MicroRNA-1 down-regulates proliferation and migration of breast cancer stem cells by inhibiting the Wnt/β-catenin pathway

Abstract

We investigated the miRNA profiles of breast cancer stem cells (CSCs) and non-CSC tumor cells by miRNA microarray and determined the effect of altered miR-1 expression on proliferation and migration of breast CSCs. The potential targets of miR-1 in the Wnt/β-catenin signaling were characterized by bioinformatics analysis and luciferase assay. We found that 14 miRNAs were up-regulated and 13 were down-regulated in the ESA+CD44+CD24-lineage- CSCs, related to ESA+CD44-CD24+lineage- non-CSC tumor cells. The miR-1 expression was associated inversely with aggressiveness of breast cancers. Furthermore, enhanced miR-1 expression decreased the percentages of SKBR3/CSCs and miR-1 inhibition increased the percentages of MCF-7/CSCs. Enhanced miR-1 expression significantly reduced the Frizzled 7 and Tankyrase-2 (TNKS2)-regulated luciferase activity in 293T cells and decreased Frizzled 7, TNKS2, c-Myc, octamer-binding transcription factor 4 (Oct4) and Nanog expression and the ratios of nuclear to cytoplasmic β-catenin as well as β-catenin-dependent luciferase activity in breast CSCs in vitro. miR-1 inhibited proliferation, migration and wound healing of breast CSCs in vitro. Enhanced miR-1 expression inhibited the growth of implanted MCF-7/CSCs while miR-1 inhibition promoted the growth of implanted MCF-7/CSCs in vivo. Our data indicate that miR-1 down-regulates breast CSC stemness, proliferation and migration by targeting the Frizzled 7 and TNKS2 to inhibit the Wnt/β-catenin signaling.

Keywords: breast cancer stem cells; miRNA profile.

Figure 1. The expression of miRNAs in breast cancer

A. The miRNA expression profile in six paired of breast CSC and non-CSC samples was characterized by miRNA microarray. The relative levels of miR-1 expression in different types of breast cancer tissue B. and serum samples C. were determined by quantitative RT-PCR. D. The relative levels of serum miRNA in patients (n = 18) with or without lymph node metastasis (n = 27). E. The levels of miR-1 expression in different types of breast cancer cell lines. Data are representative profile of clustered miRNA in individual samples and expressed as the mean ± SD of each group of samples from at least three separate experiments. N = 20 for luminal A; 10 for luminal B; 5 for Her2+; 10 for basal-like. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2. Altered miR-1 expression changes the percentages of breast CSC in breast cancer cell lines

MCF-7 and SKBR3 cells were transfected with, or without, miR-1 NC, miR-1 inhibitor, or miR-1 mimic for 24 or 48 h. The relative levels of miR-1 expression in different groups of cells were determined by quantitative RT-PCR A. The frequency of CD44+CD24− CSCs was determined by flow cytometry B–C. Data are representative FACS charts and expressed as the mean ± SD of each group of cells from three separate experiments. *p < 0.05, **p < 0.01 vs. the control CSCs.

Figure 3. miR-1 inhibits the stemness of breast CSCs by targeting the Wnt/β-catenin signaling

293T cells were transfected with luciferase reporter plasmids containing the 3UTR sequences of the fizzled 4, 5, 7, or TNKS2 and miR-1 or control miR-1NC and the effect of miR-1 on luciferase activity was determined A. The luciferase activity in the control cells that had been transfected with miR-1NC were considered as 100%. CD44+CD24− CSC from MCF-7 and SKBR3 cells were sorted and transfected with, or without, miR-1NC, miR-1inhibitor, or miR-1mimic for 24 or 48 h. The relative levels of fizzled 7, TNKS2 and c-Myc to control GAPDH were determined by Western blot B. In addition, the relative levels of cytoplasmic and nuclear β-catenin to GAPDH or lumen B1 in each group of cells were determined by Western blot C. Moreover, MCF-7/CSC and SKBR3/CSC were transfected with miR-1mimic, miR-1NC or miR-1inhibitor for 24 h. The cells in each group were transfected with TopFlash and pRL-TK for 48 h. The relative levels of β-catenin-dependent firefly luciferase activity in different groups of CSCs were determined by the dual luciferase assay D. The relative levels of β-catenin-dependent luciferase activity in the control cells that had been transfected with miR-1NC, TopFlash and pRL-TK were designated as 1. Finally, the relative levels of Oct4 and Nanog expression in each group of cells were determined by Western blot E. Data are representative images and expressed as the mean ± SD of each group of cells from three separate experiments. *p < 0.05; **p < 0.01.

Figure 4. miR-1 inhibits breast CSC proliferation in vitro

CD44+CD24− CSCs from MCF-7 and SKBR3 cells were transfected with, or without, miR-1 NC, miR-1 inhibitor, or miR-1 mimic for 24 or 48 h. Their proliferation was determined by MTT assay. Data are expressed as the mean rates ± SD of each group of cells from three separate experiments and proliferation of untransfected control CSCs was designated as 1. *p < 0.05; **p < 0.01 vs. the miR-1 NC and control CSCs.

Figure 5. miR-1 inhibits the migration and wound healing of CSC in vitro

CD44+CD24− CSC from MCF-7 and SKBR3 cells were transfected with, or without, miR-1 NC, miR-1 inhibitor, or miR-1 mimic. Four hours after transfection, the migration of different groups of CSC for the indicated time points was determined by transwell migration assay A. In addition, two days after transfection, the wound healing of different groups of cells within 24 or 48 h was tested B. Data are representative images and expressed as the mean ± SD of each group of cells from three separate experiments. The number of migrated SKBR3/CSCs was similar to those of SKBR3/miR-1NC CSCs and the migration distance of control MCF-7CSC and SKBR3/CSC was similar to those of MCF-7/miR-1NC and SKBR3/miR-1NC CSCs, respectively (data not shown). *p < 0.05; **p < 0.01 vs. the miR-1 NC-transfected CSCs.

Figure 6. miR-1 inhibits the growth of implanted breast tumors in vivo

The relative levels of miR-1 expression in the MCF-7/miR-1mimic, MCF-7/miR-1inhibitor or control MCF-7/miR-1NC CSCs were determined by quantitative RT-PCR A. The development and growth of implanted CSC-related tumors were monitored at the indicated time points post inoculation B. At the end of this experiment (on day 26 post inoculation), the relative levels of miR-1 expression in the different groups of dissected tumors were determined C. Data are expressed as the mean ± SD of each group of mice (n = 8 per group). *p < 0.05; **p < 0.01; ***p < 0.001 vs. the tumors induced by the MCF-7/miR-1NC CSCs.