Abstract Background The high mortality rate of breast cancer and the difficulties associated with therapeutic resistance, especially in cases where targeted treatments are unavailable, make it a serious threat to women's health. This study examines the relationship between three mature microRNAs (miRNAs) that are clustered together, namely miR- 23a, miR- 27a, and miR- 24–2, as well as their potential correlation with breast cancer. Methods We identified common gene targets of miR- 23a, miR- 27a, and miR- 24–2 using computational analysis. We also checked for the levels of miR- 23a, miR- 27a, and miR- 24–2 in 26 breast tumor tissues (with their matched control) as well as MCF7 and MDA-MB- 231 cell lines using qRT-PCR. Dual-luciferase reporter assay was conducted to validate the binding site of the microRNAs in their target gene. Western blot was performed to study the expression of various breast cancer related genes in the presence of the three microRNAs. In addition, the effect of microRNAs in cancer cell metastasis and cell division was carried out using invasion and cell cycle assay. Results Computational analysis identified key genes, including GSK3β, NCOA1 and SP1, which are functionally linked to tumor progression and various other malignancies. All three microRNAs were found to be significantly downregulated in the breast cancer tissue samples in comparison to their respective controls. Kaplan–Meier plot analysis revealed that the expression levels of these genes and associated microRNAs correlates with breast cancer patient survival rates. Reduced SP1 and NCOA1 levels predicted a worse prognosis, but elevated levels of GSK3β were linked with decreased survival. Moreover, miR- 23a and miR- 24–2 specifically target GSK3β, potentially disrupting the Wnt/β-catenin pathway involved in breast cancer development. Functional tests showed that miR- 23a, miR- 27a and miR- 24–2 affect expression of EMT related genes, influencing cell invasion and migration, impacting ERK signaling and EMT, critical in the spread of breast cancer. Conclusion This study unlocks the potential of targeting the microRNA cluster as a therapeutic approach and emphasizes the complex regulatory roles of each individual members of the miR- 23a/27a/24–2 cluster in the pathogenesis of breast cancer. Supplementary Information The online version contains supplementary material available at 10.1186/s12885-025-14045-8. Keywords: Breast cancer, MicroRNA, Migration, Invasion, GSK3β, SP1 Background Breast cancer is one of the most frequently diagnosed cancers among women globally, with a complex and heterogeneous nature characterized by diverse cell compositions, proliferation rate and responses to treatments, resulting in varying patient outcomes [[42]1, [43]2]. This diversity is evident in the different molecular subtypes, defined by the expression of specific proteins such as estrogen (ER) and progesterone (PR) receptors, human epidermal growth factor receptor 2 (HER2) and the proliferation marker, Ki67 [[44]3, [45]4]. Triple Negative Breast Cancer (TNBC) which lacks the expression of ER, PR and HER2 receptors accounts for nearly 15–20% of newly diagnosed breast cancer cases and poses a significant challenge due to absence of targeted treatments, contributing substantially to breast cancer-related mortality [[46]5]. MicroRNAs are short, non-coding RNA sequences typically spanning around 18 to 25 nucleotides [[47]6]. They play a crucial role in regulating gene expression by binding to complementary sequences in the 3’ untranslated region (UTR) of target mRNAs, leading to either inhibition of translation or degradation of the mRNAs [[48]6, [49]7]. MiRNAs have a substantial role in controlling several physiological processes, such as cell division, proliferation, and apoptosis [[50]8]. Evidence suggest that distinct types of human cancers exhibit unique miRNA expression patterns, making them useful biomarkers for cancer diagnosis, progression, and prognosis [[51]9]. Our group's earlier research has demonstrated that aberrant gene expression of the miR- 23a, miR- 27a and miR- 24–2 in HEK293 cells can induce apoptosis [[52]10]. Moreover, ER stress is identified by microarray analysis as the primary initiator of the intrinsic pathway of apoptosis [[53]11]. The miR- 23a/27a/24–2 cluster encodes three mature miRNAs derived from a single transcript, which display varying expression levels across multiple disease states [[54]11]. Further, these three miRNAs' dysregulated expression has been linked to several human malignancies [[55]11]. In the current study our findings suggest the roles and relationships of individual members of miR- 23a/27a/24–2 cluster in breast cancer pathogenesis, focusing on their effects on common target genes (NCOA1, GSK3β, and SP1), along with their influence on epithelial-mesenchymal transition (EMT), cancer cell invasion and cell cycle. The study seeks to explore the potential of targeting these miRNAs for therapeutic interventions in breast cancer. Materials and methods In-silico prediction of microRNAs TargetScan 8.0 ([56]www.targetscan.com) was used to predict the targets of hsa-miR- 23a- 3p, hsa-miR- 24- 3p, and hsa-miR- 27a- 3p. Highly conserved targets of these three miRNAs were taken into consideration and Venny 2.1.0 was used to find the common genes between them ([57]https://bioinfogp.cnb.csic.es/tools/venny/). Pathway enrichment analysis of these common target genes was done using the DAVID pathway analysis tool ([58]https://david.ncifcrf.gov/tools.jsp; v2023q3) [[59]12, [60]13] The list of highly conserved predicted target genes of hsa-miR- 23a- 3p, hsa-miR- 24- 3p, and hsa-miR- 27a- 3p is provided in Supplementary Table 1. Construction of the protein–protein interaction (PPI) network To study the interaction of the common target genes with other genes, we used STRING 12.0 ([61]https://string-db.org/). The protein–protein interactions score was set to a medium confidence of 0.400 (default setting) and were searched against the organism Homo sapiens. Plasmid constructs miRBase was used to retrieve the miRNA sequences ([62]http://microrna.sanger.ac.uk/sequences/). Primers were designed using Primer 3 software ([63]http://frodo.wi.mit.edu). Hsa-miR- 23a, hsa-miR- 24-2, and hsa-miR- 27a were cloned in pSilencer 4.1 vector (Ambion, Austin, TX, USA) and now onwards designated as miR- 23a, miR- 24-2 and miR- 27a. ENSEMBL was used to retrieve the 3’ UTR of the GSK3β sequence ([64]www.ensembl.org/) and human genomic DNA was used to amplify it. Two different regions encompassing the binding sites for miR- 23a, miR- 24-2, and miR- 27a were amplified and cloned in psiCHECK2 reporter vector backbone (Promega, Madison, WI, USA) between Not1 and Xho1. The clones were designated as GSK3β 3’UTR 1 (containing seed sequence of miR- 23 and miR- 27, 840 bp) and GSK3β 3’UTR 2 (containing the seed sequence of miR- 24-2, 141 bp). The accuracy of the all the clones was confirmed by sequencing. Primers used are listed in Table [65]1. Table 1. Primer sequences used in this study Gene name Primer type Primer sequence hsa-miR- 23 - 3p Stem loop 5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGAATGTGAT- 3′ Forward primer 5′-ACACTCCAGCTGGGATCACATTGCCAGGG- 3′ hsa-miR- 24 - 3p Stem loop 5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGCTGAGCCA- 3′ Forward primer 5′-ACACTCCAGCTGGGTGGCTCAGTTCAGCAG- 3′ hsa-miR- 27 - 3p Stem loop 5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGACTGTGAA- 3′ Forward primer 5′-ACACTCCAGCTGGGTTCACAGTGGCTAAG- 3′ miR- 23 - 3p, miR- 24 - 3p, miR- 27 - 3p Universal reverse primer 5′-GTGTCGTGGAGTCGGCAATTC- 3′ hsa-sno- 234 Stem loop 5'-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAG- 3’ Forward primer 5′-GGCTTTTGGAACTGAATCTAAGT- 3′ Reverse primer 5′-GAGGTATTCGCACCAGAGGA- 3′ U6 Forward primer 5’-CTCGCTTCGGCAGCACA- 3’ Reverse primer 5’-AACGCTTCACGA ATTTGC GT- 3’ GSK3β 3’UTR 1 Forward primer 5’-CCGCTCGAGTTGCCTTTGGCATGTTGGTG- 3′ Reverse primer 5’-ATAAGAATGCGGCCGCTCTTAACTGGGTGTGGGGGA- 3′ GSK3β 3’UTR 2 Forward primer 5’-CCGCTCGAGACACTTTTCTCCCCTGTGTGG- 3′ Reverse primer 5’-ATAAGAATGCGGCCGCGCCTCTCAAAGGTAGATCTCAGT- 3′ [66]Open in a new tab Cell culture and transfections Human breast adenocarcinoma cell lines, MCF- 7 and MDA-MB- 231 were procured from the National Centre for Cell Science (NCCS), Pune, India. Cells were cultured using DMEM-high glucose media (D7777, MERCK, Darmstadt, Germany) supplemented with 10% fetal calf serum, 100 units/ml penicillin and 100 µg/ml streptomycin, and 0.25 μg/ml amphotericin (Gibco, Thermo Fisher Scientific, MA, USA) and grown in a humidified chamber supplemented with 5% CO[2] at 37° C. For overexpression studies, cells were transfected with Empty pSilencer vector 4.1 (pSil) or miR- 23a or miR- 24-2 or miR- 27a using Lipofectamine 3000™ (Invitrogen, CA, USA) reagent as recommended by the manufacturers. Anti-miR- 23 - 3p, anti- miR- 24 - 3p, and anti-miR- 27 - 3p were used at 100 nM (Ambion, TX, USA) wherever indicated and anti-miR inhibitor negative control (NC) (AM17010, Ambion) was used as a control. The cells were harvested 24 h post transfection and used for all experiments as described in our previous lab publication [[67]10]. Collection of primary tissues and patient characteristics In our study primary- paired tissues from 26 breast cancer patients were obtained after surgical resection from the Surgery Department at All India Institute of Medical Sciences, Delhi, India. The paired adjacent tissue was at least 2 cm away from the tumor. The research complies with all the relevant national regulations, institutional policies and is in accordance with the tenets of the Helsinki Declaration and has been approved by the Institutional Ethical Committee of CSIR-IGIB, Delhi and AIIMS, New Delhi (IEC- 726/01.10.2021, RP- 67/2022). A written and signed informed consent form was obtained from all the participants. Bioinformatics based assessment of prognostic significance of genes Correlation of genes with their expression and overall-survival rates in breast cancer patients was assessed by Kaplan–Meier plots generated through Kaplan–Meier plotter ([68]www.kmplot.com) [[69]14]. METABRIC database was used over TCGA for its longer follow-up duration and better cohort. GSK3β [Affymetryx ID 209945_s_at], SP1 [Affymetryx ID 1553685_s_at], and NCOA1 [Affymetryx ID 209105_at] were used to generate Kaplan–Meier plots for genes as well as for microRNA hsa-miR- 23a, hsa-miR- 24 and hsa-miR- 27. We excluded biased arrays from our analysis. Luciferase assay MCF- 7 cells were seeded in 24-well plate at a density of 30,000 cells per well and co transfected with GSK3β 3’UTRs plasmid (150 ng) and either with 400 ng of miR- 23a or miR- 24–2 or miR- 27a or 40 nm of anti-miR- 23 - 3p, anti- miR- 24 - 3p, and anti-miR- 27 - 3p along with their respective controls using Lipofectamine 3000 (Invitrogen,CA, USA). Cells were lysed to measure the activity of firefly and renilla luciferase using the dual-luciferase reporter assay (Promega, MA, USA) according to the manufacturer’s instructions. Renilla reporter was used to normalize the firefly luciferase activity. These experiments were performed in triplicates. Total RNA isolation and miRNA quantification RNA was isolated from cells (MCF- 7 and MDA-MB- 231) and primary breast cancer tissues using Trizol reagent (Invitrogen, CA, USA) and miReasy Tissue/cells Advanced micro kit (Qiagen, Hilden, Germany) as per manufacturer’s protocol respectively. Reverse transcription was carried out with 500 ng of RNA using M-MuLV reverse transcriptase High-capacity cDNA reverse transcription kit (Applied Biosciences, USA) as per the instructions provided by the manufacturer. miRNA quantification was done using stem-loop primers as mentioned in Table [70]1. Sno- 234 expression was used for normalization and U6 primers were used for breast cancer patient tissue samples. Results were analysed using Pfaffl's method [[71]15]. Western blot Protein isolation was done using RIPA lysis buffer supplemented with protease inhibitor cocktail mix (Merck, Darmstadt, Germany). Protein estimation was done by BCA (Sigma, USA). 40–60 ug protein was subjected to electrophoresis using 12% SDA-PAGE at 100 V for 3 h and transferred onto PVDF membrane (MDA, Advanced Microdevices, India). Blots were probed with primary antibodies overnight at 4 °C and were developed using enhanced chemiluminescence assays (ECL detection system; G Biosciences, USA). The integrated densitometric values (IDV) were measured using Alphaimager software (Alpha Innotech Corporation, CA, USA). GSK3β (D5 C57), β-catenin (D2U84), NCOA1 (ab1093281), claudin- 1 (ab101430), Keratin8/18 (4546S), Fibronectin (E5H6X) and GAPDH (D1GH11) antibodies were purchased from Cell Signalling Tech (Cell Signalling Tech, Danver, MA, USA). SP1 (ab308364) and cyclin D1 (ab101430) antibody were procured from Abcam (Abcam, MA, USA). E- cadherin (sc- 7870), pERK (sc7383), total ERK (sc- 514302) and c-myc (sc- 764) was from Santa Cruz Biotechnology (Santa Cruz Biotechnology, CA, USA). GAPDH was used as house-keeping gene control. Migration and invasion assay BioCoat™ matrigel invasion chambers (BD Biosciences, NJ, USA) were used to assess the invasive capacity of cells after miRNA or AM transfections. Briefly, invasion chambers were kept at room temperature and hydrated with serum-free media before adding the cells. 24 h post treatment with miR- 23a, miR- 24–2, and miR- 27a and their respective Anti-miRs, cells were trypsinized, counted and suspended in an incomplete (serum-free) medium. Approximately 40,000 cells per well of a 12 well plate were seeded in the upper chamber and were allowed to invade for 24 h. After completion of the incubation period, non-invaded cells were wiped off using cotton bud and invaded cells were stained with crystal violet and images were taken as described previously. Migration and invasion of cells was analysed through colorimetric assay by dissolving the membrane in 10% acetic acid and absorbance was taken at 570 nm. Fold change was calculated for same by dividing absorbance of samples by absorbance of controls for both the cell lines. Cell cycle analysis For cell cycle distribution analysis, cells were initially fixed for 2 h using ice-cold 70% ethanol. Subsequently, they were treated with RNase A and labelled with propidium iodide (Sigma, USA) for 1 h in darkness. The data was acquired through a flow cytometer (BD FACSAria™ III, USA) with a minimum of 10,000 events counted. Statistical analysis Statistical analysis of the results was performed using unpaired student's t-test using GraphPad Prism Software (v.10.2.2 GraphPad, Inc., CA, USA). We performed an independent student’s two-tailed, paired t test for the analysis of qRT-PCR of breast cancer patients along with their adjacent control and for cell cycle analysis. Unpaired t test was used for the analysis of western blots and qRT-PCR of miRNAs in MCF- 7 and MDA-MB- 231 cells. Mean of n = 3 ± SEM was plotted. P-value ≤ 0.05, < 0.01, < 0.001 and < 0.0001 corresponds to *, ** and ***, ****respectively, on plots. Results Target prediction and pathway enrichment analysis of miR- 23a, miR- 24-2 and miR- 27a target genes Derived from gene duplication within the human genome, the miR-23–27-24 cluster has 2 paralogs: miR- 23a cluster (miR- 23a/27a/24–2) situated on chromosome 9q22 and miR- 23b cluster (miR- 23b/27b/24–1) located on chromosome 19p13 [[72]11, [73]16]. MiR- 23a cluster codes for miR- 23a, − 27a, and − 24–2 and all these are known to be involved in various cellular processes, like proliferation, cell cycle, differentiation and several human diseases including cancer [[74]11, [75]17]. In the current study, we retrieved highly conserved predicted targets of the hsa-miR- 23a- 3p (1342 genes), hsa-miR- 24 - 3p (761 genes), and hsa-miR- 27a- 3p (1421 genes) using target prediction database TargetScan version 8.0 as shown in Supplementary Table 1. We next used Venny 2.1.0 tool and found a total of 42 common predicted gene targets between the three miRNAs (Fig. [76]1A and Table [77]2). Functional enrichment analysis of 42 common genes was next done using DAVID software [[78]12, [79]13]. Results of the analysis demonstrated enrichment of a total of four KEGG pathways, namely, MAPK signaling, Hedgehog signaling, Cushing syndrome, and Breast cancer as shown in Fig. [80]1B [[81]18, [82]19]. Breast cancer showed significant enrichment (p-value = 0.027) with three core genes SP1, GSK3 [MATH: β :MATH] , and NCOA1. Simultaneously we also checked the protein–protein interaction network of these 42 common target genes using STRING version 12.0. The interaction score was set to 0.400 and within the PPI network GSK3 [MATH: β :MATH] , SP1, and NCOA1 emerged to have a significant network with APC (Adenomatous polyposis coli) and CSNK1A1 (Casein kinase 1 alpha) (FDR = 0.0097) as shown in Fig. [83]1C. APC and CSNK1A1 are important components of the destruction complex and are known to negatively regulate canonical Wnt-signaling by phosphorylating β-catenin in the absence of Wnt ligands, leading to its ubiquitination and ultimate proteasomal destruction [[84]20]. Abnormal activation of Wnt-signaling is known to be linked with cell proliferation, evasion of apoptosis, and metastasis [[85]21, [86]22]. Fig. 1. [87]Fig. 1 [88]Open in a new tab Bioinformatic analysis and association of the components of miR- 23a/27a/24–2 cluster in Breast Cancer using string analysis and Kaplan–Meier survival curve. (A) Venn diagram and list of highly conserved predicted target genes of miR- 23a, miR- 24–2, and miR- 27a using TargetScan (B) Bubble plot represents the pathway enrichment analysis of common predicted target genes using DAVID tool. Gene count is represented by the size of the bubbles and colour gradation represents the negative log10 of p-value. (C) PPI network of common predicted target genes was constructed using STRING database (D) Kaplan–Meier curves of GSK3β, SP1 and NCOA1 showing survival probabilities of the two contrasting groups in breast cancer Table 2. Forty-two common gene targets of individual members of miR- 23a/24–2/27a cluster and their function in cancer Gene symbols Gene name Function in cancer Reference NEK6 NIMA related kinase 6 Regulates cancer cell invasion and metastasis [[89]23] NR6A1 Nuclear Receptor Subfamily 6, Group A, Member 1 Overexpressed in prostate cancer cases [[90]24] ZNF395 Zinc finger protein 395 It promotes progression of cancer [[91]25] SOCS6 Suppressor of cytokine signalling 6 It is downregulated in various carcinomas [[92]26] EBF3 EBF Transcription factor 3 They regulate developmental pathways including B cell differentiation, and development of bone [[93]27] MEIS2 Meis Homeobox 2 It is involved in ontogeny and differentiation as well as organ development [[94]28] DCUN1D5 Defective In Cullin Neddylation 1 Domain Containing 5 Play a role in DNA damage response to genotoxic stress [[95]29] PLEKHH2 Pleckstrin Homology, MyTH4 and FERM Domain Containing H2 Promotes the malignancy of non-small cell lung cancer [[96]30] HIC1 HIC ZBTB Transcriptional Repressor 1 It is a tumor suppressor [[97]31] NCOA1 Nuclear Receptor Coactivator 1 Promotes metastasis of breast cancer [[98]32] CELF2 CUGBP, Elav-like family member 2 It negatively regulates breast cancer cell invasion [[99]33] NFIB Nuclear Factor I/B Promotes angiogenesis at the site of metastasis [[100]34] RAP1B Ras associated protein 1 Promotes invasiveness and tumorigenicity of breast cancer [[101]35] C17orf85 Nuclear Cap Binding protein 3 Also called NCPB3, it is an mRNA protein [[102]36] REPS2 RALBP1 associated Eps domain containing 2 Dysregulated during prostate cancer progression [[103]37] GSK3β Glycogen Synthase Kinase 3 Beta It negatively regulates breast cancer tumorigenicity [[104]37] SMAD5 SMAD family member 5 It induces chemoresistance in breast cancer [[105]38] CBX5 Chromobox 5 It functions as an oncogene [[106]39] UBN2 Ubinuclein 2 It promotes tumor progression in colorectal cancer [[107]40] KPNA3 Karyopherin subunit alpha 3 It positively regulates EMT in triple negative breast cancer [[108]41] CSNK1G1 casein kinase 1 gamma 1 It is overexpressed in triple negative breast cancer [[109]42] MBD5 methyl-CpG-binding domain (MBD) 5 It promotes tumor cell growth [[110]43] STRN Striatin It forms signalling hubs with different kinases and phosphatases and other signalling proteins called STRIPAK [[111]44] CLCN3 Chloride Channel 3 It prevents proliferation of ovarian cancer cells [[112]45] RSBN1L Round Spermatid Basic Protein 1 Like Its overexpression is involved with better prognosis of breast cancer [[113]46] PLAG1 Pleomorphic adenoma gene 1 It promotes cell proliferation in various cancers [[114]47] DTNA Dystrobrevin Alpha It is involved in cardiovascular malignancies [[115]47] ADAM19 A disintegrin and metalloproteinase 19 It is overexpressed in breast invasive carcinoma [[116]48] SHROOM2 Shroom family member 2 It plays a role in EMT and cancer cell metastasis inhibition [[117]49] LIMK2 LIM domain Kinase 2 It is overexpressed in triple negative breast cancer [[118]50] MIDN Midnolin Involved in kinase binding and plays a role in Parkinson’s Disease [[119]51] TAOK1 Thousand and one amino acid kinase 1 It promotes proliferation and invasion of non-small cell lung cancer [[120]52] PDE3A Phosphodiesterase 3 A It is upregulated in breast cancer [[121]53] WDFY2 WD repeat and FYVE domain-containing protein 2 It is a tumor suppressor gene [[122]54] NLK Nemo like kinase It is involved in cancer cell cycle progression and tumorigenesis [[123]55] IPMK Inositol Polyphosphate Multikinase It has tumor -suppressive role [[124]56] SP1 Specificity Protein 1 Overexpressed in numerous cancer [[125]57] NRP2 Neuropilin 2 Associated with poor prognosis in breast cancer [[126]58] PURA Purine Rich Element Binding Protein A Overexpressed in tumor [[127]59] BNIP3L BCL2 Interacting Protein 3 Like Upregulated expression in breast cancer [[128]60] UBE2K Ubiquitin Conjugating Enzyme E2 K It promotes tumor malignancies [[129]61] WISP1 WNT1 Inducible Signaling Pathway Protein 1 It positively regulates cancer cell progression and metastasis [[130]62] [131]Open in a new tab In a step forward to understanding the diagnostic significance of the core target genes in breast cancer patients we plotted the Kaplan Meier (KM) survival curve for GSK3 [MATH: β :MATH] , SP1, and NCOA1 respectively using KMplotter. Glycogen synthase kinase 3 beta (GSK3β) is known to play a role in invasion, migration, survival, proliferation, and chemo resistance [[132]22]. Suppression of GSK3β reduces the chemoresistance, invasive, and stemness properties of cancer cells [[133]63, [134]64]. As shown in Fig. [135]1D, the Kaplan–Meier curves depict poor relapse-free survival with high expression of GSK3β (p = 0.017). In some cases GSK3β is associated with cancer progression by reducing autophagic flux, whereas in other cases it is known to harbor an antiproliferative effect by inhibiting β-catenin and c-Myc [[136]65]. TNBC patients showed higher expression of GSK3β which is correlated with poor prognosis of TNBC patients [[137]66]. SP1 (Specificity protein 1) belongs to the SP transcription factor family which regulates the transcription of housekeeping genes by recruiting the transcription machinery in the absence of TATA box [[138]67]. The survival analysis curve for SP1 shows that low expression of SP1 is related to worse relapse-free survival of breast cancer patients (Fig. [139]1D). The dysregulation of nuclear receptor coactivator 1 (NCOA1) has also been found in various cancer types including breast cancer [[140]32, [141]68]. NCOA1 is overexpressed in breast cancer and is known to promote breast cancer metastasis [[142]32]. As shown in Fig. [143]1D, the Kaplan–Meier curves depicted high relapse-free survival with high expression of NCOA1 (p < 0.0001). All these findings suggest the importance of GSK3β, SP1 and NCOA1 in breast cancer. Differential expression and prognostic significance of the members of miR- 23a/27a/24–2 cluster in breast cancer The METABRIC database which includes a relative study of gene expressions in 1262 primary breast cancer cases was used to generate the Kaplan–Meier plot to study the correlation between miR- 23a, miR- 24 and miR- 27a expression levels and the overall survival rates. As shown in Fig. [144]2B, high expression of miR- 24 is responsible for significantly low survival probability of breast cancer patients (p value < 0.0001) with a cut-off value of 12.53 (expression range: 11–15). On the other hand, high expression levels of miR- 23a and miR- 27a indicated better overall survival of breast cancer patients with a p-value of 0.054 and 0.073 respectively (Fig. [145]2A, C). The cut-off value for miR- 23a was 13.39 (expression range: 10–16) and for miR- 27a was 13.48 (expression range: 11–15). Fig. 2. [146]Fig. 2 [147]Open in a new tab The role of miR- 23a, miR- 24–2 and miR- 27a in breast cancer patients. Kaplan–Meier curves showing survival probabilities of the two contrasting groups based on (A) miR- 23a, (B) miR- 24, and (C) miR- 27a expression. (D) The log2 Fold change expression of miR- 23a/miR- 24–2/miR- 27a of 26 paired primary tumor tissue samples quantified using qRT-PCR. The log2 fold change was measured as Mean ± SEM. p-value < 0.05 was considered to statistically significance We also collected breast tumor tissues along with their adjacent controls from 26 breast cancer patients. Table [148]3 shows the clinical and pathological characteristics of breast cancer patients. Around 46% of the women were below the age of 50 whereas the rest 54% were above the age of 50. We have classified the patients in TNBC (n = 3) and DPBC (n = 13) groups. Any patients not falling in either of the two groups were part of a separate group (n = 10). We evaluated the expression of the mature miRNAs, miR- 23a, miR- 24–2, and miR- 27a in breast cancer tissue and matched normal control tissues using qRT-PCR. Comparative analysis of the expression levels of these miRNAs between breast cancer and matched normal breast tissues demonstrated significant downregulation in the levels of miR- 23a, miR- 24–2, and miR- 27a breast cancer tissue in comparison to their adjacent control (Fig. [149]2D). The log2 fold change values of miR- 23a were (− 1.53 ± 0.69) fold, miR- 24–2 was (− 3.47 ± 0.63) while miR- 27a was (− 2.45 ± 0.90). This provides clues to the tumor-suppressive nature of miR- 23a, miR- 24–2 and miR- 27a in breast cancer. Table 3. Baseline characteristics of breast cancer subjects Characteristics Number % Age (years)  ≤ 50 12 46  > 50 14 54 Laterality  Left 13 50  Right 13 50 Tumor size  ≤ 9 cm^2 13 50  > 9 cm^2 12 46 Tumor grade  1 1 4  2 17 65  3 4 15  4 2 8 ER status  Positive 18 69  Negative 7 27 PR status  Positive 15 58  Negative 10 38 Her2 status  Positive 5 19  Negative 25 96 Lymph node status  N0 17 65  N1 9 35 Therapy  Post chemotherapy 8 31  Therapy naïve 18 69 Diagnosis  Invasive Carcinoma of Breast 20 77  Invasive Ductal Carcinoma 4 15  Invasive Lobular Carcinoma 1 4  Papillary Ductal Carcinoma In-situ 1 4 [150]Open in a new tab Overexpression of the members of miR- 23a/27a/24–2 provide clues towards GSK3β as a direct target To evaluate the impact of individual members of this cluster in breast cancer, we overexpressed and down-regulated miR- 23a, miR- 24–2, and miR- 27a using their plasmids (as described in Materials and Methods) and/antisense mimics (AMs) respectively in MCF- 7 and MDA-MB- 231 breast cancer cell lines. 24 h post transfection with miRNA plasmids led to an increase in mature microRNA levels in MCF- 7 cell lines as observed by Quantitative real-time PCR analysis (Fig. [151]3A). Specifically, miR- 27a was upregulated by 4.29-fold, while miR- 23a and miR- 24–2 were upregulated by 1.25-fold and 1.26-fold respectively in MCF- 7 cells. However, we did not find any significant change in the miR- 23 and miR- 27a levels but a slight decrease in miR- 24–2 levels were observed after treatment with respective AMs in MCF- 7 cells. Similarly, in MDA-MB- 231 cells, we observed an overexpression of miR- 23a by 1.31 fold (p-value = 0.04), miR- 24–2 by 1.4 folds and miR- 27a by 2.27 fold (p-value = 0.02) upon treatment with miRNAs (Fig. [152]3B). However, upon AM treatment there was decrease in mir- 23a levels by 0.67 folds, no change in the levels of miR- 24–2 and a slight increase in miR- 27a. Fig. 3. [153]Fig. 3 [154]Open in a new tab Members of miR- 23a/27a/24–2 cluster targets GSK3β. Estimation of miR- 23a, miR- 24–2 and miR- 27a levels upon miR and anti-miR treatments in (A) MCF- 7 and (B) MDA-MB- 231 breast cancer cells using qRT-PCR. (C) Predicted binding sites of miR- 23a/miR- 24–2/miR- 27a in the 3’UTR of GSK3β (D) Dual-luciferase assay to determine the direct binding of miR- 23a/miR- 24–2/miR- 27a in 3’ UTR sequence of GSK3β. Translational quantification of GSK3β (E) in MCF-7, and (F) MDA-MB- 231 cells and (G) β-catenin in MCF-7 cells by western blot analysis 24 h post transfection. GAPDH was used as a loading control. Data is representative of the mean ± SEM of 3 independent experiments. p-value < 0.05 was considered to indicate significance Interestingly our in silico analysis identified highly conserved binding sites of all three miRNAs in the 3′UTR of GSK3β (miR- 23a- 3p at position 1001–1008, miR- 24 - 3p at position 4111–4118 and miR- 27a- 3p at position 920–927) (Fig. [155]3C), in 3′UTR of NCOA1 (miR- 23a- 3p at position 197–204, miR- 24 - 3p at position 441–447 and miR- 27a- 3p at position 1218–1224) (Supplementary Fig. 1 A) and in 3′UTR of SP1 (miR- 23a- 3p at position 3774–3781, miR- 24 - 3p at position 542–548 and miR- 27a- 3p at position 1663–1669) genes (Supplementary Fig. 1B). Hence, we constructed two reporter plasmids containing the 3′UTR region of GSK3β mRNA for miR- 27a and miR- 23a (UTR1), and miR- 24–2 (UTR2) to perform the dual-luciferase reporter assay. As represented in Fig. [156]3D, we noted a marked reduction in luciferase activity by 0.58-fold in the case of miR- 23a (p-value = 0.0035), 0.27-fold in the case of miR- 24–2 (p value = 0.03), and significant upregulation of 1.15-fold in the case of miR- 27a (p value = 0.04) upon miRNA transfections in MCF- 7 cells. However, upon anti-miR transfection, we found a significant decrease in luciferase activity by 0.87-fold in antimiR- 24 (p-value < 0.0001) and 0.50-fold in antimiR- 27 (p-value = 0.0004). Along with GSK3β, we also performed luciferase assay for the 3’ UTR of NCOA1 but we did not observe any significant change in the luciferase activity upon miRNA and AM treatment (data not shown). GSK3β and β-catenin are known to play a critical role in the regulation of various cellular processes via the Wnt/β-catenin signaling pathway. Herein, we also evaluated the expression of GSK3β and β-catenin at the translational level. As shown in Fig. [157]3E the western blot analysis showed a significant increase in GSK3β protein levels after miR- 23a (1.19-fold, p-value = 0.002) as well as miR- 24–2(1.14-fold, p-value = 0.01) overexpression in MCF- 7 cells. However, no change in the GSK3β protein levels was observed upon miR- 27a overexpression. On anti-miR treatment, we observed no significant difference in the levels of GSK3β in the case of anti-miR- 23, anti-miR- 24, and anti-miR- 27 treatment in MCF- 7 cells. As shown in Fig. [158]3F, we also performed GSK3β western blot in MDA-MB- 231 cells and observed no significant changes in the protein levels upon transfection with all three microRNAs. Upon treatment with anti-miRs, we observed little to no changes in the protein levels. However a significant decrease by 0.74 folds was observed after treating the cells with anti-miR- 27 (p-value = 0.009). On the other hand, β-catenin levels upon overexpression of all three miRNAs were found to be slightly downregulated but were not statistically significant (Fig. [159]3G). A significant increase in the levels of β-catenin protein was observed after anti-miR- 23 and anti-miR- 24 by 1.22-fold (p-value = 0.04) and 1.39-fold (p-value = 0.01) respectively in MCF- 7 cells. (Fig. [160]3G). Together these results show that GSK3β is a direct target of miR- 23a and miR- 24–2. Cells may activate compensatory mechanisms to maintain GSK3β expression levels in response to decreased β-catenin levels. These compensatory mechanisms could involve feedback loops or alternative signaling pathways that upregulate GSK3β expression to counterbalance the effects of β-catenin downregulation [[161]69, [162]70]. Further research in this area is warranted. Individual members of miR- 23a/24–2/27a cluster downregulates SP1 in breast cancer cells SP1 is a transcription factor that controls the transcription of housekeeping genes, as well as genes involved in the progression of cell cycle and differentiation [[163]71] and is often found to be overexpressed in several cancers, including breast cancer [[164]57, [165]71, [166]72]. NCOA1 is the crucial regulator of steroid hormone receptor and is linked to the promotion of metastasis [[167]32]. Extracellular signal-related kinase (ERK) plays a major role in cell proliferation, EMT as well as cell migration, and its elevated expression is found in various cancers [[168]73]. Keeping this in mind, we next assessed the levels of SP1, NCOA1, and ERK after transfection with miRNAs and their respective anti-miRs in MCF- 7 cells. As illustrated in Fig. [169]4A, the overexpression of miR- 24–2 and miR- 27a led to a decrease in the expression of SP1 by 0.69 and 0.77 folds. However, there was a decrease in the level of protein level by 0.95 and 0.83 folds upon treatment with anti-miR- 24 (p value = 0.007) and anti-miR- 27 respectively. On the other hand there was no change in SP1 levels upon miR- 23a (0.94 folds) as well as anti-miR- 23 (0.98 folds) treatments. NCOA1 did not show any significant alterations upon treatment with both miRNAs and anti-miRNAs (Fig. [170]4B). On the other hand, overexpression of all three miRNAs increased the expression of total ERK by 1.15 folds (miR- 23a), 1.47 folds (miR- 24–2), and 1.35 folds (miR- 27a) respectively (Fig. [171]4C). Conversely, there was no change in total ERK expression upon treatment with anti-miR- 23 (1.00 fold) and anti-miR- 27 (0.96 folds). However, there was an increase in its levels by 1.23 folds upon treatment with anti-miR- 24. As shown in Fig. [172]4D, treatment with miR- 23a, miR- 24–2 and miR- 27a did not alter pERK expression levels, though miR- 23a showed a statistically significant p-value of 0.03. In contrast, inhibition of miR- 23a and miR- 24–2 led to increased pERK levels, while anti-miR- 27 treatment resulted in a decreased expression. Together, SP1, NCOA1, and ERK coordinate cellular responses, ensuring proper cell growth, differentiation, and tissue homeostasis. Fig. 4. [173]Fig. 4 [174]Open in a new tab Regulation of genes in cell proliferation via miR- 23a, miR- 24–2 and miR- 27a in MCF- 7 cells. Western blot technique was used to quantify the expression of (A) SP1, (B) NCOA1, (C) Total ERK, (D) p-ERK, at translational level in MCF- 7 cells upon miRs and anti-miRs treatment. GAPDH was used as a loading control. Data is represented as Mean ± SEM for 3 independent experiments. p-value < 0.05 was considered to indicate significance Overexpression of individual components of miR- 23a/27a/24–2 cluster supresses migration and invasive capacity of breast cancer cells Transwell invasion assay was performed to determine the invasive capacity of cells after treatment with individual components of miR- 23a/27a/24–2 cluster. As shown in Fig. [175]5A all three miRNAs miR- 23a/miR- 24–2/miR- 27a showed inhibition of the invasive capacity of the cells by 0.90, 0.80 and 0.63 folds in MCF-7 cells respectively as compared to pSil. Among them miR- 24–2 showed a statistically significant decrease in the invasive capacity. The migration of MDA-MB- 231 cells was also inhibited upon treatment of miR- 23a by 0.70 folds and miR- 24–2 by 0.90 folds whereas we did not see any changes in miR- 27a (Fig. [176]5B) as compared to the pSil control. In contrast AM treatment by anti-miR- 23/anti-miR- 24/anti-miR- 27 enhanced the invasion capacity of both MCF-7 and MDA-MB- 231 cells respectively in comparison to negative control. This evidence supports the intricate regulatory roles of individual miRNA components in controlling cell invasion and hold tremendous potential as miRNA-based therapeutics for breast cancer. Fig. 5. [177]Fig. 5 [178]Open in a new tab Regulation of Invasion and EMT pathway by miR- 23a, miR- 24–2, and miR- 27a in breast cancer cells. Cell invasion assay was used to assess the invasion capacity in (A) MCF-7, and (B) MDA-MB-231 breast cancer cells after overexpression or downregulation of individual components of miR- 23a/27a/24–2 cluster. The images were captured at 10 × magnification. (C) E- Cadherin, (D) Claudin 1, (E) Keratin 8/18, and (F) Fibronectin expressions were quantified for epithelial-mesenchymal transition using western blot. GAPDH was used as a loading control. Data is representative of the mean ± SEM for three independent experiments. p-value < 0.05 was considered to indicate significance In order to understand the molecular mechanism underlying regulated invasion of breast cancer cells, we further investigated the involvement of miR- 23a, miR- 24–2 and miR- 27a in regulating EMT (Epithelial-mesenchymal transition) markers. Our findings demonstrated that overexpression of all three microRNAs increased the levels of epithelial markers, including E-cadherin, claudin 1, and Keratin 8/18 as shown in Fig. [179]5C-E. MiR- 23a overexpression led to significant upregulation of E-cadherin by 1.18 fold and insignificant changes in E-Cadherin were observed after miR- 24–2 and miR- 27a over expression in MCF7 cells by 1.11 and 1.13 fold change respectively. Overexpression of miR- 23a, miR- 24–2 and miR- 27a led to increase in claudin 1 levels by 1.46, 1.48 and 1.33-fold respectively, whereas their anti-miRs led to a decrease in claudin 1 levels in MCF- 7 cells. In case of Keratin 8/18 significant upregulation was seen upon overexpression of miR- 27a, by 1.18-fold and it was downregulated in the case of anti-miR- 23a, 24 and 27a by 0.85, 0.85 and 0.97-folds, respectively in MCF- 7 Cells. Fibronectin, a mesenchymal marker, was also examined as shown in Fig. [180]5F. Of these, miR- 23a overexpression exhibited a significant upregulation by 1.14-fold, while miR- 24–2 and miR- 27a did not show any change. On transfection with anti-miR- 23, 24 and 27, we observed a significant upregulation in fibronectin expression by 1.24, 1.19, and 1.32-fold, respectively. These findings demonstrate that individual members of the miR- 23a/27a/24–2 cluster suppresses migration, invasion, and metastasis in breast cancer cells. Members of miR- 23a/27a/24–2 cluster modulates cell cycle in breast cancer cells Cell cycle disruption is the defining characteristic of cancer, resulting in uncontrolled cell division. Hence, it is important to investigate the impact of miR- 23a/27a/24–2 cluster on cell cycle progression. The cell cycle assay in MCF-7 cells was studied using Propidium Iodide (PI) staining. As shown in Fig. [181]6A, we observed a decreased number of cells in G1 phase as well as an increase in cell population in S phase upon treatment with miR- 23a, miR- 24–2 and miR- 27a respectively as compared to pSil. This is followed by the decrease in number of cells in G2 phase upon overexpression of individual members of cluster. In addition, as shown in Fig. [182]6B, upon overexpression of miR- 23a, 24–2, and 27a, we observed an upregulation in the expression of cyclin D1 by 1.16-folds, 1.22-folds, and 1.14-folds, respectively. Conversely, upon treatment with their respective anti-miRNAs, cyclin D1 was downregulated by 0.87, 0.77 and 0.9-folds, respectively. However, we did not observe any changes in c-Myc expression levels upon miR and anti-miR treatment (Fig. [183]6C). The above result suggests that these miRNAs promote cell cycle progression from G1 to S phase, followed by the inhibition of transition to G2 phase. Fig. 6. [184]Fig. 6 [185]Open in a new tab Modulation of cell cycle progression by the members of miR- 23a/24–2/27a cluster. Cell cycle distribution of MCF-7 cells transfected with miR- 23a, miR- 24–2, and miR- 27a (A) After PI staining, flow cytometry of MCF- 7 cells was performed to assess cell cycle progression 24 h post transfection. (B) Cyclin D1 (C) c-Myc expressions were quantified for cell proliferation using western blot. GAPDH was used as a loading control. Data is representative of the mean ± SEM. p-value < 0.05 was considered to indicate significance Discussion Breast cancer poses a notable risk to women’s health, primarily due to its high mortality rate, often stemming from therapy failures [[186]1]. Resistance to chemotherapy and the risk of relapse present major challenges in breast cancer management [[187]74]. Whilst miRNAs are involved in the control of apoptosis and the prevention of cancer, changes in their expression may also be linked to the hallmarks of carcinogenesis, such as activation, invasion, and metastasis, as well as the encouragement of tumor growth and proliferation and the suppression of apoptosis [[188]75]. Herein, we discuss the association between three mature microRNAs in the miR- 23a/27a/24–2 cluster and breast cancer. In brief, miR- 23a/27a/24 − 2 cluster is a polycistronic miRNA, encoding miR- 23a, miR- 27a and miR- 24–2 but their biological functions are distinct due to variations in expression regulation. The expression of miRNAs is regulated by several factors (transcriptional and post-transcriptional) and molecular mechanisms causing oncogenic or tumour suppressive functions. Some studies have shown that upregulation of miR- 23a cluster enables it to serve as potential biomarker for diagnostic and progression of disease/metastasis for lung cancer, colorectal cancer, gastric cancer and for other metabolic diseases [[189]17, [190]76, [191]77]. In breast cancer research, findings have been inconsistent. A study by Li et al. (2023), reported that the expression levels of miR- 23a, miR- 24–2 and miR- 27a were significantly elevated in breast cancer with lymph node metastasis, compared to those without metastasis or normal and played an oncogenic role [[192]78]. However, in the current study our findings from both patient samples and in vitro models suggest that the miR- 23a/27a/24–2 cluster may have a tumor-suppressive role in breast cancer. In silico analysis of the overlapping predicted target genes of the cluster revealed NCOA1, GSK3β, and SP1 to be interacting with each other in a hub. While direct interactions between all three genes might not be extensively documented, their individual roles suggest potential interconnectedness. GSK3β dysregulation has been linked to several malignancies, including breast cancer, where it had an impact on the survival and proliferation of cells [[193]79, [194]80]. In cancer, SP1 is frequently overexpressed and has a role in tumor development, angiogenesis, and metastasis [[195]57]. NCOA1 by acting as a transcriptional coactivator frequently plays a role in progression of various cancers [[196]81]. The development and spread of tumors may be influenced by the interactions between these variables. Herein, Kaplan–Meier plots were used to evaluate the correlation between the expression of these genes as well as miRNAs and the overall survival rates of patients with breast cancer. GSK3β and NCOA1 were among the significant genes. Compared with healthy mammary cells, breast tumors having increased expression of GSK3β are correlated with a reduced life expectancy. On the other hand, higher expression of NCOA1 and SP1 in breast tumors was associated with better relapse-free survival rate. According to the log-rank test while plotting the Kaplan–Meier graph, miR- 24–2 was found to be particularly significant. Hsa-miR- 24–2 could be a novel biomarker for breast cancer patients as it is found highly expressed in FFPE tissues of breast cancer patients. It's interesting to note that a higher survival rate for breast cancer patients is linked to increased expression of miR- 23a and miR- 27a. We have further shown that GSK3β is a direct target of miR- 23a and miR- 24–2 in breast cancer cells. The Wnt/β-catenin pathway is one of those pathways that are particularly important for the development of breast cancer [[197]22]. Overexpression of individual miRNAs in the miR- 23a/27a/24–2 cluster stimulated Wnt/β-catenin signaling in breast cancer. Simultaneously, SP1 is frequently overexpressed in breast cancer and encourages tumor development and metastasis. While it is frequently overexpressed in multiple cancers and is associated with poor clinical outcomes, its function remains complex, as it can both promote and suppress the expression of oncogenes and tumor suppressors. Due to its involvement in critical cancer pathways, SP1 has been considered a potential therapeutic target; however, its diverse regulatory roles pose significant challenges for targeted treatment strategies [[198]57, [199]71]. In our investigation, we observed SP1 downregulation upon overexpression of the three members of the cluster. Further research is required to confirm SP1 as the target of the cluster. Overexpression/inhibition of individual members in the miR- 23a/27a/24–2 cluster, modulated the expression of proteins involved in the ERK pathway, resulting in the upregulation of total ERK levels. Research has shown that the upregulated ERK signaling can promote EMT in breast cancer cells by activating key transcription factors [[200]82]. The potential of GSK3β to impede the development of EMT is demonstrated by the reduction in the expression of mesenchymal phenotypic markers upon inhibition. Functional assays, such as the invasion assay, have frequently been used to demonstrate the enhancement of the migratory potential of the cells, which contributes to the highly aggressive and metastatic nature of the cells that have undergone EMT [[201]83]. Various studies have highlighted the significance of miR- 23a in regulating migration and invasive capacity of breast cancer by targeting CDH1 gene which plays a pivotal role in EMT [[202]84]. On the other hand, miR- 27a targets FBXW7 (F-box and WD repeat domain containing 7) gene, which positively impacts the breast cancer survival rate by inhibiting EMT thus influencing the migration of cancer cells [[203]85]. miR- 24–2 overexpression is also known to inhibit ING5 (member of Inhibitor of Growth family) resulting in increased proliferation and invasion of breast cancer cells [[204]86]. Although the role of these miRNAs individually has been studied in the context of breast cancer, their collective impact remains unexplored. As represented in Fig. [205]7, the overexpression of miR- 23a/24–2/27a cluster in breast cancer cell line induced several molecular and phenotypic changes. We observed an increase in GSK3β expression and an associated decrease in β-catenin as well as SP1 expression. According to literature, elevated GSK3β levels promotes the phosphorylation and subsequent proteasomal destruction of β-catenin and SP1. Reduced SP1 and increased GSK3β expression suggests a decrease in the transcriptional activation of genes involved in cell growth and proliferation. These observations were accompanied by a decrease in epithelial-to-mesenchymal transition (EMT), evidenced by the increased expression of epithelial markers such as E-cadherin, Claudin 1 and Keratin 8/18 followed by a decrease in mesenchymal marker, Fibronectin. The reduced EMT is associated with a decrease in migration and invasive capacities of cancer cells. Also, the over-expression of the cluster led to an S phase arrest in MCF-7 cells along with an increase in cell cycle marker, Cyclin D1. Fig. 7. [206]Fig. 7 [207]Open in a new tab Proposed mechanism of individual components of miR-23a/27a/24–2 cluster in various pathways in breast cancer Several studies including our current study suggested that expression of all three miRNA in miR- 23a cluster, both individually and as part of the cluster are implicated in breast cancer via association with metastasis, migration, and invasion [[208]11, [209]16, [210]84, [211]86]. Conclusion Our study clarifies the important role of the individual members of the miR- 23a/27a/24–2 cluster in breast cancer providing new therapeutic approach. In our study we observed that the breast cancer tissues have decreased expression of individual members of miR- 23a/27a/24–2 compared to control. However, breast cancer being heterogeneous in nature and having different subtypes, would require validation of the same in large number of breast cancer samples. The study also highlights the important role of miR- 23a/27a/24–2 cluster in regulating key genes linked to breast cancer progression, namely, GSK3β, NCOA1 and SP1. Our observations regarding the dysregulation of miRNA expression in breast cancer tissues underscore the importance of further research to validate these findings across diverse patient populations and tumor subtypes. This work opens avenues for future investigations into personalized treatment strategies and the clinical utility of targeting the miR- 23a/27a/24–2 cluster in breast cancer therapy. Supplementary Information [212]Supplementary Material 1.^ (47.6KB, xlsx) [213]Supplementary Material 2^ (92.5MB, pptx) [214]Supplementary Material 3^ (264KB, pptx) [215]Supplementary Material 4^ (10.4MB, pptx) Acknowledgements