Advertisement

Toll-Like Receptor 3 Overexpression Induces Invasion of Prostate Cancer Cells, whereas Its Activation Triggers Apoptosis

      Toll-like receptor 3 (TLR3) is an endosomal receptor expressed in several immune and epithelial cells. Recent studies have highlighted its expression also in solid tumors, including prostate cancer (PCa), and described its role mainly in the proinflammatory response and induction of apoptosis. It has been found up-regulated in some castration-resistant prostate cancers. However, the role of TLR3 in prostate cancer progression remains largely unknown. We have experimentally demonstrated that exogenous TLR3 activation in PCa cell lines leads to the significant induction of secretion of the cytokines IL-6, IL-8, and interferon-β, depending on the model and chemoresistance status. Transcriptomic analysis of TLR3-overexpressing cells revealed a functional program that is enriched for genes involved in the regulation of cell motility, migration, and tumor invasiveness. Increased motility, migration, and invasion in TLR3-overexpressing cell line were confirmed by several in vitro assays and using an orthotopic prostate xenograft model in vivo. Furthermore, TLR3-ligand induced apoptosis via cleavage of caspase-3/7 and poly (ADP-ribose) polymerase, predominantly in TLR3-overexpressing cells. We conclude that TLR3 may be involved in prostate cancer progression and metastasis; however, it might also represent an Achilles heel of PCa, which can be exploited for targeted therapy.
      Prostate cancer (PCa) remains nowadays one of the most frequent cancer types affecting men in developed countries.
      • Sung H.
      • Ferlay J.
      • Siegel R.L.
      • Laversanne M.
      • Soerjomataram I.
      • Jemal A.
      • Bray F.
      Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
      Given the dependency of prostate cells on androgenic hormones, one common curative approach aimed to maintain disease under control is the androgen-deprivation therapy, which consists of androgen depletion via chemical or surgical castration and the use of androgen receptor–targeting molecules, such as enzalutamide and abiraterone acetate. Despite initial sensitivity to anti–androgen receptor therapy, patients eventually develop resistance to these therapeutic agents, progressing into a condition known as castration-resistant PCa associated with further development of the tumor.
      • Davies A.
      • Conteduca V.
      • Zoubeidi A.
      • Beltran H.
      Biological evolution of castration-resistant prostate cancer.
      In addition to endocrine therapy, treatment strategies include the use of chemotherapeutic agents, such as docetaxel, with limited efficiency, which ultimately may develop resistance, leading to a metastatic condition.
      • Marin-Aguilera M.
      • Codony-Servat J.
      • Kalko S.G.
      • Fernandez P.L.
      • Bermudo R.
      • Buxo E.
      • Ribal M.J.
      • Gascon P.
      • Mellado B.
      Identification of docetaxel resistance genes in castration-resistant prostate cancer.
      Recently, more attention switched toward poly (ADP-ribose) polymerase inhibitors, and immunotherapy, to allow long-term suppression of tumor growth and improve clinical benefits.
      • Fay E.K.
      • Graff J.N.
      Immunotherapy in prostate cancer.
      An important role in the immune response is covered by toll-like receptors (TLRs) and, in the last decade, an increasing number of studies have been focusing on TLRs as potential players in cancer progression or regression.
      • Hossain D.M.
      • Pal S.K.
      • Moreira D.
      • Duttagupta P.
      • Zhang Q.
      • Won H.
      • Jones J.
      • D'Apuzzo M.
      • Forman S.
      • Kortylewski M.
      TLR9-targeted STAT3 silencing abrogates immunosuppressive activity of myeloid-derived suppressor cells from prostate cancer patients.
      ,
      • Ou T.
      • Lilly M.
      • Jiang W.
      The pathologic role of toll-like receptor 4 in prostate cancer.
      In particular, TLR3 was shown to be expressed in several cancer types and associated with context-dependent prognoses in patients.
      • Muresan X.M.
      • Bouchal J.
      • Culig Z.
      • Soucek K.
      Toll-like receptor 3 in solid cancer and therapy resistance.
      TLR3 is an endosomal receptor that binds double-stranded RNAs (dsRNAs), and therefore is intensively studied for its involvement in the antiviral response and pro-apoptotic effects.
      • Gao D.
      • Ciancanelli M.J.
      • Zhang P.
      • Harschnitz O.
      • Bondet V.
      • Hasek M.
      • Chen J.
      • Mu X.
      • Itan Y.
      • Cobat A.
      • Sancho-Shimizu V.
      • Bigio B.
      • Lorenzo L.
      • Ciceri G.
      • McAlpine J.
      • Anguiano E.
      • Jouanguy E.
      • Chaussabel D.
      • Meyts I.
      • Diamond M.S.
      • Abel L.
      • Hur S.
      • Smith G.A.
      • Notarangelo L.
      • Duffy D.
      • Studer L.
      • Casanova J.L.
      • Zhang S.Y.
      TLR3 controls constitutive IFN-beta antiviral immunity in human fibroblasts and cortical neurons.
      Recent investigations demonstrated that TLR3 is expressed also in PCa, and its exogenous activation by synthetic dsRNA poly(I:C) leads to apoptosis of PCa cells.
      • Gambara G.
      • Desideri M.
      • Stoppacciaro A.
      • Padula F.
      • De Cesaris P.
      • Starace D.
      • Tubaro A.
      • Del Bufalo D.
      • Filippini A.
      • Ziparo E.
      • Riccioli A.
      TLR3 engagement induces IRF-3-dependent apoptosis in androgen-sensitive prostate cancer cells and inhibits tumour growth in vivo.
      • Kourko O.
      • Smyth R.
      • Cino D.
      • Seaver K.
      • Petes C.
      • Eo S.Y.
      • Basta S.
      • Gee K.
      Poly(I:C)-mediated death of human prostate cancer cell lines is induced by interleukin-27 treatment.
      • Palchetti S.
      • Starace D.
      • De Cesaris P.
      • Filippini A.
      • Ziparo E.
      • Riccioli A.
      Transfected poly(I:C) activates different dsRNA receptors, leading to apoptosis or immunoadjuvant response in androgen-independent prostate cancer cells.
      On the contrary, there is new evidence that shows that TLR3 stimulation may affect PCa cells' metabolism and sustain tumor progression.
      • Magnifico M.C.
      • Macone A.
      • Marani M.
      • Bouzidi A.
      • Giardina G.
      • Rinaldo S.
      • Cutruzzola F.
      • Paone A.
      Linking infection and prostate cancer progression: toll-like receptor3 stimulation rewires glucose metabolism in prostate cells.
      These findings suggest that TLR3 can be important in both PCa progression and regression, although its function in either pathway remains to be defined.
      Herein, we show for the first time that TLR3 expression is associated with chemoresistance and migratory phenotype in PCa. Moreover, we found that, based on the activation status of the receptor, it could serve as a therapeutic target in PCa.

      Materials and Methods

      Cell Culture

      Parental DU145 cells were acquired from ATCC (Manassas, VA) and grown in RPMI 1640 medium [Gibco, Thermo Fisher Scientific (TFS), Waltham, MA] supplemented by 10% fetal bovine serum (FBS; Gibco, TFS). Therapy-resistant DU145, PC3, and LAPC4 prostate cancer cell lines were derived as previously reported.
      • Puhr M.
      • Hoefer J.
      • Schafer G.
      • Erb H.H.
      • Oh S.J.
      • Klocker H.
      • Heidegger I.
      • Neuwirt H.
      • Culig Z.
      Epithelial-to-mesenchymal transition leads to docetaxel resistance in prostate cancer and is mediated by reduced expression of miR-200c and miR-205.
      ,
      • Hoefer J.
      • Akbor M.
      • Handle F.
      • Ofer P.
      • Puhr M.
      • Parson W.
      • Culig Z.
      • Klocker H.
      • Heidegger I.
      Critical role of androgen receptor level in prostate cancer cell resistance to new generation antiandrogen enzalutamide.
      The chemotherapy-naïve PC346C and PC339 xenografts and their resistant derivatives were established, as described previously.
      • de Morree E.S.
      • Bottcher R.
      • van Soest R.J.
      • Aghai A.
      • de Ridder C.M.
      • Gibson A.A.
      • Mathijssen R.H.
      • Burger H.
      • Wiemer E.A.
      • Sparreboom A.
      • de Wit R.
      • van Weerden W.M.
      Loss of SLCO1B3 drives taxane resistance in prostate cancer.
      ,
      • van Weerden W.M.
      • Bangma C.
      • de Wit R.
      Human xenograft models as useful tools to assess the potential of novel therapeutics in prostate cancer.
      Detailed information about resistant cell models is described in Supplemental Table S1.
      • van Soest R.J.
      • de Morree E.S.
      • Kweldam C.F.
      • de Ridder C.M.A.
      • Wiemer E.A.C.
      • Mathijssen R.H.J.
      • de Wit R.
      • van Weerden W.M.
      Targeting the androgen receptor confers in vivo cross-resistance between enzalutamide and docetaxel, but not cabazitaxel, in castration-resistant prostate cancer.
      All cell lines and derived sublines were cultivated at 37°C in a humidified atmosphere. The AmpFLSTR Identifiler PCR Amplification Kit (Applied Biosystems, TFS) was used to confirm the origin of cell lines. Cell lines were regularly tested for mycoplasma contamination using PCR.

      Generation of Constructs and Stable Cell Lines

      For TLR3 overexpression (OE), DU145 cells in suspension were electroporated with the hTLR3-pcDNA3 construct (a gift from Saumen Sarkar; Addgene plasmid number 32712
      • Zhu J.
      • Smith K.
      • Hsieh P.N.
      • Mburu Y.K.
      • Chattopadhyay S.
      • Sen G.C.
      • Sarkar S.N.
      High-throughput screening for TLR3-IFN regulatory factor 3 signaling pathway modulators identifies several antipsychotic drugs as TLR inhibitors.
      ) using Neon Electroporation System (TFS) and immediately seeded to antibiotic-free media. After at least 24 hours, the medium was replaced by a fresh complete medium, containing G418 (1 mg/mL; Roche, Basel, Switzerland) for selection. For dominant-negative TLR3 overexpression, DU145 cells were transfected with pUNO1-hTLR03-DN (InvivoGen, San Diego, CA), and pUNO1 empty vector (EV) served as an appropriate control. Blasticidin-selected pools were further single-cell cloned by limiting dilution. Protein overexpression was detected by Western blot analysis. For knockout (KO) clones, single-guide RNAs targeting human TLR3 were subcloned into lentiCRISPRv2 blast (a kind gift from Dr. Feng Zhang, Massachusetts Institute of Technology, Boston, MA). The human-specific single-guide RNA sequence was 5′-CACCGcaactttattgggactaaag-3′. CRISPR construct was combined with third-generation lentiviral system for the generation of lentiviral particles, with which parental DU145 cells were transduced in presence of polybrene (8 μg/mL; Sigma-Aldrich, Merck, St. Louis, MO) for 8 hours, followed by medium replacement to the fresh one. After 24 hours, selection with blasticidin (1 μg/mL; Invitrogen, TSF) was started. After 5 to 7 days, selection antibiotics (G418 or blasticidin) were removed, and single-cell colonies were picked or derived by limiting dilution in a 96-well plate. After expansion, overexpression and KO were validated by Western blot analysis, and the Cas9-mediated mutation was verified by DNA sequencing (data not shown). Clones were validated for the activity of TLR3 by stimulation with poly(I:C) (InvivoGen).

      Generation of shRNA Stable Clones

      TLR3-overexpressing clones were transduced overnight with control or TLR3 shRNA lentiviral particles (catalog number sc-36685-V; Santa Cruz Biotechnology, Dallas, TX) in presence of polybrene (5 μg/mL). The medium was replaced with a fresh one, and cells were recovered for 24 hours, then the selection was started with the addition of puromycin (1 μg/mL; InvivoGen). After 3 to 4 days, puromycin was removed, and single-cell colonies were picked. After expansion, knockdown was verified by Western blot analysis. Clones were then validated for TLR3 activation by stimulation with poly(I:C).

      RNA Isolation from Cell Cultures, Reverse Transcription, and RT-qPCR Analysis

      RNA was isolated using a High Pure RNA Isolation Kit (Roche) and quantified on BioSpectrometer (Eppendorf, Hamburg, Germany). cDNA synthesis was performed with equal amounts of RNA using a High-Capacity RNA-to-cDNA Kit (Applied Biosystems, TFS). mRNA levels were measured with gene-specific primers targeting TLR3 (forward, 5′-AGAGTTGTCATCGAATCAAATTAAAG-3′; and reverse, 5′-AATCTTCCAATTGCGTGAAAA-3′) in combination with UPL hydrolysis probe number 80 using Roche LightCycler 480 system (Roche), and relative expression levels were normalized to reference gene TBP. For the analysis of gene expression, real-time quantitative PCR was performed in 384-well customized plates (RealTime Ready real-time quantitative PCR assay; Roche). The genes of interest evaluated in the expression screening comprised all TLRs and TLR downstream markers (Supplemental Table S2). The results of real-time quantitative PCR analyses were calculated with the ΔΔthreshold cycle method in comparison to all housekeeping genes (GAPDH, ALAS1, and SDHA
      • Schmittgen T.D.
      • Livak K.J.
      Analyzing real-time PCR data by the comparative C(T) method.
      ). Heat map analysis and hierarchical clustering (one minus Pearson correlation) were performed by the use of Morpheus software (Broad Institute, Cambridge, MA).

      Patient Samples and Immunohistochemistry

      Archival tissue samples from prostate cancer patients were obtained at the University Hospitals in Olomouc and Ostrava during the years 2008 and 2017 (Supplemental Tables S3 and S4). The study was approved by the Ethical Committee of the University Hospital and Faculty of Medicine and Dentistry, Palacký University in Olomouc (reference number 121/11). Immunohistochemical staining was performed both for tumor tissues as well as for cells with known TLR3 expression. The pellet of cells was resuspended in the mixture of 225 μL plasma (blood was collected in sodium citrate vacuum tubes from volunteers and spun for 5 minutes at 3000 × g, and the supernatant was frozen in aliquots) with 5.65 μL of 1 mol/L calcium chloride. A clot was formed by adding 22.5 μL thrombin (120 NIHU/mL; Sigma Aldrich, Merck) and transferred by pipette tip into Tissue-Tek Paraform Sectionable Cassette for small samples (Sakura Finetek, Alphen aan den Rijn, the Netherlands). Formalin fixation and paraffin embedding were performed according to standard protocol. The sections were immunostained with monoclonal TLR3 antibody [clone 40F9.6 (Innate Pharma); dilution 1:200; antigen retrieval FLEX, pH 9 (Dako)]. We have also tested commercial antibodies [clone TLR3.7 (Santa Cruz Biotechnology); clone 40C1285.6/IMG-315A (TFS); and polyclonal PK-AB718-3643 (Promokine; Promocell, Heidelberg, Germany)]; however, none of them gave reasonable immunohistochemical results with formalin-fixed and paraffin-embedded control cells.

      TLR3 Stimulation and Inhibition

      Cells were seeded (20,000/cm2) for 24 hours, followed by antibiotics-free media replacement. A mixture of Lipofectamine 3000 (2.5 μL/mL; TFS) and poly(I:C) (2.5 μg/mL; InvivoGen) in OptiMEM (50 μL/mL; Gibco, TFS) was added to the medium. Cells were gently shaken and then incubated at 37°C. In case of inhibition of TLR3 stimulation, cells were pretreated for 1 hour at 37°C with bafilomycin A1 (100 nmol/L; InvivoGen), and then medium was replaced with fresh one and stimulation with poly(I:C) was performed as described above.

      Enzyme-Linked Immunosorbent Assay

      Levels of secreted cytokines and chemokines were quantified using commercial kits [Human IL-8 Uncoated ELISA (TFS); Human IL-6 ELISA DuoSet (R&D Systems, Bio-Techne, Minneapolis, MN); and Human IFN-β ELISA DuoSet (R&D Systems, Bio-Techne)] from serum-free culture supernatants on 8 hours (PC3) and 24 hours (DU145) stimulation or not with poly(I:C), following manufacturer's instructions. Levels of cytokines were determined directly in the supernatants, in technical duplicate.

      RNA Sequencing: Library Construction, Sequencing, and Data Processing

      Total RNA extracted from DU145 EV clones (numbers 1 and 2) and TLR3 OE (numbers 1 and 2) clones was sequenced using the Lexogen Quantseq FWD kit (Lexogen, Vienna, Austria). Samples were checked for PCR duplicates (6 bp UMI) in the cDNA library. Raw reads were quality checked (FastQC, MultiQC, minion, and swan), preprocessed (Trimmomatic, FastQC, and MultiQC), and mapped (STAR, Samtools, and MultiQC) to the reference genome with gene annotation (genome version: Ensembl GRCh38; gene annotation: Ensembl version 94). FASTQ files of raw data are deposited in Gene Expression Omnibus database under accession number GSE185953 (https://www.ncbi.nlm.nih.gov/geo, last accessed June 2022). Mapped reads were counted and summarized to genes (featureCounts). Differential gene expression was calculated by two separate tools: edgeR and DESeq2. A reduced list of genes differentially expressed between DU145 EV clones and DU145 TLR OE clones is shown in Supplemental Table S5 (log fold change >1.5 or <−1.5; adjusted P < 0.05) Data were analyzed by Ingenuity Pathway Analysis software for pathway enrichment and prediction of upstream regulators (z-score >2 or <−2). Normalized gene counts were analyzed via the Gene Set Enrichment Analysis tool to identify significant (false discovery rate q value < 0.05) enrichment of molecular signatures deposited in the Molecular Signature Database (systematic names: M27608, M15511, and M40389; Gene Ontology terms: 0016477, M11069, and M2572).
      Only pathways, upstream regulators, and molecular signatures passing the criteria of −log10(P value) > 1.3, P < 0.5, and false discovery rate q-value < 0.05, respectively, were considered as statistically significant (Supplemental Tables S6–S8).

      Electrophoresis and Immunoblot Analysis

      Cells were washed twice in phosphate-buffered saline (PBS) and harvested in radioimmunoprecipitation assay buffer (50 mmol/L Tris–HCl, pH 7.4, 1% Nonidet P-40, 0.25% sodium deoxycholate, and 150 mmol/L NaCl) supplemented with protease inhibitors (Serva, Heidelberg, Germany) and phosphatase inhibitor cocktail set II (Serva). The protein concentration was determined using a detergent-compatible protein assay (Bio-Rad, Hercules, CA). Cell lysates were mixed and boiled with 3× Laemmli buffer (240 mmol/L Tris–HCl, pH 6.8, 6% SDS, 0.02% bromophenol blue, 30% glycerol, and 3% β-mercaptoethanol). Equivalent protein quantities (30 μg) were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Millipore, Merck). The membranes were blocked at room temperature for 1 hour in tris-buffered saline (20 mmol/L Tris–HCl, pH 7.6, and 140 mmol/L NaCl) containing 0.1% Tween 20 and 5% nonfat milk. The membranes were incubated with specific primary antibodies overnight at 4°C, followed by incubation with secondary antibodies for 1 hour at room temperature. Detection of antibody reactivity was performed using Immobilon Western HRP Substrate (Millipore, Merck) and visualized on X-ray films (Agfa, Mortsel, Belgium). Densitometry quantification and analyses were performed using ImageJ software version 1.53c (NIH, Bethesda, MD: http://imagej.nih.gov/ij, last accessed June 26, 2020). The primary antibodies used were rabbit anti-TLR3 [number 6961; Cell Signaling Technologies (CST), Danvers, MA; 1:1000 in 5% nonfat milk], rabbit anti–phosphorylated S386-IRF3 [number 37829; CST; 1:1000 in 5% bovine serum albumin (BSA)], rabbit anti-IRF3 (number 11904; CST; 1:1000 in 5% BSA), rabbit anti–phosphorylated epidermal growth factor receptor (EGFR; Tyr1068; number 3777; CST; 1:1000 in 5% BSA), rabbit anti-EGFR (Santa Cruz Biotechnology; sc-03; 1:1000 in 5% nonfat milk), and mouse anti–α-tubulin (T9026; Sigma-Aldrich, Merck; 1:8000 in 5% nonfat milk). The secondary antibodies used were anti-rabbit IgG1 horseradish peroxidase (NA9340; Amersham, Merck; 1:3000 in 5% nonfat milk) and anti-mouse IgG1 horseradish peroxidase (NA931; Amersham, Merck; 1:4000 in 5% nonfat milk). For membrane staining, a solution of Amido Black 10B (Sigma-Aldrich, Merck; 0.1% in a mixture of acetic acid/methanol/H2O, 1:3:6) was used for 10 minutes. Staining excess was removed by washing with acetic acid/methanol/H2O solvent, then H2O.

      Cell Migration and Invasion

      Migration and invasion of the cells were performed using the xCELLigence system (OMNI Life Science GmbH, Bremen, Germany). Briefly, cells were starved overnight in a growth medium with 0.5% FBS, then detached by trypsin, subsequently neutralized by trypsin inhibitor, and seeded (30,000 cells/well) in the upper compartment of an electronically integrated Boyden chamber. The membrane surface was previously coated with fibronectin (20 ng/mL; 30 minutes at room temperature; Corning Inc., Corning, NY) for migration and with Cultrex BME GF-reduced (0.5 mg/mL; 4 hours at 37°C; Trevigen, Gaithersburg, MD) for invasion. HFF-1 fibroblast (ATCC; SCRC-1041) conditioned medium + 0.5% FBS was used in the lower chamber as a chemoattractant. Cell index was measured every 30 minutes as an indicator of cells that passed through the pore membrane.
      • Slabáková E.
      • Kharaishvili G.
      • Smějová M.
      • Pernicová Z.
      • Suchánková T.
      • Remšík J.
      • Lerch S.
      • Straková N.
      • Bouchal J.
      • Král M.
      • Culig Z.
      • Kozubík A.
      • Souček K.
      Opposite regulation of MDM2 and MDMX expression in acquisition of mesenchymal phenotype in benign and cancer cells.

      Transwell Migration Assay

      A total of 50,000 overnight-starved cells (0.5% FBS) were seeded in the upper compartment of fibronectin-coated (20 ng/mL) transwells with 8-μm pores (Falcon, BD Biosciences, Franklin Lakes, NJ). Fibroblast-conditioned medium containing 0.5% FBS was used in the lower compartment as a chemoattractant. After 24 hours, cells were fixed with 4% paraformaldehyde and stained with crystal violet (0.1% in water), and the remaining cells in the upper part of the transwell were removed with cotton swabs. Migrated cells on the lower part of the transwell were imaged with an Olympus IX70 microscope, 4× objective.

      Random Motility

      Cells were starved overnight in a growth medium containing 0.5% FBS, then detached and seeded on fibronectin-coated (20 ng/mL) μ-slide chambered coverslip (Ibidi, Gräfelfing, Germany) in a complete growth medium with 10% FBS, 5000 cells per well. Cells were let to adhere for 3 hours, then the medium was exchanged to remove unattached cells and live imaging was started on a confocal microscope (Olympus FV10i), with scanning every 5 minutes for 24 hours. Cells' motility was analyzed in ImageJ software by the use of the Manual Tracking plugin.

      Scratch Healing Assay

      Cells were seeded on an ImageLock 96-well plate (Incucyte; Sartorius AG, Göttingen, Germany) precoated with fibronectin (20 ng/mL). Cells were grown for 24 hours, then starved overnight in a medium with 0.5% FBS, and finally one scratch was performed in each well on fully confluent cells, using a WoundMaker tool (Incucyte; Sartorius AG). Detached cells were washed out, then complete growth medium containing mitomycin C (2.5 μg/mL for OE cells and 1 μg/mL for KO cells; Sigma-Aldrich, Merck) was added. Scratches and cells around the scratches were scanned in the bright field by the Incucyte imaging system (Sartorius) every hour. Wound closure at each time point was analyzed by Incucyte software.

      Orthotopic Implantation and in Vivo Imaging

      Two stable clones of DU145 cells with TLR3 overexpression and two empty vector control clones were transduced with pLenti PGK V5-luc Puro (w543-1; a gift from Eric Campeau and Paul Kaufman; Addgene plasmid number 19360
      • Campeau E.
      • Ruhl V.E.
      • Rodier F.
      • Smith C.L.
      • Rahmberg B.L.
      • Fuss J.O.
      • Campisi J.
      • Yaswen P.
      • Cooper P.K.
      • Kaufman P.D.
      A versatile viral system for expression and depletion of proteins in mammalian cells.
      ) using the third-generation lentiviral system and selected by 1 μg/mL puromycin for 1 week. Luciferase activity in vitro was validated with the Luciferase Assay System (Promega, Madison, WI; E1500). Immunodeficient 6- to 8-week–old male NOD.Cg-Rag1tm1Mom Il2rgtm1Wjl/SzJ (The Jackson Laboratory, Bar Harbor, ME) mice were shaved and anesthetized with ketamine/xylazine solution, applied intraperitoneally (100 μL/10 g of weight). A total of 10 μL of cell suspension of 30,000 cells containing 10% bromophenol blue for visualization was injected into the anterior prostate lobe in the left seminal vesicle. Mice were anesthetized using 2% to 3% of isoflurane in air administered through a nasal cone. Live imaging was performed by IVIS Lumina XR imaging system (Caliper Life Sciences, Waltham, MA) once a week after i.p. luciferin (Goldbio, St. Louis, MO) injection. Living Image software version 4.7.2 was used to generate images representing light intensity. All mice were sacrificed on day 35. Animal experiments were approved by the Academy of Sciences of the Czech Republic (AVCR 18/2018), supervised by the local ethical committee, and performed by the certified individuals (M.P., R.V., O.V., and K.S.).

      Docetaxel Dose-Response

      Cells were seeded in a flat white 96-well plate (Falcon; 20,000/cm2) and grown for 24 hours. A fresh medium containing different concentrations of docetaxel in the range from 0 to 333 nmol/L (Selleckchem, Houston, TX) was added, and cells were treated for 48 hours. At the end of the treatment, cell viability was evaluated by CellTiter Glo viability assay (Promega), according to the manufacturer's indications.

      Flow Cytometry Analysis

      A total of 1 × 106 cells of each cell line were first stained for viability for 20 minutes at 4°C in 100 μL of LIVE/DEAD Far Red (1:1000 in PBS; TFS), then washed with PBS. After viability staining, cells were fixed for 15 minutes at room temperature with 4% paraformaldehyde/PBS, then permeabilized for 15 minutes at room temperature with 0.25% Triton X-100/PBS. Next, cells were incubated with primary antibody mouse anti–cleaved poly (ADP-ribose) polymerase (1:400 in 1% BSA/PBS; CST) overnight at 4°C, then washed and incubated for 1 hour at room temperature with anti-mouse Alexa Fluor 488 (1:1000 in 1% BSA/PBS; TFS). Cells were then washed, resuspended in 1% BSA/PBS buffer, and analyzed by FACSVerse (BD Biosciences). Results were evaluated by the use of FlowJo software version 10.0.7 (TreeStar, BD Biosciences).

      Apoptosis Kinetics

      Apoptosis kinetics were performed by the use of a Cellevent Casp-3/7 Green flow kit (TFS). Briefly, cells were seeded in a 96-wells plate and stimulated with poly(I:C), as mentioned above. In addition, a caspase 3/7 probe (1:500) was added to the medium at the beginning of the stimulation. Scanning of the cells was immediately started and was performed every hour for 24 hours. The scanning was performed by Incucyte microscopy system (Sartorius), and the number of fluorescent cells at each hour was analyzed by Incucyte software.

      Liquid Chromatography–Tandem Mass Spectrometry Analysis of the Secretome

      Stable clones of DU145 cells transfected with empty vectors (EV clones 1 and 2) or with construct overexpressing TLR3 (TLR3-OE clones 1 and 2) were cultivated for 24 hours in serum-free conditions, and media were harvested under sterile conditions, centrifuged at 500 × g for 5 minutes to remove cellular debris, and stored in −80°C until further processed for protein analysis of secretome performed by liquid chromatography–tandem mass spectrometry in Proteomics Core Facility CEITEC (Brno, Czech Republic). The lysates (approximately 100 μg of total protein) were used for filter-aided sample preparation (10-kDa cutoff cartridges), as described elsewhere,
      • Wisniewski J.R.
      • Zougman A.
      • Nagaraj N.
      • Mann M.
      Universal sample preparation method for proteome analysis.
      using 1 μg of trypsin (sequencing grade; Promega). The resulting peptides were analyzed by nanoElute system (Bruker) connected to timsTOF Pro spectrometer (Bruker) using 60 minutes long gradient and diaPASEF acquisition mode. DiaPASEF data were processed in DIA-NN version 1.8
      • Demichev V.
      • Messner C.B.
      • Vernardis S.I.
      • Lilley K.S.
      • Ralser M.
      DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput.
      in library-free mode against the modified cRAP database (based on http://www.thegpm.org/crap, last accessed November 22, 2018) and UniProtKB protein database for Homo sapiens (https://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/reference_proteomes/Eukaryota/UP000005640/UP000005640_9606.fasta.gz, last accessed June 2021, number of protein sequences: 20,600). Protein MaxLFQ intensities reported in the DIA-NN output file (Supplemental Table S9) were filtered on the basis of log fold change (>1 in TLR3-OE clones 1 and 2) and adjusted P values (<0.075) to obtain the final list of protein hits enriched in media collected from DU145 clones overexpressing TLR3 when compared with their control counterparts (DU145 EV clones) (Supplemental Table S10).

      Statistical Analysis, Software, and Repeatability of Experiments

      Statistical analyses were performed in GraphPad Prism version 9 (GraphPad Software, San Diego, CA) by two-tailed t-test and one- or two-way analysis of variance, as indicated within figure legends. The number of independent biological replicates is three, if not otherwise indicated in figure legends. The results are shown as means ± SD if not otherwise indicated.

      Results

      TLR3 Expression Is Deregulated in Castration-Resistant PCa and Therapy-Resistant PCa

      Analysis of different studies from cohorts of patients affected by hormone-naïve or hormone-refractory metastatic and nonmetastatic PCa underlined contrasting gene expression of TLR3 in all these conditions (Supplemental Figure S1A
      • Best C.J.
      • Gillespie J.W.
      • Yi Y.
      • Chandramouli G.V.
      • Perlmutter M.A.
      • Gathright Y.
      • Erickson H.S.
      • Georgevich L.
      • Tangrea M.A.
      • Duray P.H.
      • Gonzalez S.
      • Velasco A.
      • Linehan W.M.
      • Matusik R.J.
      • Price D.K.
      • Figg W.D.
      • Emmert-Buck M.R.
      • Chuaqui R.F.
      Molecular alterations in primary prostate cancer after androgen ablation therapy.
      • Varambally S.
      • Yu J.
      • Laxman B.
      • Rhodes D.R.
      • Mehra R.
      • Tomlins S.A.
      • Shah R.B.
      • Chandran U.
      • Monzon F.A.
      • Becich M.J.
      • Wei J.T.
      • Pienta K.J.
      • Ghosh D.
      • Rubin M.A.
      • Chinnaiyan A.M.
      Integrative genomic and proteomic analysis of prostate cancer reveals signatures of metastatic progression.
      • Tomlins S.A.
      • Mehra R.
      • Rhodes D.R.
      • Cao X.
      • Wang L.
      • Dhanasekaran S.M.
      • Kalyana-Sundaram S.
      • Wei J.T.
      • Rubin M.A.
      • Pienta K.J.
      • Shah R.B.
      • Chinnaiyan A.M.
      Integrative molecular concept modeling of prostate cancer progression.
      • Tamura K.
      • Furihata M.
      • Tsunoda T.
      • Ashida S.
      • Takata R.
      • Obara W.
      • Yoshioka H.
      • Daigo Y.
      • Nasu Y.
      • Kumon H.
      • Konaka H.
      • Namiki M.
      • Tozawa K.
      • Kohri K.
      • Tanji N.
      • Yokoyama M.
      • Shimazui T.
      • Akaza H.
      • Mizutani Y.
      • Miki T.
      • Fujioka T.
      • Shuin T.
      • Nakamura Y.
      • Nakagawa H.
      Molecular features of hormone-refractory prostate cancer cells by genome-wide gene expression profiles.
      • Goswami C.P.
      • Nakshatri H.
      PROGgeneV2: enhancements on the existing database.
      ), pointing to an insufficiently understood correlation between TLR3 and PCa progression. Survival analysis in two data sets (Gene Expression Omnibus database, https://www.ncbi.nlm.nih.gov/geo, last accessed June 2022; accession numbers GSE70768 and GSE70769
      • Ross-Adams H.
      • Lamb A.D.
      • Dunning M.J.
      • Halim S.
      • Lindberg J.
      • Massie C.M.
      • Egevad L.A.
      • Russell R.
      • Ramos-Montoya A.
      • Vowler S.L.
      • Sharma N.L.
      • Kay J.
      • Whitaker H.
      • Clark J.
      • Hurst R.
      • Gnanapragasam V.J.
      • Shah N.C.
      • Warren A.Y.
      • Cooper C.S.
      • Lynch A.G.
      • Stark R.
      • Mills I.G.
      • Gronberg H.
      • Neal D.E.
      • CamCa P.S.G.
      Integration of copy number and transcriptomics provides risk stratification in prostate cancer: a discovery and validation cohort study.
      ) showed a trend between high TLR3 and poor relapse-free survival in patients affected by PCa (Supplemental Figure S1B). It is well known that PCa progression and metastatic occurrence are linked to therapy-resistance onset, and, to better understand the correlation of these events with TLRs, we screened gene expression of all human TLRs, together with their downstream effectors (Supplemental Table S2), in therapy-naïve and therapy-resistant PCa cell models (Supplemental Table S1). Hierarchical clustering showed that, in most of the models, the differences in expression of TLR pathway markers did not result in a separation between sensitive and resistant PCa cell lines. However, the TLR3 gene appeared to be up-regulated in most resistant cells (Supplemental Figure S2A). We next evaluated whether TLR3 gene up-regulation was translated also at a protein level, and we observed higher TLR3 protein expression in drug-resistant clones of three cell models, in particular the full-length isoform at 130 kDa (Figure 1A). The mRNA levels of the receptor observed in the gene screen were confirmed by the use of a different set of primers in LAPC4 and PC3 models, although not in DU145 doc cells (Supplemental Figure S2B).
      Figure thumbnail gr1
      Figure 1Toll-like receptor 3 (TLR3) high expression is associated with resistance in prostate cancer (PCa). A: TLR3 expression was evaluated in sensitive and drug-resistant PCa cell lines by Western blot analysis; TLR3 isoforms are signed by arrows; α-tubulin was used as loading control. B and C: Immunohistochemical analysis with the antibody 40F9.6. Validation of the antibody was performed by staining of formalin-fixed, paraffin-embedded cells with low expression (DU145 TLR3 empty vector 2; B) and with TLR3 overexpression (OE; DU145 TLR3 OE 2; see also ; C). D: Prominent positivity was also observed in the epithelium of the urethra. EG: Representative examples of low and medium TLR3 expression of prostate cancer (E and F, respectively) as well as heterogeneous staining of benign prostatic hyperplasia (G) are provided. HJ: Castration-resistant prostate cancer tissues obtained by TURP. I: Note aberrant mitoses. H-score of all samples is shown in and (see for details). BJ: Higher-magnification images of the boxed areas (dashed lines) are shown (solid lines). Scale bars = 50 μm (BJ). CRPC, castration-resistant PCa; Ctrl, therapy responsive; Doc, docetaxel resistant; Enza, enzalutamide resistant.
      Given the variability we observed in TLR3 expression levels in patient cohorts (Supplemental Figure S1A) and PCa models of therapy resistance (Supplemental Figure S2A), we aimed to immunohistochemically evaluate the protein expression in patients affected by hormone-naïve PCa and after endocrine therapy (Supplemental Tables S3 and S4). We tested three commercial antibodies (clone 40C1285.6 from TFS, clone TLR3.7 from Santa Cruz Biotechnology, and polyclonal antibody from Promokine), which had been used in previous studies (reviewed by Muresan et al
      • Muresan X.M.
      • Bouchal J.
      • Culig Z.
      • Soucek K.
      Toll-like receptor 3 in solid cancer and therapy resistance.
      ); however, they did not pass our validation using cells with known TLR3 expression (Supplemental Figure S3). Finally, Innate Pharma provided us with their licensed antibody (clone 40F9.6).
      • Salaun B.
      • Zitvogel L.
      • Asselin-Paturel C.
      • Morel Y.
      • Chemin K.
      • Dubois C.
      • Massacrier C.
      • Conforti R.
      • Chenard M.P.
      • Sabourin J.C.
      • Goubar A.
      • Lebecque S.
      • Pierres M.
      • Rimoldi D.
      • Romero P.
      • Andre F.
      TLR3 as a biomarker for the therapeutic efficacy of double-stranded RNA in breast cancer.
      We optimized the TLR3 staining protocol in control cells, in TLR3-overexpressing cancer cells, and in the urethra (Figure 1, B–D) and examined the receptor's expression in formalin-fixed, paraffin-embedded patient samples (Figure 1, E–J). Despite the heterogeneity between samples, we observed a trend toward higher H-scores in TURP samples of advanced cancer, including castration-resistant PCa, in comparison to primary cancer obtained by prostatectomy (Supplemental Tables S3 and S4). This pilot immunohistochemical study as well as other pieces of evidence provided above clearly show the importance of TLR3 up-regulation in PCa progression, at least in a subset of patients.

      TLR3 in Therapy-Resistant PCa Cells Is Functional

      TLR3 is an endosomal protein known to exert its function on activation binding of dsRNA ligands. Thus, we assessed whether the up-regulation of TLR3 expression we found in chemoresistant cells corresponded to a higher ability of the cells to activate the receptor. We first verified that baseline levels of TLR3 observed in drug-resistant cells did not lead to intrinsic activation of TLR3 receptor in terms of IRF3 phosphorylation (Supplemental Figure S4A). Next, we stimulated DU145 and PC3 cell models with TLR3 ligand, poly(I:C), and evaluated TLR3 activation kinetics as IRF3 phosphorylation. The stimulation showed the trend of higher phosphorylated IRF3 levels in docetaxel-resistant cells than in control, in particular at 3 and 6 hours (Figure 2A and Supplemental Figure S4B). Given that poly(I:C) has the capacity to induce IRF3 phosphorylation and cytokine production by binding not only endosomal TLR3 but also cytoplasmic MDA5 and RIG-I,
      • Dauletbaev N.
      • Cammisano M.
      • Herscovitch K.
      • Lands L.C.
      Stimulation of the RIG-I/MAVS pathway by polyinosinic:polycytidylic acid upregulates IFN-beta in airway epithelial cells with minimal costimulation of IL-8.
      we verified whether the increase we observed was dependent on TLR3 binding. For this purpose, we pretreated the cells with bafilomycin A1, a commonly used inhibitor of TLR3 endosomal activation, then stimulated them with poly(I:C) for several time points. DU145 sensitive and resistant cells presented lower phosphorylated IRF3 levels in the case of pretreatment with bafilomycin A1, whereas PC3 cells maintained relatively high levels of TLR3 activation, regardless of the use of the inhibitor (Figure 2A and Supplemental Figure S4B). In parallel, we stimulated the cells with poly(I:C) and examined the production and secretion of interferon (IFN)-β and cytokines IL-6 and IL-8. We observed a strong secretion of IL-6 and IL-8 in PC3, with higher levels of IL-6 in chemoresistant cells after 8 hours of stimulation. On the contrary, IFN-β secretion was increased in PC3 sensitive control compared with chemoresistant cells (Figure 2B). In DU145 cells, poly(I:C) stimulation led to lower and later (at 24 hours) cytokine production than in PC3; contrary to the PC3 model, IL-6 was secreted more in DU145 control cells than chemoresistant line; differently, IL-8 and IFN-β were produced at higher levels in docetaxel-resistant cells (Figure 2B). Interestingly, we observed higher IL-8 basal levels in both DU145 and PC3 chemoresistant lines compared with sensitive control, in absence of dsRNA exogenous treatment (Figure 2B).
      Figure thumbnail gr2
      Figure 2Toll-like receptor 3 (TLR3) activation leads to the secretion of cytokines, and its expression is associated with chemoresistance. A: TLR3 activation on stimulation with poly(I:C) for several time points was evaluated by Western blot analysis as levels of IRF3 phosphorylation; total IRF3 was used as loading control; representative of three biological repetitions. B: Basal and poly(I:C)-induced secretion levels of chemokines and interferon (IFN)-β are shown as pg/mL normalized to cell number; results are presented as a box with minimum to maximum whiskers. C: Cell viability on docetaxel treatment is shown as percentage of relative luminescence units (RLUs) of each dose in comparison to dimethyl sulfoxide (DMSO)–treated control (100% viable cells); DU145 Doc cell line was used as a positive control for docetaxel resistance. ∗P < 0.05 by one-way analysis of variance. Ctrl, control; DL, detection limit; EV, empty vector control; OE, TLR3 overexpressing.

      High TLR3 Sustains Resistance, Migration, and Invasion of PCa Cells

      Because we found TLR3 protein up-regulated in several therapy-resistant PCa models, we verified whether this receptor's levels are directly associated with resistance to docetaxel. We also investigated how TLR3 influences biological mechanisms related to PCa progression and therapy resistance, such as migration and invasion. For this purpose, we generated DU145 therapy-responsive models with ectopic TLR3 OE (Supplemental Figure S5A). We also generated models in which we down-regulated TLR3 levels in TLR3 OE clones by specific shRNA (Supplemental Figure S5C) and restored basal TLR3 expression. In parallel, we knocked out the receptor in parental DU145 cells (Supplemental Figure S5B).
      For the analysis of the TLR3 effect on chemoresistance, we treated control clones transfected with EV and OE clones with different doses of docetaxel and examined the cell viability after 48 hours. We observed that TLR3 OE clones were less susceptible to docetaxel treatment compared with EV clones (Figure 2C), suggesting the involvement of TLR3 in docetaxel resistance.
      Next, we evaluated the role of TLR3 in migration and invasion. The first approach was to observe the scratch closure ability, and results showed that TLR3 OE cells had a higher ability of scratch healing in 24 hours compared with EV (Figure 3A). The effect was the opposite in TLR3 KO clones, which closed the scratch with lower efficiency compared with corresponding EV cells (Supplemental Figure S6B). In parallel, we investigated cell motility in two dimensions and observed that TLR3 OE cells move on longer trajectories compared with EV clones (Figure 3B), whereas TLR3 KO cells travel shorter distances compared with EV controls (Supplemental Figure S6D). TLR3-positive effect on migration was confirmed also by other assays, such as the transwell assay, which showed migration of a higher number of TLR3 OE cells compared with EV clones (Supplemental Figure S6A). A similar pattern was observed also using the real-time cell analysis (xCELLigence), by which we tracked cell migration up to 24 hours. Also, in this case, TLR3 OE cells presented a greater ability to migrate (Figure 3C). Given the fact that the selection of single-cell–derived clones led us to DU145 TLR3 OE cells with a slightly bigger diameter compared with DU145 TLR3 EV cells, we wanted to exclude any potential clonal-dependent bias. Therefore, we down-regulated the receptor's expression from OE clones and, in this case, we obtained control-shRNA and TLR3-shRNA clones, which differed in TLR3 expression, but not in cell diameter. We performed a migration assay by xCELLigence system on these cells and observed that TLR3-shRNA cells had significantly lower migratory ability compared with control-shRNA cells (Figure 3D and Supplemental Figure S6C), confirming our previous observations of the effect of TLR3 levels on cell migration. In parallel, we generated also LAPC4 cells with ectopic TLR3 overexpression on which we investigated cell migration and motility. No conclusive results were obtained because LAPC4 cells did not present any migratory ability (data not shown). However, this could underline the need for intrinsic migratory capacity of the cells for TLR3-dependent motility increase.
      Figure thumbnail gr3
      Figure 3Toll-like receptor 3 (TLR3) expression favors cell migration and invasion. A: Right panel: Time-lapse scratch healing was plotted as mean percentage of relative wound density; performed by two-way analysis of variance. Left panels: Representative images of scratch closure at 24 hours. B: Left panel: Cell motility is shown as the distance traveled by cells on a two-dimensional surface; performed by one-way analysis of variance. Right panels: Representative cell trajectories for each clone are shown. CE: Cell migration and invasion are shown as cell index measured by xCELLigence system; performed by two-way analysis of variance. Results presented as means ± SEM (A and CE). ∗P < 0.05. Scale bar = 400 μm (A). Ctrl, control; EV, empty vector control; OE, TLR3 overexpressing.
      Next, we assessed whether TLR3 also influences cell invasion. Using the xCELLigence system, we tracked invasion in EV and TLR3 OE cells up to 33 hours, and we observed that TLR3 OE cells had an increased invasive ability than EV control cells (Figure 3E). The implication of canonical TLR3 signaling in DU145 motility and chemoresistance was assessed in DU145 cells expressing TLR3-DN (ΔTIR): dominant-negative TIR-less TLR3 gene.
      • Galli R.
      • Paone A.
      • Fabbri M.
      • Zanesi N.
      • Calore F.
      • Cascione L.
      • Acunzo M.
      • Stoppacciaro A.
      • Tubaro A.
      • Lovat F.
      • Gasparini P.
      • Fadda P.
      • Alder H.
      • Volinia S.
      • Filippini A.
      • Ziparo E.
      • Riccioli A.
      • Croce C.M.
      Toll-like receptor 3 (TLR3) activation induces microRNA-dependent reexpression of functional RARβ and tumor regression.
      ,
      • Galli R.
      • Starace D.
      • Busà R.
      • Angelini D.F.
      • Paone A.
      • De Cesaris P.
      • Filippini A.
      • Sette C.
      • Battistini L.
      • Ziparo E.
      • Riccioli A.
      TLR stimulation of prostate tumor cells induces chemokine-mediated recruitment of specific immune cell types.
      Contrarily to the overexpression of wild-type TLR3, overexpression of truncated TLR3 failed to increase cell motility or resistance to docetaxel (Supplemental Figure S7, A–D).
      To exclude that the cellular response was exclusive to in vitro conditions, we ectopically introduced luciferase in DU145 TLR3 EV and OE cells and injected the cells orthotopically into the prostate of immunodeficient mice. Tumor formation and dissemination were monitored over 4 weeks using in vivo imaging. As illustrated in Figure 4, there was a significantly higher incidence of intra-abdominal metastasis in mice injected with TLR3 OE cells compared with mice carrying EV tumors.
      Figure thumbnail gr4
      Figure 4Toll-like receptor 3 (TLR3) overexpression is linked to increased metastatic potential. Noninvasive monitoring of orthotopic prostate xenograft in NOD.Cg-Rag1tm1Mom Il2rgtm1Wjl/SzJ mice. The images show a luciferase signal in each mouse 4 weeks after injection of the bioluminescent cells; stars indicate the mice with metastasis; mice are divided into two groups [empty vector control (EV) and TLR3 overexpression (OE)], each containing two clones. The table shows counts of metastasis frequency and proportion; the statistical difference between EV and OE groups was calculated by the N-1 χ2 test. n = 11 mice per group.
      To better understand potential mechanisms underlying TLR3-dependent migratory phenotype, we performed RNA-sequencing analysis on DU145 TLR3 EV and OE cells. Gene enrichment analysis confirmed the positive correlation between TLR3 overexpression and genes belonging to several migration-related data sets (Figure 5, A and C). Moreover, Ingenuity Pathway Analysis revealed in TLR3 OE cells the up-regulation of pathways, such as IL-8 signaling, bone morphogenetic protein signaling, ephrin receptor signaling, and androgen signaling. In parallel, TLR3 overexpression corresponded to an increase of upstream regulators involved in the inflammatory cascade (Figure 5B). Taken together, our results demonstrate a clear correspondence between high TLR3 expression and cell migration ability.
      Figure thumbnail gr5
      Figure 5High toll-like receptor 3 (TLR3) expression is associated with migratory and inflammatory pathways. A: The volcano plot represents genes with down-regulated, up-regulated, or unchanged expression in TLR3 overexpression (OE) cells. The dashed lines show the set threshold values: log fold change >1.5 or <−1.5, adjusted P < 0.05. B: Ingenuity Pathway Analysis (IPA) of the top five differentially regulated pathways (left panel) and of the top five upstream regulators (right panel) predicted as activated (red) or inhibited (green) in TLR3 OE cells. C: Gene Set Enrichment Analysis was performed to compare online migration data sets and data obtained by RNA sequencing from DU145 empty vector (EV) control and TLR3 OE clones. BMP, bone morphogenetic protein; CTCF, CCCTC-binding factor; EGFR, epidermal growth factor receptor; PDGF, platelet-derived growth factor; PI3K, phosphatidylinositol 3-kinase.
      To propose functional candidates controlling motility and migration of TLR3-overexpressing cells, liquid chromatography–tandem mass spectrometry–based proteomic analysis of media harvested from control DU145 cells (EV clones 1 and 2) and DU145 cells overexpressing TLR3 (TLR3-OE clones 1 and 2) was performed. The results indicated high baseline similarity between the tested groups (Supplemental Table S9), yet several statistically distinct protein hits were revealed and are summarized in Supplemental Table S10. Interestingly, the analysis did not reveal a stimulatory effect of ectopically expressed TLR3 on the content of secreted IL-8, IL-6, or IFN-β (Supplemental Table S9).
      Because our results have shown potentiated migration and motility in TLR3-OE DU145 clones, we wanted to suggest an underlying mechanism. Functional analysis of transcriptome (Figure 5B and Supplemental Tables S6–S8) identified, among others, epidermal growth factor and its receptor (EGFR) as one of the most deregulated upstream regulators in TLR3-OE clones. EGFR is a known potent driver of cell migration and motility in prostate cancer34, thus, RNA-sequencing data revealed the possible mechanism of a negative feedback loop recruited by TLR3-OE clones to tune their increased migratory potential. In DU145 cells, TLR3 protein expression correlated with the level of EGFR phosphorylation (Supplemental Figure S7E), and experiments with EGFR inhibitor lapatinib confirmed the role of EGFR signaling in the regulation of migration in both EV and TLR3-OE DU145 clones (Supplemental Figure S7F).

      Exogenous Activation of TLR3 Leads to Apoptosis of Chemoresistant PCa Cells

      Interestingly, TLR3-dependent migratory phenotype occurred in absence of any additional stimulation of the receptor with exogenous poly(I:C). At the same time, previous studies showed that TLR3 activation by dsRNA leads to cell death in PCa. Given that TLR3 is up-regulated and functional in chemoresistant PCa cells, we wanted to reveal the fate of the cells in case of activation by poly(I:C). We treated sensitive and docetaxel-resistant DU145 and PC3 cells with different endogenous levels of TLR3 expression with poly(I:C), and we observed an induction of cell apoptosis in both cell models. There was a higher percentage of cleaved poly (ADP-ribose) polymerase positivity in docetaxel-resistant cells compared with sensitive control cells (Figure 6A) in all cases, even with relatively high variability between biological repetitions. We also stimulated DU145 EV and TLR3 OE cells with poly(I:C) and, as expected, we detected significantly higher cleaved poly (ADP-ribose) polymerase positivity in TLR3 OE cells compared with EV control (Figure 6B). Moreover, we monitored the kinetics of poly(I:C)-induced apoptosis in a time lapse of 24 hours, and we observed a higher number of cells positive for caspase 3/7 in TLR3 OE condition (Figure 6C), whereas in parallel, poly(I:C) stimulation induced lower caspase 3/7 positivity in TLR3 KO cells compared with empty vector control cells in 24 hours (Supplemental Figure S8). These data highlight a direct correlation between TLR3 levels and poly(I:C)-induced apoptotic response.
      Figure thumbnail gr6
      Figure 6Toll-like receptor 3 (TLR3) activation promotes prostate cancer cell apoptosis. A: DU145 cells were stimulated or not with poly(I:C) for 24 hours. A and B: PC3 (A) and DU145 empty vector (EV) control and TLR3 overexpression (OE) clones (B) were stimulated for 8 hours; the percentage of cleaved poly (ADP-ribose) polymerase (cPARP) on stimulation was evaluated by flow cytometry on single, compact cells; performed by one-way analysis of variance. C: Right panel: Apoptosis kinetics on poly(I:C) stimulation are presented in the plot as numbers of cells positive for caspase 3/7 (Casp3/7); performed by two-way analysis of variance. Left panels: Representative of cells positive for Casp3/7 (green fluorescence) at 24 hours after poly(I:C).
      • Muresan X.M.
      • Bouchal J.
      • Culig Z.
      • Soucek K.
      Toll-like receptor 3 in solid cancer and therapy resistance.
      TLR3 overexpression induces invasion of prostate cancer cells, whereas its activation triggers apoptosis. ∗P < 0.05. Ctrl, control; Doc, docetaxel resistant.

      Discussion

      TLR3 is a receptor belonging to the TLR family, whose expression was assessed lately in multiple tumor types, including PCa.
      • Muresan X.M.
      • Bouchal J.
      • Culig Z.
      • Soucek K.
      Toll-like receptor 3 in solid cancer and therapy resistance.
      Tumor progression and loss of response to existing therapies continue to be frequent medical challenges in patients affected by PCa. Through analysis of public data sets, we showed that high TLR3 expression in PCa is associated with poor relapse-free survival. In line with these data, a previous investigation demonstrated that high TLR3 levels in PCa positively correlated with biochemical recurrence.
      • Gonzalez-Reyes S.
      • Fernandez J.M.
      • Gonzalez L.O.
      • Aguirre A.
      • Suarez A.
      • Gonzalez J.M.
      • Escaff S.
      • Vizoso F.J.
      Study of TLR3, TLR4, and TLR9 in prostate carcinomas and their association with biochemical recurrence.
      In our study, we revealed an up-regulation of TLR3 in docetaxel- and enzalutamide-resistant PCa cell models compared with drug-responsive cells. Such up-regulation was observed at both protein and mRNA levels, the later with exception of DU145 cells. TLR3 is under a complex network of regulations that includes protein trafficking, cathepsin-mediated proteolytic processing, and the contribution from distinct endosomes.
      • Paone A.
      • Starace D.
      • Galli R.
      • Padula F.
      • De Cesaris P.
      • Filippini A.
      • Ziparo E.
      • Riccioli A.
      Toll-like receptor 3 triggers apoptosis of human prostate cancer cells through a PKC-alpha-dependent mechanism.
      ,
      • Qi R.
      • Singh D.
      • Kao C.C.
      Proteolytic processing regulates toll-like receptor 3 stability and endosomal localization.
      In parallel, we demonstrated through exogenous stimulation with a specific dsRNA ligand that TLR3 in docetaxel-resistant cells is functional; indeed, higher TLR3 levels corresponded to higher IRF3 phosphorylation, and this effect was diminished by inhibitor bafilomycin A1. PC3 cells appeared less responsive to TLR3 inhibition compared with DU145 cells. This might be due to the differences in TLR3 isoform expression, which were shown to influence the response to poly(I:C).
      • Toscano F.
      • Estornes Y.
      • Virard F.
      • Garcia-Cattaneo A.
      • Pierrot A.
      • Vanbervliet B.
      • Bonnin M.
      • Ciancanelli M.J.
      • Zhang S.Y.
      • Funami K.
      • Seya T.
      • Matsumoto M.
      • Pin J.J.
      • Casanova J.L.
      • Renno T.
      • Lebecque S.
      Cleaved/associated TLR3 represents the primary form of the signaling receptor.
      In parallel, TLR3 functionality in docetaxel-resistant cells was confirmed also by increased secretion of IFN-β, IL-6, and IL-8, which are known to be downstream products of the TLR3 activation pathway.
      • Gao D.
      • Ciancanelli M.J.
      • Zhang P.
      • Harschnitz O.
      • Bondet V.
      • Hasek M.
      • Chen J.
      • Mu X.
      • Itan Y.
      • Cobat A.
      • Sancho-Shimizu V.
      • Bigio B.
      • Lorenzo L.
      • Ciceri G.
      • McAlpine J.
      • Anguiano E.
      • Jouanguy E.
      • Chaussabel D.
      • Meyts I.
      • Diamond M.S.
      • Abel L.
      • Hur S.
      • Smith G.A.
      • Notarangelo L.
      • Duffy D.
      • Studer L.
      • Casanova J.L.
      • Zhang S.Y.
      TLR3 controls constitutive IFN-beta antiviral immunity in human fibroblasts and cortical neurons.
      ,
      • Lundberg A.M.
      • Drexler S.K.
      • Monaco C.
      • Williams L.M.
      • Sacre S.M.
      • Feldmann M.
      • Foxwell B.M.
      Key differences in TLR3/poly I:C signaling and cytokine induction by human primary cells: a phenomenon absent from murine cell systems.
      Cytokine secretion on poly(I:C) was faster in PC3 cells, in line with previously reported strong ability of PC3 cell line to produce proinflammatory cytokines.
      • Kourko O.
      • Smyth R.
      • Cino D.
      • Seaver K.
      • Petes C.
      • Eo S.Y.
      • Basta S.
      • Gee K.
      Poly(I:C)-mediated death of human prostate cancer cell lines is induced by interleukin-27 treatment.
      Interestingly, we observed increased basal levels of secreted IL-8 in both DU145 and PC3 chemoresistant lines compared with sensitive control, in absence of dsRNA exogenous treatment. Although IL-8 is a downstream marker of the TLR3 pathway, we detected no basal activation of this receptor in our chemoresistant PCa models, indicating that most likely TLR3 up-regulation only is not responsible for IL-8 basal levels; up to date, no study was published on IL-8 and TLR3 in absence of activation. However, it was previously shown that IL-8 sustains androgen-independent PCa
      • Araki S.
      • Omori Y.
      • Lyn D.
      • Singh R.K.
      • Meinbach D.M.
      • Sandman Y.
      • Lokeshwar V.B.
      • Lokeshwar B.L.
      Interleukin-8 is a molecular determinant of androgen independence and progression in prostate cancer.
      , therefore, increased basal IL-8 could be associated with the chemoresistance in our cell models. In DU145 cells, the positive effect of TLR3 overexpression on metastasis frequency is not reflected in increased IL-6 and IL-8 mRNA levels (Supplemental Table S5) or their secretion (Supplemental Tables S9 and S10); thus, we hypothesize that these particular cytokines are not ultimate drivers of increased migratory potential observed in prostate cells with higher expression levels of TLR3. By overexpression of TLR3 in PCa cells, we confirmed that this receptor can partially modulate chemoresistance, given that overexpressing cells were less responsive to docetaxel treatment up to higher doses; yet, TLR3 up-regulation cannot be considered a leading mechanism. Up to date, only one other study showed a relation between TLR3 activation, but not expression, and resistance to cisplatin, although in a different cancer type.
      • Chuang H.C.
      • Chou M.H.
      • Chien C.Y.
      • Chuang J.H.
      • Liu Y.L.
      Triggering TLR3 pathway promotes tumor growth and cisplatin resistance in head and neck cancer cells.
      In our study, we showed that TLR3 influences cell migration. A previous investigation demonstrated TLR3’s ability to promote cell migration in response to specific ligand binding, through activation of a so-called non-transcriptional branch. Indeed, authors showed that TLR3 stimulation in endosomes by dsRNA leads to the association of the receptor with EGFR and Src, inducing Src and FAK phosphorylation.
      • Chattopadhyay S.
      • Sen G.C.
      Tyrosine phosphorylation in toll-like receptor signaling.
      ,
      • Yamashita M.
      • Chattopadhyay S.
      • Fensterl V.
      • Saikia P.
      • Wetzel J.L.
      • Sen G.C.
      Epidermal growth factor receptor is essential for toll-like receptor 3 signaling.
      However, we demonstrated that TLR3 expression alone can affect migration. Indeed, we observed that TLR3 OE PCa cells presented higher cell motility, migration, and invasion capacities without any stimulation with external specific ligands, and we revealed that the presence of TLR3 TIR domain and EGFR signaling activity underlie this effect. In line with our hypothesis, TLR3 KO cells had lower motility. In parallel, a decrease of TLR3 in OE clones by shRNA also decreased cells' migratory phenotype together with activity of EGFR. TLR3-dependent invasive capacity was demonstrated also in vivo, given that tumors formed by TLR3 OE cells led to a higher incidence of intra-abdominal metastasis. By RNA-sequencing analysis, we demonstrated that TLR3 overexpression in PCa cells leads to up-regulation of several pathways, such as androgen signaling, bone morphogenetic protein signaling pathway, ephrin receptor signaling, or IL-8 signaling, which have all been previously associated with PCa progression and metastasis.
      • Araki S.
      • Omori Y.
      • Lyn D.
      • Singh R.K.
      • Meinbach D.M.
      • Sandman Y.
      • Lokeshwar V.B.
      • Lokeshwar B.L.
      Interleukin-8 is a molecular determinant of androgen independence and progression in prostate cancer.
      ,
      • Astin J.W.
      • Batson J.
      • Kadir S.
      • Charlet J.
      • Persad R.A.
      • Gillatt D.
      • Oxley J.D.
      • Nobes C.D.
      Competition amongst Eph receptors regulates contact inhibition of locomotion and invasiveness in prostate cancer cells.
      • Dai C.
      • Heemers H.
      • Sharifi N.
      Androgen signaling in prostate cancer.
      • Darby S.
      • Cross S.S.
      • Brown N.J.
      • Hamdy F.C.
      • Robson C.N.
      BMP-6 over-expression in prostate cancer is associated with increased Id-1 protein and a more invasive phenotype.
      More important, up-regulation of epidermal growth factor gene expression is part of the IL-8 and ephrin receptor signaling pathways that we identified, as well as of the Ingenuity Pathway Analysis upstream regulators EGFR and phosphatidylinositol 3-kinase in TLR3 OE cells. TLR3 overexpression led to the activation of upstream regulators, such as the CCCTC-binding factor, which have been linked to poor prognosis in PCa
      • Hoflmayer D.
      • Steinhoff A.
      • Hube-Magg C.
      • Kluth M.
      • Simon R.
      • Burandt E.
      • Tsourlakis M.C.
      • Minner S.
      • Sauter G.
      • Buscheck F.
      • Wilczak W.
      • Steurer S.
      • Huland H.
      • Graefen M.
      • Haese A.
      • Heinzer H.
      • Schlomm T.
      • Jacobsen F.
      • Hinsch A.
      • Poos A.M.
      • Oswald M.
      • Rippe K.
      • Konig R.
      • Schroeder C.
      Expression of CCCTC-binding factor (CTCF) is linked to poor prognosis in prostate cancer.
      or are involved in inflammation.
      • Stik G.
      • Vidal E.
      • Barrero M.
      • Cuartero S.
      • Vila-Casadesus M.
      • Mendieta-Esteban J.
      • Tian T.V.
      • Choi J.
      • Berenguer C.
      • Abad A.
      • Borsari B.
      • le Dily F.
      • Cramer P.
      • Marti-Renom M.A.
      • Stadhouders R.
      • Graf T.
      CTCF is dispensable for immune cell transdifferentiation but facilitates an acute inflammatory response.
      Interestingly, previous studies showed that TLR3-induced inflammatory response can affect invasiveness in metastatic intestinal cells
      • Bugge M.
      • Bergstrom B.
      • Eide O.K.
      • Solli H.
      • Kjonstad I.F.
      • Stenvik J.
      • Espevik T.
      • Nilsen N.J.
      Surface toll-like receptor 3 expression in metastatic intestinal epithelial cells induces inflammatory cytokine production and promotes invasiveness.
      and even promote premetastatic niche in lung tissue.
      • Liu Y.
      • Gu Y.
      • Han Y.
      • Zhang Q.
      • Jiang Z.
      • Zhang X.
      • Huang B.
      • Xu X.
      • Zheng J.
      • Cao X.
      Tumor exosomal RNAs promote lung pre-metastatic niche formation by activating alveolar epithelial TLR3 to recruit neutrophils.
      In conclusion, the experiments performed demonstrated the importance of EGFR signaling in the phenotype of increased migration in TLR3 OE cells and suggested the role of ERBB2. Enriched processes of phosphatidylinositol 3-kinase signaling, integrin binding, and positive regulation of protein tyrosine kinase activity strongly support increased migratory potential in prostate cancer cells overexpressing TLR3. Therefore, potentiated EGFR and ERBB2 signaling may represent possible mechanisms reported previously (ie, how TLR3 overexpression contributes to the migration, motility, and promotion of prostate cancer).
      • Day K.C.
      • Lorenzatti Hiles G.
      • Kozminsky M.
      • Dawsey S.J.
      • Paul A.
      • Broses L.J.
      • Shah R.
      • Kunja L.P.
      • Hall C.
      • Palanisamy N.
      • Daignault-Newton S.
      • El-Sawy L.
      • Wilson S.J.
      • Chou A.
      • Ignatoski K.W.
      • Keller E.
      • Thomas D.
      • Nagrath S.
      • Morgan T.
      • Day M.L.
      HER2 and EGFR overexpression support metastatic progression of prostate cancer to bone.
      It has been shown that poly(I:C)-induced stimulation of TLR3 pathway activates mitogen-activated protein kinases and protein kinase C-α, and this leads to inhibition of proliferation as well as caspase-dependent apoptosis in PCa cells.
      • Paone A.
      • Starace D.
      • Galli R.
      • Padula F.
      • De Cesaris P.
      • Filippini A.
      • Ziparo E.
      • Riccioli A.
      Toll-like receptor 3 triggers apoptosis of human prostate cancer cells through a PKC-alpha-dependent mechanism.
      In our study, we demonstrated that, indeed, poly(I:C) stimulation promotes apoptosis in a TLR3 expression-dependent manner; docetaxel-resistant DU145 and PC3 cells underwent higher apoptotic response to poly(I:C) compared with therapy-responsive cells. In line with our results, it was demonstrated that exogenous TLR3 activation could improve chemotherapy-driven cytotoxicity in paclitaxel-resistant colon cancer.
      • Zhao J.
      • Xue Y.
      • Pan Y.
      • Yao A.
      • Wang G.
      • Li D.
      • Wang T.
      • Zhao S.
      • Hou Y.
      Toll-like receptor 3 agonist poly I:C reinforces the potency of cytotoxic chemotherapy via the TLR3-UNC93B1-IFN-beta signaling axis in paclitaxel-resistant colon cancer.
      Moreover, the strategy of TLR3 activation by poly(I:C) was already shown to be effective in several cancer types.
      • Vanbervliet-Defrance B.
      • Delaunay T.
      • Daunizeau T.
      • Kepenekian V.
      • Glehen O.
      • Weber K.
      • Estornes Y.
      • Ziverec A.
      • Djemal L.
      • Delphin M.
      • Lantuejoul S.
      • Passot G.
      • Gregoire M.
      • Micheau O.
      • Blanquart C.
      • Renno T.
      • Fonteneau J.F.
      • Lebecque S.
      • Mahtouk K.
      Cisplatin unleashes toll-like receptor 3-mediated apoptosis through the downregulation of c-FLIP in malignant mesothelioma.
      ,
      • Wu Y.
      • Huang W.
      • Chen L.
      • Jin M.
      • Gao Z.
      • An C.
      • Lin H.
      Anti-tumor outcome evaluation against non-small cell lung cancer in vitro and in vivo using polyI:C as nucleic acid therapeutic agent.
      Therefore, our data may be relevant for clinical practice in PCa. Indeed, invasive chemoresistant PCa cells with high TLR3 levels are more susceptible to being targeted and eliminated through the use of specific activating ligands. However, it should be kept in mind that TLR3 may act to stimulate carcinogenesis through the induction of metabolic reprogramming, so its use in clinical practice should be considered in light of all available results in a particular cancer type.
      • Matijevic Glavan T.
      • Cipak Gasparovic A.
      • Verillaud B.
      • Busson P.
      • Pavelic J.
      Toll-like receptor 3 stimulation triggers metabolic reprogramming in pharyngeal cancer cell line through Myc, MAPK, and HIF.
      In conclusion, our study showed that TLR3 is up-regulated in drug-resistant PCa and demonstrated that its presence has both detrimental and beneficial roles in PCa, based on its activation status. High levels of this receptor are, on one hand, associated with chemoresistance and induce cell migration. On the other hand, targeting TLR3 with an exogenous ligand leads to apoptosis, with higher toxicity in TLR3-high chemoresistant PCa cells.

      Uncited Reference

      • Gan Y.
      • Shi C.
      • Inge L.
      • Hibner M.
      • Balducci J.
      • Huang Y.
      Differential roles of ERK and Akt pathways in regulation of EGFR-mediated signaling and motility in prostate cancer cells.
      .

      Acknowledgments

      We thank Dr. Ján Remšík for help with Ingenuity Pathway Analysis; Iva Lišková, Martina Urbánková, Kateřina Svobodová, Alena Mičková, Eva Szczyrbová, Monika Levková, Michala Bezděková, and Gabriela Kořínková for technical assistance; Prof. Radek Vrtěl for cell line authentication; Dr. Petr Beneš for help with CRISPR/Cas9 approach; Dr. Vladimír Študent, Jr., for help with clinical samples; Dr. Feng Zhang for lentiCRISPRv2; Eric Campeau and Paul Kaufman for pLenti PGK V5-LUC Puro (w543-1); and Innate Pharma (Marseille, France) for kindly providing the anti–toll-like receptor 3 antibody for immunohistochemistry (clone 40F9.6).

      Author Contributions

      X.M.M. performed experiments and analyzed the data, interpreted the data, and wrote and reviewed the article; E.S., R.F., J.P., T.S., and S.D. helped with in vitro experiments and analyses; J.P. and V.H. performed RNA-sequencing and data analysis; M.Pu. and W.M.v.W. established chemoresistant cell lines; J.B., D.K., M.K., T.H., V.S., and M.P.S. provided clinical samples and performed their analyses; K.S., O.V., M.Pí., and R.V. performed in vivo experiments; E.S., J.P., W.M.v.W., Z.C., and J.B. interpreted the data and wrote and reviewed the article; D.P., V.P., and Z.Z. performed proteomic analysis and initial data evaluation; and K.S. conceptualized and designed the study, interpreted the data, wrote and reviewed the article, and supervised the study. All authors read and approved the final version of this article.

      Supplemental Data

      • Supplemental Figure S1

        Toll-like receptor 3 (TLR3) expression in prostate cancer (PCa) types and association with poor prognosis. A: Changes in TLR3 mRNA were analyzed from the Oncomine database (Compendia Bioscience, Ann Arbor, MI). Data extracted from the Best,

        • Best C.J.
        • Gillespie J.W.
        • Yi Y.
        • Chandramouli G.V.
        • Perlmutter M.A.
        • Gathright Y.
        • Erickson H.S.
        • Georgevich L.
        • Tangrea M.A.
        • Duray P.H.
        • Gonzalez S.
        • Velasco A.
        • Linehan W.M.
        • Matusik R.J.
        • Price D.K.
        • Figg W.D.
        • Emmert-Buck M.R.
        • Chuaqui R.F.
        Molecular alterations in primary prostate cancer after androgen ablation therapy.
        Varambally,
        • Varambally S.
        • Yu J.
        • Laxman B.
        • Rhodes D.R.
        • Mehra R.
        • Tomlins S.A.
        • Shah R.B.
        • Chandran U.
        • Monzon F.A.
        • Becich M.J.
        • Wei J.T.
        • Pienta K.J.
        • Ghosh D.
        • Rubin M.A.
        • Chinnaiyan A.M.
        Integrative genomic and proteomic analysis of prostate cancer reveals signatures of metastatic progression.
        Tomlins,
        • Tomlins S.A.
        • Mehra R.
        • Rhodes D.R.
        • Cao X.
        • Wang L.
        • Dhanasekaran S.M.
        • Kalyana-Sundaram S.
        • Wei J.T.
        • Rubin M.A.
        • Pienta K.J.
        • Shah R.B.
        • Chinnaiyan A.M.
        Integrative molecular concept modeling of prostate cancer progression.
        and Tamura
        • Tamura K.
        • Furihata M.
        • Tsunoda T.
        • Ashida S.
        • Takata R.
        • Obara W.
        • Yoshioka H.
        • Daigo Y.
        • Nasu Y.
        • Kumon H.
        • Konaka H.
        • Namiki M.
        • Tozawa K.
        • Kohri K.
        • Tanji N.
        • Yokoyama M.
        • Shimazui T.
        • Akaza H.
        • Mizutani Y.
        • Miki T.
        • Fujioka T.
        • Shuin T.
        • Nakamura Y.
        • Nakagawa H.
        Molecular features of hormone-refractory prostate cancer cells by genome-wide gene expression profiles.
        data sets. B: TLR3-dependent relapse-free survival data were extracted from GSE70768 (left panel) and GSE70769 (right panel) data sets (https://www.ncbi.nlm.nih.gov/geo, last accessed June 2022) using PROGeneV2 (http://www.progtools.net/gene).
        • Goswami C.P.
        • Nakshatri H.
        PROGgeneV2: enhancements on the existing database.
        P < 0.05 by t-test. H.N., hormone naive; H.R., hormone refractory; Met, metastatic PCa.

      • Supplemental Figure S2

        Toll-like receptor 3 (TLR3) gene expression in therapy-resistant prostate cancer. A: Gene expression of TLRs and downstream markers was performed by custom real-time quantitative PCR (qPCR) screen on parental and drug-resistant PCa PDX models and cell lines; data are representative of two biological repetitions assayed in technical duplicate each; GAPDH, ALAS1, and SDHA were used as housekeeping genes. Arrow indicates TLR3 gene. B: TLR3 gene levels in LAPC4 control (ctrl) and enzalutamide-resistant (enza) cells, as well as in DU145 and PC3 ctrl and docetaxel-resistant (doc) cells, were validated by the use of different sets of primers by qPCR; TBP was used as housekeeping gene. ∗P < 0.05 by t-test of each cell line pair. Id, identifier; max, maximum; min, minimum.

      • Supplemental Figure S3

        Immunohistochemistry testing of commercial anti–toll-like receptor 3 (TLR3) antibodies on formalin-fixed, paraffin-embedded cells and tissues. None of the antibodies provided staining concordant with the Western blot analysis of the DU145 cells. A, D, and G: Empty vector (EV). B, E, and H: Overexpression (OE). See also Supplemental Figure S4A. AC and GI: Furthermore, ThermoFisher monoclonal antibody (dilution 1:20; antigen retrieval in microwave; pH 6; AC) and Promokine polyclonal antibody (number PK-AB718-3643; dilution 1:1000; antigen retrieval in microwave; pH 6; GI) provided nuclear staining in virtually all cancer cells, which is not expected according to our current knowledge. DF: Granular cytoplasmic positivity was obtained with Santa Cruz Biotechnology antibody (dilution 1:20; harsh antigen retrieval using Ventana OptiView Amplification kit; DF); however, all cells were positive, and there was no difference between DU145 cells with empty vector and TLR3 overexpression (D and E, respectively). C, F, and I: Prostatectomy samples. Scale bar = 50 μm (AI).

      • Supplemental Figure S4

        Basal activation and induced activation of toll-like receptor 3 (TLR3) in prostate cancer cell lines. A: Basal levels of TLR3 activation were assessed by Western blot analysis as IRF3 phosphorylation (1- and 5-minute exposure times shown), in normally cultured and non-stimulated cells; total IRF3 and α-tubulin were used as loading controls. B: Plots show Western blot analysis quantification of phosphorylated IRF3 (p-IRF3) kinetics (Figure 2A) in three biological repetitions of DU145 and PC3 control (ctrl) and docetaxel-resistant (doc) cells stimulated with poly(I:C) in the absence or presence of bafilomycin. p-IRF3 levels were normalized to total IRF3. Enza, enzalutamide resistant.

      • Supplemental Figure S5

        Validation of toll-like receptor 3 (TLR3) expression in genetically modified DU145 clones. AC: Western blot analyses are representative of three biological repetitions; α-tubulin was used as a loading control. Arrows indicate position of specific bands. ctrl, control; EV, empty vector control; OE, TLR3 overexpressing.

      • Supplemental Figure S6

        Toll-like receptor 3 (TLR3)–dependent cell migration. A: Images represent stained cells migrated during transwell migration assay. B: Migration of empty vector control (EV) and TLR3 knockout (KO) cells is shown as percentage of relative wound density; performed by two-way analysis of variance. C: TLR3-overexpressing (OE) control (ctrl)-shRNA and TLR3-shRNA clones are shown as cell index measured by xCELLigence system; performed by two-way analysis of variance. D: Cell motility in EV and TLR3 KO clones is shown as distance traveled by cells on the two-dimensional surface; performed by one-way analysis of variance. Results shown as means ± SD (B) or means ± SEM (C). n = 2 (A). ∗P < 0.05. Original magnification, ×4 (A).

      • Supplemental Figure S7

        TIR domain of toll-like receptor 3 (TLR3) and epidermal growth factor receptor (EGFR) signaling in enhanced migration and motility of TLR3-overexpressing (OE) cells. A: Cell migration shown as cell index measured by xCELLigence system. B: Cell motility is shown as distance traveled by cells on a two-dimensional surface. C: Cell viability on docetaxel treatment is shown as percentage of relative luminescence units (RLUs) of each dose in comparison to dimethyl sulfoxide (DMSO)–treated control (100% viable cells); DU145 DOC cell line was used as a positive control for docetaxel resistance. D: Overexpression of dominant-negative (DN) TLR3 in DU145 cells detected by Western blot analysis; α-tubulin was used as loading control. White rectangles highlight blasticidin-resistant pools and single-cell clones G4 [empty vector control (EV)] and E11 (DN TLR3) used in functional experiments. E: EGFR phosphorylation (P-EGFR) was detected by Western blot analysis in TLR3 OE cells with control (ctrl) shRNA and TLR3 shRNA, and in EV and TLR3 OE cells treated with 0.3 μmol/L EGFR inhibitor lapatinib. F: Cell migration shown as cell index measured by xCELLigence system of EV and TLR3 OE cells pretreated with lapatinib for 2 hours. Results presented as means ± SEM (A and F). ∗P < 0.05 by one-way analysis of variance. WT, wild type.

      • Supplemental Figure S8

        Toll-like receptor 3 (TLR3)–dependent cell apoptosis. Right panel: Apoptosis kinetics on poly(I:C) stimulation of empty vector control (EV) and TLR3 knockout (KO) clones are presented as numbers of cells positive for caspase 3/7 (Casp3/7). Left panels: Images that are representative of cells positive for Casp3/7 (green fluorescence) at 24 hours after poly(I:C) stimulation. n = 2 (right panel). ∗P < 0.05 by two-way analysis of variance. Scale bar = 400 μm (left panels).

      References

        • Sung H.
        • Ferlay J.
        • Siegel R.L.
        • Laversanne M.
        • Soerjomataram I.
        • Jemal A.
        • Bray F.
        Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
        CA Cancer J Clin. 2021; 71: 209-249
        • Davies A.
        • Conteduca V.
        • Zoubeidi A.
        • Beltran H.
        Biological evolution of castration-resistant prostate cancer.
        Eur Urol Focus. 2019; 5: 147-154
        • Marin-Aguilera M.
        • Codony-Servat J.
        • Kalko S.G.
        • Fernandez P.L.
        • Bermudo R.
        • Buxo E.
        • Ribal M.J.
        • Gascon P.
        • Mellado B.
        Identification of docetaxel resistance genes in castration-resistant prostate cancer.
        Mol Cancer Ther. 2012; 11: 329-339
        • Fay E.K.
        • Graff J.N.
        Immunotherapy in prostate cancer.
        Cancers (Basel). 2020; 12: 1752
        • Hossain D.M.
        • Pal S.K.
        • Moreira D.
        • Duttagupta P.
        • Zhang Q.
        • Won H.
        • Jones J.
        • D'Apuzzo M.
        • Forman S.
        • Kortylewski M.
        TLR9-targeted STAT3 silencing abrogates immunosuppressive activity of myeloid-derived suppressor cells from prostate cancer patients.
        Clin Cancer Res. 2015; 21: 3771-3782
        • Ou T.
        • Lilly M.
        • Jiang W.
        The pathologic role of toll-like receptor 4 in prostate cancer.
        Front Immunol. 2018; 9: 1188
        • Muresan X.M.
        • Bouchal J.
        • Culig Z.
        • Soucek K.
        Toll-like receptor 3 in solid cancer and therapy resistance.
        Cancers (Basel). 2020; 12: 3227
        • Gao D.
        • Ciancanelli M.J.
        • Zhang P.
        • Harschnitz O.
        • Bondet V.
        • Hasek M.
        • Chen J.
        • Mu X.
        • Itan Y.
        • Cobat A.
        • Sancho-Shimizu V.
        • Bigio B.
        • Lorenzo L.
        • Ciceri G.
        • McAlpine J.
        • Anguiano E.
        • Jouanguy E.
        • Chaussabel D.
        • Meyts I.
        • Diamond M.S.
        • Abel L.
        • Hur S.
        • Smith G.A.
        • Notarangelo L.
        • Duffy D.
        • Studer L.
        • Casanova J.L.
        • Zhang S.Y.
        TLR3 controls constitutive IFN-beta antiviral immunity in human fibroblasts and cortical neurons.
        J Clin Invest. 2021; 131e134529
        • Gambara G.
        • Desideri M.
        • Stoppacciaro A.
        • Padula F.
        • De Cesaris P.
        • Starace D.
        • Tubaro A.
        • Del Bufalo D.
        • Filippini A.
        • Ziparo E.
        • Riccioli A.
        TLR3 engagement induces IRF-3-dependent apoptosis in androgen-sensitive prostate cancer cells and inhibits tumour growth in vivo.
        J Cell Mol Med. 2015; 19: 327-339
        • Kourko O.
        • Smyth R.
        • Cino D.
        • Seaver K.
        • Petes C.
        • Eo S.Y.
        • Basta S.
        • Gee K.
        Poly(I:C)-mediated death of human prostate cancer cell lines is induced by interleukin-27 treatment.
        J Interferon Cytokine Res. 2019; 39: 483-494
        • Palchetti S.
        • Starace D.
        • De Cesaris P.
        • Filippini A.
        • Ziparo E.
        • Riccioli A.
        Transfected poly(I:C) activates different dsRNA receptors, leading to apoptosis or immunoadjuvant response in androgen-independent prostate cancer cells.
        J Biol Chem. 2015; 290: 5470-5483
        • Magnifico M.C.
        • Macone A.
        • Marani M.
        • Bouzidi A.
        • Giardina G.
        • Rinaldo S.
        • Cutruzzola F.
        • Paone A.
        Linking infection and prostate cancer progression: toll-like receptor3 stimulation rewires glucose metabolism in prostate cells.
        Anticancer Res. 2019; 39: 5541-5549
        • Puhr M.
        • Hoefer J.
        • Schafer G.
        • Erb H.H.
        • Oh S.J.
        • Klocker H.
        • Heidegger I.
        • Neuwirt H.
        • Culig Z.
        Epithelial-to-mesenchymal transition leads to docetaxel resistance in prostate cancer and is mediated by reduced expression of miR-200c and miR-205.
        Am J Pathol. 2012; 181: 2188-2201
        • Hoefer J.
        • Akbor M.
        • Handle F.
        • Ofer P.
        • Puhr M.
        • Parson W.
        • Culig Z.
        • Klocker H.
        • Heidegger I.
        Critical role of androgen receptor level in prostate cancer cell resistance to new generation antiandrogen enzalutamide.
        Oncotarget. 2016; 7: 59781-59794
        • de Morree E.S.
        • Bottcher R.
        • van Soest R.J.
        • Aghai A.
        • de Ridder C.M.
        • Gibson A.A.
        • Mathijssen R.H.
        • Burger H.
        • Wiemer E.A.
        • Sparreboom A.
        • de Wit R.
        • van Weerden W.M.
        Loss of SLCO1B3 drives taxane resistance in prostate cancer.
        Br J Cancer. 2016; 115: 674-681
        • van Weerden W.M.
        • Bangma C.
        • de Wit R.
        Human xenograft models as useful tools to assess the potential of novel therapeutics in prostate cancer.
        Br J Cancer. 2009; 100: 13-18
        • van Soest R.J.
        • de Morree E.S.
        • Kweldam C.F.
        • de Ridder C.M.A.
        • Wiemer E.A.C.
        • Mathijssen R.H.J.
        • de Wit R.
        • van Weerden W.M.
        Targeting the androgen receptor confers in vivo cross-resistance between enzalutamide and docetaxel, but not cabazitaxel, in castration-resistant prostate cancer.
        Eur Urol. 2015; 67: 981-985
        • Zhu J.
        • Smith K.
        • Hsieh P.N.
        • Mburu Y.K.
        • Chattopadhyay S.
        • Sen G.C.
        • Sarkar S.N.
        High-throughput screening for TLR3-IFN regulatory factor 3 signaling pathway modulators identifies several antipsychotic drugs as TLR inhibitors.
        J Immunol. 2010; 184: 5768-5776
        • Schmittgen T.D.
        • Livak K.J.
        Analyzing real-time PCR data by the comparative C(T) method.
        Nat Protoc. 2008; 3: 1101-1108
        • Slabáková E.
        • Kharaishvili G.
        • Smějová M.
        • Pernicová Z.
        • Suchánková T.
        • Remšík J.
        • Lerch S.
        • Straková N.
        • Bouchal J.
        • Král M.
        • Culig Z.
        • Kozubík A.
        • Souček K.
        Opposite regulation of MDM2 and MDMX expression in acquisition of mesenchymal phenotype in benign and cancer cells.
        Oncotarget. 2015; 6: 36156-36171
        • Campeau E.
        • Ruhl V.E.
        • Rodier F.
        • Smith C.L.
        • Rahmberg B.L.
        • Fuss J.O.
        • Campisi J.
        • Yaswen P.
        • Cooper P.K.
        • Kaufman P.D.
        A versatile viral system for expression and depletion of proteins in mammalian cells.
        PLoS One. 2009; 4e6529
        • Wisniewski J.R.
        • Zougman A.
        • Nagaraj N.
        • Mann M.
        Universal sample preparation method for proteome analysis.
        Nat Methods. 2009; 6: 359-362
        • Demichev V.
        • Messner C.B.
        • Vernardis S.I.
        • Lilley K.S.
        • Ralser M.
        DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput.
        Nat Methods. 2020; 17: 41-44
        • Best C.J.
        • Gillespie J.W.
        • Yi Y.
        • Chandramouli G.V.
        • Perlmutter M.A.
        • Gathright Y.
        • Erickson H.S.
        • Georgevich L.
        • Tangrea M.A.
        • Duray P.H.
        • Gonzalez S.
        • Velasco A.
        • Linehan W.M.
        • Matusik R.J.
        • Price D.K.
        • Figg W.D.
        • Emmert-Buck M.R.
        • Chuaqui R.F.
        Molecular alterations in primary prostate cancer after androgen ablation therapy.
        Clin Cancer Res. 2005; 11: 6823-6834
        • Varambally S.
        • Yu J.
        • Laxman B.
        • Rhodes D.R.
        • Mehra R.
        • Tomlins S.A.
        • Shah R.B.
        • Chandran U.
        • Monzon F.A.
        • Becich M.J.
        • Wei J.T.
        • Pienta K.J.
        • Ghosh D.
        • Rubin M.A.
        • Chinnaiyan A.M.
        Integrative genomic and proteomic analysis of prostate cancer reveals signatures of metastatic progression.
        Cancer Cell. 2005; 8: 393-406
        • Tomlins S.A.
        • Mehra R.
        • Rhodes D.R.
        • Cao X.
        • Wang L.
        • Dhanasekaran S.M.
        • Kalyana-Sundaram S.
        • Wei J.T.
        • Rubin M.A.
        • Pienta K.J.
        • Shah R.B.
        • Chinnaiyan A.M.
        Integrative molecular concept modeling of prostate cancer progression.
        Nat Genet. 2007; 39: 41-51
        • Tamura K.
        • Furihata M.
        • Tsunoda T.
        • Ashida S.
        • Takata R.
        • Obara W.
        • Yoshioka H.
        • Daigo Y.
        • Nasu Y.
        • Kumon H.
        • Konaka H.
        • Namiki M.
        • Tozawa K.
        • Kohri K.
        • Tanji N.
        • Yokoyama M.
        • Shimazui T.
        • Akaza H.
        • Mizutani Y.
        • Miki T.
        • Fujioka T.
        • Shuin T.
        • Nakamura Y.
        • Nakagawa H.
        Molecular features of hormone-refractory prostate cancer cells by genome-wide gene expression profiles.
        Cancer Res. 2007; 67: 5117-5125
        • Goswami C.P.
        • Nakshatri H.
        PROGgeneV2: enhancements on the existing database.
        BMC Cancer. 2014; 14: 970
        • Ross-Adams H.
        • Lamb A.D.
        • Dunning M.J.
        • Halim S.
        • Lindberg J.
        • Massie C.M.
        • Egevad L.A.
        • Russell R.
        • Ramos-Montoya A.
        • Vowler S.L.
        • Sharma N.L.
        • Kay J.
        • Whitaker H.
        • Clark J.
        • Hurst R.
        • Gnanapragasam V.J.
        • Shah N.C.
        • Warren A.Y.
        • Cooper C.S.
        • Lynch A.G.
        • Stark R.
        • Mills I.G.
        • Gronberg H.
        • Neal D.E.
        • CamCa P.S.G.
        Integration of copy number and transcriptomics provides risk stratification in prostate cancer: a discovery and validation cohort study.
        EBioMedicine. 2015; 2: 1133-1144
        • Salaun B.
        • Zitvogel L.
        • Asselin-Paturel C.
        • Morel Y.
        • Chemin K.
        • Dubois C.
        • Massacrier C.
        • Conforti R.
        • Chenard M.P.
        • Sabourin J.C.
        • Goubar A.
        • Lebecque S.
        • Pierres M.
        • Rimoldi D.
        • Romero P.
        • Andre F.
        TLR3 as a biomarker for the therapeutic efficacy of double-stranded RNA in breast cancer.
        Cancer Res. 2011; 71: 1607-1614
        • Dauletbaev N.
        • Cammisano M.
        • Herscovitch K.
        • Lands L.C.
        Stimulation of the RIG-I/MAVS pathway by polyinosinic:polycytidylic acid upregulates IFN-beta in airway epithelial cells with minimal costimulation of IL-8.
        J Immunol. 2015; 195: 2829-2841
        • Galli R.
        • Paone A.
        • Fabbri M.
        • Zanesi N.
        • Calore F.
        • Cascione L.
        • Acunzo M.
        • Stoppacciaro A.
        • Tubaro A.
        • Lovat F.
        • Gasparini P.
        • Fadda P.
        • Alder H.
        • Volinia S.
        • Filippini A.
        • Ziparo E.
        • Riccioli A.
        • Croce C.M.
        Toll-like receptor 3 (TLR3) activation induces microRNA-dependent reexpression of functional RARβ and tumor regression.
        Proc Natl Acad Sci U S A. 2013; 110: 9812-9817
        • Galli R.
        • Starace D.
        • Busà R.
        • Angelini D.F.
        • Paone A.
        • De Cesaris P.
        • Filippini A.
        • Sette C.
        • Battistini L.
        • Ziparo E.
        • Riccioli A.
        TLR stimulation of prostate tumor cells induces chemokine-mediated recruitment of specific immune cell types.
        J Immunol. 2010; 184: 6658-6669
        • Gan Y.
        • Shi C.
        • Inge L.
        • Hibner M.
        • Balducci J.
        • Huang Y.
        Differential roles of ERK and Akt pathways in regulation of EGFR-mediated signaling and motility in prostate cancer cells.
        Oncogene. 2010; 29: 4947-4958
        • Gonzalez-Reyes S.
        • Fernandez J.M.
        • Gonzalez L.O.
        • Aguirre A.
        • Suarez A.
        • Gonzalez J.M.
        • Escaff S.
        • Vizoso F.J.
        Study of TLR3, TLR4, and TLR9 in prostate carcinomas and their association with biochemical recurrence.
        Cancer Immunol Immunother. 2011; 60: 217-226
        • Paone A.
        • Starace D.
        • Galli R.
        • Padula F.
        • De Cesaris P.
        • Filippini A.
        • Ziparo E.
        • Riccioli A.
        Toll-like receptor 3 triggers apoptosis of human prostate cancer cells through a PKC-alpha-dependent mechanism.
        Carcinogenesis. 2008; 29: 1334-1342
        • Qi R.
        • Singh D.
        • Kao C.C.
        Proteolytic processing regulates toll-like receptor 3 stability and endosomal localization.
        J Biol Chem. 2012; 287: 32617-32629
        • Toscano F.
        • Estornes Y.
        • Virard F.
        • Garcia-Cattaneo A.
        • Pierrot A.
        • Vanbervliet B.
        • Bonnin M.
        • Ciancanelli M.J.
        • Zhang S.Y.
        • Funami K.
        • Seya T.
        • Matsumoto M.
        • Pin J.J.
        • Casanova J.L.
        • Renno T.
        • Lebecque S.
        Cleaved/associated TLR3 represents the primary form of the signaling receptor.
        J Immunol. 2013; 190: 764-773
        • Lundberg A.M.
        • Drexler S.K.
        • Monaco C.
        • Williams L.M.
        • Sacre S.M.
        • Feldmann M.
        • Foxwell B.M.
        Key differences in TLR3/poly I:C signaling and cytokine induction by human primary cells: a phenomenon absent from murine cell systems.
        Blood. 2007; 110: 3245-3252
        • Araki S.
        • Omori Y.
        • Lyn D.
        • Singh R.K.
        • Meinbach D.M.
        • Sandman Y.
        • Lokeshwar V.B.
        • Lokeshwar B.L.
        Interleukin-8 is a molecular determinant of androgen independence and progression in prostate cancer.
        Cancer Res. 2007; 67: 6854-6862
        • Chuang H.C.
        • Chou M.H.
        • Chien C.Y.
        • Chuang J.H.
        • Liu Y.L.
        Triggering TLR3 pathway promotes tumor growth and cisplatin resistance in head and neck cancer cells.
        Oral Oncol. 2018; 86: 141-149
        • Chattopadhyay S.
        • Sen G.C.
        Tyrosine phosphorylation in toll-like receptor signaling.
        Cytokine Growth Factor Rev. 2014; 25: 533-541
        • Yamashita M.
        • Chattopadhyay S.
        • Fensterl V.
        • Saikia P.
        • Wetzel J.L.
        • Sen G.C.
        Epidermal growth factor receptor is essential for toll-like receptor 3 signaling.
        Sci Signal. 2012; 5: ra50
        • Astin J.W.
        • Batson J.
        • Kadir S.
        • Charlet J.
        • Persad R.A.
        • Gillatt D.
        • Oxley J.D.
        • Nobes C.D.
        Competition amongst Eph receptors regulates contact inhibition of locomotion and invasiveness in prostate cancer cells.
        Nat Cell Biol. 2010; 12: 1194-1204
        • Dai C.
        • Heemers H.
        • Sharifi N.
        Androgen signaling in prostate cancer.
        Cold Spring Harb Perspect Med. 2017; 7a030452
        • Darby S.
        • Cross S.S.
        • Brown N.J.
        • Hamdy F.C.
        • Robson C.N.
        BMP-6 over-expression in prostate cancer is associated with increased Id-1 protein and a more invasive phenotype.
        J Pathol. 2008; 214: 394-404
        • Hoflmayer D.
        • Steinhoff A.
        • Hube-Magg C.
        • Kluth M.
        • Simon R.
        • Burandt E.
        • Tsourlakis M.C.
        • Minner S.
        • Sauter G.
        • Buscheck F.
        • Wilczak W.
        • Steurer S.
        • Huland H.
        • Graefen M.
        • Haese A.
        • Heinzer H.
        • Schlomm T.
        • Jacobsen F.
        • Hinsch A.
        • Poos A.M.
        • Oswald M.
        • Rippe K.
        • Konig R.
        • Schroeder C.
        Expression of CCCTC-binding factor (CTCF) is linked to poor prognosis in prostate cancer.
        Mol Oncol. 2020; 14: 129-138
        • Stik G.
        • Vidal E.
        • Barrero M.
        • Cuartero S.
        • Vila-Casadesus M.
        • Mendieta-Esteban J.
        • Tian T.V.
        • Choi J.
        • Berenguer C.
        • Abad A.
        • Borsari B.
        • le Dily F.
        • Cramer P.
        • Marti-Renom M.A.
        • Stadhouders R.
        • Graf T.
        CTCF is dispensable for immune cell transdifferentiation but facilitates an acute inflammatory response.
        Nat Genet. 2020; 52: 655-661
        • Bugge M.
        • Bergstrom B.
        • Eide O.K.
        • Solli H.
        • Kjonstad I.F.
        • Stenvik J.
        • Espevik T.
        • Nilsen N.J.
        Surface toll-like receptor 3 expression in metastatic intestinal epithelial cells induces inflammatory cytokine production and promotes invasiveness.
        J Biol Chem. 2017; 292: 15408-15425
        • Liu Y.
        • Gu Y.
        • Han Y.
        • Zhang Q.
        • Jiang Z.
        • Zhang X.
        • Huang B.
        • Xu X.
        • Zheng J.
        • Cao X.
        Tumor exosomal RNAs promote lung pre-metastatic niche formation by activating alveolar epithelial TLR3 to recruit neutrophils.
        Cancer Cell. 2016; 30: 243-256
        • Day K.C.
        • Lorenzatti Hiles G.
        • Kozminsky M.
        • Dawsey S.J.
        • Paul A.
        • Broses L.J.
        • Shah R.
        • Kunja L.P.
        • Hall C.
        • Palanisamy N.
        • Daignault-Newton S.
        • El-Sawy L.
        • Wilson S.J.
        • Chou A.
        • Ignatoski K.W.
        • Keller E.
        • Thomas D.
        • Nagrath S.
        • Morgan T.
        • Day M.L.
        HER2 and EGFR overexpression support metastatic progression of prostate cancer to bone.
        Cancer Res. 2017; 77: 74-85
        • Zhao J.
        • Xue Y.
        • Pan Y.
        • Yao A.
        • Wang G.
        • Li D.
        • Wang T.
        • Zhao S.
        • Hou Y.
        Toll-like receptor 3 agonist poly I:C reinforces the potency of cytotoxic chemotherapy via the TLR3-UNC93B1-IFN-beta signaling axis in paclitaxel-resistant colon cancer.
        J Cell Physiol. 2019; 234: 7051-7061
        • Vanbervliet-Defrance B.
        • Delaunay T.
        • Daunizeau T.
        • Kepenekian V.
        • Glehen O.
        • Weber K.
        • Estornes Y.
        • Ziverec A.
        • Djemal L.
        • Delphin M.
        • Lantuejoul S.
        • Passot G.
        • Gregoire M.
        • Micheau O.
        • Blanquart C.
        • Renno T.
        • Fonteneau J.F.
        • Lebecque S.
        • Mahtouk K.
        Cisplatin unleashes toll-like receptor 3-mediated apoptosis through the downregulation of c-FLIP in malignant mesothelioma.
        Cancer Lett. 2020; 472: 29-39
        • Wu Y.
        • Huang W.
        • Chen L.
        • Jin M.
        • Gao Z.
        • An C.
        • Lin H.
        Anti-tumor outcome evaluation against non-small cell lung cancer in vitro and in vivo using polyI:C as nucleic acid therapeutic agent.
        Am J Transl Res. 2019; 11: 1919-1937
        • Matijevic Glavan T.
        • Cipak Gasparovic A.
        • Verillaud B.
        • Busson P.
        • Pavelic J.
        Toll-like receptor 3 stimulation triggers metabolic reprogramming in pharyngeal cancer cell line through Myc, MAPK, and HIF.
        Mol Carcinog. 2017; 56: 1214-1226