Ethics statement and study cohort

The study received MRC-01–19–142 ethical approval from Hamad Medical Corporation (HMC) and QBRI-IRB 2020–09–035 approval from Qatar Biomedical Research Institute (QBRI). Clinicopathological features of the study cohort are presented in Table 1. Detailed description of the study cohort and inclusion/exclusion criteria can be found in our recent publication [19].

Table 1 Clinical characteristics of study cohort​Total RNA isolation and next-generation sequencing (NGS)

Total RNA isolation from FFPE tissues was conducted as we described before [19]. For miRNA library preparation, 100 ng of total RNA was used as input for 3′ ligation, followed by 5′ ligation and reverse transcription using the QIAseq miRNA library kit (QIAGEN, Hilden, Germany). The resulting cDNA libraries were quantified using the Qubit dsDNA HS assay kit and assessed for size distribution using the Agilent 2100 Bioanalyzer DNA1000 chip. Pooled libraries underwent sequencing on the Illumina platform. The miRNA transcriptomic data were deposited in the SRA repository under BioProject number PRJNA953015. For miRNA analysis, FASTQ files were mapped to the miRBase v22 database and the miRNA expression (total counts) were calculated using the small RNA analysis workflow in CLC Genomics Workbench 21.0.5. Expression data were then imported into iDEP.951, normalized (CPM, count per million), and log-transformed using EdgeR (log2(CPM + c)). miRNAs with a minimum expression of 5 CPM in at least 10 samples were retained. Hierarchical clustering and identification of differentially expressed miRNAs (DEMs) in relation to PAM50 intrinsic subtypes were conducted in iDEP.951 using 1.5-fold change (FC) and false discovery rate (FDR) p < 0.05.

Survival analysis

To identify the set of miRNAs correlating with ten-year patients’ RFS, the normalized miRNA expression data (log2 transformed) underwent RFS analysis using RStudio 2021.09.2. The ‘survival’ package in R was employed to compute the log-rank p values and hazard ratio (HR). Initial identification of potential miRNA candidates associated with RFS was conducted through univariate survival analysis and subsequently, these identified miRNA candidates underwent multivariate cox regression survival analysis using IBM SPSS Statistics v26 to adjust for potential confounding factors (molecular subtype, tumor grade, and age). The survival plot was generated by stratifying the patient cohort into high and low groups based on the median hsa-miR-5683 expression. Log-rank p value was used for curve comparison.

Validation of miR-5683 expression and survival analysis in additional breast cancer datasets

To validate the expression of hsa-miR-5683 in additional breast cancer cohorts, we investigated its expression in different breast cancer subtypes compared to 104 normal controls from the ExplORRnet database (https://mirna.cs.ut.ee/). Survival analyses based on hsa-miR-5683 expression were subsequently validated using the KM Plotter database (https://kmplot.com/analysis/index.php?p=background). Correlation between hsa-miR-5683 and its target mRNA expression were calculated using the ENCORI database (https://rnasysu.com/encori/).

Cell culture and transfection

The human TNBC cell lines (MDA-MB-231 and BT-549) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with D-glucose at a concentration of 4500 mg/L, 2–4 mM L-glutamine, 10% fetal bovine serum, and 1× penicillin-streptomycin (Pen-Strep), all purchased from GIBCO-Invitrogen (Waltham, MA, USA). Cells were maintained in humidified CO2 (5%) incubator at 37 °C. The hsa-miR-5683 mirVana miRNA mimic (Assay ID: MC22887) and miRNA negative control (scrambled) were purchased from Thermo Fisher Scientific (Thermo Fisher Scientific, Waltham, MA). In the current study, we utilized the reverse transfection protocol where miRNA mimic and control were diluted in 50 µL of Opti-MEM (GIBCO, Carlsbad, CA, USA), and 1.5 µL of Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA) was diluted in 50 µL of Opti-MEM. The resulting miRNA and Lipofectamine 2000 mixtures were combined and incubated at room temperature for 20 min. The cell count per well was 1.68 × 105 cells (MDA-MB-231) and 8.4 × 104 cells (BT-549) in 2.4 mL transfection medium (complete DMEM without Pen-Strep). This was followed by the addition of 0.8 mL of the transfection mixture to the 6-well tissue culture plate resulting in final concentration of 30 nM. After 24 h, the transfection cocktail was topped up with complete DMEM. Transfection of the MCF7 HR + breast cancer model was carried as detailed above, using 0.75 µl of lipofectamine 2000.

Cell line authentication by STR

Genomic DNA (gDNA) extracted from MDA-MB-231 and BT-549 TNBC lines was used as input for STR profiling using AmpFLSTR Identifiler PCR amplification kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) following the manufacturer’s protocol. Briefly, starting from 1 ng of gDNA, DNA was subjected to PCR amplification. Positive and negative controls were run in parallel with the samples. After amplification, PCR products were prepared for electrophoresis by adding Hi-Di Formamide and size standard mixture to each sample and allelic ladder. Electrophoresis was performed in the Genetic Analyzer 3500xl DX system. Allelic calls analysis is performed by Gene Mapper software from Applied Biosystems/Thermo Fisher Scientific. Reference STR profiles were retrieved from ATCC (https://www.atcc.org/).

Reverse transcriptase quantitative polymerase chain reaction (RT-qPCR)

To validate hsa-miR-5683 overexpression efficiency, total RNA was extracted 72 h post transfection using miRNeasy Mini Kit (Qiagen Inc., Hilden, Germany). The concentration and quality of extracted RNA was measured using NanoDrop 2000 (Thermo Scientific, DE, USA). Following the RNA extraction, reverse transcription PCR was performed using the miRCURY LNA RT Kit (Qiagen Inc., Hilden, Germany). The generated cDNA was then used to perform qPCR using the miRCURY LNA SYBR Green PCR kit (Qiagen Inc., Hilden, Germany) to measure the level of hsa-miR-5683 expression in mimic and control transfected TNBC cells. The relative FC in miRNA expression was calculated using the 2−ΔΔCt method, where the average of ΔCt values for the target amplicon was normalized to that of SNORD44 endogenous control and compared to negative control transfected samples.

The candidate genes identified as bona fide targets for hsa-miR-5683 were validated in both MDA-MB-231 and BT-549 post miRNA mimic transfection using RT-qPCR. RNA extracted from both TNBC cell lines (500 ng) were reverse transcribed to cDNA using the High-Capacity cDNA Reverse Transcription kit (Thermo Fisher Scientific, Waltham, MA). Subsequently, qPCR was performed using specific primer pairs as detailed in Table S1 and the PowerUp SYBR Green Master Mix (Thermo Fisher Scientific, Waltham, MA) on QuantStudio 6 Flex qPCR system (Applied Biosystems). The mRNA transcript levels of the target genes were determined based on their respective CT values, normalized against β-actin (ACTB) transcript levels, and presented as FC using the 2−ΔΔCt method compared to control cells.

Proliferation assay

MDA-MB-231 and BT-549 cells were transfected with hsa-miR-5683 mimic or negative miRNA control as described above. Subsequently, the ramifications of hsa-miR-5683 expression on TNBC cells were assessed using proliferation assay. On day 7, cells were washed twice using phosphate-buffered saline (PBS). Subsequent staining with crystal violet (0.1% in 10% ethanol) was conducted and plates were placed on a shaker for 2–3 h. Plates were then air-dried at room temperature before imaging, and quantification by dissolving crystal violet in 10% SDS, and subsequent measurement of absorbance at 590λ.

Three-dimensional (3D) and spheroid culture

For the generation of 3D cultures, cells under different treatment conditions were pelleted at a concentration of 0.25 × 106 cells/mL and were subsequently combined with Matrigel (Corning; 356231; Growth Factor Reduced Basement Membrane Matrix). Subsequently, several drops of the cell suspension (one drop/well, approximately each drop containing 10,000 cells in 40 µl of Matrigel) were dispensed into pre-warmed (37 °C) Ultra-Low Attachment 24-well Culture plates (Corning Corp., Bebford MA, USA). The plates were then inverted and placed in a 37 °C, 5% CO2 cell culture incubator for 20 min to allow the droplets to solidify and form dome structures. Subsequently, 1–2 mL of expansion medium was added to cover the dome. Organoid formation under different experimental conditions were observed using microscope on the indicated dates. For spheroid formation, MDA-MB-231 and BT-549 under different treatment conditions were trypsinized, and re-suspended in culture media, and were then seeded in 60 mm low cell binding dishes (Corning Corp., Bebford MA, USA) at a density of 6 × 104 cells/dish, following our previously described protocol [21]. On day 10, multicellular tumor spheroids were observed using an inverted microscope (Axio Observer-A1, Carl Zeiss, Germany) at 4X magnification.

Dead/live cells staining using AO/EtBr fluorescent microscopy

Fluorescence microscopy was employed for the detection of cell death under different experimental conditions. Briefly, the AO/EtBr fluorescence staining method was used to assess cell death on day 5 post transfection of TNBC cell with hsa-miR-5683 mimic or negative control. Subsequently, cells were washed twice using PBS before staining with a dual fluorescent solution containing 100 µg/mL AO and 100 µg/mL EtBr (AO/EtBr, Sigma Aldrich, St. Louis, MO, USA) for 2 min. The stained wells were then observed and imaged using an Olympus IX73 fluorescence microscope (Olympus, Tokyo, Japan). AO staining was used to visualize nuclei, while EtBr-positive cells indicated the presence of necrotic cells.

Cell cycle analysis using flow cytometry

Flow cytometry was employed for cell cycle analysis of MDA-MB-231 and BT-549 post-transfection with hsa-miR-5683 mimic or negative control. On day 4 post-transfection, both floating and adherent cells were collected and were then washed, fixed with ice-cold 70% ethanol, and stored at -20 °C. The cells were washed twice using PBS and were subsequently incubated in RNase A (100 µg/mL) followed by addition of propidium iodide (PI; 50 µg/mL). Stained cells were run through a BD LSRFortessa X-20 flow cytometer (BD Biosciences, CA, USA) and events were recorded using the FL3 channel. Data were subsequently analyzed using FlowJo software (FlowJo 10.7.2, BD Biosciences, CA, USA).

Identification of potential gene targets for hsa-miR-5683

To identify potential gene targets for hsa-miR-5683, MDA-MB-231 and BT-549 were transfected with hsa-miR-5683 mimic or negative control. At 72 h, total RNA was extracted from transfected cells, and subsequently library preparation was done using the TruSeq Stranded Total RNA Library (Illumina Inc., San Diego, CA, USA) followed by NGS using the Illumina HiSeq 4000 at approximately 30 million paired end reads (2 × 75 bp) per sample as we described before [22]. The generated FASTQ files were subsequently aligned and mapped to the hg38 reference genome using the CLC Genomics Workbench v21.0.5. Subsequently, iDEP.951 was employed for differential expression analysis to identify differentially expressed genes (DEGs) in hsa-miR-5683 compared to negative control transfected TNBC cells using 2.0 FC and p (FDR) < 0.1. To identify bona fide gene targets for hsa-miR-5683, the TargetScan (version 80) in silico miRNA target prediction database was employed. The microRNA Target Filter in Ingenuity Pathway Analysis (IPA) was utilized to construct the hsa-miR-5683-mRNA network. The identified targets resulting from this integration were further validated using RT-qPCR.

Statistical analysis

For DEGs and DEMs, analyses were performed in iDEP.951. A FC of 1.5 and an FDR-adjusted p-value < 0.05 served as the cutoff unless otherwise specified. Graphing and pairwise statistical analyses were executed in GraphPad Prism v9 and GraphPad Prism v10.