miR-9, a MYC/MYCN-activated microRNA, regulates E-cadherin and cancer metastasis

Abstract

MicroRNAs (miRNAs) are increasingly implicated in regulating the malignant progression of cancer. Here we show that miR-9, which is upregulated in breast cancer cells, directly targets CDH1, the E-cadherin-encoding messenger RNA, leading to increased cell motility and invasiveness. miR-9-mediated E-cadherin downregulation results in the activation of beta-catenin signalling, which contributes to upregulated expression of the gene encoding vascular endothelial growth factor (VEGF); this leads, in turn, to increased tumour angiogenesis. Overexpression of miR-9 in otherwise non-metastatic breast tumour cells enables these cells to form pulmonary micrometastases in mice. Conversely, inhibiting miR-9 by using a 'miRNA sponge' in highly malignant cells inhibits metastasis formation. Expression of miR-9 is activated by MYC and MYCN, both of which directly bind to the mir-9-3 locus. Significantly, in human cancers, miR-9 levels correlate with MYCN amplification, tumour grade and metastatic status. These findings uncover a regulatory and signalling pathway involving a metastasis-promoting miRNA that is predicted to directly target expression of the key metastasis-suppressing protein E-cadherin.

Figure 1. miR-9 directly targets CDH1 and increases cell motility and invasiveness

(a) Left panel: predicted duplex formation between human CDH1 3′UTR and miR-9. Right panel: sequence of the miR-9 binding site within the CDH1 3′UTR of human (hs), mouse (mm), and rat (rn). (b) Phase contrast images of HMLE cells infected with the miR-9-expressing, miR-10b-expressing, or empty vector. Cells were plated on 10 cm dishes at the same density (1 × 10⁶ cells in 10 ml medium). Two days after plating, images were taken and cells were then counted (cell numbers are shown in parentheses). Magnification: x200. (c) Immunoblotting of E-cadherin and vimentin in HMLE and SUM149 cells infected with the miR-9-expressing, miR-10b-expressing, or empty vector. (d) Luciferase activity of the wild-type or mutant CDH1 3′UTR reporter gene in SUM149 cells infected with the miR-9-expressing or empty vector. (e, f) Transwell migration assay and Matrigel invasion assay of miR-9-transduced or mock-infected HMLE (e) and SUM149 (f) cells with or without ectopic expression of E-cadherin. A representative experiment is shown in triplicate along with s.e.m. in d–f.

Figure 2. miR-9 increases VEGF levels in an E-cadherin- and β-catenin-dependent manner

(a) Immunofluorescence staining of β-catenin (red) in mock-infected or miR-9-expressing SUM149 cells demonstrates differential localization. Right panels are the overlay of β-catenin and nuclear 4′,6-diamidino-2-phenylindole (DAPI; blue) staining of the same field. White arrows indicate cells positive for nuclear β-catenin. Insets: blow-up images of particular cells. Magnification: x200. (b) Immunoblotting of phospho-β-catenin (Ser33/37/Thr41, GSK-3β phosphorylation sites) and β-catenin in SUM149 cells infected with the miR-9-expressing or empty vector. (c) Topflash reporter assay in HMLE, SUM149, and SUM159 cells infected with the miR-9-expressing or empty vector. (d) Real-time RT-PCR of total VEGFA mRNA in SUM149 and SUM159 cells infected with the miR-9-expressing or empty vector. (e) Immunoblotting of E-cadherin and β-catenin in SUM149 cells infected with E-cadherin siRNA (si-Ecad) or ΔN90β-catenin (ΔN90), and in SUM149-miR-9 cells infected with E-cadherin (Ecad) or β-catenin siRNA (si-βcat). Vec: the pLKO-puro vector with a scrambled sequence that does not target any mRNA. (f) Real-time RT-PCR of total VEGFA mRNA in the same cells described in e. A representative experiment is shown in triplicate along with s.e.m. in c, d and f. (g–i) Plasma levels of VEGF (g), primary tumor weight (h), and normalized VEGF levels (i) in mice that received orthotopic injection of miR-9-transduced or mock-infected SUM149 cells, at week 12 after transplantation. Data are presented as mean ± s.e.m. (n = 8 mice per group).

Figure 3. miR-9 induces angiogenesis, mesenchymal marker expression, and metastasis of the SUM149 epithelial tumors

(a, b) Left panels: Ki-67- (a) and MECA-32-stained sections (b) of primary mammary tumors formed by mock-infected or miR-9-transduced SUM149 cells, at week 12 after orthotopic transplantation. The circle and arrow indicate pyknotic nuclei. Magnification: x200 for Ki-67; x400 for MECA-32. Right panels: counting of Ki-67-positive cells (a) and intratumoral vessels (the number of microvessels per field, b). Data are presented as mean ± s.e.m. (we counted three fields per section and analyzed four mice per group). (c) Human-specific vimentin staining of primary mammary tumors formed by mock-infected or miR-9-transduced SUM149 cells, at week 12 after orthotopic transplantation. Both intratumoral regions (center) and tumor-stroma interfaces (edge) are shown. Magnification: x200. n = 3 mice per group were analyzed for vimentin. (d) Numbers of lung micrometastases per section in individual mice that received orthotopic injection of miR-9-transduced or mock-infected SUM149 cells, at week 12 after transplantation. Data are presented as mean ± s.e.m. (each data point represents a different mouse; n = 5 mice per group). (e, f) H&E- (e) and AE1/AE3 (a cocktail of two distinct anti-cytokeratin monoclonal antibodies, f)-stained sections of lungs isolated from mice that received orthotopic injection of miR-9-transduced or mock-infected SUM149 cells, at week 12 after transplantation. Circles indicate clusters of micrometastatic cells. Arrows indicate normal bronchial epithelium. Magnification: x200 in left columns; x600 in right columns.

Figure 4. Inhibiting miR-9 suppresses metastasis

(a) Enhanced activity of miR-9-regulated reporter by infection of 4T1 cells with the miR-9 sponge. A representative experiment is shown in triplicate along with s.e.m. (b) Primary tumor weight in Balb/c mice that received orthotopic injection of 4T1 cells infected with the miR-9 sponge or control sponge, at 4 weeks after transplantation. (c) Bright field imaging and H&E staining of lungs isolated from mice that received orthotopic injection of 4T1 cells infected with the miR-9 sponge or control sponge, at 4 weeks after transplantation. Arrows indicate metastatic nodules. Magnification: x8 for bright field imaging; x40 for H&E staining. (d, e) Numbers of visible lung metastases (d) and metastasis index (= number of metastases/primary tumor weight, e) in mice that received orthotopic injection of 4T1 cells infected with the miR-9 or control sponge, at 4 weeks after transplantation. Data in b, d and e are presented as mean ± s.e.m. (n = 8 mice per group).

Figure 5. miR-9 expression is activated by MYC/MYCN

(a) Real-time RT-PCR of miR-9 in HMLE, MCF7, MCF7-RAS, and MDA-MB-231 cells. (b) Real-time RT-PCR of MYC, mature miR-9, mir-9-1, mir-9-2, and mir-9-3 in mock-infected or MYC-transduced HMLE cells. (c) Mature miRNA expression in SH-EP-MYCN-ER cells upon MYCN induction with 4-OHT, 48 h after treatment. MiRNA expression values were rescaled relative to the control (no 4-OHT). The miR-17-92 expression value represents the expression of miR-20a, a miRNA residing within the miR-17-92 cluster representative for miR-17-92 expression. Data in a–c are presented as mean ± s.e.m. of triplate samples. (d) ChIP-on-chip data showing occupancy of the mir-9-3 genomic sequence by MYCN and MYC in Kelly (MYCN-amplified) and SJ-NB-12 (MYC-amplified) neuroblastoma cells, respectively. The genomic positions for probes and their enrichment ratios are displayed on the X and Y axes, respectively. Smoothed local enrichment ratios are given for either MYCN or MYC at the mir-9-3 locus in Kelly and SJ-NB-12 cells. The red line indicates median enrichment ratio for MYCN or MYC versus input as calculated of all probes for chromosome 15.

Figure 6. miR-9 levels correlate with MYCN amplification and metastatic status in human cancers

(a) Expression of mature miR-9 in MYCN-normal copy (n = 22) and MYCN-amplified (n = 23) neuroblastoma tumor samples. (b) Expression of mature miR-9 in primary breast tumor samples from metastasis-free and metastasis-positive patients. Data in a and b are presented as mean ± s.e.m. (c) Model for miR-9-mediated pathway in cancer metastasis.