Effect of Melatonin in Epithelial Mesenchymal Transition Markers and Invasive Properties of Breast Cancer Stem Cells of Canine and Human Cell Lines

Cancer stem cells (CSCs) have been associated with metastasis and therapeutic resistance and can be generated via epithelial mesenchymal transition (EMT). Some studies suggest that the hormone melatonin acts in CSCs and may participate in the inhibition of the EMT. The objectives of this study were to evaluate the formation of mammospheres from the canine and human breast cancer cell lines, CMT-U229 and MCF-7, and the effects of melatonin treatment on the modulation of stem cell and EMT molecular markers: OCT4, E-cadherin, N-cadherin and vimentin, as well as on cell viability and invasiveness of the cells from mammospheres. The CMT-U229 and MCF-7 cell lines were subjected to three-dimensional culture in special medium for stem cells. The phenotype of mammospheres was first evaluated by flow cytometry (CD44+/CD24low/- marking). Cell viability was measured by MTT colorimetric assay and the expression of the proteins OCT4, E-cadherin, N-cadherin and vimentin was evaluated by immunofluorescence and quantified by optical densitometry. The analysis of cell migration and invasion was performed in Boyden Chamber. Flow cytometry proved the stem cell phenotype with CD44+/CD24low/- positive marking for both cell lines. Cell viability of CMT-U229 and MCF-7 cells was reduced after treatment with 1mM melatonin for 24 h (P<0.05). Immunofluorescence staining showed increased E-cadherin expression (P<0.05) and decreased expression of OCT4, N-cadherin and vimentin (P<0.05) in both cell lines after treatment with 1 mM melatonin for 24 hours. Moreover, treatment with melatonin was able to reduce cell migration and invasion in both cell lines when compared to control group (P<0.05). Our results demonstrate that melatonin shows an inhibitory role in the viability and invasiveness of breast cancer mammospheres as well as in modulating the expression of proteins related to EMT in breast CSCs, suggesting its potential anti-metastatic role in canine and human breast cancer cell lines.

Introduction an E-cadherin repressor, via activation of kinase protein GSK3β. Nowadays some studies have shown inhibitory effect of melatonin in cancer stem cells. Thus, previous studies demonstrated that the treatment with melatonin was able to decrease the cell proliferation and induced the cell death by apoptosis and autophagy of colorectal and glioma CSCs [34,35].
However, the specific action of melatonin in cancer stem cells, which result from EMT, has been underexplored [30,36]. Therefore, the objectives of this study were to evaluate the effects of melatonin treatment on modulation of molecular markers: OCT4, E-cadherin, N-cadherin and vimentin, as well as, in the cell viability and invasiveness of the cell mammospheres.

Cell cultures
The canine mammary cancer cell line CMT-U229, previously cultivated [37] was kindly provided by Dr. Eva Hellmén. The histological type this canine cell lineage it is a benign mixed mammary tumor according to WHO (World Health Organization). The human breast cancer cell line MCF-7 was obtained from ATCC (American Type Culture Collection, Manassas, VA, USA).

Mammospheres culture assay
Mammosphere culture was performed as previously described [38]. Cells were grown in the MammoCult medium (Stem Cell Technologies, Vancouver, Canada) supplemented with MammoCult Proliferation Supplements (Stem Cell Technologies, Vancouver, Canada) and plated in ultra low attachment individual wells of 8.96 cm 2 at a density of 10,000 viable cells/ mL and grown for 7 days. The mammosphere formation rate was calculated as previously described [39] using the following equation: (number of mammospheres per well/number of cells seeded per well) × 100%.

Flow cytometry analysis
By using a Guava easyCyte flow cytometer (Millipore), the expression of stem cell markers in a breast cancer panel was distinctly evaluated in cells from mammospheres. The antibodies used were phycoerythrin (PE)-conjugated anti-CD24 (BDbiosciences) and fluorescein isothiocyanate (FITC)-conjugated anti-CD44 (BDbiosciences). Staining was done according to the instructions of the manufacturer and analysed in flow cytometer. For the analysis, 80,000 cells were detected and quantified by percentage (%).

MTT cell viability assay
The cell viability potential of cells treated or not with 1 mM melatonin (Sigma-Aldrich, St. Louis, MO, USA) was measured by MTT assay (3(4,5-dimetiliazol-2-il)-2,5difeniltetrazolium bromide), which is based on the ability of live cells to convert tetrazolium salt into purple formazan.
To MTT assay, the mammospheres were disaggregated by the action of the enzyme trypsin and the cells were plated in a 96 well plate. Thus, the MTT assay was performed with adherent cells provided from mammospheres.
Regarding treatment, melatonin was diluted with 50% of PBS (Phosphate Buffered Saline) solution and 50% of ethanol and the cells received treatment in the absence of light, because the hormone is photosensitive. Melatonin has been shown to act decreasing the cell viability of breast cancer cell lines [40,41]. It should be emphasized here that the concentration of 1 mM melatonin used for the treatment of the cells was defined according to the literature. This is the pharmacological concentration used in several studies about the effects of melatonin in neoplastic cells [42][43][44][45][46][47].
Briefly, individual well of 0.31 cm 2 was inoculated with 100 μL of supplemented medium containing 5 x 10 4 cells. After 24 h, 10 μL MTT stock solution (5 mg/mL, Sigma Chemical Co) was added to each well, and the plates were further incubated for 4 h at 37°C. To solubilize the MTT formazan crystals, the cells were incubated with SDS-HCl (10 mM) (Invitrogen Life Technologies, Carlsbad, CA, USA) at 37°C for 4 h. The absorbance at a wavelength of 570 nm was measured by the ELISA reader (Thermo Plate, Waltham, MA,USA). For the analysis, the cell viability (%) was calculated for all groups compared to control samples. All experimental samples were done in triplicate.

Immunofluorescence staining
Cells from mammospheres, attached to 8-well chamber slides (Sarstedt, Newton, NC, USA) and incubated or not with 1 mM melatonin at 37°C for 24 h, were fixed immediately in 4% paraformaldehyde and permeabilized 0.

Immunofluorescence staining evaluation
Two different photomicrographs were taken at 40X magnification under bright field and the intensity of the staining was quantified by Image J Software (NIH, Bethesda, MD, USA). Each photograph was divided into four quadrants and 20 spots (small circular ROI) were randomly selected in each photomicrography. A negative control section of the corresponding staining was used to measure background activity in S1 Fig

Invasion Assay
The invasion assay was performed with the adherent cells provided from mammospheres.
The invasiveness of CMT-U229 and MCF-7 cells treated or not with melatonin was assessed using transwell chambers with 8 μM pore size membrane (Corning Matrigel Invasion Chamber; Bedford, MA, USA). In the upper compartment of the chamber, approximately 2.5 x 10 4 cells/insert were added into the culture medium without serum, while to the other compartment were added 750 μL of culture medium with the chemoattractant (0.5% and 10% FBS to negative and positive control's, respectively) and 10% FBS was associated with 1 mM melatonin.
The cells were incubated at 37°C in 5% of CO 2 for 24 h. After incubation, the transwell membranes were washed and impermeabilized. The invading cells were stained with hematoxylin and the nonmigrating cells were removed from the upper surface of the transwell membrane by a cotton swab. The counting was made with an inverted optic microscope by putting the insert over a plate containing glycerol at 50%.
Cell invasion (%) of each group was calculated from the average values of cells that migrated and invaded the matrigel membrane in duplicate. The positive control group was taken as 100%. The result was calculated from the difference compared to the positive control group.

Statistical Analysis
The cell lines were separated in two groups, according to the treatments performed (control and melatonin). First, the results were submitted to analysis of normal distribution using Column statistics and the Gaussian distribution test. The average of densitometric analysis, cell viability and the percentage of total cells with migratory and invasive capacity were subjected to Student's t test. All values were expressed as the average ± Standard Deviation (SD). The value of P<0.05 was considered statistically significant. All analyses were performed using GraphPad PRISM5 software (GraphPad Software, Inc., La Jolla, CA).

CMT-U229 and MCF-7 cell lines mammospheres
The mammospheres were generated from the canine mammary cancer cell line CMT-U229 and human breast cancer cell line MCF-7 in MammoCult TM medium (StemCell Technologies) and efficiently formed compact mammospheres (Fig 1). CMT-U229 and MCF-7 cells were continuously capable of forming mammospheres through repeated subcultures in Mammo-Cult TM medium (StemCell Technologies). The mammosphere formation rate was 0.5.

Relationship of CMT-U229 and MCF-7 mammospheres and cancer stem cell population
To confirm the phenotype of breast cancer stem cells (CD44 + / CD24 -/ low ) flow cytometry was performed in mammospheres. For reaction control monolayer of CMTU-229 and MCF-7 cells were used. Our results confirmed the phenotype of breast cancer stem cells. The results indicated that the mammary stem cells population for CMTU-229 cell line was consisted of 26.89% of cell positive for the anti-CD44 antibody and 6.75% for the anti-CD24 antibody (

Melatonin reduces cell viability of CMT-U229 and MCF-7 mammospheres
Thereby, it was performed an MTT assay to estimate the number of viable cells after the treatment with 1 mM melatonin for 24 h. The viability of cells from mammospheres, in both cells lines, was significantly decreased after melatonin treatment when compared to control groups (P = 0.0147, P = 0.0286, respectively; Fig 3).

Melatonin induces differential proteins expression in EMT mammospheres process
In order to examine the effects of melatonin on breast cancer stem cells from canine and human cell lines, the protein expression of the cancer stem cell marker OCT4, the epithelial marker E-cadherin, the mesenchymal markers N-cadherin and vimentin, were measured in cells from mammospheres.
In CMT-U229 cells treated with 1 mM of melatonin, OCT4 protein expression was significantly decreased compared to the control group (P = 0.0426; Fig 4A and 4E) and E-cadherin protein expression increased after melatonin treatment compared to the control group (P = 0.0002; Fig 4B and 4E). On the other hand, N-cadherin and vimentin protein expression were significantly decreased in the melatonin treated cells compared to the control groups (P = 0.0002; Fig 4C and 4E and P = 0.0411; Fig 4D and 4E, respectively).
For MCF-7 cells, OCT4 protein expression was also decreased in melatonin treated cells compared to the control group (P = 0.0001; Fig 5A and 5E) and E-cadherin protein expression

Melatonin decreased migration and invasion of CMT-U229 and MCF-7 mammospheres
To confirm the above-mentioned effect of melatonin on canine and human mammospheres, CMT-U229 and MCF-7 cells were treated with 1 mM of melatonin for 24 h and it was verified reduction in migration and invasion in both cell lines (40.4% for CMT-U229 and 56.3% for MCF-7), when compared to the positive control (P = 0.0017, P = 0.0377, respectively; Fig 6A-6D).

Discussion
The main purpose of this study was to evaluate the effects of melatonin treatment in breast CSC of canine and human cell lines and the results demonstrated that melatonin has similar activity in the canine CMT-U229 and human MCF-7 cell lines. This fact reinforces the concept that breast tumors in women and female dogs have similar biological characteristics and dogs tumors can be used to better understand the pathogenesis of this disease and valid comparative models for the study of breast tumors [48,49].
The presence of stem cells in canine [19,37,[50][51][52][53][54] and human [19,52] breast cancer cell lines have been demonstrated in several studies. CSC exhibit surface markers, whereby the subpopulation of stem cells can be separated from non-stem cells [50]. According to Dontu et al. [12] stemness' of tumor cells is measured in vitro by its ability to form mammospheres. As can be seen in this study, canine and human breast cancer efficiently formed mammospheres, consisting of CD44 + /CD24 low/cells, which confirms the cancer stem cell phenotype. The selection of CSCs from both lineages was performed in serum-free culture medium permissive for growing of stem cells.
Breast cancer cells with CD44 + /CD24 low/surface phenotype have tumor initiating properties with pluripotency characteristics and invasive capacity. The transition from an epithelial phenotype (CD44 -/CD24 + ) to a mesenchymal phenotype (CD44 + /CD24 low/-) enables the cell to move from the primary tumor to metastatic site [55]. Ponti et al. [10] found that breast cancer cell line that grown as spheroids also had CD44 + /CD24 low/phenotype and expressed the transcription factor OCT4. OCT4 is a member of the POU family, expressed in embryonic stem cells, germ cells and human stem cells and it is responsible for maintaining an undifferentiated state in the cells [56]. This transcription factor is an essential regulator for self-renewal and pluripotency of embryonic stem cells. Since CSCs share features with embryonic stem cells, it is suggested that transcription factors are expressed commonly in both cells [57]. The expression of OCT4 has an important role in carcinogenesis and provides a possible mechanism by which cancer cells acquire or maintain the therapy resistance phenotype [58]. Linn et al. [58] related overexpression of OCT4 with drug resistance in prostate cancer cell line. Furthermore, overexpression of this gene has also been associated with metastasis and poor prognosis in several types of cancer, including colorectal [59], lung [60] and glioma [61]. In our study, it was showed the expression of OCT4 protein in canine and human breast cancer cell lines, and this expression was decreased after 1 mM of melatonin treatment. Pang et al. [62] and Ferletta et al. [37] also noted OCT4 expression in mammospheres of canine is not yet understood. The CSC have an invasive and metastatic phenotype and can be generated by epithelial mesenchymal transition mechanism (EMT) [17]. Previous studies showed that EMT activation of human neoplastic mammary epithelial cells is associated with enrichment of cells with stemlike properties [55,62]. Thus, due to the relationship between EMT and cancer stem cell, we verified the presence of EMT molecular markers in CMT-U229 and MCF-7 stem cells. Also, we analyzed the action of melatonin treatment on these EMT markers. Our results showed that the CSCs from both cell lines, CMT-U229 and MCF-7, presented decrease in E-cadherin expression and increase in N-cadherin and vimentin expression. Furthermore, after melatonin treatment, there was an increase of E-cadherin expression and a decrease of N-cadherin and vimentin expression. Some studies have demonstrated low expression of E-cadherin associated with metastasis, lymph nodes involvement and poor prognostic in breast [22,65] and liver cancer [66]. On the other hand, overexpression of N-cadherin and vimentin is correlated with poor prognosis in several types of cancer, including colorectal [67], bladder [32] and hepatocellular carcinoma [66].
However, few studies have evaluated the action of melatonin on EMT markers. According to our results, Cos et al. [68] found high expression of E-cadherin in MCF-7 cells treated with melatonin. Ma et al. [69] also verified an increase of E-cadherin and survival, in mice with mammary tumor treated with melatonin. According to Mao et al. [33], melatonin has inhibited EMT in MCF-7 cells because it induces the degradation of β-catenin, an E-cadherin repressor, via activation of kinase protein GSK3β. The β-catenin can translocate to the nucleus and complex with the transcription T-cell factor/lymphocyte enhancer factor (TCF/LEF) to induce the transcription of Wnt target genes, including the zinc finger protein Snail, a transcriptional repressor of E-cadherin [70].
Cos et al. [68] found that there was no vimentin expression in both treated and untreated MCF-7 cells after melatonin treatment, probably due to the fact that vimentin is a specific Effectiveness of Melatonin in Breast Cancer Stem Cells marker of mesenchymal cells. In our study it was observed vimentin expression in both cell lines and its expression decreased after treatment with melatonin. According to Gilles et al. [71] and Mao et al. [33], a signaling model of Wnt/β-catenin also induces the expression of vimentin. Therefore, melatonin may decrease vimentin expression by GSK3β activation and consequential β-catenin degradation. Besides that, melatonin action on N-cadherin expression has been little explored in the literature, but the Wnt/β-catenin signaling model also induces the expression of N-cadherin [72], therefore, melatonin can suppress N-cadherin expression by a similar mechanism to that described above for vimentin.
It was also observed that melatonin decreased the viability of stem cells in both cell lines. Currently, some studies have shown inhibitory effect of melatonin on CSC [34,35]. Kannen et al. [35] by using colorectal CSCs demonstrated that melatonin reduces cell proliferation and promotes apoptosis of cancer stem cells. Similar results were found by Martin et al. [34], which showed that melatonin induces cell death by autophagy and simultaneously increased the effect of chemotherapy in glioma stem cells.
However, the exact mechanism through which melatonin inhibits cell growth in both in vitro and in vivo models is not fully understood, although some mechanisms have been proposed. These include induction of apoptosis, changes in the lipidic metabolism and an increase in the activity of natural killer cells, as well as stimulation of cytokines production, such as interleukins IL-2, IL-6, IL-12 and IFN [73].
Some studies also indicate that melatonin is able to reduce migration and invasion in some cancer types, such as glioblastoma [74], hepatocarcinoma [75], lung [76] and breast cancer [68,77,78]. According to Cos et al. [68], melatonin treatment reduces the invasiveness of MCF-7 cells, causing a decrease in cell attachment and cell motility, probably by interacting with the estrogen-mediated mechanisms of MCF-7 cells. Ortíz-López et al. [77] verified that melatonin inhibits the migration process and cell invasion in MCF-7 cells via ROCK-1 protein by modulating dynamic cytoskeleton. In turn, Mao et al. [78] found that melatonin exerts an inhibitory effect in breast cancer cell invasion through down-regulation of the p38 pathway, and inhibition of metalloproteinase 2 and 9 expression and activity. According to that, we also found that 1 mM of melatonin reduced the invasiveness of stem cells in both cell lines, CMT-U229 and MCF-7, and this fact may be associated with the anti-metastatic properties of melatonin described above. This result is of great interest since the invasiveness is one of the main characteristics of tumor stem cells.
In conclusion, our results show that the breast cancer stem cells, in both a canine and a human cell line, are responsive to melatonin treatment, reducing the viability and the invasiveness cellular capacity, as well as, the expression of stem cell and EMT markers. Few studies have investigated the melatonin effects on breast cancer stem cells, making this study the first one to simultaneously evaluate canine and human cell lines. In addition, due to the breast CSCs being highly resistant to chemotherapy, drugs that act successfully on this subpopulation can represent an effective therapeutic option for the breast cancer patient.