ORIGINAL RESEARCH published: 02 November 2020 doi: 10.3389/fonc.2020.594023 Cell Plasticity-Related Phenotypes and Taxanes Resistance in Castration-Resistant Prostate Cancer Natalia Jiménez 1,2, Òscar Reig 1,2,3, Ruth Montalbo 1,2, Maria Milà-Guasch 1, Lluis Nadal-Dieste 1,3, Giancarlo Castellano 4, Juan José Lozano 5, Iván Victoria 1,2,3, Albert Font 6, Alejo Rodriguez-Vida 7, Joan Carles 8, Cristina Suárez 8, Montserrat Domènech 9, Núria Sala-González 10, Pedro Luis Fernández 11, Leonardo Rodrı́guez-Carunchio 12, Sherley Dı́az 12, Aleix Prat 1,2,3,13, Mercedes Marı́n-Aguilera 1,2,3* and Begoña Mellado 1,2,3,13* 1 Translational Genomics and Targeted Therapeutics in Solid Tumors Lab, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 2 Translational Genomics and Targeted Therapeutics in Solid Tumors Lab, Fundació Edited by: Clı´nic per a la Recerca Biomèdica, Barcelona, Spain, 3 Medical Oncology Department, Hospital Clı´nic, Barcelona, Spain, Fred Saad, 4 Genomic Unit, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 5 Bioinformatics University of Montreal Hospital Centre Platform, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Barcelona, (CRCHUM), Canada Spain, 6 Medical Oncology Department, Institut Català d'Oncologia, Hospital Germans Trias i Pujol, Badalona, Spain, Reviewed by: 7 Medical Oncology Department, Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Hospital del Mar, Barcelona, Eugenio Zoni, Spain, 8 Vall d’Hebron Institute of Oncology, Vall d’Hebron University Hospital, Barcelona, Spain, 9 Medical Oncology University of Bern, Switzerland Department, Fundació Althaia Manresa, Barcelona, Spain, 10 Oncology Department, Institut Català d’Oncologia, Hospital Leena Latonen, Universitari Doctor Josep Trueta, Girona, Spain, 11 Department of Pathology, Hospital Germans Trias i Pujol, IGTP and University of Eastern Finland, Finland Universidad Autonoma de Barcelona, Badalona, Spain, 12 Department of Pathology, Hospital Clı´nic, Barcelona, Spain, 13 Department of Medicine, University of Barcelona, Barcelona, Spain *Correspondence: Begoña Mellado bmellado@clinic.cat The prostatic tumor cells plasticity is involved in resistance to hormone-therapy, allowing Mercedes Marı´n-Aguilera mmarin1@clinic.cat these cells to survive despite androgen receptor inhibition. However, its role in taxanes resistance has not been fully established. Gene expression of plasticity-related Specialty section: phenotypes such as epithelial-mesenchymal transition (EMT), stem cell-like and This article was submitted to Genitourinary Oncology, neuroendocrine (NE) phenotypes was studied in vitro, in silico, in circulating tumor cells a section of the journal (CTCs) (N=22) and in tumor samples (N=117) from taxanes-treated metastatic castration- Frontiers in Oncology resistant prostate cancer (mCRPC) patients. Docetaxel (D)-resistant cells presented a Received: 12 August 2020 Accepted: 05 October 2020 more pronounced EMT phenotype than cabazitaxel (CZ)-resistant cells. In silico analysis Published: 02 November 2020 revealed ESRP1 down-regulation in taxane-exposed mCRPC samples. Cell plasticity- Citation: related changes occurred in CTCs after taxanes treatment. Tumor EMT phenotype was Jiménez N, Reig Ò, Montalbo R, associated with lower PSA progression-free survival (PFS) to D (P<0.001), and better to Milà-Guasch M, Nadal-Dieste L, Castellano G, Lozano JJ, Victoria I, CZ (P=0.002). High ESRP1 expression was independently associated with longer PSA- Font A, Rodriguez-Vida A, Carles J, PFS (P<0.001) and radiologic-PFS (P=0.001) in D and shorter PSA-PFS in the CZ cohort Suárez C, Domènech M, Sala-González N, Fernández PL, (P=0.041). High SYP expression was independently associated with lower PSA-PFS in D Rodrı´guez-Carunchio L, (P=0.003) and overall survival (OS) in CZ (P=0.002), and high EZH2 expression was Dı´az S, Prat A, Marı´n-Aguilera M and associated with adverse OS in D-treated patients (P=0.013). In conclusion, EMT profile in Mellado B (2020) Cell Plasticity-Related Phenotypes and primary tumor is differentially associated with D or CZ benefit and NE dedifferentiation Taxanes Resistance in Castration- correlates with adverse taxanes clinical outcome. Resistant Prostate Cancer. Front. Oncol. 10:594023. Keywords: cell plasticity, EMT—epithelial-mesenchymal transition, neuroendocrine, castration-resistant prostate doi: 10.3389/fonc.2020.594023 cancer, docetaxel, cabazitaxel, taxanes resistance Frontiers in Oncology | www.frontiersin.org 1 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance INTRODUCTION authenticated using Human 9-Marker STR Profile and Interspecies Contamination Test by IDEXX BioAnalytics. Cell plasticity refers to the ability of cancer cells to switch their DU-145 and PC-3 cell lines were cultured in RPMI 1640 phenotype in response to environmental conditions, facilitating medium (Gibco, Life Technologies, Paisley, UK) and in F-12K therapy fail. Specifically in prostate cancer (PC) cell plasticity nutrient mixture medium (Gibco), respectively, both allows cancer cells to reprogram and survive despite androgen supplemented with 10% fetal bovine serum (FBS). D and CZ receptor (AR) inhibition, being considered one of the mechanisms drugs (MedChem Express, Monmouth Junction, NJ, USA) were involved in androgen deprivation therapy (ADT) resistance (1). dissolved in 10 mM in DMSO. D and CZ-resistant cells were Throughout this process, tumor cells may develop epithelial- maintained in 2 nM of D or CZ-containing medium, respectively. mesenchymal transition (EMT), and/or evolve towards an stem cell-like (SCL) or neuroendocrine (NE) phenotypes, and it is Cell Viability Assays possible that they acquire mixed or intermediate phenotypes (2, 3). Cells were seeded at a density of 3,000 cells/well in a 96-well ADT, novel hormone-therapies consisting in AR signaling microtiter plate. After 24 h, cells were exposed to D or CZ inhibitors (ARSI; abiraterone or enzalutamide), and taxanes, (MedChem Express, Monmouth Junction, NJ, USA) for an chemotherapy agents that act by blocking microtubules additional 72 h. Cell viability was assessed by using Cell Titer depolymerization, are the most used therapies in metastatic PC 96 Aqueous One Solution Cell Proliferation Assay kit (Promega, (4–8). Today, a high percentage of patients with disseminated PC Madison, WI, USA) according to instructions protocol. receive novel hormone-therapies at early stages of disease: in combination with ADT in non-castrate PC, in non-metastatic Cell Adhesion and Proliferation castration-resistant PC or as first-line treatment for metastatic Measurement castration-resistant PC (mCRPC). Taxanes are used after ARSI Quantitative monitoring of cell adhesion and proliferation was progression but its clinical activity may be eventually compromised performed using the xCELLigence real-time cell analysis (RTCA) by the development of cell plasticity-related phenotypes (9, 10). system, which measures the change in electrical impedance Furthermore, different degrees of NE dedifferentiation (from pure, expressed as the Cell Index (CI). Experiments were carried out in mixed or intermediate adenocarcinoma/NE phenotypes) have been a RTCA DP Instrument (ACEA Bio-sciences, San Diego, CA, USA) observed in biopsies from mCRPC patients who progressed to placed in a humidified incubator maintained at 37 °C with 5% CO2 abiraterone and its presence is associated with an adverse clinical according to manufacturer instructions. Briefly, 50 µl of cell-free outcome (11). Moreover, no specific clinical strategies are defined medium with 10% FBS was added to the wells of E-16 plates for based on the presence of these phenotypes and no targeted therapies background impedance detection, then 5,000 cells/well were seeded. have demonstrated benefit in mCRPC patients. Cell Index was monitored for 90 h, the first 4 h every 1 min, then On the other hand, although there is growing evidence that once every 30 min. Monitoring of the cells over the first hours tumor cell plasticity is a relevant biological event involved in provides adhesion data, while over the subsequent hours provides therapeutic resistance and aggressive evolution of PC, the proliferation data. Each cell line was tested in triplicate. The rate of presence of these cell plasticity related phenotypes already in adhesion and proliferation were determined by calculating the the primary tumor and its impact in the therapeutic response averaged slopes of the curves between two given time points. and clinical outcome has not been fully described. Data were analyzed by RTCA 2.0 software (ACEA Bio-sciences). In the present study we investigated the presence and potential role in predicting tumor evolution of EMT, SCL and Cell Migration Assay NE cell plasticity-related phenotypes in CRPC cell lines, in silico, Cell migration was performed using the Cultrex 96 Well Cell in circulating tumor cells (CTCs) and in tumor biopsies from Migration Assay (Trevigen, Gaithersburg, MD, USA) for 24 h and mCRPC taxanes-treated patients. quantified using Calcein-AM according to the manufacturer’s instructions. Each assay was performed in sextuplicate and the experiment was repeated twice, independently. MATERIALS AND METHODS Western Blot Analysis Establishment of Taxane-Resistant Cells Whole-cell extracts were prepared and Western blot analysis DU-145 and PC-3 cells were purchased from American Type performed as described previously (12). Nitrocellulose Culture Collection (ATCC). They were converted to docetaxel- membranes were blocked, incubated and washed following the resistant (DU-145DR and PC-3DR, respectively) and to Odyssey System (Li-Cor Biosciences, Lincoln, NE, USA) cabazitaxel-resistant cells (DU-145CZR and PC-3CZR, recommendations. Blots were scanned with an Odyssey respectively) by exposing them to an initial dose of 1.3 nM of Infrared Imaging System (Li-Cor Biosciences). docetaxel (D) and 0.6 nM of cabazitaxel (CZ), respectively, and Antibodies used were Aurora A (D3E4Q) (AURKA) (ref. culturing surviving cells with stepwise increasing doses in an 14475), AXL (C89E7) (ref. 8661), CD44 (156-3C11) (ref. 3570), intermittent regimen until a concentration of 6 nM during one E-cadherin (CDH1) (ref. 4065), N-cadherin (CDH2) (ref. 4061), year approximately. A subset of parental cells was cultured N-MYC (D4B2Y) (ref. 51705), Synaptophysin (D35Ea) (SYP) alongside the resistant ones as a control. All cell lines were (ref. 5461), Vimentin (R28) (VIM) (ref. 3932), b-Catenin (6B3) Frontiers in Oncology | www.frontiersin.org 2 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance (CTNNB1) (ref. 9582) purchased from Cell-Signaling Affymetrix WT PLUS Reagent Kit user guide. Following Technology (Leiden, The Netherlands). Actin (ref. A2066), fragmentation and terminal labelling, 5.2 µg of ss-cDNA were CHGA (ref. AMAb90525), ESRP1 (ref. HPA023719), and hybridized for 16 h at 45 °C on GeneChip Human Gene 2.0 ST Monoclonal Anti-a-Tubulin clone B-5-1-2 (TUB) (ref. T5168) Arrays (Affymetrix, Santa Clara, CA, USA). Arrays were washed were purchased from Sigma-Aldrich (St. Louis, MO, USA). ZEB1 and stained in the Affymetrix Fluidics Station 450. GeneChips were (H-102) antibody (ref. sc-25388) was purchased from Santa Cruz scanned using the Affymetrix GeneCHip Scanner 3000 System. Biotechnology (Santa Cruz, CA, USA). Raw expression data from microarrays were normalized using the robust multiarray algorithm (13) with a custom probe set RNA Extraction definition that mapped probes to Entrez Gene IDs Total RNA from cell lines and CTCs samples was isolated using the (hugene20st_Hs_ENTREZG) (14). After this step, a filtering Trizol Reagent (Tri-reagent solution; Invitrogen) according to the was done to obtain the 5% genes with the lowest coefficient of manufacturer’s protocol. For microarray hybridizations, RNA from variation. Finally, to identify differentially expressed genes cell lines was purified (from aqueous phase of Trizol lysis) using between the different microarray study groups, we employed RNeasy Micro kit (QIAGEN, Hilden, Germany). In case of LIMMA (15) to estimate moderated t-statistics and to select formalin-fixed paraffin-embedded (FFPE) tumor samples, total statistically differentially expressed genes. RNA was isolated using RecoverAll Total Nucleic Acid Isolation Significant genes were selected if they accomplished a fold Kit (Invitrogen) according to the manufacturer's instructions. RNA change (|FC|)≥1.5 and false discovery rate (FDR)<0.05. was quantified by NanoDrop Spectrophotometer ND-1000 Differentially expressed genes of resistant versus parental cells (Thermo Scientific, Wilmington, MA, USA) and RNA quality for were mapped against the Kyoto Encyclopedia of Genes and microarrays was assessed using a Bioanalyzer 2100 (Agilent, Santa Genomes (KEGG) pathway database (16) and analysis was Clara, CA). performed using clusterProfiler (v3.8.1) R package. Gene interactions were studied using the Ingenuity Pathway Gene Expression Analysis Analysis (IPA) software (QIAGEN). One mg of total RNA from cell lines and 0.5 mg from tissue We defined signatures for epithelial-mesenchymal transition biopsies and CTCs was reverse transcribed using the High (EMT) (345 EMT-related genes from the dbEMT database (17) Capacity cDNA Archive Kit (Life Technologies), following plus ESRP1), stem cell-like (SCL) [659 genes from five expression manufacturer’s instructions. In case of tissues and CTCs profiles (18–21) using StemChecker online (22)] and samples, cDNA samples were pre-amplified using TaqMan neuroendocrine (NE) prostate cancer (NEPC) [70 NE-related PreAmp Master Mix kit following manufacturer's instructions genes from the integrated NEPC score (23) plus ASCL1, AURKB, (Applied Biosystems, Foster City, CA, USA), except that the final CHGA, and SYP (24)], and tested them in silico in our reaction volume of the reaction was 12.5 ml. microarray data. Hierarchical cluster analysis of these Real-time quantitative reverse-transcription PCR (qRT-PCR) signatures was performed using Cluster 3.0 (25) and results was performed in duplicate using a StepOne Plus Real-Time PCR were visualized in Java TreeView (26). system (Applied Biosystems), according to the manufacturer’s recommendations. Data were acquired using Step One Software Circulating Tumor Cells Enrichment v2.2.2. Expression values were based on the quantification cycle Blood circulating tumor cells (CTCs) enrichment was performed (Cq) from target genes relative to the Cq of the housekeeping using the IsoFlux System (Fluxion Biosciences, South San genes GUSB in cell lines and tissue samples and ACTB in CTCs Francisco, CA, USA). Two 10 mL EDTA tubes were collected (DCq). Minus DCq values were considered as the expression from patients before taxanes initiation and after three cycles or at level in tissue samples. Relative expression with respect to each progression. One of the tubes was used for CTCs counting with reference group studied was reported as fold change or log2ratio. the IsoFlux CTC Enumeration Kit, according to manufacturer’s Target genes were amplified using commercial primers and instructions. CTCs were defined as nucleated cells, probes (Applied Biosystems): ACTB (Hs99999903_m1), morphologically intact, cytokeratin positive and CD45 negative AURKA (Hs01582073_m1), AXL (Hs01064444_m1), CD44 cells. The second EDTA tube was used for gene expression (Hs01075861_m1), CDH1 (Hs01023895_m1), CDH2 analysis. Briefly, peripheral blood mononuclear cells were (Hs00169953_m1), CHGA (Hs00900375_m1), CTNBB1 isolated by Ficoll (GE Healthcare, Uppsala, Sweden) gradient (Hs00170025_m1), ESRP1 (Hs00214472_m1), EZH2 and IsoFlux Rare Cell Enrichment Kit was used to incubate them (Hs00544830_m1), GUSB (Hs99999908_m1), MYCN with customized anti-epithelial cell adhesion molecule (Hs00232074_m1), SYP (Hs00300531_m1), TC2N (EPCAM), prostate stem cell antigen (PSCA) and N-cadherin (Hs01120134_m1), TWIST1 (Hs01675818_s1), VIM (CDH2)-coated beads for 2 h at 4 °C. Cell-beads complexes were (Hs00185584_m1), ZEB1 (Hs01566407_m1). loaded into Isoflux cartridges to be isolated automatically by the instrument. Cells returned by the instrument were stained or Microarrays Hybridization and Differential total RNA was extracted. Expression Analysis RNA samples from parental and resistant cell lines were processed In Silico Analysis in Prostate Cancer using WT PLUS chemistry (Affymetrix, Santa Clara, CA, USA). The alteration status of genes involved in EMT, SCL, and NE cell Fragmented and labelled ss-cDNA was prepared according to plasticity related phenotypes was studied in an in silico analysis Frontiers in Oncology | www.frontiersin.org 3 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance including seven PC studies summing a total of 1,131 samples (23, between both agents. Phenotypically, DR differed from CZR 27–32) obtained from cBioportal for Cancer Genomics platform cells when compared to parental cells: DR cells were more (http://cbioportal.org) (33, 34). elongated, with lower adhesion and proliferation, and higher RNA-seq data from Abida et al. (35) were also downloaded migration rates, while CZR cells were more rounded and similar from cBioportal and analyzed. Boxplots representing log FPKM to parental cells, with similar (DU-145 model) or higher (PC-3 (Fragments Per Kilobase of exon per Million fragments mapped) model) adhesion and lower proliferation rates, and without were generated by R (v.3.6.3) software. significant differences in migration rate (Figures 1B–E). These patterns suggest that DR cells have more motility abilities than Patients and Samples CZR cells. We retrospectively collected tumor samples from patients diagnosed with mCRPC treated with D (75 mg/m2 iv every 3 Differentially Expressed Genes in Taxane- weeks) or CZ (25 mg/m2 iv every 3 weeks or 10 mg/m2 iv Resistant Versus Parental Cell Lines weekly), both in association with prednisone (10 mg/day p.o), We performed cDNA microarrays to study the gene expression with available FFPE tumor tissue from several institutions. changes related to taxanes-acquired resistance in the above Patients under the same treatment regimen with the characterized cell lines. The microarray data analysis revealed possibility of extracting CTCs samples were prospectively 625 and 147 differentially deregulated genes in DR and CZR included as a part of an ongoing prospective biomarker study versus parental cells (|FC|≥1.5 and FDR<0.05), respectively in our institution. Blood samples were extracted before starting (Figure 2A and Tables S1 and S2). KEGG pathway treatment and after 3 cycles and/or at time of progression enrichment and Ingenuity network analysis showed the Treatment-response criteria and progressive-disease relevance of EMT and NFKb pathway in both D and CZ- definitions followed Prostate Cancer Working Group 2 resistance, among other deregulated pathways. Of note, ESRP1 criteria (36). appeared highly underexpressed in DR models (Figure S1A–C). Twenty-three genes were commonly deregulated in both DR Statistical Analysis and CZR versus parental cells (|FC|≥1.5 and FDR<0.05) (Figure In vitro experiments data were expressed as mean ± SD and 2A and Table S3). Among them, ZEB1 and ITGB3 were up- analyzed by Student t-test. All tests were 2-sided and P-values regulated and BMP4 was down-regulated. Ingenuity network <0.05 were considered significant. Optimal cut-offs for gene analysis showed that the most relevant network had two expression were assessed using maximally selected log-rank significant nodes around EGFR and ITGB3 genes and the statistics (Maxstat package) (37). PSA progression-free survival EMT-related markers, ZEB1 a nd TRPC1, app ea red (PFS), radiologic-PFS (RX-PFS), and overall survival (OS) were overexpressed (Figure S1D). calculated from the date of taxanes initiation to PSA progression, radiologic progression, and death or last follow-up visit, Differential Expression Analysis in CZ respectively, and were evaluated by log-rank test. Univariate Versus D-Resistant Cells analysis of gene expression levels and other clinical variables was From the microarray data analysis we found 559 genes performed by Cox regression; P<0.1 was required for inclusion in significantly commonly deregulated in CZR models versus DR multivariate analysis. Test of interaction was performed by models (|FC|≥1.5 and FDR<0.05) (Table S4). ESRP1, TC2N, entering into proportional hazard models selected MPZL2, CDH1, OCLN, and EPCAM were epithelial-related multiplicative interaction terms between two binary variables: genes overexpressed in CZR respect to DR cell lines, whereas treatment (D or CZ) and ESRP1 expression (high or low). AXL was down-regulated. Ingenuity network analysis centered Changes in gene expression in CTCs samples before and after the most relevant network on CDH1 and EZH2 (Figure S1E). treatment were analyzed by Wilcoxon signed-rank test. SPSS These results were validated by qRT-PCR and Western Blot. 12.0 and R (v.3.6.3) softwares were used for statistical analyses. By these techniques it was observed an overall increased expression of mesenchymal markers such as ZEB1 and VIM and a reduced expression of epithelial markers CDH1 and ESRP1 RESULTS lead by taxanes. This validation supports a more pronounced EMT profile in DR cells (Figure S2). Phenotypic Characterization of Docetaxel We defined, based on literature and published gene and Cabazitaxel-Resistant Cells expression data (as described in Materials and Methods), EMT, Docetaxel (D) and cabazitaxel (CZ) resistant cell lines (DR and SCL, and NE-signatures and tested them in silico in our CZR, respectively) were generated from CRPC cell line models microarray data. We found 33 EMT, 24 SCL and 4 NE-related DU-145 and PC-3. DU-145DR and PC-3DR cells acquired levels genes differentially expressed in CZR versus DR cells (|FC|≥1.5 of resistance to D that were 1.5 to 2.1 times higher than their and FDR<0.05). Hierarchical cluster analysis revealed a different parental cells. The levels of resistance to CZ for DU-145CZR and deregulation pattern of EMT (more pronounced in DR cells) and PC-3CZR were 2.2 to 4.5 times higher than their parental cells SCL-related genes between D and CZ-resistance and more (Figure 1A). DR and CZR cells were also resistant to CZ and D, expression of NE markers in CZR than in DR cells (Figures respectively, suggesting the existence of a cross resistance 2B–D). Frontiers in Oncology | www.frontiersin.org 4 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance A B C E D FIGURE 1 | Taxane-resistant cells characterization. (A) Cell-viability assays (MTS) of docetaxel (D) and cabazitaxel (CZ)-treatment in D-resistant (DR) and CZ- resistant (CZR) cells and their respective parental cells (DU-145 and PC-3). IC50 value of each cell line from these MTS assays is indicated. Results are represented as mean of % viability ± SD. Each treatment was carried out in triplicate for 72 h. (B) Cell adhesion curves of parental and resistant cells. Data is plotted as mean of Cell Index. Adhesion rates (bar-graphs) were determined by analyzing the slope of the lines between the 2 h and 6 h interval. (C) Cell proliferation curves of parental and resistant cells. Data is plotted as mean of Cell Index ± SD. Proliferation rates (bar-graphs) were determined by analyzing the slope of the lines between the 12 h and 24 h interval. (D) Migration rate quantification of DR and CZR cells versus their respective parental cells. Results are represented as mean of % migration ± SD. Significant differences are indicated as *P < 0.05 and **P < 0.001 (Student t-test). (E) Light microscopy images (100X) showing the morphology of DR and CZR cells (cultured with 2 nM of D or CZ, respectively) and their respective parental cells. Frontiers in Oncology | www.frontiersin.org 5 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance A B C D FIGURE 2 | Differentially expressed genes in docetaxel-resistant (DR) and cabazitaxel-resistant (CZR) cell lines from microarray data. (A) Venn Diagram of the differentially expressed genes between DR and CZR cell lines versus their respective parental cells (PAR). (B) Hierarchical clustering expression heatmap for expression values of the epithelial-mesenchymal transition (EMT)-signature in DR and CZR cell lines. (C) Hierarchical clustering expression heatmap for expression values of the neuroendocrine (NE)-signature in DR and CZR cell lines. (D) Hierarchical clustering expression heatmap for expression values of the NE-signature in DR and CZR cell lines. Taxanes Dose-Response Experiments dependent manner, restored ESRP1 expression in DU-145DR In order to evaluate whether EMT and NE patterns were model and decreased AXL levels in DR models but not in CZR modifiable in a dose-dependent manner we treated cell lines cells (Figure S3). D and CZ-treatment also increased the with increasing doses of taxanes and evaluated EMT and NE expression of NE markers SYP, MYCN (only detected in PC-3 gene expression by qRT-PCR. We observed that D and CZ- model), and CHGA in a dose-response manner in both resistant treatment induced changes in the expression of EMT and NE models (Figure S4). All together these results confirm the markers at different degrees. Notably, CZ exposure significantly plasticity towards EMT and NE patterns after a short-term increased ZEB1 and VIM expression in PC-3 models in a dose- drug exposure in vitro. Frontiers in Oncology | www.frontiersin.org 6 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance In Silico Analysis of Cell Plasticity-Related Amplification of ZEB1 (7.9%) and AXL (10.5%), both found in Genes in Prostate Cancer NEPC, were also frequent alterations. Otherwise, CDH1 showed a A subgroup of cell plasticity related genes was selected among those variety of alterations, but most notably was amplified in NEPC significantly deregulated between CZR and DR cell lines (and (3.5%) and deleted in primary tumors and in metastatic PC (0.5%– validated by qRT-PCR and Western Blot) to be analyzed in silico 4.5%) (Figure 3A and Table S5). in seven PC studies obtained from cBioportal: three studies of We also investigated gene expression changes in tumors from primary tumors with 333 (30), 240 (31), and 57 samples (32); three mCRPC patients treated with taxanes, by using RNA-seq data studies of metastatic PC with 61 (27), 176 (28), and 150 samples from the study of Abida et al. (35) Within ARSI-naïve patients, (29); and one study of NEPC with 114 samples (23). ESRP1 gene, ESRP1 expression was significantly lower in taxane-exposed one of the most differentially deregulated genes between DR and respect to taxane-naïve patients (P=0.007) (Figure 3B). Median CZR, was mostly amplified, being this alteration detected in around expression value for AXL, VIM, ZEB1, AURKA, EZH2, MYCN, 5% of primary tumors, 20% of metastatic PC, and in 40% of NEPC. and SYP was higher and for CDH1 was lower in taxane-exposed A B FIGURE 3 | In silico analysis from public datasets. (A) Bar graphs representing alterations’ frequency of cell plasticity related genes in prostate cancer from seven studies (23,27–32) through cBio Cancer Genomics Portal (http://cbioportal.org) platform. (B) Boxplot representing ESRP1 expression (log FPKM (Fragments Per Kilobase of exon per Million fragments mapped)) from RNA-seq data in taxane-naïve versus taxane-exposed mCRPC tumor samples from ARSI-naïve cohort of Abida et al. (35). Welch’s t-test was used for means comparison. Frontiers in Oncology | www.frontiersin.org 7 November 2020 | Volume 10 | Article 594023 Jiménez et al. Cell Plasticity and Taxanes Resistance versus taxane-naïve patients, although these differences were not patients, 17 of them with PSA-progression at post-treatment statistically significant (Figure S5). These results confirm the effect extraction, and performed qRT-PCR analyses. Patients’ of taxanes in the ESRP1 down-regulation and EMT up-regulation characteristics’ are shown in Table 1. Gene expression in a clinical sample setting. patterns in CTCs before and after taxanes treatment showed changes in expression levels of all EMT and NE markers, with Cell Plasticity in CTCs high intra- and inter-patient heterogeneity (Figures 4A, B and Beyond the in silico analysis, we wondered whether molecular Figure S6). CDH1 expression was significantly higher in post- plasticity changes could be observed in CTCs samples. Then we treatment samples (P=0.034). Response was not correlated with collected 47 CTCs-enriched blood samples from 22 mCRPC pre-treatment expression values or with the difference obtained TABLE 1 | Clinical characteristics of patients with CTCs-enriched blood samples. Total DOCETAXEL CABAZITAXEL P-value Patients, N (%) 22* 16 (66.7) 8 (33.3) (*2 included in both D and CZ cohorts, 1 has 1 sample post-D and pre-CZ) Age* (years) Median (range) 70 (41.6 - 87.1) 70.1 (41.6 - 87.1) 67.9 (43.2 - 83.3) Post-treatment samples, N (%) end of treatment 15 (62.5) 10 (62.5) 5 (62.5) after progression 6 (25) 4 (25) 2 (25) during treatment 3 (12.5) 2 (12.5) 1 (12.5) PSA response in post-treatment samples, N (%) Stable disease 4 (16.7) 3 (18.8) 1 (12.5) Partial response 3 (12.5) 1 (6.3) 2 (25) Progression 17 (70.8) 12 (75) 5 (62.5) CTCs pre-treatment Mean (range) 17 (0 - 100) 23 (1 - 100) 5 (0 - 22) 0.009 CTCs post-treatment Mean (range) 21 (0 - 182) 23 (0 - 182) 16 (1 - 44) 0.713 Stage at diagnosis, N (%)