Polyphenol-rich Avicennia marina leaf extracts induce apoptosis in human breast and liver cancer cells and in a nude mouse xenograft model

Avicennia marina is the most abundant and common mangrove species and has been used as a traditional medicine for skin diseases, rheumatism, ulcers, and smallpox. However, its anticancer activities and polyphenol contents remain poorly characterized. Thus, here we investigated anticancer activities of secondary A. marina metabolites that were purified by sequential soxhlet extraction in water, ethanol, methanol, and ethyl acetate (EtOAc). Experiments were performed in three human breast cancer cell lines (AU565, MDA-MB-231, and BT483), two human liver cancer cell lines (HepG2 and Huh7), and one normal cell line (NIH3T3). The chemotherapeutic potential of A. marina extracts was evaluated in a xenograft mouse model. The present data show that EtOAc extracts of A. marina leaves have the highest phenolic and flavonoid contents and anticancer activities and, following column chromatography, the EtOAc fractions F2-5, F3-2-9, and F3-2-10 showed higher cytotoxic effects than the other fractions. 1H-NMR and 13C-NMR profiles indicated that the F3-2-10 fraction contained avicennones D and E. EtOAc extracts of A. marina leaves also suppressed xenograft MDA-MB-231 tumor growth in nude mice, suggesting that EtOAc extracts of A. marina leaves may provide a useful treatment for breast cancer.


IntroductIon
Cancer causes a wide variety of diseases and is characterized by uncontrolled growth and spread of abnormal cells and subsequent disruption of organ and tissue functions. Treatments for cancer usually include surgery followed by radiotherapies and chemotherapies [1] that inhibit cancer cell growth and proliferation. However, in contrast with localized surgery and radiotherapy, chemotherapy has systemic effects [2][3][4]. Among chemotherapeutic agents, phytochemicals, such as paclitaxel from the Pacific yew tree and vincristine from Vinca rosea Linn, are widely used to treat cancers and have demonstrated pro-apoptotic effects [5,6]. Various chemopreventive and chemotherapeutic agents have been shown to induce apoptotic cell death in both in vitro and in vivo studies, suggesting that apoptosis plays a crucial role in cancer treatment [7]. Accordingly, the widely used chemotherapeutic drug 5-fluorouracil (5-FU) inhibits tumor cell growth in animal models by inducing apoptotic activation of the CD95/CD95L system [8]. The chemopreventive agent curcumin predominantly induces apoptosis via mitochondria-mediated pathways in various cancer cell types [9]. Apoptosis, or programmed cell death, is a physiological process that eliminates abnormal, misplaced, or nonfunctional cells and is critical for maintenance of tissue homeostasis [10]. Excessive apoptosis causes organ atrophy and dysfunction, whereas failure of apoptosis results in accumulation of abnormal cells, potentially leading to tumor development. Apoptosis is controlled at multiple molecular levels and involves pro-and anti-apoptotic members of the Bcl-2

Research Paper
protein family [11]. Many studies have shown that dietary phytochemicals induce apoptosis in cancer cells, suggesting potential for development as cancer therapeutic agents [12].
Mangrove forests are economically and ecologically important and are rich in medicinal and non-medicinal edible plants. In particular, mangroves produce a wide variety of structurally novel natural agents with biochemical profiles [13]. Avicennia marina is a mangrove species of the Acanthaceae family, and discoveries of its chemical compounds have received much attention [14]. A. marina has been used as a traditional medicine for the treatment of skin diseases, rheumatism, ulcers, and smallpox. In vitro antimalarial, antibacterial, analgesic, and cytotoxic activities of A. marina have been reported [15]. Hence, A. marina is considered a valuable source of chemical constituents with medicinal potential. Among these, luteolin 7-O-methylether, chrysoeriol 7-O-glucoside, isorhamnetin 3-O-rutinoside, tannin lapachol, and naphthoquinone analogs have been isolated, and their biological activities have been described [16]. However, few studies demonstrate in vitro and in vitro anti-cancer activities of A. marina, and its chemical constituents have not been extensively examined.
Here we evaluated the biomedical potential of A. marina plants from a mangrove forest and established the chemical composition of A. marina extracts. In particular, total phytopolyphenol contents were separated using chromatography, and A. marina fractions were evaluated for anti-cancer effects in in vitro and in vivo models.

Polyphenol contents and anticancer activities against breast and liver cancer cell lines
Plant extracts that are rich in polyphenols have been safely used as traditional Chinese medicines for many centuries. Thus, water (H 2 O), ethanol (EtOH), methanol (MeOH), and ethyl acetate (EtOAc) extracts of A. marina were evaluated for phenol and flavonoid contents. As shown in Table 1A (Table 1B). The absence of detectable As, Cu, Cd, Cr, and Ag in the present A. marina leaves was considered favorable for clinical application without toxicity.
To determine whether high phenol and flavonoid contents were associated with anticancer activities, cytotoxic effects of H 2 O, EtOH, MeOH, and EtOAc extracts of A. marina leaves were compared using MTT assays in normal NIH3T3 cells, and in breast (AU565, MDA-MB-231 and BT483) and liver (HepG2 and Huh7) cancer cell lines ( Figure 1B). In these experiments, EtOH and EtOAc extracts of A. marina leaves inhibited cell growth in cancer cell lines more than in normal NIH 3T3 cells. Inhibition of cancer cell growth was greater with EtOAc extracts than with EtOH and MeOH extracts, but was dose-dependent in all cases, and similar observations were made after treatments with EtOAc extracts of A. marina seeds (data not shown). However, whereas growth inhibition by H 2 O extracts of A. marina leaves was significant at 1000 μg/mL, H 2 O extracts had no effect at 200 μg/ mL (data not shown). These data suggest that EtOAc extracts of A. marina leaves have the highest polyphenol contents and anticancer activities. Accordingly, EtOAc extracts did not inhibit proliferation of NIH 3T3 cells at 40-80-μg/ mL, but significantly inhibited cell proliferation in AU565, BT483, HepG2, and Huh7 cancer cells. Thus, subsequent analyses were performed after treatments with EtOAc extracts at 40-80 μg/mL.
Soft agar colony formation assays can be used to separate cancer cells and to confirm phenotypes, and were used to determine the effects of EtOAc extracts of A. marina leaves on anchorage-independent growth in breast (AU565 and BT483) and liver (HepG2 and Huh7) cancer cells. After 14-21 days of treatment with the EtOAc extract, cultured cells showed reduced soft agar colony formation, with decreased colony sizes and a smaller numbers of AU565, BT483, HepG2, and Huh7 cancer cell colonies ( Figure 2).

Identification and separation of EtOAc extracts of A. marina leaves
To identify bioactive components of A. marina leaves, EtOAc extracts were purified using column chromatography ( Figure 4A). The plant material (20 kg) of A. marina was immersed in 200 L distilled water for 30 min and then boiled at 95℃ for 30 min. The dried extract was partitioned between H 2 O and EtOAc to yield 35.69 g of a dried EtOAc extract and an aqueous residue. The EtOAc extract was subjected to silica gel medium pressure column chromatography (gradient mixtures, from 100% CH 2 Cl 2 to 100% MeOH). The eluents were pooled to form 6 fractions on the basis of TLC analysis. Fraction 3 (8.29 g) was further purified on a silica gel column chromatography (hexane-EtOAc-Acetone, 10:3:3) to give fractions F3-1 to F3-6. Fraction F3-2 (224.7 mg) was subjected to Sephadex LH-20 column chromatography (MeOH) to give fractions F3-2-1 to F3-2-12.
The bioactive fraction F3-2-10 further purified by HPLC to yield 1 and 2 (4.9 mg). The chemical structures of compounds 1 and 2 were identified by NMR spectroscopy. Compounds 1 and 2, which were chromatographically inseparable could be obtained only as a mixture (1:1, according to 1 H & 13 C NMR data). The identical molecular formula of 1 and 2 was established as C 12 H 6 O 4 by ESIMS at m/z 213.0 [M-H]and NMR data. A naphthoquinone skeleton was suggested by the absorption bands of UV spectrum at l max 267, 307, and 350 nm. One set of 1 H NMR signals of compound 1 indicated three aromatic protons (d H 8.01, 7.47, 7.10) in an AMX spin system and two AB-type olefinic protons (d H 7.96, d, J = 1.9Hz, d H 6.99, d, J = 1.9 Hz) of a furan ring (Table 3).
In the 13 C NMR spectrum, two sets of signals, each representing 12 carbons, could be distinguished. The NMR data of the mixture of compounds 1 and 2 indicated them to carry a hydroxyl substituent on the aromatic ring. Based on above data and with those reported in the literature, 1 was determined to be 6-hydroxynaphtho [2, 3-b] furan-4, 9-dione and 2 as 7-hydroxynaphtho [2, 3-b] furan-4, 9-dione, and these compounds were named avicennone D and avicennone E, respectively. In contrast with previously reported 1 H-NMR and 13 C-NMR data, the F3-2-10 fraction was identified as a mixture of avicennone D and E (Table 3). To our knowledge, this is the first study to demonstrate apoptotic mechanisms of avicennone D and E from a HPLC fraction from an EtAOc extract of A. marina leaves in breast and liver cancer cells. Major HPLC peaks from F2-5, F3-2-9, and F3-2-10 showed retention times of 36-66 minutes, and corresponded with luteolin, quercetin apigenin, kaempferol, and chrysin standards ( Figure 3B and Table 2B). Thus, the anticancer properties of these polyphenols were compared with those of F2-5 and F3-2-9 in AU565 and BT483 cells. In these experiments, anticancer activities were ranked in the same decreasing order of F2-5 > F3-2-9 > luteolin > quercetin, apigenin, kaempferol, and chrysin in both cell lines ( Figure 4C). In comparisons with paclitaxel, tamoxifen, doxorubicin, and rapamycin, antiproliferative activities decreased in the order paclitaxel > doxorubicin > F2-5 > F3-2-9 or rapamycin > tamoxifen in both cell lines. Similar results were also observed in BT483 cells ( Figure 4D). Taken together, these data indicate the efficacy of F2-5 and F3-2-9 as treatments for breast cancer.
Flow cytometry and DNA fragmentation data indicated that treatments with EtOAc extracts trigger   Figure 6D). In further investigations of caspasedependent mechanisms ( Figure 6E), pretreatment of AU565 cells with the caspase inhibitor Z-VAD-FMK blocked EtOAc extract-mediated apoptosis, but did not block apoptosis in BT483 cells. These results suggest roles of both caspase-dependent and -independent mechanisms in BT483 cells.

Anti-motility activities of etoAc extracts of A. marina leaves in MdA-Mb-231 cells
The effects of EtOAc extracts of A. marina leaves on cell motility were examined using wound-healing assays. In these experiments, significant cytotoxic activity was observed in MDA-MB-231 cells after treatment with 80-μg/mL EtOAc extracts for 48 h. To decrease interference in assessments of motility, wound-healing assays were performed after 24-48 h exposure to 10-40 μg/mL EtOAc extracts ( Figure 7A). These assays showed inhibition of motility but not growth in MDA-MB-231 cells. Because cell motility has been associated with expression of specific cell adhesion proteins such as MMP2, MMP9, and EMT-related proteins, expression levels of these and fibronectin, snail protein, slug protein, vimentin, LRP6, and cyclin B were determined in MDA-MB-231 cells. In these experiments, treatments with 10-20-μg/mL EtOAc extracts decreased MMP2, MMP9, cyclin B, vimentin, and snail protein expression in MDA-MB-231 cells after 24 h ( Figure 7B).

etoAc and etoH extracts, and F2-5 from A. marina leaves inhibited growth of MdAMb-231 xenograft tumors in nude mice
To determine whether A. marina extracts inhibit breast cancer cell growth in vivo, human MDAMB-231 breast cancer cells were xeno-transplanted into the back skin of BALB/c nude mice and the effects of EtOAc extracts and EtOH extracts of A. marina leaves were investigated (Figure 8). In these animals, mean tumor volumes and weights were decreased by treatments with EtOAc and EtOH extracts at 200 mg/kg/day compared with DMSO-treated controls on day 23. However, no significant changes in relative weights of hearts, livers, spleens, or kidneys were observed in comparisons of EtOAc and EtOH extract-treated animals and the DMSO-treated control group. Finally, no differences in serum aspartate aminotransferase (AST) and creatinine values were found between treatment and control groups (data not shown), indicating limited side effects of A. marina extracts.

dIscussIon
Mangroves grow under environmental conditions of high visible and UV radiation [17]. Both visible and UV radiation have been shown to be harmful to the photosynthetic apparatus in plants. Accordingly, higher plants adapt to high irradiance by accumulating UV-Babsorbing phenolic compounds such as flavonoids. Such flavonoids and phenolic acids also have antioxidant properties and health benefits in mammals, and flavonoids are the most abundant polyphenols in plant foods.    Previous studies show that higher intakes of flavonoidrich foods are associated with lower risks of various diseases, including cardiovascular disease and cancer [18]. A. marina is the most abundant and dominant mangrove species and has a long history of use as a traditional medicine for the treatment for skin diseases, rheumatism, ulcers, and smallpox. However, few previous studies have characterized the relationship between anticancer activities and polyphenol contents. The present study shows anticancer activities of four different extracts of A. marina in three human breast cancer cell lines, two liver human cancer cell lines, and in one normal cell line, and in a mouse xenograft model. Taken together, these data indicate chemotherapeutic potential of A. marina extracts and fractions, and the ensuing anticancer efficacies were correlated with total phenol contents. The present findings also show that treatment with A. marina leaf extracts and fractions leads to activation of PARP, caspase 3, and caspase 8 in AU565, BT483, and HepG2 cancer cell lines. EtOAc and EtOH extracts inhibited MDA-MB-231 Mangroves produce a wide variety of secondary metabolites with demonstrated anticancer properties, including alkaloids, triterpenoids, flavonoids, and tannins.
Here EtOAc extracts of A. marina leaves had the highest phenol and flavonoid contents, and also had the best anticancer activities. The present F2-5, F3-2-9, and F3-2-10 fractions showed higher cytotoxic effects than the raw EtOAc extracts. In particular, F2-5 and F3-2-9 fractions exhibited strong inhibitory effects on AU565 cells, with IC 50 values of 0.75 μg/mL and 2.1 μg/mL, respectively.
Because the criteria of the American National Cancer Institute consider IC 50 values of < 30 μg/mL indicative of a promising extract [19], F2-5 and F3-2-9 can be considered as potential sources of anticancer drugs. However, whether EtOAc extracts of A. marina leaves contain other secondary metabolites, such as alkaloids, triterpenoids, and tannins, requires confirmation in further studies.
The present HPLC data showed that F3-2-9 contains two major peaks with retention times close to those of luteolin and quercetin standards. Luteolin is present in a fraction from a MeOH extract of A. marina that had high anticancer activity [20]. However, the present EtOAc extracts had higher cytotoxic effects than MeOH extracts. Hence, further structural comparisons of these fractions are warranted to elucidate anticancer mechanisms. Here reversed-phased HPLC of the F3-2-10 fraction yielded avicennones D and E, which were previously isolated from in 2007 by Li et al. [21]. However, corresponding anticancer activities of these compounds remain uncharacterized. Hence, the present data are the first to indicate apoptotic mechanisms of avicennone D and E in HPLC fractions from EtOAc extracts of A. marina leaves in breast and liver cancer cells.
In summary, EtOAc extracts of A. marina leaves induce apoptosis and inhibit migration of breast and liver cancer cells. These results warrant further investigation of A. marina leaf extracts to identify candidate drugs for the treatment of breast and liver cancers, either alone or in combination with other drugs.

Preparation of A. marina extracts and ethyl acetate (etoAc) fractions
Raw A. marina materials were collected from the Xinfeng mangrove conservation area in Taiwan. To prepare water (H 2 O) extracts, 100 g of dried A. marina leaves or seeds were immersed in 1000 mL of distilled water for 30 min and were then boiled at 95°C for 30 min. After cooling, the mixture was filtered through a filter paper and water extracts were obtained after evaporation to dryness in a vacuum rotary evaporator at 80°C. To prepare methanol (MeOH) and ethanol (EtOH) extracts, 100 g of A. marina dried leaves or seeds were immersed in 1000 mL of MeOH or EtOH for 24 h at 4°C and were then filtered using filter papers. Filtrates were then evaporated under vacuum at 55°C. To prepare ethyl acetate (EtOAc) extracts, water extracts were reconstituted in distilled water and were then partitioned sequentially with equal volumes of EtOAc as indicated in Figure 1A. EtOAc soluble fractions were then fractionated using various column chromatography protocols as indicated in Figure 4A. Tested concentrations were expressed according to dry weights of extracts as indicated in Table 1. Furthermore, the chemical structures were identified by NMR spectroscopy.

cell viability assays
Cells were treated with A. marina extracts for 48 h and cell viability was determined using MTT [3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] assays. Briefly, MTT-formazan crystals were formed in metabolically viable cells and were dissolved in 500 μl of DMSO for spectrophotometric determinations at a wavelength of 550 nm.

Determinations of total phenol and flavonoid contents
Total phenol contents were determined spectrophotometrically using gallic acid as a standard according to a previously described Folin-Ciocalteu method [22]. Total flavonoid contents were also determined spectrophotometrically using quercetin as a standard, according to a previously reported method [23].

determination of trace metal elements in A. marina leaves using inductively coupled plasma atomic emission spectroscopy (IcP-Aes)
ICP-AES was used for rapid, precise, and accurate determinations of trace metal elements in digested A. marina leaves. Briefly, 0.5-g leaf extracts were digested by heating in a sealed teflon microwave digestion vessel (MARS Xpress TFM high pressure reaction vessel) containing 10 mL of concentrated nitric acid. Teflon vessels were then incubated at 180°C for 10 min and after cooling to the room temperature, contents of vessels were thoroughly transferred into 50 mL polypropylene test tubes and were diluted to 10 mL in deionized water. Sample solutions were then filtered and concentrations of trace metal elements were measured directly using ICP-AES (PerkinElmer ICP Optima 21000bv). www.impactjournals.com/oncotarget

soft agar colony formation assays
Equal volumes of single-cell suspensions were treated with EtOAc extracts of A. marina leaves in DMEM containing 20% fetal bovine serum (FBS), and were then mixed with 0.7% agarose. Mixtures were plated on 6-cm culture dishes on top of a base layer of 0.7% agarose containing 10% FBS, and were allowed to gel. After 14-21-days incubation, colonies of > 60 μm were counted using a light microscope.

Flow cytometry
Flow cytometric analyses were performed as previously described [24], and fluorescent propidium idodide (PI)-DNA complexes were quantified using FAC-Scan cytometry.

trypan blue exclusion assays
Trypan blue dye exclusion assays were used to determine cytotoxic effects of A. marina leaf extracts in AU565, BT483, and HepG2 cells. Briefly, cells were seeded in six-well plates at a density of 5 × 10 4 cells in DMEM containing 10% FBS, and were then incubated for 24 h at 37°C. Culture medium was replaced and cells were incubated with or without the indicated concentrations of A. marina leaf extracts for 24-72 h in DMEM containing 2% FBS. Cells were then harvested by trypsinization and were mixed with trypan blue solution, and dead stained cells were counted using a hemocytometer under a microscope.

dnA fragmentation assays
DNA fragmentation assays were performed as previously described [19]. Briefly, cells were incubated with reagents at various concentrations for 48 h and were then lysed using lysis buffer. DNA was then extracted using phenol/chloroform/isoamyl alcohol and fragments were visualized using 2% agarose gel electrophoresis.

Analyses of nuclear morphology
Cells were plated on coverslips in six-well plates. After treatment with A. marina leaf extracts for 48 h, cells were fixed with 4% formaldehyde for 20 min and were then incubated in 4-mg/mL Hoechst 33258 for 30 min. Coverslips were washed and mounted in Vectashield (Vector Laboratories, Burlingame, CA) and nuclear chromatin morphology was viewed under a microscope.

Western blot analyses
Western blot analyses were performed as previously described [19], and immunoblots were visualized using enhanced chemiluminescent reagent.

Animals and treatments
Female BALB/c nude mice (6-8 weeks old) were purchased from the National Laboratory Animal Breeding and Research Center (Taipei, Taiwan). A total of 2 × 10 6 MDA-MB-231 cells were subcutaneously implanted into the legs of female nude mice. Mice were then randomly divided into three groups (four mice per group) for intraperitoneal treatments with DMSO, EtOH (200 mg/ kg/ day), and EtOAc extracts (200 mg/kg/day) for 23 days. Tumor volumes were calculated using the following formula: Volume (mm 3 ) = A × B 2 /2, where A is the longest diameter (mm) and B is the shortest diameter (mm).

statistical analyses
All data are expressed as means ± standard errors of the mean (SE) of at least three separate experiments for each group. Differences were identified using Student's t-tests and were considered significant when *p < 0.05, **p < 0.01, or ***p < 0.001.