Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Dec 1;126(12):4585-4602.
doi: 10.1172/JCI86505. Epub 2016 Nov 7.

MicroRNA-424 Impairs Ubiquitination to Activate STAT3 and Promote Prostate Tumor Progression

Free PMC article

MicroRNA-424 Impairs Ubiquitination to Activate STAT3 and Promote Prostate Tumor Progression

Cecilia Dallavalle et al. J Clin Invest. .
Free PMC article

Abstract

Mutations and deletions in components of ubiquitin ligase complexes that lead to alterations in protein turnover are important mechanisms in driving tumorigenesis. Here we describe an alternative mechanism involving upregulation of the microRNA miR-424 that leads to impaired ubiquitination and degradation of oncogenic transcription factors in prostate cancers. We found that miR-424 targets the E3 ubiquitin ligase COP1 and identified STAT3 as a key substrate of COP1 in promoting tumorigenic and cancer stem-like properties in prostate epithelial cells. Altered protein turnover due to impaired COP1 function led to accumulation and enhanced basal and cytokine-induced activity of STAT3. We further determined that loss of the ETS factor ESE3/EHF is the initial event that triggers the deregulation of the miR-424/COP1/STAT3 axis. COP1 silencing and STAT3 activation were effectively reverted by blocking of miR-424, suggesting a possible strategy to attack this key node of tumorigenesis in ESE3/EHF-deficient tumors. These results establish miR-424 as an oncogenic effector linked to noncanonical activation of STAT3 and as a potential therapeutic target.

Conflict of interest statement

The authors have declared that no conflict of interest exists.

Figures

Figure 1
Figure 1. miR-424 is elevated in prostate tumors and is associated with EMT and stem cell–like features.
(A) Unsupervised hierarchical clustering (UHC) of primary prostate tumors from the Biella data set according to their miRNA profile. ESE3lo tumors (blue rectangles) cluster together and have similar miRNA profile. (B) miR-424 log2 intensity levels evaluated by microarray (left) and relative expression levels evaluated by qRT-PCR (right) in prostate tumors. (C) miR-424 level evaluated by qRT-PCR in indicated prostate cell lines. The results were normalized to RNU6 internal control. (D) Prostate samples divided according to miR-424 expression in 2 independent primary prostate tumor data sets, The Cancer Genome Atlas (TCGA) project and the in-house Biella data set (see Methods for details). (E) Gene set enrichment analysis (GSEA) comparing miR-424hi with miR-424lo prostate tumors in the 2 indicated microarray data sets (total = 545), using the indicated stem cell–like and EMT gene signatures. (F) Venn diagram crossing genes upregulated in miR-424hi signatures extracted from Biella and TCGA data sets and showing a significantly high convergence (P < 0.00001). (G) Functional analysis of miR-424hi signatures in Biella and TCGA data sets using ENCODE pathway analysis tools. Signatures were extracted by comparison of miR-424hi versus miR-424lo samples. See Methods for experimental details.
Figure 2
Figure 2. ESE3/EHF occupies MIR424 promoter region and represses miR-424 transcription.
(A) Schema of MIR424 promoter and position of the evaluated ETS binding site relative to the 5′ end of pri-miR-424 (RefSeq NR_029946). (B) ESE3/EHF occupancy on MIR424 promoter evaluated by ChIP and quantitative PCR in indicated cell lines. The results are represented relative to input. Bottom panels: Immunoblots of ESE3 in indicated cell lines. (C) miR-424 level in DU145 with stable ESE3/EHF (pESE3) expression compared with control cells (pcDNA). The results were normalized to RNU6 and are represented as miR-424 expression relative to pcDNA. Bottom panel: Immunoblot of ESE3. (D) miR-424 level evaluated by qRT-PCR in DU145 following transfection with pESE3 or pcDNA at the indicated time points (top), and immunoblot of ESE3/EHF (bottom). *P < 0.05; **P < 0.01.
Figure 3
Figure 3. miR-424 promotes transforming phenotypes in prostate epithelial cells.
miR-424 precursor (miR-424) or negative control (Ctrl) was transiently transfected in RWPE1 cells. (A) Colony formation in soft agar. (B and C) Cell migration by wound healing (WH) assay (B) and Boyden chamber assay (C). (D) Sphere-forming efficiency (SFE) (top) and representative images of spheroids (bottom). **P < 0.01. Scale bar: 200 µm.
Figure 4
Figure 4. miR-424 inhibition reverts transforming phenotypes in transformed cell models.
(AF) Colony formation in soft agar (A and D), WH (B and E), and SFE and representative images of spheroids (C and F) in ESE3kd cell line models following inhibition of miR-424 by anti–miR-424 (Anti-424) or scrambled control (Scr). (GJ) Colony formation in soft agar (G), WH assay (H), SFE and self-renewal potential (I), and representative images of spheroids (I, right). (J) Evaluation of CD44hiCD24lo fraction by FACS following inhibition of miR-424 in DU145. Data show mean ± SD (n = 3) of 1 representative experiment. *P ≤ 0.05, **P ≤ 0.01 by 2-tailed Student’s t test. Scale bars: 200 μm.
Figure 5
Figure 5. miR-424 targets the E3 ubiquitin ligase COP1, leading to multiple oncogenic TF activation.
(A) Venn diagram of predicted targets (miRWalk) and genes repressed by miR-424 in RWPE1 cells. (B) Functional annotation analysis by DAVID of the putative miR-424 target genes. (C) Level of COP1 by qRT-PCR (top) and immunoblotting (bottom) in prostate cell lines. (D) Diagram of 3′-UTR COP1 reporter construct with WT and mutated (MUT) sequence of the miR-424 binding site (left) and relative luciferase activity (RLA) following transfection of miR-424 in LNCaP cells (right). (E) COP1, c-Jun, and ETV1 expression in RWPE1 cells following transient transfection of miR-424 and LNCaP cells stably overexpressing miR-424. Replicate samples were run on separate gels for blotting with the different antibodies. (F) COP1, c-Jun, and ETV1 expression following transfection of siRNAs targeting COP1 (siCOP1#1 and #2) or control siRNA (siGL3) in RWPE1 and LNCaP cells. (G) COP1, c-Jun, and ETV1 expression following transfection of Anti-424 or Scr in RWPE-ESE3kd and DU145 cells. (HK) Immunoblot of COP1 (H), colony formation in soft agar (I), cell migration by WH (J), and SFE and representative images of spheroids (K), following knockdown of COP1 by siRNA (siCOP1#1) and control (siGL3) in RWPE1 cells. Data show mean ± SD of 3 independent experiments. *P < 0.05; **P < 0.01. Scale bar: 200 µm.
Figure 6
Figure 6. miR-424 induces STAT3 stability and activity by targeting COP1.
(A) Pie chart showing the number of STAT3 targets among the genes induced by miR-424 extracted with Enrich (A, top) and functional annotation analysis by DAVID (A, bottom) of genes induced by miR-424 in RWPE1 cells. (B) COP1, STAT3, and p-STAT3 level (left) and relative luciferase activity (RLA) of STAT3 reporter (right) following transfection of miR-424 or control (Ctrl) in RWPE1 cells. (C) Immunoblot (IB) of COP1, STAT3, and p-STAT3 in indicated cell lines. (D) STAT3 level in LNCaP-424 stable cells by IB (bottom) and immunocytochemistry (ICC; top). (E) IB of COP1 and STAT3 in LNCaP cells after transient transfection of miR-424 or control (Ctrl) at the indicated time points. (F) IB of COP1, STAT3, and p-STAT3 in control (EV) and LNCaP-424 monoclonal cells with or without IL-6 (10 ng/ml) stimulation. (G) RLA of STAT3 reporter in cells transfected with negative control (Ctrl) or miR-424 with or without IL-6 stimulation. (H) IB of COP1, STAT3, and p-STAT3 (left) and RLA of STAT3 reporter (right) in DU145 cells following transfection with Anti-424 or Scr. (I and J) IB of COP1, STAT3, and p-STAT3 in RWPE1 (I) and LNCaP (J) cells after COP1 silencing. (K) IB of COP1, STAT3, and p-STAT3 in DU145 cells after transfection with full-length FLAG-COP1 (p-COP1) or empty vector (EV). (L) IB of DU145 cells transfected with p-COP1 with or without treatment with proteasome inhibitor PS-341 (10 μM) for 5 hours. c-Jun used as control. (M) IB of STAT3 and p-STAT3 in DU145 transfected with EV, p-COP1, or a deletion mutant (pRING). *P ≤ 0.05, **P ≤ 0.01 by 2-tailed Student’s t test. The experiments were performed in triplicate, and data are shown as mean ± SD.
Figure 7
Figure 7. COP1 interacts with STAT3 and mediates STAT3 ubiquitylation and degradation.
(A) Immunoprecipitation of STAT3 and IB of COP1 and STAT3 (B) Immunoprecipitation of COP1 (p-COP1) with anti-FLAG and IB with anti-STAT3 in cells treated with PS-341 (10 μM for 6 hours). Asterisk: aspecific band. TUB, tubulin. (C) IB of STAT3 and COP1 in HEK293T cells transfected with indicated constructs. IP of COP1 with anti-FLAG and IB using indicated antibodies. Cells were treated with IL-6 (10 ng/ml for 45 minutes) and with 10 μM PS-341 for 3 hours. (D) STAT3 ubiquitination in RWPE1 cells transfected with siCOP1#1 or siGL3 for 48 hours. Lysates were immunoprecipitated with anti-STAT3 or IgG followed by IB with indicated antibodies. UBI,ubiquitin. (E) STAT3 ubiquitination in DU145 cells transfected for 48 hours and treated with PS-341 (10 μM for 3 hours). Lysates were immunoprecipitated with anti-STAT3 or IgG followed by IB with indicated antibodies. (F) IB of STAT3 and p-STAT3 following sequential transfection of miR-424 or control and p-COP1, the mutant construct (pRING), or empty vector (EV). (G) SFE in cells transfected as described in F. (H and I) RLA of STAT3 reporter in RWPE1 cells following transfection of siCOP1#1 or negative control (NC) (H) and in DU145 cells transfected with p-COP1 or EV (I). (J) STAT3 reporter activity following sequential transfection of miR-424 or control and p-COP1, mutant construct (pRING), or EV. (K) STAT3 reporter activity following transfection with siCOP1#1 or NC with or without IL-6. (L) IB of indicated proteins following IL-6 with and without COP1 knockdown. (M) STAT3 reporter activity following transfection of miR-424 or control and pCOP1, pRING, or EV in presence of IL-6. *P ≤ 0.05, **P ≤ 0.01 by 2-tailed Student’s t test. The experiments were performed in triplicate, and data are shown as mean ± SD.
Figure 8
Figure 8. miR-424 promotes tumor initiation recapitulating in vivo miR-424/COP1/STAT3 axis.
(AC) IB of COP1, STAT3, and p-STAT3 (A), colony formation in soft agar (B), and SFE (C) in RWPE1 cells following cotransfection with miR-424 or Ctr and siRNA targeting STAT3 (siSTAT3) or control siRNA (siGL3). (DF) IB of COP1, STAT3, and p-STAT3 (D), colony formation in soft agar (E), and SFE (F) in RWPE1 cells 48 hours after transfection with miR-424 or Ctr and treatment with 5 μM NVP or DMSO for 16 hours. (G and H) Tumor growth (G), tumor weight (H), and representative images of tumors (bottom) of subcutaneous xenografts in nude mice, from RWPE1 cells transfected with miR-424 (n = 4) or Ctr (n = 4). (I) H&E and IHC stain and scores (percentage of positive cells) of STAT3, p-STAT3, COP1, and Ki67 stain of control and miR-424 induced xenografts described above. Scale bars: 10 μm. (J) Schematic of the experimental plan (left) and tumor volume (right). *P < 0.05; **P < 0.01.
Figure 9
Figure 9. miR-424 inhibition disrupts the miR-424/COP1/STAT3 axis in vivo.
(A) Schematic of the experimental plan. (B) Tumor incidence in control-treated (Scr n = 10) and anti–miR-424–treated (Anti-424 n = 9) DU145 cells. Significance was calculated by Fisher exact test. Two-tailed P = 0.023. (C and D) Tumor growth monitored by tumor size (C) and in vivo bioluminescence (D) of subcutaneous xenografts (mean ± SEM, Scr n = 6 and Anti-424 n = 2) of DU145-Luc cells pretreated in vitro with anti–miR-424 or Scr. (E) Ex vivo SFE from Scr or Anti-424 dissociated tumor xenografts. (F) IHC staining for STAT3, p-STAT3, COP1, and Ki67 in tumor xenografts of DU145 cells treated with Anti-424 or Scr. Scale bars: 10 μm. (G) qRT-PCR evaluation of selected STAT3 targets (top) and selected cancer stem cell markers (bottom) in tumor xenografts from control (Scr) and miR-424–ablated (Anti-424) xenografts. (H) Schematic of the experimental plan. (I) Lung metastasis from DU145-Luc cells pretreated in vitro with Anti-424 or Scr and injected into the tail vein of nude mice. Representative bioluminescent images (left); quantification after necropsy of lung metastatic foci (right). (J) Representative images of H&E, immunostaining, and IHC scores (percentage of positive cells). Areas within the boxes in top panels are magnified in the lower samples. Scale bars: 100 μm. *P < 0.05; **P < 0.01.
Figure 10
Figure 10. Prognostic impact of COP1 loss and STAT3 activation in prostate tumors.
(A) Representative IHC images of total STAT3 in prostate tumors (left) and STAT3/miR-424 high/low association in prostate tumors (right). The t test P value is shown. Scale bars: 100 μm. (B) IHC score of COP1 and STAT3 and p-STAT3 Tyr705 in TMA of primary prostate tumors (n = 136) (top) and representative IHC images (bottom). Scale bars: 100 μm. (C) STAT3/COP1 and COP1-negative/STAT3/p-STAT3 association analysis in the TMA. The t test P value is shown. (D) Association between p-STAT3 Tyr705–negative and –positive tumors and COP1 and STAT3 staining (t test P < 0.00001). (E) Survival plots showing significant increase in biochemical recurrence in COP1-negative/STAT3-high tumors compared with the other tumors in the TMA cohort analyzed above. Log-rank test P value is shown.
Figure 11
Figure 11. Proposed model for induction of miR-424 and establishment of the miR-424/COP1/STAT3 oncogenic axis.

Similar articles

See all similar articles

Cited by 22 articles

See all "Cited by" articles

MeSH terms

Feedback