Key words
Clusiaceae - estrogenic activity - hyperforin -
Hypericum perforatum
- Proteomics - St. Johnʼs wort
Abbreviations
17β-E2:
17β-estradiol
AKR1C3:
aldo-keto reductase family 1 member C3
CAT:
catalase
CCND1:
cyclin D1
CTSD:
cathepsin D
ERE:
estrogen response element
ERK:
extracellular signal-regulated kinases
ESR:
estrogen receptor
IL6ST:
interleukin-6 signal transducer
NOX:
NADPH oxidase
NUMA:
nuclear mitotic apparatus protein 1
PRDX1:
peroxiredoxin 1
PRDX2:
peroxiredoxin 2
PRDX6:
peroxiredoxin 6
PTM:
post-translational modification
SJW:
St. Johnʼs wort
Introduction
Estrogens play important roles in reproductive physiology and multiple diseases, including
breast and endometrial cancers, cardiovascular disease, osteoporosis, and Alzheimerʼs
disease [1], [2]. The most active form of endogenous estrogen, 17β-E2, is mostly synthesized in the ovaries of premenopausal women. To exert its effect,
17β-E2 binds to specific ESR1 and ESR2, after which the activated estrogen-ESR complex
is transported to the nucleus, where it binds to the ERE in the promoter regions of
estrogen-dependent genes, thereby altering gene expression [3]. The maintenance of appropriate levels of estrogen is important for reducing post-menopausal
symptoms; however, the hypoestrogenic state is associated with an increased risk of
cardiovascular disease, osteoporosis, and complications in pregnancy and lactation,
while the hyperestrogenic state promotes cancer cell proliferation [4], [5].
Estrogen replacement therapy (ERT) appears to reduce the risk of osteoporosis, colorectal
cancer, and the severity of postmenopausal symptoms [6], [7], [8], [9]. Nevertheless, the safety of ERT is a major concern due to the higher incidence
of breast cancer, cardiovascular disease, and stroke among postmenopausal women undergoing
therapy [8], [9], [10]. In this respect, phytoestrogens, a diverse group of plant-derived nonsteroidal
compounds that mimic 17β-E2 because of their structural similarity to mammalian estrogens, could be alternatives
to conventional ERT [11]. Phytoestrogens, including glycinol in soybean and resveratrol in grapes, are reported
to exert estrogenic effects [12], [13]. Lignans are thought to exert protective effects against cancer; however, the anticancer
effects of glabridin, genistein, quercetin, and resveratrol are inconsistent [14], [15], [16]. Furthermore, phytoestrogens act as both estrogen agonists and antagonists, and
differ in their levels of estrogenic activity [17], [18]. In addition, herbal plants, such as kudzu root, chasteberry, red clover, hops,
and agrimony, show estrogenic activities in the uterus and in breast cancer cell lines
[19], [20], [21]. Among those, hops and red clover contain 8-prenylnaringenin and genistein, respectively,
as the major estrogenic compounds [21], [22]. Ginseng, which contains ginsenosides as active compounds, prevents postmenopausal
osteoporosis and cancer cell proliferation by modulating ESR1 activity [23], [24].
SJW (Hypericum perforatum L., Clusiaceae) contains several bioactive compounds, the main ones being hyperforin
and hypericin. SJW extract has been reported to have a therapeutic effect on mild-to-moderate
depression, relieve the psychological and psychosomatic symptoms of menopause, and
possess antimicrobial, anti-inflammatory, antioxidant, and free radical scavenging
properties [25], [26], [27], [28]. Hypericin has been found to show antiretroviral activity, while hyperforin has
antidepressant and anti-inflammatory activities [29], [30], [31]. Moreover, several studies evaluated the role and mechanism of SJW for the treatment
of estrogen-mediated menopausal symptoms. A previous study reported that SJW regulates
the genes related to antidepressant activity [32]. Liu et al. presented the possible mechanisms of SJW extract for the reduction of
menopausal symptoms, including depression and hot flashes [33]. Although many studies have dealt with the estrogen-related activities of SJW and
its constituents, the mechanisms of the activities of hyperforin have not yet been
fully characterized.
In the present study, we assessed the estrogenic activities of hyperforin by measuring
evoked ERE-luciferase activity and cell proliferation in MCF-7 cells. In addition,
we used proteomic approaches to examine the modulation of proteomic profiles and their
post-translational modifications in cells treated with hyperforin and 17β-E2 to identify the underlying mechanism.
Results and Discussion
The components of SJW extract were separated as a total ion chromatogram using UFLC-MS
(Fig. 1S A, Supporting Information). To confirm the identification, the electrospray ionization
(ESI) mass spectra of the compound was obtained from the sample under similar operating
conditions, and was found to be comparable to the spectra of the authentic reference
sample. Among the different peaks, the largest peak area was identified as hyperforin
(denoted by the green chromatogram) and had a retention time of 9.0. The mass spectra
for hyperforin showed ionic species at m/z 535.4 (Fig. 1S A, B, Supporting Information), which confirms its presence in SJW. The amount of hyperforin
and hypericin in the extract was 3.5 mg/g and 2.6 mg/g, respectively.
The estrogenic activities of SJW extract and hyperforin in MCF-7 cells transfected
with ESR1 or ESR2 were tested using ERE-luciferase activity assays. The results are
expressed as percentages of relative luciferase activity (RLA) compared to the untreated
control, and for comparison, a positive control was also prepared by treating the
transfectants with 10−2 µM 17β-E2. The RLA of the transfected cells after treatment with SJW extract (0.2 to 20 µg/mL)
and hyperforin (10−5 to 10 µM) is shown in [Fig. 1]. Both samples significantly increased the RLA in cells transfected with ESR1 or
ESR2 in a concentration-dependent manner. At a concentration of 20 µg/mL, SJW extract
elicited responses that were 1.3-fold smaller (RLA: 311.2 % vs. 412.9 %) or 2.0-fold
larger (297.8 % vs. 146.7 %) than those elicited by 10−2 µM 17β-E2 in ESR1 or ESR2 transfectants, respectively ([Fig. 1 A, B]). In contrast, irrespective of whether the cells were transfected with ESR1 or ESR2,
RLA levels elicited by 10−2 µM hyperforin (RLA: ESR1, 252.8 % and ESR2, 63.4 %) were lower than those elicited
by the same concentration of 17β-E2 (RLA: ESR1, 412.9 % and ESR2, 146.7 %) ([Fig. 1 C, D]). This finding is consistent with that of Harris et al. who observed a difference
between RLAs elicited by 17β-E2 and phytoestrogen in breast cancer cells transfected with either ESR1 or ESR2
[34].
Fig. 1 Effects of St Johnʼs wort (SJW) extract, hyperforin (HF), and 17β-estradiol (17β-E2) on ERE-luciferase activity in MCF-7 cells. The luciferase activities in MCF-7
cells transfected with ERα (A, C) or ERβ (B, D) and an ERE reporter plasmid were measured after treatment with 0.2 to 20 µg/mL SJW
extract (A, B) or 10−5 to 10 µM HF (C, D). A positive control was prepared by treating the cells with 10−2 µM 17β-E2. Relative luciferase activities were calculated as percentages of the induced
luminance relative to control after normalization to the Renilla luciferase activity. Bars represent the averages of triplicate determinations. Asterisks
indicate significant differences from the control (Ctrl) determined using Dunnettʼs
multiple comparison t-test (*p < 0.05 and *** p < 0.001).
[Fig. 2] shows the concentration-dependent effects of the SJW extract, hyperforin, and 17β-E2 on the proliferation of MCF-7 cells. SJW extract induced cell proliferation with
a 50 % effective concentration (EC50) of 2.1 µg/mL and maximum response (Emax) of 124.7 % of the control ([Fig. 2 A]). The EC50 of hyperforin and 17β-E2 was 3.3 × 10- 3 µM and 1.8 × 10−2 µM, respectively, indicating that the EC50 of hyperforin was lower than that of 17β-E2. Similarly, the Emax of hyperforin was lower than that of 17β-E2 (Emax: hyperforin, 112.7 % vs. 17β-E2, 128.5 %). These results indicated that hyperforin possessed lower potency (EC50 value) and efficacy (Emax value) to that of 17β-E2 for inducing cell proliferation ([Fig. 2 B]).
Fig. 2 Concentration-response curves of St. Johnʼs wort (SJW) extract, hyperforin, and 17β-estradiol for MCF-7 cell proliferation. MCF-7 cells were cultured in estrogen-free
medium, treated with 0.2–20 µg/ml SJW extract or with 10−5 – 10 µM of hyperforin or 17β-estradiol. Cell proliferation was determined using the Cell Counting Kit-8, and calculated
as percentage of cell proliferation relative to untreated control. The curves of cell
proliferation versus logarithm of concentration of SJW extract (A), hyperforin, or 17β-estradiol (B) were plotted. The curves represent the averages of five determinations expressed
as percentages.
To determine whether the cell proliferation induced by SJW extract and hyperforin
reflected their estrogenic activities, we next examined the effects of the ER antagonist
ICI 182,780 on the response. The proliferation of cells treated with SJW extract (20 µg/mL)
and hyperforin (10 µM) increased significantly when compared to the control cells
in the absence of ICI 182,780 (p = 0.000 and p = 0.001, respectively). However, we
found that the ER antagonist ICI 182,780 significantly inhibited cell proliferation
induced by the same concentration of SJW extract and hyperforin by 16.9 % (p = 0.000)
and 6.6 % (p = 0.007), respectively, when compared to the corresponding sample without
ICI 182,780 ([Fig. 3]), indicating that the induced cell proliferation was related to their estrogenic
activities. Notably, ICI 182,780 had a 1.2-fold and 3.1-fold stronger effect on 17β-E2-induced proliferation than it did on SJW-induced and hyperforin-induced proliferation,
respectively. The inhibitory effect of ICI 182,780 on 17β-E2-induced cell proliferation has also been reported in the rat uterus, which could
be due to multiple steps, including binding to the ESR by ICI 182,780 followed by
disrupting ESR nuclear localization, and reducing binding of ESR to ERE [35], [36].
Fig. 3 Inhibitory effect of ICI 182,780 on MCF-7 cell proliferation induced by St. Johnʼs
wort (SJW) extract, hyperforin (HF), or 17β-estradiol (17β-E2). MCF-7 cells were cultured for 96 h in estrogen-free medium and treated with
20 µg/mL of SJW extract, 10 µM HF, or 10 µM 17β-E2 in the presence and absence of ICI 182,780. Cell counts were normalized to an
untreated control sample. Bars represent the averages of triplicate determinations
expressed as percentages. Asterisk (*) indicates significant differences when compared
to the untreated control (without ICI 182,780) determined using Dunnettʼs multiple
comparison-test (***p < 0.001), and hash (#) indicates significant difference when
compared to the corresponding sample without ICI 182,780 (##p < 0.01 and ###p < 0.001)
determined using Studentʼs t-test.
Proteomic profiles of three samples, including control, 17β-E2-treated, and hyperforin-treated cells, were generated by nano-UPLC-MS analysis.
A total of 453 proteins were identified in the three samples, and among them, 282
proteins were differentially modulated in the hyperforin-treated cells compared to
the 17β-E2-treated cells (Table 1S, Supporting Information). The proteomic data were further integrated into a knowledge
database supported by Ingenuity pathway analysis (IPA) to visualize the interaction
network composed of identified and predicted proteins. [Fig. 4] shows the networks of proteins that were differentially expressed in response to
hyperforin treatment relative to 17β-E2 treatment in MCF-7 cells. Mainly, hyperforin was predicted to be more inhibitive
of ESR1 activity than 17β-E2, leading to the downregulation of cell proliferation. However, hyperforin was
predicted to activate ESR2 activity, leading to the reduced production of reactive
oxygen species (ROS).
Fig. 4 Network derived with comparing proteins expressed in hyperforin-treated and 17β-estradiol (17β-E2)-treated MCF-7 cells. The proteins shown in Tables 1S and 2S, Supporting Information, were imported into IPA, and specific proteins related to
cell proliferation and ROS generation were selected for the network using the Ingenuity
knowledge database. Significantly upregulated and downregulated (p value < 0.05) proteins
in hyperforin-treated cells compared to 17β-E2-treated cells are shown in red and green, respectively. The proteins, which were
predicted to be activated and inhibited, are shown in orange and blue, respectively.
Post-translational modification, phosphorylation (P), and acetylation (A) of proteins
are indicated by enclosed text. (Color figure available online only.)
In our study, cell proliferation-related proteins, mainly CCND1, ERK, IL6ST, and AKR1C3,
were downregulated by hyperforin treatment compared to 17β-E2 treatment, indicating that hyperforin less likely leads to cancer progression.
A previous study reported that activation of ESR1 by 17β-E2 treatment subsequently triggered ERK signaling, which supported the upregulation
of the expression of CCND1 and further induced cell proliferation [37], [38]. Similar to CCND1, CTSD, a prognostic biomarker of cancer cell proliferation, was
downregulated by hyperforin compared to17β-E2 [39]. Downregulation of IL6ST by hyperforin suggests that hyperforin abrogates cancer
cell proliferation by impairing CCND1 and ERK expression. Furthermore, 17β-E2-stimulated AKR1C3 also was downregulated by hyperforin, which suggests that cell
proliferation is inhibited by controlling the activities of estrogen and progesterone
[40]. AKR1C3 induces cell proliferation by catalyzing the reaction either to reduce estrone
to the more potent 17β-E2 or to reduce progesterone to the less potent pregnanediol [41]. AKR1C3 controls concentrations of estrogens and progesterone, which further regulates
the activities of ESR1 and ESR2.
To protect against the harmful consequences of oxidative stress, IPA predicted that
a number of prooxidant and antioxidant enzymes, including CAT, NOX, PRDX1, and PRDX6
were modulated by hyperforin and 17β-E2 through the activation of ESR2. Hyperforin and 17β-E2 affect oxidative stress by increasing CAT activity, which is dependent on the
ratio of ESR1 and ESR2 in the cancer cell. Moreover, hyperforin, similar to other
phytoestrogens, acts as an inhibitor of NOX, resulting in the decreased production
of ROS via ESR2 [42]. IPA also showed that the activity of NOX is dependent on PRDX6 [43]. Hyperforin suppressed PRDX6 expression, which subsequently decreased NOX activity.
Additionally, IPA indicated that the hyperforin-induced activation of ESR2 reduced
oxidative stress relative to 17β-E2. A previous study also reported that a low ESR1/ESR2 ratio is a hallmark for the
protection of cancer cells against oxidative stress [44].
Along with protein expression, ER function is also regulated by PTMs of the proteins.
In this study, proteomic analysis identified 59 post-translationally modified proteins.
Among the PTMs, phosphorylation, acetylation, and oxidation were observed in 47, 15,
and 3 proteins, respectively. Proteins that were phosphorylated or acetylated included
heat shock protein 90B (HSP90B), PRDX2, and isoform 2 of the NUMA, which are all involved
in cell proliferation and ROS production (Fig. 4S and Table 2S, Supporting Information). However, notably, phosphorylation of NUMA was significantly
inhibited by hyperforin treatment when compared to 17β-E2 treatment (hyperforin/17β-E2 ratio, 0.867). NUMA is crucial for mitotic spindle formation during progression
of the cell cycle. Localization of NUMA to the mitotic spindle is regulated by its
phosphorylation, which transforms NUMA into a soluble component of the mitotic pole
at the onset of mitosis [45]. Our PTM analysis showed lower levels of phosphorylated NUMA in hyperforin-treated
cells than in 17β-E2-treated cells, indicating that cancer cell proliferation was inhibited by blocking
mitosis.
A limitation of this study is that our in vitro findings represent a portion of the animalʼs metabolic system, and does not address
the bioavailability of the compounds. The effects of a compound in an in vitro study may not be directly replicated in an animal model. Therefore, these compounds
that showed estrogenic potential in vitro need to be tested in vivo for their efficacy.
In conclusion, hyperforin can mimic the estrogenic activity of 17β-E2 by showing comparable effects on ERE-luciferase activity in vitro, in MCF-7 cells. In this study, for the first time, we attempted to identify comparative
protein expression in hyperforin and 17β-E2 treatment. The expression of proteins related to cell proliferation and ROS production
was differentially regulated. Hyperforin treatment induced less cancer cell proliferation
than 17β-E2 treatment, possibly by downregulating CCND1 and ERK expression through ESR1. At
the same time, hyperforin-treated cells showed lower rates of ROS production than
17β-E2-treated cells, possibly by the downregulation of NOX through ESR2. Hence, this
study will aid in the identification and development of a safe alternative method
for estrogenic regulation for the purpose of reducing postmenopausal symptoms in women.
Materials and Methods
Sample preparation
The leaves and flowers of SJW obtained from a local producer (Cooperative of Daeho-dong,
Naju, Korea) were used for this study because the compounds, including hyperforin,
hypericin, chlorogenic acid, hyperoside, and quercetin, are mainly present in those
parts of the plant [46]. A voucher specimen (identification number: JBF-FRI-B-2012–0001) of the plant used
in our experiment was deposited at the Jeonnam Biofood Technology Center, Korea. First,
500 g of dried leaves and flowers of SJW were extracted in 5 L of 75 % ethanol for
8 h and extracted again in 2.5 L 75 % ethanol for 5 h, with stirring in the dark.
The resultant mixture was filtered through a bag filter (1 µm), after which the solvent
was evaporated at 42 °C using a rotary vacuum evaporator (Daesin Machine Industry).
The remaining extract was then freeze-dried, packed into screw-capped vials under
nitrogen, and stored at − 20°C for later use. Stock solutions of SJW extract and 17β-E2 (Sigma Aldrich, purity ≥ 98 %) were prepared in DMSO (purity ≥ 99.7 %, Sigma-Aldrich)
at concentrations of 200 µg/mL and 10 mM, respectively. Stock solutions of hyperforin
(Cayman Chemical Co., purity ≥ 90 %) were prepared in methanol and stored at − 20 °C.
Transient transfection and luciferase activity assay
For transient transfection, MCF-7 cells (Korean Cell Line Bank) were cultured in DMEM/Nutrient
Mixture F-12 medium (DMEM/F-12, Gibco Life Technologies) supplemented with 10 % fetal
bovine serum, 100 U/mL penicillin, 100 U/mL streptomycin, and 1 % bovine insulin as
previously described [20]. The cells were plated in 96-well plates at a density of 2.0 × 104 cells/well. After 24 h, the cells was transfected using FuGENEHD transfection reagent
(Promega) according to the manufacturerʼs protocol. Briefly, Opti-MEM was used to
dilute ERα (pEGFP-C1-ERα, Addgene), ERβ (pcDNA Flag ERβ, Addgene), ERE (3× ERE TATA luc, Addgene), and Renilla luciferase (pRL-SV40, Promega) prior to the addition of the transfection reagent.
The cells were incubated for 24 h at 37°C under a humidified 5 % CO2 atmosphere [47], [48] in phenol red-free DMEM/F-12 containing 5 % charcoal-dextran stripped fetal bovine
serum with 0.2, 2.0, or 20 µg/mL SJW extract or 10−5, 10−2, or 10 µM hyperforin. Following the incubation, the cells were washed twice with
PBS and lysed in 20 µL of passive lysis buffer (Promega). Luciferase activity was
then measured using Luciferase Assay Reagent (Promega) in a GloMax Multi Microplate
Luminometer (Promega) according to the manufacturerʼs protocol. The RLA was calculated
as previously described [49]. The positive control was prepared by treating the cells with 10−2 µM 17β-E2, while the negative control was prepared by using the vehicle solvent only.
Cell proliferation assay
MCF-7 cells were seeded into 96-well microplates at a density of 5 × 103 cells/well in culture medium. After 24 h, the medium was replaced with estrogen-free,
phenol red-free DMEM/F-12 containing 5 % charcoal-dextran-stripped fetal bovine serum.
Different concentrations of SJW extract or hyperforin were added to the medium as
described in the luciferase activity assay, and the cells were cultured for 96 h.
In addition, to investigate the effect of an estrogen antagonist, duplicate test samples
were prepared using 20 µg SJW extract, 10 µM hyperforin, and 10 µM 17β-E2, with or without the ER-antagonist ICI 182,780 (10−3 µM, Santa Cruz Biotechnology, purity ≥ 98 %). The positive control was prepared by
treating the cells with 10−5, 10−2, and 10 µM 17β-E2, while the negative control was prepared by using the vehicle solvent only [20]. Cell proliferation was assessed using a Cell Counting Kit-8 (Dojindo) as described
by Nakagawa et al. [50]. The proliferation was expressed as a percentage compared to the negative control.
Concentration-response curves were plotted and the EC50 and Emax were estimated using GraphPad Prism version 5.01 (GraphPad Software, Inc.).
Protein extraction and tube-gel protein digestion
Proteins were extracted as previously described [51]. The protein was quantified using a Detergent Compatible Protein Assay Kit (Bio-Rad).
A tube-gel digestion protocol was adopted as described previously [51].
Nano-UPLC-HDMSE
Tryptic peptide mixtures were separated using nano-ACQUITY ultra-performance liquid
chromatograph (UPLC) equipped with a Synapt G2-Si HDMS System (Waters Corp.), a previously
described method with optimization of the mobile phase system [51]. A gradient elution program was conducted for chromatographic separation with mobile
phase A (0.1 % formic acid in water) and mobile phase B (0.1 % formic acid in acetonitrile)
as follows: 97 % mobile phase A initially, 90 % mobile phase A for 3 min, 65 % mobile
phase A for 150 min, gradual decrease to 20 % mobile phase A over 160 min, and a sharp
increase to 97 % mobile phase A for the last 10 min.
Identification and relative quantification of protein
For identification of proteins, MS spectra of peptides were aligned using Progenesis
QI for Proteomics (QIP) version 2.0 (Nonlinear Dynamics), a previously adopted method
with modifications in criteria for protein identification [51], and the spectra were matched to human proteins using the International Protein
Index (IPI) human database (v.3.87). The criteria for protein identification were
set as follows: ≥ 3 fragment per peptide, ≥ 7 fragments per protein, and ≥ 2 peptides
per protein. Carbamidomethylation of cysteine was set as fixed, and oxidation of methionine
and phosphorylation of serine/threonine/tyrosine were set as variable modifications.
Bioinformatics analysis
Ingenuity pathway analysis (IPA version 9.0; Ingenuity Systems, Inc.) was used to
perform knowledge-based network analysis of proteomics data.
Data analysis
Data are shown as the mean ± SD. All experiments were performed in triplicate. Statistical
significance among multiple treatment groups was determined by one-way analysis of
variance (ANOVA) followed by Dunnettʼs multiple comparison test. Studentʼs t-test
was performed for comparison between two treatment groups. The analysis was performed
using SPSS Version 21 (IBM). A difference was considered statistically significant
at p < 0.05.
Supporting information
A list of proteins and mean relative ratios, a list of PTM, target peptide sequence
of modified proteins, and mean relative ratios, ion chromatogram and mass spectra
of compounds present in St. Johnʼs wort extract, networks derived comparing proteins
expression, and three-dimensional spectra are available as Supporting Information.
Acknowledgements
This research was financially supported by the Ministry of Trade, Industry, & Energy
(MOTIE), the Korea Institute for Advancement of Technology (KIAT), the Honam Institute
for Regional Program Evaluation through the Leading Industry Development for Economic
Region, and Priority Research Centers Program through the National Research Foundation
of Korea (NRF), funded by the Ministry of Education, Science, and Technology (Grant
No. 2010–0 020 141).
