BD Matrigel™ was purchased from BD Biosciences. Hoechst dye was obtained from Santa Cruz Biotechnology, Inc. CellMask™ Orange plasma membrane stain was obtained from Invitrogen (Thermo Fisher Scientific, Inc.). Tetramethylrhodamine (TRITC)-conjugated phalloidin, mitomycin C, Annexin V and Triton™ X-100 were from Sigma-Aldrich (Merck KGaA). PP2 and PP3 were from Merck KGaA.

MDA-MB-231 and MCF-7 breast cancer cells were obtained from the American Type Culture Collection (ATCC) and cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.), with 5% FBS (ByProductos), 3.7 g/l sodium bicarbonate and antibiotics under a humidified atmosphere with 5% CO₂ and 95% air at 37°C.

Human mammary non-tumorigenic epithelial cells (MCF12A) were obtained from ATCC and cultured in DMEM/F12 medium (Gibco; Thermo Fisher Scientific, Inc.) with 5% FBS, 0.5 µg/ml hydrocortisone (Sigma-Aldrich, Merck KGaA), 20 ng/ml epidermal growth factor (EGF; Sigma-Aldrich, Merck KGaA), 10 µg/ml insulin (Sigma-Aldrich, Merck KGaA) and antibiotics under a humidified atmosphere with 5% CO₂ and 95% air at 37°C.

MDA-MB-231 and MCF-7 cells were serum-starved in DMEM for 24 h, and MCF12A cells were serum-starved in DMEM/F12 for 18 h before treatment with inhibitors and/or EVs.

The patients enrolled between August 2014 and December 2018 consisted of 32 unrelated women who resided in Mexico City (median age, 57.7 years; age range, 38–80 years) from the First October Regional Hospital-ISSSTE (Mexico) with biopsy-diagnosed breast cancer at clinical stages II and III, and without receiving therapy. The control group consisted of 20 healthy women (median age, 42.7 years; age range, 16–86 years) who did not have any family history of breast cancer. All study participants provided signed informed consent, and the protocol was approved by The Committee of Research, Ethics and Biosafety of the First October Regional Hospital-ISSSTE, and was conducted in accordance with the Declaration of Helsinki. The pathological and clinical information of patients with breast cancer is shown in Table I.

In total, 4 ml peripheral blood was collected into polypropylene tubes containing sodium citrate (Vacutainer System; BD Biosciences). Whole blood samples were centrifuged at 1,500 × g for 15 min at 4°C, and plasma samples were obtained. Isolation of EVs was performed as described previously (17). Then, one volume of 1 ml plasma was centrifuged at 3,000 × g for 30 min at 4°C to remove platelets. Plasma was obtained and centrifuged at 10,000 × g for 30 min at 4°C to remove apoptotic bodies and platelet EVs, and supernatants were aliquoted and frozen at −80°C until further analysis. Frozen samples were thawed on ice and centrifuged at 110,000 × g for 70 min at 4°C and pellets were reconstituted in PBS or DMEM (EV fractions). The absolute number of EVs was determined by flow cytometry using BD Trucount™ tubes (BD Biosciences) (18).

TEM of EV fractions was performed as described previously (33). EV fractions were adsorbed for 5 min on carbon-coated copper grids with mesh formvar (0.3%) at room temperature. The grids were stained for 30 sec at room temperature with 2% uranyl acetate solution and excess fluid was removed. Grids were air-dried and analyzed using a JEM-1400 TEM (JEOL, Ltd.), operated at 80 kV and coupled with a digital camera Veleta (Olympus SIS).

NTA was used to assess the size distribution of EV fractions. EV fractions were diluted in 10 ml filtered PBS and analyzed with a NanoSight NS300 (Malvern Instruments, Ltd.), which was equipped with a 488 nm laser and a sCMOS camera. Then, three videos of each sample were captured at a duration of 60 sec and data were analyzed with NTA v3.0 software (Malvern Instruments, Ltd.).

Cultures of MDA-MB-231 cells (1.5×10⁶ cells/dish) were serum-starved for 24 h, washed twice with PBS and then stimulated at different time periods with EV fractions (20,000 EVs/1.5×10⁶ cells/experimental condition) from patients with breast cancer and healthy women. Scratch-wound assays were performed for 48 h; invasion assays for 72 h; proliferation assays for 48; and zymography assays for 3, 6, 9, 12 and 24 h.

Cells and EVs were solubilized in 0.1 ml of ice-cold RIPA buffer (50 mM HEPES pH 7.4, 150 mM NaCl, 1 mM EGT4, 1 mM sodium orthovanadate, 100 mM NaF, 10 mM sodium pyrophosphate, 10% glycerol, 1% Triton X-100, 1% sodium deoxycholate, 1.5 mM MgCl₂, 0.1% SDS and 1 mM PMSF). Protein concentration of each sample was determined by using the Bradford protein assay (Bio-Rad Laboratories, Inc.). Equal amounts of protein (30 µg/lane) were separated by SDS-PAGE on a 10% gel and proteins were transferred to nitrocellulose membranes. Nitrocellulose membranes were blocked for 2 h with 5% non-fat dried milk in PBS (pH 7.2/0.1% Tween 20; wash buffer) at room temperature. The membranes were incubated with the following primary antibodies at 4°C overnight: Anti-Flotillin-2 (Flot-2) antibody (Ab; Mouse monoclonal; cat. no. 610383; 1:1,000; BD Biosciences), anti-CD9 Ab C-4 (Mouse monoclonal; cat. no. sc-13118; 1:300; Santa Cruz Biotechnology, Inc.), anti-FAK Ab D-1 (Mouse monoclonal; cat. no. sc-271126; 1:1,000; Santa Cruz Biotechnology, Inc.), anti-c-Src Ab 17AT28 (Mouse monoclonal; cat. no. sc-130124; 1:500; Santa Cruz Biotechnology, Inc.), anti-CD81 Ab EPR4244 (Rabbit monoclonal; cat. no. ab109201; 1:1,000; Abcam), anti-phosphorylated-specific Ab to Tyr-397 of FAK (anti-p-FAK Ab; Rabbit polyclonal; cat. no. 44-624G; IF 1:250; WB 1:1,000; Thermo Fisher Scientific, Inc.), anti-phosphorylated-specific Ab to Tyr-418 of Src (anti-p-Src Ab; rabbit polyclonal IgG; cat. no. AF2685; 1:1,000; R&D Systems, Inc.), anti-vinculin Ab (Rabbit polyclonal; cat. no. V4139; IF 1:250; WB 1:500; Sigma-Aldrich; Merck KGaA) and anti-actin Ab (Mouse polyclonal; 1:1,000) was provided by Dr Manuel Hernandez (Cinvestav-IPN). Subsequently, membranes were washed three times with wash buffer and incubated with horseradish peroxidase-conjugated secondary Ab (cat. nos. G21040 and G21234; 1:5,000; Thermo Fisher Scientific, Inc.) for 2 h at room temperature, and washed again with wash buffer three times. ECL detection reagent (cat. no. sc-2048; Santa Cruz Biotechnology, Inc.) and images were used for visualization of immunoreactive bands. Bands were analyzed by using the ImageJ software v1.52e (National Institute of Health).

EV uptake assays were performed as described previously (34,35). EV fractions (20,000 EVs/condition) were stained via incubation for 30 min at 4°C with CellMask Orange dye solution (2.5 µg/ml). EV fractions were then washed with PBS, centrifuged at 110,000 × g for 70 min at 4°C and reconstituted in 100 µl DMEM. MDA-MB-231 cells (1.5×10⁶ cells/condition) were incubated for 4 h at 37°C with stained EVs, and after incubation, cells were washed twice with PBS and fixed for 20 min at room temperature with a solution of 4% paraformaldehyde. Next, cells were incubated for 10 min at room temperature with a solution of trypsin (0.1%), washed twice with PBS and re-suspended in PBS-1% BSA (Santa Cruz Biotechnology, Inc.). Controls of inhibition of EVs uptake were included, and were obtained by treatment of EVs with 200 ng/ml Annexin V for 30 min at 4°C, before staining of EVs with CellMask Orange dye solution for 30 min at 4°C. Cells were analyzed using a BD FACSCalibur™ flow cytometer (BD Biosciences) and data analysis was performed with Summit v4.3 software (Beckman Coulter, Inc.).

MDA-MB-231 cells (1.5×10⁵ cells/condition) were incubated with stained EVs (20,000/condition) for 4 h at 37°C. After incubation, cells were washed twice with PBS, fixed for 20 min at room temperature with a solution of paraformaldehyde (4%) and washed twice again with PBS. Next, cells were counterstained with Hoechst for 20 min at room temperature and mounted on glass slides using Vectashield. Preparations were analyzed by confocal microscopy (Model TCS SP2; Leica Microsystems, Inc.).

MDA-MB-231 cells (1.5×10⁶ cells) were cultured until they reached confluency and pretreated for 2 h at 37°C with 12 µM mitomycin C to inhibit proliferation during the experiment. Cell cultures were scratched, washed with PBS and supplemented with serum-free DMEM with or without inhibitors and/or EV fractions (20,000 EVs/condition) at 37°C. After treatment for 48 h, the cultures were imaged using an inverted microscope coupled to a camera (FSX100; Olympus Corporation; magnification, ×100). Images from ≥3 fields per experimental condition were acquired and analyzed using ImageJ software v1.52e (National Institutes of Health). In total, one control of cell migration was included (5% FBS).

Inserts of 24-well plates (Costar; Corning, Inc.) were covered with 50 µl Matrigel (3 mg/ml) and incubated at 37°C for 2 h. Then, 1.2×10⁵ MDA-MB-231 cells in serum-free DMEM were plated on Matrigel of each insert in the upper chamber, and the lower chamber contained EV fractions (20,000 EVs/condition) in 600 µl DMEM. The plates with the inserts were incubated at 37°C for 72 h under a humidified atmosphere with 5% CO₂ and 95% air. After incubation, cells and Matrigel on the upper surface of the membrane were removed, and cells on the lower surface of the membrane were washed with PBS followed by fixation for 12 min at room temperature with 4% paraformaldehyde. After fixation, invaded cells were imaged using an inverted microscope couple to a camera (FSX100, Olympus Corporation; magnification, ×400). Quantification of invaded cells was calculated by staining membranes with 0.5% crystal violet for 15 min at room temperature and elution of the dye with 300 µl 30% acetic acid. Absorbance of the collected solution was measured at 600 nm. Cells treated with 5% FBS were included as an invasion control.

Conditioned media from TNBC MDA-MB-231 cells (1.5×10⁶ cells) treated with EV fractions (20,000 EVs/condition) at different time points (3, 6, 9, 12 and 24 h) at 37°C from women with breast cancer and healthy controls were concentrated using 5,000 Da Centricon^® filters (EMD Millipore). An equal volume (12 µl) of non-heated conditioned medium and sample buffer (2% sucrose, 2.5% SDS, 4 µg/ml phenol red) was mixed and the samples were loaded onto 8% acrylamide gels copolymerized with gelatin (1 mg/ml). Next, gels were rinsed three times for 30 min with Triton X-100 (2.5%) and incubated in assay buffer (50 mM Tris-HCl pH 7.4, 5 mM CaCl₂) for 48 h at 37°C. After incubation, gels were stained for 1 h at room temperature with a solution of Coomassie Brilliant Blue G-250 (0.25%) dissolved in acetic acid (10%) and methanol (30%). Proteolytic activity was identified as white zones on a blue background. Controls of MMP-2 and MMP-9 secretions were included, which were obtained by treatment of MDA-MB-231 cells with 400 mg/dl ethanol and 100 ng/ml phorbol-12,13-dibutyrate for 24 h at 37°C, as treatment of cells with these compounds induces the expression and secretion of MMP-2 and MMP-9, respectively (36,37).

MDA-MB-231 cells (1.5×10⁵ cells) were grown on coverslips, washed with PBS, equilibrated in DMEM and treated with EV fractions (20,000 EVs/condition) for 30 min at 37°C. Cells were fixed with paraformaldehyde (4%) for 20 min at room temperature, permeabilized with 0.5% Triton X-100 and then blocked for 30 min at room temperature with FBS (10%) dissolved in PBS. Next, cells were incubated overnight at 4°C with anti-p-FAK Ab (1:250) followed by FITC-labeled anti-mouse secondary Ab (cat. no. 115-095-003; Jackson ImmunoResearch, Inc) for 2 h at 4°C. Staining of focal contacts was performed by incubation of cells with anti-vinculin Ab for 12 h at 4°C, while staining of cells for fibrillar actin was performed by incubation for 2 h at 4°C with TRITC-conjugated phalloidin (cat. no. R415; Thermo Fisher Scientific, Inc.) Cells then were analyzed by confocal microscopy (Model TCS SP2; Leica Microsystems, Inc.).

MDA-MB-231 cells (30,000 cells per well) were treated for 48 h at 37°C with EV fractions (20,000 EVs/condition). Next, 10 µl WST-1 reagent (cat. no. ab65473; Abcam) was added to the cells in each well and microplates were incubated at 37°C for 2 h. The absorbance of each well was measured at 450 nm, and one control for proliferation was included, which was prepared by treatment of cells with 5% FBS for 48 h at 37°C.

Data are presented as the mean ± SD of ≥3 independent experiments. Data were analyzed by one-way ANOVA followed by Tukeys post hoc test for ≥3 groups, while Students t-test was used to analyze groups of two. P<0.05 was considered to indicate a statically significant difference.

The present study characterized EV fractions using NTA, TEM and western blotting against Flot-2, CD9 and CD81, which are molecular markers associated with EVs (7). NTA and TEM results identified a population of spherical vesicles with sizes between 30–300 nm in healthy women, while women with breast cancer had vesicles between 50–600 nm (Fig. 1A and B). Moreover, NTA showed that the number of EVs was significantly higher in women with breast cancer (1.542×10¹¹ EVs/ml) than in healthy women (2.14×10¹⁰ EVs/ml). Western blotting results demonstrated the presence of Flot-2, CD9 and CD81 protein expression in EV fractions from healthy women (Ctrl EVs) and EV fractions from women with breast cancer (BC EVs; Fig. 1C).

Next, whether BC EVs induced migration in TNBC MDA-MB-231 cells was investigated. Cell migration assays were performed using scratch-wound assays with MDA-MB-231 cells treated with EVs from 22 patients with breast cancer and from 15 healthy women. The results indicated that BC EVs induced increased migration compared with Ctrl EVs in MDA-MB-231 cells (Fig. 1D). In addition, the association between the migration induced by BC EVs in MDA-MB-231 cells and the expression levels of estrogen, progesterone and Her2/neu receptors in the mammary tumors of the women from where the plasma EVs were obtained was analyzed. It was found that the migration induced by BC EVs was not related with the expression levels of estrogen, progesterone and Her2/neu receptors in the tumors of women with breast cancer (Figs. 1E and S1A). Furthermore, migration of MDA-MB-231 cells induced by BC EVs was not related with the stages II and III of the women with breast cancer (Figs. 1F and S1B).

To further examine these findings, whether BC EVs induced migration and/or invasion was determined in another breast cancer cell line (MCF-7) and mammary non-tumorigenic epithelial cells (MCF12A). The results showed that treatment with BC EVs did not induce migration in MCF-7 cells, and it did not induce migration and invasion in MCF12A cells (Fig. 2A-C).

It was studied whether Ctrl EVs and BC EVs were taken up by MDA-MB-231 cells. Ctrl EVs and BC EVs were labeled with CellMask orange dye, MDA-MB-231 cells were incubated with unstained EVs and stained EVs, and the fluorescence intensity was analyzed by flow cytometry. It was identified that fluorescence intensity was higher in MDA-MB-231 cells treated with stained Ctrl EVs and stained BC EVs compared with MDA-MB-231 cells treated with unstained EVs (Figs. 3A and B and S2). In addition, the comparison of fluorescence intensity was not significantly different between the value obtained from MDA-MB-231 cells treated with stained BC EVs and the value obtained from MDA-MB-231 cells treated with stained Ctrl EVs (Fig. 3C).

EVs express PS on their surface, and Annexin V has a strong and specify affinity for PS (38). Controls of uptake inhibition were included, and were obtained by treatment of EVs with 200 ng/ml Annexin V, before staining of EVs with CellMask Orange dye. The results indicated that treatment of stained Ctrl EVs with Annexin V inhibited the increase of fluorescence intensity induced by stained Ctrl EVs (Fig. 3A). However, treatment of BC EVs with Annexin V did not inhibit the increased fluorescence intensity induced by stained BC EVs (Fig. 3B).

To further examine whether EVs are taken up by breast cancer cells, MDA-MB-231 cells treated with stained Ctrl EVs, stained BC EVs and unstained BC EVs were analyzed by confocal microscopy. It was found that MDA-MB-231 cells treated with stained BC EVs and stained Ctrl EVs showed a red staining, while cells treated with unstained BC EVs did not show any color (Fig. 3D).

It was determined whether treatment of MDA-MB-231 cells with BC EVs induced gelatinase secretion. MDA-MB-231 cells were stimulated for 3, 6, 9, 12 and 24 h with EVs from three patients with breast cancer, and stimulated for 24 h with EVs from three healthy women. Conditioned media were obtained, concentrated and analyzed by gelatin zymography. It was identified that stimulation of MDA-MB-231 cells with BC EVs induced increased secretion of MMP-2 and MMP-9 at 12 and 24 h of treatment (Fig. 4A).

Since treatment of MDA-MB-231 cells with BC EVs for 24 h induced MMP-2 and MMP-9 secretion, MDA-MB-231 cells were treated for 24 h with EVs from three patients with breast cancer and EVs from three healthy women, and supernatants were analyzed by gelatin zymography. The results indicated that treatment of MDA-MB-231 cells with Ctrl EVs induced a small increase in MMP-9 secretion, but stimulation with BC EVs induced a significant increase in both MMP-9 and MMP-2 secretion (Fig. 4B-D).

In order to determine whether treatment of MDA-MB-231 cells with BC EVs induced Src activation, which is initiated by its phosphorylation at Tyr-418 (p-Src), MDA-MB-231 cells were treated for 20 min with three Ctrl EV samples and three BC EV samples, and cell lysates were analyzed by western blotting with anti-p-Src Ab. It was identified that treatment of MDA-MB-231 cells with Ctrl EVs induced a small level of phosphorylation of Src at Tyr-418, while treatment with BC EVs induced a strong phosphorylation of Src at Tyr-418 (Fig. 5A).

Next, whether Ctrl EVs and BC EVs contain p-Src and Src kinase was determined by western blotting with p-Src Ab and Src Ab. The results showed that Ctrl EVs contained a low amount of p-Src, while BC EVs contained a significantly larger amount of p-Src (Fig. 5B). However, it was demonstrated that Ctrl EVs and BC EVs contained a similar amount of Src kinase (Fig. 5B).

The present study determined whether BC EVs induced FAK activation, which is induced by its phosphorylation at Tyr-397 (p-FAK). Lysates of MDA-MB-231 cells treated for 20 min with three Ctrl EV samples and three BC EV samples were analyzed by western blotting with anti-p-FAK Ab. It was identified that treatment with Ctrl EVs induced low levels of phosphorylation of FAK at Tyr-397, whereas treatment with BC EVs induced a significantly higher level of phosphorylation of FAK at Tyr-397 (Fig. 6A).

Since Src is able to induce maximal FAK activation mediated by G-protein-coupled receptors (GPCRs) and tyrosine kinase receptors (27), the role of Src in FAK activation was investigated. The role of Src was examined using PP2, which is a specific inhibitor of Src family members (39). To determine whether PP2 inhibited the activity of Src, MDA-MB-231 cells were untreated and treated for 1 h with 10 µM PP2 and then treated for 20 min with three BC EVs, and lysed. Cell lysates were analyzed by western blotting with anti-p-Src Ab, and it was found that PP2 inhibited the increase of p-Src induced by BC EVs (Fig. 5C). Next, MDA-MB-231 cells were untreated and treated for 1 h with 10 µM PP2 and stimulated for 20 min with BC EVs. It was identified that treatment with BC EVs induced an increase of p-FAK via a Src-dependent pathway in MDA-MB-231 cells (Fig. 6B).

The present study also assessed whether p-FAK and FAK were localized in EV fractions, and three Ctrl EVs and four BC EVs were analyzed by western blotting with p-FAK Ab and FAK Ab. The results showed that Ctrl EVs and BC EVs expressed variable levels of p-FAK and FAK, and there was no significant difference in p-FAK and FAK expression levels between Ctrl EVs and BC EVs (Fig. 6C)

It was examined whether BC EVs induced a redistribution of p-FAK, as well as the assembly of focal adhesions and the role of Src in the assembly of focal adhesions. MDA-MB-231 cells cultured on coverslips were untreated or treated for 1 h with 10 µM PP2 and stimulated with Ctrl EVs and BC EVs. The redistribution of p-FAK was analyzed by immunofluorescence with anti-p-FAK Ab. Moreover, the number of focal adhesions was analyzed by immunofluorescence with anti-vinculin Ab, as vinculin is a cytoplasmic actin binding protein enriched in focal adhesions (40). It was demonstrated that BC EVs induced the redistribution of p-FAK at the edges of cells, and increased the number of focal adhesions (Fig. 7A and B). Furthermore, focal adhesion assembly induced by BC EVs was dependent on Src activity (Fig. 7B). In addition, the present study determined whether vinculin was localized in EV fractions, and three Ctrl EVs and four BC EVs were analyzed by western blotting with anti-vinculin Ab; the results indicated that Ctrl EVs and BC EVs did not express vinculin (Fig. 7C). One control of vinculin expression was included (Fig. 7D).

The role of Src in migration was examined using PP2 and its inactive analog (PP3) (39). Cultures of MDA-MB-231 cells were treated for 1 h with 10 µM PP2 or 10 µM PP3, scratch-wounded and stimulated with Ctrl EVs and BC EVs. It was demonstrated that migration induced by BC EVs was dependent on Src activity in MDA-MB-231 cells (Fig. 8A).

Next, it was studied whether BC EVs induced invasion, and the role of Src was assessed using invasion assays with MDA-MB-231 cells treated with Ctrl EVs and BC EVs. It was identified that BC EVs induced invasion of MDA-MB-231 cells (Fig. 8B). Moreover, invasion assays performed in the presence of PP2 demonstrated that invasion required Src activity (Fig. 8C).

In addition, whether BC EVs induce proliferation in MDA-MB-231 cells was examined, and proliferation assay results found that BC EVs and Ctrl EVs did not induce proliferation in MDA-MB-231 cells (Fig. 8D).