Cytotoxic Triterpenes and Sterols from Pipturus arborescens ( Link )

1 Biology Department, De La Salle University Science & Technology Complex, Leandro V. Locsin Campus, Biñan City, Laguna 4024, Philippines. 2 Biology Department, De La Salle University, 2401 Taft Avenue, Manila 0922, Philippines. 3 Center for Natural Science and Ecological Research, De La Salle University, 2401 Taft Avenue, Manila 0922, Philippines. 4 Chemistry Department, De La Salle University, 2401 Taft Avenue, Manila 0922, Philippines. 5 National Research Institute of Chinese Medicine, Ministry of Health and Welfare, 155-1, Li-Nong St., Sec. 2, Taipei 112, Taiwan. 6 Chemistry Department, De La Salle University Science & Technology Complex, Leandro V. Locsin Campus, Biñan City, Laguna 4024, Philippines.


INTRODUCTION
Pipturus arborescens (Link) C.B. Rob.belongs to the family, Urticaceae, and is widely distributed in the Philippines where it is locally known as "dalunot" or "handalamay".It is dioecious, reaches up to 5 m tall and commonly grows in thickets and secondary forests at low and medium altitudes in China, Taiwan, Japan and Borneo, but ascends to 2000 m asl in the Philippines (Chen et al., 2004;Pelser et al., 2011).Aside from being a source of edible fruits, scrapings from the bark are used externally as poultice for boils while the leaves are used for the treatment of herpes (Van Valkenburg, 2001;Esperanza and Kitche, 2005;Quisumbing, 1978).Previous works using ethyl acetate extracts of P. arborescens leaves showed antibacterial activity against Bacillus subtilis (Rosal, 1995).Methanol extracts of the plant exhibited potency against Staphylococcus aureus and Escherichia coli (Enerio, 2007).Another study evaluating the cytotoxic and anti-oxidant properties of the plant showed that the crude methanol extract of the leaves exhibited an LC 50 of 57.5 μg/mL in the brine shrimp lethality assay and an EC 50 of 838 μg/mL in the DPPH free radical scavenging assay (Peteros and Uy, 2010).
A phytochemical screening of P. arborescens leaves indicated the presence of tannins, saponins, flavonoids and alkaloids from the chloroform extracts (Uy and Rivera, 2011).
Other previous studies using P. arborescens leaves yielded the triterpenes, glutinone, friedelin and glutinol, and a mixture of sterols, campesterol, stigmasterol and sitosterol from the hexane extracts (Gabona, 2000) and ficaprenol-10 and squalene from the methanol extracts (Peteros, 2010).There are few reports on the genus Pipturus.Only two other species, P. albidus and P. argenteus, have been studied.Studies on P. albidus yielded three phenolic acids, catechin, chlorogenic acid and rutin (Kartika et al., 2007) and led to the discovery of its potential antimicrobial activity against S. aureus and S. pyogenes (Locher et al., 1995).Both ethanol and water extracts from the leaves and bark of P. albidus exhibited a highly selective inhibition of the replication of the human immunodeficiency Virus Type-1 (HIV-1) with low cytotoxicity on normal cells (Locher et al., 1996).A study on the extracts of P. argenteus, along with Phyllanthus pulcher and/or Piper betel led to the development of a skin aging preventing cream (Mitani et al., 2000a).Another study on P. argenteus, Rauwolfia serpentine and Oxalis corniculata showed the usefulness of the extracts as melanin production inhibitor (Mitani et al., 2000b).

Sample collection
Samples of leaves and twigs of Pipturus arborescens (Link) C.B. Rob were collected from the De La Salle University-Science and Technology Complex (DLSU-STC) riparian forest in February 2014.The samples were authenticated and deposited at the De La Salle University Herbarium with voucher specimen # 921.

Isolation and structure elucidation
The isolation and structure elucidation of 1-4b from the leaves and twigs of P. arborescens were reported previously (Ragasa et al., 2014).

Preparation of compounds
The triterpenes (1-3b) and sterols (4a and 4b) from P. arborescens were dissolved in dimethyl sulfoxide (DMSO) to make a 4 mg/mL stock solution.Working solutions were prepared in complete growth medium to a final non-toxic DMSO concentration of 0.1%.

Maintenance and preparation of cells
The bioactivity of the dichloromethane (CH 2 Cl 2 ) extracts from P. arborescens was tested on the following human cell lines: breast cancer (MCF-7), colon cancer (HCT-116 and HT-29), and human dermal fibroblast, neonatal (HDFn)(ATCC, Manassas, Virginia, U.S.A.) which are routinely maintained at the Cell and Tissue Culture Laboratory, Molecular Science Unit, Center for Natural Science and Ecological Research, De La Salle University.
Upon reaching 80% confluence, the monolayer cultures were washed with phosphate-buffered saline (PBS, pH 7.4, Gibco ® , USA), trypsinized with 0.05% Trypsin-EDTA (Gibco ® , USA) and resuspended with complete fresh media.Cells were counted following standard trypan blue exclusion method using 0.4% Trypan Blue Solution (Gibco ® , USA).Cells were later seeded in 100µL aliquots into 96-well microtiter plates (Falcon TM , USA) using a final inoculation density of 1 x 10 4 cells/well.The plates were further incubated overnight at 37°C with 5% CO 2 in a 98% humidified incubator until complete cell attachment was achieved.These plates were used for the bioassay as described below.

Cell viability assay
The cytotoxicity of the P. arborescens compounds was determined in a cell viability test using PrestoBlue ® (Molecular Probes ® , Invitrogen, USA).The bioassay is based on the presence of mitochondrial reductase in viable cells that converts the resazurin dye (blue and nonfluorescent) to resorufin (red and highly fluorescent).The conversion is proportional to the number of metabolically active cells and is determined quantitatively using absorbance measurements.To the monolayers in the microtiter plate, 100 µL of filter-sterilized 1-4b were added to corresponding wells at two-fold serial dilutions to make final screening concentrations of 50, 25, 12.5, 6.25, 3.12, 1.56, 0.78, and 0.39 µg/mL.
The treated cells were further incubated for 4 days at 37°C in 5% CO 2 and 98% humidity.Twenty microliters of PrestoBlue ® was added to each well.The cells were incubated for 1 hr at 37°C in 5% CO 2 and 98% humidity.Wells with no sample added served as negative controls, wells with Zeocin TM (Gibco ® , USA) served as positive controls.Absorbance measurements were carried out using BioTek ELx800 Absorbance Microplate Reader (BioTek ® Instruments, Inc.) at 570 nm and normalized to 600 nm values (reference wavelength).Absorbance readings were used to calculate for the cell viability for each sample concentration following the equation below.
Nonlinear regression and statistical analyses were done using GraphPad Prism 6.05 (GraphPad Software, Inc.) to extrapolate the half maximal inhibitory concentration, IC 50 (the concentration of the compound which resulted in a 50% reduction in cell viability).
The cytotoxicity of 1-4b was expressed as IC 50 values.All tests were performed in triplicates and data were shown as mean ± SEM.The extra sum-of-squares F-test was used to evaluate the differences in the best-fit parameters (half maximal inhibitory concentration) among data sets (treatments) and to determine the differences among dose-response curve fits according to the software's recommended approach.One-way ANOVA (p < 0.05) was also conducted to determine significant differences among group variables, followed by the multiple comparison, Tukey's post hoc test (p < 0.05), to compare different pairs of data sets.
Results were considered significant at p < 0.05.

RESULTS AND DISCUSSION
This study presents investigations on the cytotoxic activities of squalene (1), friedelin (2), 2:3 ratio of ursolic acid (3a) and oleanolic acid (3b) and 2:1 ratio of β-sitosterol (4a) and stigmasterol (4b) (Figure 1), previously isolated from the dichloromethane extracts of P. arborescens leaves and twigs (Ragasa et al., 2014), against three human cancer cell lines, breast (MCF-7) and colon (HT-29 and HCT-116), and a human normal cell line, human dermal fibroblast, neonatal (HDFn).Zeocin, a known anti-cancer drug, was used as positive control.Figures 2  and 3 shows the % cell viability as a function of the logarithmic values of sample concentration.The plots mostly follow the typical sigmoidal curve characteristic of an inhibitory doseresponse relationship.Figure 2 compares the anti-proliferative effect of all samples per cell line, while Figure 3 compares the effect on cell viability of a sample against all the cell lines used.The corresponding IC 50 values are summarized in Table 1.were extrapolated from dose-response curves generated from nonlinear regression analysis done using GraphPad Prism 6.05 (GraphPad Software, Inc.).For each cell line, one-way ANOVA was conducted to determine significant differences among data sets.The results are: MCF-7, F(5,126) = 159.6 p < 0.0001; HCT-116, F(5,126) = 98.14, p < 0.0001; HT-29, F(5,122) = 266.6,p < 0.0001; HDFn, F(5,115) = 0.0006892, p > 0.9999.
The breast cancer cell line (MCF-7) is most susceptible to 1 and 4a and 4b, with IC 50 values of 5.92 and 8.62 μg/mL, respectively, followed by 2 and 3a and 3b, with IC 50 values of 16.51, and 23.97 μg/mL, respectively.Tukey's multiple comparison post hoc test showed that there was no significant difference between the IC 50 values for 1 and 4a and 4b (p > 0.05), as well as between 2 and 3a and 3b (p > 0.05).However, significant differences were established between other pairs of treatments (p < 0.0001).The colon cancer cell line, HCT-116, was most susceptible to 4a and 4b and 2, with IC 50 values of 1.14 and 1.22, μg/mL, respectively, followed by 3a and 3b and 1, with IC 50 values of 1.66, and 4.21 μg/mL, respectively.The IC 50 values for 4a, 4b and 2 are comparable (p > 0.05) while those between 3a, 3b and 1 are statistically different (p < 0.0001).The growth of the other colon cancer cell line, HT-29, was inhibited only at higher concentrations of the compounds, with IC 50 values of 11.97, 29.46, 33.63, and 52.52 μg/mL for 4a and 4b, 3a and 3b, 2 and 1, respectively.The IC 50 value for 4a and 4b is significantly different from those of the other compounds (p < 0.0001).The normal cell line, HDFn cell, did not exhibit cytotoxicity to any of the samples, with IC 50 values of >100 μg/mL in all cases (p > 0.05).On the other hand, all the cell lines exhibited cytotoxicity to Zeocin, with IC 50 values of 1.19, 0.36, 0.36, and 3.99 μg/mL for MCF-7, HCT-116, HT-29 and HDFn, respectively.Statistical analysis comparing dose-response curves for all treatments showed significant differences among best-fit values (half maximal inhibitory concentration) for all cell lines and treatments (Figures 3 and 3).Overall, comparing the three human cancer cell lines, HCT-116 was the most reactive to 1-4b, requiring the least half maximal inhibitory concentrations.This was followed by MCF-7 cell line which was moderately affected by 1-4b.The cell line HT-29 reacted the least to 1-4b.Other studies showed similar reaction where HCT-116 cells were found to be more reactive to cytotoxic compounds compared to MCF-7 (Abeer et al., 2011;Kumar et al., 2015;Malek et al., 2011;Mohammed et al., 2011;Sung et al., 2015).Compound 1 was most effective against colon cancer cells HCT-116 and breast cancer cells MCF-7, with IC 50 values of 4.21 and 5.92 μg/mL, respectively.Triterpene 2, 3a and 3b, and 4a and 4b exhibited high cytotoxicity against HCT-116 cells (IC 50 values of 1.22, 1.66 and 1.14 μg/mL, respectively).Only the mixture of 4a and 4b was found to be moderately inhibitory against HT-29 (IC 50 value of 11.97 μg/mL).None of the samples was cytotoxic to the normal cells HDFn (>100 μg/mL in all cases).The US National Cancer Institute has defined the active cytotoxic limits of natural products as 20 μg/mL or less for crude extracts and 4 μg/mL or less for pure compounds (Geran et al., 1972).Pure compounds that exhibit active cytotoxicity may have some potential for drug development (Jacinto et al., 2011).The results showed that 1-4b from Pipturus arborescens can be further developed for the treatment of cancer against the human colorectal type, HCT-116, and human breast, MCF-7.
The study showed that the cytotoxic activity of 1-4b depended on the specific type of cancer cell being targeted.When the two colon cancer cell lines (HCT-116 and HT-29) are compared, the IC 50 values of 1-4b for HCT-116 were lower, implying that HCT-116 is more responsive to anti-cancer treatments using the samples tested.It was reported that differences in the expression profiles of several genes associated with drug sensitivity between HCT-116 and HT-29 could be an important factor in how the cells respond to different inhibitory compounds (Makizumi et al., 2008).A related study using four human colon cancer cells (HCT-116, HT-29, HCT-15, KM-12) showed that gene expression profiling after inhibition of signal transduction by 17-allylamino-17-demethoxygeldanamycin, an inhibitor of the hsp90 molecular chaperone, could explain why cells responded differently under similar treatment conditions (Clarke et al., 2000).
Squalene (1) was reported to significantly suppress colonic aberrant crypt foci (ACF) formation and crypt multiplicity in laboratory rats which strengthened the hypothesis that it possesses chemopreventive activity against colon carcinogenesis (Rao et al., 1998).
In a study evaluating the bioactivity of tocotrienols, carotenoids, squalene and coenzyme Q10 from palm oil with respect to the anti-proliferative effects on two human breast cancer cell lines, MDA-MB-231 and MCF-7, it was found that there was a suppression of nuclear factor kappa-light-chainenhancer of activated B-cells (NF-κB) protein in the breast cancer cells exposed briefly to tumor necrosis factor-alpha (TNF-α) (Loganathan et al., 2013;Loganathan et al., 2015).The preventive and therapeutic effects of squalene-containing compounds on tumor promotion and regression have been reported (Desai et al., 1996).Recent reviews on the bioactivities of squalene have been provided (Ronco and De Stéfani, 2013;Chudzik et al., 2015).Thus, 1 was reported to exhibit cytotoxic properties against colon and breast cancer cell lines which corroborate our findings that 1 showed high cytotoxicity against colon cancer cells HCT-116 and breast cancer cells MCF-7, with IC 50 values of 4.21 and 5.92 μg/mL, respectively.Friedelin (2) showed significant anti-proliferative effects against both human cervical cancer cell line (HeLa) and human cutaneous squamous carcinoma cell line (HSC-1) (Prabhu et al., 2013).In another study, 2 exhibited the strongest inhibitory activity against HeLa cancer cells with an IC 50 of 3.54 μg/ml (Utami et al., 2013).It also displayed anti-proliferative properties against human melanoma cells (A375), mouse lung epithelial tumor cells (L929), human cervical tumor cells (Hela), and human macrophage tumor cells (THP-1) in a time-and dose-dependent manner (Lu et al., 2010).Triterpene 2 was found to exhibit growth inhibitory activities against MBA-MD-231 human breast cancer cells (Ee et al., 2005).It was reported that the synergistic effects of 2, 28-hydroxy-3-friedelanone and 7-methoxy-coumarin inhibited the growth of acute promyelocytic leukemia cells lines NB4 and HT93A (Sangsuwon et al., 2013).Thus, 2 was reported to exhibit cytotoxic effect against human breast cancer cells which corroborate the cytotoxic property of 2 found in our study.Our study also showed that 2 was highly effective against colon cancer cells HCT-116 with an IC 50 value of 1.22 μg/mL.Ursolic acid (3a) was found to induce apoptosis in tumor cells by activation of caspases and modulation of pathways affecting cell proliferation and migration (Neto, 2011).It also decreased proliferation and induced apoptosis in gastric cancer cell line BGC-803 and hepatocellular cancer cell H22 xenograft, both in vivo and in vitro (Wang et al., 2011).Previous studies showed that 3a exhibited anti-tumor activity against human colon carcinoma cell line HCT15 (Li et al., 2002) and inhibited the growth of colon cancer-initiating cells by targeting STAT3 (Wang et al., 2013).Triterpene 3a and betulinic acid were found responsible for the anti-estrogenic effects suggesting its potential use as therapeutic agents against estrogen-dependent tumors (Kim et al., 2014a).Furthermore, 3a has potential therapeutic use also against prostate cancer through its anti-proliferative and apoptotic effects (Kassi et al., 2007).A recent study reported that 3a inhibited cell growth and proliferation of Jurkat leukemic T-cells, inhibiting PMA/PHA induced IL-2 and TNF-α production in a concentration and time dependent manner (Kaewthawee and Brimson, 2013).A study on cervical cancer cells TC-1 reported that ursolic acid-activated autophagy induced cytotoxicity and reduced tumor growth in a concentration-dependent manner Leng et al., 2013).Another study evaluated the antitumor activities of 3a on U87MG brain cancer cells and found that both G1-phase arrest and autophagy were induced by the compound (Shen et al., 2014).In a study evaluating the anticancer properties of ursolic acid and three flavonoids, daidzein, baicalein, and hesperidin, it was found that 3a and baicalein inhibited the proliferation of MCF-7 breast cancer cells induced by PhIP, a food-derived carcinogen with estrogenic activity (Lee et al., 2010).The anticancer potential of 3a in different berries has been provided (Neto, 2011).Oleanolic acid (3b) was found to be anti-mutagenic and anti-tumor, inhibiting proliferation of gastric, colon, and liver cancer cells by inducing apoptosis and necrosis (Zhang et al., 2011).Triterpene 3b was found to inhibit mouse skin tumor (Oguro et al., 1998) and exhibited significant anti-tumor activity against human colon carcinoma cell line HCT 15 (Li et al., 2002).A recent study identified 3b as an anti-tumor compound able to suppress aerobic glycolysis in MCF-7 breast cancer cells by inducing a metabolic switch in the PKM2 to PKM1 ratio which is important in cancer development (Liu et al., 2014).Thus, 3a and 3b were reported to exhibit cytotoxic properties against colon and breast cancer cell lines which corroborate our findings that a mixture of 3a and 3b showed high cytotoxicity against colon cancer cells HCT-116 with an IC 50 value of 1.66 μg/mL and moderate cytotoxicity against breast cancer cells MCF-7, with an IC 50 value of 23.97 μg/mL.
β-Sitosterol (4a) was observed to influence the programmed cell death pathway in human breast cancer cells (MCF-7) and human adenocarcinoma cells (MDA-MB-231), inhibiting tumor proliferation by promoting apoptosis (Awad et al., 2007).It also inhibited cell proliferation of human colon cancer cell line (HT-29) (Jayaprakasha et al., 2007).In vitro studies showed that 4a inhibited the growth of human colon cancer cells (COLO 320 DM) with an IC 50 of 266.2 μM, inducing apoptosis by scavenging oxidants and attenuating β-catenin and PCNA expression while in vivo studies proved that 4a reduced the number of aberrant crypt and crypt multiplicity in DMH-initiated rats in a dose-dependent manner (Baskar et al., 2010).It significantly reduced the expression of Niemann-Pick C1-like 1 (NPC1L1) in human small intestine epithelial cell line (FHs 74 Int) to reduce intestinal cholesterol absorption at the cellular level (Jesch et al., 2009).It also induced apoptosis mediated by the activation of ERK and the downregulation of Akt in murine fibrosarcoma cells (MCA-102) (Moon et al., 2007).Sterol 4a .inhibited the growth of HT116 human colon cancer cells by a number of mechanisms including the activation of caspase-3 and caspase-9 accompanied by proteolytic cleavage of poly(ADPribose)-polymerase and the reduction of the expression of the antiapototic Bcl-2 protein and mRNA and a subsequent increase of the pro-apototic Bax protein and mRNA (Choi et al., 2003).Stigmasterol (4b) decreased tumor volume and cell viability, increasing the life span of Ehrlich ascites carcinoma (EAC)bearing mice (Ghosh, 2011).It was reported that 4a and 4b exhibited anti-proliferative activities against human prostate cancer cells (DU-145) by increasing p53 protein expression and inhibiting carcinoma development by decreasing p21 and p27 protein expression (Scholtysek et al., 2009).Other studies reported that 4b showed cytostatic activity against HEp-2 and McCoy cells (Gómez et al., 2001), significantly inhibited tumor promotion in mouse skin two-stage carcinogenesis experiments (Kasahara et al., 1994), and demonstrated antimutagenic (Lim et al., 2005), and anti-oxidant (Panda et al., 2009) properties.A recent report showed that 4b induced apoptosis in hepatocarcimona cells (HepG2) by up-regulating the expression of pro-apoptotic gene expressions (Bax, p53) while down-regulating the anti-apoptotic genes (Bcl-2), activating caspase-8 and caspase-9 in the process (Kim et al., 2014b).Thus, 4a and 4b were reported to exhibit cytotoxic properties against several cancer cell lines.β-Sitosterol was reported to possess cytotoxicity against colon and breast cancer cell lines which corroborate our findings that a mixture of 4a and 4b showed high cytotoxicity against colon cancer cells HCT-116 and breast cancer cells MCF-7, with IC 50 values of 1.14 and 8.62 μg/mL, respectively.