MiR-1287-5p inhibits triple negative breast cancer growth by interaction with phosphoinositide 3-kinase CB, thereby sensitizing cells for PI3Kinase inhibitors

In this study, we used the triple negative BC cell lines SUM159, BT549, MDA-MB-231, MDA-MB-468, and HCC1937; the endocrine receptor positive T47D, MCF7, and BT474; and the HER2 receptor positive HCC1937 and SKBR3 [20]. The BC cell line SUM159 was purchased from Asterand (Detroid, MI, USA); KPL-1 was from DMSZ (Braunschweig, Germany); BT549, MDA-MB-231, MDA-MB-468, MCF-7, BT474, T47D, HCC1419, HCC1937, SKBR3, and the embryonic kidney cell line HEK-293 were obtained from American Type Culture Collection (ATCC; Manassas, CA, USA). BT549, T-47D, HCC-1419, and HCC1937 cells were maintained in RPMI 1640 (with l-glutamine, Gibco, Darmstadt, Germany) 10% fetal bovine serum (FBS) gold (Biochrom, Cambridge, UK) and 1% penicillin/streptomycin (for all used cell lines: penicillin: 10000 units/ml, streptomycin: 10.000 μg/ml, Gibco); SUM159 cells were grown in Ham’s F12 containing 1 mmol/L l-glutamine (GE Health Care Life Sciences, Pittsburgh, USA), 2 mmol/L HEPES buffer (Gibco), 5 μg/ml insulin actrapid (Novo Nordisk, Vienna, Austria), 1 μg/ml hydrocortisone (Sigma-Aldrich, Vienna, Austria), 1% penicillin/streptomycin (Gibco), and 5% FBS gold (Biochrom). MDA-MB-231, MDA-MB-468, KPL-1, and HEK-293 cells were maintained in high-glucose DMEM (Gibco), 10% FBS gold (Biochrom), and 1% penicillin/streptomycin (Gibco). MCF-7 cells were grown in MEM with Earle’s salts containing 2 mmol/L l-glutamine (PAA, Pasching, Austria), 1% sodium pyruvate (Gibco), 1% penicillin/streptomycin (Gibco), and 10% FBS gold (Biochrom). BT474 were cultured in RPMI 1640 (with l-glutamine, Gibco), 20% FBS gold (Biochrom), 1% penicillin/streptomycin (Gibco), and 10 μg/ml insulin Actrapid (Novo Nordisk). SKBR3 were grown in McCoy’s 5A modified Medium (Gibco; w/o l-glutamine, 2,2 g/L sodium bicarbonate) 1% penicillin/streptomycin (Gibco) and 10% FBS gold (Biochrom). All cell lines were grown in a 5% CO₂ humidified incubator at 37 °C. The BC cell lines were authenticated at the Cell bank of the Core Facility of the Medical University of Graz, Austria, by performing a STR profiling analysis (Kit: Promega, PowerPlex 16HS System; Cat.No. DC2101, last date of testing: March 3, 2016). Mycoplasma testing was performed using the Venor GeM Mycoplasma Detection Kit (Minerva Biolabs, Berlin, Germany). After obtaining a confluence of approximately 70%, total RNA was isolated following a standard TRIzol protocol (Invitrogen, Thermo Fisher Scientific, Waltham, MA USA) according to the manufacturer’s instructions.

To identify novel miRNAs associated in breast carcinogenesis, we generated tumor spheres as previously described with slight modifications [19]. In detail, the adherent growing BC cell lines were dissociated into single cells using trypsin/EDTA and 2,000 single cells per well seeded in ultra-low attachment six-well plates (Corning, NY, USA) using serum-free MEBM (Lonza, Basel, Switzerland) medium (SFM) supplemented with 1xB27 supplement (Gibco), 20 ng/ml human epidermal growth factor (EGF; Peprotech, Hamburg, Germany), 10 ng/ml human basic fibroblast growth factor (FGF; Peprotech), 20 IU/ml Heparin (Baxter, Vienna, Austria), and 1% antibiotic/antimycotic solution (Thermo Fisher Scientific, containing 10,000 units/mL of penicillin, 10,000 μg/mL of streptomycin, and 25 μg/mL of Gibco Amphotericin B). Spheres and corresponding adherent growing cells were harvested and RNA was extracted for the microarray analysis using the miRNeasy Kit (Qiagen, Hilden, Germany).

To assess differentially expressed miRNAs in adherent growing cells compared to mammospheres in three different BC cell lines in biological triplicates, total RNA was isolated using the miRNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Total RNA was checked on Bioanalyzer BA2100 (Agilent; Foster City, CA) for excellent quality. All samples showed a RIN (RNA integrity number) > 9. The whole transcriptome analysis was performed on Affymetrix GeneChip miRNA Arrays v3 (Affymetrix; Santa Clara, CA, USA) as instructed by the manufacturer’s protocol. Hybridizations were done at the Core Facility Molecular Biology at the Centre of Medical Research at the Medical University of Graz. Raw data is available at the Gene Expression Omnibus (GEO; accession number GSE103218). Heat map for the miRNA expression levels was generated by R software.

For comparison of matched normal breast and cancer tissue samples, a cohort of 131 BC patients was provided by the Laboratory of Oncology, IRCCS Casa Sollievo della Sofferenza, Viale Padre Pio, 71013 San Giovanni Rotondo, FG, Italy. Ethical approval and informed consent were obtained to fulfill the institutional requirements. RNA preparation and quantitative PCR were performed as previously described [21]. The relative expression levels of miR-1287-5p were determined by qRT-PCR, and expression differences in cancer tissue were compared against normal tissue. To perform a confirmation in a second cohort, we downloaded and analyzed data of publicly available patients from the Cancer Genome Atlas Project (TCGA; https://​cancergenome.​nih.​gov/​) for BC patients (Download date: December 2018). In detail, level 3 Illumina miRNASeq (Illumina Sequencing technology: Genome Analyzer) were used for miRNA expression analysis. We derived the “reads_per_million_miRNA_mapped” values for mature forms for each miRNA from the “isoform_quantification” files. We downloaded patient clinical information for the TCGA patients with breast invasive carcinoma (BRCA) from cbioPortal (http://​www.​cbioportal.​org/). For the miRNA-Seq data for primary tumors, we derived the ‘reads_per_million_miRNA_mapped’ values for the mature form of hsa-miR-1287-5pMIMAT0005878 hsa-miR-4521 MIMAT0019058 hsa-miR-27a-5p MIMAT0004501 hsa-miR-3150b-3p from the “Isoform Expression Quantification” files from Genomic Data Commons Data Portal (https://​portal.​gdc.​cancer.​gov/​). The log2-transformation was applied to miRNASeq data. We ended up with a number of 916 primary tumor cases and 93 normal adjacent tissue with miRNA data. Among them there are 92 matched tumor-normal pairs. To be able to determine the expression difference for each miRNA between normal and tumor tissue, we first employed a Shapiro-Wilk test to verify if the data follows a normal distribution. Accordingly, the t-test, respectively the nonparametric Mann Whitney Wilcoxon test, was applied to assess the relationship between miRNA expression and tissue type. We performed paired and unpaired comparisons. A box-and-whisker plot (Box plot represents first (lower bound) and third (upper bound) quartiles, whiskers represent 1.5 times the interquartile range) was used to visualize the data. Analyses were carried out in R statistical environment (version 3.4.1) (http://​www.​r-project.​org/​). All tests were two-sided and considered statistical significant at the 0.05 level. For testing the prognostic significance of miR-1287-5p, we made use of the publicly available online tool (http://​kmplot.​com/​analysis/​index.​php?​p=​service&​cancer=​breast_​mirna) to analyze 1262 patient data from different cohorts [22]. For analysis of the prognostic value of PI3KCB, we used two publicly available online tools (http://​www.​oncolnc.​org/​ for the TCGA data set), and for validation cohort, the Kaplan-Meier plotter (http://​kmplot.​com/​analysis/​index.​php?​p=​service&​cancer=​breast) as previously reported [23].

For quantification of miRNA levels in BC cell lines, 1 μg of total RNA was reverse transcribed by the miScript II RT Kit (Qiagen) according to the manufacturer’s protocol. The following miScript Primer Assays (Qiagen) were used to validate miR-1287-5p in adherent growing cells compared to mammospheres from the miRNA microarray: Hs_miR-1287_1 miScript Primer Assay and Hs_RNU6-2_1 miScript Primer Assay. The miScript Primer Assays were applied on a LightCycler 480 Real-Time PCR System (Roche Diagnostics, Mannheim, Germany) using the miScript SYBR Green PCR Kit (Qiagen) according to the protocol. Measurements were carried out in technical and biological triplicates and RNU6-2 was used as a housekeeper. Relative miRNA expression levels were calculated using the 2^−ΔΔCT method according to Livak and Schmittgen [24].

For detection of mRNA expression levels, 1 μg of total RNA was reverse transcribed by using QuantiTect Reverse Transcription Kit (Qiagen) according to the manufacturer’s protocol. Quantitative RT-PCR was carried out in technical duplicates of biological triplicates using primers specific for PIK3CB, LAYN, RAP2B, SMAD2, PLD5, CORO2A, GAPDH, and U6. Primer sequences are listed in Additional file 1: Table S1. Quantitative RT-PCR was done on a LightCycler® 480 Real-Time PCR System (Roche Diagnostics) using the QuantiTect SYBR Green PCR Kit (Qiagen) according to the manufacturer’s standard protocol. The arithmetic mean of the housekeeping genes GAPDH and U6 was used for normalization, and relative gene expression levels were calculated using a standard 2^−ΔΔCT method [24]. Each experiment was performed in three independent biological replicates.

Total proteins from BC cells were extracted using radioimmunoprecipitation assay (RIPA) buffer (Sigma-Aldrich). Of total cellular proteins, 20 μg were resuspended in 4x Laemmli buffer (BioRad, Hercules, CA, USA) and heated at 95 °C for 10 min. Proteins were separated by a 4–15% Mini-PROTEAN® TGX™Precast Gel (BioRad), transferred onto a nitrocellulose membrane (BioRad), and the membrane was blocked for 1 h with 3% non-fat dry milk in 1xTris buffered saline (TBS; BioRad)/0.1% Tween-20 (Sigma-Aldrich). Immunoblotting was performed, and antibodies specific for PIK3CB (mAb #3011, CellSignaling, diluted 1:1000 in 1% non-fat dry milk in Tris buffered Saline/0.1% Tween-20) and β-actin (AC-15, Sigma-Aldrich, diluted 1:5000 in 1% non-fat dry milk in Tris buffered Saline/0.1% Tween-20) were detected using HRP-conjugated anti-mouse (Dako, Glostrup, Denmark, dilution 1:5000) or anti-rabbit (Santa Cruz, dilution 1:1000) antibodies. Visualization was performed using an enhanced chemiluminescence detection system (Super Signal West Pico, Thermo Scientific, Rockford, IL) on a BioRad ChemiDoc Touch device. Relative quantification of protein expression was performed using the ImageJ (NIH, Bethesda, Maryland) software. Therefore, the band density of the protein of interest was measured and divided by the density of the loading control beta actin.

To achieve transient overexpression or reduction of miR-1287-5p expression in BC cell lines, the miR1287-5p mimic (Syn-hsa-miR-1287-5p miScript miRNA Mimic, 50 nM), inhibitor (Anti-hsa-miR-1287-5p miScript miRNA Inhibitor, 50 nM), and recommended negative control (miScript Inhibitor Negative Control and AllStars Negative Control, 50 nM) were used according to protocol recommendation of the manufacturer (Qiagen). Cells in 6-well plates were transfected using the fast-forward transfection protocol; cells in 96-well plates were transfected using the reverse transfection protocol using the HiPerFect Transfection Reagent protocol (Qiagen) according to the manufacturer’s instructions. To confirm the reached levels of overexpression or silencing, quantitative RT-PCR was used in comparison to the respective controls.

MDA-MB-231 and SUM159 cells were seeded and incubated overnight in complete growth medium in 24-well plates. After 24 h, the medium was replaced with complete growth medium containing ViralPlus Transduction Enhancer (1:200, ABM, Richmond, BC, Canada) and 8 μg/ml polybrene (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Cells were infected by adding 10 μl miR-1287 precursor (LentimiRa-GFP-hsa-miR-1287precursor Virus, ABM), miR-1287-5p lentiviral particles (LentimiRa-GFP-hsa-miR-1287-5p Virus, ABM), or control lentiviral particles (Lenti-III-mir-GFP Control Virus ABM). Stably transfected cells were continuously selected with 1 μg/ml puromycin dihydrochloride (Gibco) for 4 weeks and expression levels were determined by qRT-PCR.

To test whether manipulation of miR-1287-5p expression levels influences cellular growth rate of BC cells, we measured the short-term effects (96 h) by applying the WST-1 proliferation assay (Roche, Applied Science, Mannheim, Germany). Cell lines were seeded in 96-well plates and transiently transfected with the mimic or inhibitor using the reverse transfection protocol and HiPerFect Transfection Reagent (Qiagen) in six technical replicates. Stable lentiviral-transduced cells were also seeded in 96-well plates. Cells were incubated from 24 to 96 h, and every 24 h, the WST-1 proliferation reagent was added in the wells according to the manufacturer’s recommendations. Colorimetric changes were measured using a SpectraMax Plus (Molecular Devices, Germany) at a wavelength of 450 nm with a reference wavelength at 620 nm. Three independent biological replicates were performed each.

To confirm the cellular growth changes after altered miR-1287-5p expression levels by a second independent method, a clonogenic (i.e., colony formation assay) was performed. Transient transfected cells were trypsinized 24 h after transfection. After trypsinization, cells were counted and seeded for colony formation assay in six-well plates at 100–500 cells/well, depending on the cell line. Cells were cultured at 37 °C and 5% CO₂ and after 10–21 days, cells were fixed as well as stained with 0.01% (w/v) crystal violet (Sigma-Aldrich) in 20% methanol and PBS. The number of colonies was counted, and each experiment was carried out in biological and technical triplicates.

To measure the effects of miR-1287-5p in vivo, we injected SUM159 cells stably overexpressing miR-1287-5p or control transduced cells into 5-week-old female nude mice (NU/NU Crl:NU-Fox1nu, Charles River Laboratories; Sulzfeld, Germany). Briefly, 1 × 10⁶ cells were resuspended in phosphate-buffered saline (PBS; 1:1 mixed with matrigel, Corning) and subcutaneously injected into the mammary fat pad of the mice. Cells with miR-1287-5p overexpression were injected in the left mammary fat pad, control cells in the right. A total number of seven mice were used for the in vivo experiments. Tumor growth was monitored every few days by caliper measurements, and animals were sacrificed before tumors reached a diameter of 10 mm. Tumors were harvested for histological analyses, and tumor volumes were calculated by the equation V(mm³) = (width)² × length/2. All animal work was done in accordance with a protocol approved by the Institutional Animal Care and Use Committee at the Austrian Federal Ministry for Science and Research (BMWF) (BMWFW-66.010/0046-WF/V/3b/2016). For experiments with the miR-1287 precursor overexpressing cells, xenografts were independently generated by an external company. This experiment has been performed by a commercially external facility (EPO Berlin-Buch GmbH, Berlin, Germany) which was completely blinded to the other results of this study. 1 × 10⁶ SUM159 cells were injected subcutanously in mammary fad-pad in a volume of 20 μl to NMRI:nu/nu mice (Janvier Labs, Paris, France). Per group, seven mice were inoculated with cells. All animal experiments were performed under the guidelines of the German Animal Protection Law and with approval by the local responsible authorities. Mice were observed daily for their health status. Mice were sacrificed after 4 weeks after cell inoculation, and the size of tumors relative to the start volume was measured.

Caspase-Glo 3/7 assay (Promega, Madison, WI, USA) was applied to measure the activity of caspase 3 and 7 according to the manufacturer’s instructions. For transient transfection experiments, BC cell lines were seeded in 96-well plates and transfected using the reverse transfection protocol and HiPerFect Transfection Reagent (Qiagen) in five technical replicates. Caspase 3 and 7 activity was measured 48 h after transient transfection. Lentiviral stable-transduced cells were also seeded in 96-well plates, and apoptosis was measured after 48 h. After adding the substrate, luminescence was measured using a luminometer (LumiStar, BMG Labtech, Ortenberg, Germany).

A number of 100,000 cells were seeded and transfected using the fast-forward protocol (Qiagen) in six-well plates and after 48-h cell cycle analysis was performed. Cells were fixed in 75% ethanol overnight at 4 °C, resuspended in 0.2% FBS/PBS, RNAse A treated (Qiagen, 100 μg/mL), stained with propidium iodide (PI; Sigma-Aldrich) at a final concentration of 40 μg/mL, and analyzed by flow cytometry at a BD LSRII Flow Cytometer. All analyses were performed in triplicates, and 10,000 gated events per sample were counted.

The efficiency of colony formation of cells with altered miR-1287-5p expression in soft agar was determined by plating 2500 cells in complete growth medium containing 0.35% low gelling temperature agarose (Sigma-Aldrich) over 2 ml of growth medium containing 0.5% agar (Sigma-Aldrich) in a 35-mm dish. Cells were cultured at 37 °C and 5% CO₂ for up to 4 weeks. Colonies were stained with 0.005% crystal violet (Sigma-Aldrich) in 20% methanol, and the number of colonies was counted using a microscope.

BC cell lines were transiently transfected with a short-interfering siRNA for PIK3CB (phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit beta) (Hs_PIK3CB_5, Qiagen, 20 nM) to knockdown the gene of interest using the fast-forward transfection method in six-well plates and the reverse transfection method in 96-well plates according to the HiPerFect Transfection Reagent protocol (Qiagen). All Stars Negative Control siRNA (Qiagen, 20 nM) was used as negative control; AllStars Cell Death Control siRNA (Qiagen, 20 nM) was used to confirm transfection efficiency.

To detect putative miR-1287-5p target mRNAs, we applied mRNA microarrays in stable miR-1287 overexpressing SUM159 cells. Total RNA from lentiviral-transduced SUM159 miR-1287-5p overexpressing cells versus control cells in biological triplicates was isolated using the miRNeasy Mini Kit (Qiagen; Hilden, Germany; Cat No. 217004) according to the manual. The whole transcriptome analysis was performed on Affymetrix Human Gene 2.0 ST mRNA Arrays (Affymetrix; Santa Clara, CA, USA) as previously described. In detail, 250 ng of the total RNA was amplified with Affymetrix WT PLUS Reagent Kit (Affymetrix; Santa Clara, CA; Cat No. 703147) as instructed by the manufacturer’s protocol. Additionally, the cDNA was quality checked on the BioAnalyzer BA2100 (Agilent, Foster City, CA) using the RNA 6000 Nano LabChip (Agilent; Foster City, CA; Cat.No. 5065-4476). An examination of ~ 250 ng generated cRNA showed a fragment size > 2000 nt which was satisfying for further processing. The hybridization cocktail was prepared as suggested by the manual and hybridized on the arrays for 18 h at 45 °C while rotating in the hybridization oven. Washing and staining (GeneChip® HT hybridization, Wash and Stain Kit; Affymetrix, Santa Clara, CA; Cat No. 900720) was done with the Affymetrix Genechip® fluidics station 450 according to the manual (protocol on fluidics station: FS450_0002). Arrays were scanned with the Affymetrix GeneChip scanner GCS3000.

The evaluation of the hybridization controls and pre-analysis was done with Affymetrix Expression Console EC 1.3.1. Hybridizations were done at the Division Core Facility Molecular Biology at the Centre of Medical Research at the Medical University of Graz. Data pre-processing and filtering was performed using Partek Genomics Suite, v.6.6 (RMA (background correction, quantile normalization across all chips in the experiment, log2 transformation, median polish summarization)). Raw data are available at the Gene Expression Omnibus (accession number GSE103388).

We performed a bioinformatics target prediction based on global gene expression analysis and comprising to distinct approaches as previously described [25, 26]. Gene expression data was filtered for transcripts that were at least 2-fold downregulated by miR-1287-5p (versus control cells). This subset was merged with an array of miRNA-target predictions (comprising 12 distinct algorithms) obtained from the miRWalk 2.0 database [27]. Then, 3′ UTR sequences of the subset of transcripts were obtained from ENSEMBL. These sequences were subsequently screened for the presence of distinct miR-1287-5p seed match types (8mer, 7mer-A1, 7mer-m8, 6mer, and offset 6mer sites) [8].

To confirm the direct interaction of miR-1287-5p and the putative target PIK3CB, a 65 nt region of the predicted 3′ UTR binding site of PIK3CB was cloned into a luciferase containing pEZX-MT06 Vector (Genecopoeia, Rockville, MD, USA), either the wild-type miR-1287-5p target sequence or the mutated sequence. An empty control plasmid (Genecopoeia) was used as a reference control. For the Luciferase assay, HEK cells were seeded in 24-well plates on the day prior to the transfection. After 24 h, cells were co-transfected with 200 ng pEZ-MT06miRNA reporter vector (wild-type = 5′ TGGGTGATCTCTCTGAGTCCTGGCAACATCCAGCAAAACTACTGCTTATTCTCCAAAGAATATTGG 3′ CS-HmiT011246-MT06–01; GeneCopoeia), mutated plasmid (5′ TGGGTGATCTCTCTGAGTCCTGGCAACATTTATTAAAACTACTGCTTATTCTCCAAAGAATATTGG 3′ CS-HmiT011246-MT06-01; GeneCopoeia) or negative control (empty plasmid CS-HmiT011246-MT06–01; GeneCopoeia), and 50 nM miR-1287-5p Mimic or AllStars Negative Control (Syn-hsa-miR-1287-5p miScript miRNA Mimic or All Stars Negative Control, Qiagen), using Lipofectamine 2000 Transfection Reagent (Thermo Scientific, Waltham, MA USA) and Opti-MEM Reduced Serum Medium (Thermo Scientific). Cells were harvested 24 h after transfection and the Luc-Pair Luciferase Assay Kit 2.0 was performed according to the user manual. Luminescence was measured using a luminometer (LUMIStar Omega, BMG LabTech, Ortenberg, Germany) in three independent biological replicates. The ratio of luminescence from the firefly luciferase to the Renilla luciferase was calculated, and the empty control plasmid was used to normalize the luciferase activity.

For selective PI3K inhibitor experiments, untreated or transiently transfected BC cells were additionally treated with 2 μM and 10 μM BYL719 (Alpelisib, a selective PI3Kα inhibitor) and 25 μM and 50 μM CAL-101 (Idelalisib, a selective p110δ inhibitor) in a 96-well plate. All inhibitors were purchased via Selleckchem.​com (Eubio, Vienna, Austria). Exposure time was 96 h before applying the WST-1 assay and compared to untreated control cells.

All statistical analyses were performed using SPSS version 23 software (SPSS Inc., Chicago, IL, USA). Unpaired Student’s t test or Mann-Whitney U test was applied. A two-sided p < 0.05 was considered statistically significant.