was
shown that chasteberry extract had a selective binding to ERβ, and
bioassay-guided fractionation of the crude extract led to the isolation of apigenin
([Table 1 ], [Fig. 3 H ]), the most selective ERβ ligand in this plant [137 ]. Choi et al. [138 ]
also observed ERβ selectivity with apigenin in the competitive binding assay
in addition to estrogenic activity in the yeast-based assay and MCF-7 (ERα +)
cell growth. Based on these studies, apigenin can preferentially activate
ERβ -dependent responses; but can also stimulate ERα -dependent effects
at higher concentrations or in certain tissues. Apigenin could also induce
progestogenic activity (discussed in progestogenic effects section) [11 ]. Jarry et al. [139 ]
isolated penduletin ([Table 1 ], [Fig. 3 H ]) from chasteberry, which was also an ERβ selective
agonist in the ER binding assay (IC50 : 0.31 µM). However, the presence of
ERβ ligands in chasteberry in addition to the progestogenic effects of
apigenin did not oppose the proliferative responses of chasteberry extract in
vivo
[136 ], which could be associated with the amounts of
these compounds in the plant extract and their insufficient bioavailability.
Hu et al. [140 ] reported MCF-7 (ERα +) cell
proliferation with four different species of Vitex . They reported that the
essential oil of Vitex rotundifolia , which was mainly composed of linoleic
acid, strongly stimulated MCF-7 (ERα +) cell proliferation, the effect which
was inhibited by ICI 182,780, demonstrating an ERα -dependent activity of the
linoleic-rich fractions [141 ]. Additionally, they
found that Vitex rotundifolia and its components agnuside ([Fig. 3 H ]) and rotundifuran ([Fig. 3 H ]) induced MCF-7 (ERα +) cell proliferation, EST1
(ERα ), PgR , and Tff1 mRNA dose-dependently and the effects
were inhibited by ICI 182,780 [142 ]. Therefore,
according to distinct chemical profiles and biological activities of different
Vitex species, identification of the species is very important,
especially for the standardization of the botanical supplements. In summary,
Vitex species have estrogenic properties, and compounds such as apigenin
and penduletin are their ERβ -selective compounds, while rotundifuran and
agnuside have been reported to activate ERα -dependent responses.
Dong quai (Angelica sinensis , Apiaceae) is another popular botanical
for managing menopausal symptoms as well as womenʼs health issues in general [42 ]. The estrogenic activity of dong quai is still
controversial [143 ]. For example, Amato et al. [119 ] reported that dong quai had proliferative effects
in MCF-7 (ERα +) cells but did not activate ERα /ERβ -dependent
luciferase transcription in transfected HeLa cells and did not exert uterotrophic
effects in CD-1 mice. However, Circosta et al. [144 ]
observed increased uterine weight, modified vaginal cytology, and reduced
luteinizing hormone levels in female Wistar rats treated with an ethanolic extract
of dong quai. Similarly, cell based investigations revealed controversial results.
Liu et al. [99 ] reported that dong quai methanolic
extract did not bind to ERs, induce alkaline phosphatase activity in Ishikawa cells,
or induce estrogen sensitive genes (PgR and Tff1 ) mRNA in Ishikawa and
S30 cells, respectively. Similarly, Zhang et al. [123 ]
published that an ethanolic extract of dong quai was not estrogenic in a yeast-based
assay over the concentration range of 0.1–1000 µg/mL. A recent study showed that
wine-processed dong quai extract at 1 mg/mL (very high concentration) had no
proliferative effect on MCF-7 (ERα +) cells but it induced ERE-luciferase
[145 ]. On the other hand, Lau et al. [146 ] observed proliferation of MCF-7 (ERα +)
cells with dong quai concentrations > 100 µg/mL, which could be blocked by
4-hydroxytamoxifen, demonstrating a weak ER-dependent estrogenic activity.
Interestingly, Rosenberg-Zand et al. [147 ] showed that
an ethanolic extract of dong quai blocked Tff1 mRNA induction in BT474 breast
cancer cells, demonstrating an antiestrogenic effect by this herb. Similarly,
Godecke et al. [148 ] reported that the lipophilic
fraction of a methanolic extract (rich in ligustilide) of dong quai at 20 µg/mL
significantly inhibited alkaline phosphatase induction in the presence of estradiol
in Ishikawa cells, suggesting an antiestrogenic potential. To date, there have been
no reports of a purified compound which could be responsible for the observed
estrogenic/antiestrogenic properties of dong quai. These studies demonstrate that
additional studies with well-defined extracts are needed in order to delineate the
estrogenic/antiestrogenic potential of dong quai as well as the active compound. One
reason for the contradictory results regarding the estrogenic activities of dong
quai could be associated with the instability of its phthalide fractions, in
particular ligustilide [149 ]. Conclusive data on the
relative estrogenic effects of dong quai are currently unavailable.
Black cohosh (Cimicifuga racemosa , Ranunculaceae) is the most popular
botanical for menopausal symptom relief in the United States [42 ]. However, reports of its estrogenic activities are controversial
[150 ]. Liu et al. [99 ] showed that black cohosh did not bind to ERs, did not induce alkaline
phosphatase in Ishikawa cells, or induce PgR and Tff1 mRNA in Ishikawa
and S30 cells, respectively. Amato et al. [119 ]
confirmed these results when they reported that black cohosh did not induce MCF-7
(ERα +) cell proliferation, ERE-luciferase in tranfected HeLa cells, and
uterotrophic effects in animal models. Bodinet et al. and Freudenstein et al. [151 ], [152 ] showed that
standardized isopropyl extract of black cohosh inhibited proliferation of
estrogen-dependent MCF-7 (ERα +) cells, dose-dependently. Similarly, Gaube et
al. [153 ] showed that a dichloromethane extract of
black cohosh inhibited MCF-7 (ERα +) cell proliferation dose-dependently
(IC50 : 14.7 µg/mL), and the majority of proliferation control and
proapoptotic genes were down-regulated. Lupu et al. [154 ] also did not observe any estrogenic effect with black cohosh in an
array of estrogenic assays in estrogen responsive cells. Similarly, Zierau et al.
[155 ] reported that ethanolic and isopropyl
extracts of black cohosh did not enhance MCF-7 (ERα +) cell proliferation and
did not show estrogenic effects in the yeast-based assay and ERE-luciferase assay
in
MCF-7 (ERα +) cells. Interestingly, the extracts inhibited estradiol-induced
MCF-7 (ERα +) cell proliferation and mRNA expression, demonstrating possible
antiestrogenic effects of the black cohosh extracts. It was also shown that
isopropyl extract of black cohosh did not influence the uterine weight and vaginal
cytology in Sprague-Dawley rats in the presence or absence of estradiol [156 ]. Mercado-Feliciano et al. [157 ] also did not observe any estrogenic or antiestrogenic effects in the
uterus of female B6C3F1/N mice treated with different doses of an ethanolic extract
of black cohosh for 3 days. Another study also demonstrated no classical estrogenic
effects in ovariectomized rats after three months of treatment with an ethanolic
extract of black cohosh [158 ]. Ruhlen et al. [159 ] observed a reduction of hot flashes in women
taking a black cohosh extract containing 2.5 % triterpenes for 12 weeks. The effect
returned to baseline after a 12-week washout period. The extract did not have any
effect on serum estrogenic markers, pS2 expression levels, and cellular
morphology in the womenʼs nipple aspirate fluids, demonstrating no detectable
estrogenic effect on the breast tissue [159 ]. In
contrast, Liu et al. [160 ] observed a significant
MCF-7 (ERα +) cell growth and ER upregulation in response to a black cohosh
treatment, but they did not mention the type of the extract. Bolle et al. [161 ] also reported a weak estrogenic activity for black
cohosh in the yeast estrogenic assay; however, they did not observe a uterotrophic
effect with the extract. It is still not clear whether black cohosh has estrogenic
activity or other mechanisms of action (i.e., serotonergic discussed below) even
though it is currently the most popular dietary supplement used by menopausal women
[162 ].
Botanicals with Serotonergic Activity
Some botanicals such as black cohosh are widely used for managing menopausal
symptoms, although there is little evidence for their hormonal effects [150 ], [162 ]. Since there
have been reports of relief of hot flash intensity and frequency by SSRIs [167 ], botanicals have been studied for potentially
similar mechanisms.
Black cohosh (Cimicifuga racemosa , Ranunculaceae) probably does not
have estrogenic activity as discussed above. However, it remains one of the most
popular botanical supplements for menopausal symptoms implying alternative efficacy
mechanisms. Therefore, black cohosh was studied for its potential serotonergic
effects [156 ]. A black cohosh isopropyl extract was
shown to inhibit the binding of [3 H] lysergic acid diethylamide to
5-HT7 , one 5-HT-subtype that is associated with thermoregulation in
the hypothalamus (IC50 = 2.4 µg/mL) [156 ].
In the same study, Burdette et al. showed that a methanolic extract of black cohosh
elevated cAMP levels in 5-HT7 transfected HEK 293 cells, suggesting that
the extract acted as a partial agonist of the receptor. This effect was reversed in
the presence of methiothepin, the antagonist of 5-HT7 , suggesting a
receptor-mediated process. Powell et al. [12 ] also
showed that the methanolic extract of black cohosh displayed 5-HT7
binding activity and induced cAMP production. Using bioassay-guided fractionation,
N
ω
-methylserotonin ([Fig. 6 E ]) was identified as a potent ligand for the 5-HT7
receptor (IC50 = 23 pM) which induced cAMP production
(EC50 = 22 nM) and blocked serotonin reuptake (IC50 = 490 nM).
Nadaoka et al. [168 ] observed that black cohosh
attenuated the conversion of 5-HT to its metabolite 5-hydroxyindoleacetic acid in
the hypothalamus, hippocampus, and cortex of the mice subjected to an acute
immobilization stress, demonstrating elevated levels of 5-HT upon black cohosh
administration. Although the clinical trials on the efficacy of black cohosh in
relieving menopausal symptoms are not conclusive [169 ], [170 ], the positive effects reported
by some women might be attributed to its ability to activate serotonin receptors and
block serotonin reuptake, leading to enhanced serotonergic activity. More animal
model studies are necessary to provide further evidence for serotonergic activity
of
black cohosh.
Fig. 6 Chemical structures of serotonergic compounds found in
botanicals.
Kudzu (Pueraria lobata , Fabaceae) methanolic extract at a high
concentration (2 g/kg) as well as puerarin (1–30 mg/kg) ([Fig. 6 B ]) systemic administration in male Sprague Dawley rats induced
hypothermia [13 ]. The same effect was observed with
i. c. v. injection of 50–100 µg of puerarin. A significant correlation between the
measured hypothermia and reduced 5-HT efflux in the rat hypothalamus was observed,
indicating the role of the serotonergic system in inducing hypothermia by kudzu
extract and puerarin as its active compound [13 ]. The
well-known components of kudzu, soy, and red clover, genistein (10 nM – 10 µM), were
shown to stimulate [3 H] serotonin uptake in transfected COS-7 cells,
demonstrating its potential serotonergic effects [171 ].
Kava (Piper methysticum , Piperaceae) was shown to have
neurotransmitter-like effects [172 ]. The extract of
kava leaves was reported to have GABAA receptor binding activity
(IC50 = 3 µg/mL), dopamine D2, opioid (µ and δ ), and
histamine (H1 and H2 ) receptor binding activity
(IC50 = 1–100 µg/mL) as well as a weak binding to serotonin
(5-HT6 and 5-HT7 ) and benzodiazepine receptors [172 ]. The active principles of kava with serotonergic
activities were reported to be the kavalactones including, kavain,
7,8-dihydrokavain, methysticin, 7,8-dihydromethysticin, yangonin, and
5,6-demethoxyyangonin ([Fig. 6 F ]).
Licorice (Glycyrrhiza glabra , Fabaceae) also was reported to contain
compounds with serotonergic effects. Glabridin, 4′-O -methylglabridin, and
glabrene ([Fig. 6 C ]) from Glycyrrhiza glabra
inhibited the reuptake of radioactive serotonin in HEK-293 cells at 50 µM, with
glabridin having a dose-dependent effect [173 ].
Dong quai (Angelica sinensis , Apiaceae) has been shown to have
serotonin-like activity in the 5-HT7 serotonin receptor binding assay,
and its most active compound was p -hydroxyphenethyl trans -ferulate
(IC50 = 47.6 µM) ([Fig. 6 D ]) [174 ]. Deng et al. [174 ]
reported that p -hydroxyphenethyl trans -ferulate,
Z -butylidenephtalide, 11(S ),
16(R )-dihydroxyoctadeca-9Z ,17-diene, 8-hydroxy-1-methoxy-,
Z -9-heptadecene-4,6-diyn-3-one, and imperatorin ([Fig. 6 D ]) isolated from the methanolic extract of dong quai weakly bound
to the 5-HT7 receptor. These data suggest that dong quai might have
serotonergic activity.
