Ethyl acetate extract of Elephantopus mollis Kunth. induced apoptosis, senescence, and inhibited metastasis-associated traits of human lung cancer cells Authors: Tran Thanh Long, Dang Minh Phuc, Phan Dang Khoi, Dang Thi Phuong Thao, Nguyen Thi My Trinh Published: 2026 ABSTRACT Background and purpose:Elephantopus mollis Kunth. (EM) is an attractive candidate for cancer treatment research due to its long history of use in traditional cancer remedies. As lung cancer is the most widespread and among the most lethal carcinomas, we assessed the toxicity and anti-metastatic ability of EM extracts on the lung cancer NCI-H1299 cells.Experimental approach:The cytotoxicity of EM extracts was evaluated via the MTT assay. Phytopolyphenol constituents of the potent extract were evaluated by HPLC. The underlying mechanisms of this toxicity were scrutinized by cellular and nuclear morphologies, cleaved caspase-3/8/9 staining, senescence-associated beta-galactosidase activity, and gene expression analysis. The scratch, invasion, adhesion assays, and RT-qPCR were employed for the anti-metastasis study.Findings/Results:The ethyl acetate fraction (EMEA) was the most potent extract against NCI-H1299 cells, with a low IC50 and an impressive selective effect. Most attractively, this is the very first study showing that EMEA induced both apoptosis and senescence in a cancer cell line. The underlying mechanisms involved the extrinsic apoptotic pathway, DNA damage, p21-dependent cell-cycle arrest, and oxidative stress response. EMEA could also strongly inhibit the migration, invasion, and collagen I adhesion abilities of NCI-H1299 cells. This might stem from the reduction of two crucial metastatic MMP2 and VIM genes. As interesting as the cytotoxic findings, this is among the surprisingly sparse studies on the anti-metastasis activity of EM extracts and compounds.Conclusion and implications:Our research contributed considerably to the knowledge of anticancer properties of EM and cemented its role as a promising source for novel cancer treatment therapies. INTRODUCTION Lung carcinoma is certainly the most prevalent of all cancers, according to the 2022 report of GLOBOCAN. Intimidatingly, with a mortality rate nearly three times larger than the second-most lethal cancer, lung malignancy was also the leading cause of cancerous deaths worldwide(1). Conventionally, surgery, radiation therapy, and chemotherapy are the methods of choice for lung cancer treatment(2). These methods, unfortunately, are often accompanied by various drawbacks. On the one hand, surgery is an invasive approach and might not be effective for managing spread or metastatic tumors. On the other hand, radiation and chemotherapies usually damage healthy cells and tissues in the vicinity or distant body parts relative to the lung tumor sites. Thus, these non-specific damages can lead to numerous mild to severe or even lethal health problems(2). Seeking alternative, effective, and safe regimens for lung cancer is therefore an urgent task. With a plethora of members and a vast bioactive chemical space, herbal plants are among the most promising sources for potential candidates. Species of the Elephantopus family have long been employed in various folk medicines as well as subjects for relatively profound studies, especially on their anticancer effects(34). Scientific research from E. scaber and E. tomentosus, the family’s most extensively studied species, strongly cemented the anticancer prospects of Elephantopus’ phytochemicals(345). Consequently, more and more attention has been directed toward other members of this group, especially E. mollis Kunth. (EM). In detail, the EM methanol extract showed high cytotoxicity toward the MCF-7 breast cancer cell, while ethyl acetate and petroleum ether fractions were proven to strongly hamper liver (HepG2), lung (A549), leukemia (HL-60), stomach (AGS), and invasive breast (MDA-MB-231) cancer cell lines(678910). In addition to the extracts, several compounds from EM, most notably sesquiterpene lactone, exhibited substantially robust cancer inhibition with promising underlying mechanisms. Serial research by Hongyu Li et al. on the EM23 compound isolated from EM discovered the cytotoxicity effects of this substance on leukemia and cervical cancer cells. The low IC50 values (ranging from 1.9 to 10.8 μM) of EM23 were found to result from apoptosis induction via JNK-p38/MAPK and aSk activation in leukemia cells, as well as the reduction of XIAP, Trx/TrxR, and reactive oxygen species (ROS) accumulation in cervical cancer cell lines(1112). Based on firmly established scientific evidence, we studied the impact of EM extracts on the viability and metastasis-related traits of the highly proliferative and highly invasive lung cancer cell line NCI-H1299. Subsequently, we also initially elucidated the underlying mechanisms of the EM extract on this cell line. MATERIALS AND METHODS Plant sample and extract preparation The EM whole plant was collected in Bidoup-Núi Bà Vietnam National Park, Lam Dong Province, Vietnam. An expert in this area confirmed the scientific name of the collected plant, Dang Le Anh Tuan, Department of Ecology and Evolutionary Biology, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam. The sample was preserved with a voucher specimen: PHH0004880. The EM extracts were prepared following the procedures of previous publications(67). Briefly, whole-plant samples were air-dried and blended into a fine powder. The EM powder (75 g) was immersed in 70% ethanol (Xilong, PRC; 750 mL) at room temperature. After 24 h, we separated the liquid and the solid phases. The soaking process was repeated three times. The collected solution was filtered, and ethanol was removed by a rotary evaporator at 40 °C. Half of the result mixture was dried to a solid via a freeze-drier to obtain the ethanol extract (EMEtOH). The constituents in the other half of the mixture were extracted sequentially with petroleum ether (PE) (Xilong, PRC) and then ethyl acetate (EA) (Xilong, PRC) solvents (1:1 v/v) using a separatory funnel at room temperature. These two fractions were subsequently concentrated and dried by the rotary evaporator and the freeze-drier, respectively. The obtained solids were named petroleum ether extract (EMPE) and ethyl acetate extract (EMEA). The remaining aqueous phase after blending with ethyl acetate was freeze-dried to obtain the water extract (EMH2O). For stock solutions preparation, EMEtOH and EMH2O were dissolved in 50% dimethyl sulfoxide (DMSO; Merck KGaA, Germany), whilst the other two extracts were dissolved in 99% DMSO. These stock extracts were diluted by phosphate-buffered saline (PBS) to the intended concentrations, then filtered through sterile membranes and stored for further experiments. Cell lines and culture Human lung cancer NCI-H1299 cell (ATCC, CRL-5803) and hTERT-immortalized fibroblast BJ-5ta cell (ATCC, CRL-4001) were cultured in Roswell Park Memorial Institute (RPMI) 1640 (Himedia, India) and Dulbecco’s Modified Eagle medium (DMEM; Himedia, India) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich, USA). Cells were kept in the incubator at 37 °C and 5% CO2. In vitro cytotoxic assay The optimized 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich, USA) method was employed to assess the toxicity of EM extracts on NCI-H1299 and BJ-5ta cells(7). In general, cells were detached by trypsin (Sigma-Aldrich, USA), and a volume of 100 μL of 104 suspended cells was seeded into a 96-well plate and cultured for 16-18 h. The EM extracts, as well as doxorubicin (Ebewe Pharma, Austria) as a positive control, were diluted in PBS before cell treatment. To evaluate the cytotoxic effect of the extracts against the NCI-H1299 cells, all four EM extracts were tested at 40 μg/mL. To determine the IC50 of EMEA on NCI-H1299 and BJ-5ta cells, as well as the IC50 of doxorubicin on the NCI-H1299 cells, the cells were treated with EMEA (1.56-200 μg/mL) or doxorubicin (0.05-50 μg/mL), which were prepared by a two-fold sequential dilution from the maximum concentrations (200 and 50 μg/mL for EMEA and doxorubicin, respectively). Groups exposed to 0.1% (v/v) DMSO were chosen as the negative control. At planned time points, MTT solution (5 μL, 5 mg/mL) was added to each well and incubated for 4 h. Then, 60 μL lysis solution (30% (w/v) SDS (Himedia, India) and 0.03 N HCl (Himedia, India)) and 90 μL DMSO (99.9% (v/v)) were added. The 96-well plates were subsequently shaken for 15 min to dissolve the formazan precipitates. The optical density (OD) was measured at a wavelength of 550 nm. Cell inhibition (CI) was calculated utilizing the following equation: The fraction that prohibited more than 50% of the NCI-H1299 population after 48 h of treatment was considered for subsequent experiments. The concentration for 50% inhibition (IC50) values was interpolated from the dose (logarithm of concentration)-response (CI%) curves by GraphPad Prism 8 software. In addition to IC50, we also calculated the selectivity index (SI) of the selected extract. The SI expresses the quantitative divergence between the extract’s toxicity against healthy cells and efficacy toward cancer cells. SI was measured using the equation below: SI ≥ 2 indicates a selective effect against cancer cells whilst SI < 1 suggests the extract is more toxic to normal cells than cancer cells(9). HPLC analysis of polyphenols The polyphenolic contents of EMEA were assessed by a UV-Vis DAD-equipped HPLC system following the optimized procedure(13). Effects of EMEA on cell growth and morphology NCI-H1299 cells were seeded into 96-well plates (104 cells/100 μL/well) and treated with either intended concentrations of EMEA (15, 20, and 30 μg/mL), doxorubicin, as the positive control, at IC50 (1.8 μg/mL), and 0.1% (v/v) DMSO (as negative control). The cell populations of treated and untreated groups were quantified via the MTT method at 0, 24, 48, and 72 h time points. In parallel, NCI-H1299 cell morphologies were observed by the inverted microscope Nikon Eclipse TiU (Nikon, Japan) at respective time points. Nuclear DAPI staining Morphologies of nuclei were visualized by 4’,6-diamidino-2-phenylindole (DAPI; Thermo Fisher Scientific, USA) staining. NCI-H1299 cells were seeded (104 cells/100 μL/well of a 96-well plate) and treated with EMEA (15 μg/mL) or 0.1% (v/v) DMSO (negative control). At 24, 48, and 72 h time points, cells were washed with PBS and fixed in 4% paraformaldehyde (Scharlau, Spain) for 10 min at room temperature and subsequently permeabilized with 0.05% Triton-X100 (Pancreac AppliChem, Spain-Germany-Italy). DAPI solution (5 mg/mL) was added to each well and incubated for 15 min at room temperature. Wells were rinsed twice with PBS solution and finally visualized by a fluorescence microscope. Activated caspase-3/8/9 immunocytochemistry The immunofluorescence staining procedure was conducted similarly to the DAPI staining experiment (previous section) from the seeding to the permeabilization step. After that, the tested wells were blocked with 10% FBS for 30 min. Following washing of FBS, cells were incubated with anti-cleaved caspase-3 (1:400, v/v; #9661, Cell Signaling, USA), or caspase-8 (1:400, v/v; PA5-99435, Thermo Fisher Scientific, USA), or caspase-9 (1:500, v/v; #20750, Cell Signaling, USA) antibody for 1 h. Subsequently, the non-specific-binding primary antibodies were rinsed with PBS, and the cells were incubated in DAPI (5 mg/mL) solution containing secondary antibody (#A-11008, Invitrogen, USA) for another 1 h. The activated caspase-3 was observed under a fluorescence microscope after washing the secondary antibodies and DAPI co-staining. Senescence-associated-β-galactosidase activity assay NCI-H1299 cells (104 cells/well) were seeded into 96-well plates. After treatment with EMEA (15 μg/mL), doxorubicin (0.05 μg/mL, as the positive control), and DMSO (0.1% v/v, as the negative control) for 24 or 48 h, cells were fixed and stained following the optimized procedure of Debacq-Chainiaux et al.(14). Scratch assay NCI-H1299 cells were seeded in the 24-well plate with a density of 1.5 × 105 cells/well (approximately 90% confluency). Mitomycin C (6 μg/mL; Abcam, UK) was added for 2 h to halt cell growth. Next, the culture medium was discarded, and fresh medium was added to each well. By using a 200 μL sterile tip, scratches were created on the confluent monolayer in each well. The NCI-H1299 cells were exposed to EMEA at concentrations of 5, 10, or 20 μg/mL, or 0.1% DMSO (v/v) (as the negative control group). Images of the scratch’s width were captured at three different fields right after scratching (0 h time point), as well as 24 and 48 h after extract treatment. The areas of the scratch were measured by ImageJ, and the cell migration was reflected via gap closure using the equation below: where, GA stands for gap area. Invasion assay NCI-H1299 cells were first added to 24-well plates at a density of 1.5 × 105 cells/600 μL/well. Cell populations were processed in the same manner as the scratch assay. After 48 h, cells were detached, quantified, and resuspended in serum-free RPMI media (2 × 105 cells/mL). Then, the suspension (200 μL) was added to the Geltrex-precoated (10% v/v; Thermo Fisher Scientific, USA) polycarbonate-membrane (8 mm pore size) transwell. The wells containing the transwell were filled with DMEM-10% FBS as a chemoattractant. Cells were then incubated in 5% CO2 and 37 °C. After 24 h, the non-invasive cells and basement membrane gel were removed with sterilized cotton swabs. The invasive cells were fixed with 70% ethanol (v/v) (Merck KGaA, Germany) for 15 min and stained with 0.5% crystal violet (Sigma-Aldrich, Germany). The stained invasive cells were subsequently observed and quantified by a brightfield microscope. Adhesion assay Collagen type I (C7661-5MG, Sigma-Aldrich, Germany) was coated in the wells of a 96-well plate (10 μg/cm2) following the guidance of the supplier. NCI-H1299 cells were pretreated with EMEA (5, 10, or 20 μg/mL) for 48 h, then trypsinized and suspended in DMEM media. Cells were then allowed to rest in the incubator for 30 min, and 104 cells/100 μL/well suspension were seeded in collagen I precoated plates and cultured for 20 min. Subsequently, the non-specific binding cells were washed twice with PBS, and the adherent cells were fixed in paraformaldehyde (2% (v/v)) for 15 min at room temperature, then stained with Crystal Violet (0.1% (w/v)) for 15 min at room temperature. The attached dye was dissolved in 10% acetic acid, and the OD was measured at 550 nm. The percentages of adherent cells were calculated using normalization so that the untreated group was set to 100%. Analysis of transcription levels NCI-H1299 cells were seeded in 24-well plates at a density of 0.6 or 1.5 × 105 cells/well for cytotoxicity (apoptosis, senescence markers, DNA damage response, cell cycle regulation, oxidative stress response) or metastasis-involved genes, respectively. Subsequently, cells were treated with 20 μg/mL EMEA for analyzing the expression of metastasis-involved genes and 15 μg/mL EMEA for apoptosis, senescence, and oxidative stress response genes. The negative group was exposed to 0.1% DMSO (v/v). After 48 h of treatment, RNA isolation, reverse transcription reaction, and quantitative real-time polymerase chain reaction (qRT-PCR) processes were conducted following the guidance of the kit’s supplier (Vazyme, PRC). Relative mRNA levels were processed by the formula 2-ΔΔCt. The sequences of primers targeting relevant genes in this work are presented in Table S1. ACTB was chosen as the reference gene. Statistical analysis In this work, statistical datasets were calculated by GraphPad Prism 8 software and presented as the mean ± SD of at least three independent experiments. The differences among mean values were assessed by one-way ANOVA following Dunnett’s post-hoc test or two-tailed unpaired Student’s t-test (two-tailed). The P-values ≤ 0.05 were considered statistically significant. RESULTS Cytotoxic effect of EMEA against NCI-H1299 cells To identify the promising EM extracts among the four fractions (EMEtOH, EMPE, EMEA, and EMH2O), the morphology of the treated cells was assessed, and the CI of each extract at 40 μg/mL on NCI-H1299 cells was evaluated. The abnormal morphology (swelling or disrupted cells) of the EMEA-treated group was clearly observed in comparison with cells from the control, as well as the other three treated groups (Fig. 1A). Cytotoxicity of Elephantopus mollis extracts on tested cell lines. (A) The morphology of NCI-H1299 cells after 48 h of exposure to the four Elephantopus mollis extracts (40 μg/mL) in comparison with the respective non-treated group (control). (B) The percentages of NCI-H1299 cell inhibition were also displayed. (C and D) The IC50 values of EMEA were deduced from the dose-response curve of EMEA on NCI-H1299 and BJ-5ta cell lines. Data are presented as the mean ± SD of three independent experiments. EMEtOH, Ethanol extract of Elephantopus mollis; EMPE, petroleum ether extract of Elephantopus mollis; EMEA, ethyl acetate extract of Elephantopus mollis; EMH2O, water extract of Elephantopus mollis. On the one hand, we found that EMEtOH, EMPE, and EMH2O showed fairly low inhibition effects on the NCI-H1299 cell line, with CI ranging from 5.33 to almost 10% (Fig. 1B). On the other hand, the EMEA extract possessed a much higher cytotoxic effect (CI = 63.25%) than the other three fractions. These results prompted us to further elucidate the toxicity of EMEA toward NCI-H1299 cells. By interpolating the dose-response curve (Fig. 1C), the IC50 of EMEA in NCI-H1299 cells was found to be 28.58 μg/mL. We also evaluated the IC50 of EMEA on normal human fibroblast BJ-5ta cells; the IC50 was greater than 104.95 μg/mL (Fig. 1D). The SI value of EMEA on NCI-H1299 was therefore greater than 2 (SI = 3.67). Based on these findings, EMEA was identified as an effective and selective subject against lung cancer NCI-H1299 cells, leading to subsequent experiments. HPLC analysis of polyphenols in EMEA Since EMEA exhibited the strongest cytotoxicity against NCI-H1299, as well as the exclusive presence of tannins in Vietnamese EM(7). We, therefore, initially evaluated several polyphenol-related compounds in this extract via the HPLC method. The semi-quantification (Table 1) revealed that the tannin-related epicatechin and epigallocatechin gallate were the most abundant, with concentrations of 3.68 and 1.49 μg/mg, respectively. Hence, this finding strongly emphasized the consistent and unique presence of tannins in EM from Vietnam, as well as suggesting the potent bioactivity of EMEA extract. Table 1Quantified results of selected polyphenols inCompoundsConcentration (μg/mg of EMEA) Rutin 0.937 Quercetin 0.087 Kaempferol 0.295 Gallic acid 0.059 Epicatechin 3.68 Epigallocatechin gallate 1.49 Epicatechin gallate 0.21EMEA, Ethyl acetate extract of Elephantopus mollis. Quantified results of selected polyphenols in EMEA, Ethyl acetate extract of Elephantopus mollis. EMEA inhibited NCI-H1299 cell growth and altered cellular morphology To assess the mode of action of EMEA on NCI-H1299 cells, we first measured the cell densities and observed the morphological changes of exposed populations. We found that the untreated group showed strong growth from the start till the 72-h time point, while treatment with EMEA remarkably inhibited the cell growth (Fig. 2A). Among the tested EMEA concentrations (15, 20, and 30 μg/mL), the NCI-H1299 cell densities treated with the two lower concentrations remained unchanged throughout the course of tracking. In contrast, populations exposed to 30 μg/mL of EMEA were significantly reduced by half at the end of the experiment (after 72 h). These changes were therefore reflected in the CI of all treated groups, which increased gradually over the experiment time points (Fig. 2B). EMEA affected the NCI-H1299 cell density and morphology. (A) The cell densities (presented as ΔOD550 nm) and (B) the respective percentages of inhibition of NCI-H1299 cells after 24, 48, and 72 h of EMEA (15, 20, and 30 μg/mL) treatment. (C) The notable changes in the morphology of NCI-H1299 cells after treatment with EMEA (15 μg/mL) or doxorubicin, and cells that received no treatment served as the control group. Data are shown as the mean ± SD of three independent. **P ≤ 0.01 and ***P ≤ 0.001 indicate significant differences between the designated groups. Blue arrowheads indicate shrunken and blebbing-membrane cells; red arrowheads indicate enlarged and flattened cells. EMEA, ethyl acetate extract of Elephantopus mollis. In tandem with the negative effects on cell growth, we also observed noticeable abnormalities in morphology. At high concentrations (20 and 30 μg/mL), EMEA caused membrane disruption and cell swelling (hallmarks of necrosis) as early as the 24-h time point (Fig. S1)(15). Conversely, the lower concentration of EMEA (15 μg/mL) induced two different characteristics. Briefly, some NCI-H1299 cells were shrunk and possessed rough or blebbing membranes, but some other cells were notably flattened and enlarged, together with the accumulation of interior vacuoles (Fig. 2C). To our knowledge, the shrunk and blebbing morphology bears a striking similarity to the apoptotic cell death(15). Whilst the enlarged phenotype may indicate the irreversible cell cycle arrest of senescence(16). Fig. 2C also indicates similar cellular morphologies in the NCI-H1299 population treated with doxorubicin at the IC50 value(1.8 μg/mL, interpolated from the dose-response curve in Fig. S2), an anticancer drug that was shown to induce apoptosis and senescence in NCI-H1299 cells(1718). Next, we set out to further analyze the hallmarks of apoptosis and senescence in EMEA-treated NCI-H1299 cells. The effect of EMEA on the appearance of nuclei and caspase-3 activation in NCI-H1299 cells Chromatin condensation and fragmentation are important characteristics of apoptosis(19). Therefore, we observed the alterations of cellular nuclei after 24, 48, and 72 h exposure to EMEA at 15 μg/mL via DAPI staining. At the 24-h time point, there were signs of condensed nuclei in treated groups (Fig. 3A). Subsequently, the fragmented chromatin occurred at 48 h and at 72 h. In addition, the condensed and fragmented nuclei clearly occupied a considerable portion of the populations (Fig. 3A). These atypical nuclei were also associated with cells exhibiting membrane shrinkage and blebbing. The control group, in contrast, displayed no marked changes after 72 h (Fig. 3B). Following that, we investigated the occurrence of activated caspase-3, a crucial trait of apoptosis(15). As was shown in Fig. 4A, the immunocytochemistry staining with anti-activated caspase-3 antibody demonstrated clear signs (i.e., green fluorescence in the EMEA-treated group in contrast with the absence of signal in the control group) after 24 h of EMEA (15 μg/mL) treatment. In addition, the relative numbers of positive cells appeared to rise at the 48- and 72-h time points (Fig. 4B and C, respectively). The effect of EMEA on the nuclear morphology of NCI-H1299 cells. DAPI staining of NCI-H1299 cells (A) with or (B) without treatment with EMEA (15 μg/mL). Red arrowhead indicates an abnormal nucleus (i.e., condensed or fragmented nucleus). Non-treated cells served as the control group. EMEA, ethyl acetate extract of Elephantopus mollis; BF, brightfield images. The effect of EMEA (15 μg /mL) on the activation of caspase-3 in NCI-EH299 cells at (A) 24, (B) 48, and (C) 72 h of treatment. Control: non-treated group. Non-treated cells served as the control group. EMEA, ethyl acetate extract of Elephantopus mollis. The scrutiny of nuclear-cellular morphologies and caspase-3 activation indicated the induction of apoptosis in NCI-H1299 cells by EMEA. Hence, we went on to further examine crucial factors of the apoptotic pathways in this cell line under EMEA influence. The effect of EMEA on the apoptotic pathways in NCI-H1299 cells In this study, we evaluated the activation of caspase-8 and caspase-9, which are specific indicators of the intrinsic and extrinsic apoptotic pathways, respectively(15). Figure 5A shows the absence of cleaved caspase-9 in 15 μg/mL EMEA-treated NCI-H1299 cells. In contrast, Fig. 5B illustrates the activation of caspase-8 in this cell line, highlighting the involvement of the extrinsic apoptosis induction by EMEA. The subsequent gene expression analysis showed a decent increase in the death receptors Fas and TRAIL, as well as an alteration in BCL2 and BAK1 transcription (Fig. 5C) after 48 h of EMEA treatment. Therefore, this cumulative evidence strongly implied the induced apoptosis in NCI-H1299 cells by EMEA. The induced apoptotic pathway in NCI-H1299 cells. EMEA (15 μg/mL) did not activate (A) caspase-9; however triggered (B) caspase-8 activation in the NCI-H1299 cell line after 48 h of treatment. (C) The expression of apoptotic genes was also examined in NCI-H1299 cells after 48 h of EMEA exposure. Non-treated cells served as the control group. Data are shown as the mean ± SD of three independent. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001 indicate the significant differences between the designated groups. EMEA, Ethyl acetate extract of Elephantopus mollis. The effect of EMEA on NCI-H1299 cell senescence As mentioned above, the EMEA fraction (15 μg/mL) also caused senescence-like morphology in the NCI-H1299 population. Therefore, we examined the crucial senescence-associated beta-galactosidase (SA-βga) activity, which is a hallmark and the most commonly examined indicator of senescence(20). Our report found the obvious appearance of blue-stained cells, a hallmark of SA-βga activity (Fig. 6A), in the treated group. Moreover, the quantification of SA-βga- positive cells displayed a stark rise in the EMEA-treated group compared to the control group in both 24 and 48-h time points (3.8- and 5.7-fold increases, respectively) (Fig. 6B). This clearly proved a significantly increased expression of SA-βga by the EMEA effect. EMEA provoked senescence in NCI-H1299 cells. (A) EMEA (15 μg/mL) treatment led to the appearance of blue-stained cells, a hallmark of SA-βga activity; (B) SA-βga activity after 24- and 48-h treatment; (C) altered senescence gene marker expression; (D) as well as causing changes in DNA-damage and cell cycle control genes in NCI-H1299 cells. Non-treated cells served as the control group. Data are shown as the mean ± SD of three independent. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001 indicate the significant differences between the designated groups. EMEA, Ethyl acetate extract of Elephantopus mollis; Dox, doxorubicin. Besides SA-βga expression, we also evaluated the changes in levels of several marker genes of senescence, including IL1A, TIMP1, CCND1, and MKI67 after 48 h of EMEA treatment. Among the tested genes, only the IL1A transcriptional level remained unchanged, while TIMP1 and CCND1 mRNA showed a significant increase, and MKI67 transcription was reduced considerably (Fig. 6C). These changes were relevant to the observed morphology (Fig. 2C) and SA-βga activity recorded in the EMEA-exposed cells (Fig. 6A and B). Hence, we confirmed the senescence induction in a subpopulation of NCI-H1299 cells by EMEA at a concentration of 15 μg/mL. The effect of EMEA on DNA damage and cell cycle arrest-involved gene expressions After confirming the senescence events in EMEA-treated NCI-H1299 cells, we initially elucidated the underlying mechanism of the extract’s effect. In this research, we quantified the relative transcription of genes involved in the DNA damage response and cell cycle regulation, including ATM, ATR, CDKN1A, CDKN1B, and RB1. Results presented in Fig. 6D indicate that under the influence of EMEA, the transcription of DNA- damage sensor ATR and cDkN1A (p21) increased significantly, in contrast to the reduced double-strand-sensor ATM transcription after 48 h. This finding implied that DNA damage is likely the cause of the cytotoxic effect of EMEA (15 μg/mL) on the NCI-H1299 cell line. The effect of EMEA on the expression of genes involved in oxidative stress Besides DNA damage and cell cycle regulation, we also elucidated the transcription of genes involved in redox balance and oxidative stress in EMEA-treated NCI-H1299 cells. Our findings (Fig. 7) showed that all the tested genes were significantly regulated after 48 h of extract exposure. This indicated that N CI-H1299 cells might experience oxidative stress caused by EMEA. EMEA (15 μg/mL) activated the expression of oxidative stress response genes in NCI-H1299 cells. Non-treated cells served as the control group. Data are shown as the mean ± SD of three independent. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001 indicate the significant differences between the designated groups. EMEA, Ethyl acetate extract of Elephantopus mollis. The effect of EMEA on the metastasis of NCI-H1299 cells Identification of suitable conditions for metastasis-involved experiments This study utilized the scratch assay to inspect the in vitro mobility of NCI-H1299 cells. Either the proliferation or the death of cells can interfere with the migration ability assessment. Hence, preliminary experiments were conducted to identify the conditions that ensured cell growth hindrance, as well as allowed cells to migrate effectively. We found that the optimal condition for assessing the migration ability of NCI-H1299 cells is culture medium RPMI supplemented with 10% FBS and 6 μg/mL Mitomycin C (Fig. S3 and Table S2). Next, we also identified the EMEA concentration threshold, which did not significantly affect NCI-H1299 cell morphology and density (Fig. S4). The EMEA concentrations that are equal to or lower than 20 μg/mL were found to be suitable for NCI-H1299 cells in metastatic-involved experiments. EMEA impeded NCI-H1299 migratory activity Therefore, in this work, we evaluated the effects of 5, 10, and 20 μg/mL EMEA on the migration, invasion, and adhesion of NCI-H1299 cells. As shown in Fig. 8A, cellular migration followed a downward trend as the tested EMEA concentrations rose. Moreover, this trend was sustained at both 24 and 48-h time points and was most prominent in groups treated with the highest EMEA concentration (20 μg/mL). Thus, our result initially indicated the potent anti-metastatic property of EMEA alongside cytotoxicity. EMEA inhibited metastatic traits of NCI-H1299 cells. The (A) migration, (B) invasion, (C) collagen I adhesion abilities, as well as (D) the expressions of crucial metastatic genes of NCI-H1299 cells were significantly decreased after EMEA (20 μg/mL) treatment. Non-treated cells served as the control group (0 concentration of EMEA in parts A-C). Data are shown as the mean ± SD of three independent measurements. *P ≤ 0.05, **P ≤ 0.01, and ***P ≤ 0.001 indicate significant differences in comparison with the control group; ###P ≤ 0.001 indicate the significant differences between the designated groups. EMEA, Ethyl acetate extract of Elephantopus mollis. EMEA reduced the invasiveness and collagen I-adhesion capacity of NCI-H1299 cells The subsequent crucial step of cancerous metastasis is invasion through the basement membrane to infiltrate surrounding tissues. In addition, cancer cells possess considerable affinity toward extracellular matrix components. Therefore, we elucidated these two abilities by harnessing the GeltrexTM-coating transwell assay and collagen I-adhesion method, respectively. Under the effect of EMEA, the invasiveness of NCI-H1299 cells was significantly reduced at all treated concentrations (Fig. 8B), with percentages of invasive cells gradually shrinking from 75.4% (5 μg/mL EMEA) to 28.6% at 20 μg/mL EmEa. The adhesion ability, although exhibiting a downward trend, only showed a marked reduction (approximately 50%) in groups exposed to the highest EMEA concentration (20 μg/mL) after a 48-h treatment (Fig. 8C). The effect of EMEA on the expression of migration and invasion involving genes After firmly identifying the effects of EMEA on metastatic traits, we initially examined the underlying anti-migration and anti-invasion mechanisms of EMEA by assessing changes in the expression of several key genes, including MMP2, VIM, TWIST1, and ZEB2 (Fig. 8). The results in Fig. 8D indicated sizeable suppression of the transcriptions of the metalloproteinase MMP2 and the intermediate filament Vimentin, with mRNA levels reduced by more than half. This finding was therefore relevant to the respective decrease in migration, invasion, and adhesion above. DISCUSSION Elephantopus mollis (EM) has been a continuously promising source of anticancer research since the beginning of the 80s. With the undeniable threat of lung cancer on health systems worldwide, assessing this scientifically well-established candidate is a reasonable strategy for novel anti-lung-cancer subjects. The screening result obviously indicated that the potent active compound(s) against NCI-H1299 cells mostly accumulated in the ethyl acetate fraction (EMEA). Subsequently, the IC50 of EMEA on NCI-H1299 in this work (28.58 μg/mL) was relatively close to the values of EM extracts (ethyl acetate and methanol fraction) on another lung cancer cell line (A549) reported in previous publications(78). Moreover, the guideline of the US National Cancer Institute (NCI), which stated that the concentration for 50% inhibition (IC50) of a potential anticancer crude extract should be equal to or less than 30 μg/mL on the sensitive cell line(21). This implies that EMEA, evaluated in this study, is a strong anti-lung cancer subject. Besides a low IC50 value, a potential anticancer candidate should be significantly less toxic to normal cells than cancer cells. Hence, the selectivity index (SI) of 2 or higher indicates a selective anticancer agent(7). This study’s findings underscore the consistent presence of active compounds within our EM extract, aligning with prior research. Furthermore, the obtained results highlight the high and selective cytotoxicity of EM phytoconstituents against lung cancer cells. The HPLC analysis results specifically proved the significant presence of several tannin-related compounds in this extract. Favorably, colleagues have identified the potent anticancer effects of tannins(22). Therefore, our findings were not only in agreement with the unique discovery of tannins in EM from Vietnam(7) but also implied its promising toxicity against NCI-H1299 cells. Subsequent assessments revealed the induction of both apoptosis and senescence in NCI-H1299 cells by EMEA (15 μg/mL). Since these specific and controlled cell fates are major mechanisms of many prescribed chemotherapies(720), our findings strongly expanded and cemented the potential of the EM’s anticancer constituents. The cysteine-aspartic protease (caspase) family plays a vital role in the initiation and execution of both the intrinsic and extrinsic pathways. Caspase-3, after being activated by initiator caspases 8/9, will directly or indirectly cause DNA fragmentation, cross-link or breakdown of nuclear and cytoplasmic proteins, and induce phagocytosis(2324). Our results, hence, indicated apoptosis in EMEA-treated NCI-H1299 cells. Furthermore, the caspase-8 cleavage, together with increased expression of Fas and TRAIL receptors, clearly indicated the activation of extrinsic apoptosis generated by EMEA. Moreover, the catalytically active caspase-8 not only switches on caspase-3 but also can cleave the Bid protein, which subsequently activates pro-apoptotic proteins Bak, Bax, and suppresses anti-apoptotic protein Bcl-2(25). Although these proteins belong to the intrinsic apoptosis pathway, in some circumstances, they may be involved in the release of Endo G from mitochondria, leading to caspase-9-independent apoptosis(26). If this is the case, it may explain why the increase in BAK1 and reduction in BCL2 genes didn’t cause caspase-9 activation (Fig. 5). Interestingly, our findings are of great relevance to some colleagues’ research. Ooi et al. reported that the ethyl acetate fraction of Malaysian EM strongly inhibited the liver cancer HepG2 cells via both extrinsic and mitochondrial apoptotic pathways in the lung cancer A549 cells(9). Further phytochemical analysis and bioactivity studies have found several promising compounds from EM. Ooi et al. proved that the 3,4-di-O caffeoylquinic acid from EM significantly suppressed lung cancer NCI-H23 cells via the apoptosis mechanism(27). Sesquiterpene lactones, however, are the specific and most attractive phytochemical group in EM. A prominent example is EM-23, which was found in the ethyl acetate extract(28), induced apoptosis in leukemia cells (K562 and HL-60) as well as cervical carcinomas SiHa and CaSki lines. The underlying modes of action were found to involve multiple signaling pathways, including the JNK-p38/MAPK activation, TNF-α-dependent NF-κB activation, and Akt/mTOR suppression(1112). If this is the case, these substances might be present in our EMEA and help explain the consistent apoptosis observed in our NCI-H1299 cells and others’ studies. Besides apoptosis, we also found induced senescence in EMEA (15 μg/mL)-treated NCI-H1299 cells based on the SA-βga assay and marker genes assessment (Fig. 6). Senescence is a stable and irreversible proliferation arrest, even after the removal of inhibitors(16). With the mentioned characteristics, it is obvious that the senescent state is a complete opposition to the immortality and uncontrolled growth of cancer. In fact, apart from apoptosis, induced senescence and the subsequent selective elimination of senescent cells (i.e., senolysis) is increasingly considered a reasonable anticancer strategy(20). Therefore, the observed senescence effect of EMEA on NCI-H1299 presents a notable advancement due to several key factors explained as follows: First, although there are numerous published anticancer reports of EM, only one regarded the senescence induction of EM constituents (the petroleum extract of EM collected in Vietnam)(10). A previous study showed that saponins and tannins were only found in Vietnamese EM(7). We, therefore, implied that the exclusive senescence state of cancer cells induced by EM from Vietnam might involve these two metabolite groups. Moreover, an accumulated body of evidence supports this proposition. For instance, Saikosaponin A from Radix Bupleurum caused senescence in breast cancer cells by inhibiting the PI3K/Akt signaling pathway(29), or gallotannin from green tea, transformed the colon cancer HCT116 cells into the senescent state via ROS production(30). Second, and most importantly, whilst there are separate reports on the apoptosis and senescence induction effects of EM constituents on several cancers, this is the first work regarding two different cytotoxic effects of EM phytochemicals on a particular cancer cell population. The dual effect of provoking apoptosis and senescence was identified in various anticancer agents; however, many cases were found to be dose-dependent(313233). The primary explanation for the induced senescence and apoptosis is the different degrees of damage caused by mild and lethal doses of anticancer subjects, respectively. In contrast, our results demonstrated that the concentration of 15 μg/mL EMEA caused both effects on the NCI-H1299 cell population. We assumed that the cell cycle phases and single-cell growth rate might play a decisive role in this choice of fate. In fact, the G1/S checkpoint was found to be more sensitive to senescence than other stages(34). In addition, the rates of metabolism and exchange of substances with the environment greatly fluctuate throughout the cell cycle phases(3536). This could affect the uptake and interaction of EM’s active compounds with NCI-H1299 cells, leading to different cellular responses. Finally, the proliferation status of single cells in a population could determine either apoptosis or senescence choice under the same stressor exposure(37). In fact, the IGFBP7 reagent and temozolomide, like our EMEA, showed the one-dose-two-fate effect on their target cells(3839). Another example is the renowned doxorubicin drug, which induced senescence or mitotic catastrophe, which could eventually lead to apoptosis, in liver cancer Huh-7 cells(40). In tandem with the cytotoxicity, the expression changes of ATM, ATR, and CDKN1A (p21) genes implied that the DNA damage caused by EMEA had upregulated the CDK inhibitor p21, which then initiated cell cycle arrest and senescence(20). Since redox imbalance and oxidative stress are closely related to apoptosis, senescence, and DNA damage, the subsequent redox gene analysis was therefore relevant. The oxidative stress involved regulators cJUN, NFκB, and NRF2, as well as the antioxidative enzymes Trx, TrxR, GR, and GRx, which are crucial factors in this aspect(414243). The stark transcriptional increase of these proteins helped explain the previous apoptosis, senescence, and DNA damage identified in EMEA-treated NCI-H1299 cells in the current study. The sesquiterpene lactone EM23 found in EM was a strong inducer of ROS as well as a modulator of the thioredoxin reductase in cancer cells; hence, the finding in our work was highly reasonable(1112). Although being a remarkable anticancer candidate for decades, except for a report on the anti-invasion of the petroleum ether extract on breast cancer MDA-MB-231 cells, the anti-metastasis property of EM is largely uncharted(10). Notably, more than 90% of cancer deaths stem from metastasis(10). Hence, we also elucidated this property of EMEA in NCI-H1299 cells in this study. The findings of the current study indicate that EMEA strongly inhibited migration, invasion, and collagen I adhesion of NCI-H1299 cells, which are all crucial aspects of the metastasis process. This was an attractive and promising but not unexpected finding since several compounds or phytoconstituents of EM and the Elephantopus family possess antimetastatic ability. The polyphenol 3,4-di-O-caffeoylquinic acid mentioned earlier can halt the PI3K/Akt and JNK cascades, leading to the migration inhibition(44). Other notable examples include the sesquiterpene lactones deoxyelephantopin, scabertopin, and WBDC-1, which strongly suppressed metastatic activities of melanoma, lung, bladder, and liver cancer cells, respectively(454647). Although these sesquiterpene lactones have been found in other Elephantopus species, they share the same pharmacophore a-methylene-y-lactone with the EM-23, EM-12, and EM-5 compounds from EM, suggesting an akin structure-function relationship among these compounds(124849). The subsequent gene expression analysis revealed decreased expression of MMP2 and VIM, which was reasonable since MMP2 facilitates extracellular matrix degradation during cancer invasion, whilst vimentin is vital in protecting cellular organelles during migration and invasion(5051). CONCLUSION Lung cancer is among the most threatening maladies worldwide. With traditional medicine practice and decades of accumulated knowledge, EM is a rising candidate for novel anticancer therapy. Here in this study, we proved the robust and selective toxicity of EMEA against the non-small cell lung cancer, NCI-H1299 line. The nature of this cytotoxic effect is apoptosis and senescence induction, which was not reported in any EM study. Furthermore, DNA damage and oxidative stress might be the source of EMEA anticancer effects. In addition, the metastasis-associated traits of NCI-H1299 cells were also substantially hindered by EMEA. Therefore, our work plays a significant role in expanding the knowledge of EM’s anticancer ability as well as highlighting the importance of traditional medicine experiences. Finally, although this study involved an in vitro model, in vivo research is expected to be conducted in further work to assess the therapeutic potential of EM extracts. Authors’ contributions TT Long and NTM Trinh designed the study. TT Long, DM Phuc, and PD Khoi performed the experiments. TT Long, DM Phuc, and PD Khoi analyzed the data. TT Long and NTM Trinh, DTP Thao performed interpretations of the data. TT Long wrote and revised the manuscript. NTM Trinh had primary responsibility for the final content. All authors have read and approved the finalized article. Each author has fulfilled the authorship criteria and affirmed that this article represents honest and original work. Conflict of interest statement All authors declared no conflict of interest in this study. AI declaration AI-assisted design tools were employed for the preparation of the graphical abstract, followed by manual refinement by the authors. Data availability The data supporting the findings of this study are documented and available from the corresponding author upon reasonable request. Researchers interested in accessing the data should contact the corresponding author to discuss the terms and conditions of access. Supplementary materials The supplementary material was uploaded and archived on https://github.com/Long-Tran-hcmus/Ethyl-acetate-extract-of-Elephantopus-mollis-Kunth.