Isodeoxyelephantopin, a Sesquiterpene Lactone Induces ROS Generation, Suppresses NF-κB Activation, Modulates LncRNA Expression and Exhibit Activities Against Breast Cancer

The sesquiterpene lactones, Isodeoxyelephantopin (IDET) and Deoxyelephantopin (DET) are known to exhibit activities against some cancer types. The activities of these lactones against breast cancer and the molecular bases is not known. We examined the efficacy of lactones in breast cancer preclinical model. Although both lactones exhibited drug like properties, IDET was relatively effective in comparison to DET. IDET suppressed the proliferation of both invasive and non-invasive breast cancer cell lines. IDET also suppressed the colony formation and migration of breast cancer cells. The assays for Acridine Orange (AO)/Propidium Iodide (PI) staining, cell cycle distribution, phosphatidylserine externalization and DNA laddering suggested the apoptosis inducing potential of IDET. The treatment with IDET also induced an accumulation of cells in the sub-G1 and G2/M phases. The exposure of breast cancer cells to the lactone was associated with a depolarization in mitochondrial membrane potential, and cleavage of caspase and PARP. The lactone induced reactive oxygen species (ROS) generation in breast cancer cells. Further, the use of N-acetyl cysteine (NAC) suppressed IDET induced ROS generation and apoptosis. The NF-κB-p65 nuclear translocation induced by okadaic acid (OA) was suppressed by the sesquiterpene. IDET also suppressed the expression of NF-κB regulated tumorigenic proteins, and induced the expression of proapoptotic gene (Bax) in cancer cells. While the expression of oncogenic lncRNAs was suppressed, the tumor suppressor lncRNAs were induced by the sesquiterpene. Collectively, the modulation of multiple cell signaling molecules by IDET may contribute to its activities in breast cancer cells.

Cell cycle analysis. The cells were stained with PI to examine if IDET affects different cell cycle phases. After treatment with various concentrations of IDET, the cells were washed with PBS and fixed with 70% chilled methanol. The RNaseA was used for the treatment of cells followed by staining with PI. We used flow cytometer to assess the percentage of cells and the Cell Quest software (Becton Dickinson) for the analysis.
Measurement of mitochondrial membrane potential (ΔΨ). The effects of IDET on mitochondrial membrane potential was examined using a previously described method 26 . In brief, cells were exposed to 10-50 μM IDET, washed and incubated in the dark (at 37 °C for 20 minutes) with 10 μg/mL JC-1. Cells were then washed and imaged under fluorescence microscopy. Whereas green fluorescence is an indicator of depolarized mitochondria, intact mitochondria produce red fluorescence.
Western blot analysis. The western blot analysis was performed to examine the effects of IDET on the expression of tumorigenic proteins 27 . Briefly, the whole cell lysate was prepared from normal and IDET treated cells. After separation on the SDS-PAGE and transferring onto nitrocellulose membrane, the proteins were probed with primary and secondary antibodies. Finally, the ECL reagent was used for the detection of the protein signals.
Immunocytochemistry for the NF-κB p65 cellular localization. For this, we used a previously described method 27 . Briefly, paraformaldehyde and PBST were used for the fixing and permeabilization of the cells, respectively. After probing with antibodies (primary and secondary) and counterstaining with DAPI, the cells were imaged under fluorescence microscope.
Cell migration assay. Whether IDET affects cell migration was examined by scratch (wound healing) assay 28 . Briefly, at the 70% confluency, the monolayer cells were wounded with a sterile culture tip. After washing the debris, IDET was applied over cells. The wounded area was examined at 0, 9, 24 and 48 hrs by phase contrast microscope. The image J software was used to calculate the healed area and wound size at each time point.
Estimation for ROS generation. The potential of IDET to generate ROS in breast cancer cells was examined by flow cytometry 29 . Briefly, the control and treated cells were stained with 10 µM H2DCFDA for 1 hr in the dark. We used flow cytometry to examine the stained cells and Cell Quest software (Becton Dickinson) for the data analysis.
Semi-quantitative and quantitative RT-PCR. We performed semiquantitative RT-PCR to examine the IDET's effects on the expression of mRNA transcripts of cyclin D1, survivin and Bax. The quantitative real-time PCR was performed to examine the IDET's effects on the lncRNA expression 30 . Table 1 lists the primer sequences used for the amplification of the gene of interest.
The trizol reagent was used for the isolation of total RNA by following the manufacturer's instructions (Invitrogen). The high capacity cDNA synthesis kit was used for reverse transcription.  Table 1. The sequences of the primers used in the semi-quantitative and quantitative RT-PCR.
We used 1.5% agarose gel for the electrophoresis of the PCR product. The densitometry and ImageJ software were used for the quantification of DNA bands. The PCR product for the gene of interest were normalized to GAPDH. The Maxima SYBR Green/ROX qPCR Master Mix and Applied Biosystems 7500 Real-Time system was used for the quantitative real-time PCR analyses of lncRNA expression. The data analysis was performed using a method as described previously 31 . The values for house-keeping genes (ACTB and 5SrRNA) were used for the normalization of the data.
In silico analysis. The drug like properties of IDET (Pub Chem ID: 38359583) and DET (Pub Chem ID: 6325056) were examined by analyzing Lipinski's rule of five (http://www.molinspiration.com//cgi-bin/properties) and ADMET (absorption, distribution, metabolism, excretion and toxicity) 32 . SMILE IDs of IDET and DET were obtained from Pub Chem database. CORINA 3D server was used to convert the SMILE ID to .pdb files. PDB ID 1NFI was used to procure 3D structure of NF-κB-p65 (Chain A), NF-κB-p50 (Chain B) and IκBα (Chain E). 2EVA and 5TQY was used to procure TAK-1 and IKKα, respectively. Auto Dock Tool 4 was used for the identification of binding affinities and poses of ligands and proteins 33,34 . Statistical analysis. Different end points were performed in control and treated groups. We performed the unpaired Student's t-test for the comparison between the two groups. Statistical significance was calculated at a value of P < 0.05.

Results
In this study, we examined the relative potency of IDET and DET in breast cancer cells (Fig. 1A). Preliminary experiments were performed with both IDET and DET. However, most experiments were performed with IDET. We examined the anti-tumorigenic and anti-inflammatory activities of IDET. Because of availability of multiple variants, MDA-MB-231 was used for most experiments. We also used other cell lines such as T-47D and MCF-7 to examine the IDET's specificity. The underlying mechanism for the anti-carcinogenic activities of IDET was examined.

iDet exhibit stronger anti-proliferative activities in breast cancer cells as compared to Det.
First, the relative anti-proliferative activities of IDET and DET was examined. We exposed MDA-MB-231 cells to 1-100 µM IDET and DET for 12-72 hrs. The mitochondrial reductase activity was measured using the MTT substrate. The anti-proliferative activities of IDET was stronger as compared to DET (Fig. 1B). For example, the cell proliferation was suppressed by 36% when cells were exposed to 10 µM DET for 72 hrs (Fig. 1C). However, the cell proliferation was suppressed by 50% after exposure of cells to 10 µM IDET for 72 hrs. Similarly, a respective 10% and 27% suppression in cell proliferation was observed after exposure of cells for 12 hrs to 25 µM DET and 25 µM IDET, respectively.
Whether IDET and DET possess drug like properties was examined by in silico tools. Using Lipinski's rule of five and ADMET analysis, we found that both IDET and DET exhibited similar characteristics ( Table 2). The predicted lipophilicity (log P), topological polar surface area, molecular weight, hydrogen bond acceptor, hydrogen bond donor, and rotatable bonds were found to be 2.25, 78.92, 344.36, 6, 0, and 3, respectively. Furthermore, both IDET and DET were permeable to blood brain barrier and intestine without any evidence of carcinogenic and genotoxic effects.
Because the anti-proliferative activities of IDET was relatively stronger as compared to DET, we used IDET for most of the other experiments.
iDet suppresses the proliferation of multiple breast cancer cells. Whether the anti-proliferative activities of IDET is specific to one cell line was examined. For this, we used multiple breast cancer cell lines. T47D, MCF-7, MDA-MB-468 and MDA-MB-453 cells were treated with 1-100 µM IDET for 72 hrs. The cell viability was decreased in a dose dependent manner ( Fig. 2A). Similarly, when cells were exposed to 25 µM IDET for 24-72 hrs, the viability of cells was suppressed in a time dependent manner (Fig. 2B). These observations suggest that the effects of IDET is not cell-type specific.
iDet suppresses the colony formation of breast cancer cells. Whether IDET affects the colony formation of breast cancer cells was examined. We exposed MCF-7 (Fig. 3A) and MDA-MB-231 (Fig. 3B) cells to 1-25 µM IDET for 24 hrs. The IDET was then washed off and the colony formation was examined after 7 days. At a concentration as low as 2.5 µM IDET, a drastic decrease in the colony formation was observed.
Breast cancer cells are sensitized to doxorubicin by iDet. Doxorubicin is a commonly used chemotherapy for breast cancer. However, patients develop resistance over time. Whether the sensitivity of breast cancer cells to doxorubicin can be enhanced by IDET was examined. We exposed MCF-7 cells to different concentrations of IDET before treatment with doxorubicin. Both IDET and doxorubicin suppressed the viability of cancer cells (Fig. 3C). However, when cells were pretreated with IDET, the sensitivity of cells to doxorubicin was significantly increased. For example, at 0.25 µM doxorubicin, 7% reduction in the viability of cells was observed. However, the pretreatment of cells with 1 µM and 10 µM IDET before 0.25 µM doxorubicin reduced the viability by 17% and 58%, respectively. iDet induces apoptosis in breast cancer cells. One possibility for the reduction in the viability of cells after IDET treatment may be due to induction of apoptosis. A variety of assays were performed to examine the apoptosis inducing potential of IDET. First, AO/PI dual staining was carried out to accurately determine the cell viability. After dual staining, the live nucleated cells fluoresce green and the dead nucleated cells fluoresce red. An increase in the IDET concentration was associated with a decrease in the viability of MCF-7 cells and increased (2019) 9:17980 | https://doi.org/10.1038/s41598-019-52971-3 www.nature.com/scientificreports www.nature.com/scientificreports/ number of dead cells (Fig. 4A, left). A similar trend in the viability of MDA-MB-231 cells was observed after treatment with IDET ( Fig. 4A, right). The membrane blebbing and nuclear condensation, which are characteristics of early apoptosis was also observed after IDET treatment. Next, we examined the effects of lactone on the cell cycle distribution by flow cytometry. While the percentage of cells in the sub-G1 and G2/M phase was increased, a decrease in the population of cells in the G1 and S phase was observed after IDET treatment (Fig. 4B). For example, 3.5 folds increase in the sub-G1 population was observed at 25 µM IDET in comparison to control. Similarly, www.nature.com/scientificreports www.nature.com/scientificreports/ 2 folds increase in the G2/M phase was observed at 25 µM IDET in comparison to control. Overall, these results suggest that IDET induces cell cycle arrest at sub-G1 and G2/M phase in MDA-MB-231 cells. A key feature of early apoptosis is the PS externalization from the inner surface to the outer surface of plasma membrane that disrupts the membrane symmetry. While Annexin V has a high affinity for PS, PI binds to DNA 35 . Thus, dual staining with Annexin V and PI can be used to distinguish cells undergoing early apoptosis and late apoptosis. In control group, 1.7% cells were stained with Annexin V. However, 13.1% Annexin V positive cells were observed at 10 µM IDET (Fig. 4C). The late stages of apoptosis are associated with cleavage of DNA into 180-200 base pair fragments known as DNA ladders. Exposure of cells to IDET induced DNA laddering in a concentration dependent manner (Fig. 4D). Next, we examined if IDET can modulate the expression of tumorigenic proteins. The expression of anti-apoptotic (Bcl-xL, Bcl-2) and invasive (MMP-9) proteins was significantly suppressed while an induction in caspase and PARP cleavage was observed by IDET (Fig. 4E). The lactone also suppressed the expression of mRNA transcript of genes involved in cell survival (survivin) and proliferation (cyclin D1) (Fig. 4F)   www.nature.com/scientificreports www.nature.com/scientificreports/ expression of mRNA transcript of proapoptotic Bax was also induced by IDET (Fig. 4F). Collectively, IDET can induce apoptosis in breast cancer cells.
iDet disrupts mitochondrial membrane potential and induces RoS generation in breast cancer cells.
We used the fluorochrome JC-1 to examine if IDET induced apoptosis in breast cancer cells require mitochondria. www.nature.com/scientificreports www.nature.com/scientificreports/ www.nature.com/scientificreports www.nature.com/scientificreports/ In control cells with intact mitochondria, the fluorochrome produces red fluorescence. However, with the depolarization of mitochondria and the reduction of the mitochondrial membrane potential, the intensity of the green fluorescence is increased. The staining of control cells with JC-1 produced prominent red fluorescence and minimal green fluorescence (Fig. 5A). However, the treatment of cells with IDET produced a reduction in the red www.nature.com/scientificreports www.nature.com/scientificreports/ fluorescence and an increase in the green fluorescence (Fig. 5A). Overall, these results suggest that IDET induces depolarization in mitochondrial membrane potential.
Whether IDET can induce ROS generation in breast cancer cells was examined. MDA-MB-231 cells were exposed to 10 µM IDET for 1 hr and ROS generation was examined by staining the cells with H2DCFDA. As shown in Fig. 5B, the treatment of cells with 10 µM IDET produced 2.8 folds increase in ROS generation. Furthermore, the use of NAC almost completely suppressed the ROS generation induced by IDET. Similarly, the PS externalization induced by IDET was also suppressed by the use of NAC (Fig. 5C). Overall, these results suggest that IDET can induce ROS generation, which is required for apoptosis induction in breast cancer cells.
iDet reduces the migration of breast cancer cells. We examined if IDET can suppress the motility of breast cancer cells which is required for the invasion and metastasis. MDA-MB-231 cells were wounded at 70% confluency and cultured in the presence of IDET for 9-48 hrs. The wound area after 48 hrs in control, 1 µM and 2.5 µM IDET groups was found to be 14%, 28%, and 40%, respectively (Fig. 6A). Similarly, the wound area was significantly healed in the control cells over time. However, IDET significantly suppressed the healing potential of cancer cells. For example, the healed area after 48 hrs was found to be 85%, 71%, and 59%, in the control, 1 µM and 2.5 µM IDET groups, respectively. In conclusion, IDET can suppress the motility of breast cancer cells. www.nature.com/scientificreports www.nature.com/scientificreports/ IDET inhibits NF-κB activation and interacts with NF-κB associated proteins. The pro-inflammatory transcription factor, NF-κB plays a crucial role in the survival, proliferation, migration and chemoresistance of breast cancer cells. Whether, the lactone can reverse the NF-κB activation induced by okadaic acid (OA) was examined. We treated MDA-MB-231 cells either with OA and IDET alone or with IDET followed by OA. The NF-κB p65 cellular localization was examined by immunocytochemistry. An induction in the p65 nuclear translocation was observed after the treatment of cells with OA. IDET alone did not affect the localization of p65. However, by pretreatment of cells with IDET, OA induced p65 nuclear translocation was significantly suppressed.
iDet modulates the lncRnAs expression in breast cancer cells. The lncRNAs are known to modulate multiple steps of tumor development. Some lncRNAs such as NKILA and H19 can also cross talk with NF-κB. Whether the expression of lncRNAs is modulated by IDET was examined. The lactone up-regulated the expression of growth arrest specific 5 (GAS5), NKILA, ANRIL, tumor necrosis factor α-induced protein 3 (lincRNA-Tnfaip3) and HOTAIR in a concentration dependent manner (Fig. 8). For example, the expression of GAS5 was induced by 1.3 folds, 2.3 folds and 3.2 folds at 1, 2.5 and 5 µM IDET, respectively. Similarly, a respective 4.7 folds, 5.1 folds and 7.5 folds increase in NKILA expression was observed by 1, 2.5 and 5 µM IDET, respectively. Conversely, H19 expression was reduced by IDET in a concentration dependent manner. Collectively, IDET can modulate the lncRNAs expression pattern in breast cancer cells.

Discussion
The multigenic breast cancer is the leading cause of cancer associated deaths in women globally. The pro-inflammatory NF-κB is known to regulate multi-steps of breast tumor development and chemoresistance. Therefore, the suppression of NF-κB activation can prevent or delay the onset of breast tumor development. The Mother Nature has been a gold mine for the discovery of anti-cancer agents. Roughly 50% of the anticancer agents approved between 1940 and 2014 were derived from Nature. In this study, we examined the activities of sesquiterpene lactones such as IDET and DET against breast cancer. The molecular mechanism by which the lactones exhibit activities was also examined.
IDET was more effective in reducing the proliferation of breast cancer cells as compared to DET. Both the lactones contain α-methylene-γ-lactone which can contribute to their cytotoxic activities 36   www.nature.com/scientificreports www.nature.com/scientificreports/ oxygen atom at C-2 is in β-orientation and α-orientation in DET and IDET, respectively. The presence of C 11 -C 13 exocyclic methylene in conjugation with γ-lactone can contribute to the cytotoxicity of both IDET and DET. In agreement with these observations, Helenain, a sesquiterpene lactone with reactive α, β-unsubstituted cyclopentenone ring has been reported to exhibit potent cytotoxicity in cancer cell lines 37 . Why IDET is more potent than DET remains to be elucidated.
Apoptosis and necrosis are two common modes of cell death. While apoptosis is programmed mode of cell death, necrosis may result in inflammation which is a precursor to several chronic diseases including cancer 38 . Therefore, agents with potential to selectively induce apoptosis in cancer cells are preferred 39 . The presence of membrane blebbing, nuclear condensation, phosphatidylserine externalization and DNA laddering by IDET suggest the potential of this lactone in inducing apoptosis in breast cancer cells. We observed that IDET suppresses the expression of cell survival (Bcl-xL, Bcl-2) and invasive (MMP-9) proteins. Furthermore, an induction in caspase cleavage was observed by IDET. Additionally, the lactone reduces the expression of mRNA transcripts www.nature.com/scientificreports www.nature.com/scientificreports/ of genes involved in survival (survivin) and proliferation (cyclin D1). The lactone also induces the expression of proapoptotic gene, Bax. The early phases of apoptosis is associated with phosphatidylserine externalization, disruption of mitochondrial membrane potential (ΔΨ), insertion of proapoptotic proteins into the membrane and the cytochrome c release from mitochondria to the cytoplasm 40 . The ability of IDET to induce loss in ΔΨ suggest that the mitochondria is involved in the apoptosis induction by this lactone. NF-κB is known to regulate the expression of over 500 tumorigenic genes and proteins. The inhibitory effects of IDET on the tumorigenic genes and proteins may be due to the negative regulation of NF-κB activation pathway by this lactone. The mechanism of NF-κB inactivation by IDET in breast cancer cells was not examined. However, a previous study demonstrated the inhibitory effects of IDET on IKK, a central kinase in the NF-κB signaling pathway 22 . It is likely that the reduction in the NF-κB activity by IDET in breast cancer cells is due to its inhibitory effects on IKK activity.
Cancer cells are normally characterized by dysregulation in cell cycle 41 . Thus, targeting cell cycle could be a potential strategy for breast cancer therapy. We observed an accumulation of cells in the G2/M phase. Similar to these observations, IDET has been demonstrated to induce G2/M cell cycle arrest in A549 lung carcinoma cells 42 . Previous studies have demonstrated that Cdc2/Cyclin B1 complex regulate G2/M cell cycle arrest under oxidative stress 43 . IDET was found to induce ROS generation in breast cancer cells that was reversed by NAC. The chemotherapeutic agents and phytochemicals work through generation of ROS 44 . The mechanistic association between ROS generation and G2/M cell cycle arrest by IDET remains to be elucidated.
The cancer cell migration is an important step prior to the invasion and metastasis. The majority of the breast cancer deaths are because of the potential of the tumor to metastasize to other organs. In our observations, IDET significantly reduced the migration of MDA-MB-231 cells. Consistence with these observations, the expression of MMP-9, an invasive protein regulated by NF-κB was also suppressed. The reduction of NF-κB activation by IDET may be responsible for the suppression in MMP-9 expression and the inhibition of migration of MDA-MB-231 cells.
Dysregulation in the lncRNAs expression plays a crucial role in several human malignancies including breast cancer 45 . An increase in the GAS5 and NKILA expression was observed after IDET treatment. Originally identified in breast cancer and located in the cytoplasm, NKILA masks the phosphorylation motifs of IκB by interacting with p65-IκB subunits thereby suppressing NF-κB activation 46 . While NKILA is abundantly expressed in the normal breast epithelia, its low expression associates with breast cancer metastasis and poor patient prognosis. Further, the gene silencing of NKILA results in significant phosphorylation and degradation of IκB leading to enhanced NF-κB activation 46 . An up-regulation in NKILA expression by IDET may suppress IκB phosphorylation and NF-κB activation. Similar to these observations, the anti-invasive activities of TGF-β 47 and anti-carcinogenic activities of baicalein 48 are mediated through NKILA. Figure 8. The lncRNA expression in breast cancer cells is modulated by IDET. MDA-MB-231 cells were exposed to indicated concentrations of IDET for 24 hrs. RNA was isolated, cDNA was synthesized and the lncRNAs expression pattern was assessed by quantitative RT-PCR. The oncogenic H19 expression was reduced while the tumor suppressive lncRNAs (GAS5, NKILA) expression was increased. *Shows the significance level in comparison to untreated group; P < 0.05.
GAS5 is known to function as tumor suppressor in a number of cancer types 49 . This lncRNA can also induce apoptosis and reduce the tumor cells proliferation 50 . GAS5 is frequently decreased in the breast cancer tissues as compared to the adjacent non-tumor tissues 51 . The chemo-resistant breast cancer cells also exhibit significantly lower expression of GAS5 51 . The development of chemoresistance by tumor cells is a major roadblock to cancer therapy. In our observations, IDET enhanced the breast cancer cells sensitivity to doxorubicin. While GAS5 expression was low under normal conditions in breast cancer, its expression was significantly upregulated by IDET. The upregulation in GAS5 expression by IDET may be responsible for its anti-proliferative, apoptosis inducing and chemosensitization activities. The chemosensitization properties of IDET are significant as the breast cancer cells develop resistance to chemotherapeutic drugs over time. Similar to these observations, downregulation in GAS5 expression decreases the therapeutic effects of dendrosomal curcumin in breast cancer cells 52 . Similarly, gambogic acid induced GAS5 expression can produce pro-apoptotic effects in bladder cancer cells 53 . The oncogenic H19 is constitutively expressed in multiple cancer types including breast cancer 28,54 . H19 expression also correlate with NF-κB activation 54 and paclitaxel resistance 55 . The suppression in H19 expression by IDET may be another possibility for the inhibition of NF-κB activation and sensitization of breast cancer cells to doxorubicin. The oncogenic role of ANRIL and HOTAIR is reported in the breast cancer model 56 . An upregulation in the expression of ANRIL and HOTAIR by IDET may be a compensatory mechanism in response to the suppressed expression of other oncogenic lncRNAs and upregulation of tumor suppressor lncRNAs. Although the lincRNA-Tnfaip3 is an early response gene controlled by NF-κB in murine macrophages 57 , its role in the breast cancer model remains to be elucidated.
The In Silico data revealed the drug like properties of both IDET and DET. The Lipinski's 'rule of five' suggest that most drug-able compounds have molecular weight ≤500, LogP ≤ 5, number of hydrogen bond donors ≤5, and number of hydrogen bond acceptors ≤10 58 . Both IDET and DET obeyed these rules. The pharmacokinetics analyses revealed that both lactones can cross the blood brain barrier and have better intestinal absorption. The lack of any evidence for carcinogenic and genotoxic properties further support that both lactones can be good drug candidate.
In conclusion, IDET exhibit anti-carcinogenic, pro-apoptotic and anti-proliferative activities in breast cancer cells. Further, IDET is potent as compared to DET. The NF-κB inactivation and the modulation in lncRNA expression may be underlying mechanism for the anti-carcinogenic activities of IDET. However, whether NF-κB regulate lncRNA expression or later regulate the former in response to IDET is unclear. Whether the micromolar concentrations of IDET used in the current study are physiologically relevant remains to be elucidated. However, a previous study demonstrated that IDET exhibit anti-inflammatory activities only in cancer cells but not in normal lymphocyte 59 . Future studies should examine the activities of the lactone in the breast cancer animal model. The thorough pharmacokinetics and pharmacodynamics studies in animal models should also be performed before IDET can be tested in humans by clinical trial.