YMe was generated by aqueous extraction as describe in our previous work (Garcia-Lazaro et al., 2020). Briefly, YM leaves were macerated and then, the mixture was concentrated until 25° Brix, using maltodextrin (MD). Immediately, the resulting solution was incorporated into a pilot scale spray dryer (Galaxie, model 1,612). The powder was collected and stored in polyethylene bags at room temperature and protected from light. Prior to use, the extracts were dissolved in double-distilled water (ddH₂O) and filtered with a 0.22 µm. Extracts were standardized to total phenolic content, antioxidant activity, and Chlorogenic Acid content.

Protein, moisture, lipid, total ash, and dietary fiber were determined using the Association of Official Analytical Chemists (AOAC) methods (Horwitz and Latimer, 2006). The total carbohydrate content was calculated as the difference between 100 and the sum of the percentages of moisture, protein, lipid, ash and dietary fiber. Sugars were determined using AOAC methods, and are the sum of individual monosaccharides (glucose and fructose) and disaccharides (sucrose and maltose). Energy values were obtained by applying factor 4–4–9 kcal/g for protein, carbohydrate, and lipid, respectively (Bertë et al., 2011).

The cells were cultured at 37°C in a humidified atmosphere containing 5% CO₂. 4T1 (ATCC: CRL-2539) from mouse mammary tumor was grown in Roswell Park Memorial Institute (RPMI) 1,640 medium (Life Technologies, United States). A sarcomatoid mammary carcinoma cell line F3II is a highly invasive and metastatic variant established from a clone of a spontaneous, hormone-independent BALB/c mouse mammary tumor (Alonso et al., 1996), MCF-7 (ATCC: HTB-22, human breast adenocarcinoma) ER/PR + human BC cells and MDA-MB 231 (ATCC: HTB-26, human breast adenocarcinoma) a triple-negative human BC cells were grown in high-glucose Dulbecco’s Modified Eagle’s Medium (DMEM) (Life Technologies, United States). The characteristics of cell lines are shown in Table 1. All cultures contained 10% fetal bovine serum (FBS, Gibco, United States) and 40 μg/ml gentamicin (Fada Pharma, Argentina). In addition, cells lines were routinely tested to rule out Mycoplasma infection of cells.

4T1, F3II, MCF-7, and MDA-MB 231 BC cells were seeded into 96-well plates at 2.5×10³ cells per well and incubated for 24 and 72 h at 37°C, 5% CO₂. YMe concentrations ranging from 0.03 to 2.5 mg/ml were added to the wells in complete medium. After incubation, the medium was removed and the plates were washed with PBS. Then, attached cells were fixed with methanol for 15 min and stained with 0.5% crystal violet for another 15 min. The excess dye was removed by washing with rinse water. The dye in the cells was dissolved in methanol-acetic solution (10%/5%, V/V) and the absorbance was measured at 595 nm in a 96-well plate reader (ASYS Hitech Gmbh, Austria). Each condition was assayed sextuplicate in three different experiments, and SD was determined. Inhibitory concentration 50 (IC₅₀) values were calculated from the growth curves for 72 h.

Detection of apoptosis in all samples were provided using a TUNEL (Terminal deoxynucleotidyl transferase dUTP nick end labeling) assay Kit (Promega) according to manufacturer’s protocol for detection of DNA fragmentation, a prominent hallmark of apoptosis. Briefly, F3II cells growing in glass coverslips were treated with 0.15 mg/ml of YMe. The cells treated with PBS were considered as negative control and the cells treated with Camptotecin (10 µM) for 4 h were considered as positive control. After 24 h of treatment, TUNEL assay was carried out. F3II cells were fixed for 10 min at room temperature in a solution containing 4% (W/V) of paraformaldehyde in PBS (pH 7.4). Then, cells were washed 3 times with PBS. Subsequently, cells were permeabilized with 0.2% Triton X-100 for 5 min on ice. After washing with PBS, cells were incubated with equilibration buffer for 10 min at room temperature. TUNEL reaction mixture (50 µL) containing the TdT enzyme was added to the cells. After that, the cells were incubated in a humidified box in the dark for 1 h at 37°C. Then, the cells were washing with PBS 3 times. To detect the nuclei, the samples were counterstained using 4, 6-diamidino-2-phenylindole (DAPI) for 2 min at room temperature in the dark and then washed with PBS 3 times. The TUNEL stained sections were viewed under fluorescence microscope (Cytation 5—Biotek) at 2 different wavelengths for fluorescein isothiocyanate (FITC) and DAPI respectively, and images of 10 randomly selected fields were captured at 10X magnification for each slide. The apoptotic index (AI) was then calculated by the following formula: AI= (Number of TUNEL positive cells/Total number of cells) x 100.

Cell adhesion assay was carried out by a colorimetric method based on staining cells with the dye crystal violet. Briefly, 4T1, F3II, MCF-7, and MDA-MB 231 BC cells in 200 µL medium with FBS were seeded into 96-well plate at 4×10⁴ cells/well. The cells were treated with YMe with concentrations ranging from 0.03 to 2.5 mg/ml for 2 h. The plates were incubated at 37°C, 5% CO_2. After that, PBS was added to each well and then aspirated to remove non-adhered cells. Then the cells were fixed and stained with 0.1% crystal violet, 20% methanol in PBS for 20 min. The excess dye was removed by washing and dye in the cells was dissolved in methanol-acetic solution (10%/5%, V/V). The absorbance was measured at 595 nm in a 96-well plate reader (ASYS HitechGmbh, Austria).

4T1, F3II, MCF-7, and MDA-MB 231 BC cells were seeded at a density of 1×10⁵ cells/well in a 6-well plate. When cells reached 80% confluence, wounds were made in the cell monolayer with a pipette tip and photographed (time 0) at 40X using an inverted microscopy. Cells were incubated during 12 h in presence of IC₅₀ of YMe. After that, cells were washed with PBS, fixed with methanol and stained with 0.5% crystal violet. Then, cells were photographed again at 3 randomly selected sites per well. Photographs were taken using a camera connected to the inverted microscope (Nikon, NIS elements software), and invasion area was quantified using ImageJ software. Data are expressed as mean ± SD. The treatments were carried out in triplicate and each experiment was repeated 3 times independently.

4T1 and F3II BC cells (2×10⁵) were plated in 100 µL of medium without FBS in the upper chamber of 8 µm Transwells (Costar Inc.), coated with 100 µL of 0.1 mg/ml Matrigel. Cells were incubated during 24 h in presence of IC₅₀ of YMe. Complete medium (with FBS as chemoattractant) was placed in the lower chamber. After the incubation period, cells that remained in the upper chamber were removed using a cotton swab. Cells adhered in the transwell lower chamber were washed with PBS, fixed and stained with 0.1% crystal violet, 20% methanol in PBS for 20 min. Cells that invaded and migrated to the lower chamber were photographed (Nikon, NIS elements software) and quantified. The number of cells were evaluated in five random areas using a phase contrast inverted microscope (40X magnification).

Four-week old pathogen-free female BALB/c mice were obtained from La Plata University. The mice were housed in 12 h of light and dark cycle. Food and water were provided ad libitum, and general health status of the animals was monitored daily. Animals with an average weight of 20 g were used.

Mice were randomly assigned into 2 groups (n per group = 11). Control group drank MD, the excipient of extract and treated group drank YMe 10 mg/ml. The YMe was administered to BALB/c mice in a dose of 1.6 g/kg/day via the drinking water before (1 month) and after the inoculation of F3II tumor cells. Mouse mammary cancer model was established through subcutaneously inoculating 1 × 10⁵ F3II cells into the left fourth mammary fat pad (MFP) in mice.

Mice were randomly assigned into 2 groups (n per group = 6). Control group drank MD and treated group drank YMe 10 mg/ml. The YMe was administered to BALB/c mice via the drinking water before (1 month) and after the inoculation of F3II tumor cells. The animals were inoculated with 5×10⁴ cells/mice on the right flank of female BALB/c mice.

1 week after inoculation of F3II cells, tumors were measured 3 times per week using a digital caliper. The greatest longitudinal diameter (length) and the greatest transverse diameter (width) were measured and volume was calculated using the following formula: tumor volume (mm³) = ½ (length × width²). When the tumor volume reached about 1800 mm³, mice were sacrificed by cervical dislocation due to ethical considerations and tumors were resected from mice. Tumors were weighed, and the mean tumor weight was calculated. For spontaneous metastasis, lungs were collected and fixed in Bouin’s solution. The number of metastasis nodules in lung surface was manually counted using a dissecting microscope.

At the end points, mice were necropsied and the tumors and lungs were harvested, fixed in 10% formaldehyde (Anedra) and embedded in paraffin, cut into 5-mm sections, and stained with hematoxylin and eosin (H&E). Metastatic tumor nodules present on lungs were quantified by counting 3 sections per lung sample. Images were taken by an inverted microscope (Cytation 5—BioTek).

All animal protocols have been carried out in accordance with the Guide for the Care and Use of Laboratory Animals as adopted by the U.S. National Institutes of Health. Protocols were approved by our institutional Animal Care Committee UNQUI-CICUAL (Resolution CD CyT Nº075/14).

All data analyses were performed using GraphPad Prism version 6.00 (GraphPad Software, San Diego California, United States). Samples were examined for normality with Kolmogorov–Smirnov test. Results were expressed as mean ± standard deviation (SD) or standard error of mean (SEM), and differences were analyzed with Student’s t test or ANOVA with a Tukey’s post-test, accordingly. Survival curves were plotted according to the Kaplan–Meier method. Statistical significance was calculated using log-rank test. p < 0.05 was considered statistically significant.

The chemical composition of spray-dried extract is shown in Table 2. Considering the nutritional components of our extract in addition to the high antioxidant capacity exerted by their polyphenolic compounds and the bioactivities against colon tumor cells (Garcia-Lazaro et al., 2020), we became interested in studying the anti-tumor effects of YMe in different in vitro and in vivo breast cancer models.

To study the sensitivity of BC cells to YMe, we first evaluated its effect on cell proliferation of murine and human BC cell lines. As shown in Figure 1, 4T1 (Figure 1A), F3II (Figure 1B), MCF-7 (Figure 1C), and MDA-MB 231 (Figure 1D) cells were treated with different concentrations of the extract ranging from 0.03 to 2.50 mg/ml for 24 and 72 h. YMe showed a statistically significant decrease in growth of murine and humane cells in vitro between the control and all treated cells. IC₅₀ values were calculated from the growth curves for 72 h. The IC₅₀ values obtained were 0.06 mg/ml to 4T1, 0.15 mg/ml to F3II, 0.6 mg/ml to MCF-7 and 0.15 mg/ml to MDA-MB 231. These values had no cytotoxic effects when assayed for 24 h on semi-confluent monolayers. Since Maltodextrin was used as a carrier agent during spray-drying process, we evaluated whether this vehicle modulates cell proliferation. We observed that cell proliferation was not affected when the tumor cells were treated with this excipient (data not shown).

To understand the effect of YMe on cell proliferation, the mechanism of cellular apoptosis was investigated. A TUNEL assay was conducted to evaluate the cell death following treatment with the IC₅₀ concentration of YMe for 24 h. As shown in Figure 2A the AI was significantly elevated in F3II cells treated with YMe compared to the untreated cells (*p < 0.05). Cells were stained with TUNEL (green) and counterstained with DAPI (blue) respectively. Blue and green stains in the merged image represent the TUNEL-positive apoptotic cells.

We investigated whether YMe can act through the blocking of cell attachment using a cell adhesion assay. The cells were seeded and treated with increasing concentrations of YMe. Cell adhesion of 4T1 (Figure 3A), F3II (Figure 3B), MCF-7 (Figure 3C), and MDA-MB 231 (Figure 3D) cells lines decreased significantly in a dose-dependent manner in response to treatment with YMe (**p < 0.01, ***p < 0.001 vs. control).

The effect of YMe on cell motility was evaluated using wound healing assays using a non-cytotoxic concentration of the extract. After scratching confluent monolayers of 4T1, F3II, MCF-7, and MDA-MB 231 cells to create a wound-like gap, the cells were treated with IC₅₀ of the extract. As shown in Figure 4, after 12 h of treatment with the extract, cell motility was inhibited by YM. The wounds of treated groups remained cracked with some differences. The treatment with YMe reduced the motility of 4T1 (Figure 4A) and MDA-MB 231 (Figure 4D) by 25%, F3II cell line by 18% (Figure 4B) and MCF-7 by 82% (Figure 4C). The motility reduction in the F3II cell line was not statistically significant.

A transwell assay was used to evaluate whether YMe affects the invasion of BC cells on the ability of cells to migrate in a chemoattractant gradient. As shown in Figures 4E,F, the invasiveness of 4T1 and F3II tumor cells treated with YMe was decreased in comparison with the control group. This reduction was statistically significant in the F3II cell line (**p < 0.01 vs. control cells) but no in the 4T1 cell line.

The F3II cell/syngeneic mouse model was used to investigate the in vivo anti-tumor effects of YMe. Orthotopic and heterotopic BC models in BALB/c mice were used and the tumor progression was observed over time (Figures 5A, 6A respectively). In both models, the control group received a MD solution via the drinking water, while the treated group received YMe at a dose of 1.6 g/kg/day. Solutions were administered 1 month before the inoculation of F3II tumor cells and until the end of the protocol.

Throughout both studies, the body weight of mice in the control group and the YM group was monitored. Mice in all groups treated with MD or YMe gained weight progressively. As shown in Table 3, there were no significant differences in initial body weight, final body weight, body weight gain, and growth rate in all groups, p > 0.05 vs. control group. In addition, during the experiments, there were no signs of systemic toxicity, no behavioral abnormality, or animal death observed.

The effect of YMe on the growth of orthotopic F3II primary tumors is shown in Figure 5B. The tumors first became palpable 7 days after injection and tumor take was 100% in groups 15 days after inoculation. By day 48, the mean tumor volume in both groups reached 1800 mm³; no statistically significant difference in mammary tumor volume between the two groups was observed.

The effect of YMe on the growth of heterotopic F3II primary tumors is shown in Figure 6B. The tumors became palpable 7 days after injection and tumor take was 100% in groups 15 days after inoculation. By day 48, we found that treatment with YMe significantly led to suppression of F3II tumor volumes when compared with the control group. Tumor volume in the control animals was the highest, reaching 1,355.29 ± 808.44 mm³ at the end of the experiment. In contrast, tumor volume in the YM group was significantly reduced, 521.34 ± 243.02 mm³. The oral administration of YMe at a dose of 1.6 g/kg/day resulted in a statistically significant decrease of tumor growth in the heterotopic model (p = 0.02).

At the end of the experiments, tumors were excised from each animal for examination of tumor weight and assessment of the anti-tumor effect of YMe. As shown in Figures 5C, 6C, both in the orthotopic model and in the heterotopic model respectively, we observed that tumor weight in the animals treated with YMe tended to be more reduced, although the differences compared with the control group were not statistically significant (p > 0.05). Unpaired Student’s t test was used for statistical analysis.

We further evaluated the effect of YMe on the survival time of F3II tumor-bearing mice. From day 48, the global survival analysis began in both models. It was observed that consumption of YMe significantly increased the survival of the animals (*p < 0.05) in both orthotopic and heterotopic models. As seen in Figure 5D, in the orthotopic model, the log-rank analysis showed that at day 59, 100% of the control group had been sacrificed while in the treated group only 62% of the experimental individuals had been sacrificed. As shown in Figure 6D, in the heterotopic model, the log-rank analysis indicate that at day 69, 100% of the control group had been sacrificed while in the group treated with the YMe only 50% of the experimental individuals had been sacrificed, which demonstrated that YMe prolongs the survival time of F3II tumor-bearing mice.

Also, we evaluated the effect of YMe on spontaneous lung metastasis. Lung tissues from mice of both BC models were dissected and fixed with Bouin’s solution to observe tumor metastasis. We could clearly see white nodules on the surfaces of lung lobes.

The incidence, (defined as the number of mice with metastases related to the total number of inoculated ones), is represented in Table 3. In the orthotopic model, only 28% of mice in the yerba mate group (3 of 11) had surface metastases, whereas that 56% of the mice in the control group (6 of 11) had surface lesions. However, the difference between the control and YM groups was not statistically significant (p > 0.05). The size of lung metastatic tumors was measured with a dissecting microscope and tumors were stratified into 3 groups based on size (diameter < 1 mm small size, 1–2 mm medium size and > 2 mm, large size) with the objective to do a qualitative analysis. As can be seen in Figure 5E, the size of lung nodules of un-treated mice was larger than to the ones of the animals treated with YMe. Interestingly, the pulmonary nodules in the mice treated with YMe did not exceed 2 mm.

The incidence of lung metastases in the heterotopic model is represented in Table 3. Whereas only 16% of mice in the YM group (1 of 6) had surface metastases, 33% of the mice in the control group (2 of 6) had surface lesions. The incidence of lung metastases was not statistically different between the control and Yerba Mate groups (p > 0.05). As Figure 6E shows, the number of medium lung metastases was reduced in treated mice compared to control animals. These results were further confirmed by H&E staining. Representative images of lung metastasis are shown in Figures 5F, 6F.