In vivo screening of traditional Chinese herbal medicines for estrogenic activities and the mechanistic studies of their effects on breast cancer cells


Student thesis: Doctoral Thesis

View graph of relations


  • Li LI

Related Research Unit(s)


Awarding Institution
Award dateApr 2015


Hormone replacement therapy (HRT) is commonly applied to relieve menopausal symptoms, but increased exposure to hormones exacerbates the risk of cancers, particularly breast cancer. Phytoestrogens are plant-derived compounds found in a wide range of traditional Chinese herbal medicines (TCHMs), which show much lower affinities to estrogen receptors (ERs) than estradiol (E2). Moreover, phytoestrogens may act as selective estrogen receptor modulators (SERMs). Thus, these compounds are considered as alternatives to HRT. However, previous studies revealed inconsistent results; as such, whether phytoestrogens can affect the risk of breast cancer development remains unclear. Therefore, TCHMs should be screened to determine estrogenic activities; the effects of these herbs on breast cancer should also be investigated. ln this study, TCHMs and their active ingredients were screened using an in vivo system; the effects of these substances on breast cancer growth were investigated in vitro.

We used transgenic medaka for estrogenic activity screening, in which green fluorescent protein (GFP) is expressed in the liver upon exposure to estrogenic substances. Among 36 herbs and 38 active ingredients, Fructus Psoraleae (Bu Gu Zhi in Chinese) and its active ingredient bakuchiol showed estrogenic activities. LC50 and EC50 of Fructus Psoraleae were 3.1 and 1.0 mg/mL, respectively. LC50 and EC50 of bakuchiol were 1.1 and 0.3 μg/mL, respectively. Two other ingredients of Fructus Psoraleae, namely, psoralen and isopsoralen, did not show estrogenic activities. Considering that GFP can be induced in a dose-dependent manner, we quantified GFP signal intensity to compare the estrogenicity of Fructus Psoraleae and bakuchiol with E2. Results suggested that the estrogenicity of Fructus Psoraleae and bakuchiol was weaker than that of E2. We also investigated the expression levels of ERα and ERβ in fish treated with Fructus Psoraleae, bakuchiol, and ethinylestradiol (EE2). As a strong synthetic estrogen, EE2 exhibited much stronger preference to ERα than to ERβ; by comparison, ERβ was preferentially activated by Fructus Psoraleae. Bakuchiol did not significantly change the expression levels of ERα and ERβ; besides, bakuchiol did not show preference to either of the two receptors.

The effects of Fructus Psoraleae and bakuchiol on breast cancer cell growth were evaluated in ER-positive breast cancer cell line MCF-7 and ER-negative cell line MDA-MB-231. Fructus Psoraleae stimulated MCF-7 cell proliferation at low concentrations and inhibited cell growth at high concentrations in a dose-dependent manner. Fructus Psoraleae inhibited MDA-MB-231 cells at IC50 of 42.8, 30.1, and 25.6 mg/mL for 24, 48, and 72 h of exposure, respectively. Fructus Psoraleae also induced S phase arrest at low concentrations and G0G1 phase arrest at high concentrations. Furthermore, this herb induced MCF-7 and MDA-MB-231 cell apoptosis dose-dependently. Considering that one herb is cornposed of various ingredients, the production condition-induced variation in herb cornponents is usually ascribed to inconsistent resuIts of studies on such herbs. Thus, we focused on the active ingredient bakuchiol in the following studies.

Bakuchiol elicited a biphasic growth effect on MCF-7 cells; this effect was similar to that of Fructus Psoraleae. The IC50 values of bakuchiol in MCF-7 cells were 12.1, 11.0, and 9.3 μg/mL for 24, 48, and 72 h, respectively. Furthermore, it inhibited MDA-MB-231 cell proliferation, with IC50 of 13.6, 10.4, and 8.9 μg/mL for 24, 48, and 72 h, respectively. In both cell lines, bakuchiol caused S phase arrest. Bakuchiol also significantly increased the mRNA levels of ataxia-telangiectasia, mutated (ATM) and p21. Although ATM and Rad3-related (ATR) upregulation patterns were observed, no significant difference was detected between control and treatment groups. Bakuchiol also upregulated the protein levels of P-Cdc2 (Tyr15) and Cyclin B1, as well as Mytl, P-Wee1 (Ser642), and p21. Pretreatment with caffeine, a well-known inhibitor of ATM/ATR, rescued S phase arrest caused by bakuchiol-induced increase in P-Cdc2 (Tyr15), Chk1, and Chk2 levels were also rescured by pretreatment with caffeine. Furthermore, sip21-induced p21 depletion marginally rescued bakuchiol-induced inhibitory effect with lower level of S phase cell accumulation in sip21 cells than in siCtrl cells. p21 knockdown increased Chkl and Chk2 expressions; bakuchiol further enhanced Chk2 expression in sip21 cells. The depletion of Cdc2 downregulated P-Cdc2 (Tyr15) but did not affect bakuchiol-induced S phase arrest. Upon 0 μg/mL to 10 μg/mL bakuchiol exposure apoptotic morphological characteristics, including more rounded, shrunk, and unattached cells, as well as more apoptotic bodies, were observed in MCF-7 cells but not in MDA-MB-231 cells. Bakuchiol induced cell apoptosis and altered mitochondrial membrane potential in MCF-7 cells but not in MDA-MB-231 cells. Upon exposure to bakuchiol, Caspase-7, Caspase-9, and poly ADP ribose polymerase (PARP) were cleaved. Furthermore, bakuchiol induces the expressions of Bak monomers and oligomers, Bax, Bim-l, and Bim-s.

Breast cancer stem cells are responsible for the development, relapse, and metastasis of mammary tumor; these stem cells are accounted for tumor resistance to current regimens. To further confirm the inhibitory effects of bakuchiol on breast cancer cells, we investigated the effects of bakuchiol on breast cancer stem cells, which were defined by CD44+/CD24-/low phenotype. We found that bakuchiol reduced the mammosphere formation in adherent MCF-7 cells. We then sorted CD44+/CD24-/low cells from MCF-7 cells and determined the direct effects of bakuchiol on this subpopulation. Our results showed that bakuchiol inhibited mammosphere formation in CD44+/CD24-/low cells with decreased size and number and loose morphological characteristic. Bakuchiol also induced S phase arrest of CD44+/CD24-/low cells at low concentrations and G0G1 phase arrest of these cells at high concentrations; this cell cycle arrest was accompanied by increased levels of cell cycle regulators, including Chk2 and p21. Bakuchiol induced apoptosis and downregulated mitochondrial membrane potential of CD44+/CD24-/low cells. The apoptosis-induction effects of bakuchiol was further approved by the upregulated expression of apoptotic regulators, including Bcl-2/adenovirus E1B 19kDa interacting protein 3 (BNIP3), death-associated protein kinase 2 (DAPK2), Cleaved Caspase 7, and PARP. Besides, the intracellular lipid content of CD44+/CD24-/low cells was upregulated by 4 μg/mL of bakuchiol and downregulated by 7 μg/mL of bakuchiol, and same trend was also observed in the mRNA expression level of fatty acid synthase (FSA), suggesting that high dose of bakuchiol may induce apoptosis through inhibition of adipogenesis. Besides, bakuchiol enhanced the reactive oxygen species (ROS)/Superoxide production of CD44+/CD24-/low cells, and reduced Notch 3 gene expression, suggesting its role in overcoming radioresistance of breast cancer stem cells. Although the mRNA level of breast cancer resistance protein (ABCG2) in CD44+/CD24-/low cells, which is related to chemoresistance in previous reports, was upregulated by bakuchiol, the intracellular doxorubicin was not decreased when co-treated with bakuchiol and doxorubicin.

Tamoxifen is used for the endocrine therapy of ER-positive breast cancer. However, tamoxifen is characterized by several limitations, including failure in treatment of ER-negative breast cancer and tamoxifen-induced menopausal symptoms; these limÏtations have prompted researchers to investigate combined treatments to eliminate adverse effects and enhance the efficacy of tamoxifen. In this study, the combined effects of tamoxifen and bakuchiol were investigated. Low-dose bakuchiol (≤4 μg/mL) negated the inhibitory effect of low tamoxifen concentration (10 -7 M to 10 -6 M) on MCF-7 cell growth; by contrast, high-dose bakuchiol (>4 μg/mL) combined with 10 -5 M tamoxifen synergistically inhibited MCF-7 cells. Furthermore, 4 μg/mL of bakuchiol combined with 10 -7 M of tamoxifen significantly induced ERα expression in MCF-7 cells, whereas 7 μg/mL of bakuchiol combined with 10 -5 M of tamoxifen significantly reduced ERα expression. In MDA-MB-231 cells, synergistic inhibitory effects of bakuchiol and tamoxifen were observed. The combined treatment of 7 μg/mL of bakuchiol and 10 -5 M of tamoxifen significantly induced cell apoptosis and decreased the mitochondrial membrane potential of ER-positive and ER-negative cells; this combined treatment also upregulated the expressions of Cleaved Caspase 7 and Cleaved PARP.

In summary, the traditional herb Fructus Psoraleae and its active ingredient bakuchiol exhibited estrogenic activities in in vivo and in vitro systems. These substances acted as ER agonist or antagonist depending on dose and tissue, suggesting their potential role as SERMs and alternatives to HRT. The inhibitory effects of bakuchiol on breast cancer cells, particularly on breast cancer stem cells, may be of great interest in clinical treatments of breast cancer. Furthermore, the combined treatment of bakuchiol and tamoxifen may provide a new regimen for breast cancer patients with tamoxifen-induced menopausal symptoms. However, the abrogation of tamoxifen-induced cancer cell death by low-dose bakuchiol in ER-positive breast cancer cells should be considered as a risk of breast cancer development.