CABA I cell line was established from the ascitic fluid of an ovarian carcinoma patient not undergoing drug treatment (26). Cells were grown as monolayers in Roswell Park Memorial Institute-1640 (RPMI-1640) supplemented with 5% (v/v) heat-inactivated FBS, 1× penicillin/streptomycin, and 2 mM of l-glutamine.

Normal human dermal fibroblasts (NHDF) cell line was purchased from Lonza (Walkersville, MD, USA) and grown as a monolayer in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% (v/v) heat-inactivated FBS, 2 mM of l-glutamine, penicillin, and streptomycin. Cells were subcultured and used within the 15th doubling, as suggested by Lonza’s protocols.

Human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical cord veins; the study was conducted in accordance with the Declaration of Helsinki, approved by the Internal Review Board of L’Aquila University (protocol code 07/2018, February 2018), and informed consent was obtained from all subjects involved. Endothelial cells were grown on 1% gelatin-coated flasks in DMEM supplemented with 10% (v/v) heat-inactivated FBS, 10% (v/v) heat-inactivated newborn calf serum (NCS), 20 mM of HEPES [N- (2-hydroxyethyl) piperazine-N′- (2-ethane sulfonic acid)], 6 U/ml of heparin, 2 mM of l-glutamine, 50 µg/ml of endothelial cell growth factor (ECGF), penicillin, and streptomycin. These cells were used within the fifth passage.

All cell lines were cultured at 37°C in a humidified atmosphere with 5% CO₂, and experiments were carried out on sub-confluent (except for wound-healing assays) and mycoplasma-negative cells.

FBS, RPMI, DMEM, glutamine, penicillin, and streptomycin were purchased from Euroclone (Euroclone SpA, Milan, Italy); Hepes and ECGF were from Sigma-Aldrich (St. Louis, MO, USA); and NCS was from Gibco (Gibco, Thermo Fisher Scientific, Waltham, MA, USA).

The protocol to isolate the two different EV subpopulations, used to stimulate NHDF, had been previously set (25). Briefly, CABA I cells were starved in serum-free medium for 24 h to avoid EV release and subsequently stimulated with 5% of 40-nm-filtered FBS HyClone (Thermo Scientific, Rockford, IL, USA) in RPMI-1640; conditioned media (CMs) containing EVs were collected in sterile working conditions after 30 min and 18 h from the HyClone supplement.

To isolate EVs, these CMs were firstly centrifuged at 4°C at 600×g for 15 min and then at 1,500×g for 30 min to remove cells and large debris, respectively. The resulting supernatants were centrifuged at 100,000×g (Rotor 70Ti, Quick-Seal Ultra-Clear tubes, k_adj 221, brake 9) for 2 h at 4°C in an Optima XPN-110 Ultracentrifuge (Beckman Coulter, Brea, CA, USA). For each preparation, the EVs were derived from a starting cell number of 4,500,000–5,000,000 cells for 30-min collection and 7,500,000–9,000,000 for the 18-h collection. Isolated vesicles were resuspended in Dulbecco’s phosphate-buffered saline (PBS) (EuroClone, Milan, Italy), and the determination of vesicle quantification was carried out by measuring the vesicle-associated protein levels using the Bradford method (27) (Bio-Rad, Milan, Italy) with bovine serum albumin (BSA; Sigma-Aldrich, St. Louis, MO, USA) as the standard.

The EV subpopulations obtained with this experimental protocol have already been previously characterized by markers, NanoSight assay, and transmission electron microscopy (25). Hereinafter, EVs from CMs collected after 30 min and 18 h from the HyClone supplement will be indicated, respectively, as EVs_30′ and EVs_18h.

NHDF were administered with EVs_30′ and EVs_18h by supplying 1 µg of EVs/ml every day for up to 5 days, in a cumulative way: EVs were added every 24 h without replacing the medium for the entire duration of the treatment, so as to mimic the continuous release of EVs by cancer cells within the tumor microenvironment and the persistent exposure of fibroblasts to EVs. Treatments were performed by adding the EVs to culture media supplemented with a reduced percentage of FBS (2%) to limit the serum stimulatory effect while ensuring fibroblast survival.

Hereinafter, NHDF treated with EVs_30′ and EVs_18h will be, respectively, indicated as NHDF_30′ and NHDF_18h. Untreated fibroblasts will be indicated as NHDF.

To verify the NHDF activation into a CAF-like state, 48 h after the end of a 5-day treatment with EVs, NHDF, NHDF_30′, and NHDF_18h were washed three times with PBS and lysed in radioimmunoprecipitation assay (RIPA) Lysis Buffer, containing 50 mM of Tris-HCl, pH 7.5, 150 mM of NaCl, 0.5% sodium deoxycholate, 1% Triton-X, 0.1% sodium dodecyl sulfate (SDS), 5 mM of EDTA, 100 mM of sodium fluoride (NaF), 2 mM of sodium orthovanadate (Na₃VO₄), 10 mM of sodium pyrophosphate (NaPPi), 1 mM of phenylmethylsulfonyl fluoride (PMSF), 1 μg/ml of leupeptin, 1 μg/ml of aprotinin, and 100 μg/ml of trypsin inhibitor (Sigma, St. Louis, MO, USA). Fibroblasts’ protein content was determined by the Bradford method, as described above. Fibroblast activation protein (FAP) and α-smooth muscle actin (α-SMA) expression were identified in samples containing 12 and 15 µg of protein (for FAP and α-SMA, respectively) resolved by 7.5% and 12.5% SDS–polyacrylamide gel electrophoresis (SDS-PAGE) (for FAP and α-SMA, respectively) under reducing conditions and with heating. Separated proteins were then blotted onto a nitrocellulose membrane (Whatman-GE Healthcare Life Sciences, London, UK), and non-specific binding sites were blocked for 2 h in 10% non-fat dry milk in TBS containing 0.5% Tween-20 (TBS-T) at room temperature.

Blots were then probed with the specific primary antibody at 4°C overnight: FAP (rabbit monoclonal, 1:1,000 dilution, ab207178, Abcam, Cambridge, UK) and α-SMA (rabbit monoclonal, 1:5,000 dilution, ab32575, Abcam, Cambridge, UK). GAPDH (mouse monoclonal, 1:5,000 dilution; MA5-11114; Thermo Scientific) was used as a normalizer. After several washes in TBS-T, the membranes were incubated in appropriate horseradish peroxidase (HRP)-conjugated secondary Abs: goat anti-mouse IgG-HRP, dilution 1:10,000 (sc-2005, Santa Cruz Biotechnology, Dallas, TX, USA) or goat anti-rabbit IgG-HRP, dilution 1:7,500 (sc-2204, Santa Cruz Biotechnology) for 1 h. All the antibodies were diluted in blocking buffer (TBS-T containing 1% non-fat dry milk). After being washed in TBS-T, the reactive bands were visualized with a chemiluminescence detection kit (SuperSignal West Femto Chemiluminescent Substrate, Thermo Scientific).

Images were recorded and analyzed with the gel documentation system Alliance LD2 (Uvitec, Cambridge, UK).

To verify if treated NHDF modify their secretome, after the 5 days of cumulative treatment with 1 μg of EVs/ml, cells were washed with serum-free DMEM and then incubated for 24 h in a complete medium in which FBS was replaced with 0.2% Lactalbumin Enzymatic Hydrolysate (LEH; Sigma, St. Louis, MO, USA) to remove the contribution of enzymes/growth factors from the serum. Parallelly, CM was prepared in the same manner from untreated fibroblasts (controls). Cells and cell debris were removed by centrifugation at 600–1,550×g from all the CMs. Then, CMs were concentrated using Centricon Ultracel YM-10 filters (Amicon Bioseparations; Millipore Corporations, MA, USA; cutoff, 10 kDa) to be analyzed by casein–plasminogen zymography assays or were used unconcentrated for tests such as proliferation, migration, and invasion assays, in addition to gelatin zymography assays.

Hereinafter, CMs obtained from NHDF, NHDF_30′, and NHDF_18h will be indicated, respectively, as CM NHDF, CM NHDF_30′, and CM NHDF_18h.

NHDF (1,000 cells/well) were seeded onto a 96-well plate, incubated for 24 h in complete medium to enable cell adhesion and spreading, and then treated with EVs_30′ and EVs_18h as explained above (1 µg of EVs/ml every day for 5 days). The effects of EVs on NHDF proliferation were evaluated by the XTT assay on the 5th day, i.e., at 96 h from the beginning of the EV treatment, while treatment was still in progress. Untreated fibroblasts, grown in the same medium but without EVs, were used as control.

For experiments with CM, NHDF (1,000 cells/well), ovarian cancer cells CABA I (1,500 cells/well), and HUVECs (1,000 cells/well) were seeded into 96-well plates (gelatin-coated for HUVECs), allowed to adhere and spread for 24 h at 37°C and 5% CO₂, and then cultured for 96 h (NHDF) or 72 h (CABA I and HUVECs) with the CM NHDF_30′ and CM NHDF_18h. At the end of each specified interval, the proliferation was assessed with the XTT assay.

CMs for experiments on NHDF were supplemented with 1% FBS to ensure fibroblast survival, without stimulating their growth; for the same reason, CMs for HUVEC experiments were supplemented with 5% FBS, 5% NCS, HEPES, heparin, and ECGF; CABA I cells were incubated with unsupplemented CM. Cells incubated with CM NHDF were used as controls.

For the proliferation assay, 1 mg/ml of XTT [2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxamide] (Sigma, St. Louis, MO, USA) and 1.53 mg/ml of phenazine methosulfate (PMS; Sigma-Aldrich) were mixed, and 50 μl of this solution was added to each well. Plates were incubated for 4 h at 37°C, 5% CO₂; after this interval, the optical density (OD) of the colored, non-toxic, water-soluble formazan originated by the metabolic reduction of XTT mixed with PMS by mitochondria of living cells was measured by an ELISA reader at 450 nm. Values obtained in the absence of cells were considered as background and subtracted from the OD values of the samples. XTT tests were performed before the cells reached confluence to prevent any possible artifact decrease in the results due to contact inhibition.

Each experiment was performed in triplicate and repeated at least twice. The data are expressed as the means ± SDs.

The wound-healing assay is one of the earliest developed tests to study directional cell migration in vitro, and it is based on the observation of cell migration into a scratch “wound” created on a cell monolayer.

NHDF were cultured in 24-well microplates and treated as previously explained. The scratch was performed at 48 h after the end of 5 days’ treatment with EVs_30′ and EVs_18h when the cells had reached the full confluency; a previously sterilized 200-µl plastic tip was drawn across the cellular stratum to produce a wound, floating cells were removed, and wells were washed 3 times with PBS to remove debris and to smooth the edge of the wound. During the migration into the wound, cells were maintained in an FBS reduced culture medium (2% FBS) that avoided scratch closure by means of cell growth.

The status of the scratch wounds was monitored up to 48 h using a contrast-phase microscope; representative images were collected at the beginning of the assay and at regular intervals. The surface of the wounded area in each image was quantified with the ImageJ software, and the data were reported as % of wound closure (compared to 100%, conventionally assigned to the original scratch area).

The study of cell invasiveness was accomplished using modified Boyden chambers, separating the upper and lower compartments with filters (8-μm pore size polycarbonate polyvinylpyrrolidone-free Nucleopore filters) coated with a thin layer of Matrigel^® Growth Factor reduced (Beckton Dickinson, Franklin Lakes, NJ, USA) diluted in serum-free medium to a concentration of 0.5 mg/ml.

Briefly, NHDF, NHDF_30′, and NHDF_18h (1,000 cells/well) were added to the upper chamber in 45 μl of serum-free medium, and their motility abilities were tested using as chemoattractant some DMEM containing 10% FBS, which was added into the lower chamber; NHDF were used as controls.

In experiments with ovarian cancer cells, CABA I cells (1,000 cells/well) were added to the upper chamber in 45 μl of serum-free medium, and in the lower chamber were added the serum-free CM NHDF_30′ and CM NHDF_18h to test their effect as chemoattractant; cells invading in response to CM NHDF were used as controls.

The cells were allowed to invade the Matrigel^® for 24 h at 37°C, 5% CO₂. The non-invading cells on the upper surface of the 8-μm pore filters were removed with a cotton swab. The invading cells on the filters’ lower surface were fixed and stained in 1% crystal violet in methanol. Invading cells in five random microscope fields for each well were counted at 20× magnifications.

Serum-free CM NHDF, CM NHDF_30′, and CM NHDF_18h were subjected to both gelatin and casein–plasminogen zymography assays. Gelatin zymography was performed using 7.5% SDS-PAGE copolymerized with 1 mg/ml of gelatin type B (Sigma, St. Louis, MO, USA); the CMs were diluted in SDS-PAGE sample buffer and analyzed under non-reducing conditions without heating. After electrophoresis, the gels were washed three times, 15 min each, at room temperature, in a washing buffer containing 50 mM of Tris-HCl (pH 7.4) and 2.5% Triton X-100 (Sigma-Aldrich); they were, then, incubated overnight in an activation Tris buffer (50 mM of Tris-HCl, pH 7.4, 5 mM of CaCl₂, and 120 mM of NaCl) at 37°C. To visualize the lytic bands, the gels were stained with Coomassie Blue R 250 (Bio-Rad, Hercules, CA, USA) dissolved in a mixture of methanol:acetic acid:water (4:1:5) for 30 min and then destained in the same solution without dye.

The plasminogen activators (PAs) in the concentrated culture CM were examined using the casein–plasminogen zymography under non-reducing conditions and without heating. Proteins were separated by electrophoresis in 10% SDS-PAGE copolymerized with 0.2% casein (Sigma-Aldrich, St. Louis, MO, USA) and 10 mg/ml of human plasminogen (Sigma-Aldrich, St. Louis, MO, USA). After electrophoresis, the gel was washed in the same buffer used for the gelatinase assay and then incubated for 48 h at 37°C in 50 mM of Tris-HCl, pH 7.4 + 0.02% NaN₃. Staining and destaining were performed as previously described. Activities of gelatinases and PAs appeared as clear and distinct bands, which indicated proteolysis of the substrate, on a blue background: those digestion bands were quantified by ImageJ software.

Scanning electron microscopy (SEM) analysis was performed on fibroblasts treated with EVs_30′ and EVs_18h for up to 5 days. Forty-eight hours after the end of this treatment, NHDF, NHDF_30′, and NHDF_18h were detached, washed, and allowed to grow to subconfluence on coverslips for an additional 96 h; then cells were fixed in 2% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) in PBS for 3 min.

After being dehydrated with a graded scale of ethanol (30% to 100%) and critical point-dried, the samples were glued onto stubs, coated with gold in an SCD040 Balzer Sputterer, and detected with Philips 505 SEM at 20 kV.

The migration of normal fibroblasts and CABA I cells was tested in response to CM NHDF_30′ and CM NHDF_18h (added as a chemoattractant in the lower chambers, the same volume for each sample). Cells migrated in response to CM NHDF were used as controls. Briefly, cells were detached, washed three times in serum-free medium, and seeded on the upper wells (5,000 cells/wells in serum-free medium) of the modified Boyden chamber. Gelatin-coated polycarbonate membranes with 8-µm pores were used to separate the upper wells from the lower ones. Each condition to be tested was analyzed in triplicate. The Boyden chambers were incubated for 24 h at 37°C in a CO₂ incubator, and then migrated cells were visualized as described for the invasion assay. The number of cells, migrated to the lower surface of the polycarbonate membrane, was counted in five random 20× fields within each well, under a microscope. The mean number of cells per field was calculated as cell counts.

This in vitro test measures the ability of endothelial cells to invade, migrate, organize, and differentiate into capillary-like tubular structures within a three-dimensional matrix constituted by Matrigel^® Growth Factor Reduced 10 mg/ml (BD56230, Franklin Lakes, NJ, USA). Briefly, Matrigel^® was plated on the bottom of 96-well plates and allowed to gel at 37°C for 1 h. HUVECs were detached, counted, washed in serum-free medium, and resuspended in serum-free CM NHDF, CM NHDF_30′, and CM NHDF_18h. Then, 20,000 cells/well were seeded on Matrigel^® and incubated at 37°C, 5% CO₂; the tube formation was observed at 7 h after cell seeding. Several images were acquired per well and processed using the Angiogenesis Analyzer plugin with ImageJ software (28) downloadable from the National Institutes of Health website. The total length of the capillary-like structures, the number of nodes, and the number of segments normalized per area were used for data analysis.

Data are expressed as the mean ± standard error. Comparisons between the means of control groups and treated groups were performed using the one-way ANOVA followed by Tukey’s post-test; results were considered statistically significant when p < 0.05 (*), p < 0.01 (**).

Potential NHDF morphological changes, a typical signature of fibroblast activation, induced by EVs_30′ and EVs_18h were observed by an inverted optical microscope.

EVs_30′ and EVs_18h are EVs isolated from CMs collected after 30 min and 18 h, respectively. As mentioned above and discussed further below, in previous work (25), we highlighted that the human ovarian cancer cell line CABA I releases two specific subpopulations of sEVs “(EVs30’) and lEVs+sEVs (EVs18h).

Such morphological changes were visible in NHDF treated with ovarian cancer EVs_30′ and EVs_18h starting after 72 h at the beginning of treatment ( Figure 1 ), whilst untreated fibroblasts exhibit typical elongated and spindle-shaped morphology, some NHDF_30′ and NHDF_18h underwent a morphological change, acquiring the typical morphology of activated fibroblasts (NHDF_30′ and NHDF_18h are, respectively, NHDF treated with EVs_30′ and EVs_18h): they appeared very spread with many visible stress-contractile fibers inside the cytoplasm.

To confirm the cell activation, at the end of the EV treatment, NHDF, NHDF_30′, and NHDF_18h were lysed as described, and protein extracts were analyzed to detect the expression of typical markers of CAFs: FAP and α-SMA ( Figure 2 ).

The quantitative analysis detected a statistically significant increase in the expression of α-SMA (calculated molecular weight: ~44 kDa) in both NHDF_30′ and NHDF_18h when compared to NHDF (1.44 and 1.35, respectively) ( Figure 2 ). FAP was also increased in both NHDF_30′ and NHDF_18h when compared to NHDF (3.3 and 4.5, respectively).

Proliferation rate alteration induced by EV treatments was evaluated. It was tested while the treatment was still ongoing on the 5th day of the EV treatment (96 h from the beginning of treatments) ( Figure 3 ): EVs_18h did not induce any significative increase, while the treatment with EVs_30′ resulted in a significant increase when compared to the untreated cells NHDF (+15%).

The motility induced by EVs_30′ and EVs_18h treatments was tested with the scratch wound assay ( Figure 4 ). Migration was observed at different time intervals (24, 32, and 48 h), and the most significant changes were captured after the beginning point (time zero); the observation of the wounded area showed that NHDF_30′ and NHDF_18h have a greater tendency to close the wound by migrating inside it compared to NHDF ( Figure 4A ); to quantify this ability to close the wound, the wounded area (i.e., the area uncovered from the cells) was measured with ImageJ software at the intermediate time intervals, 24 h ( Figure 4B ) and 32 h ( Figure 4C ). The 100% value was conventionally assigned to the wounded area of the original scratch (time zero). NHDF_18h migrated with the same trend as NHDF, while NHDF_30′ exhibited higher motility: indeed, after 24 h, the area not yet covered was quite comparable in NHDF and NHDF_18h (respectively 69% and 71% with respect to the original wound), but it was significantly lower (53% with respect to the original wound) in NHDF_30′. After 32 h, the scratch area of NHDF and NHDF_18h was again comparable (respectively 61% and 60% with respect to the original wound), but it was significantly lower in NHDF_30′ (44% compared to the original wound). After 48 h, the trend was substantially maintained, but since proliferative effects could begin to occur at this time interval, it was not considered (despite that the experiment conducted in the presence of a low concentration of serum has certainly avoided the scratch closure by means of cell growth) (data not shown).

The invasion assay performed with the modified Boyden chamber showed that both fibroblasts treated with EVs_30′ and EVs_18h showed a trend to a greater invasiveness capacity (respectively +101% and +30%) as compared to NHDF ( Figure 5A ), but only NHDF_30′ had a statistically significant greater ability if compared to NHDF. To estimate if the invasion ability induced by the EV treatment could be supported by an increased secretion of proteolytic enzymes, CMs from EV-treated fibroblasts were normalized according to the same volume and assayed to evaluate the gelatinolytic and PA activities by employing zymographic techniques. The gelatinase assay ( Figure 5B ) revealed that both EVs_30′ and EVs_18h induced in NHDF the expression of pro-MMP-2: +41% and +24%, respectively, in NHDF_30′ and NHDF_18h compared to NHDF (calculated molecular weight 70 kDa). The casein–plasminogen zymography ( Figure 5C ) similarly highlighted a trend to a higher release of the high-molecular-weight urokinase-type PA (HMW-uPA) (calculated molecular weight 48–55 kDa), particularly in NHDF_30′ (+51% in NHDF_30′ compared to NHDF).

NHDF_30′ and NHDF_18h cell surface was also observed by SEM to verify if EV-mediated activation stimulated, in turn, the EV release, particularly of MVs from the cell surface, whilst the shedding of MVs was very sporadic in untreated NHDF; in EV-treated fibroblasts, the extent of the MV release was more evident and involved large membrane areas ( Figure 6 ).

After the end of the EV treatment, the CMs of NHDF_30′ and NHDF_18h (representing the cell secretome and containing both EV-associated and soluble molecules) were used as stimuli to evaluate their effect on the cells normally present in the tumor microenvironment, such as fibroblasts, endothelial cells, and tumor cells. CM from untreated NHDF was used as a control.

Fibroblast proliferation rate was not at all affected by CM ( Figure 7A ). On the contrary, CM NHDF_30′ and CM NHDF_18h exerted a considerable chemotactic effect, stimulating the migration ability of normal fibroblasts ( Figure 7B ): migration of fibroblasts toward the secretome of EV-treated fibroblasts was almost 2-fold increased with respect to the migration toward the CM NHDF (+93% and + 81%, respectively for CM NHDF_30′ and CM NHDF_18h), even if no differences were appreciable between CM NHDF_30′ and CM NHDF_18h.

The effects of activated fibroblasts’ secretome on ovarian cancer cells were also analyzed evaluating the migration and invasion abilities, in addition to their proliferative capacity ( Figure 8 ). CABA I cells cultured in the presence of CM NHDF, CM NHDF_30′, and CM NHDF_18h showed no significant change in their proliferation ( Figure 8A ). On the contrary, their motility ( Figure 8B ) and invasiveness ( Figure 8C ) were significantly promoted: they were higher in response to CM NHDF_30′ and CM NHDF_18h than in response to CM NHDF, with no significant differences between CM NHDF_30′ and CM NHDF_18h (motility, +140% and +116% compared to CM NHDF in CM NHDF_30′ and CM NHDF_18h, respectively; invasion, +158% and +176% compared to CM NHDF in CM NHDF_30′ and CM NHDF_18h, respectively).

HUVECs, too, were stimulated by CM NHDF, CM NHDF_30′, and CM NHDF_18h to assess their proliferation response ( Figure 9 ): although the cell number appeared to increase in endothelial cells treated with CM from EV-treated fibroblasts, this increase was not statistically significant ( Figure 9A ). On the contrary, the tube formation assay highlighted the pro-angiogenic potential of CM. The test revealed that the differentiation of HUVECs into primitive capillary-like structures occurred in response to both CM NHDF_30′ and CM NHDF1_8h; the number of nodes, the total length of formed tubes, and the number of segments were significantly higher in HUVECs treated with CM NHDF_30′ and CM NHDF_18h than with CM NHDF ( Figure 9B ) (number nodes/area: 14.6 in HUVECs treated with control NHDF CM; 56.7 and 39.6 in CM NHDF_30′- and CM NHDF_18h-treated HUVECs, respectively. Total length/area: 1,078 pixels in HUVECs treated with control CM NHDF; 1,751.6 and 1545.8 pixels in CM NHDF_30′ and CM NHDF_18h treated HUVECs, respectively. Number segments/area: 2.12 in HUVECs treated with control CM NHDF; 19.7 and 14 in CM NHDF_30′ and CM NHDF_18h treated HUVECs, respectively).