Quinacrine inhibits cMET-mediated metastasis and angiogenesis in breast cancer stem cells

A trans-membrane receptor tyrosine kinase, cMET, belonging to the MET proto-oncogene family, is responsible for cancer metastasis and angiogenesis. But not much is known about the role of cMET in growth and progression of cancer stem cells (CSCs). Earlier studies have shown that Quinacrine (QC), a bioactive agent, has anti-CSCs activity. Here, the role of QC in deregulation of cMET-mediated metastasis and angiogenesis has been systematically evaluated in vitro in highly metastatic breast CSCs (mBCSCs), ex vivo in patient-derived breast cancer stem cells (PDBCSCs) and in vivo in xenograft mice model systems. Cell proliferation, migration, invasion and representative metastasis markers were upregulated in cMET-overexpressed cells and QC exposure inhibited these processes in both mBCSCs and PDBCSCs. Interestingly, metastasis was significantly inhibited by QC in cMET-overexpressed cells but comparatively lesser significant alteration of the process was noted in cMET-silenced cells. Increase in vascularization (in in ovo CAM assay), and cell–cell tube formation (in HUVECs), and enhanced MMP9 and MMP2 enzymatic activities (in gelatin zymography) were noted after cMET overexpression but these processes got reversed after cMET knockdown or QC treatment in cMET-overexpressed cells. QC inhibited angiogenesis significantly in cMET-overexpressed cells, but lesser significant change was observed in cMET-silenced cells. Reduction in tumor volume and decreased expression of metastatic and angiogenic markers were also noted in xenograft mice after QC treatment. Furthermore, QC inhibited cMET activity by dephosphorylation of its tyrosine residues (Y1234 and Y1356) and downregulation of its downstream cascade. Thus, QC inhibited the cMET-mediated metastasis and angiogenesis in in vitro, in ovo, in vivo and ex vivo model systems. Ligand (HGF) binding leads to receptor dimerization and phosphorylation of tyrosine kinase domain of cMET. This activates the cMET signaling cascade. The representative downstream metastasis and angiogenesis-related proteins get upregulated and induce the metastasis and angiogenesis process. But after the QC treatment, cMET get dephosphorylated and inactivated. As a result, the downstream signaling proteins of cMET along with the other representative metastatic and angiogenic factors get downregulated. These lead to inhibition of cMET-mediated metastasis and angiogenesis. (Created with BioRender.com).


Introduction
Breast cancer is the second most common cancer worldwide and the leading cause of cancer related death in women (DeSantis et al. 2019) (Song and Farzaneh 2021).It has been commonly observed that even after successful therapeutic interventions, the breast cancer patients often experience cancer relapse (Ahmad 2013).This recurrence of cancer takes place due to the presence of cancer stem cells (CSCs), which comprise of a small subpopulation of cells within a tumor, and possess high self-renewal, drug efflux and DNA repair capabilities (Saeg and Anbalagan 2018).CSCs also promote tumor progression, highlighting their roles in different stages of cancer, which include metastasis (Krishnapriya et al. 2019), angiogenesis (Huang and Rofstad 2017) and resistance to chemotherapy (Ayob and Ramasamy 2018).Conventional cancer treatments have not been as effective as desired in part due to intratumor heterogeneity and the survival of CSCs (Ayob and Ramasamy 2018).Hence, researchers are currently focussing on breast cancer stem cells (BCSCs) as a potential target for the development of novel breast cancer therapies (Song and Farzaneh 2021) (Saeg and Anbalagan 2018).cMET(MET proto-oncogene) is a cell-surface protein tyrosine kinase, aberrantly expressed in various solid tumors, including breast cancer (Tchou et al. 2017).It comprises 50 kDa alpha-chain and 140 kDa betachain and is located on chromosome 7q21-31.It has a natural ligand, hepatocyte growth factor (HGF), which binds to cMET, resulting in cMET dimerization and trans-phosphorylation of the tyrosine residues in the intracellular domain (Gaule et al. 2014).Autophosphorylation of these domains leads to activation of downstream signaling cascade (Puccini et al. 2019).Catalytic activation of downstream proteins regulates several cellular processes, including cancer cell survival, proliferation, motility, invasion, epithelial-to-mesenchymal transition (EMT), and angiogenesis (Gaule et al. 2014) (Puccini et al. 2019) (Ho-Yen et al. 2015).Although cMET is involved in tumor metastasis and angiogenesis in breast cancer (Mitra et al. 2020), studies regarding the role of cMET in CSCs are limited.Among the different ways to inhibit cMET, inhibition of autophosphorylation of cMET has emerged as one of the major ways to develop an anticancer therapy (Puccini et al. 2019).
Quinacrine (QC) is a plant-based, safe, bioactive compound, derived from quinine and has shown anti-cancer activity in different types of cancer, including breast cancer (Das and Kundu 2021) (Oien et al. 2021) (de Souza et al. 1997).Previous studies suggest that QC inhibits cancer cell growth through various mechanisms such as the S-phase cell cycle arrest via inhibition of the topoisomerase activity, causing apoptosis, regulation of autophagy, inhibition of NF-kβ and Wnt-TCF signaling through adenomatous polyposis coli (APC) gene (Das and Kundu 2021) (Oien et al. 2021).QC is a smart and highly efficient molecule and it specifically targets CSCs without harming normal cells (Das and Kundu 2021) (Dash et al. 2021) (Dash et al. 2021) (Siddharth et al. 2016b).Early reports also suggested that QC has anti-metastatic and anti-angiogenic potentialities in cancer (Das and Kundu 2021) but it needs further research to explore the detailed mechanism.
Earlier, conventional cMET-targeted chemotherapies exhibited a lot of side effects such as hypertension, loss of appetite, leukocytopenia, fatigue, and anorexia (Bouattour et al. 2018) (Mo and Liu 2017).Therefore, targeting cMET with a natural, safe and bioactive drug would be a plausible approach for an effective anti-CSCs therapy.Interestingly, a previous study indicated that QC is a tyrosine kinase inhibitor (Guo and Stark 2011) and hence we hypothesize that there will be an effective inhibition of cMET in BCSCs upon treatment with QC.However, the role of QC in inhibition of cMET-mediated metastasis and angiogenesis in breast cancer has not been examined yet.Here, we have studied and analyzed the underlying biochemical basis for the effect of QC on cMET-mediated metastasis and angiogenesis in breast cancer using in vitro (highly metastatic breast cancer stem cells; mBCSCs), ex vivo (patient-derived breast cancer stem cells; PDBCSCs), and in vivo (xenograft mice), model systems.

Development of metastatic model in breast cancer cell lines
The highly metastatic breast cancer model was developed to generate CSCs according to the protocol mentioned earlier (Siddharth et al. 2016a) with little modification.In the current study, MCF-7 and MDA-MB-231 cells were trypsinized and re-suspended in serum-free media supplemented with 10 ng/ml basic fibroblast growth factor (bFGF), 20 ng/ ml epidermal growth factor (EGF), 5 μg/ml insulin, 0.4% bovine serum albumin (BSA) and 200 μM CoCl 2 (to induce a hypoxic condition).The cells were found to grow as adherent monolayer and remained stable till 3-5 passages.These adherent monolayer cells were reported as the quiescent (Q) phase cells.Then, the 'Q' cells were cultivated in fresh serum-free media (supplemented with above-mentioned growth factors and CoCl 2 ) on an ultra-low attachment plate and these cells could grow for the next 7-15 days.Under the serum and oxygen deprived conditions, most of the cancer cells died while a small population of cancer cells with mesenchymal stem cell properties survived and floated as viable clusters of spheres termed as mammospheres (MAMMO).These MAMMO cells were centrifuged at 850 rpm for 5 min and dissociated by using trypsin-EDTA followed by giving little mechanical force using a Pasteur pipette.Finally, the resulting single suspension cells were grown in regular serum-containing media (DMEM for MCF-7 and RPMI-1640 for MDA-MB-231).These cultured adherent monolayer cells were termed as Post-Epithelial-to-Mesenchymal-Transition (PEMT) cells, which had first undergone epithelial-to-mesenchymal transition and then mesenchymal-to-epithelial transition.These cells were characterized and they were found to be a CSCs-rich population, possessing highly metastatic and aggressive properties compared to parental cells.These highly metastatic breast cancer stem cells (mBCSCs) were used for further experimental purposes.

Development of CSCs enriched metastatic model from patient-derived breast cancer cells
Breast primary tumor tissues were collected and processed for the formation of the metastatic model following the protocol mentioned earlier (Siddharth et al. 2016a) (Pradhan et al. 2021).Briefly, breast tumor samples (invasive duct carcinoma) were collected from Acharya Harihar Regional Cancer Centre, Cuttack, Odisha, as per the Hospital Review Board, under ethical guidelines of the hospital.Tumor tissues were processed for the formation of the metastasis model.Briefly, the tumor tissues were washed in 1XPBS and chopped into tiny pieces in a media which contained a mixture of antibiotics (0.14 mM ampicillin, 0.26 µM Amphotericin B and 7.54 µM ciprofloxacin).Then, the tissue fragments were treated with 0.1% collagenase along with 50U/ mL dispase.Next, the sample was incubated in 37 °C water bath for at least 2 h with continuous rotating motion.Then, the cells were sieved through 40 µm cell strainer to remove the fat cells, centrifuged at 1000 rpm for 10 min, seeded onto 60 mm dishes in DMEM-F12 containing 20% FBS, 1.5 mM l-glutamine and 2% antibiotic (100U/mL penicillin and 10 mg/mL streptomycin) and allowed to grow for 5 to 8 days under regular observation.Then, the metastatic model from patient-derived breast cancer cells was developed using the procedure described above and characterized before using them for further experimentation.

Preparation of conditioned media (CM)
According to the protocol discussed earlier, conditioned media were collected from the cultured MCF-7-PEMT, MDA-MB-231-PEMT, HUVECs, etc. (Chatterjee et al. 2021).Briefly, 1 × 10 6 cells were seeded in 60 mm sterile cell culture dishes and incubated for 48 h without any drug treatment.On the other hand, another set of cells were seeded and treated with QC for 48 h.Then, the media from these two sets of cells were collected from different culture conditions and centrifuged at 1800 rpm at 4 °C for 3-5 min.The supernatant was kept in a fresh tube and concentrated by using Eppendorf Concentrator plus (Eppendorf, Hamburg, Germany).The final concentrated liquid product was used as conditioned media (CM) in different biochemical assays and preserved at − 20 °C.

Measurement of cell proliferation and viability by MTT assay
The cell viability of QC-treated cells was measured by performing an MTT assay as described earlier (Pradhan et al. 2021).Briefly, exponentially growing MCF-7-PEMT, MDA-MB-231-PEMT, and primary PEMT cells (8000-10,000 cells/well) were seeded in 96 well culture plates and allowed to grow for 70-80% confluency.After the treatment with increasing concentrations of QC (0-10 µM) for 48 h, 0.05% MTT was added to each well and incubated at 37 °C for 5-7 h to form formazan crystals.Next, the formazan crystals were dissolved with 0.2% NP-40 detergent.Then, the color intensity was measured by a microplate spectrophotometer (Berthhold, Germany) at 570 nm wavelength.Similarly, the proliferation rate of cMET overexpressed (OE) MCF7-PEMT and cMET knockdown (cMET-KD) MDA-MB-231-PEMT cells was evaluated and compared them to their respective control.

ELISA to measure the proliferation, metastatic and angiogenic markers
According to our previously described protocol (Pradhan et al. 2021), an indirect enzyme-linked immunosorbent assay (ELISA) was performed to determine the expression of the cell proliferation marker Ki-67, the metastatic and stem cell markers (ALDH-1, CXCR-4, CD-44 and Oct-4) and angiogenesis markers (Ang1, Ang2, VEGF-A and HIF-1α) in respective cell lines after treating with increasing concentrations of QC for 48 h.Briefly, 30 μg of protein antigen (CM) is well mixed with coupling buffer, which was coated onto a 96 well microplate (#3679, Corning, NY, USA).Excess antigen-binding sites are blocked by using a blocking solution.Specific primary antibodies were added to detect the desired antigens followed by HRP conjugated secondary antibodies.The color was developed by adding ABTS substrate solution and the color intensity was measured at 405 nm using a microplate reader (Berthold, Germany).

Western blot analysis
Western blot experiment was performed according to the protocol described earlier (Nayak et al. 2019).The tissue or cell lysate was prepared by using RIPA lysis buffer; then, SDS-PAGE was used to separate 80 µg of protein samples which were then transferred to PVDF membrane for detection.Specific antibodies were applied to detect the protein of interest.Relative fold change in protein expression levels was measured by densitometry analysis by using a UVP GelDoc-IT® 310 system and represented as numerical values above each protein band panel.In all the experiment of western blotting, GAPDH was used as loading control.The numbers in the figures above each blot represent fold changes in protein expression analyzed by ImageJ software.

Matrigel invasion assay to investigate the invasive potentiality of cells
Matrigel invasion assay was performed according to the protocol described previously (Pradhan et al. 2021).In short, 3 × 10 5 cells were suspended in 100 µL of serumfree media and seeded in a 24 well transwell plate (#3422, Corning, NY, USA), coated with 20 µL of matrigel (#356,234,BD Biosciences,CA,USA). Next, serumcontaining media was added to the lower chamber and incubated for 24 h.Accordingly, cells were treated with respective IC 50 value of QC for 48 h prior to performing the experiment.The non-invaded cells were removed with cotton swabs and the invaded cells were fixed with 4% paraformaldehyde followed by DAPI staining.The stained cells were counted at 20× magnification in 5 different microscopic fields under the inverted fluorescence microscope (Nikon, Tokyo, Japan).The percentage of invaded cells was calculated and represented graphically.

Wound healing assay to check the cell migration
Cell migration potentiality in cMET-OE-MCF-7-PEMT and cMET-KD-MDA-MB-231-PEMT cells with or without the exposure of QC was examined by using a wound healing assay as described earlier (Pradhan et al. 2021).Briefly, the cells were cultured in 35 mm tissue culture dishes until 90% confluent.A sterile 20µL micro tip was used to make a straight-line scratch in the cell monolayer across the center of the plate.The cells were treated in fresh media with respective concentrations of QC and then the cells were incubated for 30 h to observe the migratory potentiality of the cells.Photographs were captured at 10× magnification under an inverted bright field microscope (Nikon, Japan) at different time points.

In ovo chick chorioallantoic membrane (CAM) assay
In ovo CAM assay was performed to study angiogenesis according to the protocol mentioned earlier (Chatterjee et al. 2021).Briefly, fertilized RIR (Rhode Island Red) chick eggs were purchased from the Central Poultry Development Organization (CPDO, Bhubaneswar).Eggs were incubated in a humidified atmosphere at 37 °C, and after 60 h from fertilization time, a window was made in the egg shell.Two sets of eggs were used in this experiment.In the first set of eggs, the CAM membranes were exposed to different types of CM (100 μg per 5 ml egg fluid) on a sterilized filter.On the other hand, in the other set of eggs, CAM membranes were exposed to CM of QC-treated cancer cells (cells were treated with the respective IC 50 concentration of QC).Both the sets of eggs were incubated for another 24 h.The changes in vascularization in CAM were studied and pictures were taken photographically.Graphical representation of average vascularization on CAM membrane was analyzed using AngioTool software (NIH).The graphical representation was made with GraphPad Prism 5.0 software.

Tubulogenesis assay in HUVECs to check tube formation
Tube formation assay was performed in HUVECs according to the protocol described earlier (Chatterjee and Kundu 2020).Briefly, 2 × 10 3 HUVECs were seeded in a 24-well matrigel coated plate and then two sets of HUVECs were used in this experiment.In the first set, the HUVECs were exposed to different types of CM (100 μg/ 3 ml, contains most of angiogenic stimuli).On the other hand, the other set was exposed to CM of QC-treated cancer cells (cells were treated with the respective IC 50 concentration of QC).Both the sets were incubated for another 48 h.To clearly visualize the formation tube-like structures, the cells were stained with acridine orange.Photographs were captured in different microscopic fields at 40× magnification (Evos Fluorescence Microscope, Thermo Fisher Scientific, MA, and USA).Average tube lengths were analyzed by using AngioTool software (NIH).The graphical representation was made with GraphPad Prism 5.0 software.

Gelatin zymography to check the expression of matrix metallopeptidases
As previously reported, gelatin zymography was performed to check the extracellular matrix metallopeptidase (MMP-9 and MMP-2) activity (Chatterjee et al. 2021).Briefly, gelatin works as a substrate for MMP-9 and MMP-2 which are gelatinase in nature.In the first set of experiment, 1 × 10 6 HUVECs alone and CM supplemented HUVECs were grown in a 6-well plate for 48 h.In another set, HUVECs were incubated with CM of QC-treated cancer cells to check the effect of QC.Then, the supernatants from both the sets of experiment were collected and each sample containing 40 µg of protein was separated by performing SDS-PAGE containing gelatin co-polymerized with polyacrylamide gel to detect MMP-9 and MMP-2 expression.After that, gels were washed twice with washing buffer (2.5% Triton X-100, 5 mM CaCl 2 , 50 mM Tris HCL, 1 μM ZnCl 2 in dH 2 O), incubated in 37 °C shaker incubator with incubation buffer (1% Triton X-100, 5 mM CaCl 2 , 50 mM Tris HCL, 1 μM ZnCl 2 in dH 2 O) for 16-21 h and stained with 1% CBBR-250.After destaining the gels in the destaining solution, the regions, where MMP-9 and MMP-2 were active, appeared as white bands in contrast to the dark background.

Immunofluorescence assay
Intracellular phospho-cMET (Y1234 & Y1356) was detected using fluorescence-based immunocytochemistry according to our previous method (Pradhan et al. 2021).In brief, cells were seeded on coverslips and treated with QC for 24 h.Next, the cells were fixed with acetone:methanol (1:1,v/v) for 15 min at − 20 °C.After fixation and blocking, the cells were incubated overnight with the respective primary antibody at 4 °C followed by washing with 1XPBS.Then, TRIT-C conjugated secondary antibody was added and incubated for 2 h at room temperature (RT).After washing twice with 1XPBS, nuclei were counterstained with DAPI.The images were taken using an inverted fluorescence microscope (Nikon, Tokyo, Japan) at 20× magnification.

Effect of QC in tumor regression, metastasis and angiogenesis by using xenograft mice model
Tumor induction in mice and all related experiments were performed as per the protocol discussed previously (Pradhan et al. 2021) (Chatterjee et al. 2022).Briefly, the Institutional Animal Ethical Committee approved all the animal related work and experimental protocols (IAEC, KIIT Deemed to be University, Bhubaneswar, India).Briefly, 6-7-week-old female BALB/c mice were taken for the experiments (Dash et al. 2021).PEMT cells developed from patient-derived breast cancer cells (1 × 10 7 in 200 µL of freshly prepared sterile PBS) were injected into the right mammary fat pads of the mice.After tumor formation, the tumor containing mice were treated with QC (20 mg/kg/every alternate day) by oral administration with a feeding needle.The animals were physically monitored to keep track of their health, body weight and tumor growth, etc.After completion of treatment period, the animals were sacrificed and the samples (breast fat pad including tumor) were collected for further experimentations.

Immunohistochemistry (IHC) of mice and patient tumor tissues
H&E staining and IHC were performed as per protocol described previously (Sethy et al. 2021) (Sinha et al. 2022).Briefly, paraffin-embedded specimens were sectioned (at 5 µm thick) and then mounted on charged slides.After heating at 60 °C for 30-40 min, the slides were de-waxed with xylene and rehydrated by immersing in decreasing series of alcohols concentrations (100%, 90% and 70%, respectively).Then, the sections were immersed in hematoxylin followed by eosin staining and washed in rinsing water.After that the sections were dehydrated by dipping in increasing concentrations of alcohol (70%, 90%, and 100%) followed by xylene and acetone for 2 min each.Then, the slides were mounted with DPX and coverslips.The images were taken in a brightfield microscope at 20× magnification (Leica DM200, USA).
In IHC, rehydrated tissue sections (3 µm thick) were washed in 1X PBS, and then, the antigen was retrieved by citric acid buffer (pH 6).After that, nonspecific sites and endogenous peroxide activity were blocked in the slide by 5% fetal bovine serum (FBS) and hydrogen peroxide.
Then, the sections were incubated with respective primary antibodies at 4 °C for overnight.The primary antibody was removed and washed 3 times with 1X PBS.Next, the slides were incubated with HRP-conjugated secondary antibody at room temperature for 1 h.Then, slides were washed with 1X PBS and immunoreactivity was observed using 3,3-Diaminobenzidine (DAB) peroxidase substrate kit (SK-4100, Vector Laboratories, CA, USA) followed by counterstaining with hematoxylin.The sections were mounted with DPX and left for air drying.The images were taken at 20X magnification using a brightfield microscope (Leica DM200, USA).Protein expression in IHC of mice tumor tissue was quantitatively analyzed by Fiji (ImageJ-2) software.

Statistical analysis
The given images and data in the figures have been used as one of the representatives of three independent experiments.Statistical analysis was carried out using Graph Pad Prism 5.0 software (USA).Results represented here were the mean ± standard deviation (SD) of three independent experiments.Data were analyzed using one-way ANOVA followed by Bonferroni's all pair comparison test.Statistical significance was designated as 'ns' (non-significant, P > 0.05), along with *P < 0.05, **P < 0.005 and ***P < 0.0001 represent significant data.

Cytotoxic potentiality of QC in highly metastatic breast cancer stem cells (mBCSCs)
To check the cytotoxic potentiality of QC in mBCSCs, the metastatic model from breast cancer cell lines, MCF-7 and MDA-MB-231, was developed.The formation of post-epithelial-to-mesenchymal-transition (PEMT) cells was pictorially represented for both the cell lines (Fig. 1Ai and Aii).Next, to validate the metastatic model and CSCs properties of MCF-7-PEMT and MDA-MB-231-PEMT cells, the expression of the stemness markers was assessed by western blot analysis.There was a complete loss in the expression of E-cadherin (an epithelial marker) and increase in the expression of Vimentin (a mesenchymal marker) in mammospheres (MAMMO).Additionally, high E-cadherin and abolishment of Vimentin expressions in PEMT cells confirmed the development of a true metastatic model (Fig. 1Bi and Bii).Moreover, a significant increase in the expression of stemness markers such as N-cadherin, CD-44 and CD-133 was observed in both MCF-7-PEMT and MDA-MB-231-PEMT cells compared to the parental cells (Fig. 1Bi and Bii).The invasive potential of MCF-7-PEMT and MDA-MB-231-PEMT cells was found to be increased up to 60% and 75%, respectively (P < 0.0001) compared to their parental cells (Fig. 1 Ci).Next, the ELISA result showed a significant increase (4-fold, P < 0.0001) in the expression of proliferation marker, Ki-67 in both PEMT cells in comparison with their parental cells (Fig. 1Cii).Further, the expression of stemness markers (CD-44 and CXCR-4) in both PEMT cells was found to be higher (P < 0.0001) compared to their respective parental cells (Fig. 1Ciii and Civ).The chemical structure of QC was obtained from PubChem (Fig. D).Next, MTT cell viability assay was performed and IC 50 values of QC in MCF-7-PEMT and MDA-MB-231-PEMT cells were found to be 8 µM and 9 µM, respectively (Fig. 1Ei and Eii).
Next, to know the effect of QC on the metastatic and angiogenic markers, the cells were treated with respective IC 50 concentrations of QC.When cMET-OE-MCF-7-PEMT cells were treated with 8 µM QC, the expression of representative metastatic markers (CXCR-4 and CD-44) and angiogenic markers (Ang1 and VEGF-A) were decreased approximately 1.6-fold in comparison with untreated control (Fig. 2E).On the other hand, when cMET-KD-MDA-MB-231-PEMT cells were exposed to 9 µM QC, the expression of the metastatic markers was found to be decreased up to 1.4-fold compared to untreated control (Fig. 2E).

QC deregulated cMET-mediated migration, invasion and angiogenic properties of BCSCs
A systematic study was carried out to determine the role of QC on cMET-mediated metastasis and angiogenesis.Since migration and invasion are the major hallmarks of cancer (Hanahan and Weinberg 2011), cell migration assay and matrigel invasion assay were performed to investigate the metastatic potentiality of cMET-OE-MCF-7-PEMT cells and cMET-KD-MDA-MB-231-PEMT cells after QC treatment.It was observed that increasing concentration of QC reduced the migration rate of these cells.The cMET-OE-MCF-7-PEMT cells exhibited better migration ability than MCF-7-PEMT cells (Fig. 2Gi).On the other hand, cMET-KD-MDA-MB-231-PEMT cells showed comparatively less migration ability than MDA-MB-231-PEMT cells (Fig. 2Gii).But cell migration ability of the cells was reduced with increasing concentration of QC in all conditions (Fig. 2G).
Then, in ovo CAM assay was performed to check the blood vessel formation following the addition of different CM (100 μg/5 ml egg fluid) to the CAM (Fig. 2I).A significantly higher amount of vascularization was observed when eggs were incubated with the CM of cMET-enriched cells (cMET-OE-MCF-7-PEMT and MDA-MB-231-PEMT), in comparison to CM of cMET-KD-MDA-MB-231-PEMT and MCF-7-PEMT cells.But when eggs were incubated with CM of QC-treated cells, the average vascularization was not significantly induced (Fig. 2Ii-Iix).The bar graph represents the average vascularization in different conditions (Fig. 2Ix).
It is a well-known fact that endothelial cell proliferation increases in response to angiogenic stimuli.Therefore, to check the role of QC in cMET-induced tube formation, HUVECs were incubated with different types of CM (100 μg/3 ml) as mentioned in Fig. 2J.A characteristic growth pattern of HUVECs was found after 72 h of CM addition.It was noticed that the average tube length was significantly increased (approximately 6-fold, P < 0.0001) when HUVECs were incubated with CM of cMET-enriched cells (cMET-OE-MCF-7-PEMT and MDA-MB-231-PEMT), in comparison to untreated control.Additionally, CM from MCF-7-PEMT cells and cMET-KD-MDA-MB-231-PEMT cells showed a 3.8-fold and 2-fold increase in average tube length of HUVECs, respectively, in comparison to untreated control (Fig. 2J).Then, to check the effect of QC on tube formation ability of HUVECs, CM of QC-treated cells was added.It was noted that tube formation was not significant when HUVECs were incubated with QC-treated CM, in comparison to untreated control (Fig. 2J).The bar graph represents the average tube length in different types of CM treated conditions (Fig. 2Ji).
Next, the expression of representative angiogenic markers, MMP-9 and MMP-2, was monitored by performing gelatin zymography.The CMs were collected from different types of CM-induced HUVECs and subjected to gelatin zymography as mentioned in Fig. 2K.Higher expression of MMP-9 and MMP-2 was noticed when HUVECs were incubated with CM of cMET-OE-MCF-7-PEMT and MDA-MB-231-PEMT cells in comparison to that observed when incubated with the CM of cMET-KD-MDA-MB-231 and 1 3 MCF-7-PEMT cells (Fig. 2Ki).The expression of MMP-9 and MMP-2 was not significantly changed when HUVECs were incubated with the CM of QC-treated cells (Fig. 2Kii).

Molecular mechanism of cMET inhibition by QC
Till now, the effect of QC in cMET-mediated metastasis and angiogenesis was studied.Next, to study the underlying mechanism of QC-mediated cMET inhibition in BCSCs, specific experiments were carried out.Phosphorylation of cMET is the key step for activating the downstream proteins in this signaling cascade (Zhang et al. 2018).Therefore, the expression of cMET, its intracellular domain phospho-cMET (p-cMET-Y1234 and p-cMET-Y1356) and the downstream signaling proteins were assessed by western blotting after treating the MDA-MB-231-PEMT cells with increasing concentrations of QC (Fig. 3A).Interestingly, unaltered expression of cMET at the basal level was observed after treatment with QC (Fig. 3A).However, a 3.5-and 5-fold decrease in the levels of p-cMET-Y1234 and p-cMET-Y1356, respectively, were observed after exposure with 9 µM QC.Additionally, other downstream proteins of cMET signaling cascade also got downregulated after QC exposure.Similar results were obtained when cMET-OE-MCF-7-PEMT cells were exposed to QC in a dose-dependent manner (Fig. 3B).
Next, immunofluorescence assay of p-cMET after QC treatment was also performed to confirm the above findings.Upon QC treatment, a dose-dependent reduction (P < 0.0001) in the expressions of p-cMET-Y1234 and p-cMET-Y1356 was noted in MDA-MB-231-PEMT cells (Fig. 3C).The bar graph represents the reduction (P < 0.0001) in the relative mean intensity of p-cMET expression with increasing concentrations of QC, in comparison to control (Fig. 3Ci and Cii).Similar observation was also noted in cMET-OE-MCF-7-PEMT cells after QC treatment (Fig. 3D, Di, and Dii).

Clinical significance of cMET and downregulation of metastatic and angiogenic markers by QC in PDBCSCs
To delineate the clinical significance of cMET in breast cancer, malignant human breast cancer tissues of different grades (Grade-I, II and III) along with their adjacent normal tissues were collected and processed for experimentation.The hematoxylin and eosin (H&E) stained slides suggested the cancerous nature or invasive histology with increasing grades of cancer tissue compared to uniform well differentiated histology of normal tissue (Fig. 4Ai).The immunohistochemistry (IHC) result showed a significant increase (P < 0.0001) in cMET, p-cMET-Y1234 and p-cMET-Y1356 expression with the increase in cancer grades (Fig. 4Aii, Aiii, and Aiv).Expressions of these proteins were quantified with Fiji-ImageJ2 software and represented in a table (Fig. 4Av).Data of the quantified protein expressions were also represented by bar diagram with statistical significance (Fig. 4Avi).An increased expression of cMET and both the p-cMET residues with increasing grades of cancer were also observed in western blot analysis.Notably, more than twofold, increase in cMET and p-cMET (Y1234 and Y1356) expression were observed in Grade-III tissue lysate compared to Grade-I tissue lysate (Fig. 4B).
Next, the isolated cells from Grade-III tumor tissues were cultured and grown as adherent monolayer following the development of the metastatic model to generate PDBCSCs (Fig. 4C).To characterize the metastatic model, expressions of E-cadherin and Vimentin as well as other stem cell markers were checked at different stages of the model in patient-derived breast cancer primary cells.Notably, very low expression of E-cadherin and a higher expression of Vimentin were observed in MAMMO (Fig. 4D).Additionally, the expression of other representative stemness markers (N-cadherin, CD-44 and CD-133) was significantly increased (> 2-fold) in breast cancer primary PEMT cells (Fig. 4D).Moreover, the rate of invasion (4-fold) and proliferation (2-fold) as well as the expressions of CXCR4 (2-fold) and CD-44 (2-fold) were also found to be elevated in breast cancer primary PEMT cells, in comparison to control (Fig. 4Ei-iv).These breast cancer primary PEMT cells were referred to as PDBCSCs.Next, these cells were exposed to increasing concentrations of QC and IC 50 concentration was found to be 5 µM (Fig. 4F).
Finally, to examine the expression of several metastatic and angiogenic markers as well as cMET, p-cMET and the downstream proteins in PDBCSCs, the cells were treated with increasing concentrations of QC (0, 1, 3, 5, 7 and 9 µM) and processed for western blot analysis.Even after treatment with increasing concentrations of QC, the basal cMET expression remained unchanged in PDBC-SCs.Interestingly, a 10-fold decrease in the expression of p-cMET (Y1234 and Y1356) was observed after 5 µM QC exposure with respect to untreated control (Fig. 4G).Similarly, the expression of other representative metastatic markers (CXCR-4, ALDH-1), angiogenic markers (Ang1, VEGF-A) and downstream signaling proteins (STAT3, FAK, PI3K, AKT, mTOR) of cMET got downregulated in PDBCSCs after treatment with QC in a dose-dependent manner as compared to untreated control (Fig. 4G).

QC reduces the expression of metastatic and angiogenic markers in mice xenograft model
To check the anti-metastatic and anti-angiogenic potentiality of QC in vivo, the mice xenograft model was developed using female BALB/c mice.Tumor formation was noticed after 10 days of breast cancer primary PEMT cells implantation.Day by day, the body weight of the mice significantly decreased and tumor volume was increased (P < 0.0001) (Fig. 5A and B).Notably, after QC treatment, a recovery in lost body weight and reduction of tumor volume was found (P < 0.0001) (Fig. 5A and B).Then, after sacrificing the mice, the tumor tissues were collected for further experiments.H&E staining revealed normal and cancerous histology of mice tissue (Fig. 5C).Next, in IHC, increased expression of cMET, p-cMET (Y1234 & Y1356), metastatic markers (CD-44 and CXCR-4) and angiogenic markers (Ang1 and VEGF-A) in tumor tissues was observed in comparison with breast-fat-pad tissue of PBS injected mice.However, upon QC treatment, the expression of the above-mentioned metastatic and angiogenic markers got reduced, in comparison to untreated tumor tissue but the expression of remained unchanged cMET (Fig. 5D).Protein expression in IHC of mice tumor tissue was quantitatively analyzed by Fiji-ImageJ-2 software (Fig. 5Di).

Discussion
Although cMET has been used as a potential target for cancer therapy, numerous side effects and lack of specificity of the drugs limited the effectiveness of cMETtargeted anti-cancer therapy.QC, a plant-based bioactive agent, possess anti-CSCs potentiality and has been found to exhibit anti-tyrosine kinase activity (Das and Kundu 2021) (Guo and Stark 2011).Here, it was found that QC inhibited the cMET-induced metastasis and angiogenesis by dephosphorylating autoregulatory domain of cMET and downregulating the downstream signaling cascade.To corroborate this observation, we have systematically performed several biochemical assays using in vitro, in ovo, ex vivo, and in vivo pre-clinical model systems.
First, the highly metastatic BCSCs (mBCSCs) model of MCF-7 and MDA-MB-231 cells was developed and characterized by several biochemical assays.Similar to the earlier reports (Siddharth et al. 2016b), our data also suggested that the PEMT cells exhibit CSCs properties and can be referred to as CSCs (Fig. 1A-C).Next, decreased expression of representative metastatic markers (CXCR-4, CD-44, ALDH-1, and Oct-4) and angiogenic markers (Ang1, Ang2, VEGF-A and HIF-1α) after QC treatment in MDA-MB-231-PEMT cells as well as MCF-7-PEMT suggested that QC might have anti-metastatic and antiangiogenic potentialities in mBCSCs (Fig. 1D-G and Supplementary Fig. A, B).
Overexpression of cMET is significantly associated with poor survival of breast cancer patients, especially in the triple-negative breast cancer (TNBC) subgroup (Yan et al. 2015).Due to the increased expression of cMET in MDA-MB-231 cells, these cells are highly invasive, compared to MCF-7 cells (Parekh et al. 2018) (Parr and Jiang 2001).In our findings, the relatively lower expression of cMET in MCF-7-PEMT and higher expression in MDA-MB-231-PEMT cells support the above agreement.To understand the detailed role of cMET in cancer progression through metastasis and angiogenesis, several experiments were performed after overexpressing cMET in MCF-7-PEMT and knocking down it in MDA-MB-231-PEMT cells.The proliferation rates, and the expression of representative metastatic and angiogenic markers were higher in cMET-OE cells than in its parental cells, whereas opposite results were found in cMET-KD cells (Fig. 2A-D).After QC treatment, significant reduction in the expressions of the metastatic and angiogenic markers was found in cMET-OE condition but a lesser significant result was noted in cMET-KD condition (Fig. 2E and F).This suggested that QC is plausibly interfering with the metastatic and angiogenic processes through the inhibition of cMET.QC also inhibited cell migration of cMET-OE cells more significantly than that observed in cMET-KD cells (Fig. 2G and H).Thus, it appeared that cMET might be playing a pivotal role in cell migration and invasion of mBCSCs, which can be inhibited by QC.
It is a well-known fact that cMET promotes the secretion of important angiogenic factors and inhibition of cMET is believed to exert an anti-angiogenic effect on tumor cells (Wang et al. 2016).Cancer cells can directly produce angiogenic induction to endothelial cells by secreting several angiogenic factors.These factors activate the endothelial cells and stimulate the generation of a new blood vessel from an existing one (Lopes- Bastos et al. 2016).Endothelial cells also secrete MMPs (mainly MMP9 and MMP2), which exhibit invasive and morphogenic events during angiogenesis (Taraboletti et al. 2002).To investigate the role of QC on cMET-mediated angiogenesis, a series of experiments were conducted.In in ovo CAM assay, CM from cMET-enriched cells (cMET-OE-MCF-7-PEMT and MDA-MB-231 PMET) significantly induced the vascularization in CAM, whereas comparatively less vascularization was found when CM from cMET-KD-MDA-MB-231-PEMT and MCF-7 PMET cells were added.On the other hand, CM from QC-treated cancer cells failed to exert any significant change in vascularization on CAM (Fig. 2 IIi-Iix).Similarly, the results of tube formation assay (Fig. 2J and Ji) and gelatin zymography (Fig. 2Ki and Kii) also supported the above findings.The expression of Ki-67 (ii), CXCR-4 (iii).and CD-44 marker was measured by ELISA (iv).F Cell viability assay.G Expression of cMET, phospho-cMET, representative metastasis, angiogenesis markers and downstream signaling proteins of cMET were assessed after treatment with increasing concentrations of QC in patient-derived breast cancer PEMT cells The above results indicated that QC has a crucial role in the deregulation of cMET-mediated angiogenesis.
Next, it was important to understand how QC is inhibiting cMET.The experimental results revealed that the cMET level was unchanged and the expression of p-cMET (Y1234 & Y1356) and downstream signaling proteins was decreased with the increasing concentrations of QC (Fig. 3A and B).The immunofluorescence data also supported the above observation (Fig. 3C and D).This implied that QC inhibits the phosphorylated form of cMET without altering its basal level expression.Taken together, data suggested that QC inhibited the metastasis and angiogenesis process by inhibiting the activation of cMET in in vitro BCSCs cells.Now, the question arises that whether QC also affects these processes in the tumor microenvironment (TME, where many types of cell population are present), similar to its effects on BCSCs.To address this question, we carried out multiple experiments using ex vivo model of  5D) was analyzed by Fiji-ImageJ2 software.E Expression of cMET, phospho-cMET, representative metastasis, and angiogenesis markers, and downstream signaling proteins of cMET patient-derived breast cancer cells.The clinical significance of cMET was also analyzed and it was observed that the expression of cMET and phospho-cMET (Y1234 and Y1356) was enhanced with increasing grades (Grade-III > Grade-II > Grade-I) of breast cancer patient samples (Fig. 4A and B).Next, a patient-derived-breast-cancerstem-cells (PDBCSCs) model was developed and validated, where PEMT cells were found to exhibit CSCs properties (Fig. 4C-E).A significant reduction was observed in the expression of metastatic and angiogenic markers, p-cMET (Y1234 and Y1356), and downstream signaling proteins of cMET after treating the PEMT cells with increasing concentrations of QC while the basal cMET level was unaltered (Fig. 4F and G).Thus, QC was not only found to effectively inhibit cMET-mediated metastasis and angiogenesis in BCSCs, but also it affected these processes in PDBCSCs as well.
To further investigate these findings in an animal model, some of the experiments were carried out in in vivo xenograft mice model.In vivo data also showed reduction of tumor volume along with other metastatic and angiogenic markers, p-cMET (Y1234 and Y1356) and unaltered  expression of cMET after QC treatment (Fig. 5).Thus, on the basis of the results obtained in different biochemical assays, it was concluded that QC exhibits anti-metastatic and anti-angiogenic effects in in vitro, ex vivo as well as in vivo model systems through inhibition of cMET.
Based on the above observations, we have schematically depicted the role of QC in inhibiting cMET and cMETmediated metastasis and angiogenesis (Fig. 6).Briefly, ligand binding leads to receptor dimerization and activation of cMET.After that, the representative downstream metastasis and angiogenesis-related proteins are upregulated, and ultimately, the metastasis and angiogenesis processes are induced.But after the administration of QC, it dephosphorylates and inactivates cMET.As a result, the downstream signaling proteins of cMET, along with the metastatic and angiogenic factors, are downregulated and the dephosphorylated cMET fails to activate the metastasis and angiogenesis processes.
The current study showed a promising approach to inhibit breast cancer progression through QC-mediated cMET inhibition but more detailed research is required before this drug can be used in clinical settings.Previous studies have reported that QC might inhibit cancer progression in several other ways (Oien et al. 2021) (Das and Kundu 2021), but cMET inhibition is one of the ways to stop breast cancer progression.All the data was generated using the breast cancer cell lines so before further processing it is utmost necessary to conduct experiments using other cancer cells or even animal/human system.Moreover, based on the current findings, we could not apply this drug in a clinical setting as this study is based on pre-clinical model systems.Thus, several systematic research studies are required to be conducted before QC can be used in the clinic.Hence, further research on the anti-cancer action of QC will help us to elucidate how this drug can be used as a monotherapy or a part of combination therapy in cancer patients in future.

Conclusion
In this study, the role of QC in cMET-mediated metastasis and angiogenesis in breast cancer was investigated in pre-clinical breast cancer model systems and it was found that QC has excellent anti-metastatic and anti-angiogenic capabilities through inhibition of cMET.Although this study explores the detailed role of QC's inhibitory action on cMET-mediated metastasis and angiogenesis, more work has to be done in clinical settings before this medication may be safely used to halt cancer progression in humans.

Fig. 1
Fig. 1 Effect of QC on the expression of metastatic and angiogenic markers in mBCSCs.A Morphology of metastatic model at different stages (i and ii).B Validation of metastatic model (i and ii).C Matrigel invasion assay (i).The expression of proliferation marker Ki-67 (ii), metastatic and CSCs marker CD-44 (iii), and CXCR-4 (iv) in metastatic models was measured by ELISA.D Chemical structure of Quinacrine (obtained from PubChem).E IC 50 of QC in MCF-7-

Fig. 2
Fig. 2 cMET-induced metastatic and angiogenic markers got downregulated by QC.A Comparative expression of cMET in MCF-7-PEMT and MDA-MB-231-PEMT.B Overexpression of cMET in MCF-7-PEMT cells and knockdown of cMET in MDA-MD-231 PEMT cells.C Cell proliferation assay.D Expression of representative metastasis and angiogenic regulators.E Expression of representative metastasis and angiogenesis markers upon QC treatment.F Percentage of cell proliferation after QC treatment.G Wound healing potentiality of cMET-OE-MCF-7-PEMT cells (i) and cMET-KD-MDA-MB-231-PEMT cells (ii) upon QC treatment.H Invasive potentiality of the cells with and without QC treatment.I In ovo CAM assay (Ii-ix) Graphical presentation of average vascularization on CAM (Ix).J In vitro tube formation assay.The graphical representation of average tube length (Ji).K Gelatin zymography to check the expression of MMP-9, and MMP-2 after incubating with different CMs in HUVECs (i) and effect of QC treatment (ii) ◂

Fig. 4
Fig.4Effect of QC on the expression of cMET-induced metastatic and angiogenic markers in PDBCSCs.A H&E staining of normal and different grades (I-III) of breast cancer tissue (i).Expression of cMET and phospho-cMET (Y1234 and Y1356) was analyzed by IHC (ii, iii and iv).Quantification of the protein expressions in IHC data (A ii, iii and iv) was analyzed by Fiji-ImageJ2 software and represented in a table(v).Data of the quantified protein expressions were represented by bar diagram with statistical significance (vi).B Expression of cMET and phospho-cMET (Y1234 and Y1356) in different grades of human breast tumor tissue was analyzed by western

Fig. 5
Fig. 5 QC reduces the expression of metastatic and angiogenic markers in mice xenograft model.A Graph representing the average body weight of mice before and after the treatment with QC.B Graph representing the average tumor volume of mice.C H&E staining of normal and cancer breast fat pad tissue of mice.D Immunohistochemical expression of Ang1, VEGF-A, CD-44, CXCR-4, cMET, and

Fig. 6
Fig. 6 Schematic diagram depicting the deregulation of cMET-mediated metastasis and angiogenesis by QC by targeting phospho-cMET.1 Binding of HGF as specific ligand allows receptor dimerization and activation of cMET. 2 Activation of cMET occurs through the phosphorylation of its tyrosine kinase domain (Y1234 and Y1356).3, & 4 Phosphorylation of cMET upregulates other downstream signaling proteins as well as several metastatic and angiogenesis markers.

5, & 6
Induction of metastasis and angiogenesis processes.7, & 8 QC treatment dephosphorylates cMET.9 Dephosphorylation leads to inactivation of whole downstream signaling cascade.10 QC downregulates the downstream signaling proteins of cMET as well as different metastatic and angiogenesis markers.11, & 12 As a result significant reduction of metastasis and angiogenesis take place (Created with BioRender.com)