Breast cancer-associated fibroblast-derived exosomal miR-1290 promotes triple-negative breast cancer cell proliferation

Abstract

Breast cancer is a major cause of cancer-associated mortality in females. Although therapeutic strategies targeting subtype-specific receptors have been developed for various breast cancer subtypes, effective targeted therapies for triple-negative breast cancer (TNBC) are lacking. Recently, the role of the tumor microenvironment (TME), particularly cancer-associated fibroblasts (CAFs), in the progression and metastasis of breast cancer, especially TNBC, has attracted increasing attention. However, the role of exosomes secreted by CAFs in TNBC cell proliferation is poorly understood. Thus, in this study, we analyzed exosomes derived from both CAFs and MDA-MB-231 cells and identified exosomal miR-1290 as a key component. When delivered to TNBC cells (BT-549 and Hs578T), exosomal miR-1290 directly suppressed the expression of the putative tumor suppressor TBC1D4, which was accompanied by reduced TSC1 expression and activation of the mTOR pathway. The promotion of TNBC cell growth by miR-1290 through direct targeting of TBC1D4 was validated using 2D culture models, 3D spheroid models, and in vivo xenograft models. Furthermore, we demonstrated that treatment with the mTOR inhibitor rapamycin and the FDA-approved rapamycin analog everolimus effectively suppressed miR-1290-induced TNBC cell proliferation, suggesting a potential therapeutic strategy for TNBC. Moreover, in patient cohorts, elevated levels of circulating miR-1290 significantly distinguished patients with breast cancer from healthy individuals and correlated with poor survival outcomes. Therefore, miR-1290 and its direct target TBC1D4 have potential value in both targeted treatment and diagnosis of TNBC.

Introduction

Despite significant advancements in target identification and drug development, breast cancer remains a leading cause of cancer-related death among women worldwide.^1–3 Breast cancer is classified into several molecular subtypes on the basis of the expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), and subtype is a key factor in prognosis evaluation and treatment selection.^1,2 Among these subtypes, triple-negative breast cancer (TNBC), characterized by the absence of ER, PR, and HER2 expression, accounts for 10–15% of all breast cancers.^1,2 TNBC is characterized by aggressive behavior with an earlier age of onset, higher grade tumors, an increased likelihood of distant recurrence within 5 years of diagnosis, and poorer overall survival than other breast cancer subtypes.^1,2 Owing to the lack of targetable receptors, TNBC patients cannot benefit from conventional endocrine or HER2-targeted treatments; thus, chemotherapy remains the primary treatment option.^1,2 Despite initial responsiveness, TNBC patients frequently develop resistance, and patients without a pathologic complete response have significantly worse outcomes.^1,2 In recent years, several new treatment modalities have been introduced for TNBC, including immune checkpoint inhibitors for programmed death-ligand 1 (PD-L1)-positive tumors and poly(ADP-ribose) polymerase (PARP) inhibitors for patients harboring BRCA1/2 mutations.^1,2 Although these approaches have expanded the therapeutic landscape and improved outcomes for selected patients, their clinical benefits are largely restricted to specific molecular or biomarker-defined subgroups.^1,2 Thus, a substantial proportion of patients with TNBC still lack broadly applicable targeted treatment options, and the prognosis of TNBC remains poor, with a high risk of early distant relapse.^1,2 Collectively, these limitations underscore the urgent need to identify novel therapeutic targets and diagnostic biomarkers for this aggressive subtype. However, such advances require a deeper understanding of the cellular and molecular mechanisms that drive TNBC progression and contribute to its poor treatment response, particularly within the tumor microenvironment.

The high heterogeneity of characteristics among breast cancers, which originates from diverse cell types within the tumor microenvironment (TME), presents major challenges in the treatment of this disease.^4–6 The TME comprises various cell types, including cancer cells, fibroblasts, immune cells, and endothelial cells, as well as extracellular matrix components.^7,8 Cancer-associated fibroblasts (CAFs) play crucial roles in promoting tumor progression, metastasis, and therapeutic resistance through bidirectional communication with cancer cells,^9–11 and breast cancer-associated fibroblasts (BCAFs) significantly contribute to tumor growth, invasion, and metastasis.^12–14 Intercellular communication within the TME occurs through a range of mechanisms, including direct cell–cell contact, soluble factors, and extracellular vesicles.^15,16 Exosomes, small membrane-bound vesicles ( ~ 50–150 nm in diameter), are key mediators of this communication and carry diverse cargos, including proteins, lipids, and microRNAs (miRNAs).¹⁷ Although research on exosomes has been conducted in accordance with the increasing importance of CAFs in breast cancer, especially TNBC, studies on which factors within exosomes influence the promotion of TNBC growth are lacking.

miRNAs are short noncoding RNAs, typically ~20–22 nucleotides in length, that act at the posttranscriptional level by pairing with the 3’ untranslated region (UTR) of their target mRNAs, resulting in reduced protein production or decreased stability of the target transcripts.^18,19 Studies increasingly indicate that miRNAs regulate diverse cellular processes, including cell proliferation, metabolic regulation, differentiation, and apoptosis, whereas altered miRNA expression has been linked to a broad spectrum of diseases, including cancer.^18,19 However, research on exosomal miRNAs involved in CAF-induced TNBC growth remains insufficient. Among the various cancer-associated miRNAs, miR-1290 has emerged as a key player in tumor progression and metastasis. Recent studies have demonstrated that exosomal miR-1290 mediates critical intercellular communication within the TME across multiple cancer types. In lung adenocarcinoma, COX-2 overexpression induces cancer cell-derived exosomal miR-1290, which promotes CAF activation through the CUL3-Nrf2 pathway, creating a tumor-supportive microenvironment.²⁰ In the context of breast cancer brain metastasis, extracellular vesicles carrying miR-1290 promote astrocyte activation through the FOXA2-CNTF axis, thereby contributing to metastatic progression in the brain.²¹ Similarly, in ovarian cancer, CAF-derived exosomal miR-1290 targets OGN to increase cancer cell proliferation and invasion.²² These findings highlight the crucial role of exosomal miR-1290 in tumor–stroma interactions. However, despite these advances, the specific role of BCAF-derived exosomal miR-1290 in TNBC progression and its downstream molecular mechanisms remain poorly understood.

Thus, in this study, we identified miR-1290 in exosomes from BCAFs and MDA-MB-231 (MB-231) cells via exosomal RNA-seq. Exosomal miR-1290 derived from BCAFs directly downregulated the expression of its target, TBC1 domain family member 4 (TBC1D4), which was associated with decreased TSC Complex Subunit 1 (TSC1) expression and increased mTOR activity, thereby promoting the growth of TNBC cell lines. We propose that targeting the miR-1290-mediated regulation of TBC1D4, which affects mTOR activation, is a highly promising therapeutic strategy for TNBC. Furthermore, treatment with the mTOR inhibitor rapamycin and the FDA-approved everolimus effectively suppressed miR-1290-induced TNBC cell proliferation both in vitro and in vivo. Moreover, large-scale data analysis revealed elevated miR-1290 levels in the blood of breast cancer patients (TNBC cohorts) compared with healthy individuals, suggesting its potential as a diagnostic biomarker. These findings highlight the importance of BCAFs in TNBC progression and reveal novel therapeutic targets, providing new insights into the complex intercellular communication between breast cancer and the TME.

Results

Identification of exosomal miR-1290 in BCAFs and TNBC cells and its clinical relevance

To identify TNBC growth-promoting factors derived from BCAFs, we hypothesized that exosomal transfer of miRNAs may promote oncogenic phenotypes in recipient breast cancer cells from BCAF (a key stromal component of the TME) and aggressive TNBC cells (MB-231 cells; a highly aggressive TNBC cell line). Exosomes isolated from BCAFs and MB-231 cells were characterized by nanoparticle tracking analysis (NTA); the measured mean diameters were 130.1 ± 73.8 nm and 154.6 ± 71.1 nm, respectively, which were consistent with the expected size range of exosomes ( ~ 50–150 nm) (Fig. 1a). Transmission electron microscopy (TEM) further confirmed the typical cup-shaped morphology of the isolated exosomes (Fig. 1b). Western blot analysis confirmed the enrichment of established positive exosomal markers (CD63, CD81, and flotillin-1) in the exosome fractions, whereas negative cellular markers (GM130 and H3) were detected only in whole-cell lysates, confirming the purity of the isolated exosomes (Fig. 1c). Furthermore, we confirmed the robust expression of CAF markers (α-SMA and FAP) in BCAFs, confirming the identity and characteristics of these fibroblasts (Fig. 1c). To confirm exosome uptake, BT-549 and Hs578T cells were incubated with BCAF-derived exosomes labeled with PKH67 (green fluorescence). After 1 h, efficient uptake was confirmed by the generation of green fluorescent signals within the recipient cells (Fig. 1d), demonstrating the functional capacity of the BCAF-derived exosomes to deliver their cargo to TNBC cells.

Fig. 1.

miR-1290 is upregulated in breast cancer and BCAF/MB-231-derived exosomes. a Nanoparticle tracking analysis of exosomes isolated from BCAFs and MB-231 cells showing their size distribution. The mean size and standard deviation (SD) are indicated for each sample. b TEM images of exosomes isolated from BCAFs (upper) and MB-231 cells (lower) showing the characteristic morphology of the exosomes. Scale bar, 100 nm. c Western blot analysis of exosome markers (CD63, CD81, and flotillin-1), CAF markers (α-SMA and FAP) and cellular markers (GM130 and H3) in BCAF cells, MB-231 cells and their derived exosomes. d BCAF-derived exosome uptake by TNBC cells. PKH67-labeled BCAF-derived exosomes (green) were incubated with BT-549 and Hs578T cells for 1 hour. The cell nuclei were stained with Hoechst 33342 (blue). Scale bar, 20 μm. e Heatmap showing miRNA expression profiles in exosomes derived from BCAFs and MB-231 cells. Red indicates high expression, and blue indicates low expression. miR-1290 (highlighted in red) is highly expressed in both exosome types. f Volcano plots showing differentially expressed miRNAs between breast cancer or TNBC samples and normal controls in five independent GEO datasets (GSE37963, GSE45666, GSE45498, GSE38167, and GSE61438). Red and blue dots represent upregulated and downregulated miRNAs, respectively, selected using fold-change and nominal p value cutoffs (|log2(fold change)| > 0.58 and nominal p < 0.05, two-tailed Welch’s t test). g Venn diagram analysis of upregulated miRNAs across the five GEO datasets revealed four commonly upregulated miRNAs. h Violin plot showing miR-1290 expression levels in serum samples from normal individuals (N, n = 2686) and breast cancer patients (P, n = 1280) from the GSE73002 dataset. The expression of miR-1290 was significantly upregulated in the breast cancer patient samples (***p < 0.001, Mann‒Whitney U test). i Analysis of serum exosomal miR-1290 levels in normal (N, n = 20) and TNBC patient (P, n = 39) samples collected at Keimyung University Dongsan Hospital. The results support the potential of miR-1290 as a candidate diagnostic marker in an independent cohort (***p < 0.001, Mann‒Whitney U test). j Kaplan‒Meier plot of overall survival for TNBC patients based on serum exosomal miR-1290 expression levels from Keimyung University Dongsan Hospital (p = 0.006, log-rank test)

Next, to identify growth-promoting exosomal miRNAs from BCAFs and MB-231 cells, we performed exosomal RNA-seq analysis and ultimately selected miR-1290 for further study (Fig. 1e). In a previous study, we suggested that, in hepatocellular carcinoma (HCC), miR-1290 acts as an oncogenic miRNA, and that exosomal miR-1290 expression is increased in HCC samples compared with normal samples.²³ To validate the overexpression of miR-1290 in breast cancer, we analyzed five independent GEO datasets representing diverse breast cancer pathophysiologies: GSE37963 (42 breast cancer tissues and 5 matched normal tissues), GSE45666 (101 breast tumor tissues and 15 adjacent normal tissues), GSE45498 (169 primary TNBC tissues and 59 adjacent normal tissues), GSE38167 (31 primary TNBC tissues and 23 matched normal tissues), and GSE61438 (44 grade III breast tumor tissues and 13 normal tissues). Volcano plot and Venn diagram analyses revealed miR-1290 as a commonly upregulated candidate miRNA in breast cancer/TNBC tissues compared with normal tissues across five independent GEO datasets representing different breast cancer contexts, including TNBC and grade III disease (Fig. 1f and g). In addition, in the GSE73002 dataset (serum samples from 2,686 normal individuals and 1,280 breast cancer patients), we found that miR-1290 expression was significantly elevated in breast cancer patients (p < 0.001, Mann‒Whitney U test) (Fig. 1h). To validate these findings in an independent cohort, we analyzed the serum levels of exosomal miR-1290 in samples collected at Keimyung University Dongsan Hospital. We observed significantly higher miR-1290 levels in breast cancer patient samples (TNBC samples) than in normal control samples (p < 0.001, Mann–Whitney U test) (Fig. 1i). Additionally, TNBC patients with high serum miR-1290 levels exhibited a poorer prognosis (p = 0.006, log-rank test) (Fig. 1j). To determine whether the other three commonly upregulated miRNAs (miR-130b, miR-18a, and miR-425) identified across the five GEO datasets are also clinically relevant, we analyzed their expression levels in both the GSE73002 dataset and our independent TNBC cohort. While the expression of all four miRNAs was significantly greater in breast cancer patients than in healthy controls in the GSE73002 dataset (Fig. 1h and Supplementary Fig. 1a), the results in our TNBC patient cohort were inconsistent across the four miRNAs. The expression of miR-1290 and miR-18a significantly increased, whereas the expression of miR-130b and miR-425 did not significantly differ between TNBC patients and normal controls (p = 0.129 and p = 0.819, respectively; Fig. 1i and Supplementary Fig. 1b). However, survival analysis revealed that among these four miRNAs, only elevated miR-1290 expression was significantly associated with poor prognosis in TNBC patients (p = 0.006, log-rank test), whereas the expression of miR-130b (p = 0.545, log-rank test), miR-18a (p = 0.138, log-rank test), and miR-425 (p = 0.738, log-rank test) was not significantly associated with patient survival (Supplementary Fig. 1c). These findings demonstrate that while miR-18a is also upregulated in TNBC patients, it lacks prognostic value, highlighting the unique clinical relevance of miR-1290 as both a diagnostic biomarker and a prognostic biomarker. This dual significance supports our focus on miR-1290 for further mechanistic studies.

BCAF- and MB-231-derived exosomes and miR-1290 promote the proliferation of recipient TNBC cell lines

To assess whether BCAF- and MB-231-derived exosomes affect TNBC growth, we first isolated exosomes from BCAFs and MB-231 cells and coincubated them with BT-549 and Hs578T cells, two lines that express miR-1290 at lower levels than MB-231 cells (Supplementary Fig. 2). As shown by crystal violet (CV) staining and cell counting kit-8 (CCK-8) assays, treatment with the BCAF-derived exosomes significantly increased the proliferation of both TNBC lines (BT-549 and Hs578T). Interestingly, the recipient TNBC cell lines presented increased miR-1290 levels in response to treatment with BCAF-derived exosomes (Fig. 2a). Additionally, we found that treatment with MB-231 exosomes enhanced cell proliferation and increased miR-1290 expression in the recipient TNBC cell lines (Fig. 2b).

Fig. 2.

miR-1290 promotes breast cancer cell proliferation. a Crystal violet staining of BT-549 and Hs578T cells treated with DPBS or BCAF-derived exosomes (left). Quantification of cell numbers (middle). qRT‒PCR analysis of miR-1290 expression in recipient cells (right). b Crystal violet staining of BT-549 and Hs578T cells treated with DPBS- or MB-231-derived exosomes (left). Quantification of cell numbers (middle). qRT‒PCR analysis of miR-1290 expression in recipient cells (right). Scale bar, 500 μm. c Crystal violet staining of BT-549 and Hs578T cells transfected with the miR-1290 mimic or mimic-NC (left). Quantification of cell numbers (middle). qRT‒PCR analysis of miR-1290 expression (right). Scale bar, 500 μm. d Representative images of xenograft tumors derived from BT-549 cells transfected with the miR-1290 mimic or mimic-NC. e Growth curves of xenograft tumors in NOD/Shi-scid/IL-2Rγ^null mice inoculated with BT-549 cells transfected with the miR-1290 mimic or mimic-NC (n = 6 per group). The tumor volume was compared between the two groups at the experimental endpoint. f Tumor weight measurements at the endpoint of the xenograft experiment shown in (e) (n = 6 per group). g Representative immunocytochemistry images showing the localization and expression of Ki67 in BT-549 and Hs578T cells treated with BCAF exosomes or MB-231 exosomes and transfected with the miR-1290 mimic or mimic-NC. The nuclei were counterstained with DAPI. Scale bar, 150 μm. The data are shown as the means ± SDs of three independent experiments unless otherwise indicated, and statistical significance was determined using Student’s t tests (*p < 0.05, **p < 0.01, ***p < 0.001)

To further validate the functional role of BCAFs in promoting TNBC cell proliferation through miR-1290 transfer, we established a Transwell coculture system to enable paracrine communication between BCAFs and BT-549 cells without direct cell–cell contact (Supplementary Fig. 3a). After 48 hours of coculture, both crystal violet staining and a CCK-8 assay confirmed the significantly enhanced proliferation of BT-549 cells in the presence of BCAFs (Supplementary Fig. 3b and c). Importantly, qRT‒PCR analysis revealed that miR-1290 expression was significantly elevated in BT-549 cells following coculture with BCAFs (Supplementary Fig. 3d), which is consistent with our findings that BCAF-derived exosomes containing miR-1290 are efficiently internalized by recipient TNBC cells (Fig. 1d). These results demonstrate that in the Transwell coculture system, BCAF-derived exosomal miR-1290 is transferred to BT-549 cells through paracrine communication, thereby promoting cell proliferation.

To investigate whether BCAFs can promote tumor growth in vivo and to directly address the contribution of BCAF-derived miR-1290, we performed xenograft experiments in which BT-549 cells were coinjected with BCAFs. NOD/Shi-scid/IL-2Rγ^null mice were subcutaneously injected with either BT-549 cells alone or BT-549 cells mixed with BCAFs. Body weight monitored throughout the experiment did not differ significantly among the groups, indicating that coinjection caused no systemic toxicity (Supplementary Fig. 3e). Notably, compared with those in the BT-549-alone group, the tumor volume and weight in the BT-549 + BCAF coinjection group significantly increased (p < 0.001; Student’s t test; n = 5 per group) (Supplementary Fig. 3f–h). Furthermore, qRT‒PCR analysis of tumor tissues revealed that compared with those in the BT-549-alone group, miR-1290 expression in the BT-549 + BCAF group was significantly elevated (p < 0.01, Student’s t test; n = 5 per group) (Supplementary Fig. 3i), providing direct in vivo evidence that BCAFs contribute to increased miR-1290 levels in the tumor microenvironment.

Immunohistochemistry and qRT‒PCR analyses confirmed significantly stronger α-SMA and FAP expression in tumors from the BT-549 + BCAF coinjection group than in those from the BT-549 alone group (p < 0.001 for all comparisons except for the FAP qRT‒PCR result, p = 0.025; Student’s t test, n = 5 per group) (Supplementary Fig. 3j–l). Together, these results demonstrate that BCAFs are successfully incorporated into the tumor microenvironment, where they promote tumor growth and increase miR-1290 levels in vivo.

Next, to determine whether miR-1290 drives the promotion of cell proliferation, we transfected BT-549 and Hs578T cells with a miR-1290 mimic or mimic-NC as a negative control. The upregulation of miR-1290 expression by the miR-1290 mimic significantly increased cell proliferation (p < 0.001, Student’s t test), which is consistent with the results of the BCAF and MB-231 exosome treatments (Fig. 2c). In addition, to confirm the relationship between miR-1290 expression and cell growth in vivo, we used a xenograft model to evaluate the effect of miR-1290 expression on cell proliferation. As shown in Fig. 2d, compared with control cells, BT-549 cells transfected with the miR-1290 mimic formed significantly larger tumors, with approximately 2-fold increases in tumor volume and weight, confirming that miR-1290 promotes tumor growth in vivo (p < 0.001 and p = 0.022, respectively; Student’s t test, n = 6 per group) (Fig. 2e and f, and Supplementary Fig. 4). Moreover, we clearly observed the induction of the expression of Ki67, an indicator of proliferation, after treatment with either exosomes derived from BCAFs and MB-231 cells or the miR-1290 mimic via immunocytochemistry (Fig. 2g, quantification shown in Supplementary Fig. 5). Thus, these results suggest that exosomes containing miR-1290 derived from BCAFs and MB-231 cells may be delivered to surrounding cancer cells and that the transferred miR-1290 likely promotes their growth.

miR-1290 promotes TNBC cell proliferation by directly targeting TBC1D4

To identify potential targets of miR-1290, we performed RNA-seq analysis with miR-1290-overexpressing BT-549 cells (both exosome-treated and miR-1290 mimic-transfected) and intersected the downregulated genes with putative miR-1290 targets from TargetScan, revealing 12 candidate genes: KAZALD1, MARVELD2, ADAMTS14, DSP, SLFN13, EFCAB13, STON2, GRIN2A, DSC2, MATN2, C1orf115, and TBC1D4 (Fig. 3a). We systematically analyzed all the candidates for their clinical relevance and functional impact. In the Cancer Genome Atlas (TCGA) data portal, we observed that TBC1D4 expression is significantly downregulated in breast cancer patients compared with normal tissues (p < 0.001, Mann‒Whitney U test; Fig. 3b), while the expression of several genes is altered in breast cancer. However, only TBC1D4 was significantly downregulated and associated with poor prognosis (Fig. 3c and Supplementary Fig. 6). To further validate the specificity of TBC1D4 as a key mediator of miR-1290-induced proliferation, we also performed siRNA knockdown experiments on other candidate genes whose expression was downregulated in breast cancer tissues according to TCGA analysis (Supplementary Fig. 6). Strikingly, knockdown of these other candidate genes did not result in growth-promoting effects; specifically, knockdown of SLFN13, STON2, GRIN2A, DSC2, and C1orf115 did not significantly affect cell proliferation, whereas knockdown of EFCAB13 and MATN2 resulted in growth inhibition, in contrast to that of TBC1D4 (Supplementary Fig. 7). These results confirm that TBC1D4 is the only downregulated miR-1290 target candidate that negatively regulates TNBC cell proliferation.

Fig. 3.

miR-1290 promotes breast cancer cell proliferation by directly targeting TBC1D4. a Venn diagram analysis of genes downregulated according to the RNA-seq data of MB-231 exosome-treated (FPKM ratio < 0.7 versus DPBS; n = 981) and miR-1290 mimic-transfected BT-549 cells (FPKM ratio < 0.5 versus mimic-NC; n = 676) with predicted miR-1290 targets from TargetScan (n = 4601), identifying 12 common candidate genes (left). Expression of TBC1D4 under the indicated conditions (right). b Expression levels of TBC1D4 in normal (n = 114) and breast cancer (n = 1097) tissues from the TCGA database (***p < 0.001, Mann‒Whitney U test). c Kaplan‒Meier plot of the overall survival of breast cancer patients from the TCGA database. The survival rate of the high-TBC1D4 subgroup was significantly greater than that of the low-TBC1D4 subgroup (p = 0.003, log-rank test). d qRT‒PCR analysis of TBC1D4 expression in BT-549 and Hs578T cells treated with the indicated conditions. e Western blot analysis of TBC1D4 expression in BT-549 and Hs578T cells under the indicated conditions. ACTB was used as a loading control. f Representative immunocytochemistry images showing the localization and expression of TBC1D4 in BT-549 and Hs578T cells treated with BCAF exosomes and transfected with the miR-1290 mimic or mimic-NC. The nuclei were counterstained with DAPI. Scale bar, 150 μm. g Dual-luciferase reporter assay in HEK293 cells cotransfected with wild-type or mutant TBC1D4 3’-UTR luciferase reporter vectors and miR-1290 mimic or mimic-NC (six technical replicates) (n.s. not significant). h qRT‒PCR analysis of TBC1D4 expression in BT-549 and Hs578T cells transfected with siTBC1D4 or siCont. i Western blot analysis of TBC1D4 expression in BT-549 and Hs578T cells transfected with siTBC1D4 or siCont. ACTB was used as a loading control. j Crystal violet staining of BT-549 and Hs578T cells transfected with siTBC1D4 or siCont and quantification of cell numbers. Scale bar, 500 μm. k Representative immunocytochemistry images showing the localization and expression of Ki67 in BT-549 and Hs578T cells transfected with siTBC1D4 or siCont. The nuclei were counterstained with DAPI. Scale bar, 150 μm. The data are shown as the means ± SDs of three independent experiments, and statistical significance was determined using Student’s t tests (***p < 0.001)

Thus, we selected TBC1D4 as a direct target of miR-1290, through which it affects TNBC cell proliferation. TBC1D4, a Rab GTPase-activating protein, is reported to be hyperphosphorylated in breast cancer,²⁴ resulting in the inactivation of its Rab GAP function,^25,26 and its hyperphosphorylation is correlated with increased tumor size and increased expression of the cell proliferation marker Ki67 in breast cancer.²⁴

To confirm that the TBC1D4 expression level was reduced by miR-1290, we treated TNBC cell lines with BCAF-/MB-231-derived exosomes and a miR-1290 mimic. As shown in Fig. 3d and e, the expression levels of TBC1D4 significantly decreased at both the RNA and protein levels following treatment with the BCAF-/MB-231-derived exosomes and the miR-1290 mimic. Consistent with these findings, immunocytochemistry revealed reduced TBC1D4 staining intensity under the same conditions in TNBC cell lines (Fig. 3f, quantification shown in Supplementary Fig. 8). Next, to confirm whether TBC1D4 is a direct target of miR-1290, a luciferase assay was performed using constructs containing either the wild-type or the mutant type of the miR-1290 binding site in the 3’-UTR of TBC1D4. As a result, luciferase activity was reduced by the miR-1290 mimic in the wild-type construct but not in the mutant construct (Fig. 3g). These results indicate that TBC1D4 is a direct target of miR-1290 and that the miR-1290-mediated downregulation of TBC1D4 expression may reduce its ability to suppress cell proliferation in TNBC. To assess cell proliferation by reducing TBC1D4 expression in TNBC cell lines, we transfected TNBC cell lines with a TBC1D4-specific siRNA. qRT‒PCR and Western blot analysis revealed that TBC1D4 expression was decreased by siTBC1D4 (Fig. 3h and i), and downregulation of TBC1D4 expression induced the proliferation of TNBC cell lines, as shown by CCK-8 assays and CV staining (Fig. 3j).

Moreover, a reduction in TBC1D4 expression resulted in the upregulation of Ki67 expression in TNBC cell lines (Fig. 3k; quantification is shown in Supplementary Fig. 8). Together, the increased level of miR-1290, which is derived from either BCAFs or neighboring TNBC cells, led to a decrease in the expression of TBC1D4, a direct target of miR-1290, thereby promoting the growth of TNBC cells. These findings suggest that miR-1290 functions as an oncogenic miRNA that contributes to TNBC progression.

miR-1290 activates mTORC1 signaling via TBC1D4 downregulation

TBC1D4 is a Rab GTPase-activating protein known to regulate GLUT4 or GLUT1 translocation in response to insulin signaling.^25,26 First, we performed immunocytochemistry after TBC1D4 knockdown in TNBC cell lines. However, TBC1D4 downregulation did not affect either GLUT4 or GLUT1 localization (Supplementary Fig. 9). Thus, we hypothesized that the promotion of cell growth induced by the miR-1290-mediated downregulation of TBC1D4 is likely linked to additional downstream signaling pathways.

To identify the novel function of TBC1D4 in TNBC cell growth, we performed RNA-seq analysis after TBC1D4 knockdown and miR-1290 mimic treatment. Using 629 genes downregulated by both the miR-1290 mimic and siTBC1D4, we performed gene ontology (GO) analysis. Biological process terms such as “transforming growth factor beta receptor signaling pathway”, “SMAD protein signal transduction”, “activation of GTPase activity” and “cell proliferation” were enriched in both the miR-1290 mimic and siTBC1D4 groups compared with the control groups (Fig. 4a). In addition, we performed a phospho-array with the lysates of cells transfected with the miR-1290 mimic and siTBC1D4 and detected increases in the expression of phospho-P70S6 kinase 1 (P70S6K), a direct downstream target of mTORC1, in both the miR-1290 mimic and siTBC1D4 groups compared with each control group (Fig. 4b). Previous studies have established complex regulatory relationships between the TGF-β and mTOR signaling pathways. Gabriel et al. reported that TGF-β signaling negatively regulates mTOR activity.²⁷ Thus, we hypothesized that TBC1D4 downregulation by miR-1290 may be associated with the activation of the mTOR signaling pathway for TNBC cell proliferation. Next, to identify the TBC1D4 target for the regulation of mTOR activity, we used a candidate approach with GO term results. Finally, we selected TSC1, listed in the “activation of GTPase activity” term. The TSC1/TSC2 complex is a negative regulator that inhibits mTORC1.^28–30 Thus, on the basis of these findings, we propose that upregulation of miR-1290 and downregulation of TBC1D4 may increase TNBC growth in association with reduced TSC1 expression and increased mTOR signaling activity, including high phosphorylation of P70S6K.

Fig. 4.

miR-1290 activates mTORC1 signaling through the downregulation of TBC1D4. a Gene Ontology analysis of commonly downregulated genes in miR-1290-overexpressing and TBC1D4-knockdown BT-549 cells (log2(FPKM) + 1 ratio < 0.95 versus the respective controls). b Representative images of phospho-kinase arrays performed using lysates from BT-549 cells transfected with the miR-1290 mimic or mimic-NC and siTBC1D4 or siCont (left). The relative intensity of p-P70S6K is shown in the graph (right). c Expression of TSC1 in BT-549 cells treated as indicated. d Expression levels of TSC1 in normal (n = 114) and breast cancer (n = 1097) tissues from the TCGA database (***p < 0.001, Mann‒Whitney U test). e Analysis of the correlation between TBC1D4 and TSC1 expression in breast cancer tissues from the TCGA database (r = 0.251, p < 0.001). f Representative immunocytochemistry images showing the localization and expression of TBC1D4, TSC1, and p-P70S6K in BT-549 and Hs578T cells transfected with the miR-1290 mimic or mimic-NC and with siTBC1D4 or siCont. The nuclei were counterstained with DAPI. Scale bar, 150 μm. g Western blot analysis of the indicated proteins in BT-549 and Hs578T cells transfected with the miR-1290 mimic or mimic-NC and with siTBC1D4 or siCont. ACTB was used as a loading control

The RNA-seq results revealed that after treatment with the miR-1290 mimic or siTBC1D4, the expression level of TSC1 decreased compared with that after treatment with mimic-NC or siCont (Fig. 4c). Analysis of TCGA data revealed that TSC1 expression was significantly lower in breast cancer samples than in normal samples (p < 0.001, Mann‒Whitney U test; Fig. 4d). Furthermore, we observed a positive correlation between TBC1D4 and TSC1 expression levels in breast cancer tissues (r = 0.251, p < 0.001) (Fig. 4e). To validate this correlation at the tissue level in clinical TNBC specimens, we performed immunohistochemistry on a TNBC tissue microarray containing 46 patient-derived cores (Supplementary Table 1). H-score analysis confirmed a significant positive correlation between TBC1D4 and TSC1 protein expression in TNBC tissues (r = 0.381, p = 0.009) (Supplementary Fig. 10), which is consistent with the TCGA transcriptomic data and further supports the functional link between TBC1D4 and TSC1 in TNBC. Immunocytochemistry revealed significant decreases in the expression of TSC1 after treatment with the miR-1290 mimic and siTBC1D4. Additionally, the expression of phospho-P70S6K (p-P70S6K) clearly increased in response to the upregulation of miR-1290 expression and the downregulation of TBC1D4 expression (Fig. 4f, quantification shown in Supplementary Fig. 11). In addition, qRT‒PCR analysis revealed that TSC1 expression levels decreased upon treatment with BCAF-/MB-231-derived exosomes, the miR-1290 mimic, or siTBC1D4 (Supplementary Fig. 12). These results suggest that exosomal miR-1290 suppresses TSC1 expression indirectly by downregulating TBC1D4 expression.

Moreover, western blot analysis revealed that p-P70S6K expression and Ki67 expression were elevated in the miR-1290 mimic and siTBC1D4 groups, confirming that miR-1290-mediated TBC1D4 downregulation drives mTOR pathway activation and promotes cell proliferation (Fig. 4g). Treatment with BCAF-/MB-231-derived exosomes led to increases in p-P70S6K levels, similar to those observed with miR-1290 mimic and siTBC1D4 treatment, as confirmed by both western blotting and immunocytochemistry (Supplementary Fig. 13). Thus, we propose that exosome-mediated upregulation of miR-1290 directly downregulates TBC1D4 expression and is accompanied by reduced TSC1 expression, increased mTOR activity, and the subsequent promotion of TNBC cell proliferation.

To further validate the necessity of miR-1290 in the regulation of mTOR, we performed loss-of-function experiments using antagomir-1290 in BT-549 and Hs578T cells. The inhibition of miR-1290 expression significantly reduced cell proliferation, as demonstrated by crystal violet staining and CCK-8 assays (Fig. 5a and b). At the molecular level, antagomir-1290 treatment led to the recovery of TBC1D4 and TSC1 expression, with a concomitant decrease in p-P70S6K and Ki67 levels, as confirmed by qRT‒PCR, western blot, and immunocytochemistry analyses (Fig. 5c–e). These loss-of-function results demonstrate that miR-1290 is not only sufficient but also necessary for the activation of mTOR signaling and the promotion of TNBC cell proliferation.

Fig. 5.

Verification of the loss of function of miR-1290 in TNBC cell lines. a Crystal violet staining of BT-549 and Hs578T cells transfected with anta-NC or anta-1290 (left) and quantification of cell numbers (right). Scale bar, 500 μm. b Growth analysis by a CCK-8 assay in BT-549 and Hs578T cells transfected with anta-NC or anta-1290. c qRT‒PCR analysis of miR-1290, TBC1D4, and TSC1 expression in BT-549 and Hs578T cells transfected with anta-NC or anta-1290. d Western blot analysis of TBC1D4, P70S6K, p-P70S6K, Ki67, and ACTB in BT-549 and Hs578T cells transfected with anta-NC or anta-1290. ACTB was used as a loading control. e Representative immunocytochemistry images showing the expression of TBC1D4, TSC1, p-P70S6K, and Ki67 in BT-549 and Hs578T cells transfected with anta-NC or anta-1290. The nuclei were counterstained with DAPI. Scale bar, 150 μm. The quantification of the fluorescence intensity per cell is shown below each panel. The data are shown as the means ± SDs of three independent experiments, and statistical significance was determined using Student’s t tests (*p < 0.05, **p < 0.01, ***p < 0.001)

Rapamycin, an mTOR inhibitor, inhibits miR-1290-induced cell proliferation through mTORC1 signaling blockade

To confirm whether the miR-1290-induced downregulation of TBC1D4 and TSC1 promotes breast cancer cell proliferation through mTORC1 signaling, we used rapamycin, a highly selective and potent mTORC1 inhibitor^31,32 that inactivates S6K1 and 4E-BP1.^32,33 Cotreatment of TNBC cell lines with rapamycin and the miR-1290 mimic revealed that rapamycin effectively suppressed miR-1290-induced TNBC cell proliferation (Fig. 6a). At the molecular level, overexpression of miR-1290 via mimic transfection led to reduced expression of TBC1D4 and TSC1 (Fig. 6b and c). Western blot analysis revealed that the increase in p-P70S6K induced by miR-1290 was attenuated by rapamycin treatment. Additionally, the expression of the proliferation marker Ki67 was elevated by miR-1290 but decreased following rapamycin treatment (Fig. 6d). These results suggest that the miR-1290-induced activation of mTOR signaling can be modulated by rapamycin, leading to the inhibition of cell proliferation. Next, we incorporated 3D spheroid models to provide additional verification in a more physiologically relevant model. Using ultralow-attachment microplates, we established a TNBC spheroid model. These findings were further validated in a 3D spheroid model, in which the miR-1290 mimic similarly increased spheroid size and cell growth, both of which were suppressed by rapamycin treatment (p < 0.001 and p = 0.003, respectively; Fig. 6e and f). Consistent with the 2D results, miR-1290 decreased TBC1D4 and TSC1 expression, and rapamycin attenuated the increase in p-P70S6K and Ki67 levels induced by miR-1290 (Fig. 6g and h). Thus, elevated levels of miR-1290 in TNBC led to increased mTOR activity and TNBC cell proliferation, suggesting that miR-1290 plays a critical role in TNBC progression. These findings indicate that targeting miR-1290 and its downstream effector TBC1D4 may serve as a potential therapeutic target for TNBC treatment, as their dysregulation drives mTOR activation.

Fig. 6.

Rapamycin inhibits miR-1290-induced cell proliferation by blocking mTORC1 signaling. a Crystal violet staining of BT-549 and Hs578T cells treated with rapamycin and mimic-NC or miR-1290 mimic (left). Quantification of cell numbers (right). Scale bar, 500 μm. b qRT‒PCR analysis of miR-1290 expression in BT-549 and Hs578T cells under the indicated conditions. c qRT‒PCR analysis of TBC1D4 expression (upper) and TSC1 expression (lower) in BT-549 and Hs578T cells under the indicated conditions. d Western blot analysis of the indicated proteins in BT-549 and Hs578T cells under various treatment conditions with rapamycin. ACTB was used as a loading control. e Representative 3D spheroid images of BT-549 cells treated with rapamycin and mimic-NC or miR-1290 mimic (left). Quantification of spheroid diameter (right) (n = 7 per group). Scale bar, 500 μm. f Cell growth analysis of the spheroids shown in (e) via the CCK-8 assay (n = 7 per group). g qRT‒PCR analysis of miR-1290 expression (left), TBC1D4 expression (middle), and TSC1 expression (right) under the indicated conditions. h Western blot analysis of the indicated proteins in BT-549 3D spheroids under various treatment conditions. ACTB was used as a loading control. The data are shown as the means ± SDs of three independent experiments unless otherwise indicated, and statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test (**p < 0.01, ***p < 0.001)

Therapeutic targeting of miR-1290-induced mTOR activation with everolimus

Everolimus is an FDA-approved rapamycin analog used for the treatment of multiple cancers, particularly hormone receptor-positive and HER2-negative breast cancer,^33–37 but its application in TNBC has not been explored. Given our finding that miR-1290 upregulation increases mTOR activation in TNBC cell lines, we investigated the potential applicability of everolimus in this context.

Consistent with the rapamycin results, everolimus effectively suppressed miR-1290-induced cell proliferation (Fig. 7a), reversed the downregulation of TBC1D4 and TSC1 at the gene expression level (Fig. 7b and c), and attenuated the miR-1290-induced phosphorylation of P70S6K (Fig. 7d). In the 3D spheroid model, everolimus similarly suppressed miR-1290-induced cell growth and spheroid size (both p < 0.001; Fig. 7e and f) and reversed the molecular changes observed in 2D, including TBC1D4 and TSC1 downregulation and p-P70S6K elevation (Fig. 7g and h). Thus, these findings suggest that TNBC cells with elevated mTOR signaling driven by upregulated miR-1290 expression may be effectively targeted by everolimus, highlighting its potential as a novel therapeutic agent for TNBC.

Fig. 7.

Everolimus effectively suppresses miR-1290-mediated breast cancer cell proliferation. a Crystal violet staining of BT-549 and Hs578T cells treated with everolimus and mimic-NC or miR-1290 mimic (left). Quantification of cell numbers (right). Scale bar, 500 μm. b qRT‒PCR analysis of miR-1290 expression in BT-549 and Hs578T cells under the indicated conditions. c qRT‒PCR analysis of TBC1D4 expression (upper) and TSC1 expression (lower) in BT-549 and Hs578T cells under the indicated conditions. d Western blot analysis of the indicated proteins in BT-549 and Hs578T cells under various treatment conditions with everolimus. ACTB was used as a loading control. e Representative 3D spheroid images of BT-549 cells treated with everolimus and mimic-NC or miR-1290 mimic (left). Quantification of spheroid diameter (right) (n = 9 per group). Scale bar, 500 μm. f Cell growth analysis of the spheroids shown in (e) via the CCK-8 assay (n = 9 per group). g qRT‒PCR analysis of miR-1290 expression (left), TBC1D4 expression (middle), and TSC1 expression (right) under the indicated conditions. h Western blot analysis of the indicated proteins in BT-549 3D spheroids under various treatment conditions. ACTB was used as a loading control. The data are shown as the means ± SDs of three independent experiments unless otherwise indicated, and statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test (**p < 0.01, ***p < 0.001)

Everolimus effectively suppresses miR-1290-induced tumor growth in a xenograft model

To validate the therapeutic potential of everolimus against miR-1290-driven breast cancer progression in vivo, we established a xenograft model using NOD/Shi-scid/IL-2Rγ^null mice. BT-549 cells transfected with the miR-1290 mimic or mimic-NC were inoculated subcutaneously, followed by oral administration of everolimus or the DMSO vehicle starting 7 days after inoculation. Body weights were comparable across all treatment groups throughout the experimental period, indicating the absence of significant systemic toxicity from either miR-1290 upregulation or everolimus treatment (Supplementary Fig. 14). Consistent with the in vitro findings, compared with vehicle-treated control cells, miR-1290 mimic-treated BT-549 cells formed significantly larger tumors, both in size and weight (Fig. 8a–c). Notably, everolimus administration attenuated the increase in tumor size and weight caused by upregulated miR-1290 expression (Fig. 8a–c). Hematoxylin and eosin (H&E) staining revealed dense, viable tumor cells with minimal necrotic areas in tumor sections from both the control and miR-1290 groups under vehicle treatment. In particular, the miR-1290 group presented increased cellularity, indicating enhanced tumor growth (Fig. 8d). In contrast, both everolimus-treated groups (control and miR-1290) presented extensive areas of tumor cell death characterized by reduced cellular density, nuclear condensation, and an increased number of necrotic regions (Fig. 8d), suggesting that everolimus treatment effectively counteracted the tumor-promoting effects of miR-1290. Moreover, immunohistochemical analysis revealed that compared with control tumors, vehicle-treated miR-1290-overexpressing tumors presented markedly diminished TBC1D4 and TSC1 immunoreactivity, with decreased staining intensity and fewer positive cells (TBC1D4, p = 0.001; TSC1, p < 0.001; Fig. 8e). These tumors also exhibited robust p-P70S6K immunopositivity, which was intense and more widespread throughout the tumor cells, indicating constitutive mTORC1 pathway activation (Fig. 8e). Additionally, Ki67 expression was significantly elevated in vehicle-treated miR-1290-overexpressing tumors (p < 0.001; Fig. 8e), confirming the correlation between TBC1D4 inactivation and increased Ki67 expression in breast cancer. Surprisingly, everolimus treatment effectively neutralized the increase in p-P70S6K immunoreactivity induced by the miR-1290 mimic, confirming the specific inhibition of mTORC1 kinase activity (p < 0.001; Fig. 8e). This intervention markedly reduced the number of Ki67-positive cells (p < 0.001; Fig. 8e). These findings provide definitive in vivo evidence that pharmacological mTORC1 inhibition is an effective strategy to counteract miR-1290-induced TNBC tumor growth.

Fig. 8.

Everolimus suppresses miR-1290-mediated tumor growth in vivo. a Growth curves of xenograft tumors from the indicated treatment groups (n = 4 per group). b Representative images of xenograft tumors from the indicated treatment groups. c Tumor weight measurements at the endpoint of the xenograft experiment (n = 4 per group). d Representative H&E staining images of tumor sections. Scale bar, 200 μm. e Immunohistochemical analysis of TBC1D4, TSC1, p-P70S6K, and Ki67 expression in tumor sections from the indicated treatment groups (left) and quantification of staining intensity (right) (n = 4 per group). Scale bar, 200 μm. The data are shown as the means ± SDs, and statistical significance was determined using two-way repeated-measures ANOVA with Tukey’s multiple comparisons test for the tumor growth curves (a) and two-way ANOVA with Tukey’s multiple comparisons test for the remaining analyses (c, e) (*p < 0.05, **p < 0.01, ***p < 0.001)

Discussion

The TME that surrounds cancer cells comprises various immune cells, CAFs, vascular endothelial cells, and other stromal components.^7,8 Numerous factors derived from the TME, including exosomes, have been reported to promote tumor progression and contribute to resistance to anticancer therapies.^7,8 Among these components, CAFs constitute a significant proportion of the TME and function as key stromal cells. For example, CAFs secrete growth factors such as TGF-β and PDGF, as well as proangiogenic factors such as VEGF, thereby facilitating tumor growth and metastasis.^9,10 Recently, increasing evidence has highlighted the critical role of exosome-mediated communication from CAFs in supporting tumor growth and survival, and extensive research has been conducted on miRNAs, proteins, and other molecules contained within these exosomes.^9–11 Therefore, in this study, we performed exosomal RNA-seq to analyze the miRNA profiles of CAF-derived exosomes. To investigate the role of CAF-derived exosomes, we selected TNBC as our target breast cancer subtype. While other breast cancer subtypes, such as hormone receptor-positive or HER2-positive tumors, can be treated with targeted therapies such as tamoxifen or lapatinib, TNBC lacks specific therapeutic targets.^1,2 To identify exosomal miRNAs potentially involved in tumor progression, we analyzed miRNAs that were highly expressed in both the CAF-derived and the MB-231-derived exosomes. Through this approach, we identified miR-1290 as a key candidate miRNA. Since exosomal miR-1290 is upregulated in both CAFs and TNBC cells, it is a promising dual target for therapeutic intervention. Intriguingly, a recent study reported the oncogenic role of miR-1290 in hepatocellular carcinoma,²³ suggesting that it may play a similar oncogenic role in TNBC.

In this study, we demonstrated that exosomal miR-1290 from BCAFs and aggressive MB-231 breast cancer cells can be transferred to the recipient TNBC cell lines BT-549 and Hs578T, leading to significantly increased proliferation. The high expression of miR-1290 in the derived exosomes suggested a potential autocrine/paracrine signaling loop that could amplify the oncogenic function of miR-1290, further enhancing its tumor-promoting effects. Mechanistically, we identified TBC1D4 as a direct target of miR-1290 and demonstrated its role as a putative tumor suppressor in breast cancer. TBC1D4 is recognized primarily for its ability to regulate glucose transporter trafficking in response to insulin signaling.^25,26 However, TBC1D4 downregulation did not affect GLUT4 or GLUT1 localization, indicating that mTOR activation by TBC1D4 downregulation may be independent of its canonical role in glucose metabolism. We aimed to identify the novel function of TBC1D4 in TNBC; in this study, using RNA-seq analysis and phosphoprotein array profiling, we confirmed that TBC1D4 is associated with mTOR activation. These findings provide new insight into the novel role of TBC1D4 in TNBC and the mechanism by which its downregulation may contribute to tumor progression. While the luciferase reporter assay confirmed that TBC1D4 is a direct target of miR-1290, the link between TBC1D4 suppression and the subsequent reduction in TSC1 expression and mTOR activation is currently supported by correlative and functional evidence rather than by a fully defined molecular mechanism. A consistent positive correlation between TBC1D4 and TSC1 expression was observed in both the TCGA cohort and the TNBC tissue microarray data, and TBC1D4 knockdown reproducibly reduced TSC1 expression, whereas miR-1290 inhibition restored both TBC1D4 and TSC1 expression. Nevertheless, the intermediate regulatory node connecting these events has not yet been identified. Notably, TSC1 expression was reduced at the transcript level (Fig. 4c), suggesting that this regulation may occur primarily at the transcriptional level rather than through direct physical interaction. Thus, these findings raise the possibility that miR-1290-mediated downregulation of TBC1D4 reshapes a broader transcriptional program that indirectly represses TSC1, although the specific regulator(s) bridging TBC1D4 loss to TSC1 transcription remain to be defined. We therefore interpret the TBC1D4-TSC1 relationship as indirect, and delineating these intermediate mechanisms will be important for future work that may provide deeper insight into the broader oncogenic effects associated with miR-1290-mediated TBC1D4 dysregulation in TNBC.

In the cohort analyses, we observed that the level of exosomal miR-1290 was significantly greater in the blood of breast cancer patients than in that of normal controls, and likewise, the exosomal miR-1290 level was also greater in samples from TNBC patients than in those from normal controls (Fig. 1h and i). These findings suggested that elevated exosomal miR-1290 expression promoted TNBC progression, indicating the potential applicability of miR-1290 and TBC1D4 as therapeutic targets for TNBC treatment. Furthermore, analysis of patient-derived exosomes revealed that exosomal miR-1290 may also serve as a candidate diagnostic biomarker for breast cancer. Thus, exosomal miR-1290 has potential not only as a therapeutic target but also as a candidate diagnostic biomarker in breast cancer, although further validation in larger clinical cohorts is warranted.

In TNBC cell lines, miR-1290-driven suppression of TBC1D4 expression was associated with mTOR activation. Treatment with rapamycin, an mTOR inhibitor, effectively suppressed TNBC cell proliferation induced by miR-1290, indicating that miR-1290 promotes tumor growth via mTOR activation. These findings suggest that the inhibition of mTOR signaling, as achieved with rapamycin, may serve as a key therapeutic strategy for TNBC. Everolimus, an FDA-approved rapamycin analog, is currently used for the treatment of hormone receptor-positive and HER2-negative breast cancer, but it has not been utilized in the treatment of TNBC patients. In this study, we demonstrated that exosomal miR-1290 delivery increased mTOR activity in TNBC cells, suggesting the potential applicability of everolimus for TNBC therapy. Thus, patients with elevated levels of miR-1290 may exhibit increased mTOR activity and may benefit from everolimus treatment. However, to establish clinical relevance, further clinical studies investigating the efficacy of everolimus in TNBC patients are warranted.

Indeed, the mechanisms of resistance to everolimus in TNBC are multifactorial and have been repeatedly reported. First, while everolimus effectively inhibits mTORC1, it does not suppress mTORC2, leading to AKT reactivation and the restoration of downstream signaling, thereby reducing drug efficacy.^38–40 Second, activation of receptor tyrosine kinases (RTKs), such as EGFR and IGF-1R, or parallel activation of the RAS/RAF/MEK/ERK MAPK pathway can bypass mTOR inhibition and sustain tumor growth.^41–44 Furthermore, elements of the TME, including CAFs and immune cells, have been reported to reactivate the mTOR pathway and may counteract the inhibitory effects of everolimus.^45–48 Collectively, these findings suggest that resistance to everolimus in TNBC is a multifaceted phenomenon involving both pathway reactivation and microenvironmental factors. In this study, we demonstrated that exosomal miR-1290 derived from CAFs and surrounding cancer cells promotes cancer cell growth through enhanced mTOR signaling. Thus, mTORC1 signaling activated by elevated levels of miR-1290 may represent a major therapeutic target of everolimus, enabling more efficient suppression of TNBC cell growth. Accordingly, increased exosomal miR-1290 levels detected by liquid biopsy may provide a rationale for the clinical use of everolimus. Moreover, in patients with high miR-1290 expression, combined treatment with everolimus and MAPK inhibitors may overcome resistance mechanisms and allow for more accurate prediction of therapeutic responses.

Our findings regarding the role of BCAF-derived exosomal miR-1290 in TNBC progression align with and extend recent discoveries in other cancer types. The bidirectional transfer of miR-1290 between BCAFs and TNBC cells observed in our study revealed a complex intercellular communication network within the TME. While Bai et al. demonstrated that cancer cell-derived exosomal miR-1290 activates CAFs in lung adenocarcinoma,²⁰ our study revealed that the complementary process—CAF-derived miR-1290—promotes cancer cell proliferation. This bidirectional exchange suggests a positive feedback loop that amplifies tumor progression through continuous cross-talk between cancer cells and stromal components. Furthermore, our identification of TBC1D4-TSC1-mTOR regulation as a downstream target of miR-1290 provides new insights into its oncogenic mechanisms. Previous studies have identified different targets: FOXA2 in brain-metastatic breast cancer²¹ and OGN in ovarian cancer.²² This diversity of targets across cancer types suggests that miR-1290 functions as a versatile oncogenic regulator that adapts its effects depending on cellular context and tissue specificity. Notably, while Sirkisoon et al. focused on the role of miR-1290 in promoting brain metastasis,²¹ our study demonstrated its fundamental role in primary tumor growth.

Our investigation of miR-1290 function across multiple cancer types revealed that the miR-1290-mediated regulation of TBC1D4 is not restricted to TNBC but may represent a common oncogenic mechanism operating in various malignancies. The pro-proliferative effects of miR-1290 and its direct targeting of TBC1D4 were consistently observed in renal, colon, and lung cancer cell lines (Supplementary Fig. 15), all of which presented elevated miR-1290 levels in patient samples according to the GSE61741 dataset (Supplementary Fig. 16). This pancancer relevance of miR-1290-mediated TBC1D4 regulation has several important implications. First, therapeutic strategies targeting this pathway, such as mTOR inhibitors, including everolimus, may have broader applications beyond TNBC. Second, it raises the possibility that circulating miR-1290 may have biomarker utility beyond TNBC, although this would require dedicated validation in each cancer type. Third, this study establishes TBC1D4 as a putative tumor suppressor with relevance across multiple cancer types, warranting further investigation into its role in different oncogenic contexts. Future studies should explore whether the therapeutic efficacy of mTOR inhibition in miR-1290-high tumors extends to other cancer types beyond breast cancer.

In conclusion, RNA sequencing analysis of exosomes isolated from BCAFs and MB-231 cells revealed elevated levels of miR-1290. This upregulation of miR-1290 led to a reduction in TBC1D4 expression in recipient TNBC cells, which was accompanied by decreased TSC1 expression. The miR-1290-driven suppression of TBC1D4 was accompanied by mTOR activation and significantly enhanced TNBC cell proliferation in both 2D and 3D culture models. Consistent with these findings, in vivo experiments with miR-1290 mimic-treated cells revealed increased tumor size, indicating that miR-1290 plays a critical role in promoting TNBC progression. Furthermore, treatment with rapamycin or everolimus suppressed the miR-1290-induced activation of mTOR signaling and effectively inhibited cell growth. These findings demonstrated a novel oncogenic function of miR-1290 in TNBC and suggested that miR-1290 and its direct target TBC1D4 may serve as promising therapeutic targets in this subtype, for which effective targeted therapies are lacking. Additionally, elevated serum miR-1290 levels may be a candidate diagnostic biomarker and a potential predictive indicator for therapeutic response, although these findings are based on a limited patient cohort and require further clinical validation in larger, independent studies.

Materials and methods

Ethics statements

All experimental procedures involving animals were reviewed and authorized by the Institutional Animal Care and Use Committee (IACUC) of the Korea Research Institute of Bioscience and Biotechnology (KRIBB-AEC-24228). Human specimen collection and use were conducted under a protocol approved by the Institutional Review Board of Keimyung University Dongsan Hospital (DSMC 2023-12-023).

Cell culture and isolation of exosomes

Human breast cancer cell lines were obtained from the Korean Cell Line Bank (MB-231 and BT-549) and Elabscience (Hs578T). Human primary breast cancer-associated fibroblasts were acquired from Neuromics (CAF116). Cells were cultured in DMEM containing 10% FBS and 1% penicillin/streptomycin under standard humidified conditions at 37 °C with 5% CO₂. Exosomes were prepared using ExoQuick-TC (System Biosciences, EXOTC250A-1) following the manufacturer’s protocol. In brief, culture media were harvested after cells had been incubated for 48 h in DMEM supplemented with 1% exosome-depleted FBS (Thermo Fisher Scientific, A2720803). The collected supernatants were centrifuged at 3000 × g, combined with ExoQuick-TC at a 5:1 ratio, and kept at 4 °C overnight. Exosome pellets were then recovered by centrifugation at 1500 × g and resuspended in DPBS at 200 μL per 10 mL of starting supernatant. For exosome uptake analysis, purified exosomes were fluorescently labeled using the PKH67 Green Fluorescent Cell Linker Kit (Sigma, MINI67-1KT) in accordance with the supplier’s instructions. Briefly, exosome pellets were suspended in Diluent C containing PKH67 and incubated for 5 min at room temperature. The staining reaction was terminated by adding 1% BSA in DPBS. After washing by centrifugation, the labeled exosomes were resuspended in DPBS and applied to recipient cells for 1 h. Cell nuclei were then counterstained with Hoechst 33342 (Thermo Fisher Scientific, H3570).

Patients and clinical samples

A total of 39 cases were identified from the pathology reports of patients who underwent surgery for TNBC at Keimyung University Dongsan Hospital from 2017 to 2022. Among these cases, those that had negative results on both the hormonal receptor test and the HER2 test were selected. These were primarily evaluated by immunohistochemistry, and for cases with equivocal HER2 results, additional SISH (silver in situ hybridization) was performed on surgical specimens. Patients with other secondary primary tumors or who died from causes unrelated to breast cancer were excluded. Fresh samples of paired tumor and normal tissues and serum were frozen ( − 80 °C) and used for analysis. Patients’ clinical, radiological, and survival outcomes were analyzed retrospectively. Pathologic stages were assessed according to the classification system of the American Joint Committee on Cancer, 8th edition. All materials were obtained under institutional review board‑approved protocols (Approval No. DSMC 2023-12-023).

Patient exosome isolation

Serum samples collected from patients were preserved at −80 °C until exosome isolation. Before processing, frozen samples were thawed on ice and centrifuged to eliminate cellular debris. The resulting supernatants were mixed with ExoQuick-TC reagent (System Biosciences, EXOTC250A-1) in accordance with the manufacturer’s protocol. Following gentle mixing, the samples were incubated at 4 °C for 30 min to allow exosome precipitation. The precipitated exosomes were recovered by centrifugation at 1500 × g for 30 min at 4 °C. The final exosome pellets were resuspended in PBS and immediately subjected to subsequent analyses. Characterization of the isolated exosomes has been described previously.⁴⁹

Nanoparticle tracking analysis

Characterization of the isolated exosomes was performed using nanoparticle tracking analysis (NTA), which was carried out with a NanoSight NS300 system (Malvern Instruments) to determine their size distribution and concentration. NTA was used in combination with western blot-based marker validation and transmission electron microscopy to confirm that the isolated particles met the established exosome criteria. Briefly, exosome samples were diluted in sterile, filtered PBS to achieve a particle count within the recommended range for accurate detection. The instrument was aligned and calibrated according to the manufacturer’s instructions, and each sample was recorded for at least 60 s in triplicate. The Brownian motion of individual particles in the recorded videos was tracked and analyzed by NTA software, which calculates particle size and concentration on the basis of diffusion speed. The results were then averaged across measurements and reported as the mean ± standard deviation.

Negative staining and electron microscopy

For transmission electron microscopy, exosomes were subjected to negative staining by placing a formvar-coated grid onto a droplet of the exosome suspension on Parafilm for 1 min to facilitate particle adsorption. The grid was subsequently immersed in 2% (w/v) uranyl acetate for 30 s, excess stain was carefully removed using filter paper, and the grid was allowed to air dry. Images were acquired using a Talos L120C transmission electron microscope (Thermo Fisher Scientific) operated at an accelerating voltage of 120 kV.

miRNA, siRNA, and antagomir transfection

The cells were transfected with the miR-1290 mimic, mimic-NC, siTBC1D4, or siCont (Bioneer) at a final concentration of 100 nM using Lipofectamine RNAiMAX (Thermo Fisher Scientific, 13778150) for 48 h. The sequences were as follows: miR-1290: 5’-UGGAUUUUUGGAUCAGGGA-3’, siTBC1D4: 5’-CAGAAAGCUUUGAUGAACA-3’ and 5’-UGUUCAUCAAAGCUUUCUG-3’, and siCont: 5’-AUGAACGUGAAUUGCUCAA-3’ and 5’-UUGAGCAAUUCACGUUCAU-3’. For loss-of-function experiments, antagomir-1290 (5’-UCCCUGAUCCAAAAAUCCA-3’) and the corresponding negative control (5’-CAGUACUUUUGUGUAGUACAA-3’) were purchased from GenePharma and transfected into BT-549 and Hs578T cells under the same conditions described above.

3D spheroid culture

For 3D culture, ultralow attachment microplates (Corning, 7007) were used. BT-549 cells were transfected for 24 h, harvested, seeded (1 × 10⁵ cells/well), and treated with rapamycin (5 nM, MCE, HY-10219) or everolimus (10 μM, MCE, HY-10218) for 48 h. Spheroid formation and growth were monitored under a microscope (Olympus, CKX53), and the diameter was measured using imaging software. The spheroids were subsequently processed for RNA extraction or protein analysis.

Cell proliferation assay

For crystal violet staining, cells were plated in 6-well plates at a density of 2 × 10⁵ cells per well and exposed to exosomes (BCAF-derived exosomes, 150 μg/mL; MB-231-derived exosomes, 120 μg/mL), miRNA mimics, or siRNAs for 48 h. Following treatment, cells were fixed in pre-chilled methanol at −20 °C for 5 min, stained with 0.1% crystal violet solution (Sigma, C0775), rinsed thoroughly, and photographed. Cell proliferation in 3D spheroid cultures was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo Laboratories, CK04). After replacing the culture medium with a freshly prepared mixture of culture medium and CCK-8 reagent (9:1), the spheroids were incubated for 30 min, and absorbance was subsequently determined at 450 nm.

RNA extraction and quantitative real-time PCR (qRT‒PCR)

Total RNA was extracted using the RNeasy Mini Kit (Qiagen, 74106). First-strand cDNA was synthesized from 1 μg of total RNA with PrimeScript RT Master Mix (Takara, RR036A). Quantitative real-time PCR (qRT-PCR) was subsequently carried out using the SensiFAST SYBR Lo-ROX Kit (Meridian Bioscience, BIO-94050) on an AriaMx Real-Time PCR System (Agilent Technologies). The primer sequences used were as follows: TBC1D4, 5′-TTGAACGGTCCAACAGTCTT-3′ (forward) and 5′-CGGAATCCTCTTCGGGAAAC-3′ (reverse); TSC1, 5′-TACCTGGAAACCAGCTCTCA-3′ (forward) and 5′-CCCAGTAACGAGAGGATGGA-3′ (reverse); and ACTB, 5′-ACTCTTCCAGCCTTCCTTCC-3′ (forward) and 5′-CAATGCCAGGGTACATGGTG-3′ (reverse). For miRNA analysis, total miRNA was purified using the miRNeasy Mini Kit (Qiagen, 217004). Reverse transcription was performed with 1 μg of miRNA using the HB miR Multi Assay Kit System I (HeimBiotek), followed by qRT-PCR according to the manufacturer’s recommended protocol.

RNA sequencing and bioinformatics analysis

For exosomal RNA sequencing, RNA was extracted from purified exosomes using TRIzol LS reagent (Thermo Fisher Scientific, 10296028). RNA integrity and concentration were determined using an Agilent 2100 Bioanalyzer (Agilent Technologies, 5067-1514) together with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, ND-2000). Small RNA libraries were generated using the NEBNext Multiplex Small RNA Library Prep Kit (New England Biolabs, E7300). Following library preparation, PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, 28106) and PAGE gel purification. The quality of the final libraries was verified on an Agilent 2100 Bioanalyzer (Agilent Technologies, G2938A) with a High Sensitivity DNA assay (Agilent Technologies, G2938-90321). Sequencing was performed on the Illumina NextSeq500 platform (Illumina, SY-415-1002) using a single-end 75-bp sequencing strategy. For total RNA sequencing, sequencing libraries were prepared and purified from total RNA using the TruSeq RNA Sample Preparation Kit V2. The libraries were subsequently analyzed on an Illumina NextSeq 1000 platform (Illumina, 20038898) with a paired-end 2 × 100-bp configuration. Raw sequencing data were first processed with Cutadapt v1.18 to remove adapter sequences using the following parameters: ‘-a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC -A AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTA -m 50 -O 5’. Low-quality bases (Phred score <20) were trimmed using Sickle v1.33, and only reads of at least 50 bp were retained for subsequent analyses. Read quality and sequence duplication were evaluated with FastQC v0.11.4. Additional filtering to eliminate low-quality reads and sequences containing poly-N stretches was carried out using NGSQCToolkit v2.3.3. High-quality reads were aligned to the human reference genome GRCh38.97 (accession no. GCA_000001405.27) using HISAT2 v2.1.0. Gene expression levels and normalized read counts were calculated as fragments per kilobase of transcript per million mapped reads (FPKM) using StringTie v2.2.1.

Dual-luciferase reporter assay

The putative miR-1290 binding site within TBC1D4 was identified using TargetScan 7.2, and the corresponding wild-type (TBC1D4-WT) and mutant (TBC1D4-mut) sequences were inserted into the pmirGLO vector (Promega, E1330). The resulting reporter construct (50 ng) was introduced into HEK293 cells together with 100 nM miR-1290 mimic or mimic-NC by cotransfection with Lipofectamine RNAiMAX (Thermo Fisher Scientific, 13778150). At 24 h post-transfection, luciferase activity was quantified with the Dual-Luciferase Reporter Assay System (Promega, N1610).

Western blot analysis

Cells were lysed with RIPA buffer (Thermo Fisher Scientific, 89901) containing protease inhibitor cocktail (Roche, 11873580001). Cell lysates were clarified by centrifugation at 14,000 × g for 20 min at 4 °C, after which protein concentrations were measured using the Bradford protein assay. Equal amounts of protein were resolved on 4–20% precast polyacrylamide gels (Bio-Rad, #4561096) and electrotransferred onto nitrocellulose membranes (Bio-Rad, #1620145). Membranes were incubated with the following primary antibodies: CD63 (1:200; ABclonal, A5271), CD81 (1:200; ABclonal, A5270), flotillin-1 (1:1000; ABclonal, A6220), α-SMA (1:1000; ABclonal, A17910), FAP (1:1000; ABclonal, A23789), GM130 (1:1000; Abcam, ab52649), H3 (1:1000; Cell Signaling Technology, 4620S), ACTB (1:20,000; ABclonal, AC026), TBC1D4 (1:200; Cell Signaling Technology, 2670S), P70S6K (1:500; Cell Signaling Technology, 9202S), phospho-P70S6K (1:500; ABclonal, AP0564), and Ki67 (1:1000; Abcam, ab16667). An HRP-conjugated secondary antibody (sc-2357; Santa Cruz Biotechnology) was applied at a dilution of 1:5000. Immunoreactive bands were detected using a chemiluminescent substrate (Thermo Fisher Scientific, 34580) and captured with a Mini HD9 imaging system (Uvitec).

Human phospho-kinase array

Phosphorylation profiles were analyzed using the Proteome Profiler Human Phospho-Kinase Array Kit (R&D Systems, ARY003C) following the manufacturer’s recommended protocol. In brief, 400 μg of total protein from each cell lysate was applied to nitrocellulose membrane arrays and incubated overnight at 4 °C. After washing, the membranes were sequentially incubated with the supplied detection antibody cocktail and streptavidin-conjugated HRP. Chemiluminescent signals were subsequently developed, captured using a Mini HD9 imaging system (UVitec), and quantified with ImageJ software.

Immunocytochemistry

For immunofluorescence analysis, cells were fixed in 4% paraformaldehyde for 10 min, permeabilized with 0.5% Triton X-100 for an additional 10 min, and subsequently blocked with 5% BSA for 30 min. The samples were then incubated with primary antibodies against TBC1D4 (1:50; Cell Signaling Technology, 2670S), TSC1 (1:200; Proteintech, 29906-1-AP), phospho-P70S6K (1:200; ABclonal, AP0564), Ki67 (1:200; Abcam, ab16667), GLUT1 (1:500; Abcam, ab115730), or GLUT4 (1:500; Abcam, ab33780). Following washing steps, Alexa Fluor 488-conjugated secondary antibody (1:500; Thermo Fisher Scientific, A11034) was applied. Cell nuclei were counterstained with DAPI (Vector Laboratories, H-1200), and fluorescent images were captured using an EVOS M5000 imaging system (Thermo Fisher Scientific).

Xenograft tumor model

All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of the Korea Research Institute of Bioscience and Biotechnology. Animal experiments were performed in compliance with the Guidelines for the Welfare and Use of Animals in Cancer Research. Throughout the study, mice were routinely monitored, and tumor burdens remained within the recommended welfare criteria. Five-week-old female NOD/Shi-scid/IL-2Rγ^null mice (Saeronbio) were randomly allocated to the indicated treatment groups. BT-549 cells at 70–80% confluence were transfected with either miR-1290 mimic or mimic-NC (100 nM). A total of 5 × 10⁶ cells suspended in 200 μL PBS were subcutaneously injected into the upper right dorsal region of each mouse. In the initial xenograft study, tumor growth was assessed by measuring tumor volume every 2–4 days beginning on day 10 after cell implantation. Mice were sacrificed on day 34, after which tumors were excised, weighed, and photographed. For the everolimus study, drug administration was initiated 7 days after tumor cell implantation. Animals were randomly divided into four treatment groups: mimic-NC + DMSO, mimic-NC + everolimus, miR-1290 + DMSO, and miR-1290 + everolimus. Everolimus (20 mg/kg) or the corresponding vehicle (DMSO) was administered by oral gavage five times per week for 32 days. Tumor volumes were recorded at intervals of 2–5 days throughout the treatment period. At the experimental endpoint (day 38), mice were euthanized and tumor tissues were harvested for further analyses.

Histological analysis

The tumor tissues were fixed (4% paraformaldehyde, 24 h), dehydrated with a graded ethanol series, cleared in xylene, and paraffin-embedded. Sections (5 μm) were prepared for H&E staining and immunohistochemistry. For immunohistochemistry, the sections were subjected to antigen retrieval (Vector Laboratories, H-3300), blocking with endogenous peroxidase (Agilent Technologies, S2023), and protein blocking (Agilent Technologies, X0909). The primary antibodies used were as follows: TBC1D4 (1:100; Cell Signaling Technology, 2670S), TSC1 (1:200; Proteintech, 29906-1-AP), phospho-P70S6K (1:200; ABclonal, AP0564), and Ki67 (1:200; Abcam, ab16667). The sections were incubated with an HRP-conjugated anti-rabbit IgG polymer detection kit (Vector Laboratories, MP-7451), developed with DAB (Vector Laboratories, SK-4105), and counterstained with hematoxylin (Agilent Technologies, S3309).

Tissue microarray and digital IHC image analysis and H-score quantification

A triple-negative breast cancer tissue microarray (TissueArray. Com, BR1509A) containing 46 patient-derived cores with clinicopathological information (age, TNM classification, grade, and stage) was used for tissue-level validation. Sections were stained with anti-TBC1D4 (1:100, Cell Signaling Technology, 2670S) and anti-TSC1 (1:200, Proteintech, 29906-1-AP) antibodies following the immunohistochemistry protocol described above. Immunohistochemistry-stained tissue microarray images were analyzed using the Qanti® Discovery digital pathology analysis platform (AIVIS, Inc.). Individual cells were automatically detected within analyzable tissue regions, and the cell-level chromogenic staining intensity was quantified for subsequent H-score analysis. Hematoxylin and 3,3’-diaminobenzidine (DAB) signals were separated using color deconvolution⁵⁰. For each detected cell, DAB optical density was quantified within segmented cell-level regions, including nuclear and/or cytoplasmic masks, according to the analysis setting. The extracted DAB intensity values were then converted to a standardized continuous intensity scale ranging from 0 to 1 using a platform-defined normalization and transformation procedure. For H-score calculation, the detected cells were assigned to ordinal staining intensity categories corresponding to negative, weak, moderate, and strong staining using fixed platform settings applied consistently across all samples within each marker analysis. The H score was calculated using the conventional intensity-weighted formula^51,52:

$$\mathrm{H\; score}=\left(1\times P_{\mathrm{weak}}\right)+\left(2\times P_{\mathrm{moderate}}\right)+\left(3\times P_{\mathrm{strong}}\right)$$

where P_weak, P_moderate, and P_strong represent the percentages of detected cells assigned to the weak, moderate, and strong staining intensity categories, respectively. The final H score ranged from 0 to 300.

Analysis of public datasets

To identify candidate miRNAs that are consistently dysregulated in breast cancer/TNBC, we analyzed five independent GEO datasets (GSE37963, GSE45666, GSE45498, GSE38167, and GSE61438). For each dataset, miRNA expression between breast cancer/TNBC samples and normal controls was compared using a two-tailed Welch’s t test. miRNAs with |log2(fold change)| > 0.58 and nominal p < 0.05 were selected as dataset-level candidates. Because this analysis was intended as an exploratory cross-dataset candidate screening step rather than a definitive within-cohort discovery analysis, within-dataset multiple-testing correction was not applied for candidate selection. Instead, candidate robustness was strengthened by the need for concordant upregulation across all five independent cohorts, followed by independent clinical assessment and experimental validation of the selected candidate. To assess the potential of miR-1290 as a noninvasive diagnostic marker, we analyzed the GSE73002 dataset, which comprises serum miRNA profiles from 2686 normal individuals and 1280 breast cancer patients. For this serum-based comparison, as well as for the comparison in our independent Keimyung University Dongsan Hospital serum cohort, log2-transformed miR-1290 expression levels were compared between patients and normal controls using two-sided Mann‒Whitney U tests. In addition, expression comparisons between normal and breast cancer tissues in the TCGA cohort were performed using two-sided Mann‒Whitney U tests. To explore miR-1290 expression across different cancer types, we analyzed the GSE61741 dataset. Normalized miR-1290 expression levels were determined from blood samples of patients with various cancer types. Pairwise comparisons between each tumor group and the normal control group were performed using two-sided Mann‒Whitney U tests with tie-aware permutation-based p values. Multiple comparisons across the five tumor-versus-normal tests were adjusted using the Holm step-down procedure. Holm-adjusted p values less than 0.001 are reported as p < 0.001; all other adjusted p values are shown to three decimal places.

Statistical analysis

Unless otherwise specified, all experiments were independently repeated at least three times, with n representing the number of biological replicates. Data are expressed as the mean ± standard deviation (SD). The luciferase reporter assay was the only exception and was evaluated using six technical replicates. For factorial analyses involving mimic-NC- or miR-1290-transfected cells treated with DMSO or an mTOR inhibitor (rapamycin or everolimus), statistical differences were assessed by two-way analysis of variance (ANOVA) followed by Tukey’s multiple-comparison test. Tumor growth data obtained from the xenograft studies were analyzed using two-way repeated-measures ANOVA with Tukey’s post hoc multiple-comparison test. Kaplan–Meier survival curves were generated for survival analysis and compared using the log-rank test. The prognostic relevance of miR-1290 target genes in the TCGA datasets was examined using the OncoLnc platform (https://www.oncolnc.org).⁵³ Associations between variables were evaluated by Pearson’s correlation analysis. The statistical method used for each dataset is described in the corresponding figure legend. A p value below 0.05 was considered statistically significant, and significance was denoted as *p < 0.05, **p < 0.01, or ***p < 0.001. All statistical analyses were carried out using R software and GraphPad Prism.

Author contributions

K. Hur, D.-S. Kim, M.-Y. Son, T.-S. Han and H.-S. Cho conceived and designed the study. J. Lee, G. Kim, H. W. Lee, I. H. Tae, Y. Kang, J. Lee, J. Kim, T.-H. Han, H. S. Ban, C.-R. Jung, and J.H. Lim developed the methodology. T.-S. Han, T. Son, and J. Lee performed the animal experiments. S.-G. Lee, K. Park, and D.-S. Kim analyzed and interpreted the data. K. Hur, D.-S. Kim, M.-Y. Son, and H.-S. Cho wrote and reviewed the manuscript. T.-H. Han, K. Hur, D.-S. Kim, M.-Y. Son, and H.-S. Cho supervised the study. All the authors have read and approved the article.

Data availability

All reagents used in this study are commercially available or may be obtained from the corresponding author upon reasonable request. The mRNA sequencing results from this research have been deposited in the NCBI Gene Expression Omnibus (GEO, www.ncbi.nlm.nih.gov/geo) under accession number GSE338456. Complete, uncropped images of the original Western blots, including molecular weight markers, are provided in the Supplementary Materials.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41392-026-02922-y.