MiR-148a inhibits angiogenesis by targeting ERBB3

MicroRNAs (miRNAs) play an important role in carcinogenesis in various solid cancers including breast cancer. Down-regulation of microRNA-148a (miR-148a) has been reported in certain cancer types. However, the biological role of miR-148a and its related targets in breast cancer are unknown yet. In this study, we showed that the level of miR-148a was lower in MCF7 cells than that in MCF10A cells. V-erb-b2 erythroblastic leukemia viral oncogene homolog 3 (ERBB3) is a direct target of miR-148a in human breast cancer cells through direct binding of miR-148a to ERBB3 3′-UTR region. Overexpression of miR-148a in MCF7 cells inhibited ERBB3 expression, blocked the downstream pathway activation including activation of AKT, ERK1/2, and p70S6K1, and decreased HIF-1α expression. Furthermore, forced expression of miR-148a attenuated tumor angiogenesis in vivo. Our results identify ERBB3 as a direct target of miR-148a, and provide direct evidence that miR-148a inhibits tumor angiogenesis through ERBB3 and its downstream signaling molecules. This information would be helpful for targeting the miR-148a/ERBB3 pathway for breast cancer prevention and treatment in the future.


INTRODUCTION
Breast cancer is the most common cancer in women with more than 400,000 deaths each year worldwide. MicroRNAs (miRNAs) are a class of small noncoding RNAs that have been identified as a new kind of gene expression regulators through targeting the 3'-untranslated region (UTR) of mRNAs for translational repression or degradation [1,2] . MiRNAs are differentially expressed in various human cancers, functioning either as oncogenes or tumor suppressors by controlling the expression of their target genes [3,4] . Some miRNAs were related to cancer biopathological features like vascular invasion, lymph node metastasis, tumor stage and cell proliferation [3][4][5] . Recent studies have shown that reduced expression of miR-148a occurs in chronic lymphocytic leukemia, gastrointestinal cancer, and esophageal carcinoma [5][6][7][8] and is correlated with the outcome of certain malignancies [8,9] . The hypermethylation-associated miR-148a silencing has also been reported to be correlated with human cancer metastasis [10] . Overexpression of miR-148a attenuated paclitaxel resistance of hormone-refractory, drug-resistant prostate cancer PC3 cells by regulating mitogen-and stress-activated kinase 1 (MSK1) expression [11] . Known direct targets of miR-148a in-clude transcription growth factor-β-induced factor 2 (TGIF2), DNA (cytosine-5-)-methyltransferase 3β (DNMT3b), pregnane X receptor (PXR), neddylationdissociated 1, and DNA methyltransferase-1 [10,[12][13][14][15] . However, the biological roles of miR-148a and the target genes in breast cancer have not yet been defined.
V-erb-b2 erythroblastic leukemia viral oncogene homolog 3 (ERBB3) is a member of the epidermal growth factor receptor (EGFR) family, which consists of EGFR/ERBB1, ERBB2/Neu/HER2, ERBB3, and ERBB4. Since ERBB3 lacks intrinsic kinase activity, signal transduction occurs through formation of heterodimers with ERBB2 and other members. The binding of peptides of the EGF-related growth factor family to the extracellular domain of the ERBB receptors causes the formation of homo-and heterodimers. Ligand binding induces intrinsic receptor kinase activity and then stimulates intracellular signaling cascades, including the phosphatidylinositol-3 kinase (PI3K)/ AKT and MAPKs cascade [16,17] . Elevated expression of ERBB3 is frequently observed in breast cancer, which may play an important role in breast cancer progression and chemotherapy resistance [18,19] . Recent studies have demonstrated that miR-125 and miR-205 act as tumor suppressors through directly targeting ERBB3 [20,21] .
Tumor angiogenesis is required for tumor development and growth in which hypoxia-inducible factor 1α (HIF-1α) plays a pivotal role [22,23] . In the present study, we want to identify novel target(s) of miR-148a, which may be related to tumor angiogenesis, identify signaling pathways that are regulated by miR-148a, and determine the direct role of miR-148a in angiogenesis. This will provide direct evidence for the role and potential mechanism of miR-148a in regulating breast tumor angiogenesis.

Lentivirus packaging and stable cell lines establishment
The lentiviral vectors with RFP tag carrying scrambled miRNA (pLe-miR-SCR), miR-148a (pLe-miR-148a), and packaging kit were purchased from Thermo Scientific (Huntsville, AL, USA). Lentiviral packaging was performed using HEK293T cells by following the manufacturer's manual. MCF7 cells were infected with lentiviruses, and selected by puromycin to obtain stable cell lines expressing miR-SCR and miR-148a.

Transient transfection and luciferase assay
Reporter plasmids containing ERBB3 3' UTR region with miR-148a binding site (wild type) or mutation binding site (mutant) were cloned into pMIR-REPORTER. HEK293 cells were seeded in 24-well plates and cultured overnight. To investigate whether miR-148a directly regulates ERBB3 transcriptional expression by binding to the miR-148a binding site of ERBB3, pre-miR-148a or pre-miR-SCR was cotransfected with plasmid pGL4.74 expressing renilla luciferase and the ERBB3 wild type or mutant reporter plasmid. After cultured for 48 h, cells were lysed with passive lysis buffer (Promega, WI, USA), renilla and firefly luciferase activities were measured using the dual luciferase assay system (Promega WI, USA). ERBB3 transcriptional activity was normalized to renilla luciferase activity, which was an internal control for transfection efficiency. Each experiment was repeated at least three times.

Angiogenesis assay on chicken chorioallantoic membrane (CAM)
Fertilized chicken eggs were purchased from SPA-FAS (Preston, CT, USA), and incubated at 37°C with 70% humidity. On day 8, an artificial air sac was created and a small window in the shell over the artificial air sac was cut as we described [37] . MCF7-miR-148a or MCF7-miR-SCR stable cells were resuspended in serum-free medium, and mixed with equal volume of Matrigel. Then, aliquots (3×10 6 , 40 μL) of the mixture were applied onto the CAM of 9-d-old embryos. The area around the implanted Matrigel was photographed 5 d after the implantation, and the number of blood vessels was obtained by counting the branching of blood vessels. Assays for each treatment were carried out using 8 chicken embryos.

Cell proliferation assay
To investigate the effects of miR-148a on the cell growth of MCF7 cells, the stable cell MCF7-miR-SCR and MCF7-miR-148a were seeded in a 96-well plate (1,000 cells per well) and incubated at 37°C in 5% CO 2 incubator. The proliferation of the cells was measured using a proliferation assay kit, CCK8 kit (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's instruction. Results were obtained from three independent experiments.

Statistical analysis
All values in the present study were reported as mean ±SE. Student's unpaired t test was used for statistical analyses. Differences between values were considered significant at P < 0.05.

MiR-148a expression in breast cancer cells
To determine whether the expression of miR-148a is downregulated in breast cancer cells, human breast cancer cell lines MCF7 and T47D and the immortalized normal breast epithelial cells MCF10A were used to test the levels of miR-148a by TaqMan realtime RT-PCR analysis. The results showed that miR-148a expression was significantly decreased in MCF7 and T47D cells (P < 0.05, Fig. 1A). To investigate the role of miR-148a in breast cancer, MCF7 cells were infected with lentiviruses expressing miR-148a and miR-SCR control, and selected by puromycin to obtain stable cell lines. We established stable cells with fluorescent and contrast phase representative pictures of MCF7-miR-SCR and MCF7-miR-148a cells (Fig. 1B). Compared to MCF7-miR-SCR, the miR-148a level in MCF7-miR-148a was much higher (Fig.1C). These results indicated that we successfully established the stable breast cancer cell lines expressing miR-SCR and miR-148a, respectively.

ERBB3 acting as a direct target of miR-148a
To determine the direct target gene(s) of miR-148a, we identified the potential binding sites of miR-148a using the combination of PicTar, TargetScan and FindTar software, and found that ERBB3 was one of the predicted targets that contain putative conserved miR-148a binding sites within its 3'UTR. The binding site of miR-148a in ERBB3 3'-UTR region is highly conserved among several different species (Fig. 2A). To test whether miR-148a regulates ERBB3 expression at protein level, total proteins prepared from MCF7-miR-148a cells and MCF7-miR-SCR cells were analyzed by Western blotting. Overexpression of miR-148a greatly decreased ERBB3 expression in MCF7-miR-148a cells (Fig. 2B), suggesting that ERBB3 is a downstream target of miR-148a. To further determine whether ERBB3 is a direct target of miR-148a, we constructed luciferase reporters containing ERBB3 3'-UTR regions and the predicted binding site of miR-148a (WT, Fig. 3A), or the predicted binding site with the mutation in the seed sequence (Mut, Fig. 3A). MCF7 cells were transfected with the ERBB3 reporter plasmids in the absence or presence of lentivirus plasmids carrying miR-148a and miR-SCR. We found that miR-148a significantly inhibited ERBB3 wild type reporter activities, but not the mutant reporter activity (Fig. 3B). This result shows that miR-148a inhibits ERBB3 wild type reporter activities through the binding of miR-148a in the seed region. These results demonstrate that ERBB3 is a direct target of miR-148a in breast cancer cells. Fig. 2 MiR-148a targeted to ERBB3. A: Putative binding site of ERBB3 3'-UTR by miR-148a. The seed-matching sites were predicted by PicTar, TargetScan and FindTar software, and were marked in red. DNA sequences of potential miR-148a binding site predicted within the human ERBB3 3'-UTR were also highly conserved among other species. B: MCF7 cells expressing scramble miRNA precursor (SCR) or miR-148a were used for Western blot analysis to determine the levels of ERBB3 expression with GAPDH as an internal control. MiR-148a inhibited the activation of AKT, ERK, and p70S6K1 and decreased HIF-1α expression.
AKT and ERK signaling pathways are two key downstream pathways of ERBB3, and play important Fig. 3 MiR-148a interacted with ERBB3 3'-UTR to inhibit its transcriptional activation. A: Schematic illustration of ERBB3 3'-UTR wild type (WT) and mutant reporter constructs with the seed-matching sites marked in red. B: MCF7 cells were transfected with ERBB3 3'-UTR wild type (WT) and mutant reporter constructs in the presence of scramble miRNA precursor (SCR) or miR-148a, and the cells were cultured for 36 h after the transfection. Relative luciferase activities were measured and calculated as the ratio of firefly/renilla activities in the cells, and normalized to those of the SCR control. The results were presented as means±SE from three independent experiments, * P < 0.05. MiR-148a U6 roles in angiogenesis and cancer development [24,25] . To investigate the effect of miR-148a on their activation, we showed that levels of phospho-AKT (p-AKT) and phospho-ERK1/2 (p-ERK1/2) were greatly suppressed in the cells expressing miR-148a (Figs. 4A to 4D), indicating that miR-148a inhibited ERBB3 expression with functional role by decreasing downstream AKT

Overexpression of miR-148a inhibited tumor angiogenesis in vivo
In order to study whether overexpression of miR-148a inhibits tumor angiogenesis, MCF7 cells expressing miR-148 and miR-SCR were resuspended in serum-free medium and used for angiogenesis assay on the chicken CAM of 9-d-old chicken embryos. The angiogenesis responses were analyzed on the CAM 4 d after the implantation. MCF7 breast cancer cells induced the angiogenesis responses 4-fold higher than the normal angiogenesis in the CAM, and the forced expression of miR-148a significantly inhibited cancer cell-inducing angiogenesis responses by 45% (Fig.  5B). Meanwhile, cell proliferation assay indicated that the growth rate of MCF7-miR-SCR was similar to that of MCF7-miR-148a, indicating that the angiogenesis response was not due to the effect of miR-148a overexpression on cell proliferation (Fig. 5C). These results suggest that overexpression of miR-148a specifically inhibits angiogenesis.

DISCUSSION
MiRNAs act as tumor suppressors or oncogenes in a variety of cancers including breast cancer. Angiogenesis is the process by which new blood capillaries are generated from the pre-existing vasculature. Angiogenesis is important for tumor growth, development, and metastasis since tumors cannot grow larger than 1-2 mm in diameter without an efficient blood supply [26] . Growing evidence demonstrates that miRNAs such as miR-378, miR-296, and the miR-17~92 cluster can regulate angiogenesis [27][28][29] . However, the role of miR-148a in tumor angiogenesis is unknown. ERBB3 is involved in regulating cancer initiation, tumor metastasis and angiogenesis and is important in breast cancer development [30,31] . In addition, ERBB3 is involved in acquired resistance to chemotherapy [32,33] . In β-actin Fig. 4 MiR-148a inhibited AKT and ERK activation, and HIF-1α expression. Stable breast cancer cells expressing scramble miRNA precursor (SCR) or miR-148a were obtained by puromycin selection. A:The protein levels of ERBB3, β-actin, p-AKT, total AKT, p-ERK1/2, and total ERK were analyzed by Western blotting. The relative densities of total ERBB3 (B), p-AKT (C), and p-ERK1/2 (D) levels were quantified and analyzed by Image J software from three replicate experiments, * P < 0.05. E: The protein levels of p-p70S6K1, total p70S6K1, HIF-1α, and β-actin were analyzed by Western blotting. The relative densities of p-p70S6K1 (F) and HIF-1α (G) levels were quantified and analyzed by Image J software from three independent experiments, * P < 0.05. pLe-SCR pLe-miR-148a * and ERK activation. Furthermore, miR-148a overexpression inhibited p70S6K1 activation and HIF-1α expression (Fig. 4E to 4G). Given the pivotal roles of these ERBB3 downstream molecules in regulating tumor growth and angiogenesis, these results indicate that miR-148 may inhibit tumor angiogenesis via targeting ERBB3 and its downstream molecules.
this study, we demonstrated that the level of miR-148a in breast cancer cells MCF7 was lower than that in immortalized normal breast epithelial cells MCF10A, and ERBB3 is a direct target of miR-148a. This data show that miR-148a is a new regulator of ERBB3 in cancer cells, and the downregulation of miR-148a is a novel molecular mechanism for breast cancer development.
ERBB3 can heterodimerize with ERBB2 to activate AKT and ERK signaling pathways in different cells. Consistent with ERBB3 inhibition, forced expression of miR-148a in breast cancer cells inhibited the activation of ERBB3 downstream molecules including AKT, ERK1/2, and p70S6K1. MiR-148a also suppressed HIF-1α expression, which is a rate-limiting subunit for forming functional HIF-1 transcription factor. HIF-1 regulates the expression of VEGF and other angiogenesis regulators [34] . Thus, we hypothesize that miR-148a inhibits breast tumor angiogenesis. To support this hypothesis, we demonstrated that miR-148a overexpression attenuated breast tumor angiogenesis induced by MCF7 cells. In summary, this study demonstrates that miR-148 directly targets ERBB3 through the seed sequence binding site at its 3'-UTR. Reduced expression of miR-148a in breast cancer cells leads to the elevation of ERBB3 expres-sion to activate AKT and ERK1/2 signaling pathways, which may in turn increase p70S6K1 activation and HIF-1α expression. miR-148a attenuates angiogenesis likely through directly inhibiting ERBB3 for transmitting the signals to its downstream signaling molecules. Our findings suggest that miR-148a/ERBB3 pathway would be a promising therapeutic target for breast cancer in the future. A: MCF7 cells expressing lentivirus with scrambled miRNA precursor (miR-SCR) or miR-148a were used for angiogenesis assay in the CAM. The angiogenesis responses were analyzed 5 days after the implantation of the MCF7 cells. The representative photos of the CAM, the cells expressing miR-SCR or miR-148a. Bar, 2 mm. B: Relative angiogenesis responses in the CAM were calculated and analyzed from eight independent CAM tissues, and presented as means±SE ( * P < 0.05). C: Cell proliferation assay of MCF7-miR-SCR and MCF7-miR-148a cells was performed using CCK8 kit.