Anti-EMT properties of CoQ0 attributed to PI3K/AKT/NFKB/MMP-9 signaling pathway through ROS-mediated apoptosis

Background Breast cancer is the most prevalent cancer among women. In triple-negative breast cancer (TNBC) cells, a novel quinone derivative, coenzyme Q0 (CoQ0), promotes apoptosis and cell-cycle arrest. This study explored the anti-epithelial–mesenchymal transition (EMT) and antimetastatic attributes of CoQ0 in TNBC (MDA-MB-231). Methods Invasion, as well as MTT assays were conducted. Lipofectamine RNAiMAX was used to transfect cells with β-catenin siRNA. Through Western blotting and RT-PCR, the major signaling pathways’ protein expressions were examined, and the biopsied tumor tissues underwent immunohistochemical and hematoxylin and eosin staining as well as Western blotting. Results CoQ0 (0.5–2 μM) hindered tumor migration, invasion, and progression. Additionally, it caused MMP-2/− 9, uPA, uPAR, and VEGF downregulation. Furthermore, in highly metastatic MDA-MB-231 cells, TIMP-1/2 expression was subsequently upregulated and MMP-9 expression was downregulated. In addition, CoQ0 inhibited metastasis and EMT in TGF-β/TNF-α-stimulated non-tumorigenic MCF-10A cells. Bioluminescence imaging of MDA-MB-231 luciferase–injected live mice demonstrated that CoQ0 significantly inhibited metastasis of the breast cancer to the lungs and inhibited the development of tumors in MDA-MB-231 xenografted nude mice. Silencing of β-catenin with siRNA stimulated CoQ0-inhibited EMT. Western blotting as well as histological analysis established that CoQ0 reduced xenografted tumor development because apoptosis induction, cell-cycle inhibition, E-cadherin upregulation, β-catenin downregulation, and metastasis and EMT regulatory protein modulation were observed. Conclusions CoQ0 inhibited the progression of metastasis as well as EMT (in vitro and in vivo). The described approach has potential in treating human breast cancer metastasis. Electronic supplementary material The online version of this article (10.1186/s13046-019-1196-x) contains supplementary material, which is available to authorized users.


Background
Worldwide, breast cancer is the commonest cancer to affect women, and in Taiwanese women, it is the leading cause of deaths from cancer [1]. It possesses highly metastatic and invasive properties, which explain its high mortality rate [2]. Among breast cancers, triple-negative breast cancers (TNBCs) lacking the genes for estrogen receptor, HER2, and progesterone receptor have been correlated with tumor aggressiveness. TNBCs are more likely than other breast cancer types to migrate beyond the breast and to recur after chemotherapy or lumpectomy [3]. TNBC cases comprise 15-20% of all breast cancer cases. Furthermore, patients with TNBC exhibit unfavorable outcomes compared with those with other breast cancer subtypes [4]. TNBC tumor cells lack the requisite receptors, which renders some targeted or hormone therapies ineffectual. Consequently, combinations of chemotherapy medicines are typically prescribed for patients with TNBC. This approach, however, does not help patients with cancer to counter the chemotherapy-induced adverse side effects and drug resistance [5]. Thus, novel compounds with lower toxicity are urgently required for effective treatment of TNBC.
In cancer cells, polarized epithelial cells complete multifaceted changes that cause them to begin expressing a mesenchymal phenotype and undergo migration, invasion, and metastasis. This process is referred to as the epithelial-mesenchymal transition (EMT) [6]. Several factors induce EMT in vitro and in vivo, for example, TGF-β1, ROS, TNF-α, and hypoxia [7][8][9]. EMT involves AKT/GSK or NFκB-mediated expression of Snail and promotes cell invasion and migration in various cancers, such as breast, renal, and colon cancers [10,11]. The loss of E-cadherin, an adherens junction cell surface protein expressed in epithelial cells is the principal characteristic of EMT [12]. The Snail and Slug signaling cascades are among those that may be involved in EMT in cancer cells. Snail and Slug are key transcription factors that can down regulate the expression of E-cadherin. They do this by binding to E-boxes in the E-cadherin promoter, subsequently increasing MMP-9 expression to promote cell invasion [13]. However, few studies have investigated the suppression of molecular events and EMT responsible for EMT inhibition in anticancer treatment.
The Wnt/β-catenin signaling pathway contributes to cell fate decisions as well as the normal cellular response during cancer cell development [14]. Researchers have suggested that dysregulated or uncontrolled triggering of this signaling pathway promotes tumor progression and metastasis in patients with breast cancer [15]. Other attributes of the Wnt extracellular signaling pathways manage tissue architecture, proliferation, embryonic axis formation, and cell migration [16] and can be broadly classified into noncanonical and canonical pathways. Canonical pathways are activated when the relevant Wnt ligands bind to the LRP-5/6 coreceptors and Frizzled transmembrane domain receptor [17], whereas non-canonical pathways are β-catenin-independent and need Ror2/Ryk coreceptors rather than LRP-5/6 coreceptors. β-Catenin is usually aberrantly activated in breast cancer tissues. Therefore, Wnt/β-catenin pathway inhibition has the potential to reduce breast cell invasion as well as that of their EMT.
Coenzyme Q 0 (CoQ 0) also known as ubiquinone 0 and 2,3 dimethoxy-5-methyl-1,4 benzoquinone) and a member of the mitochondrial respiratory chain is a redox-active ubiquinone compound commonly present in the mitochondrion. It possesses strong antioxidant activity and prevents the mitochondrial permeability transition pore [18] from being opened calcium-dependently. CoQ 0 has demonstrated activity against the proliferation of numerous cancer cell lines (e.g., HepG2, A549, and SW480) [19,20]. Although it exhibits cytotoxic anticancer activities, it was also demonstrated to stimulate insulin secretion in pancreatic islets [21]. We described its anti-inflammatory and antiangiogenic properties in vivo and in vitro in our previous study [22]. Remarkably, administering CoQ 0 mixtures prevents oxidative damage in rodent spleen, blood, kidney, heart, and liver [23]. Our previous study on CoQ 0 found that it significantly inhibits melanoma cell growth and tumor formation by inducing apoptosis and cell-cycle arrest [24]. Additionally, it effectively promoted apoptosis by increasing ROS in MCF-7 cells that were irradiated using ultraviolet B [22]. Despite CoQ 0 's anticancer attributes, its inhibitory effect on breast cancer metastasis and EMT and the molecular mechanism that gives it its therapeutic efficacy are unclear.
To ascertain CoQ 0 's capabilities at inhibiting metastasis, EMT, and their associated changes, we designed a validated EMT and metastasis model for human TNBC (MDA-MB-231). Metastasis and EMT control levels and the principal molecular biomarkers involved were analyzed to ascertain the anti-EMT and antimetastatic attributes mediated by CoQ 0 . In addition, we sought to clarify the fundamental mechanism of TNBC cells.

MTT assay
MTT colorimetric assays were employed to ascertain cell viability [25]. Cells were grown in 12-well plates to a confluence. Subsequently, they underwent 24-h CoQ 0 incubation (2.5-20 μM). Following the MTT assay, an ELISA microplate reader (Bio-Tek Instruments Inc., Winooski, VT, United States) was utilized to ascertain the absorbance at 570 nm. The proportion of viable cells relative to vehicle-treated control cells (designated as 100%) was the basis for evaluating CoQ 0 's effect on cell viability.

In vitro wound-healing assay
The effects of CoQ 0 on cell migration were evaluated using in vitro wound-healing assay as follows: the cells were cultured (density: 1 × 10 4 ), and the standard protocol of the in vitro healing assay was followed [26]. Subsequently, the cells underwent 24-h incubation with varying CoQ 0 concentrations (0.5-2 μM) in 1% FBS medium followed by phosphate-buffer saline (PBS) wash (3 times). They were subsequently fixed using 100% methanol. Finally, they were stained using Giemsa staining solution (Merck, Darmstadt). Migrated cells were monitored, and phase-contrast microscopy at 200× was used to photograph them. Image-Pro Plus (Media Cybernetics, Inc., Rockville, MD, United States) was used to compute the wounded area's closure.

Cell invasion assay
Cell invasion assay was conducted using BD Matrigel invasion chambers (Bedford, United States), as described by Debbie and Brooks [27]. We used 10 μL of Matrigel (25 mg/50 mL) to coat 8-μm polycarbonate membrane filters. Then, we seeded the cells (density: 1 × 10 5 ) onto the Matrigel-treated filter in 200 μL of CoQ 0 (0.5-2 μM) devoid of serum. Cell migration underwent 24-h observation at 37°C. After being incubated for 24 h, the cells that did not migrate were removed from atop the membrane by using a cotton swab. On the other side of the membrane, the migrated cells were fixed for 15 min in ice-cold methanol (75%) and were washed in PBS three times. Subsequently, Giemsa stain was used to fix the cells; they were subsequently destained using PBS. Images were viewed and captured through 200× light microscopy; we quantified invading cells by using manual counting. All experiments were repeated three times.

Protein isolation and Western blotting
We seeded the cells (density: 4 × 10 6 cells/dish) into a 6-cm dish that had undergone 24-h CoQ 0 (2.5-10 μM) pre-treatment. Next, they underwent trypsinization and 3 times rinsing with ice-cold PBS. Total cytoplasmic and nuclear extracts were isolated in accordance with the suggested protocol (Pierce Biotechnology, Waltham, MA, United States). The protein concentration was ascertained using Bradford reagent, with bovine serum albumin (BSA) as standard. The lysate was separated through 12% SDS-PAGE. PVDF membranes were immunoblotted using specific primary antibodies and their corresponding horseradish peroxidase-conjugated secondary antibodies. An enhanced chemiluminescence substrate (Pierce Biotechnology, Waltham) was utilized to develop the blot, and a densitometric graph displaying band intensities was generated using AlphaEaseFC (Miami, FL, United States).

Mammosphere formation assay
We performed a mammosphere formation assay as described elsewhere with several minor modifications [29]. For mammosphere culture from MDA-MB-231 cells, we suspended cells at 1 × 10 3 cells/mL and seeded them onto ultralow attachment plates (Corning Inc., Corning, NY, United States) in MammoCult basal medium (STEMCELL Technologies, United States) supplemented with 0.2% heparin (STEMCELL Technologies, United States), 500 μg/mL cortisone (STEMCELL Technologies, United States), and 10 μL/mL mammo supplant serum (STEMCELL Technologies, United States) in an incubator supplemented with 5% CO 2 . The cells underwent dimethyl sulfoxide (DMSO) (0.1%) or CoQ 0 (0.5-2 μM) treatment and were then incubated for 1 week during which time the plate was kept stationary and the media were replenished. On day 7, the mammospheres were collected through gentle centrifugation and dissociated into single-cell suspensions; the cells then underwent repeated vehicle control (0.1% DMSO) or CoQ 0 (0.5-2 μM) treatment. The cell suspensions underwent three passages through a syringe with 26-G needles. This procedure resulted in single-cell suspensions being formed. We then adjusted cell concentration to 500 cells/mL and plated the 2-mL single-cell suspensions on a 35-mm ultralow attachment plate (Corning Inc., Corning) in triplicate. For serial passage (secondary sphere formation) and differentiation experiments, this study employed early progenitor cells as well as sphere forming, breast stemenhanced single cells. On day 14, a graticule-enabled microscope was employed to determine the number of mammospheres of diameter > 50 μm as well as the number of colony-forming units. Then, the cells were trypsinized, and the cell numbers were counted. Two independent observers not exposed to the treatment were asked to count the sphere numbers manually. All experiments were conducted three times.

Colony formation assay
MDA-MB-231 cells (density: 5 × 10 5 cells/60-mm dish) underwent 24-h CoQ 0 treatment (concentrations: 0-7.5 μM). We then trypsinized the cells and replated them (density: 3 × 10 4 cells/35-mm dish) in triplicate. They were subsequently subjected to 7-d RPMI 1640 incubation. Afterward, they were fixed at room temperature for 10 min using 10% neutral buffered formalin and stained using 20% Giemsa stain (Merck, Darmstadt). We subsequently assayed the cells for ability to form colonies and proliferation rate. The number of colonies of size > 1 mm was determined with light microscopy (40×). Colony number was ascertained based on a colony formation percentage of 100% in the absence of CoQ 0 .

Gelatin zymography assay
The zymography protease assay was used to quantify the activity of MMP-2 and 9 in the medium used to grow the MCF-10A cells. The MCF-10A cells (1 × 10 6 cells/ well) were seeded into 6-well culture dishes and grown in medium with 10% FBS to a nearly confluent monolayer. The cells were re-suspended in 1% FBS medium, and then incubated with TGF-β/TNF-α (10 ng/mL) and CoQ 0 (2 μM) for 24 h. After treatment the remains procedures were followed according to the previously mentioned work [30]. The changes in expression of MMP-2 and -9 were quantified by Matrix Inspector 2.1 software (AlphaEase, Genetic Technology, Inc., Miami, FL, USA).

Luciferase reporter assay
The transcriptional activity of NFκB, E-cadherin, and β-catenin was ascertained through dual-luciferase reporter assays (Promega, Fitchberg, WI, United States). We grew the cells to 70-80% confluence for 5 min in a 24-well plate by using serum-free DMEM devoid of antibiotics. They were subsequently transfected with NFκB, E-cadherin, or β-catenin plasmids with β-galactosidase or a pcDNA vector by using Lipofectamine 2000 (Invitrogen, Carlsbad). Afterward, they underwent 4-h CoQ 0 (0.5-2 μM) treatment. Subsequently, they were lysed, following which their luciferase activity was ascertained through luminometry (Bio-Tek, Winooski) and then standardized to their β-galactosidase activity in the cell lysates. In addition, we quantified the intensity of relative fluorescence by using a luminance ELISA reader.

Immunoprecipitation
Protein A-sepharose beads were used to preclear 1 mg of the protein sample for 1 h, which subsequently underwent 24-h incubation using 2 mg of anti-E-cadherin antibody. Radioimmunoprecipitation assay (RIPA) buffer was employed to wash the immunoprecipitated complexes. Subsequently, the complexes were denatured using SDS sample buffer. Centrifugation (14,000 rpm) was employed to clarify the cell lysate for 15 min. It was then separated through SDS-PAGE and conveyed to PVDF membranes. Skim milk (5%) was used to block the membrane for 30 min. Then, it was immunoblotted using specific primary antibodies and the corresponding HRP-conjugated secondary antibodies for 1 h and subsequently visualized with the aid of an electroluminescence reagent (Millipore, Burlington, MA, United States).

Transient transfection of siRNA targeting β-catenin
Lipofectamine RNAiMAX (Invitrogen, Carlsbad) was used to transfect the cells with β-catenin siRNA. To facilitate transfection, the cells were incubated in 10% FBS containing DMEM; subsequently, they were plated to 60% confluence on a 6-well plate during transfection. The following day, we replaced the culture medium with 500 μL of Opti-MEM, which we then subjected to transfection using 5 μL of RNAiMAX.
We mixed 250 μL of Opti-MEM and 5 μL of RNAi-MAX and subjected the mixture to incubation for 5 min at room temperature. In another tube, 100 pM siRNA in 1 mL of Opti-MEM and 250 μL of Opti-MEM were combined. Subsequently, we added siRNA solution to the diluted RNAiMAX reagent, and the prepared 500-μL siRNA/RNAiMAX mixtures underwent incubation at room temperature for 25 min to facilitate the formation of the complex.
Afterward, the cells and solution were combined. They together represented a final transfection volume of 1 mL. After 6-h incubation, 2 mL of standard growth medium was used to replace the transfection medium, which subsequently underwent culturing at 37°C. Afterward, the cells underwent 24-h CoQ 0 (1 μM) incubation. To ascertain cellular protein level, Western blotting was employed.

Cell-cycle analysis
Cellular DNA content was assessed through flow cytometry using propidium iodide (PI)-labeled cells. DMEM medium with thymidine (3 mM) was used to block AGS cells for 16 h. PBS was subsequently employed to wash cell-cycle synchronized cells, and these cells were reprogrammed so as to enter the G 1 phase for 24 h with fresh DMEM medium added that included CoQ 0 (5-15 μM). Afterward, cellular trypsinization was executed, and the cells were subsequently fixed at − 20°C overnight in 3 mL of ice-cold ethanol (70%). Their pellets were gathered through centrifugation and resuspended in 0.5 mL of PI staining buffer (1% Triton x-100, 0.5 mg/mL RNase A, 4 μg/mL PI in PBS). They were then incubated at room temperature for 30 min. FACScan cytometry (BD Biosciences, San Jose, CA, United States) adapted with a 488-nm single Ar ion laser was adopted to ascertain cell-cycle progression. ModFit (Verity Software House, Topsham, ME, United States) was utilized to analyze the cell cycles.

TUNEL assay
Apoptotic cell death was examined by applying the TUNEL method by using TdT-dUTP-fluorescein in situ cell detection tools (Roche, Mannheim, Germany) as expounded elsewhere [31]. Optical microscopy was used to view the slides.

Measurement of ROS level in cells
The cells (density: 4 × 10 5 cells/well) were grown in a 12-well plate. They underwent 5 min-1 h of CoQ 0 (5-15 μM) pre-treatment. Following, DCFH 2− (10 μM) and the culture were combined; they were subjected to incubation for 30 min at 37°C. Warm PBS was used to wash the cells, and generation of ROS was ascertained from intracellular DCF production that was the result of DCFH 2 oxidation [32]. A fluorescence microscope (Olympus, Shinjuku, Tokyo) was employed (magnification: 200×) to determine the level of DCF fluorescence.

Xenograft animal model
China Medical University's Institutional Animal Care and Treatment Committee approved all protocols involving animals and their welfare. Briefly, 5-to 7-weekold athymic nude mice (female; BALB/c-nu), acquired from the National Laboratory Animal Center (Taipei, Taiwan), were confined in a sterile environment with a 12-12 h light-dark cycle. They were fed rodent chow (Oriental Yeast Co., Tokyo) and provided unlimited access to water.
The mice were subcutaneously engrafted in the right-hind flank with 1 × 10 6 MDA-MB-231 cells. The mice were separated into two groups of five. For this experiment, cells passaged less than 20 times were used. The treatment-group mice were administered C0Q 0 (0.75 mg/kg b.w.) Intraperitoneal (IP) injection (three times/week) was performed for 12 weeks; the control group received only the vehicle (PBS). The tumors were measured on a weekly basis, and tumor volume was ascertained as width 2 × length × 0.5 (mm). The mice were sacrificed on the 12th week; this was followed by tumor removal and weighing. A veterinary pathologist examined the excised organs, including the liver, lungs, and kidneys.

In vivo anti-metastasis by bioluminescence imaging
The animals that underwent CoQ 0 (1.5 or 2 mg/kg) and MDA-MB-231-luciferase cell (1 × 10 6 cells/well) treatment were injected intravenously. Afterward, they were sedated and then IP injected with luciferin. An IVIS 200 system was employed to image them, with the images representing the luminescence (photons/s) emitted from the animal.

Immunohistochemistry analyses
Paraformaldehyde (4%) was used to fix the biopsied tumor tissues. This, in addition to segmentation and staining, were performed to facilitate observation under light microscopy. For Western blotting, the tissue samples were mixed in RIPA buffer that contained phosphatase inhibitor cocktail (1%; Sigma-Aldrich, St. Louis) and protease inhibitor cocktail (1%). SDS-PAGE gel was the site of sample separation. Subsequently, the samples were relocated onto PVDF membranes. Corresponding primary and biotinylated secondary antibodies (Zymed Laboratories, South San Francisco, CA, United States) were added to the PVDF membranes and incubated. Afterward, the membranes underwent avidin-biotin complex reagent incubation. Finally, they were stained using 3,3′-diaminobenzidine in accordance with the manufacturer's procedures (Histostain-Plus Kit; Zymed, South San Francisco).

Statistical analysis
The experiment was conducted in triplicate; the values were represented as mean ± standard error. Significance was ascertained using Dunnett's test for pair-wise comparison.
Tumor local invasion and migration comprise the first stage in cancer cell metastatic cascade, resulting in the production of typically fatal distant metastasis. The influence of CoQ 0 (0.5-2 μM) on MDA-MB-231 cell migration was investigated through an in vitro wound closure assay followed by a scratch assay. A significant dose-dependent decrease in cell migration was noted for CoQ 0 treatment ( Fig. 1c-d). Then, the ability of CoQ 0 to hinder invasion of MDA-MB-231 cells was determined using BD Matrigel chamber assay. CoQ 0 (0.5-2 μM, 24 h) significantly and dose-dependently decreased the invasion of MDA-MB-231 ( Fig. 1e-f). The foregoing results suggest that treatment with CoQ 0 has antimigratory and anti-invasive impacts on MDA-MB-231 cells.  (Fig. 1g). Western blotting revealed that VEGF expression was downregulated following CoQ 0 treatment (Fig. 1g). MMP and uPA physiological activity is associated with PAIs and TIMPs irrespective of their particular endogenous inhibitors. Accordingly, we studied TIMP and PAI expression in the MDA-MB-231 cells that underwent 24-h CoQ 0 (0.5-2 μM) treatment. Substantial upregulation in the expression of PAI-1, TIMP-1, and TIMP-2 was recorded after CoQ 0 treatment (Fig. 1g). Furthermore, CoQ 0 in MDA-MB-231 inhibited the mRNA gene expression of uPA, uPAR, MMP-9, and MMP-2 for the tested concentrations (Fig. 1h). The results suggest that MMP-9, which is a crucial factor for metastasis, was inhibited by CoQ 0 treatment.

COQ 0 attenuates mammosphere formation in MDA-MB-231 cells
A previous research depicted the development of mammospheres, which are spherical clusters of nonadherent mammary stem/progenitor cells [33]. To investigate whether CoQ 0 affects tumor cell mammosphere formation, we exposed cells to various concentrations of CoQ 0 for 14 d. Intriguingly, our results indicate that CoQ 0 inhibited nonadherent spherical breast cancer clusters in vitro so that the cells became noncompetent in generating secondary spheres and differentiating along more than one lineage ( Fig. 2a-b). As indicated in Fig. 2a, reduction in the size and formation of such spheres was observed for CoQ 0 (0.5, 1, and 2 μM) treatment. Subsequently, we determined the differentiation and proliferation capacity of these cells in the presence of CoQ 0 . A significant reduction in cell growth of MDA-MB-231 was noted (Fig. 2b). The collective outcome of the aforementioned results suggests that CoQ 0 exhibits a strong anti-mammosphere forming capability.

CoQ 0 attenuates colony formation
Subsequently, we determined whether CoQ 0 could affect soft agar-cultured MDA-MB-231 cells anchorage-independent growth, a property that tumor cells demonstrate with in vivo tumorigenesis. Dose-dependent inhibition (5 d) of the growth of anchorage-independent MDA-MB-231 cells was noted (Fig. 2c). Colony number inhibition increased with a reduction in colony size. The decreased colony formation ability with CoQ 0 indicates reduced MDA-MB-231-cell tumorigenic ability. These results reveal that CoQ 0 effectively hinders MDA-MB-231-cell survival and growth.  (Fig. 2d-g), whereas CoQ 0 showed no to minimal effect against BT549, Hs578T, MCF-7, and BT474 ( Fig. 2h-k). These results indicated that CoQ 0 was more potent against the brain and bone metastatic variants of TNBC cells than against other tumorigenic cells.

CoQ 0 inhibits breast cancer cells viability
CoQ 0 attenuates NFκB activation by suppressing I-κBα degradation in MDA-MB-231 cells The NFκB transcription factor family regulates myriad genes involved in MMP, VEGF, or uPA expression. Therefore, the effect of CoQ 0 on NFκB signaling and its regulatory proteins in MDA-MB-231 cell were studied. Immunofluorescence assay shows that CoQ 0 treatment diminished nuclear p65 protein dose dependently (Fig. 3a). The results are presented as the mean ± SD of three independent assays **p < 0.05, ***p < 0.001 significant compared to control cells NFκB activity, which was ascertained through luciferase reporter assays, was high in the control group but decreased dose dependently in cells treated using CoQ 0 (Fig. 3b). Next, we determined whether the inhibition of NFκB by CoQ 0 is related to I-κBα protein degradation, which regulates NFκB stability. CoQ 0 treatment attenuated I-κBα degradation, which in turn hindered NFκB (p65) activation (Fig. 3c). To determine whether I-κBα degradation suppression was the result of IKKα phosphorylation inhibition, we used Western blotting to explore IKKα phosphorylation. CoQ 0 pre-treatment suppressed IKKα phosphorylation dose dependently (Fig. 3c). These results indicate that CoQ 0 suppressed nuclear activation of NFκB by inhibiting I-κBα degradation.
CoQ 0 inhibits MMP-9 through the inhibition of PI3K/AKT/ NFκB pathways in MDA-MB-231 NFκB has a major role in managing the MMP expression in numerous cancer cell lines. To more profoundly comprehend the inhibitory mechanisms of CoQ 0 on the transcriptional regulation of MMP-9, we investigated NFκB and MMP-9 through Western blotting. The cells underwent celastrol (an NFκB inhibitor) pre-treatment and then 24-h CoQ 0 treatment. The results revealed that CoQ 0 lowered p65 and MMP-9 protein expression. However, inhibition was more pronounced in the case of celastrol pre-treatment (Fig. 3d).
Based on the findings that CoQ 0 inhibited NFκB and MMP-9 expression and because PI3K/AKT is the major pathway involved in NFκB/MMP-9 activation, we hypothesized that CoQ 0 regulates the PI3K/AKT pathway. We evaluated this hypothesis by appraising the impact of CoQ 0 on PI3K and AKT phosphorylation. As shown in Fig. 3e, CoQ 0 substantially decreased p-PI3K and p-AKT expression dose dependently. Subsequently, we investigated the impact of LY294002, a PI3K inhibitor, on nuclear p65 and MMP-9 expression in the absence or presence of CoQ 0 . Lower nuclear p65 and MMP-9 expression was recorded in the presence of LY294002 (Fig. 3f). These results suggest that CoQ 0 suppressed MMP-9 by inhibiting the PI3K/AKT/NFκB pathways of MDA-MB-231 cells. a Cells were grown in chamber slides and exposed to CoQ 0 (0.5-2 μM) for 2 h, fixed and permeabilized. Cells were incubated with anti-p65 antibody followed by FITClabeled secondary antibody. The subcellular localization of p65 was visualized using a confocal microscope of 40 × magnification. b NFκB activity was evaluated via luciferase reporter gene assay. Cells were transfected with the luciferase reporters and co-transfected with either NFκB or empty vector. Then, they were treated with CoQ 0 (0.5-2 μM) for 4 h, and luciferase activity was determined and normalized with β-gal activity, shown as relative luciferase activity. c CoQ 0 induces nuclear p65 and inhibits cytosolic I-κB and p-IKK degradation. Cells were treated with CoQ 0 (0.5-2 μM for 2 h). Cytoplasmic and nuclear extracts were prepared and Western blot analyses were performed. β-actin or histone H3 was used as an internal control for cytoplasmic and nuclear extracts, respectively. d CoQ 0 inhibits MMP-9 expression through the downregulation of NFκB signaling pathways. The nuclear p65 (for 4 h) and MMP-9 expression for 24 h was inhibited following the treatment with CoQ 0 (1 μM) in the presence or absence of NFκB inhibitor celastrol (0.3 μM), as shown by Western blot. For internal control β-actin was used. e Phosphorylated PI3K (p-PI3K) and AKT (p-AKT) levels were evaluated by immunoblot analysis. Cells were treated with CoQ 0 (0.5-2 μM) for 24 h. β-actin, total PI3K and AKT levels were considered as an internal control. f CoQ 0 inhibits NFκB/MMP-9 expression through the downregulation of PI3K/AKT signaling pathways. The nuclear p65 (for 4 h) and MMP-9 expression (for 24 h) was inhibited following treatment with CoQ 0 (1 μM) in the presence of absence of PI3K/AKT inhibitor LY294002 (25 μM), as shown by Western blot. β-actin was used as an internal control. The results are presented as the mean ± SD of three independent assays. **p < 0.05, ***p < 0.001 significant compared to control cells changes compared with the control group (Fig. 4a). E-cadherin, a protein that has a role in regulating EMT, was assayed in MDA-MB-231 cells. A rise in E-cadherin expression due to CoQ 0 may be because E-cadherin transcriptional activity was activated. Thus, we measured E-cadherin promoter activity on the basis of luciferase activity: the data are provided in Fig. 4b. We transfected an E-cadherin promoter construct in a pcDNA vector into MDA-MB-231 cells; we then ascertained cell luciferase activity. The findings indicated that luciferase activity from the E-cadherin promoter consistently and dose dependently improved following CoQ 0 treatment (0.5-2 μM) (Fig. 4b).
The immunofluorescence assay revealed that CoQ0 induced the expression of E-cadherin. E-cadherin antibodies were used for immunofluorescence staining, and the results indicated E-cadherin upregulation in the CoQ 0 -treated cells (Fig. 4c). Western blotting as well as RT-PCR affirmed E-cadherin upregulation by CoQ 0 . Increased dose-dependent expression of Occludin, followed by reduced Vimentin, Slug, Twist, and Snail expression, while an increase expression of E-cadherin was only observed at 0-1 μM and not at 2 μM was recorded for CoQ 0 -treated MDA-MB-231 cells (Fig. 4d). Furthermore, increased mRNA E-cadherin expression and reduced Snail and Slug expression with dose-dependent 6-h CoQ 0 treatment were observed (Fig. 4e). These outcomes indicate that CoQ 0 inhibited EMT through the upregulation of E-cadherin pathways.
CoQ 0 attenuates EMT by inhibiting Wnt/β-catenin signaling pathways in MDA-MB-231 E-cadherin/β-catenin protein complexes play an active role in EMT and are critical to cancer progression [34]. To evaluate the effects of CoQ 0 on the E-cadherin/β-catenin complex; we first measured the β-catenin expression of MDA-MB-231 cells. The results demonstrated that CoQ 0 treatment suppressed β-catenin expression in the cytoplasm and nucleus dose dependently (Fig. 5a). Moreover, the mRNA gene expression of β-catenin was subdued by CoQ 0 treatment in the MDA-MB-231 cells (Fig. 5b) b Transcriptional activity of E-cadherin was monitored by luciferase reporter assay. c Immunofluorescence analysis for E-cadherin protein expression. d CoQ 0 -induced modulation of epithelial (E-cadherin and Occludin) and mesenchymal marker proteins (Vimentin, Slug, Twist, and Snail) were monitored using Western blot analyses. e mRNA expression of E-cadherin, Vimentin, Slug, and Snail after 6 h treatment with CoQ 0 (0.5-2 μM) was measured by RT-PCR analyses. As internal control GAPDH was used. The results are presented as the mean ± SD of three independent assays. **p < 0.05, ***p < 0.001 significant compared to control cells dependently treated with CoQ 0 . By contrast, CoQ0 did not affect cells that were transfected with FOP reporter vector and used as negative control. Subsequently, enhanced E-cadherin and β-catenin association was noted (Fig. 5d). Next, we examined E-cadherin by using a luciferase reporter construct. This construct was stably transfected into MDA-MB-231 cells. A luciferase reporter assay revealed that cells treated with CoQ 0 (1 μM) caused a profound increase in E-cadherin promoter activity. Intriguingly, cells that underwent pre-treatment with the NFκB inhibitor celastrol (0.3 μM), β-catenin inhibitor iCRT5 (5 μM), or both exhibited a significant increase in E-cadherin luciferase activity relative to the control cells (Fig. 5e), suggesting that CoQ 0 induces the expression of E-cadherin by suppressing the β-catenin and NFκB signaling pathways in MDA-MB-231 cells. We transfected MDA-MB-231 cells with β-catenin siRNA to confirm this phenomenon; consequently, we observed changes in E-cadherin, Snail, and Slug proteins following CoQ 0 treatment. Through Western blotting, we determined that CoQ 0 treatment considerably increased the expression of E-cadherin and suppressed that of Slug and Snail in MDA-MB-231 cells transfected with β-catenin siRNA (Fig. 5e). These results suggest that CoQ0 would act on E-cadherin expression thus modulating EMT by not only Wnt/β-catenin, but also NFκB pathway.

CoQ 0 inhibits metastasis and EMT induced by TNF-α/TGF-β
The non-toxic concentrations of CoQ 0 (2 μM) were employed to assess the anti-EMT and antimetastatic effects in TNF-α/TGF-β-stimulated non-tumorigenic MCF-10A breast cells. MCF-10A intrusion was ascertained using a Boyden chamber assay. This assay determined cells' ability to traverse a Matrigel-coated filter's extracellular matrix layer. The findings revealed that TNF-α/TGF-β treatment enhanced MCF-10A cell invasiveness significantly compared with that of untreated cells. Additionally, CoQ 0 pre-treatment resulted in considerable inhibition of TNF-α/TGF-β-induced invasiveness of MCF-10A (Fig. 6a). Subsequently, we investigated the effect of CoQ 0 on TNF-α/TGF-β-induced stimulation of MMPs and uPA levels. Zymography and Western blotting analysis showed that CoQ 0 pre-treatment attenuated the Following transfection, the cells were incubated with or without CoQ 0 (1 μM) for 24 h. Western blot analyses was performed to measure the expression levels of β-catenin transcriptional target genes such as E-cadherin, Slug, and Snail. The results are presented as the mean ± SD of three independent assays. **p < 0.05, ***p < 0.001 significant compared to control cells MMP-2 and MMP-9 activity (Fig. 6b) and uPA protein expression in MCF-10A cells induced by TNF-α/TGF-β (Fig. 6c).

CoQ 0 induces apoptosis through generation of ROS in MDA-MB-231 cells
Studies have implicated ROS generation as a cellular apoptosis inducer [35]. To ascertain whether the generation of ROS is associated with CoQ 0 -engendered apoptosis, the intracellular ROS level in CoQ 0 -treated MDA-MB-231 cells was determined. CoQ 0 dose dependently induced apoptosis in breast cancer cells; its effect was observed through death Annexin V-FITC/PI flow cytometry and staining (Additional file 1 a-d). Incubation of cells with CoQ 0 (15 μM) for 5-60 min caused DCF fluorescence to increase time dependently, which was directly proportionate to the amount of ROS generated maximum at 15 min (Fig. 7a-b). ROS levels dose-dependently increased in CoQ 0 -treated (5-15 μM for 15 min) MDA-MB-231 cells (Fig. 7c-d). However, cells subjected to ROS inhibitor treatment (1 mM NAC for 60 min) before CoQ 0 treatment exhibited significantly reduced ROS generation (Fig. 7c-d). Another line of evidence revealed that CoQ 0 -induced MDA-MB-231 cell death did not occur in NAC-pre-treated cells (Fig. 7e). Western blot results revealed that NAC preincubation resulted in a gradual decrease in CoQ 0 -induced apoptotic Bax and p53 protein expression (Fig. 7f) In vivo growth inhibition in xenografted mouse model by CoQ 0 To determine the in vivo impact of CoQ 0 on tumor growth, xenografted nude mice were used. We xenografted MDA-MB-231 cells into them. Observations indicated that all mice were healthy and that their body weights were unaffected during CoQ 0 treatment (FiA). The MDA-MB-231 xenografted animals were treated with CoQ 0 (0.75 mg/kg three times/week) or with only vehicle. A significant time-dependent inhibition of tumor volume was observed for CoQ 0 treatment (Fig. 8a). Additionally, a reduction in tumor weight in the CoQ 0 -treated xenografted mice was observed (Fig. 8b). After 12 weeks, the animals were killed and the xenografted tumor was extracted. Concomitantly, excised tumor sections were observed under the microscope to discern the differences in nuclei and cytoplasmic morphology. The cancer cells in the xenografted mice used as controls appeared large and oval or round in form with myriad nucleoli, and expressed substantial mitotic figure and cellular activity levels (Fig. 8c). By contrast, the CoQ 0 -treated tumor-xenografted mice demonstrated less angiogenesis, and their cells appeared shrunken and condensed. Furthermore, their nuclei exhibited karyopyknosis, implying carcinoma regression or activity (Fig. 8c). These results suggest that CoQ0 promotes antitumor activity in xenografted mouse models.

Fragmentation of apoptotic DNA induced by CoQ 0 in tumorxenografted mouse model
CoQ 0 's role in apoptosis in MDA-MB-231-xenografted mice was ascertained using the TUNEL assay on cancerous sections. The presence of a greater number of TUNEL-positive cells in mouse tumors treated with CoQ 0 compared with those of the untreated controls (p < 0.001) suggested that CoQ 0 treatment was affiliated with reduced cell proliferation and increased apoptosis (Fig. 8d). In addition, immunohistochemical analysis revealed that CoQ 0 -treated mice significantly decreased Cyclin A, Cyclin B, and CDK1 expression and increased Bax, p53, and p21 expression in MDA-MB-231 xenografted tumor tissues (Fig. 8e). The results are the mean (±SE) numbers of cells/microscope field (as percentage) for 5~7 animals per group. Significant at *p < 0.05; **p < 0.01; ***p < 0.001 compared to untreated control cells the inhibition of tumor development by induction of apoptosis in TNBCs during CoQ 0 treatment was observed.

CoQ 0 attenuates in vivo lung metastasis
To validate the potential effect of CoQ 0 on metastasis, we treated mice with CoQ 0 (1.5 or 2 mg/kg) and then intravenously injected them with MDA-MB-231-luciferase cells (1 × 10 6 cells/well). The luciferase-labeled MDA-MB-231 cancer cells metastasized to the lungs after 28 d in the control mice (Fig. 9a). CoQ 0 treatment blocked MDA-MB-231 cell lung metastasis to a statistically significant level; photon flux in the mouse lungs fell by > 90% in the CoQ 0 treatment groups (1.5 and 2 mg/ kg) (Fig. 9a). These findings imply that CoQ 0 entirely suppresses highly metastatic breast cancer cell lung metastasis at concentrations of 1.5 or 2 mg/kg. To validate the mechanism by which CoQ 0 attenuates tumor metastasis, the metastasis-related proteins from control and CoQ 0 -treated (0.75 mg/kg) mouse tumors were studied. Immunohistochemical and Western blot results revealed that CoQ 0 inhibited MMP-9, MMP-2, p-AKT, p65, and β-catenin expression and increased E-cadherin expression relative to the controls (Fig. 9b-c). Comprehensive results confirmed that CoQ 0 suppresses metastasis by downregulation of MMP-2, MMP-9, p65, β-catenin, and p-AKT and upregulation of E-cadherin proteins. Furthermore, the gene expression patterns of MMP-2 and MMP-9 mRNA were substantially suppressed by CoQ 0 (Additional file 3 a). In addition, immunohistochemical analysis revealed significant decreases in uPA, uPAR, Vimentin, COX-2, Twist, VEGF, and p-mTOR in CoQ 0 -treated mice (Additional file 3 b). Subsequently, we investigated the effects of CoQ 0 on EMT regulatory proteins. CoQ 0 substantially reduced uPA, uPAR, COX-2, and Vimentin expression and promoted Occludin protein levels compared with those of the control group as discerned through Western blot analysis (Additional file 3 c). These results indicate that CoQ 0 attenuated metastasis and EMT in MDA-MB-231-xenografted nude mice.

Discussion
EMT is a physiological process that is usually activated during wound healing and embryonic development. It is a crucial step in cancerous metastatic progression [36]. During EMT, the epithelial-derived tumor cells stimulate intercellular and intracellular changes that contribute to mesenchymal cell phenotypes, including cytoskeleton reorganization, polarity alteration, extracellular matrix remodeling, and migratory ability acquisition [12]. Numerous researchers have investigated EMT's role in breast cancer. Mesenchymal EMT molecular marker overexpression in biopsies of breast cancer is correlated with increased recurrence, adverse clinicopathological characteristics, reduced survival, and tumor aggressiveness [37]. Therefore, efficacious therapeutic strategies must be established to reduce breast cancer cell tumor aggressiveness and prevent malignant growth. In our previous study, we reported that CoQ 0 exerts antimetastatic effects in melanoma carcinomas. This action may be because of the modulation of the Wnt/β-catenin signaling pathway in B16F10 melanoma cells [24]. In the current study, CoQ 0 's antimetastatic and anti-EMT abilities were characterized, and mechanisms responsible for its effects in MDA-MB-231 were studied. Additionally, E-cadherin downregulation and alterations of the EMTlinked signaling regulator indicate that MDA-MB-231 cells can commence and propagate the EMT process in cancer cells. The salubrious impact of pre-treatment with CoQ 0 was proven by the renewal of E-cadherin protein and transcriptional activity. The renewal of E-cadherin was linked to β-catenin, NFκB, and MMP-9 inhibition, a key molecular event in EMT inhibition. Increased cancer cell migration and invasion, mammosphere formation, colony formation ability, and tumor growth were effectively suppressed through CoQ 0 treatment. These findings indicate that CoQ 0 is an antimetastatic and anti-EMT substance, and the potential molecular signaling pathways that are involved in this process can be inferred. E-cadherin is an adherens junction protein expressed in normal breast tissue; it is a useful phenotypic marker in cases of breast cancer [38]. In this study, the transcriptional activity and protein levels of E-cadherin were investigated to ascertain the manifestation of EMT with TNF-α/TGF-β-stimulation in TNBC cells. The results of immunofluorescence and luciferase activity and Western blotting revealed that TNF-α/TGF-β could undermine E-cadherin junctions by governing the organization of actin in MDA-MB-231 cells. These results were supported by evidence from previous studies [33]. E-cadherin loss stimulates EMT, which plays a major role in the development of carcinomas to a metastatic state. Although the mechanism that is involved in E-cadherin inactivation in cancer cells remains vague, alterations of transcriptional levels may explain its downregulation [39]. Therefore, an effective strategy for controlling metastasis and EMT progression may be restoring or preventing E-cadherin downregulation by using TNF-α/TGF-β. In this work, the restoration of E-cadherin protein levels and transcriptional activity through CoQ 0 treatment inhibited EMT and the associated carcinoma metastasis. Restoration of E-cadherin expression at the transcription and protein level by Withaferin A was linked to metastasis and cell proliferation inhibition in breast cancer cells [33]. Furthermore, our in vivo study proved that CoQ 0 significantly increased the expression of E-cadherin in MDA-MB-231 tumors and prevented MDA-MB-231 cell lung metastasis, implying that CoQ 0 arrests EMT programming because of its antimetastatic properties in breast cancer cells.
EMT is a crucial mechanism in cancer development and in the first phase of metastasis. Retardation of E-cadherin/β-catenin may facilitate tumor invasion and metastasis [40]. Increasingly, evidence indicates that E-cadherin has a vital function in β-catenin function and stabilization. When E-cadherin expression decreased, β-catenin was able to be separated from the E-cadherin/ β-catenin complexes and could translocate to the nucleus freely. Moreover, β-catenin bound with the TCF/ a b c Fig. 9 In vivo anti-metastatic activity of CoQ 0 . a CoQ 0 inhibited lung metastasis in living MDA-MB-231-luciferase-injected mice by bioluminescence imaging. Mice were treated with CoQ 0 (1.5 or 2 mg/kg) and then the MDA-MB-231-luciferase cells (1 × 10 6 cells/well) were intravenously injected. The mice were anaesthetized, and luciferin was intraperitoneally injected. The mice were imaged using the IVIS 200 system, and the photons from the whole animal were quantified. Diagrams showing the bioluminescent signal emitted from the whole body. The color overlay on the image represents the luminescence (photons/sec) emitted from the animal, as indicated by the color scales. Photos are representative images (n = 4). b-c Tumor sections were from control animals and experimental analogues treated with CoQ 0 (0.75 mg/kg). b MMP-2, MMP-9, p-AKT, p65, E-cadherin, and β-catenin were examined using immunohistochemical staining. c MMP-2, MMP-9, p-AKT, p65, E-cadherin, and β-catenin were examined using Western blotting. The results are the mean (±SE) numbers of cells/microscope field (as percentage) for 3 animals per group. Western blotting results showing the effects of CoQ 0 on the cumulative protein content in the xenograft tumors. β-actin were used as an internal control. Relative changes in protein bands were measured by densitometric analysis with the control being 100%. The results are the mean (±SE) numbers of cells/ microscope field (as percentage) for 5~7 animals per group. Significant at *p < 0.05; **p < 0.01; ***p < 0.001 compared to untreated control cells LEF-1 element after which it activated certain promigratory genes required for EMT combined with related transcription factors [41]. Some transcription factors, including Slug and Snail, which are both among the transcriptional targets of β-catenin, may be associated with E-cadherin and EMT repression [42]. In this work, CoQ 0 induced E-cadherin and significantly decreased nuclear β-catenin, Snail, and Slug protein association, as shown by Western blot analysis. Furthermore, immunoprecipitation assays revealed that CoQ 0 increased E-cadherin and β-catenin expression relative to that of the untreated group. This result indicates that CoQ 0 may restore the formation of E-cadherin/β-catenin complexes in MDA-MB-231 cells, impeding nuclear transport of β-catenin to a greater extent, which subsequently enhances the expression of E-cadherin by inhibiting Slug. Our results are a strong indication notion that the anti-EMT impact of CoQ 0 is correlated with the governance of the formation of E-cadherin/β-catenin complexes. Matrix metalloproteases (MMPs) have a key function in extracellular matrix (ECM) remodeling and degradation [43]. MMPs play roles in all stages of breast carcinogenesis, from tumor initiation to metastasis. Among the several MMP family members, MMP-2 and MMP-9 were highly expressed in invasive breast cancer cells [44]. The present study determined that CoQ 0 pre-treatment abrogated the TNF-α/TGF-β-induced MMP-9 and MMP-2 expression levels in MDA-MB-231 cells. Therefore, MMP-9 and MMP-2 could be CoQ 0 -responsive mediators whose ECM degradation could result in ensuing cancer invasion and migration.
The MMP-9 promoter region possesses cis-regulatory elements, such as two AP-1 and one NFκB binding sites. These sites are not present in MMP-2's promoter region [45]. Therefore; we investigated the effects of CoQ 0 on NFκB, which plays a major role in the transcription of MMP-9. NFκB activation results in cell invasion, metastasis, and survival advantages and drug resistance to several cancer types [46]. Nuclear translocation and transcriptional activation of NFκB subunits are strictly governed by NFκB's inhibitory protein, I-κBα, whose phosphorylation releases NFκB subunits [47]. The data from our experiments clearly demonstrate that treatment with CoQ 0 suppressed the transcriptional activation and nuclear translocation of NFκB. This effect may have been caused by the inhibition of I-κB kinase phosphorylation and I-κBα degradation. Furthermore, pre-treatment with celastrol remarkably reduced the expression of MMP-9 and NFκB proteins. This result indicates that the NFκB pathway is the principal regulatory pathway in the suppression of MMP-9 expression by treatment with CoQ 0 . PI3K/AKT is the major pathway for tumor invasion [48]. Therefore, we sought to determine whether CoQ 0 suppresses the phosphorylation of PI3K/AKT because its signaling cascade is the main component upstream of NFκB and plays a key role in cellular adhesion, differentiation, and growth. The PI3K/AKT axis plays a principal role in metastasis and tumor invasion through activation of NFκB-mediated MMP-9 [49]. Our findings clearly demonstrate that CoQ 0 treatment suppressed PI3K/AKT phosphorylation substantially. Furthermore, NFκB activation and MMP-9 were significantly reduced by blockage of the PI3K/AKT pathway with LY249002 treatment. These findings reveal that CoQ 0 lowers the expression of MMP-9 by blocking NFκB activation through PI3K/ AKT and thus suppresses MMP-9-mediated cell intrusion in MDA-MB-231 human breast cancer cells. Our findings are consistent with those of a report that suggested that LFG-500 extracted from flavonoid inhibits cancer cell intrusion by suppressing the PI3K/AKT/ NFκB/MMP-9 signaling pathways [49].
The present paper documents the anti-EMT and antimetastatic capabilities of CoQ 0 and lists the mechanisms that may cause its effects in non-tumorigenic MCF-10A cells under stimulation induced by TNF-α/TGF-β. TGF-β enhances tumor development by activating EMT. TGF-β-induced EMT exhibited the following attributes: the loss of junctional E-cadherin localization, acquisition of fibroblastic morphology, and increased cellular motility [50]. TNF-α is a proinflammatory cytokine and plays a vital role in tumor malignancy, including motility, tumor cell invasion, and metastasis [51]. TNF-α induced EMT in renal cell carcinoma by suppressing E-cadherin expression and promoting Vimentin and MMP-9 protein expression [52]. Stimulation of TGF-β, TNF-α, or both may cause an EMT-like phenomenon, E-cadherin expression reduction, and morphological changes in Madin-Darby canine kidney cells [53]. In the present study, downregulation of E-cadherin, upregulation of β-catenin, and changes in EMT-linked signaling regulators initiated and propagated EMT in MCF-10A cells stimulated using TNF-α/TGF-β. The advantageous impact of pre-treatment with CoQ 0 was based on the renewal of transcriptional and E-cadherin promoter activity against losses induced by TNF-α/TGF-β. Additionally, the renewed E-cadherin promoter activity was linked to β-catenin, NFκB, and MMP-2/− 9 inhibition, which is a vital molecular event in the inhibition of EMT induced by TNF-α/TGF-β.
Human mammary epithelial cells forms numerous non-adherent spherical colonies know as, "non-adherent mammospheres". [54]. They comprise of several stem cells that regenerate to form mammospheres within serial passages and progenitor cells that perform multi origin differentiation. Accumulating evidence suggests that many cancers, including breast cancer, are guided by a cellular subpopulation, designated as cancer stem cells (CSCs), that mediates tumor metastasis and resistance to conventional therapies. Thus, preventing CSC growth in breast cancer is the optimal strategy for inhibiting tumor development and metastasis [55]. Therefore, research on CoQ 0 -induced molecular mechanisms that mediate CSC proliferation is vital to clarify CoQ 0 's antimetastatic and anticancer activities. Our study revealed that CoQ 0 treatment considerably lowered mammosphere formation and sphere size. These results suggest that CoQ 0 inhibits mammosphere formation.
Apoptosis induction, restriction of cell proliferation by chemical or biological agents, and cell-cycle arrest are intended to be effective strategies in cancer management, particularly of TNBCs. Apoptosis-inducing agents are under investigation as alternative tools for cancer treatment management. A study reported that CoQ 0 treatment caused the proportion of late apoptotic MDA-MB-231 cells to rise when Annexin V/PI staining and then flow cytometry were employed [56]. In the present study, the treatment of MDA-MB-231 cells with CoQ 0 , successfully inhibited anchorage-independent growth and cell proliferation. Examples of the characteristic features of apoptosis are chromatin condensation, internucleosomal DNA cleavage, caspase activation, and cellular morphological changes [57]. In the current study, we demonstrated that, by treating MDA-MB-231 cells with CoQ 0 , apoptotic cell death linked to DNA fragmentation increased considerably. In a study, treating human lung cancer cells with CoQ 0 increased the number of early and late apoptotic cells and reduced apoptotic cell death through antioxidant treatment [19]. Other studies have demonstrated that methoxy-containing analogs of CoQ 0 and quinones that have similar structures to CoQ 0 have a cytotoxic influence on human cancer cells because they induce apoptosis [58]. Researchers employed various CoQ analogs and recorded enhanced DNA fragmentation, caspase-3 activation, and apoptosis for CoQ 4 and CoQ 2 in HL 60 human leukemia cells. However, these effects were not observed for CoQ 10 or CoQ 6 . These results suggest that CoQ 0 analogs pro-apoptotic and anticancer attributes vary depending on the location of the methoxy-substitutions on the quinone nucleus and the length of the isoprenyl side chain. No matter the cell line, CoQ 0 , which possesses no isoprenoid units, suppresses cancer cell growth and triggers early and late apoptosis.
The excessive generation of ROS can induce cell-cycle arrest, oxidative stress, damaged DNA in cancer cells, cell function loss, and cellular apoptosis [59]. CoQ 0 's favorable impact on breast cancer cell lines is linked to mitochondrial dysfunction and the overproduction of intracellular ROS. ROS causes the mitochondrial permeability transition pore to open, mitochondrial proapoptotic factors to be released, and the mitochondrial membrane to depolarize during mitochondria-mediated apoptosis [60]. The present study determined that CoQ 0 treatment leads to a notable increased in intracellular ROS production in MDA-MB-231 cells. By contrast, the antioxidant, NAC, inhibited ROS production, which reduced apoptosis significantly, indicated that MDA-MB-231-cell apoptosis induced by CoQ 0 had a close link with ROS production. CoQ 0 potentially play roles as upstream signaling molecules to induce cell apoptosis mediated by mitochondria. Our findings are agreement with those of prior investigations indicating that natural compounds (e.g., celastrol and deltonin) induce MDA-MB-231 cell ROS-mediated mitochondrial apoptosis [61].
Disruption of the cancer cell cycle is a therapeutic objective of research on novel cancer drugs. This is linked to lower Cyclin A, Cyclin B, Cdc2, and Cdc25C expression and higher CDK inhibitor p21 expression. In eukaryotes, cell-cycle progression included the resultant triggering of CDKs; their activation is cyclin associated. Among CDKs, Cdc2 and CDK2 kinases are mainly triggered with Cyclin B and Cyclin A during G2/M phase progression [62]. Cdc2/Cyclin A and Bi kinase complex activity was suppressed by phosphorylating Tyr15 of Cdc2. Cdc25C phosphatase catalyzed the dephosphorylation of Tyr15 of Cdc2. This reaction was considered the rate-limiting step in their progression into mitosis [63]. P21 might facilitate G2/M cell-cycle arrest maintenance through CyclinB1/Cdc2 complex inactivation, thereby disrupting the cell nuclear antigen-Cdc25C interaction [64]. The findings suggest that Cdc25C, Cdc2, Cyclin A, and Cyclin B expression is downregulated, and the CDK inhibitor p21 increased in MDA-MB-231 cells treated with CoQ 0 , which arrests G2/M phase. The present study's data suggest that the monitored suppression of MDA-MB-231 cell proliferation linked to CoQ 0 treatment was because of G2/M-phase cell-cycle arrest and not G1 arrest. Intriguingly, our results differ from those of the previous report that indicated that treatment of MDA-MB-231 cells with CoQ 0 led to G0/G1-phase cell-cycle arrest.
To enhance CoQ 0 's antimetastatic and anticancer attributes, an in vivo investigation of CoQ 0 -treated MDA-MB-231-xenografted nude mice was executed. CoQ 0treated xenografted nude mice resulted in a significant fall in tumor volume and significantly prevented lung metastasis. The observed anticancer action is seemingly related to mitotic cell inhibition and substantial proliferation of apoptotic cells in tumors treated with CoQ 0 . Fewer mitotic-positive cells in the HL60 xenografted nude mouse tumors treated with CoQ0 indicated reduction in cell proliferation. This reduced cell proliferation may cause a lower tumor volume in nude mice [65].
Furthermore, antimetastatic activity may be linked to the upregulation of E-cadherin and the downregulation of MMP-2, MMP-9, p-AKT, p65, and β-catenin proteins. These in vivo results verify CoQ 0 's effective antimetastatic and antitumor attributes against TNBC that are in agreement with its in vitro anticancer attributes.

Conclusion
This work determined for the first time that non-cytotoxic concentrations of CoQ 0 exhibit antimetastatic and anti-EMT attributes in MDA-MB-231 breast cancer cell. EMT inhibition in MDA-MB-231 cells was linked to the renewal of transcriptional activity and E-cadherin protein. CoQ 0 inhibited mammosphere formation and exhibited its anti-invasive activity by downregulating the PI3K/AKT/NFκB/MMP-9 signaling pathways. Furthermore, CoQ 0 induced ROS-mediated apoptosis and significantly inhibited the growth of tumors in MDA-MB-231 xenografted nude mice. Our results provide new insight into the potential molecular mechanisms that underlie CoQ 0 's promising anticancer attributes. Our findings justify further preclinical and clinical assessments of CoQ 0 for metastatic breast cancer therapy.

Additional files
Additional file 1: CoQ 0 -induced apoptosis in MDA-MB-231 cells. The cells were exposed to CoQ 0 (5-15 μM for 24 h). (a-b) The TUNEL assay was performed to determine apoptotic DNA fragmentation. The green florescence indicates the number of TUNEL positive cells in the microscopic fields (400 × magnification) from three separate samples. The percentage of apoptotic cells was calculated by measuring the florescence intensity of treated cells using commercially available software. (c) Annexin V-FITC and PI staining was used to identify the early/late apoptosis or necrosis, and the data were analyzed using flow cytometry. The results in each quadrant are labeled and interpreted as follows: ( for each tumor sample, and MMP-2 and MMP-9 were examined using RT-PCR. (b) uPA, uPAR, Vimentin, COX-2, Twist, VEGF, and p-mTOR were examined using immunohistochemical staining. (c) uPA, uPAR, Vimentin, COX-2, and Occludin were examined using Western blotting. Western blot on the effects of CoQ 0 on the total protein contents in the xenograft tumors. β-actin was used as the control. Relative changes in protein bands were measured by densitometric analysis with the control being 100%. The results are the mean (±SE) numbers of cells/microscope field (as percentage) for 3 animals per group. Significant at *p < 0.05; **p < 0.01; ***p < 0.001 compared to untreated control cells. (PPTX 12155 kb) Abbreviations EMT: Epithelial-mesenchymal transition; HER2: Human epidermal growth factor receptor 2; TNBC: Triple Negative Breast Cancer Cell