MicroRNAs (miRNAs) are a group of small non-coding single-stranded RNAs molecules, the dysregulation of which plays a critical role in the initiation and biological progression of malignancies. The current study demonstrated that miR-140-5p was frequently downregulated in breast cancer stem cells (BCSCs), and miR-140-5p mimics could inhibit the proliferation of BCSCs. Moreover, Wnt1 was a direct target of miR-140-5p, as was proved by luciferase reporter assays. miR-140-5p mimics could downregulate the wnt1 mRNA and protein levels in MCF-7 and MDA-MB-231 cells. Furthermore, miR-140 mimics could enhance the sensitivity of BCSCs to doxorubicin (Dox) through the Wnt1/ABCB1 pathway both in vitro and vivo. Our findings have presented a novel miRNA-mediated regulatory network for BCSCs, which may provide a potential therapeutic target for breast cancer.
Breast cancer is one of the most common cancers in women worldwide. Like normal human organs, there are a small number of primitive stem cells in tumor called cancer stem cells (CSCs), which possess high proliferation, self-renewal, and differentiation potentials [1, 2]. The ALDH1+ phenotype of breast cancer stem cells (BCSCs) has been reported to exhibit high tumorigenesis and self-renewing capability. Several mechanisms have been reported to be involved in the progression and chemotherapy resistance of breast cancer. Nonetheless, CSCs have been pointed out in recent evidence to play critical roles in such process. On this account, it is of urgent need to develop CSC-targeting therapies, so as to inhibit the proliferation and enhance the efficacy of chemotherapy for breast cancer.
microRNAs are a class of small non-coding RNAs about 22–25 nucleotides in length, which can regulate gene expression at post-transcriptional level . In recent years, miRNAs have been recognized as potential diagnostic or prognostic markers for several cancers, which can serve as novel targets in the treatments for cancers, including breast cancer [4,5,6]. Recently, miRNA-140-5p has been demonstrated to play an important role in the initiation and progression of cancer . miR-140-5p has been shown in previous study to be abnormally expressed in preoperative breast cancer patients at various stages . Besides, other studies have found that miRNA-140 can suppress osteosarcoma growth, which acts as an anti-oncomiR in gastric cancer [9, 10]. Meanwhile, the overexpression of miR-140 inhibits tumor formation and metastasis of colorectal cancer (CRC) in vivo [11, 12]. In addition, miR-140-5p can also regulate the stem cells by suppressing the Wnt, SOX2, and SOX9 stem cell regulator pathways in breast cancer .
Wnt is a complex signaling pathway involved in regulating the crucial biological functions, such as development, proliferation, differentiation, and migration of cells (mainly stem cells), which are virtually present in all embryonic and adult tissues. Wnt1 is a member of the Wnt family, which can promote cancer progression through enhancing cell cycle, migration, and survival. Wnt1 can bind to specific Frizzled (FZD) surface receptors of target cells, so as to activate distinct intracellular pathways, which has resulted in the accumulation and nuclear localization of down-stream molecule β-catenin protein . Some clinical studies have reported that abnormal activation of the Wnt/β-catenin pathway is frequently involved in BCSCs .
The current study aimed to further identify the roles of miR-140-5p as well as the underlying molecular and cellular mechanisms during the progression of BCSCs. miR-140-5p mimics was found to inhibit the proliferation and sphere-forming capacities of BCSCs through suppressing the key target Wnt/β-catenin pathway. Furthermore, the current study also suggested that miR-140-5p could induce chemosensitivity in BCSCs to doxorubicin (Dox) through restraining ABCB1. Thus, our findings will help to elucidate the functions of miR-140-5p and its roles in breast cancer progression, which can also provide further evidence for targeting Wnt1 by miRNAs.
Materials and methods
Human breast cancer cell lines MCF-7 and MDA-MB-231 were obtained from the American Type Culture Collection (Manassas, VA, USA). All cells were cultured in DMEM medium (Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc) and 1% Penicillin/Streptomycin (Sigma-Aldrich, St. Louis, MO, USA), and maintained at 37 °C in a 5% CO2 and 95% air incubator. Dox was purchased from Sigma (St. Louis, MO, USA).
Tumor sphere formation assay
MCF-7 and MDA-MB-231 cells were suspended and diluted to a density of 105 cells/ml in the sphere-forming medium (SFM; Gibco) containing 10 ng/ml basic fibroblast growth factor (bFGF; PeproTech, Inc., Rocky Hill, NJ, USA), 20 ng/ml epidermal growth factor (EGF; PeproTech, Inc.), and 2% B27 (Thermo Fisher Scientific, Inc.) in the ultra-low attachment 6-well plates (Corning Incorporated). The medium was half-replaced every 3 days and cell passage was conducted every 7–10 days.
Luciferase reporter assay
The binding sites for miR-140-5p in the 3′-untranslated region (3′-UTR) sequence of Wnt1 were specifically synthesized. The mutant Wnt1 3′-UTR was constructed through mutating multiple nucleotides complementary to the miR-140-5p seed region. A mixture of 100 ng pmiR-RB-Report™ hWnt1 wild-type (WT) or mutant (Mut) reporter plasmid vector together with 100 nM miR-140-5p mimics or negative control (Thermo Fisher Scientific) was co-transfected into the HEK-293 cells in the 96-well plates using Lipofectamine 2000 (Thermo Fisher Scientific). Luciferase activities of the indicated cells were measured by the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA) 24 h after transfection. Typically, the firefly and Renilla luciferase activities were measured using the Luc-Pair miRNA Luciferase Assay Kit (GeneCopoeia). Activities were normalized to that of Renilla luciferase.
Cells at logarithmic growth phase were transfected with Has-miR-140-5p mimics, treated with 0.25% trypsin (Invitrogen, Carlsbad, CA, USA), and collected through centrifugation. Cells were seeded into the 24-well plates (Corning, Lowell, MA, USA) and grown overnight until they reached 60–70% confluence before transfection. miR-140-5p mimics (GenePharma, Shanghai, China) were transfected into MCF-7 and MDA-MB-231 cells at a final concentration of 50 µM using LipofectamineTM 2000 (Invitrogen, Carlsbad, CA, USA) in accordance with the manufacturer instructions. Each experiment was performed in triplicate.
Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis
Total cellular RNA was extracted with the TRIzol Reagent (Invitrogen; Thermo Fisher Scientific, Inc). mRNA was quantified through qPCR using the SYBR Premix Ex Taq (Takara Bio, Inc) and the Applied Biosystems Prism 7900HT system (Applied Biosystems; Thermo Fisher Scientific, Inc). The Wnt1 primers were as follows: Forward, 5′-CGCTCTTCCAGTTCTCAGACAC-3′, and reverse, 5′-CAGGATGGCAAAGGGTTCG-3′ . Primers for miR-140-5p were as follows: forward, 5′-GAGTGTCAGTGGTTACCGT-3′, and reverse, 5′-GCATGGTCCGAGGTATTC-3′. Primers for U6 were as shown below: Forward, 5′-CTCGGACTCAGCCTGCA-3′, and reverse, 5′-TCAATATTACCGAGCTGCGT-3′. Primers for β-actin included: forward, 5′-TTGCACGAACTCTATCTCCTG-3′, and reverse, 5′-GGCAGATAGACGCTAAGCGTCAAC-3′. The expression level of mature miRNA-140-5p was normalized to U6, while that of Wnt1 mRNA was normalized to β-actin mRNA. The fold change in expression was calculated using the 2−ΔΔCq method.
Cell viability assay
After transfection with miR-140-5p mimics, cells were seeded in the 96-well plates at a concentration of 5 × 103 cell/well. Subsequently, cells were treated with Dox (1 µM) for 0–72 h. Subsequently, 10 µl CCK-8 solution (Sigma-Aldrich) was added to each well for cell counting. The optical density (OD) was measured at 490 nm using a microplate reader.
Aldehyde dehydrogenase activity was analyzed using the Aldefluor reagent (StemCell Technologies Inc., Vancouver, BC, Canada) in accordance with the manufacturer instructions. In brief, a single cell suspension (1 × 106) was mixed with the activated ALDH substrate (StemCell Technologies Inc.), whereas diethylaminobenzaldehyde (DEAB), a specific and irreversible inhibitor of ALDH, was used as a negative control. Finally, Aldefluor+ and Aldefluor− populations were isolated under DEAB‐negative conditions through the cell sorter.
Cellular total proteins in each group were extracted using the ProteoPrep® Total Extraction Sample kit (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). The following specific primary antibodies were used, including anti-Wnt1 (1:500; Abcam, Cambridge, MA, USA), anti-β-catenin (1:500; Abcam), anti-ABCB1 (1: 800; Abcam), anti-ALDH1 (1:500; Abcam), anti-Oct4 (1:500; Abcam), anti-ki-67 (1:500; Abcam), and anti-GAPDH (1:1,000; Abcam). Blots were visualized through the enhanced chemiluminescence (Thermo Fisher Scientific, Inc.). Densitometry of the bands was performed using the Quantity One software, version 4.6.9 (Bio-Rad Laboratories, Inc)
Intracellular Dox accumulation
MCF-7 and MDA-MB-231 cells transfected with miR-140-5p mimics or control were incubated with 1 µM Dox for 72 h. Afterwards, cells were harvested and washed twice with ice-cold PBS. Finally, Dox-associated mean intracellular fluorescence intensity was determined using the FACS Caliber flow cytometer (Becton–Dickinson, USA).
To determine the effect of restoring miR-140-5p expression on reversing drug resistance in BCSCs in vivo, nine 8-week-old female nude mice were randomly divided into three groups. Group 1 and Group 2 were given an injection of 5 × 105 MCF-7 cells, while Group 3 was given an injection of 5 × 105 cells transfected with miR-140-5p mimics. Ten days after injection, when the average tumor volume had reached about 100 mm3, Group 1 was given an injection of sterile saline (0.9% NaCl), while Group 2 and Group 3 were treated with 2 mg kg−1 Dox for every 3 days. Tumor volume (mm3) was measured by a digital caliper and calculated as W2 × L/2. Subsequently, the mice were sacrificed 18 days later, and tumor tissues were collected from the above treated mice for immunohistochemistry. In brief, the tissue sections were incubated with anti-Oct4 (1:200, Abcam). Immunohistochemical staining was analyzed using the Image-Pro-Plus (IPP) 6.0 software. The nude mice were obtained from Dalian Medical University Laboratory Animal Center, and the experiments were approved by the Ethics committee of Dalian Medical University.
Each experiment was representative of at least three independent experiments. All data were expressed as the means ± S.E.M. of at least three independent experiments. Differences between groups were analyzed using Student’s t-test, while those among three and above groups were analyzed using analysis of variance (ANOVA). A difference of *P < 0.05 was deemed as statistically significant.
BCSCs expressed a low miR-140-5p level compared with the non-CSC cells
To isolate the CSCs subpopulation from MCF-7 cells, cells were incubated in the medium supplemented with B27, bFGF, and EGF. As shown in Fig. 1a, the BCSCs spheres rapidly increased in size on day 7. Ten days later, cells were separated into the BCSCs and non-CSCs by flow cytometry with the Aldefluor+ marker, so as to evaluate the effect of miR-140-5p on BCSCs. The miR-140-5p expression levels in BCSCs and non-CSCs were examined. Results of RT-qPCR (Fig. 1b) suggested that the total amount of miR-140-5p mRNA in CSCs was markedly downregulated compared with that in non-CSCs.
miR-140-5p degraded the Wnt1 mRNA in MCF-7 and MDA-MB-231 cells
The target gene of miR-140-5p was predicted using Targetscan (http://www.targetscan.org) (Fig. 2a). We had focused on Wnt1, since aberrant Wnt1 gene expression had been reported to be involved in the abnormal development and malignancy of BCSCs. Luciferase assay data showed that the miR-140-5p mimics could restrain the activity of wild-type (wt) Wnt1 construct compared with a vector control, whereas no abnormality was observed in the mutant (mut) Wnt1 construct, suggesting that miR-140-5p could directly regulate Wnt1 gene expression at post-transcriptional level (Fig. 2b). Furthermore, MCF-7 and MDA-MB-231 cells were transfected with miR-140-5p mimics and its negative control (N-control), respectively, for 48 h. The results indicated that the Wnt1 mRNA level was downregulated by miR-140-5p mimics, thereby leading to the same protein expression profiles in MCF-7 and MDA-MB-231 cells (Fig. 2c, d). Taken together, these results revealed that miR-140-5p could target Wnt1 through regulating the relevant 3′-UTR regions and repressing the corresponding translation, thereby suppressing the Wnt1 protein expression level.
miR-140-5p sensitized breast cancer cells to Dox
To determine the influence of Dox on the chemosensitivity of miR-140-5p, MCF-7 and MDA-MB-231 cells were treated with 1 μM Dox as well as miR-140-5p mimics + Dox (1 μM), respectively. Afterwards, the viability of both cell lines from 0 to 72 h was evaluated by CCK-8 assay. Our results suggested that the OD levels in MCF-7 and MDA-MB-231 cells treated with miR-140-5p mimics + Dox were evidently lower than those in the control group and Dox group (Fig. 3a, b). Furthermore, flow cytometry was conducted at 72 h to analyze the influence of miR-140-5p on cell drug resistance. The results demonstrated that after upregulating the expression of miR-140-5p, the intracellular fluorescence intensity of Dox had increased from 15.3 to 51.2% in MCF-7 cells, and 19.3 to 60.4% in MDA-MB-231 cells (Fig. 3c, d). These data proved that miR-140-5p mimics could partially reverse the resistance to Dox in MCF-7 and MDA-MB-231 cells.
MiRNA-140-5p inhibited the proliferation of BCSCs and enhanced the sensitivity of BCSCs to Dox by the Wnt1/ABCB1 pathway
First of all, the mammosphere formation assay was performed. Mammosphere was calculated as the volume of sphere-like structures with the diameter of >50 μm. As shown in Fig. 4a, the size of mammosphere was outstandingly reduced after the overexpression of miR-140-5p in MCF-7 and MDA-MB-231 cells compared with the control. Subsequently, results of the flow cytometry depicted that miR-140-5p could markedly reduce the proportion of ALDH1+ cells in MCF-7 and MDA-MB-231 cells. Interesting, we found that, Dox had little influence on ALDH1+ cells, but such influence could be enhanced by miR-140-5p mimics; meanwhile, the proportion of ALDH1+ was dramatically lowered in cells treated with Dox and miR-140-5p mimics together (Fig. 4c, d). To further explore the mechanism by which miR-140-5p induced CSCs inactivation, the expression of proteins involved in the Wnt1/β-catenin signaling pathway was examined. Moreover, the expression of several molecules related to drug resistance and self-renewal in BCSCs was also explored. Compared with the control group, BCSCs from miR-140-5p cells treated with Dox had dramatically downregulated expression of Wnt1, β-catenin, ABCB1, Oct4, and ALDH1 (Fig. 4e, f). These results suggested that the miR-104-5p/Wnt1/ABCB1 signaling pathway might play an important role in the Dox-induced inactivation of BCSCs.
MicroRNA-140-5p enhanced the sensitivity of BCSCs to Dox in vivo
In our xenograft model, 5 × 105 miR-140-5p-transfected or untreated MCF-7 cells were injected into the cleared mammary fat pads of nude mice. Our results showed that miR-140-5p preconditioning could strongly delay the growth in tumor size and tumor weight (Fig. 5a, b). In addition, the tumor growth curves demonstrated that the tumor volume in miR-140-5p combined with Dox group was notably decreased compared with that in control group and Dox group (Fig. 5a, b). Besides, remarkably lower Oct4 expression could be found in miR-140-5p group by IOD analysis, as shown in immunohistochemical staining (Fig. 5c). Furthermore, the expression of Wnt1, ABCB1, ALDH1, and ki-67 was also detected by Western blotting. Compare with the control group, the expression levels of Wnt1, β-catenin, ABCB1, and ALDH1 in Dox group showed no obvious differences (except for ki-67), regardless of the smaller tumor volume after Dox treatment (Fig. 5d, e). On the contrary, the expression levels of Wnt1, ABCB1, β-catenin, and ALDH1 were distinctly lower in miR-140-5p groups, which had supported our hypothesis that miR-140-5p could effectively target BCSCs with self-renewal and proliferation capacities in vivo.
Previous studies have demonstrated that the dysregulation of miRNA will contribute to carcinogenesis by suppressing the expression of target gene [17, 18]. miRNAs act as either oncogenes or tumor suppressor genes, some of which are strongly associated with drug resistance . miR-140 acts as a tumor suppressor gene in a variety of cancers, including hepatocellular carcinoma (HCC), CRC, and cervical cancer. Besides, recent miRNA profiling studies have demonstrated that miR-140-5p is one of the downregulated miRNAs in various differentiated cell types compared with the BCSCs. Nevertheless, its functional role has not been fully investigated in these earlier reports yet . The Wnt/β-catenin signaling pathway is a well-known pathway involved in regulating CSCs. Typically, CSCs are reported to be the origin of malignant cancers, which will result in the poor prognosis for multiple cancer types . The current study has investigated whether miRNAs play a role in the proliferation of BCSCs alongside the Wnt signaling genes. In addition, the molecular mechanism that mediates the reduction of miR-140-5p in breast cancer cells is also examined using Targetscan. A subsequent luciferase reporter assay reveals that downregulation of miR-140-5p will increase the luciferase activity controlled by the wild-type Wnt1 3′UTR construct. However, such effect is diminished when the 3′UTR of the Wnt1 sequence is mutated. Consistently, Han et al. revealed that miR-140-5p could target and inhibit the Wnt1 gene, thus highlighting the negative regulation of the Wnt/β-catenin signaling pathway in hypoxia .
CSCs play a key role in tumor initiation, progression, and resistance to chemotherapy. Recent evidence suggests that different types of miRNA also play a role in regulating the growth and replication of CSCs . For instance, overexpression of let-7a can inhibit the proliferation and mammosphere formation in BCSCs in vitro , while miR-30 can downregulate the self-renewal ability and reduce anoikis resistance through directly targeting ubiquitin-conjugating enzyme 9 (UBC9) and intergrin b3 (ITGB3) . In our study, we find that the overexpression of miR-140-5p can inhibit the mammosphere number and ALDH1+ population in vitro, suggesting that miR-140-5p can repress the expansion of BCSCs. As previously demonstrated, the overexpression of Wnt1 can enhance the tumorsphere formation efficiency, but knockdown of Wnt1 expression will result in decreased tumorigenic potential and downregulated ALDH1+ expression . Thus, taken together, these results suggest that miR-140-5p may be correlated with the differentiation of BCSCs, which is achieved through regulating Wnt. Furthermore, the expression of CSC-associated proteins (including Oct4 and ALDH1) is also detected to support the above speculation. As a result, we discover that the downregulated expression of Oct4 and ALDH1 may be attributed to miR-140-5p mimics transfection. Oct4 is a well-known transcription factor to regulate the self-renewal and pluripotency of embryonic stem cells (ESCs), and Oct4 expression is related to the maintenance and expansion of CSCs . In the present study, no direct relationship between miR-140-5p and Oct4 is found, but our previous report indicates that the downregulation of Wnt1 by siRNA will decrease the levels of pluripotency marker of gastric CSCs, including Oct4 . Consistent with this notion, our study has suggested a possible mechanism for the limited proliferation of BCSCs by the miR-140-5p/Wnt1/Oct4 pathway.
Recent studies have reported that miRNA can directly target BCSCs or increase the sensitivity of BCSCs to existing chemotherapy . In this study, we observe that the intracellular fluorescence intensity of Dox is increased after upregulating the expression of miR-140-5p in vitro. Meanwhile, we also discover that Oct4 and ALDH1 expression is markedly downregulated in the miR-140-5p + Dox treatment group, as detected by immunohistochemical staining and Western blotting in vivo. Chemotherapy resistance is a common event, which contributes to cancer-related death, since it often leads to failure in inhibiting disease progression. ABCB1, which belongs to the ABC transporter family, locates on the chromosome. The ABC family is constituted by about 12 putative drug transporters, such as MDR-associated protein-1 (MRP1) encoded by ABCC1 and MDR1 encoded by ABCB1. ABCB1 participates in the efflux of drugs from tumor cells. In addition, the overexpression of ABCB1 is found to contribute to drug resistance in various cancers . High MDR1 expression in CSCs will decrease the sensitivity of cells to chemotherapeutic agents, which will in turn lead to the failure in eliminating CSCs. In comparison, the terminal differentiated tumor cells will be removed in the presence of low MDR1 expression . In lung cancer, miRNA-140-5p can enhance the cisplatin efficacy by reducing the migration and invasion ability of cancer cells . Currently, few studies are available about the mechanism by which miR-140-5p regulates drug resistance, and the relationship between Wnt and drug resistant proteins has been partially illustrated. For instance, Huang et al. demonstrated that the Wnt/β-catenin pathway was involved in the RARγ-induced up-regulation of MDR1 in CRC, and RARγ knockdown would suppress MDR1 expression in CRC cells through the Wnt/β-catenin pathway . Additionally, Stein et al. reported that the mutation status of β-catenin could determine the ABCB1 expression in a defined cell line model and in colon cancer specimens . Taken together, we speculate that the main mechanism by which miR-140-5p increases the breast cancer sensitivity to Dox is achieved by reducing the expression of Wnt1/ABCB1 signaling.
This study has provided a first insight into the important function of restoring the miR-140-5p level in BCSCs. Moreover, it has provided a new therapeutic strategy about the combinational use of miR-140-5p and anticancer drugs to overcome the acquired multi-chemoresistance through targeting the Wnt1 signaling.
This work was supported by grants from the National Natural Science Foundation of China (No. 81502561) and Tumor Stem Cell Research Key Laboratory of Liaoning Province.