Molecular profiling of prostate cancer derived exosomes may reveal a predictive signature for response to docetaxel

Abstract

Docetaxel is a cornerstone treatment for metastatic, castration resistant prostate cancer (CRPC) which remains a leading cause of cancer-related deaths, worldwide. The clinical usage of docetaxel has resulted in modest gains in survival, primarily due to the development of resistance. There are currently no clinical biomarkers available that predict whether a CRPC patient will respond or acquire resistance to this therapy. Comparative proteomics analysis of exosomes secreted from DU145 prostate cancer cells that are sensitive (DU145 Tax-Sen) or have acquired resistance (DU145 Tax-Res) to docetaxel, demonstrated significant differences in the amount of exosomes secreted and in their molecular composition. A panel of proteins was identified by proteomics to be differentially enriched in DU145 Tax-Res compared to DU145 Tax-Sen exosomes and was validated by western blotting. Importantly, we identified MDR-1, MDR-3, Endophilin-A2 and PABP4 that were enriched only in DU145 Tax-Res exosomes. We validated the presence of these proteins in the serum of a small cohort of patients. DU145 cells that have uptaken DU145 Tax-Res exosomes show properties of increased matrix degradation. In summary, exosomes derived from DU145 Tax-Res cells may be a valuable source of biomarkers for response to therapy.

Keywords: biomarkers; docetaxel; exosomes; prostate cancer; resistance.

Figure 1. Characterization of extracellular vesicles secreted from docetaxel sensitive and resistant DU145 cells

(A) Exosomes isolated from DU145 Tax-Sen and DU145 Tax-Res cells were visualized by transmission electron microscopy (bar size: 100 nm); (B) Nanoparticle tracking analysis on an LM10 Nanosight demonstrating a mean size of 100 nm for DU145 Tax-Res and 120 nm for DU145 Tax-Sen exosomes. The size distribution and relative concentration were calculated by the Nanosight software (n = 3); (C) Western blot analysis of 10 μg lysates from DU145 Tax-Sen and DU145 Tax-Res cells and exosomes and probed for the indicated proteins (n = 2); (D) Flow cytometric analysis of the mean fluorescence intensity (MFI) for a panel of exosomal markers CD9, CD63, CD81 and CD82. Data is presented as means of triplicate experiments.

Figure 2. Comparative biochemical characterization of exosomes isolated from docetaxel sensitive and resistant DU145 cells

(A) Nanoparticle tracking analysis of sucrose gradient fractions with the indicated sucrose density. The size distribution and relative concentration were calculated by the Nanosight software (n = 3); (B) Flow cytometric analysis for the mean fluorescence intensity (MFI) of CD9-APC in the sucrose gradient fractions (n = 2); (C) Western blot analysis of the sucrose gradient fractions from DU145 Tax-Sen and DU145 Tax-Res exosomes, probed for Rab 5, TSG101 and CD9, (n = 2).

Figure 3. Bioinformatics analysis of the DU145 Tax-Sen and DU145 Tax-Res exosomal proteome

(A) Venn Diagram constructed from the unique and shared proteins identified in DU145 Tax-Sen and DU145 Tax-Res exosomes; (B) Ingenuity pathway analysis of the signaling cascades identified in the proteomics analysis; (C) Pie charts from the classification analysis (PANTHER 9.0) of the biological processes, molecular functions, cellular components and protein classes.

Figure 4. Proteomics profiling of exosomes isolated from docetaxel sensitive and resistant DU145 cells

(A) Heat map of the hierarchical clustering of the 100 most abundant protein identified. The values were mean centered and log-transformed. The relative protein abundance is colored-coded with red corresponding to a relatively high abundance, green to a relatively low abundance, and black indicating indifferent abundance values. Each exosome sample was analyzed in duplicates; (B) Western blot analysis of 10 μg of DU145 Tax-Sen and DU145 Tax-Res exosomes, probed for the indicated proteins. Rab5 and TSG101 were used as exosomal markers and AIF as a quality control of the exosomal isolation (n = 2); (C) Quantification of the abundance values for MDR-1 and PABP4; (D) Western blot analysis of the sucrose gradient fractions from DU145 Tax-Sen and DU145 Tax-Res exosomes, probed for MDR-1/3 and TSG101 (n = 2).

Figure 5. Validation of MDR-1/3 and PABP4 in the serum of castration resistant prostate cancer patients

(A) Nanoparticle tracking analysis of the mean concentration and size of extracellular vesicles (EV) isolated from the serum of 3 docetaxel resistant, castration-resistant prostate cancer (CRPC) patients and 3 docetaxel sensitive CRPC patients; The size distribution and relative concentration were calculated by the Nanosight software. (B) Protein concentration measurements of the EVs isolated from 3 docetaxel sensitive CRPC patients and 3 docetaxel resistant CRPC patients; (C) Western blot analysis of 10 μg of CRPC Tax-Sen and CRPC Tax-Res EVs, probed for MDR-1/3, PABP4 and the exosomal markers Rab5, Alix and CD9.

Figure 6. DU145 Tax-Res exosomes promote DU145 cell invasion

(A) Quantification of PKH67-labeled, DU145 Tax-Res exosome uptake by DU145 cells after 3 hours and 24 hours of incubation. For control, DU145 cells were incubated with PBS for 24 hours (ctrl). The fluorescence intensity was measured by flow cytometry and the percentage of positive cells was measured by the manufacturer's software (means ± SD, n = 3, ***P < 0.05); (B) Extracellular matrix degradation (ECM) assay on DU145 Tax-Sen cells cultured in the presence of either PBS, 10 μg/ml of DU145 Tax-Res or 10 μg/ml DU145 Tax-Sen cell derived exosomes for 24 hours. DU145 Tax-Res exosomes (10 μg/ml) were cultured together with the ECM-fluorescein for 24 hours. The regions of interest were identified and quantified as described in materials and methods. The DU145 Tax-Sen cells were co-stained with phalloidin for actin cytoskeleton and with DAPI for the nucleus, (C) The ECM degradation was quantified as described in materials and methods, (means ± SD, n = 3, ***P < 0.05); (D) Immunoblot analysis of total and phosphorylated levels of AKT and ERK1/2 in DU145 Tax-Sen cells cultured in the presence or absence of 10 μg/ml of DU145 Tax-Res cell derived exosomes for 24 hours.