Cells were cultured in medium with different concentrations of Doxorubicin (Sigma-Aldrich) and P. trimera extract for 48 h and 72 h, respectively. Briefly, after one day of pre-incubation for the attachment of VNBRCA1 cells to the culture surface, cells were treated with Doxorubicin at the concentration of 1000, 500, 250, 125, 75, 32.5, 16, 8, or 0 (μg/mL). At the same time, cells were also treated with P. trimera extract at 200, 100, 50, 25, 12.5, 6, 3, 1.5, or 0 (mg/mL). Each concentration was tested in triplicate. Cell viability was measured by MTT assay. Each experiment was repeated three times. The data were analyzed by GraphPad Prism (GraphPad Software, Inc., La Jolla, CA, USA).

After cells were treated with chemotherapeutic agents, MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide) assay was performed to determine the IC₅₀ value of Doxorubicin and P. trimera extract. Briefly, MTT agent was added into each well on the second day of Doxorubicin treatment and on the third day of plant extract treatment. Then, the plates were incubated at 37^oC for 4 h. After that, MTT agent was removed from those wells. During the next step, dimethyl sulfoxide (DMSO) was added to wells and incubated for 15 min at 37^oC. The OD value at 595 nm wavelength was measured by DTX880 machine (Beckman Coulter, USA). The IC₅₀ was determined based on the dose-response curve. The X-axis of the dose-response curve was the logarithm of the different concentrations of the agent, and the Y-axis is the inhibition percentage calculated from the OD value. From this sigmoidal curve, IC₅₀ (X value) was calculated by setting Y value as 50%.

Cell viability of VNBRCA1 at 250 µg/mL concentration was evaluated by Hoechst 33342/ PI staining. Skin human fibroblast (HF) cells were used as a control cell line. Briefly, VNBRCA1 and HF cells were treated with P. trimera extract at a concentration of 250 µg/mL; untreated cells were cultured in normal media complemented with solvent. At day 3 of treatment, cells were stained with Hoechst 33342/ PI. The final concentration of Hoechst 33342, as well as PI, was 1 µg/mL in culture media. Next, cells were incubated in an incubator with the condition of 37^oC, 5% CO₂ for 5 minutes. Stained cells were observed under a fluorescent microscope.

The culture flask of VNBRCA1 was treated with P. trimera extract at 3 concentrations of 30, 60 and 125 µg/mL, for 3 days. After that, annexin V/7-AAD staining and flow cytometry techniques were performed to determine the mode of cell death. Briefly, all cells were detached from the flask surface by trypsin and centrifuged at 300 g to collect the pellet. After that, cells were stained with annexin-V-FITC and 7-AAD. Stained cells were analyzed by a BD FACSCalibur (BD Biosciences, San Jose, CA, USA) flow cytometer. Data was analyzed by CellQuest Pro software.

Data was analyzed by Graphpad Prism software (Graphpad Software Inc.) for the determination of IC50 value. All data were presented as the mean of triplicate experiments. Statistically significant difference was set as p<0.05.

After 3 days of treatment with 10 different concentrations of P. trimera extract (2000, 1000, 500, 250, 125, 62.5, 31.25, 15.63, 7.81, or 3.91 µg/mL; control), proliferation of P. trimera extract-treated cells was measured by MTT assay. The IC₅₀ value of P. trimera extract on VNBRCA1 after 72 h was 106±10 µg/mL, while IC₅₀ of P. trimera extract on HF after 72 h was 310±8 µg/mL (Figure 1).

Similar to the method for calculating the IC₅₀ of P. trimera extract, the OD value was measured to evaluate the dose-response inhibition of Doxorubicin (1, 0.5, 0.25, 0.125, 0.0625, 0.031, 0.016, or 0 (µg/mL) on VNBRCA1 and on HF cells. The IC₅₀ value of Doxorubicin on VNBRCA1 after 48 h is 0.63±0.12 µg/mL while IC₅₀ value of Doxorubicin on HF after 48 h was 0.135±0.09 µg/mL.

Based on the IC50 value of each agent on both VNBRCA1 and HF, we determined the selectivity index (SI). This parameter reflects the selective toxicity of the agent against cancer cell but not normal cells. In this study, we calculated SI of the agent by the ratio of the IC₅₀ of the agent on HF to the IC₅₀ of the agent on VNBRCA1 (Table 1):

SI = IC50 of  agent on HFIC50 of agent on VNBRCA1

The higher the SI, the more selective the toxicity of the agent against breast cancer cells (but not HF cells). Calculations of SI_extract and SI_Doxorubicin were performed to determine the selective cytotoxicity of both the plant extract and Doxorubicin against VNBRCA1.

SI of P. trimera extract on HF/VNBRCA1:

SI_{P.trimera extract}= IC50 of extract on HFIC50 of extract on VNBRCA1=310106= 2.92

SI of Doxorubicin on HF/VNBRCA1:

SI_Doxorubicin= IC50 of Doxorubicin on HFIC50 of Doxorubicin on VNBRCA1=0.1350.63= 0.21

Difference between SI_{P.trimera extract} and SI_Doxorubicin

SI of extractSI of Doxorubicin=2.92920.2121=13.92

SI_{P. trimera extract} was nearly 14-fold higher than SI_Doxorubicin.

Hoechst 33342 is a blue-fluorescence dye and PI is a red-fluorescence dye; each has different excitation/emission wavelengths (350/461 nm and 536/617 nm, respectively). Hoechst 33342 can stain the DNA of both dead and living cells while PI is only permeable to dead cells. At a concentration of 250 µg/mL, P. trimera plant extract caused severe cell death to VNBRCA1 (red spots in Figure 2C), while P. trimera extract did not show toxicity in HF cells (Figure 2A). This indicated the selective cytotoxic ability of P. trimera extract on VNBRCA1 cells.

The total apoptotic populations (late apoptosis in upper right quadrant and early apoptosis in lower right quadrant) are shown in Figure 3; the % of apoptosis was 2.98% (for untreated cells), 4.37% (for the 30 µg/mL P. trimera extract sample), 6.99% (for the 62.5 µg/mL sample), and 20.97% (for the 125 µg/mL sample). The results also showed that the apoptosis percentage clearly increased when the concentration of P. trimera extract doubled, while the necrotic cell population was tightly augmented from 0.29% (in control sample) to 10% (in the 125 µg/mL sample). Viable cells were significantly reduced from 96% to 78%, in the 125 µg/mL P. trimera extract sample.

Figure 4

We used the plant extract at a concentration of 200 µg/mL, which is nearly double of the IC50 value of the P. trimera extract on VNBRCA1 to test effects of the extract on DNA damage. The results showed that the extract caused disintegrated nuclei in VNBRCA1 after 24 h of treatment, as compared to human fibroblast cells (where the extract did not induce much DNA disintegration) (Figure 4). This confirmed that the P. trimera extract can induce apoptosis in VNBRCA1 cells.

P.trimera has been traditionally used for a long time as a medicinal plant. The cytotoxic effects of P. trimera extract on certain cell types (such as Hela and HepG2, but not for VNBRCA1) have been reported in several studies. In this paper, we demonstrated the evidence of selective cytotoxicity of the plant extract against VNBRCA1 cells compared to normal human fibroblast. Furthermore, we provided evidence of the apoptosis-inducing effect of P. trimera extract on VNBRCA1 in vitro.

The results from the MTT assay showed that the IC₅₀ of the P. trimera plant extract on VNBRCA1 was 106 µg/mL, which is about 3-fold lower than the IC50 value of the P. trimera extract on human skin fibroblasts (HF; a non-cancerous cell line), which was 310 µg/mL. This means that the P. trimera extract has a 3-fold stronger killing effect on VNBRCA1 cells than on HF cells. Thus, the P. trimera extract killed breast cancer stem cells (VNBRCA1), but had negligible side effects on normal human skin cells. A previous study showed that the methanolic extract of P. trimera was active against HepG2 and Hela cell lines with IC50 values of 39.61 and 5.36 µg/mL, respectively. The previously reported IC50 values of P. trimera on HepG2 and Hela also showed the selective toxicity of the plant extract in comparison to its IC50 value on HF in this study (Figure 1).

Meanwhile, we also investigated the cytotoxicity of Doxorubicin on VNBRCA1 and HF. Interestingly, the IC₅₀ of Doxorubicin on HF was 0.135 µg/ml, which was almost 4.5-fold lower than the IC₅₀ of Doxorubicin on VNBRCA1 (0.63 µg/mL). Thus, Doxorubicin kills more normal cells (HF) than cancer cells (VNBRCA1) (Figure 1).

From the IC₅₀ data from Table 1, we bring up the equation called selectivity index (SI), which is the ratio of (IC50 of the agent on HF)/(IC50 of the agent on VNBRCA1). The selectivity index was previously used to figure out the selective cytotoxic activity of synthesis drugs on human breast carcinoma MCF-7 compared to normal cell line14. In this study, the agent could be either P. trimera extract or Doxorubicin. The greater the SI value is, the more toxic the agent is against VNBRCA1 (but safer against HF). Therefore, calculations of SI_extract and SI_Doxorubicin were performed to compare the selective cytotoxicity of both P. trimera extract and Doxorubicin, on cancer cells (VNBRCA1) and normal cells (HF). The SI of the P.trimera extract was nearly 14-fold higher than that for Doxorubicin (2.92>0.214). From the above results, it can be concluded that the P.trimera extract not only showed less toxicity on HF than on VNBRCA1, but also showed more selective toxicity than Doxorubicin. Because of its considerably high selective cytotoxicity, P.trimera extract could become a potential candidate as an anticancer agent.

Hoechst 33342/ PI staining results indicated that P. trimera extract showed cytotoxicity against VNBRCA1 but not HF, at the concentration of 250 µg/mL (Figure 2). In other words, at this concentration, the extract was safe on HF, while a large population of VNBRCA1 underwent cell death. Taken together, the data show that P. trimera extract primarily displays a highly selective toxicity in cancer cells. However, in order to further examine the side effect index of the P. trimera extract, it was necessary to perform the cytotoxicity assay on other normal cell lines, such as non-cancerous breast cell line or normal liver cell line. In recent studies, the side effect index was similarly performed to compare the IC₅₀ value of the agent on non-cancerous breast cell line and cancerous breast cell line15, or to determine the IC₅₀ value of the agent on Hep-G2 cell line versus normal liver cell line16. PI staining assay further showed the selective cytotoxicity of the P. trimera extract on the 2 cell lines, HepG2 and HF, but the result did not indicate whether the P. trimera extract induced apoptosis on HepG2 or not. The reason for this is that PI stained both necrotic cells and late apoptotic cells, thus PI staining only could not discriminate the two types of cell death17,18,19.

Apoptosis induction of medicinal plant extract is an important characteristic required in cancer drug screening20. It is a more favorable mode of cell death than necrosis since cell death via necrosis can cause inflammation during treatment21. Apoptosis is characterized by chromatin condensation, cell membrane shrinkage, and DNA fragmentation. In this study, the apoptotic cell population from P. trimera extract-treated cells was firstly analyzed by flow cytometry. VNBRCA1 cells were cultured in medium complemented with P. trimera extract, with various concentrations (30, 60, or 125 µg/mL) of P. trimera for 24 hours, and then the apoptotic population was identified by Annexin V staining. Meanwhile, the necrotic population was identified by 7-AAD staining. The results showed that percentage of apoptotic population increased as concentrations of P. trimera extract were increased. The early and late apoptotic population was 2.98% (for untreated sample), 4.37% (for 30 µg/mL extract sample), 6.99% (for 60 µg/mL extract sample), and 20.97% (for 125 µg/mL extract sample) (Figure 3). Thus, the apoptosis induced by P. trimera extract appears to be dose-dependent.

However, the percentage of necrotic population was not significantly increased. Even at the highest concentration of P. trimera extract (125 µg/mL), the necrotic population was only 1%, which was not considerably different from the portion of apoptotic population (0.29 %) in the control sample. Recently, another study reported the dose-dependent apoptosis of MCF-7 breast cancer cell line treated with Lepidium sativum seed extract22. However, the authors reported that extracts of Lepidium sativum seeds also caused apoptosis in human skin fibroblasts at the same level as that of MCF-7 breast cancer cells. In general, the Lepidium sativum extract did not selectively kill cancer cells in comparison with the P. trimera extract. However, in our study, the P. trimera extract was tested only on VNBRCA1 cells and HF cells, so it is necessary to examine the cytotoxicity of this plant extract on other breast cancer cell lines. Based on those results, we can conclusively compare the P. trimera extract with other extracts evaluated in other studies.

Nuclei observation was performed in order to confirm the findings from the flow cytometry results which indicated that the extract had the ability to induce apoptosis. Disintegrated nuclei appeared in the fluorescent images when cells were treated with P. trimera extract, demonstrating that the treated cells underwent apoptosis. On the other hand, only a few dead cells were observed in the control sample and those cells did not show abnormal nuclei. These observations are in agreement with the flow cytometry results which indicating that the extract can induce apoptosis on VNBRCA1 after a short period of incubation.

Overall, the results from this study suggest that P. trimera methanolic extract could be a potential anticancer agent. MTT assay results and microscopic images indicated that the cytotoxicity of the extract was dose-dependent. The P. trimera extract selectively killed the VNBRCA1 breast cancer cell line but not normal human fibroblast HF. In addition, P. trimera extract was shown to induce apoptosis in VNBRCA1 after a short term of incubation (24 hours). Further studies are warranted to investigate the antitumor properties of P. trimera extract in vivo. Identification of bioactivity compounds in P. trimera extract is also necessary. In conclusion, P. trimera could be a promising anti-proliferative agent against Vietnamese breast cancer cells.

HF: human fibroblast

P. Trimera: Paramignya trimera

PI: propidium iodide

SI: selectivity index

VNBRCA1: Vietnamese breast cancer stem cell line

Authors declare no conflict of interest.

Sinh and Hung carried out the IC50 study. Sinh helped to perform FCM assay and fluorescent of disintergrated nuclei experiment. Mai, Hai, Nhan participated in performing the extract from Paramignya trimera root. Ngoc and Kiet and Phuc helped to draft the manuscript.

This research was funded by Vietnam National University, Ho Chi Minh city, Viet Nam under grant number A2015-18-01/HD-KHCN.

Ferlay J. Soerjomataram I. Dikshit R. Eser S. Mathers C. Rebelo M. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012 Int J Cancer 2015 136 5 359 86 0020-7136 10.1002/ijc.29210 25220842 Trieu P.D. Mello-Thoms C. Brennan P.C. Female breast cancer in Vietnam: a comparison across Asian specific regions Cancer Biol Med 2015 12 3 238 45 2095-3941 26487968 Al-Hajj M. Wicha M.S. Benito-Hernandez A. Morrison S.J. Clarke M.F. Prospective identification of tumorigenic breast cancer cells Proc Natl Acad Sci USA 2003 100 7 3983 8 0027-8424 10.1073/pnas.0530291100 12629218 Gordaliza M. Natural products as leads to anticancer drugs Clin Transl Oncol 2007 9 12 767 76 1699-048X 10.1007/s12094-007-0138-9 18158980 Wall M.E. Wani M.C. Cook C.E. Palmer K.H. McPhail A.T. Sim G.A. Plant Antitumor Agents. I. The Isolation and Structure of Camptothecin, a Novel Alkaloidal Leukemia and Tumor Inhibitor from Camptotheca acuminata1,2 J Am Chem Soc 1966 88 16 3888 90 0002-7863 10.1021/ja00968a057 Chavan S. Pathak A.B. Kalkote U.R. Total synthesis of (+)-camptothecin via an intramolecular palladium-catalyzed cyclization strategy Synlett 2007 2007 17 3 0936-5214 10.1055/s-2007-991054 Kingston D.G. Newman D.J. Taxoids: cancer-fighting compounds from nature Curr Opin Drug Discov Devel 2007 10 2 130 44 1367-6733 17436548 Camidge R. The Story of Taxol: Nature and Politics in the Pursuit of an Anti-Cancer Drug BMJ 2001 323 7304 115 115 0959-8138 10.1136/bmj.323.7304.115/a Pham H.B. N.M.K., Nguyen T.B.T, Isolation and quantitative analysis of ostruthin in the medicinal plant Paramignya trimera collected in Vietnam Medicine Master 2013 18 3 173 9 Li F. Okamura Y. Dibwe D.F. Awale S. Kadota S. Tezuka Y. Anti-austerity agents from Rhizoma et Radix Notopterygii (Qianghuo) Planta Med 2012 78 8 796 9 0032-0943 10.1055/s-0031-1298463 22516931 Cuong N.M. Huong T.T. Khanh P.N. Tai N.V. Ha V.T. Son N.T. Paratrimerins A and B, Two New Dimeric Monoterpene-Linked Coumarin Glycosides from the Roots and Stems of Paramignya trimera Chem Pharm Bull (Tokyo) 2015 63 11 945 9 0009-2363 10.1248/cpb.c15-00336 26521859 Pham P.V. Phan N.L. Nguyen N.T. Truong N.H. Duong T.T. Le D.V. Differentiation of breast cancer stem cells by knockdown of CD44: promising differentiation therapy J Transl Med 2011 9 1 209 1479-5876 10.1186/1479-5876-9-209 22152097 Pham P.V. Vu B.T. Nhan P.L.C. Thuy D.T. Ceccherini-Nelli Matteoli Barbara Isolation of Breast Cancer Stem Cells by Single-Cell Sorting Biomedical tissue culture 1 Intech Open 2012 59 72 10.5772/51506 Badisa R.B. Darling-Reed S.F. Joseph P. Cooperwood J.S. Latinwo L.M. Goodman C.B. Selective cytotoxic activities of two novel synthetic drugs on human breast carcinoma MCF-7 cells Anticancer Res 2009 29 8 2993 6 1791-7530 19661306 Abu N. Akhtar M.N. Ho W.Y. Yeap S.K. Alitheen N.B. 3-Bromo-1-hydroxy-9,10-anthraquinone (BHAQ) inhibits growth and migration of the human breast cancer cell lines MCF-7 and MDA-MB231 Molecules 2013 18 9 10367 77 1420-3049 10.3390/molecules180910367 23985955 Al-Qubaisi M. Rozita R. Yeap S.K. Omar A.R. Ali A.M. Alitheen N.B. Selective cytotoxicity of goniothalamin against hepatoblastoma HepG2 cells Molecules 2011 16 4 2944 59 1420-3049 10.3390/molecules16042944 21471934 Brauchle E. Thude S. Brucker S.Y. Schenke-Layland K. Cell death stages in single apoptotic and necrotic cells monitored by Raman microspectroscopy Sci Rep 2014 4 1 4698 2045-2322 10.1038/srep04698 24732136 Denecker G. Vercammen D. Declercq W. Vandenabeele P. Apoptotic and necrotic cell death induced by death domain receptors Cell Mol Life Sci 2001 58 3 356 70 1420-682X 10.1007/PL00000863 11315185 Kroemer G. Dallaporta B. Resche-Rigon M. The mitochondrial death/life regulator in apoptosis and necrosis Annu Rev Physiol 1998 60 1 619 42 0066-4278 10.1146/annurev.physiol.60.1.619 9558479 Dixon S.C. Soriano B.J. Lush R.M. Borner M.M. Figg W.D. Apoptosis: its role in the development of malignancies and its potential as a novel therapeutic target Ann Pharmacother 1997 31 1 76 82 1060-0280 10.1177/106002809703100113 8997471 Kiaris H. Schally A.V. Apoptosis versus necrosis: which should be the aim of cancer therapy? Proc Soc Exp Biol Med 1999 221 2 87 8 0037-9727 10.3181/00379727-221-44388 10352117 Mahassni S.H. Al-Reemi R.M. Apoptosis and necrosis of human breast cancer cells by an aqueous extract of garden cress (Lepidium sativum) seeds Saudi J Biol Sci 2013 20 2 131 9 1319-562X 10.1016/j.sjbs.2012.12.002 23961228