STAT-3 inhibitor Stt was purchased from Santa Cruz Biotechnology, Inc. and used at non-toxic doses below the reported half-maximal inhibitory concentration (<10 µM) (19). Stt was dissolved in DMSO (Sigma-Aldrich; Merck KGaA) to create 50 mM stock solution and stored in aliquots at −80°C until use. The IL-6R humanized monoclonal antibody Tcz was purchased from Roche Diagnostics GmbH as Actemra/RoActemra^®.

The non-tumor immortalized human prostate epithelial cell line RWPE-1 was maintained in keratinocyte serum-free medium containing 0.5 mg/ml bovine pituitary extract, 5 ng/ml human recombinant EGF supplement and penicillin/streptomycin (100 IU/ml; Gibco; Thermo Fisher Scientific, Inc.). The prostate carcinoma cell lines 22Rv-1, LNCaP and DU-145 were maintained in RPMI-1460 culture medium supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin (100 IU/ml; all Gibco; Thermo Fisher Scientific, Inc.). All cell lines were obtained from American Type Culture Collection. All cultures were maintained at 37°C in 5% CO₂ and routinely tested for mycoplasma; cells were negative throughout the study. When PCa cells reached 80% confluence, they were harvested with Trypsin^® (Gibco; Thermo Fisher Scientific, Inc.) and seeded at different densities for subsequent experiments.

DU-145 cells were treated at 37°C for 24–72 h with Stt (3 µM), Tcz (10 µg/ml) or IL-6 (50 ng/ml) or their simultaneous combinations as follows: Stt + IL-6; Tcz + IL-6; Stt + Tcz or Stt + Tcz + IL-6. For Stt + IL-6, Tcz + IL-6 and Stt + Tcz + IL-6, addition of IL-6 was performed at 1 h following treatment with Stt and/or Tcz. An untreated control group (UCG) did not receive any treatment. The cells were cultured overnight at 37°C prior to treatment to allow cells to attach to the plates.

PCa 22Rv1, LNCaP, DU-145, and RWPE-1 cells were grown in 75-cm² flasks to 90% confluence. Cells were harvested using Trypsin solution (Gibco; Thermo Fisher Scientific, Inc.). A total of 250,000 cells was plated into a 25-cm² flask containing fresh medium. Cultures were maintained at 37°C in a humidified atmosphere with 5% CO₂. Following 12, 24 and 48 h incubation, supernatant was collected under sterile conditions, centrifuged at 4°C, 1,400 × g for 10 min, and filtered through a membrane (0.2 µm; MilliporeSigma). The supernatant was stored at −80°C until use.

Bead-based multiplex assay was used to quantify cytokines and growth factor in supernatant of prostate-derived cell lines using LEGENDplex™ HU Essential Immune Response Panel and LEGENDplex Human Growth Factor Panel (cat. no. 741061 and 740929 respectively; both BioLegend, Inc.) according to the manufacturer's instructions. Briefly, 25 µl thawed supernatant, diluted standard and blanks were added to the corresponding tubes; 25 µl pre-mixed beads and detection antibodies was added to all tubes (Growth Factor Panel Detection Antibodies, cat. no, 76303 and Human Essential Immune Response Panel Antibodies Cat. no. 750000547 both BioLegend, Inc.). Tubes were incubated for 2 h at room temperature with shaking. Then, 25 µl Streptavidin-Phycoerythrin conjugate was added, tubes were incubated for 30 min at room temperature, washed and suspended in 200 µl wash buffer. Data were acquired using an Attune Acoustic Focusing Cytometer (Thermo Fisher Scientific, Inc.) and analyzed utilizing LEGENDplex version 8.0 (BioLegend, Inc.). All results are expressed in pg/ml.

The secreted VEGF in DU-145 cell line was determined quantitatively using a commercial ELISA kit (cat. no. DVE00, Quantikine Human VEGF; R&D Systems, Inc.) according to the manufacturer's instructions. In brief, cell supernatant was diluted at 1:3 with assay diluent, 150 µl supernatant was added to wells, incubated at room temperature for 2 h and washed three times with 300 µl wash buffer. Then, 200 µl Human VEGF Conjugate was added to each well and incubated and washed as aforementioned. Finally, 200 µl substrate solution was added, color was developed for 20 min and the reaction was stopped with 50 µl STOP solution. Absorbance was measured at 450 nm using a microplate reader (Synergy HT; BioTek Instruments, Inc.); data are expressed as pg/ml (analyzed using Gene5, version 3.08, biotek.com/products/software-robotics–software/gen5-microplate-reader–and-imager-software/).

PCa cells were seeded at a density of 2×10⁵ cells/well in a 96-well plate and cultured overnight at 37°C to allow attachment. The medium was replaced with fresh medium containing Stt (1, 3, 5, 7 or 10 µM) or Tcz (10, 100, 1,000 or 10,000 ng/ml) for 24 h at 37°C, Etoposide at 5 µM was used as a positive control Then, 10 µl WST-1 reagent solution (Sigma-Aldrich; Merck KGaA) was added for 2 h to each well. Absorbance was read at 480 nm with a reference wavelength of 650 nm using a microplate reader (Synergy HT; BioTek Instruments, Inc.). A total of three independent experiments were performed in triplicate. All readings were normalized to the UCG.

xCELLigence DP (Roche Diagnostics GmbH) was used to monitor cell proliferation in real time. This system allows the measurement of impedance in real time; the higher the impedance value, the higher the number of live cells growing on the surface of each well; a maximal value is reached when the cell layer becomes confluent (22). Treated or untreated RWPE-1, 22Rv1, LNCaP, and DU-145 cells were seeded at a density of 5×10³ cells/well in quadruplicate in electronic microtiter plates (E-Plate; Roche Diagnostics GmbH) and measured with programmed signal detection every 20 min for 92 h with the xCELLigence system according to the manufacturer's instructions. Data acquisition and analysis were performed with RTCA software (version 1.2; Roche Diagnostics GmbH).

Scratch wound-healing assay was performed to determine cell migration using confluent cultures (80–90% confluence). Briefly, 1×10⁵ cells/ml were seeded in a 6-well tissue culture. After cells reached confluency, they were starved for 24 h using 0.2% serum in growth medium. At 2 h before wound healing assay cells were treated with mitomycin (5 µg/ml) at 37°C for 2 h to inhibit cell proliferation, as previously described (23). A sterile p200 pipette tip was used to create a wound on the confluent monolayer and culture medium was replenished. Images of the scratch area were recorded using a guide-dot previously drawn underneath the plates at 0, 6, 12 and 24 h. The experiments were repeated three times in triplicate. Images were captured using AxioCam ERc5s inverted phase-contrast microscope (Primo Vert, Cat. no. 415510-1101-000; Zeiss AG) at 40× magnification. ImageJ software (version 1.8.0_172; National Institutes of Health) was used to calculate wound area. Each group was compared with the UCG. Closure rate (%) was calculated as follows: (Initial wound width-final wound width)/initial wound width ×100.

Invasion capacity was determined in DU-145 cells via chemotaxis assay. Transwell chambers with 8-µm pore polycarbonate filters (Wuxi NEST Biotechnology Co., Ltd.) were coated with Matrigel diluted in RPMI serum-free medium at a final concentration of 0.25 mg/ml for 24 h at 37°C prior the beginning of the invasion assay.

Briefly, DU-145 cells were seeded at 5×10⁵ cells/ml in 6-well plates and cultured at 37°C in RPMI Supplemented medium (10% FBS and penicillin/streptomycin 100 IU/ml) for 24 h. Then, the cells were washed with PBS and cultured for another 72 h at 37°C with Stt, Tcz and/or IL-6. The medium was replaced 24 h before the beginning of the invasion assay with RPMI serum-free medium (Gibco; Thermo Fisher Scientific, Inc.) with Stt, Tcz and/or IL-6 at 37°C.

For the assay, cells were harvested with Trypsin and resuspended in RPMI serum-free medium with Stt, Tcz and/or IL-6 at 5×10⁵ cell/ml, then 150 µl cell suspension was added to a Transwell chamber. RPMI culture medium with 10% FBS (both Gibco; Thermo Fisher Scientific, Inc.) used as a chemoattractant was added to the lower chambers and cells were allowed to invade through a porous membrane coated with Matrigel and incubated at 37°C for 48 h. Invading cells on the membrane were fixed in 3.7% paraformaldehyde at room temperature for 15 min and stained for 30 min at room temperature with Sulforhodamine B (SRB) (0.057% wt/vol). A total of five fields of view were observed using a Zeiss Primo Vert microscope light microscope at 100× magnification and photographed with AxioCam ERc5s.

SRB that became attached to cells was recovered and measured as previously described by Vichai and Kirtikara (24). Briefly, the Transwell inserts were placed on a new p24 plate with 500 µl 10 mM Tris base (pH, 10.5) for 30 min at room temperature. Aliquots of 100 µl recovered SRB were transferred onto a 96-well plate and the optical intensity was measured at 510 nm using a microplate reader. Invasion (%) was normalized to the control as follows: (Absorbance in treatment group/absorbance in UCG) ×100.

DU-145 cells were harvested at exponential growth phase, seeded at 200 cells/well in 6-well plates and cultured at 37°C for 24 h before addition of Stt, Tcz and/or IL-6 for 72 h at 37°C. Following treatment, cells were harvested, and 300 cells/well were cultured in a new 6-well plate. The cells were incubated with RPMI supplemented medium for 24 h at 37°C, then medium was replaced with fresh medium containing Stt, Tcz and/or IL-6. The cells were then incubated at 37°C for 14 days. During this time, the culture medium was replaced every 3 days with fresh supplemented medium with Stt, Tcz and/or IL-6. Cells were fixed using 3.7% formaldehyde at room temperature for 15 min and stained for 30 min at room temperature with Sulforhodamine B (SRB) (0.057% wt/vol). The colonies (>50 cells) were viewed with Zeiss Primo Vert light microscope at 40X magnification to observe the cell number in the colonies and photographed with AxioCam ERc5s. Colonies was counted with the soft ImageJ software (version 1.8.0_172; National Institutes of Health).

Briefly, DU-145 cells were seeded at 4×10⁶ cells/ml in 150-mm plates and cultured at 37°C in RPMI Supplemented medium (10% FBS and penicillin/streptomycin 100 IU/ml) for 24 h. Then, the cells were washed with PBS and cultured for 120 h at 37°C with Stt, Tcz and/or IL-6. Untreated cells were seeded at 4×10⁶ cells/ml in 150 mm plates and cultured at 37°C in RPMI Supplemented medium (10% FBS and penicillin/streptomycin 100 IU/ml) for 24 h, and harvested at 90% of confluency.

For the protein extraction, cells were detached with a cell scraper and centrifuged at 4°C, 1,400 × g for 10 min. A total of 300 µl RIPA buffer (0.5% deoxycholate, 0.5% NP-40, 0.5% SDS, 50 mM Tris, pH 8.0 and 150 mM NaCl) with Protease Inhibitor Cocktail (Roche Applied Science) was added and cell suspensions were incubated on ice for 30 min. The lysate was sonicated at 20 kHz for 5 min at high intensity with 30 sec rest intervals with the Bioruptor Sonicator (Diagenode SA). Protein extract was obtained following 30 min incubation at 4°C and 12 min centrifugation at 13,300 × g and 4°C. Bradford Assay kit (HGHiring; Bio-Rad Laboratories, Inc.) was used to determine concentration of proteins in the samples. Samples containing 50 µg total protein were resolved via 10% SDS-PAGE. The proteins were transferred onto a PVDF membrane (0.2-µm pore) and blocked with LI-COR Odyssey Blocking Solution Ready to use (cat. no. 927-70001; LI-COR Biosciences) under agitation for 2 h at room temperature. Primary antibodies for STAT-3 (1:1,000; clone no. 124H6) and pSTAT-3 (1:2,000; clone no. M9C6) were acquired from Cell Signaling Technology, Inc.; antibodies for vimentin (1:2,000; clone no. V9), E-cadherin (1:2,000; clone no. G-10) and IL-6R (1:1,000; clone no. H-7) were obtained from Santa Cruz Biotechnology, Inc. Primary antibodies were incubated overnight at 4°C under agitation in the dark. The membranes were washed and probed with IRDye^® 800 secondary antibodies (1:15,000; Cat. no. 926-68028; LI-COR Biosciences) for 2 h at room temperature. Washed membranes were scanned with Odyssey™ Infrared Imaging System (LI-COR Biotechnology). Optical density was measured using Image Studio Lite ver. 5.2.5 software and normalized to β-actin coupled with Alexa Fluor^® 680 (1:2,000; clone no. C4; cat. no. sc-47778 AF680; Santa Cruz Biotechnology, Inc.) or with Revert 700 Total Protein Stain (TPS; cat. no. 926-11010; LI-COR Biosciences) according to manufacturer instructions. Briefly, the membranes were submerged in TPS solution for 5 min, washed twice for 30 sec at room temperature with wash solution (LI-COR Biosciences) and imaged. Membranes were scanned using the 700 nm channel of the Odyssey scanner. TPS stain was removed with reversal solution (LI-COR Biosciences) for 5 min at room temperature with gentle shaking. Optic density was measure with the software Image Studio Ver 4.0 (LI-COR Biosciences).

For evaluation of genes involved in EMT, RNA extraction was performed using the commercial GeneJET RNA Purification kit (Thermo Fisher Scientific, Inc.). RNA was quantified using a plate reader (Synergy HT; BioTek Instruments, Inc.). Synthesis of cDNA was performed using commercial Transcriptor First Strand cDNA Synthesis kit (Roche Diagnostics GmbH) according to the manufacturer's protocol. qPCR was performed using LightCycler^® FastStart DNA Master PLUS SYBR Green I and LightCycler version 2.0 (both Roche Diagnostics GmbH). The qPCR was performed using initial denaturation at 95°C for 10 min followed by 45 cycles of denaturation at 95°C for 15 sec; annealing/extension is shown in the table I. The primer design and qPCR condition are shown in Table I. Data were normalized via the 2^−ΔCq method (25) using the RPL32 as reference gene.

ICW assay was performed using Odyssey Imaging System (LI-COR Biotechnology). Briefly, 2×10⁴ cell/well DU-145 cells were seeded at 37°C for 24 h in a 96-well optical black wall clear bottom plate (Thermo Fisher Scientific, Inc.). When cells reached 60–70% confluence, supplemented medium with Stt, Tcz and/or IL-6 were applied for 24 h at 37°C. Cells were fixed with 100 µl/well methanol-acetone solution (3:1) for 20 min at −20°C. Next, cells were permeabilized with 0.5% Triton X-100 for 15 min at room temperature and left for 24 h at 4°C. Cells were blocked with LI-COR Odyssey Blocking Solution Ready to use (cat. no. 927-70001; LI-COR Biosciences) at room temperature for 2 h. The cells were incubated at 4°C overnight with mouse IgG against STAT-3 (1:1,000; Cat. no. 9193; Cell Signaling Technology, Inc.) and pSTAT-3 (1:400; Cat. no. 9193; Cell Signaling Technology, Inc.). Following five washes with Dulbecco's PBS (HyClone; Cytiva), cells were stained with goat anti-mouse IgG IRDye 800 antibody (1:5,000; cat. no. 926-68028; LI-COR Biosciences) at room temperature for 2 h. The microplates were scanned with the Odyssey CLx Infrared Imaging System (LI-COR Biosciences). The integrated fluorescence intensity representing the protein expression levels was acquired using software provided with the imager station (Odyssey Software version 3.0; LI-COR Biosciences). Fluorescence intensity of STAT-3 and pSTAT-3 was calculated by normalization to DRAQ5 (1:5,000; Cat. no. 4084; Cell Signaling) at room temperature for 2 h. Optical density was measure with Image Studio Ver 4.0 (Provided by LI-COR Biosciences).

All experimental procedures were performed in triplicate and repeated ≥3 times unless otherwise specified. Data were analyzed using Shapiro-Wilk test to determine normality and one-way ANOVA followed by Tukey's post hoc test to compared groups. Data are presented as the mean ± standard deviation. Statistical analysis was performed using GraphPad Prism version 9.0 (GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference. BioRender.com was used to create figures.

Secretion of cytokines, chemokines was determined with CBAs assay [transforming growth factor Transforming growth factor (TGF)-β1, C-X-C chemokine ligand (CXCL)-8 (also called IL-8), INF-γ, IL-6, C-C motif chemokine ligand (CCL)-2, CXCL-10 and IL-4] and growth factors [Vascular endothelial growth factor (VEGF), Tumor necrosis factor-α, Stem cell factor, Human Platelet-Derived Growth Factor, Macrophage Colony-Stimulating Factor, hepatocyte growth factor, Granulocyte-colony stimulating factor (G-CSF) and Erythropoietin] were assessed in PCa (22Rv1, LNCaP, and DU-145) and non-tumorigenic prostate cells (RWPE-1) at 12, 24 and 48 h. PCa cells secreted high concentrations of CXCL8, IL-6, VEGF and G-CSF at 12 (Fig. 1A), 24 (Fig. 1B) and 48 h (Fig. 1C). IL-6 and VEGF secretions were higher in DU-145 than 22Rv1, LNCaP, and RWPE-1 cells (Fig. 1D and E). These results show that IL-6 and VEGF were secreted to a greater extent by metastatic cell lines. The cytokines and transcription factors expressed by cells are presented in Figs. S1 and S2.

Expression of genes associated with EMT were assessed and an inverse association between vimentin and E-cadherin was observed; these are consistent with protein levels (Fig. 2A). DU-145 highly expressed vimentin, but LNCaP and 22Rv1 cells exhibited low expression; E-cadherin was highly expressed in LNCaP and 22Rv1 cells but its expression in DU-145 cells was low (Fig. 2A). Expression levels of E-cadherin, vimentin and other proteins associated with activation of the STAT-3 pathway (IL-6R and total and pSTAT-3) were assessed. There were different protein expression in PCa (22Rv1, LNCaP and DU-145) and RWPE-1 non-tumorigenic prostate cells (Fig. 2B). The phosphorylation was different in all cell lines: RWPE-1 and DU-145 cells exhibited increased pSTAT-3 compared with 22Rv1 and LNCaP cells (Fig. 2C). The expression of E-cadherin was lower in DU-145 compared with 22Rv1 and LNCaP cells but not significantly different compared with RWPE-1 cells; high expression of vimentin was observed in RWPE-1 and DU-145 compared with 22Rv1 and LNCaP cells. IL-6R was more highly expressed in metastatic LNCaP and DU-145 than RWPE-1 and 22Rv1 cells (Fig. 2D). The expression of vimentin was higher in cells expressing pSTAT-3; on the other hand expression of E-cadherin was lower in the same cells. These results reveal a potential association between expression of vimentin and IL-6R and downregulation of E-cadherin in a pSTAT-3- dependent matter.

Proliferation and migration capacity were assessed in cell lines. DU-145 cells exhibited a higher proliferative capacity than other cell lines (Fig. 3A); at 48 h, proliferation was higher than in RWPE-1, 22Rv1 and LNCaP cells (Fig. 3B). Likewise, the migratory capacity of cells was different at 6, 12 and 24 h (Fig. 3C and D). The wound-closure rate was higher in DU-145 cells (Fig. 3E). The proliferative and migratory capacity exhibited a similar pattern among cell lines. The order of the proliferative and migratory capacities of the cell lines was LNCaP, RWPE-1, 22Rv1 and DU-145. These results demonstrated that the metastatic androgen-independent cell line DU-145 had higher proliferative and migratory capacity than other cell lines. IL-6 secretion and IL-6R expression results showed the role of the IL-6/pSTAT-3 circuit in DU-145 PCa cells. DU-145 cells were therefore selected to evaluate the effect of inhibition of IL-6/IL-6R/STAT-3 on the migratory, invasive, clonogenic and proliferative capacity of a cell line expressing activated STAT-3.

Before evaluating the effect of STAT-3 inhibition on DU-145 cells, the effect of Stt and Tcz on cell viability was assessed. Stt decreased viability in DU-145 cells at a concentration of ≥5 µM (Fig. 4A). Moreover, Tcz (10–1,000 ng/ml) did not decrease viability of DU-145 cells (Fig. 4B). Therefore, Stt at 3 µM and Tcz 10 µg/ml were selected, based on a previous report (26). The effect of Stt and/or Tcz on STAT-3 phosphorylation was assessed. Both Stt and Tcz decreased phosphorylation of STAT-3 by 40% compared with UCG (Fig. 4C and D). Following addition of IL-6, the inhibitory effect was decreased for both treatments. However, Stt + Tcz maintained high and effective inhibition. At 24 h, STAT-3 phosphorylation levels were lower than 20% even when IL-6 was added. These results indicated that IL-6 decreased the inhibitory effect produced by Stt. However, combined Stt + Tcz was more effective in inhibiting pSTAT-3 and addition of IL-6 did not significantly increase pSTAT-3 expression.

The effect of Stt and/or Tcz, on expression of VEGF, vimentin and E-cadherin was assessed. Both Stt and Tcz alone decreased VEGF compared with the IL-6 group in DU-145 cells. Following added IL-6 to Stt or Tcz, we did not observe this effect; however, Stt + Tcz combination decreased VEGF secretion compared with the IL-6 group (Fig. 5A). A similar effect was observed for vimentin (Fig. 5B and C); Stt + Tcz significantly decreased baseline vimentin expression in DU-145 cells and reversed IL-6-mediated induction of vimentin expression. E-cadherin, (a key molecule that maintains cell-cell adherence and regulates the Wnt pathway (27), was increased by Tcz but decreased in the IL-6 group. In the Stt + Tcz group, E-cadherin expression was increased even when IL-6 was added (Fig. 5D). These results demonstrated that combined Stt + Tcz was effective in inhibiting the effect of IL-6 on tumor cells, resulting in upregulation of E-cadherin and downregulation of vimentin and VEGF in DU-145 PCa cells.

Next, proliferative capacity was assessed using xCELLigence. All groups treated with Stt exhibited decreased proliferation in a time-dependent manner from 12 to 48 h; at 72 h the effect was non-significant in comparison with UCG (Fig. 6A).

IL-6 treatment increased colony formation compared with UCG (160 vs. 100%, respectively; Fig. 6B). Stt decreased colony formation to <10% (Fig. 6C) but Stt + IL-6 treatment increased colony formation. Stt + Tcz decreased the number of colonies to <10% in comparison with UCG and Stt + Tcz + IL-6 did not increase the number of colonies significantly. This showed that Stt + Tcz was effective in decreasing colony formation capacity in the presence or absence of IL-6.

To evaluate the effect of the treatment on migration and invasion, cells were treated for 72 h and wound-healing assay was performed (Fig. 7A). IL-6 significantly increased migratory capacity in DU 145 cells (Fig. 7B and C) and Stt decreased cell migration. The effect produced by Stt is partially reversed when IL-6 was added (Stt + IL-6 group) but in the Stt + Tcz group, inhibition of the cell migration capacity was maintained following IL-6 treatment. This demonstrated that Stt + Tcz inhibited migration of DU-145 in the presence or absence of IL-6.

Next, invasive capacity of cells was evaluated via chemiotaxis assay (Fig. 7D and E). There was a similar pattern to that found in the wound-healing assay: IL-6 increased invasive capacity and Stt decreased invasion. Stt + Tcz decreased invasive capacity even when IL-6 was added. This revealed that IL-6 decreased the inhibitory effect produced by Stt alone. However, combined Stt + Tcz increased the inhibitory effect on migratory and the invasive capacity of metastatic DU-145 cells via blockade of the IL-6/IL-6R/STAT-3 axis.