microRNA-1246 Is an Exosomal Biomarker for Aggressive Prostate Cancer

Abstract

Because of high heterogeneity, molecular characterization of prostate cancer based on biopsy sampling is often challenging. Hence, a minimally invasive method to determine the molecular imprints of a patient's tumor for risk stratification would be advantageous. In this study, we employ a novel, digital amplification-free quantification method using the nCounter technology (NanoString Technologies) to profile exosomal serum miRNAs (ex-miRNA) from aggressive prostate cancer cases, benign prostatic hyperplasia, and disease-free controls. We identified several dysregulated miRNAs, one of which was the tumor suppressor miR-1246. miR-1246 was downregulated in prostate cancer clinical tissues and cell lines and was selectively released into exosomes. Overexpression of miR-1246 in a prostate cancer cell line significantly inhibited xenograft tumor growth in vivo and increased apoptosis and decreased proliferation, invasiveness, and migration in vitro miR-1246 inhibited N-cadherin and vimentin activities, thereby inhibiting epithelial-mesenchymal transition. Ex-miR-1246 expression correlated with increasing pathologic grade, positive metastasis, and poor prognosis. Our analyses suggest ex-miR-1246 as a promising prostate cancer biomarker with diagnostic potential that can predict disease aggressiveness.Significance: Dysregulation of exosomal miRNAs in aggressive prostate cancer leads to alteration of key signaling pathways associated with metastatic prostate cancer. Cancer Res; 78(7); 1833-44. ©2018 AACR.

Fig. 1. Dysregulated exosomal microRNAs in aggressive PCa

A. MicroRNA profiling was performed using Nanostring miRNA (v3a) platform in aggressive prostate tumors (n=6), normal individuals (n=3) and cases with BPH (n=3). Significantly dysregulated miRNAs are represented in the heatmap. Top 100 genes and four spike in controls were used for normalization. B. Venn diagram showing dysregulated miRNAs in aggressive PCa as compared to BPH and normal controls. C. Grouped data analyses for ex-miR-1246 signal intensities using Nanostring miRNA platform.

Fig. 2. Ex-miR-1246 is specifically upregulated in aggressive PCa and is a potential diagnostic marker

A. Relative miR-1246 expression in exosomes derived from sera of PCa patients (n=44) as compared to normal individuals (n=8) as assessed by real-time PCR. RNU6A was used as a control. Error bars represent SEM. B. Relative ex-miR-1246 expression in BPH and PCa cases. Horizontal lines represent the average in each group. C. ROC curve analyses for ex-miR-1246 (left panel) and serum PSA (right panel) as parameters to discriminate between tumor and normal samples.

Fig. 3. High expression of serum exosomal miR-1246 is associated with advanced pathological stage and localized PCa metastasis

A. Relative ex-miR-1246 expression in exosomes derived from the sera of PCa patient training cohort (cohort 2) as assessed by real-time PCR. RNU6A was used as an endogenous control. Patients with BPH (n=21) were used as calibrators. B. Correlation of serum exosomal miR-1246 expression with clinicopathological parameters in PCa patients. P-values are based on Chi square test. C. Correlation of serum exosomal miR-1246 expression with prognostic risk groups IIA-IV as defined by American Joint Committee on Cancer. P-values are based on Chi square test. D. ROC curve analyses for ex-miR-1246 expression (left panel) and PSA (right panel) as parameters to discriminate between non-metastatic and localized metastatic PCa cases Analyses was based on dCT values of ex-miR-1246 in training cohort (1 and 2, n=77) that included lymph node metastatic PCa (n=26), non-metastatic PCa (n=43) and normal (n=8) individuals.

Fig. 4. Validation of ex-miR-1246 as a PCa exosomal biomarker for aggressive PCa

A. Relative ex-miR-1246 expression in exosomes derived from sera of validation cohort of aggressive PCa patients as assessed by real-time PCR. RNU6A was used as an endogenous control. Normals (n=7) were used as calibrators. B. ROC curve analyses for ex-miR-1246 (left panel) and serum PSA (right panel) as parameters to discriminate between tumor and aggressive, metastatic PCa. C. Correlation of serum exosomal miR-1246 expression with clinicopathological parameters in validation cohort. P-values are based on Chi square test.

Fig. 5. miR-1246 is a tumor suppressor miRNA that is downregulated in PCa clinical tissues and is selectively released in exosomes

A. Relative miR-1246 expression in microdissected PCa tissues (n=36) and matched adjacent normal regions as assessed by real-time PCR. Data were normalized to RNU48 control and represented as mean ± SEM. B. Relative miR-1246 expression in prostate cell lines as assessed by real-time PCR. Data were normalized to RNU48 control and represented as mean ± SEM. C. PCa cell lines PC3 (left panels) and LNCaP (right panels) were treated with GW4869 for 48 hrs followed by real time PCR analyses of miR-1246 expression in cellular and exosomal fractions. Data were normalized to RNU6A control and represented as mean ± SEM. D. Relative miR-1246 expression in PCa tissues and corresponding serum exosomes from the same patients in a training cohort subset. Data were normalized to RNU6A for exosomes and RNU48 for tissues and represented as mean ± SEM.

Fig. 6. miR-1246 is a tumor suppressor miRNA with pleiotropic roles in PCa

A. Relative miR-1246 expression levels in PC3 cells stably transfected with control miR/miR-1246 expression construct (clone-1 and -2) as assessed by RT-PCR. Data were normalized to RNU48 control. B. Control miR or miR-1246 expressing cells (clone 1 or 2) were subcutaneously injected into the right or left flank of nude mice (n=6 for control group, n=5 each for two miR-1246 clones) to generate PCa xenograft tumors. Images of mice and extracted tumors from miR-CON/ miR-1246 groups at 5 weeks are shown. C. Tumor volumes of xenograft tumors from miR-CON and miR-1246 groups, at the indicated time points. Data represent the mean of each group ±SD. D. Real-time PCR analyses of miR-1246 expression in xenograft tumors (left panels) and corresponding serum exosomes (right panels) in control/ miR-1246 PCa xenografts (1–4). Data were normalized to RNU6A control. E. Relative miR-1246 expression in PC3 cells transiently transfected with either miR-CON/miR-1246 or mock transfected cells as assessed by real time PCR. Data were normalized to RNU48 control and are represented as mean ± SEM. F. Colony formation assay in mock/miR-CON/miR-1246 transfected PC3 cell line. Representative pictures from miR-CON/miR-1246 transfected cells are shown. G. Matrigel assay using mock/miR-CON/miR-1246 transfected PC3 cells. Representative pictures from miR-CON/miR-1246 transfected cells are shown above. H. Apoptosis assay in PC3 cells after miR-CON/ miR-1246 transfections as assessed by ANNEXIN V-FITC/7-AAD staining. Right panels: Real time PCR and immunoblot analyses of TRAIL in transfected PC3 cells. RNU48 and GAPDH were used as respective controls. I. Transwell migration and invasion assay in PC3 cells transfected with miR-CON/ miR-1246. Representative pictures are shown in left panels (* P< .05).

Fig. 7. miR-1246 inhibits EMT by direct repression of mesenchymal genes in PCa

A. Schematic representation of the CDH2, VIM and ZEB1 3’-UTRs showing the putative miR-1246 binding sites. Mutant 3’ UTRs used in luciferase reporter assays are represented below. B. Real-time PCR analyses of relative transcript levels of CDH2, VIM and ZEB1 in PC3 cells transfected as indicated. Data were normalized to GAPDH. C. Luciferase reporter assays with the indicated wt and mutated 3’ UTR constructs or control luciferase construct co-transfected with miR-CON/ miR-1246 in PC3 cells. Firefly luciferase values were normalized to Renilla luciferase activity and plotted as relative luciferase activity to control 3’ UTR construct (* P< .05). D. Immunoblots of endogenous N-cadherin and Vimentin in PC3 cells stably transfected with miR-CON/miR-1246. E. Immunoblot analyses of EGFR and phospho Akt in PC3 cells stably expressing control miRNA/miR-1246. GAPDH was used a loading control for D and E.