miR-142-3p attenuates breast cancer stem cell characteristics and decreases radioresistance in vitro

Abstract

Effectively targeting cancer stem cells, a subpopulation of tumorigenic, aggressive, and radioresistant cells, holds therapeutic promise. However, the effects of the microRNA miR-142-3p, a small endogenous regulator of gene expression on breast cancer stem cells, have not been investigated. This study identifies the influence of miR-142-3p on mammary stemness properties and breast cancer radioresistance to establish its role in this setting. miR-142-3p precursor transfection was performed in MDA-MB-468, HCC1806, and MCF-7 cells, and stem cell markers CD44, CD133, ALDH1 activity and mammosphere formation were measured. β-catenin, the canonical wnt signaling effector protein, was quantified by Western blots and cell fluorescence assays both in miR-142-3p–overexpressing and anti–miR-142-3p–treated cells. Radiation response was investigated by colony formation assays. Levels of BRCA1, BRCA2, and Bod1 in miR-142-3p–overexpressing cells as well as expression of miR-142-3p, Bod1, KLF4, and Oct4 in sorted CD44⁺/CD24^–/low cells were determined by quantitative polymerase chain reaction. miR-142-3p overexpression resulted in a strong decline in breast cancer stem cell characteristics with a decrease in CD44, CD133, ALDH1, Bod1, BRCA2, and mammosphere formation as well as reduced survival after irradiation. miR-142-3p expression was strongly reduced in sorted CD44⁺/CD24^–/low stem cells, while Bod1, Oct4, and KLF4 were overexpressed. β-catenin levels strongly decreased after miR-142-3p overexpression, but not after anti–miR-142-3p treatment. We conclude that miR-142-3p downregulates cancer stem cell characteristics and radioresistance in breast cancer, mediated by a reduced role of β-catenin in miR-142-3p–overexpressing cells. miR-142-3p might therefore help to target cancer stem cells.

Introduction

MicroRNAs (miRNAs) are small single-stranded RNA molecules of approximately 22 nucleotides. ¹ They play a crucial role in the post-transcriptional regulation of gene expression. ² Wide-ranging effects on cancer cell growth, metastasis, and the tumor microenvironment have been attributed to miRNA regulatory functions.^1,3 Due to their dysregulation in cancer, miRNAs have been discussed as diagnostic markers and therapeutic agents in this setting. ⁴ Notably, there is a strong interest in investigating miRNAs to understand their potential roles as radio- or chemosensitizers.^5,6 Several studies have identified miRNAs as important regulatory elements in breast cancer.^7,8 In particular, specific miRNAs have been linked to breast cancer stem cell (CSC) characteristics. ⁹

Breast CSCs are a cancer cell subpopulation that is known to possess high tumorigenity.^10,11 CSCs express specific markers like CD44⁺/CD24^–/low, CD133, ALDH, and multidrug resistance (MDR)-efflux systems which have been linked to increased chemo- and radioresistance.^12,13 Therefore, successful therapeutic targeting of this subpopulation could be a driving factor for outcomes in breast cancer therapy. ¹⁴

miR-142-3p is dysregulated in breast cancer compared to normal breast tissue. ¹⁵ However, its role in breast malignancy and, specifically, its influence on breast CSCs remain unclear. Contradicting studies show miR-142-3p either as upregulating the wnt signaling pathway ¹⁶ or as directly targeting—and downregulating—the wnt signaling effector protein, β-catenin, as well as suppressing the canonical wnt pathway. ¹⁷ With wnt signaling being key to breast CSC maintenance,^18,19 the implications for breast CSCs remain unclear.

In the field of radiosensitivity, studies on different tumor entities show an miR-142-3p–mediated radiosensitizing effect in renal and lung carcinoma ²⁰ but increased radioresistance in malignant pediatric brain tumors. ²¹ No data exist regarding the role of miR-142-3p on breast cancer radioresistance.

In this study, we examine the influence of the miR-142-3p on breast CSC characteristics and radiosensitivity in vitro, as well as the underlying mechanisms.

Materials and methods

Cell lines and transfection

The breast cancer cell lines MCF-7 (hormone receptor positive), MDA-MB-468, and HCC1806 (both triple-negative) were obtained from American Type Culture Collection (ATCC)/LGC Standards (Wesel, Germany). Short tandem repeat (STR) analysis confirmed cell line authenticity. Cells were cultured as previously described. ²² For transient transfection, Lipofectamine and 20 nM of pre–miR-142-3p, pre-miR negative-control, anti–miR-142-3p, and anti-miR negative-control (all Thermo Fisher Scientific, Waltham, MA, USA) were used. To verify the success of transfection, miR-142-3p levels were determined after pre–miR-142-3p transfection using quantitative polymerase chain reaction (qPCR) analysis. Compared to cells transfected with a control pre-miR, miR-142-3p expression levels were increased in all cell lines when transfected with pre–miR-142-3p (Supplementary Figure S1), similar to previous findings. ²²

qPCR analysis

mRNA isolation was performed using the RNeasy Mini Kit 48 h after miR transfection. For mRNA reverse transcription, the High-Capacity cDNA Reverse Transcription Kit was applied. Kits were used per manufacturer’s instructions. qPCR was performed on a Rotor-Gene Q machine (Qiagen, Venlo, The Netherlands). All TaqMan probes were purchased from Applied Biosystems (Foster City, CA, USA), normalizing to 18S RNA expression (Supplementary Table SI). MicroRNA isolation was performed using the miRVana miRNA Isolation Kit (Thermo Fisher). RNA quality was evaluated with a biophotometer (Eppendorf, Hamburg, Germany). A260/A280 ratios were determined. Ratios between 1.8 and 2.0 were deemed appropriate. Reverse transcription was performed using the TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher). Kits were used according to manufacturer’s instructions. qPCR of miR-142-3p was performed with probes purchased from Applied Biosystems, normalizing to RNU6B expression. For detailed information, see manufacturer’s website. Data were expressed as fold change comparing pre–miR-142-3p or anti–miR-142-3p transfected cells to cells transfected with a control pre-miR or anti-miR using the 2^–ΔΔCT method. ²³

Western blot

Western blot experiments were conducted as previously described. ²⁴ Visualization of antibody binding was performed using peroxidase substrate (Thermo Fisher). A Fusion SL System (Peqlab, Erlangen, Germany) was used for quantification.

Quantification of ALDH1 activity

Flow cytometric ALDH analysis ²⁵ was performed 24 h after transfection. A CyFlow Space flow cytometer (Sysmex Partec, Görlitz, Germany) was used. An ALDEFLUOR^® Kit (STEMCELL Technologies, Vancouver, Canada) was applied for analysis. All experiments were performed according to manufacturer’s protocol.

Quantification of cell surface markers and cell sorting

The proportion of CD44⁺/CD24^–/low was determined 48 h after transfection using CD44 APC and CD24 PE antibodies and corresponding isotype controls. Similarly, CD133 PE and its isotype (all from BD Pharmingen, Franklin Lakes, NJ, USA) were used. All experiments were performed according to manufacturer’s protocol. CD44⁺/CD24^–/low CSCs were sorted on a CyFlow Space flow cytometer. Degree of purity compared to controls was assessed (Supplementary Figure S2). Gene expression was quantified by qPCR as described. Expression in the sorted CSCs was normalized to unsorted controls.

First- and second-generation mammosphere assay

Equal numbers of control pre-miR– and pre–miR-142-3p–transfected cells were seeded 24 h after transfection using coated six-well plates (Greiner, Kremsmünster, Austria). Spheromax CSC Medium (4 mL) from Promocell (Heidelberg, Germany) was added. Spheres exceeding 50 µm in size were counted microscopically 8 days after seeding and normalized against the number of originally implanted cells. Spheroids of the first generation were trypsinized and seeded as single cells in a 96-well plate. Again, spheroids exceeding 50 µm in size were counted after 8 days of incubation.

Intracellular β-catenin staining

Cells were harvested with accutase (PAN-Biotech, Aidenbach, Germany), washed with phosphate-buffered serum (PBS) and fixed and permeabilized with Cytofix/Cytoperm solution (BD Bioscience, San Diego, CA, USA) according to manufacturer’s specifications. After washing, cells were incubated with a primary antibody against β-catenin (Mouse IgG1, Thermo Fisher) for 60 min at room temperature, washed with Perm/Wash buffer and labeled with an Alexa-488–linked secondary antibody (goat anti-mouse IgG, Invitrogen, OR, USA) for another hour in the dark. Cells were washed again and resuspended in 1X BD Perm/Wash buffer containing bisBenzimide H 33342 (Sigma-Aldrich, St. Louis, MO, USA). Cells were then visualized with a fluorescence microscope (Axiophot, Zeiss, Jena, Germany).

Cell colony formation and irradiation

Predefined numbers of cells were seeded into cell culture dishes (Nunc, Langenselbold, Germany) and grown for 10 days. Colonies with more than 50 cells were counted. The surviving fraction (SF) was calculated as %SF = PE (irradiated)/PE(control) × 100, where plating efficiency (PE) is defined as PE = colony number/number of seeded cells. Cells were irradiated using a TrueBeam linear accelerator (Varian, Palo Alto, CA, USA). Doses of 2, 4, or 6 Gy were applied at a dose rate of 4.8 Gy per minute.

Statistics

All experiments were conducted at least three times in duplicates. Fold changes are presented as the mean values ± standard error of means (s.e.m). Data were tested for significance using Student’s unpaired t-test. The level of significance was defined as p < 0.05.

Results

miR-142-3p attenuates breast CSC characteristics

The purpose of this study was to examine the influence of the miR-142-3p on breast CSC characteristics and radiosensitivity in vitro, and to study the underlying mechanisms.

Breast CSCs express numerous stem cell markers. ¹³ Subsequently, we initially tested multiple CSC-related characteristics. For consistency, we performed the same stem cell tests in two triple-negative cell lines.

In MDA-MB-468 cells, flow cytometric analysis of CD133/1 expression revealed a decrease of 40.4% 48 h after transfection in miR-142-3p–overexpressing cells compared to controls (p < 0.001). In HCC1806 cells, the same decrease was noted and equaled 34.5% (p = 0.007, Figure 1(a)). ALDH1 activity decreased by an average of 35% in flow cytometry analysis when comparing miR-142-3p–overexpressing MDA-MB-468 cells to controls (p < 0.001). However, in HCC1806 cells, only a small (7.0%), but insignificant decrease was seen (p = 0.44, Figure 1(b)). In both MDA-MB-468 and HCC1806 cells, the CD44⁺/CD24^–/low subpopulation linked to the CSC phenotype ¹⁰ was decreased (p = 0.029 and p = 0.006, respectively, Figure 1(c)).

Figure 1.

Significant reduction of stem cell markers following miR-142-3p overexpression. (a1) Reduction of CD133/1 expression after miR-142-3p transfection in MDA-MB-468 cells. (a2) Representative flow cytometric analysis of CD133/1 positivity. The proportion of CD133/1 positive cells is given in Q2-gate. (a3) Reduction of CD133/1 expression after miR-142-3p transfection in HCC1806 cells. (b1) Reduction of ALDH1 activity after pre–miR-142-3p transfection in MDA-MB-468 cells. (b2) Representative example of the flow cytometric measurement. The black graph represents the DEAB-treated control, the red graph represents control pre-miR-treated cells, and the blue graph represents miR-142-3p–overexpressing cells. (b3) ALDH1 activity after pre–miR-142-3p transfection in HCC1806 cells. (c1) Reduced proportion of cells of the CD44⁺/CD24^–/low phenotype after pre-miR-142-3p transfection in MDA-MB-468 cells. (c2) Representative example of the flow cytometric determination. The proportion of the CD44⁺/CD24^–/low cells is given in Q4. (c3) Reduced proportion of cells of the CD44⁺/CD24^–/low phenotype after pre–miR-142-3p transfection in HCC1806 cells. Cells were transfected with a control pre-miR and pre–miR-142-3p, respectively, as detailed in the methods section (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, error bars indicate s.e.m.).

With very low constitutive expression of CD133/1 ²⁶ and low ALDH activity, ²⁷ hormone receptor–positive MCF-7 cells underwent mammosphere testing as an alternative readout to quantify the CSC phenotype. In human mammary epithelial cells, first-generation mammosphere formation is comprised of 68% of stem cells. In the second generation, this ratio increases to 98%. ²⁸ We chose to test sphere formation in both generations in separate experiments. Mammosphere formation was severely repressed in both generations after pre–miR-142-3p transfection. In the first generation, the reduction averaged more than 50% (p < 0.001), whereas in the second generation, sphere formation declined by 80% compared to controls (p = 0.006, Figure 2).

Figure 2.

miR-142-3p overexpression reduces MCF-7 mammosphere formation. (a) Quantification of first-generation spheres. (b) Quantification of second-generation mammosphere (count of spheres exceeding 50 μm). (n = 3, **p < 0.01, ***p < 0.001, error bars indicate s.e.m.). (c1 and c2) Representative pictures of first-generation mammosphere formation after control pre-miR transfection. (d1 and d2) Representative pictures of pre–miR-142-3p transfected cells.

Finally, as previously shown, ²² another stem cell factor, KLF4, was significantly downregulated in miR-142-3p–overexpressing cells in all but one cell line, HCC1806 (Figure 6b). Meanwhile, cells treated with anti–miR-142-3p did not demonstrate any downregulation (Figure 6b). Given this factor is already known to be influenced by miR-142-3p, ²² we did not perform further analyses, including Western blots.

miR-142-3p is downregulated in sorted CD44⁺/CD24^–/low breast cancer cells

Based on these associations, we hypothesized that breast CSCs are characterized by a low expression of miR-142-3p. We sorted CD44⁺/CD24^–/low MDA-MB-468 cells and compared the expression of miR-142-3p, Bod1 (based on a pre-described link to miR-142-3p ²⁰ ), and the pluripotency-associated stem cell markers KLF4 and Oct4 to unsorted control cells. In unsorted cells, Oct4 possesses very low expression levels in all cell lines which led us to refrain from further experiments, including Western blots, in unsorted cells. Meanwhile, in sorted cells, we found KLF4 and Oct4 to be highly expressed (13.4-fold and 18.5-fold compared to controls, p < 0.001 and p = 0.012, respectively, Figure 3(b)) and a very low expression level (0.06-fold compared to controls, p < 0.001) of miR-142-3p in the sorted stem cell population (Figure 3(a)). Bod1, a potential target of miR-142-3p, was 2.8-fold increased in the sorted stem cells (p = 0.03, Figure 3(b)).

Figure 3.

Expression of selected genes in sorted CD44⁺/CD24^–/low MDA-MB-468 cells compared to non-sorted controls: (a) sorted stem cells show a significantly decreased miR-142-3p expression compared to the non-sorted control cells and (b) stem cell–related genes (KLF4 and Oct4) as well as the radiation-related gene Bod1 were significantly higher expressed (n = 3, *p < 0.05; ***p < 0.001, error bars indicate s.e.m.).

β-catenin is repressed in miR-142-3p–overexpressing cells

A previous study discussed β-catenin, the effector protein of the canonical wnt pathway, as a direct target of miR-142-3p in non-breast cancer cell lines. ¹⁷ We therefore investigated this regulation in MCF-7, MDA-MB-468, and HCC1806 mammary cancer cell lines. Western blot revealed a strong reduction in β-catenin levels in all cell lines after pre–miR-142-3p transfection (p < 0.001, p < 0.001, and p = 0.0099, respectively, Figure 4). Meanwhile, cells treated with anti–miR-142-3p did not show any decline in β-catenin levels (Supplementary Figure S3). Fluorescence experiments demonstrated universally downregulated β-catenin across the cell in pre–miR-142-3p transfected cells. Stronger β-catenin levels were seen in both controls and anti-miR-142-3p transfected cells, particularly striking in HCC1806 cells (Figure 4). Downregulated nuclear β-catenin leads to a canonical wnt pathway downregulation. Canonical wnt pathway inactivity is known to reduce stem cell maintenance in breast cancer,^18,19 so we propose this pathway as the underlying mechanism behind our findings.

Figure 4.

Reduction of β-catenin protein expression after pre–miR-142-3p transfection in breast cancer cell lines. (a) Quantification of Western blot results. (b) Representative examples. MDA-MB-468, MCF-7, and HCC1806 cells were transfected with a control pre-miR and pre–miR-142-3p, respectively, as detailed in the methods section (n = 3, **p < 0.01, ***p < 0.001, error bars indicate s.e.m.). (c) Representative images of cells treated with a primary antibody against β-catenin and then labeled with an Alexa-488–linked second antibody (green). Cells were also treated with bisBenzimide H 33342 to visualize cell nuclei (blue). Images are presented for cell lines HCC1806, MDA-MB-468, and MCF-7. Prior to fluorescence treatment, cells were either transfected with pre–miR-142-3p, anti–miR-142-3p, or control pre-miR, as detailed in the methods section (n = 3).

miR-142-3p leads to a decrease of radioresistance in breast cancer cells

CSCs form a strongly radioresistant subpopulation in breast cancer. ⁴ As miR-142-3p attenuates the CSC phenotype, we subsequently evaluated changes in the radiation response after pre–miR-142-3p transfection.

Subsequent to all radiation doses (2, 4, and 6 Gy), a significant decline in cell survival was noted in all three cell lines transfected with pre–miR-142-3p compared to controls (Figure 5(a)–(c), Supplementary Table SII). Generally, the effect seemed to increase with higher doses. Cell survival was lower in MCF-7 than in the triple-negative MDA-MB-468 and HCC1806 cell lines.

Figure 5.

Impact of miR-142-3p on clonogenic cell survival after irradiation in MCF-7, MDA-MB-468, and HCC1806 cells. (a) MCF-7 cells were first transfected with a control pre-miR and pre–miR-142-3p, respectively, and then underwent radiation of 0, 2, 4, or 6 Gy. Surviving fractions were determined compared to unirradiated controls. (b) Radiation response in MDA-MB-468 cells. (c) Radiation response in HCC1806 cells. (n = 3, *p < 0.05 **p < 0.01, ***p < 0.001, error bars indicate s.e.m.; detailed significances are given in supplementary table SII). (d) Morphology of control pre-miR transfected cells. (d1) Non-irradiated MDA-MB-468 cells form a cell colony. (d2) Irradiated MDA-MB-468 cells form a cell colony (dose: 2 Gy). (e) Morphology of pre–miR-142-3p transfected cells. (e1) Non-irradiated MDA-MB-468 cells form a loose cell colony. (e2) Irradiated MDA-MB-468 cells are pictured (dose: 2 Gy).

The radiosensitization effect on cell survival is additive to a previously published ²⁹ non-radiation–related proliferation repression mediated by miR-142-3p. Comparing non-radiated pre–miR-142-3p transfected and control-miR transfected cells, we detected a strong decline in colony formation across all cell lines (Supplementary Figure S4).

Finally, further qPCR results of pre–miR-142-3p transfected and control pre-miR–treated cells identified several genes that are known to play roles in DNA repair and radioresistance to be significantly regulated, including Bod1, BRCA1, and BRCA2 (Figure 6(a) and (c)). KLF4, a stem cell factor described above, is also known to be protective against irradiation-induced damage (Figure 6(b)).

Figure 6.

qPCR results in cell lines MDA-MB-468, HCC1806, and MCF-7. (a1–a3) Changes in BRCA1 and BRCA2 expression in miR-142-3p–overexpressing cells. (b1–b3) Changes in KLF4 mRNA expression in miR-142-3p–overexpressing and anti–miR-142-3p–treated cells. (c1–c3) Changes in Bod1 expression in miR-142-3p–overexpressing and anti–miR-142-3p–treated cells. MCF-7, MDA-MB-468, and HCC1806 cells were transfected with a control pre-miR, pre–miR-142-3p, and anti–miR-142-3p, respectively, as detailed in the methods section (n = 3, *p < 0.05, **p < 0.01 ***p < 0.001, error bars indicate s.e.m.).

Discussion

In this study, we identified a miR-142-3p–mediated decrease of mammary CSC characteristics and breast cancer radioresistance.

Breast CSC characteristics

CSCs are associated with a number of different quantifiable features.^10,12,13,18 We demonstrate that miR-142-3p has as a strong downregulating influence on breast CSC characteristics in all three cell lines.

CD133

We show a miR-142-3p–mediated decrease of the predicted target CD133 ³⁰ in MDA-MB-468 and HCC1806 cells. High levels of this prominent stem cell factor predict chemoresistance in breast cancer and subsequently, CD133 has been identified as a target for immunotherapy. ³¹ Accordingly, the miR-142-3p–mediated downregulation might be of clinical promise. Our results are in line with recent findings in colon and liver cancer demonstrating a decrease in CD133/1 expression due to miR-142-3p overexpression.^32,33

ALDH1

ALDH1 activity is often quantified to identify breast CSCs. ¹³ We show that ALDH1 enzyme activity is significantly reduced in miR-142-3p–overexpressing MDA-MB-468 cells. Clinically, ALDH1-positive breast cancers show a greater invasiveness, a greater tumor size, and a higher histologic grade. ³⁴ Accordingly, ALDH1 activity reduction subsequent to miR-142-3p transfection offers further explanation to what was previously described as a miR-142-3p–mediated decrease in breast cancer cell invasiveness. ²² While miR-142-3p elicits a strong change of ALDH1 in MDA-MB-468 cells, there are only small non-significant changes in HCC1806. The reason for this remains unclear but could be due to cell line specific mutations, like single-nucleotide polymorphisms (SNPs), at the miR-142-3p binding seed region in the 3’-untranslated region (UTR) of aldehyde dehydrogenase (ALDH). This phenomenon was described for BRCA1 3’UTR and miR-103 in the breast cancer field. ³⁵

CD44⁺/CD24^–/low

We also identify a decrease in the CD44⁺/CD24^–/low ratio in both MDA-MB-468 and HCC1806 after miR-142-3p overexpression. This ratio has been shown to be an independent prognostic predictor of poor prognosis in triple-negative breast cancer. ³⁶

Experiments in sorted CD44⁺/CD24^–/low MDA-MB-468 cells demonstrated that KLF4 and Oct4 were more than 10-fold increased. MiR-142-3p expression was very low in the sorted stem cells compared to the unsorted cells and was associated with an increased expression of the potentially radioprotective molecule Bod1. ²⁰ This underlines both the hypothesis of increased therapy resistance of breast CSCs as well as the antagonistic relationship between miR-142-3p levels and breast CSC characteristics.

Mammosphere formation

Mammosphere formation is a common and valid tool for identifying and quantifying breast CSCs. ³⁷ Particularly, in MCF-7, a cell population with otherwise low stem cell characteristics, mammosphere formation is frequently used. ³⁷ In our study, we show that miR-142-3p has a downregulating influence on this stem cell characteristic. Mammosphere formation has been clinically correlated with tumorigenity in mice. ³⁸

KLF4

KLF4 has previously been shown to be regulated by miR-142-3p. ²² We reconfirm this downregulation both with respect to controls and anti–miR-142-3p–treated cells. KLF4 is overexpressed in breast CSCs and has been linked to invasion and metastasis. ³⁹ The KLF4 downregulation further underlines the loss of breast CSC characteristics in miR-142-3p–overexpressing cells. There was no significant KLF4 dysregulation in HCC1806 which might be due to a differential regulation of some miR-142-3p targets that has previously been described, including specifically for KLF4. ²²

Relationship between miR-142-3p and β-catenin

miR-142-3p has previously been shown to be dysregulated in breast cancer, ¹⁵ and recently, there has been some debate about its role in breast cancer cell signaling. A study by Isobe et al. ¹⁶ demonstrated an miR-142– and miR-142-3p–mediated activation of the canonical wnt pathway. A more recent publication reported conflicting data for miR-142-3p and showed a miR-142-3p–dependent downregulation of canonical wnt signaling: ¹⁷ Hu et al. offered compelling evidence highlighting varying roles for miR-142. They hypothesized that miR-142 might be fine-tuning the wnt signaling pathway depending on context, either upregulating it by suppressing APC gene expression or downregulating it by directly targeting post-transcriptional β-catenin expression. ¹⁷

In our study, given those previously described post-transcriptional expression changes, we decided to assess β-catenin levels using Western blot to assess protein levels. Another reason for this investigation was the need to present strong protein-based evidence, considering the intriguing relationship between β-catenin, the canonical signaling pathway and CSC characteristics. Finally, previously described high canonical wnt signaling activity ⁴⁰ indicated sufficient protein levels for this analysis.

Accordingly, our study demonstrates a strong decrease in β-catenin expression across the cell in miR-142-3p–overexpressing malignant mammary cells. Nuclear β-catenin is key to canonical wnt pathway activation. ⁴⁰ Canonical wnt signaling has been strongly correlated with mammosphere formation ¹⁸ and our study underlines this finding. We hypothesize that this pathway, via its role in stem cell maintenance, may also be the underlying cause for the reduction in the non-predicted CSC-related miR-142-3p targets CD44⁺/CD24^–/low ratio and ALDH1.

In general, β-catenin is also known to be associated with reduced survival in breast cancer patients. ⁴¹ Therefore, targeting β-catenin and its activity has been discussed as a potentially promising future therapeutic option. ⁴¹

Radiosensitizing effects

With contradicting studies in different tumor entities defining miR-142-3p either as increasing radiosensitivity in renal and lung cancer, ²⁰ or radioresistance in malignant pediatric brain tumors, ²¹ we show that, consistent in all breast cancer cell lines investigated, miR-142-3p functions as a radiosensitizer. Our findings, also demonstrating a miR-142-3p–mediated Bod1 regulation, are in line with the miR-142-3p/Bod1-pathway defined by Pan et al. ²⁰ in renal and lung cancer. KLF4, a stem cell factor discussed above that is downregulated by miR-142-3p overexpression is also known to be protective against irradiation-induced damage. ⁴² Finally, two highly important DNA damage repair–associated molecules, BRCA1 and BRCA2, the latter a predicted target of miR-142-3p, ³⁰ are downregulated as well. They are also known to influence radiation response.^43,44 The inconsistent results for BRCA1, regulated in MCF-7 but not in HCC1806 and MDA-MB-468, could be related to an indirect regulation linked to BRCA2 downregulation in the first cell line. Indeed, a coregulation of BRCA1 and BRCA2 has been previously described. ⁴⁵ BRCA1 is not a predicted target, which supports this assumption.

In summary, this study shows that miR-142-3p attenuates breast CSC characteristics and decreases radioresistance in vitro. Breast CSCs are key to breast cancer tumorigenesis, ¹⁴ with a pivotal role in cancer initiation, metastatic spread, and invasiveness.^11,27 Subsequently, inactivation of breast CSCs is a primary therapeutic target. ¹⁴ However, breast CSCs possess higher radioresistance, both in primary breast cancer^4,46 as well as in recurrent and metastatic breast malignancies. ⁴⁶ Recently, combination therapy for breast CSC eradication has been discussed with particular focus on stem cell targeting agents. ⁴⁷ In this setting, means to deliver miRNAs to tumor cells for therapy in vivo have been investigated. ⁴⁸ Our studies show that miR-142-3p regulatory functions might help overcome two obstacles in breast cancer therapy: miR-142-3p reduces expression of stem cell characteristics and additionally functions as a radiosensitizer. Accordingly, we propose that miR-142-3p could have significant therapeutic potential for breast cancer.

Supplemental Material

Supplementary_material – Supplemental material for miR-142-3p attenuates breast cancer stem cell characteristics and decreases radioresistance in vitro

Supplemental material, Supplementary_material for miR-142-3p attenuates breast cancer stem cell characteristics and decreases radioresistance in vitro by Fabian M Troschel, Nicolas Böhly, Katrin Borrmann, Timo Braun, Alexander Schwickert, Ludwig Kiesel, Hans Theodor Eich, Martin Götte and Burkhard Greve in Tumor Biology