Effects of Mangosteen Pericarp Extracts Against Mammary Cancer

The pericarp of the mangosteen fruit, Garcinia mangostana Linn, has a long history of use as a medicinal plant in Southeast Asia. The actions of mangosteen extracts have been scientifically supported and it has become a popular natural health-promoting dietary supplement. Mangosteen extracts, particularly α-mangostin, induced apoptosis via the mitochondrial pathway, cell cycle arrest by p21cip1 induction and Akt dephsophorylation in mammary cancer cells; these anti-proliferative effects are associated in vivo with suppression of tumor growth and metastasis in mouse mammary cancer models. Such preclinical evidence suggests that mangosteen extracts have potential chemotherapeutic and/or chemopreventive uses. End-Labeling; vWF: Von Willebrand Factor comprised of 78% α-mangostin and 16% γ-mangostin [8]. Mangosteen extracts induce a similar anti-tumor effect in HCT 116 human colorectal carcinoma cells in nude mice [9]. α-mangostin also significantly suppresses tumor growth associated with cell cycle arrest in a xenograft mouse model of prostate cancer [10]. We recently found the anti-tumor effects of panaxanthone, a mangosteen extract comprised of 75-85% α-mangostin and 5-15% γ-mangostin, and of purified α-mangostin in a mouse metastatic mammary cancer model carrying a p53 mutation, a model which demonstrates a metastatic spectrum similar to that seen in human breast cancers [11,12]. We further analyzed the mechanisms of anti-tumor effects induced by α-mangostin in vitro, using both human and mouse metastatic mammary carcinoma cells, and in vivo in an established mouse mammary cancer model [12,13]. Here, we summarize the results of these experiments. Citation: Shibata MA, Matoba Y, Tosa H, Iinuma M (2013) Effects of Mangosteen Pericarp Extracts Against Mammary Cancer. Altern Integ Med 2: 139. doi:10.4172/2327-5162.1000139


Introduction
The mangosteen (Garcinia mangostana Linn) has been dubbed the "queen of fruit" in its native Thailand. Mangosteens are round fruits roughly 4-8 cm in size with a thick, brittle, deep purple spherical outer pericarp; the edible snow white endocarp consists of 4 to 8-segmented wedge-shaped arils ( Figure 1A). When damaged, the pericarp secretes a yellow substance ( Figure 1B) that acts to protect the fruit from bacterial infection. The secretion also apparently functions as an insect repellant.
Mangosteen pericarp has a long history of medicinal use in both Chinese and Ayurvedic medicine. For centuries, people in Southeast Asia have used dried mangosteen pericarp as an antiseptic, an antiinflammatory, an anti-parasitic, an antipyretic, an analgesic, and as a treatment for skin rashes [1]. The yellow exudate from the mangosteen pericarp contains a class of compounds called xanthones, including α-mangostin, β-mangostin, γ-mangostin, garcinone B, and garcinone E, along with mangostinone, tanins, and a flavonoid called epicatechin [1]. The chemical structures of α-mangostin, β-mangostin, γ-mangostin are shown in Figure 1C. Mangosteen products have become popular natural dietary health supplements and the effects of mangosteen extracts have had recent scientific support. Within the arena of anticarcinogenic natural products, the characteristic anti-proliferative effects of α-mangostin, an extract of the mangosteen pericarp, has received a great deal of attention. It is shown to induce cell cycle arrest and apoptosis in various types of human cancer cells [2][3][4][5], as well as to inhibit cell invasion and migration in mammary and prostate cancer cells. α-Mangostin is further associated with down-regulation of MMP-2 and MMP-9 [6,7].
Several animal cancer models demonstrate the anti-tumor effects of mangosteen extracts. Formation of aberrant crypt foci, a putative preneoplastic lesion in rat colon carcinogenesis, is significantly suppressed by dietary administration of a crude mangosteen derivative

Cell line
The BJMC3879 cell line is derived from a metastatic focus within a lymph node of a BALB/c mouse originally inoculated with purified MMTV virus; the virus was obtained from media in which Jyg-MC cells, established from mammary tumors of the Chinese wild mouse, were previously grown [14]. Mammary tumors developing in syngeneic mice injected with BJMC3879 cells show a high metastatic propensity, especially to lymph nodes and lungs [15,16]. This cell line contains a p53 mutation [17].

Abstract
The pericarp of the mangosteen fruit, Garcinia mangostana Linn, has a long history of use as a medicinal plant in Southeast Asia. The actions of mangosteen extracts have been scientifically supported and it has become a popular natural health-promoting dietary supplement. Mangosteen extracts, particularly α-mangostin, induced apoptosis via the mitochondrial pathway, cell cycle arrest by p21 cip1 induction and Akt dephsophorylation in mammary cancer cells; these anti-proliferative effects are associated in vivo with suppression of tumor growth and metastasis in mouse mammary cancer models. Such preclinical evidence suggests that mangosteen extracts have potential chemotherapeutic and/or chemopreventive uses.
We recently found the anti-tumor effects of panaxanthone, a mangosteen extract comprised of 75-85% α-mangostin and 5-15% γ-mangostin, and of purified α-mangostin in a mouse metastatic mammary cancer model carrying a p53 mutation, a model which demonstrates a metastatic spectrum similar to that seen in human breast cancers [11,12]. We further analyzed the mechanisms of anti-tumor effects induced by α-mangostin in vitro, using both human and mouse metastatic mammary carcinoma cells, and in vivo in an established mouse mammary cancer model [12,13]. Here, we summarize the results of these experiments.

Dietary administration of panaxanthone
We injected BJMC3879 cells subcutaneously into the inguinal region of female BALB/c mice. Two weeks post-inoculation, mice were fed powder diet containing 0, 0.25, or 0.5% of panaxanthone, mammary tumor weekly and calculated tumor volumes using the formula 'maximum diameter×(minimum diameter) 2 ×0.4' [18]. After 8 weeks of treatment, we euthanized the animals and the mammary tumors, selected lymph nodes (i.e. lymph nodes from axillary and femoral regions as well as any that appeared abnormal) and lungs were removed and portions of each immediately fixed in 10% phosphatebuffered formalin. Fixed tissue sections were processed through to paraffin embedding, cut at 4 µm and stained with hematoxylin and eosin (H&E) for histopathological examination or mounted unstained for immunohistochemistry.

Survival rates, tumor growth and metastasis
No treatment-related prolongation of the survival rates was observed in this study. The induced mammary tumors were uniformly moderately differentiated adenocarcinomas. As assessed by computed volume, primary tumor growth was significantly suppressed in mice fed panaxanthone at 0.25% and 0.5% from weeks 3 and 2, respectively, when compared with the control group. The numbers of metastatic lung foci >250 µm per mouse were significantly reduced in both the 0.25% and 0.5% groups as compared with the control animals, and the number of metastatic lymph nodes per mouse tended to decrease in a dose-dependent manner; however, statistical significances were not observed because of the large variations among control animals.

Blood microvessel density in mammary tumors
Unstained sections from formalin-fixed paraffin-embedded (FFPE) tumors were immunohistochemically stained with von Willebrand factor (vWF). The blood microvessel density in tumors, as inferred by the number of immunopositive blood microvessels counted, was significantly decreased in mice fed 0.5% panaxanthone when compared to that in controls.

Cell proliferation and apoptosis
Unstained tumor sections from FFPE tissues were also immunohistochemically stained with proliferating cell nuclear antigen (PCNA) and terminal deoxynucleotidyl transferase-mediated dUTP-FITC nick end-labeling (TUNEL). Tumor cell proliferation, assessed by positive PCNA staining, was significantly suppressed in mice fed panaxanthone at 0.25% and 0.5% when compared with the control group, while cellular apoptosis, assessed as the number of TUNELpositive cells, was significantly elevated in mice fed 0.5% panaxanthone as compared with tumors in the control group.

Cell line
In this study, we used the BJMC3879Luc2 mammary carcinoma cell line [19] which is generated by stable transfection of the luc2 gene (an improved firefly luciferase gene) into the parent BJMC3879 cell line described in the previous experiment. The BJMC3879Luc2 cell line is bioluminescent and allows in vivo imaging for staging and quantifying expansion of metastasis [19]. The BJMC3879Luc2 cell line also contains the p53 mutation of the parent line [12,20,21].

Osmotic-pump administration of purified α-mangostin
Since it is difficult and expensive to obtain large amounts of purified α-mangostin for dietary feeding, the compound was continuously (Alzet model 2002) at either10 or 20 mg/kg/day. BJMC3879Luc2 cells were subcutaneously implanted as previously into the inguinal region of female BALB/c mice. Three weeks later, when tumors had reached approximately 0.4-0.6 cm in diameter, mice was implanted with miniosmotic pumps calibrated to release 0, 10 or 20 mg/kg/day α-mangostin for 6 weeks. One hour prior to euthanasia at study termination, mice were injected intraperitoneally with 50 mg/kg 5-bromo-2'deoxyuridine (BrdU) as a means to quantify the degree of tumor cell proliferation through DNA synthesis. Measurements of tumor volume and histopathology examination were conducted in the same manner as in the previously described feeding study.

Survival rates and tumor growth
Survival rates were significantly higher in the 20 mg/kg/day group when compared to the control group. Tumor growth was significantly inhibited in the 20 mg/kg/day group from week 1 to 5 as compared tumor growth in control mice.

Bioluminescence imaging and metastasis
Bioluminescence imaging provides an indication of metastatic growth based on the quantity and relative signal strength of tissues expressing luciferase. Animals receiving 20 mg/kg/day α-mangostin tended to show decreased metastatic expansion overall; in particular, the number of lymph node metastases per mouse was significantly decreased in the 20 mg/kg/day group as compared to the control group (Figures 2A and 2B). In the lungs, the number of metastatic foci per mouse >1 mm tended to be lower in the 20 mg/kg/day group as compared to the control group.

Lymphatic vessel invasion
We immunohistochemically stained FFPE mammary tumor sections with podoplanin, a lymphatic endothelial marker. The number of dilated lymphatic vessels having intraluminal tumor cells (a measure of lymphatic vessel invasion) was significantly decreased in the group receiving α-mangostin at 20 mg/kg/day as compared to those of control mice ( Figures 2C and 2D).

Microvascularization density in mammary tumors
Using the endothelial cell marker CD31, immunohistochemical analysis of vascular microvessel density revealed significantly lower microvascularization in mammary tumors exposed to 10 and 20 mg/ kg/day α-mangostin as compared to untreated tumors.

Cell proliferation and apoptosis in mammary tumors
BrdU labeling indices indicated a trend to lower cell proliferation rates in mammary carcinomas exposed to 20 mg/kg α-mangostin, but this decrease was not statistically significant. However, apoptosis of tumor cells, as assessed by TUNEL staining, was significantly increased in tumors from mice treated with 20 mg/kg/day vs. control ( Figures 2E  and 2F).

Mechanistic analyses
Apoptosis: Activities of caspase-3, caspase-8, and caspase-9 were significantly increased in BJMC3879Luc2 cells treated with 12 µM α-mangostin. Furthermore, cytosolic cytochrome c was significantly elevated in α-mangostin-treated cells. This strongly suggests that α-mangostin induces apoptosis via the mitochondria-mediated pathway. However, since no Bid cleavage was seen, the mitochondriamediated apoptosis induced by α-manogstin may not be through the caspase-8-Bid pathway. The results in in vivo experiments were similar in that we observed higher expression of the active forms of caspase-3 and caspase-9 in mice mammary carcinomas exposed to α-mangostin that α-mangostin induces the same mechanism in vivo as well as in vitro.
Cell cycle analysis: As measured by flow cytometry, we found an increased number of BJMC3879Luc2 cells in the G1-phase and a decrease of cells in the S-phase after treatment with 12 µM α-mangostin.
Akt phosphorylation: Akt phosphorylation contributes not only to cell proliferation but also to anti-apoptotic cell death, cell cycle entry, angiogenesis and metastasis-all important aspects of the oncogenic process [22]. We found that treatment with α-mangostin significantly decreased phospho-Akt expression in both mammary carcinoma cells in vitro and in mammary carcinoma tissues in vivo. Given this correlation, the phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) pathway, which is activated in many solid cancer patients, may be a target in cancer therapeutics [23].

Analysis of Apoptosis and Cell Cycle Arrest in Human Breast Carcinoma Cells Treated with Purified α-Mangostin
We used MDA-MB231 human breast cancer cells for these analyses, comparing naïve cultures to those treated with 20 µM α-mangostin. Exposure to α-mangostin increased apoptosis as evaluated by ssDNA levels and increased levels of caspase-3, caspase-8 and caspase-9; along with significant cytochrome c release, this would indicates α-mangostin-induced apoptosis to be mitochondria-mediated and in agreement with our previous studies [11,12]. α-mangostin also induced G1-phase arrest and decreased PCNA levels indicated S-phase suppression. Cell cycle arrest was associated with upregulation of p21 cip1 , a cyclin-dependent kinase inhibitor, and check-point protein 2 (Chk2), a cell cycle checkpoint regulator.

Discussion
via the mini-osmotic pumps as compared to control tumors, suggesting Both panaxanthone, a crude mixture consisting of 75-85% α-mangostin and 5-15% γ-mangostin, and purified α-mangostin exerted anti-tumor effects in immunocompetent mice inoculated with a mammary cancer xenograft containing a p53 mutation that induces a metastatic spectrum similar to that seen in human breast cancer [11,12]. Other animal cancer models also demonstrate the anti-tumor effects of mangosteen extracts. For example, α-mangostin inhibited 7,12-dimethylbenz[α]anthracene (DMBA)-induced preneoplastic lesions in a mouse mammary organ culture assay [24], while dietary administration of crude α-mangostin significantly inhibited the development of preneoplastic aberrant crypt foci in rat colon carcinogenesis and is associated with decreased expression of PCNA and increased β-catenin accumulated in the crypts [8]. In addition, crude α-mangostin in the diet significantly suppressed colon tumor growth in a nude mouse xenograft model using HCT 116 human colorectal carcinoma cells [9]. Furthermore, crude α-mangostin demonstrated decreases in cell migration, cell invasion and clonogenicity, as well as induction of apoptosis via mitochondrial pathway in in vitro studies. In prostate cancer, treatment with α-mangostin by oral gavage (100 mg/kg) significantly suppressed xenograft tumor growth induced with human prostate carcinoma 22Rv1 cells in nude mice; this growth suppression is associated with cell-cycle arrest at G1-phase and apoptosis, possibly due to inhibition of cyclins/cyclin-dependent kinase 4 (Cdk4), a critical component of the G1-phase. In two different skin cancer cell lines (human melanoma SK-MEL-28 cells and human squamous cell carcinoma A431 cells), α-mangostin inhibited motility, adhesion, migration and invasion in vitro study, showing anti-metastatic ability [25].
Both α-and γ-mangostin inhibit aromatase in a dose-dependent manner, so that this is another possible mechanism by which α-mangostin exerted the anti-tumorigenic effects seen in our experiments. Estrogen and the estrogen receptor α (ERα) are widely accepted as playing a crucial role in the development and progression of hormone-dependent breast cancer. Aromatase is an estrogen synthase responsible for catalyzing the biosynthesis of estrogens from androgens, making it another attractive therapeutic target in treating breast cancer. Currently, use of third-generation aromatase inhibitors are common as standard adjuvant therapy for postmenopausal breast cancer patients having hormone-receptor positive disease and are associated with improved disease-free survival compared with the previous standard tamoxifen therapy [28,29]. Figure 3: Possible mechanistic pathway of α-mangostin's antitumorigenic effects. Akt phosphorylation contributes not only to cell proliferation but also to anti-apoptotic cell death, cell cycle entry, angiogenesis and metastasis-all important aspects of the oncogenic process. Since Akt exist up-stream of these processes, α-mangostin-induced dephosphorylation may be one possible mechanism for its anti-tumor action.

Conclusion
We similarly observed G1-phase arrest and S-phase suppression in the cell cycle and mitochondria-mediated apoptosis in mammary carcinoma cells treated with α-mangostin [12,13]. Furthermore, it has shown that transcription levels of p21 Cip1 and Chk2 are up-regulated in α-mangostin-treated mammary carcinoma cells [13]. The likely anti-tumorigenic mechanisms induced by α-mangostin are illustrated in Figure 3. p21 cip1 and Chk2 act as a break at G1-phase of cell cycle and activation of the Cdk inhibitor p21 Cip1 leads to the inhibition of the Cyclin E/A-Cdk2 complex, which causes cell-cycle arrest at the G1/S boundary. Chk2 kinase is further responsible for inhibitory phosphorylation of Cdc25A, a phosphatase required for activation of Cdks2/4/6, resulting in cell-cycle arrest.
α-mangostin also inhibits Akt phosphorylation in mammary carcinoma cells both in vitro and in vivo [12] and similarly decreases phosphorylation of Akt as well as MAPK in DLD1colorectal cancer cells [5]. Akt is considered to be the key downstream effector of PI3K oncogenic signaling. Given the crucial role of PI3K/Akt signaling in regulating processes such as cell growth, proliferation, migration, and survival and in angiogenesis and anti-apoptosis [22], it is not surprising that components of this pathway are frequently dysregulated in cancer, making the Akt kinase family members important therapeutic targets. This pathway is frequently dysregulated in human cancer and modulates many of the characteristic processes of tumorigenesis [26], including cell growth, proliferation, survival, migration, metabolism, and tumor angiogenesis, offering another pivotal target for cancer therapy [27]. In fact, our experiments demonstrated that treatment with α-mangostin showed Akt dephosphorylation [12], inhibition of tumor angiogenesis and induction of apoptosis [11,12]. Since Akt exists up-stream of these processes, α-mangostin-induced dephosphorylation may be one possible mechanism for its anti-tumor action (Figure 3).
Our studies provide correlating evidence for mangosteen extracts as potential chemopreventive or anti-cancer agents. Of particular significance is the in vivo reduction of lymph node metastasis by α-mangostin; since metastasis is the most important prognostic factor in breast cancer patients, this anti-metastatic action of certain fractions of mangosteen suggests specific clinical applications, at least in breast cancer carrying p53 mutation. Mangosteen extracts appear to, in fact, have chemopreventive qualities and might prove useful as adjunctive and/or complementary alternative treatments in human breast cancer.