Abstract
Background
Circular RNAs (circRNAs) appear to exert critical functions in breast cancer (BC). The objective of this study is to explore the usefulness of circRNAs as potential diagnostic and prognostic biomarkers of BC.
Methods
The Gene Expression Omnibus (GEO) database was referenced to identify differentially expressed circRNAs in BC. Quantitative real-time polymerase chain reaction (qPCR) was used to detect the expression levels of hsa_circ_0001756 in both BC tissue samples and BC-derived cell lines. The functions of hsa_circ_0001756 were investigated both in vitro and in vivo. The luciferase reporter and rescue assays were used to clarify the molecular mechanisms of hsa_circ_0001756. Receiver operating characteristic (ROC) curve was established to evaluate the clinical value of hsa_circ_0001756 as a serum biomarker, and to investigate its potential correlation with the clinical pathological characteristics of BC patients by Chi-squared test.
Results
Hsa_circ_0001756 expression was upregulated in BC tissues and substantially correlated with tumor size and tumor-node-metastasis (TNM) stage. Knockdown (KD) of hsa_circ_0001756 markedly inhibited the malignant potential of BC both in vitro and in vivo. Mechanistically, hsa_circ_0001756 acted as a miR-584-5p sponge to regulate TRAF6 in BC cells. Serum levels of hsa_circ_0001756 were significantly higher in pre-operative BC patients than in healthy controls, fibroadenoma patients, and post-operative BC patients. Also, serum hsa_circ_0001756 was remarkably correlated with tumor size, patient age, metastasis state, and TNM stage. The combination of the traditional tumor markers carcinoembryonic antigen (CEA) and cancer antigen 15-3 (CA15-3) with hsa_circ_0001756 significantly improved the diagnostic accuracy of BC.
Conclusions
Our findings indicated that hsa_circ_0001756 could promote BC malignant progression through the miR-584-5p/TRAF6 signaling axis. Especially, hsa_circ_0001756 in serum holds promise as a biomarker for BC screening and diagnosis.
Keywords: Hsa_circ_0001756, breast cancer (BC), miR-584-5p, TRAF6, biomarker
Highlight box.
Key findings
• The upregulation of hsa_circ_0001756 in breast cancer (BC) tissues and cell lines promotes tumor progression and epithelial-mesenchymal transition (EMT) through the miR-584-5p/TRAF6 signaling pathway.
• Serum hsa_circ_0001756 levels are elevated in BC patients and show high diagnostic accuracy, especially when combined with traditional biomarkers, carcinoembryonic antigen (CEA) and cancer antigen 15-3 (CA15-3).
What is known and what is new?
• The role of hsa_circ_0001756 has been studied in several cancers, such as ovarian cancer and colorectal cancer (CRC).
• Our findings establish hsa_circ_0001756 as a tumor-promoter in BC and highlight its potential as a novel biomarker in serum.
What is the implication, and what should change now?
• Hsa_circ_0001756 promotes the progression of BC via the miR-584-5p/TRAF6 axis.
• Serum hsa_circ_0001756 holds potential as a clinical biomarker.
Introduction
Breast cancer (BC) has emerged as the most prevalent cancer in women worldwide, threatening their lives and health and posing a huge burden to healthcare systems (1). Hence, regular screening for early diagnosis is particularly important to improve prognosis and survival. Therefore, it is essential to identify novel molecular biomarkers and elucidate the functions and mechanisms to improve the diagnostic accuracy of BC.
Circular RNAs (circRNAs) are characterized by a covalently closed continuous loop structure without a 5'-cap or 3'-poly A tail (2), and were initially regarded as aberrant splicing by-products, thus attracting little attention. However, recent advancements in molecular techniques and bioinformatics have clarified the roles of various circRNAs in carcinogenesis and tumor progression. For instance, circRNAs can act as sponges of microRNAs (miRNAs) (3) and some contain intronic sequences that have been implicated in transcription of the parental gene by interacting with RNA polymerase II (4). Additionally, circRNAs with open reading frames and internal ribosome entry sites have the potential to translate proteins (5). For example, circSMARCA5 can suppress the transcription of its parental gene by forming an R-loop that reduces Pol II occupancy (6). CircRNAs have been shown to be associated with several cancer types, such as colorectal cancer (CRC) (7) and hepatocellular carcinoma (8). However, the functions and mechanisms of numerous circRNAs implicated in the onset and progression of BC remain largely unknown.
CircHIPK2 (hsa_circ_0001756), which is derived from exon 2 of the HIPK2 gene, has been shown to regulate astrocyte activation in collaboration with autophagy by sponging microRNA-124 (MIR124-2HG) (9). Moreover, hsa_circ_0001756 is reported to contribute to the functional recovery of neuronal stem cells after ischemic stroke (10). A recent study found that hsa_circ_0001756 promoted progression of ovarian cancer via activation of the EGFR/MAPK signaling pathway (11). A previous study by our group demonstrated that lipopolysaccharide (LPS) stimulation increased the invasive and metastatic capacities of BC cells via the TLR4/NF-κB signaling pathway (12). In this study, bioinformatics analysis was conducted with reference to the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) to identify differentially expressed circRNAs associated with BC. The results confirmed that hsa_circ_0001756 expression was relatively upregulated in BC tissues as compared with adjacent non-tumor tissues. Moreover, LPS stimulation was found to enhance hsa_circ_0001756 expression in mesenchymal-like (MDA-MB-231) BC cells as compared with epithelial-like BC cells (MCF-7). Furthermore, an investigation to establish correlations between hsa_circ_0001756 in BC tissues and clinicopathological factors found that hsa_circ_0001756 expression was related to tumor size and tumor-node-metastasis (TNM) stage. Collectively, these findings indicate that hsa_circ_0001756 is associated with the progression of BC. Our data further confirmed that hsa_circ_0001756 modulated the expression of TRAF6 by competitively binding to miR-584-5p, thus promoting the proliferation, migration, and invasion of BC cells. In summary, these findings position hsa_circ_0001756 as a potential diagnostic and prognostic biomarker and a viable candidate for therapeutic targeting in BC. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1610/rc).
Methods
Clinical tissues, serum samples and BC cell lines
Surgically resected cancer tissues and adjacent non-tumor tissues (≥2 cm away from the tumor margin) were obtained from 21 BC patients, a pieces of each tissue aliquots (approximately 20–30 mg) were immersed in cold RNAlater solution (4 ℃) (Ambion, Austin, USA) for RNA extraction, others were immediately (within 30 min) frozen with liquid nitrogen and stored at −80 ℃ for further use. All tissue samples were confirmed by pathological examination. Serum samples from healthy controls, patients with fibroadenomas, and BC patients, both pre- and post-operatively, were pretreated and stored at −80 ℃, Serum samples were processed to avoid hemolysis (pinkish/red color). Serum samples for the healthy group were collected from individuals undergoing routine physical examinations, while other groups were assigned based on comprehensive clinical assessments and pathological examination results. No significant differences in age were observed across these groups. All procedures were performed using RNase-free consumables to ensure RNA integrity. All patients included in the study did not receive preoperative radiotherapy, chemotherapy, or targeted drugs. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Review Board of The Second Affiliated Hospital of Soochow University (project No. JD-LK-2021-115-01), and written informed consent was obtained from all participants. Human embryonic kidney (HEK)-293T, MDA-MB-231, and MCF-7 cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific, Grand Island, NY, USA). All cells were cultivated in a humidified incubator at 37 ℃ with 5% CO2.
Transfection
Cells were transfected with short interfering RNA (siRNA), miRNA inhibitors, and miRNA mimics (GenePharma Co., Ltd., Suzhou, China) using Lipofectamine 2000 reagent (Invitrogen Corporation, Waltham, MA, USA) in 6-well plates at 40-60% confluence. Si-RNA (si-hsa_circ_0001756#1 and si-hsa_circ_0001756#2), miR-584-5p inhibitor and miR-584-5p mimics were transfected at concentrations of 250 and 100 nM, respectively. Non-targeting siRNA and miRNA negative controls were used for all experiments. The RNA sequences used in this study are listed in Tables S1-S3.
Quantitative real-time polymerase chain reaction (qPCR)
Total RNA was extracted from BC cells and tissues using TRIzol reagent (Invitrogen Corporation) and the nuclear and cytoplasmic fractions were separated with a Cytoplasmic & Nuclear RNA Purification Kit (Norgen Biotek Corp., Thorold, ON, Canada). Total RNA was isolated from serum samples using TRIzol LS reagent (Invitrogen Corporation) in accordance with the manufacturer’s instructions. For reverse transcription, 500 ng of total RNA was used for circRNA and mRNA using the HiScript® III 1st Strand cDNA Synthesis Kit (+ gDNA wiper), and the cDNA was quantitated using AceQ® qPCR SYBR Green Master Mix (Vazyme Biotech Co., Ltd., Nanjing, China) with the primers (Sangon Biotech Co., Ltd., Shanghai, China) listed in Table S4. In a total reaction volume of 20 µL with 0.4 µM primers. The thermal cycling conditions were: 95 ℃ for 10 min; 40 cycles of 95 ℃ for 10 s and 60 ℃ for 30 s, followed by a melt curve analysis. No-template and no-reverse transcription controls were included. Primer efficiencies (90–110%) and R² values were validated using standard curves. For the quantitative detection of circRNAs, it was critical to employ primers that were specifically designed to span the unique back-splice junction (BSJ). Relative expression of mRNA was calculated using the 2−ΔΔCt method with U6 or GAPDH as an internal control. The primers of miRNA for reverse transcription and qPCR analysis were designed using MiRNA Design V 1.01 software (https://www.vazyme.com/companyfile/653.html). The sequences were listed in Tables S5,S6. All experiments were repeated three times (biological replicates, n=3, each with technical triplicates).
Vector construction and dual-luciferase reporter assay
Wild-type or mutated sequence of hsa_circ_0001756 covering miR-584-5p binding sites was subjected to insertion into the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Madison, WI, USA), named as hsa_circ_0001756-WT/Mut constructs. The inserts of hsa_circ_0001756-WT were synthetic oligonucleotides with 183 nt, including all three predicted miR-584-5p seed sites. HEK-293T cells (5×105) were plated in 12-well plates and cultured for 24 hours to reach an appropriate confluence before transfection. Then, dual-luciferase reporter vectors were co-transfected with miR-584-5p mimics or mimics NC using Lipofectamine 2000. After 24 h, firefly luciferase activity was measured using the Dual-Glo® Luciferase Assay System (Promega, Madison, WI, USA). The Firefly Luciferase/Renilla Luciferase was calculated for normalization calculations. Hsa_circ_0001756-WT and TRAF6-WT sequences, including predicted miR-584-5p seed sites, are listed in Table S7. The result of the empty pmirGLO vector served as a negative control, as shown in Figure S1. Assays were performed in three independent biological replicates, each with technical triplicates.
RNase R treatment
The resistance of circRNAs to RNase was confirmed using Ribonuclease R (Epicentre Technologies Corporation, Madison, WI, USA). Briefly, RNA (2 µg) was treated with Ribonuclease R (6 U) at 37 ℃ for 15 min. The control group was incubated under the same conditions without Ribonuclease R. Then, the enzyme was rendered inactive by heating it at 70 ℃ for 10 min. Acquired RNA products were used for qPCR or analyzed by 1.0% agarose gel electrophoresis following conventional PCR and imaged using the Bio-Rad Gel-Doc XR+ system (Bio-Rad Laboratories, Hercules, CA, USA). All experiments were performed with three technical replicates and repeated in three independent biological experiments.
Colony formation assay
At 48 hours post-transfection, cells were trypsinized and seeded into 6-well plates at a density of 1×103 cells per well. After approximately 14 days, the resulting cell colonies were fixed with 4% paraformaldehyde, stained with crystal violet [0.5% weight/volume (w/v)], washed with phosphate-buffered saline (PBS), and imaged. Each condition was assayed in triplicate wells per experiment, and the entire assay was independently repeated three times.
Cell counting kit-8 (CCK-8) assay
Transfected cells were seeded in 96-well plates at a density of 2×103 cells per well in 100 µL of DMEM. Every 24 h, 10 µL of CCK-8 reagent (Dojindo Laboratories Co., Ltd., Kumamoto, Japan) was added to the corresponding wells, and the plate was incubated at 37 ℃ in a humidified 5% CO2 incubator for 2 h. Afterward, the absorbance at 450 nm was measured by Tecan Spark 10 M (Tecan Group Ltd., Männedorf, Switzerland) spectrophotometer. The absorbance values of the blanks were subtracted from the sample readings. Each condition was assessed with five technical replicates, and the experiment was independently repeated three times.
Wound healing assay
The 2×105 transfected BC cells were plated in the wells of 6-well plates and cultured until they reached ~100% confluence. The cell monolayer was scratched in a straight line once with a sterile 200 µL pipette tip and then washed with PBS to remove detached cells. Images were observed at 0 h (immediately after scratching) and at the endpoint (24 h for MDA-MB-231 and 48 h for MCF-7) using an inverted microscope. Cells were maintained in serum-free medium and incubated at 37 ℃ with 5% CO2 during the assay. The relative wound closure area was quantified using ImageJ software (https://imagej.net/ij/), and the percentage of wound closure was calculated as: [(area at 0 h − area at 24 or 48 h) / area at 0 h] × 100. Data from three independent experiments were analyzed, and results are expressed as mean ± standard error of mean (SEM).
Transwell migration and invasion assays
Cell migration and invasion were assessed using 24-well Transwell chambers (Corning, USA). For the invasion assay, chambers were pre-coated with Matrigel (Corning) prior to use, following the manufacturer’s protocol. A total of 1×105 transfected BC cells in 200 µL of serum-free medium were seeded in the upper chamber. DMEM supplemented with 10% FBS was added to the lower compartment as a chemoattractant. After incubation for 8 h (migration) or 24 h (invasion) at 37 ℃, the cells on the upper surface of the membrane were removed with a cotton swab. The cells that had migrated or invaded the lower surface were fixed with 4% paraformaldehyde for 20 min and stained with crystal violet (0.5% w/v) for 20 min, and imaged under a microscope. The mean number of cells in five random fields was counted using ImageJ software. Each experiment was performed with duplicate chambers per condition and repeated in three independent biological replicates (n=3). Data are presented as the mean ± standard deviation (SD) from the three biological replicates.
Western blot analysis
BC cells were lysed with radioimmunoprecipitation assay (RIPA) buffer containing a protease inhibitor cocktail (phenylmethylsulfonyl fluoride). Protein concentration was determined using a bicinchoninic acid (BCA) assay kit (Beyotime, Shanghai, China). Equal amounts of protein (20–40 µg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after denaturation in Laemmli sample buffer containing β-mercaptoethanol at 95 ℃ for 5 minutes. The proteins were then transferred to a polyvinylidene fluoride (PVDF) membrane (EMD Millipore Corporation, Billerica, MA, USA). The membrane was blocked for 1 h with Tris-Buffered Saline with Tween-20 (TBST) contain 5% (w/v) skim milk, and incubated overnight at 4 ℃ with primary antibodies against E-cadherin (dilution, 1:500; Cell Signaling Technology, Inc., Danvers, MA, USA), N-cadherin (dilution, 1:1,000; Cell Signaling Technology, Inc.), vimentin (dilution, 1:1,000; Bioworld, Dublin, OH, USA), TRAF6 (dilution, 1:1,000; Cell Signaling Technology, Inc.), and GAPDH (dilution, 1:1,000; Cell Signaling Technology, Inc.). The antibody dilution is prepared using TBST containing 2.5% skim milk. After washing, the membrane was incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit or anti-mouse secondary antibodies (dilution, 1:5,000; Multisciences Biotech Co., Ltd., Hangzhou, China) for 1 h at room temperature. Afterward, the protein bands were visualized using chemiluminescent reagent and imaged. Band intensity was quantified using ImageJ software and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Data are representative of three independent biological experiments.
Immunohistochemical analysis
Tissues harvested from nude mice were fixed with formalin, sliced after being paraffin-embedded, dewaxed, rehydrated, and then boiled for 30 min in citrate buffer (10 mM, pH 6.0) for antigen retrieval. Endogenous peroxidase activity was reduced by exposure to 3% hydrogen peroxide for 10 min. To minimize background signal from endogenous biotin, an avidin/biotin blocking step was performed prior to antibody incubation. Slides were then blocked using 5% bovine serum albumin (BSA) (Boster Bioengineering, Wuhan, China), incubated with primary antibodies against epithelial-mesenchymal transition (EMT) markers and Ki-67 (dilution, 1:200; Servicebio, Wuhan, China) at 4 ℃ overnight, followed by biotin-labeled secondary antibodies at 37 ℃ for 1 h. Slides were visualized with 3,3'-diaminobenzidine (DAB) and stained with hematoxylin and observed under a microscope. Immunohistochemistry (IHC) scoring was performed by two independent pathologists who were blinded to the group allocation.
Subcutaneous xenograft model in BALB/c nude mice
Female 6-week-old BALB/c nude mice (purchased from Shanghai Sushang Biotechnology Co., Ltd., Shanghai, China) were reared under specific pathogen-free (SPF) conditions with a 12-hour light/dark cycle, controlled temperature (22±2 ℃) and humidity (50%±10%), and provided with autoclaved food and water ad libitum. After one week of acclimatization, mice were randomly assigned to groups (n=6 per group) using a random number table. Pre-transfected MDA-MB-231 cells (passage <20, mycoplasma-free, viability >95%) were harvested and resuspended at a density of 5×106 cells in 100 µL PBS, kept on ice. Under isoflurane anesthesia, the injection site (mammary fat pad) was disinfected with alcohol, and the cell suspension was injected subcutaneously. Tumor growth was monitored twice weekly by measuring the longitudinal (L) and latitudinal (W) diameters with a caliper. The investigator performing the tumor measurements was blinded to the group allocation. After 21 days, the nude mice were euthanized by cervical dislocation under 3% isoflurane deep anesthesia, and the xenografts were photographed and weighed. Tumor volume was calculated using the formula: Volume (mm3) = 0.5 × L × W2. Then, two small pieces of tissue specimens were used to extract RNA and perform immunohistochemical analysis. The remaining tissues were stored at −80 ℃. All animals in this study were euthanized at the experimental endpoint. The humane endpoints, such as decreased food and water consumption, weight loss reaching 20%, max tumor diameter >20 mm, and so on, were not used during the experiment. The animal assay was performed under a project license (No: JD-LK-2021-115-01) granted by the ethics committee of The Second Affiliated Hospital of Soochow University, in compliance with the Chinese guidelines for the care and use of animals. A protocol was prepared before the study without registration.
Statistical analysis
Statistical analysis was performed using IBM SPSS 24.0 (IBM Corporation, Armonk, NY, USA). Student’s t-test and Wilcoxon rank-sum test were used to assess differences between two groups, and multiple-group comparisons were analyzed by one-way ANOVA. Chi-squared test was used to analyze the correlation between serum hsa_circ_0001756 expression and the clinicopathological data. The predictive performance of the model was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC), with internal validation performed using the Bootstrap method (n=1,000). Differences in AUC values were compared using DeLong’s test. All statistical diagrams were generated using Prism 8.0 software (GraphPad Software, Inc., San Diego, CA, USA). Data were expressed as mean ± SD. A probability (P) value <0.05 was considered statistically significant.
Results
Verification of hsa_circ_0001756 in BC cells and tissues
Initially, the microarray data from the GEO dataset GSE182471 were reviewed to compare the expression patterns of circRNAs in BC tissues and adjacent non-tumor breast tissues. A corresponding volcano plot was constructed [fold change >1.5 and false discovery rate (FDR) <0.05] and showed that the expression of hsa_circ_0001756 was upregulated in BC tissues (Figure 1A). Subsequent comparison of 21 pairs of BC and adjacent non-tumor tissues found that the expression levels of hsa_circ_0001756 were considerably higher in BC tissues (Figure 1B). Moreover, hsa_circ_0001756 expression was considerably higher in MDA-MB-231 cells than in MCF-7 cells (Figure 1C). Correlation analysis of the clinicopathological data of 21 BC patients showed that hsa_circ_0001756 expression was positively associated with tumor size and TNM stage (Figure 1D,1E; Table S8). Collectively, these findings suggest that upregulation of hsa_circ_0001756 potentially contributes to the progression of BC.
Figure 1.
Verification of hsa_circ_0001756 in BC cells and tissues. (A) Volcano plot showing dysregulation of circRNAs determined by analysis of the GEO dataset (fold change >1.5, FDR <0.05). The red spots in the upper right and upper left represent upregulated and downregulated circRNAs, respectively. (B) Expression of hsa_circ_0001756 was confirmed in 21 pairs of BC tissues and adjacent non-tumor tissues. (C) The expression profiles of hsa_circ_0001756 in MDA-MB-231 and MCF-7 were clarified by qPCR analysis. (D,E) Expression of hsa_circ_0001756 in BC tissues was positively correlated with tumor size and TNM stage. (F) The existence of hsa_circ_0001756 from cDNA and gDNA of MDA-MB-231 cells was detected by conventional PCR analysis and agarose gel electrophoresis. Divergent primers amplified hsa_circ_0001756 from cDNA but not gDNA. (G) Expression of hsa_circ_0001756 and GAPDH were detected by agarose gel electrophoresis after amplification of RNase R-treated production. (H,I) Expression of hsa_circ_0001756 and GAPDH were detected by qPCR analyses after RNase R treatment. Data are presented as the mean ± standard deviation. *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. BC, breast cancer; cDNA, complementary DNA; circRNAs, circular RNAs; DAB, 3,3'-diaminobenzidine; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; FDR, false discovery rate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; gDNA, genomic DNA; GEO, Gene Expression Omnibus; HIPK2, homeodomain-interacting protein kinase-2; LPS, lipopolysaccharide; M, marker; NBC, non-breast cancer; qPCR, quantitative real-time polymerase chain reaction; TNM, tumor-node-metastasis.
Hsa_circ_0001756 consists of 1084 nucleotides from exon 2 of the homeodomain-interacting protein kinase-2 (HIPK2) gene that form a circular structure (Figure S2). Sanger sequencing with divergent primers identified the presence of a BSJ sequence in the conventional PCR product (Figure S2). In consideration of the fact that circRNAs are generated by back-splicing of pre-mRNAs, appropriate divergent primers were designed. The circRNA-specific primers amplify cDNA-derived products, but not genomic DNA (gDNA), as confirmed by electrophoresis results (Figure 1F). RNase digestion revealed that hsa_circ_0001756 was resistant to RNase R treatment (Figure 1G-1I). The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) linear transcript was used as a positive control for validation.
Knockdown (KD) of hsa_circ_0001756 inhibited cell growth and motility of BC cells in vitro
To explore the possible functions of hsa_circ_0001756 in vitro, BC cells were transfected with siRNA targeting the BSJ of hsa_circ_0001756. The interference efficiency of the siRNA was evaluated by detecting the expression levels of hsa_circ_0001756 (Figure S3A,S3B) and EMT markers (Figure S3C,S3D). Due to higher KD efficiency, si-hsa_circ_0001756#2 was selected for subsequent functional experiments. The qPCR results showed that si-hsa_circ_0001756#2 only targeted hsa_circ_0001756 and had little effect on linear HIPK2 (Figure S3E,S3F). Cell growth analysis demonstrated that KD of hsa_circ_0001756 reduced the proliferation of BC cells (Figure 2A-2C). The wound healing and transwell migration assays demonstrated that silencing of hsa_circ_0001756 inhibited migration of BC cells (Figure 2D-2F). Additionally, the transwell invasion assay revealed that downregulation of hsa_circ_0001756 significantly reduced the invasive capacity of BC cells (Figure 2G). Notably, although MDA-MB-231 cells have higher baseline hsa_circ_0001756 expression, the inhibitory effects appeared more pronounced in MCF-7 cells. This discrepancy might be attributed to differences in KD efficiency, proliferation rate, or baseline motility between the two cell lines. Moreover, in MDA-MB-231 cells, downregulation of hsa_circ_0001756 significantly increased E-cadherin protein levels, but had opposite effects on N-cadherin and vimentin protein levels (Figure 2H,2I).
Figure 2.
KD of hsa_circ_0001756 reduced clonogenic growth, migration, and invasion of BC in vitro. (A-C) The CCK-8 and colony formation assays were used to assess proliferation of BC cells transfected with siRNA. (D-F) The wound healing and migration assays were conducted to detect migration of BC cells transfected with siRNA. Scale bars =50 µm. (G) The transwell invasion assay was used to detect invasive capacity of BC cells transfected with siRNA. Scale bars =50 µm. (H,I) Expression profiles of proliferation (E-cadherin, N-cadherin, and vimentin) in MDA-MB-231 cells after transfection with si-hsa_circ_0001756 by western blot analysis. GAPDH was used as a housekeeping gene. Data are presented as the mean ± standard deviation from three independent experiments. (C,F,G) Stained with crystal violet (0.5% weight/volume). *, P<0.05; **, P<0.01; ***, P<0.001. BC, breast cancer; CCK-8, Cell Counting Kit-8; EMT, epithelial-mesenchymal transition; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; KD, knockdown; OD, optical density; si-NC, short interfering RNA negative control; siRNA, short interfering RNA.
KD of hsa_circ_0001756 inhibited tumor growth and EMT of BC cells in vivo
To explore the functions of hsa_circ_0001756 in vivo, BALB/c-nu mice were subcutaneously injected with MDA-MB-231 cells transfected with siRNA targeting hsa_circ_0001756 to induce tumorigenesis. The results showed that KD of hsa_circ_0001756 dramatically inhibited the growth of subcutaneous tumors (Figure 3A). The weight and size (volume, mm3) of the tumor tissues were reduced in the treatment group compared with the small interfering RNA negative control (si-NC) (Figure 3B,3C). To further confirm the anti-proliferative effect, Ki-67 IHC was performed, showing reduced proliferation in the KD group (Figure 3D). The expression levels of EMT-related genes in subcutaneous tumor tissues were assessed by qPCR analysis. The results revealed that E-cadherin expression was notably increased in the si-hsa_circ_0001756 group, while the expression levels of N-cadherin and vimentin were substantially reduced (Figure 3E). IHC results were consistent with qPCR analyses in subcutaneous tumor tissues (Figure 3F).
Figure 3.
KD of hsa_circ_0001756 inhibited tumor growth and EMT of BC cells in vivo. (A) Xenograft tumors images of the si-NC and si-hsa_circ_0001756 groups. (B,C) Downregulation of hsa_circ_0001756 decreased the weight and volume of xenograft tumors. (D) The expression levels of Ki-67 in tumors of nude mice were detected by IHC (×100 and ×400; scale bar, 50 μm). (E,F) The expression levels of E-cadherin, N-cadherin, and vimentin in tumors of nude mice were detected by qPCR and IHC analyses (×100 and ×400; scale bar, 50 μm). GAPDH was used as a housekeeping gene. Data are presented as the mean ± standard deviation. **, P<0.01; ***, P<0.001. BC, breast cancer; EMT, epithelial-mesenchymal transition; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IHC, immunohistochemical; KD, knockdown; qPCR, quantitative real-time polymerase chain reaction; si-NC, small interfering RNA negative control.
Hsa_circ_0001756 functions by sponging miR-584-5p
To elucidate the potential roles of hsa_circ_0001756 in the onset and progression of BC, a nucleocytoplasmic distribution assay was conducted to determine the intracellular localization of hsa_circ_0001756. As a result, hsa_circ_0001756 was predominantly detected in the cytoplasm of BC cells (Figure 4A,4B), which is consistent with a potential “miRNA sponge” function. Subsequently, CircBank (http://www.circbank.cn) and CircInteractome (https://circinteractome.nia.nih.gov) databases identified nine potential target miRNAs (miR-1178-3p, miR-1261, miR-338-3p, miR-516a-5p, miR-578, miR-581, miR-584-5p, miR-663b, miR-889-3p) of hsa_circ_0001756 (Figure 4C). Further analysis revealed that miR-584-5p expression levels were considerably lower in BC tissues than in adjacent non-tumor tissues. In addition, KD of hsa_circ_0001756 significantly increased miR-584-5p expression levels in MDA-MB-231 cells (Figure S4). To further validate the interaction between hsa_circ_0001756 and miR-584-5p, a wild-type dual-luciferase vector containing hsa_circ_0001756 sequences (hsa_circ_0001756 WT) and mutant (MUT) constructs with mutated miR-584-5p binding sites (hsa_circ_0001756 MUT) were established (Figure 4D). The dual-luciferase reporter assay revealed that miR-584-5p mimics reduced luciferase activity in wild-type, while no discernible relative luciferase activity change was observed in MUT (Figure 4E), indicating that miR-584-5p directly interacts with hsa_circ_0001756.
Figure 4.
Hsa_circ_0001756 acts as a miR-584-5p sponge. (A,B) Presence of hsa_circ_0001756 in the nuclear and cytoplasmic fractions of BC cells. U6 and GAPDH were used as controls for the nuclear and cytoplasmic fractions, respectively. (C) A Venn diagram showing the overlap of target miRNAs of hsa_circ_0001756 was predicted with the CircBank and CircInteractome databases. (D) Schematic model of the dual-luciferase plasmids with WT and MUT sequences containing putative binding sites of miR-584-5p. (E) The relative luciferase activities were detected after co-transfection of hsa_circ_0001756 MUT or hsa_circ_0001756 WT and miR-584-5p mimics or NC mimics, respectively. (F) Expression of miR-584-5p in MDA-MB-231 cells after transfection with miR-584-5p mimics was determined by qPCR analysis. (G) The proliferation of MDA-MB-231 cells with transfection of miR-584-5p mimics and NC mimics was measured by colony formation assays. (H-J) The migration of MCF-7 and MDA-MB-231 cells was measured by wound healing and transwell migration assays. (K) The invasive ability of MDA-MB-231 cells was measured by the transwell invasion assay. Scale bars =50 µm. (L,M) Western blot detected the expression of E-cadherin, N-cadherin, and vimentin in MDA-MB-231 cells transfected with miR-584-5p mimics and NC mimics. (N,O) The rescue experiment between miR-584-5p and hsa_circ_0001756 in proliferation, migration, and invasion of MDA-MB-231 cells were measured by colony formation, transwell migration, and invasion assays. Scale bars =50 µm. *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. BC, breast cancer; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MUT, mutant; NC, negative control; qPCR, quantitative real-time polymerase chain reaction; si-NC, small interfering RNA negative control; WT, wild-type.
To validate the functions of miR-584-5p, BC cells were transfected with miR-584-5p mimics to overexpress miR-584-5p. The transfection efficiency was confirmed by qPCR results in MDA-MB-231 cells (Figure 4F). The colony formation assay revealed that miR-584-5p inhibited clonogenic growth of both BC cell lines (Figure 4G). The wound healing and transwell assays demonstrated that overexpression of miR-584-5p significantly impaired migration and attenuated the invasive capacity of BC cells (Figure 4H-4K). Western blot analysis revealed that miR-584-5p mimics increased protein expression of E-cadherin in MDA-MB-231 cells, but had opposite effects on protein expression of N-cadherin and vimentin (Figure 4L,4M).
To further verify that hsa_circ_0001756 functions via sponging miR-584-5p, a series of rescue experiments was conducted. The results showed that miR-584-5p inhibitors reversed the attenuation of proliferation, migration, and invasion in MDA-MB-231 cells caused by KD of hsa_circ_0001756 (Figure 4N,4O).
Hsa_circ_0001756 acts as a competing endogenous RNA (ceRNA) to regulate TRAF6
CircRNAs can regulate downstream target genes by sponging miRNAs. Screening of the TargetScan (https://www.targetscan.org/vert_80/), mirDIP (https://ophid.utoronto.ca/mirDIP/) and miRWALK (http://mirwalk.umm.uni-heidelberg.de/) databases identified 531 potential target genes of miR-584-5p (Figure 5A). The results of Gene Ontology biological process (GO-BP) enrichment analysis confirmed that some terms were highly enriched, such as positive regulation of transcription by RNA polymerase II, intracellular protein transport and mitotic cell cycle. Several pathways, including “Pathways in cancer”, “PD-L1 expression and PD-1 checkpoint pathway in cancer” and “Toll-like receptor signaling pathway”, were revealed as significantly enriched by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway. Then, we selected five genes (PPP2CA, TRAF6, MDM2, KRAS, and PRKAA2) associated with cancer for further analysis. According to the expression in BC cells, we found that the expression of tumor necrosis factor receptor-associated factor 6 (TRAF6) and PRKAA2 was markedly higher in MDA-MB-231 cells than that in MCF-7 cells (Figure 5B), and TRAF6 was attenuated more significantly by miR-584-5p mimics (Figure 5C). Moreover, the protein levels of TRAF6 were decreased significantly upon downregulation of hsa_circ_0001756 in MDA-MB-231 cells (Figure 5D). Therefore, we chose TRAF6 for the follow-up study. According to the dual luciferase reporter assays, miR-584-5p mimics reduced luciferase activity significantly (Figure 5E,5F). In addition, downregulation of TRAF6 markedly impaired the mobility and invasiveness of MDA-MB-231 cells (Figure 5G). Western blot analysis showed that downregulation of TRAF6 increased E-cadherin protein levels and decreased protein levels of N-cadherin and vimentin (Figure 5H,5I).
Figure 5.
Hsa_circ_0001756 acts as a ceRNA to regulate TRAF6. (A) Venn diagram detailing the exploration of miR-584-5p downstream targets predicted by TargetScan, mirDIP and miRWalk. (B) The expression levels of five genes in two BC cells. (C) The expression levels of TRAF6 and PRKAA2 in miR-584-5p mimics and NC mimics groups. (D) The protein levels of TRAF6 in MDA-MB-231 cells transfected with si-hsa_circ_0001756. (E) Schematic model of WT and MUT sequences of putative binding sites in the 3'UTR of TRAF6. (F) The dual-luciferase reporter analysis of TRAF6 and miR-584-5p in 293T cells. (G) Transwell migration and invasion analyses of MDA-MB-231 cells transfected with si-TRAF6. Scale bars =50 µm. (H,I) Expression levels of E-cadherin, N-cadherin, and vimentin in MDA-MB-231 cells transfected with si-TRAF6 were determined by western blot analysis. (J,K) Rescue experiments identified TRAF6 as a downstream target of miR-584-5p. Scale bars =50 µm; (J,K,N,O) stained with crystal violet (0.5% weight/volume). *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. BC, breast cancer; ceRNA, competing endogenous RNA; MUT, mutant; si-NC, small interfering RNA negative control; TRAF6, tumor necrosis factor receptor-associated factor 6; UTR, untranslated region; WT, wild-type.
Rescue experiments were performed to confirm that TRAF6 is a downstream molecule of miR-584-5p. The findings show that downregulation of miR-584-5p increased mobility and invasiveness in MDA-MB-231 cells, whereas co-transfection with si-TRAF6 and miR-584-5p inhibitors partly reversed this effect (Figure 5J,5K). Collectively, these findings confirm that suppression of hsa_circ_0001756 reduced proliferation, migration, and invasion of BC cells via miR-584-5p/TRAF6 signaling axis.
Clinical application of hsa_circ_0001756 in BC
To explore the potential clinical value of hsa_circ_0001756 as a biomarker, serum levels of hsa_circ_0001756 in healthy controls, patients with fibroadenomas, and BC patients both pre- and post-operatively were measured. Intriguingly, hsa_circ_0001756 expression was notably elevated in BC patients as compared with the other three experimental groups (Figure 6A). Subsequently, ROC curves were generated to assess the diagnostic value of circ_00001756. The AUC of BC patients and healthy controls was 0.833 [95% confidence interval (CI): 0.762–0.905, Figure 6B]. The AUC of BC and fibroadenoma patients was 0.772 (95% CI: 0.695–0.849, Figure 6C). The AUC of pre- and post-operative BC patients was 0.954 (95% CI: 0.914–0.993, Figure 6D). Then, potential correlations of serum levels of hsa_circ_0001756 with the clinicopathological characteristics of 86 BC patients were investigated. The results showed that hsa_circ_0001756 expression was markedly linked with tumor size, patient age, lymph node metastases, and TNM stage (Table 1), indicating the potential of hsa_circ_0001756 as a serum biomarker for screening, diagnosis, and prognosis of BC. To further evaluate the diagnostic value of hsa_circ_0001756, serum levels of cancer antigen 15-3 (CA15-3) and carcinoembryonic antigen (CEA) were evaluated. Among the serum biomarkers evaluated, hsa_circ_0001756 showed the highest diagnostic accuracy with an AUC of 0.827 (95% CI: 0.743–0.910). This performance was significantly better than CEA (P=0.02) and comparable to CA153 (P=0.39). Statistical comparison using DeLong’s test indicated that the combined model performed significantly better than each individual biomarker, including hsa_circ_0001756 (P=0.04), suggesting complementary value among the markers (hsa_circ_0001756, CA15-3, and CEA). (Figure 6E-6G). The combination of all three indicators for diagnosis of BC had the highest AUC value (AUC =0.897, 95% CI: 0.829–0.965) with a sensitivity of 77.2% and specificity of 93.5% (Figure 6H).
Figure 6.
Clinical application of hsa_circ_0001756 in BC. (A) Serum expression levels of hsa_circ_0001756 in healthy controls, fibroadenoma patients, and pre- and post-operative BC patients, GAPDH is used for normalization. (B) ROC curve analysis of BC patients and healthy controls. (C) ROC curve analysis of BC and fibroadenoma patients. (D) ROC of pre- and post-operative BC patients. (E) DeLong’s test analysis difference of AUC between the combined model and CEA. (F) DeLong’s test analysis difference of AUC between the combined model and CA15-3. (G) DeLong’s test analysis difference of AUC between the combined model and hsa_circ_0001756. (H) ROC of CA15-3, CEA, hsa_circ_0001756, and the combination of CA15-3, CEA, and hsa_circ_0001756 as serum biomarkers of BC. The combination of CA15-3, CEA, and hsa_circ_0001756 had the highest AUC value for diagnosis of BC; (G,J,K) stained with crystal violet (0.5% weight/volume). ***, P<0.001; ns, not significant. AUC, area under the curve; BC, breast cancer; CA15-3, cancer antigen 15-3; CEA, carcinoembryonic antigen; CI, confidence interval; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ROC, receiver operating characteristic.
Table 1. Correlation analysis between clinical characteristics and expression levels of hsa_circ_0001756 in serum of 86 BC patients.
| Characteristics | No. of patients | Hsa_circ_0001756 expression | Exact P value | |
|---|---|---|---|---|
| Low | High | |||
| Tumor size (cm) | 0.005 | |||
| ≤2 | 39 | 26 | 13 | |
| >2 | 47 | 17 | 30 | |
| Age (years) | 0.01 | |||
| ≤50 | 44 | 28 | 16 | |
| >50 | 42 | 15 | 27 | |
| Pathological type | 0.58 | |||
| IDC | 70 | 34 | 36 | |
| DCIS and others | 16 | 9 | 7 | |
| Lymph node metastasis | 0.01 | |||
| 0 | 59 | 35 | 24 | |
| ≥1 | 27 | 8 | 19 | |
| TNM stage | 0.03 | |||
| 0–I | 32 | 21 | 11 | |
| II–III | 54 | 22 | 32 | |
| HER2 | 0.26 | |||
| − | 55 | 25 | 30 | |
| + | 31 | 18 | 13 | |
BC, breast cancer; DCIS, ductal carcinoma in situ; HER2, human epidermal growth factor receptor 2; IDC, invasive ductal carcinoma; TNM, tumor-node-metastasis.
The prevalence, prognosis, and current therapeutic strategies significantly differ among the four major subtypes of BC: luminal A, luminal B, human epidermal growth factor receptor 2 (HER2)-enriched, and triple-negative (TN) (13). In accordance with the criteria established by the Chinese Society of Clinical Oncology, BC was classified based on expression of the estrogen receptor, progesterone receptor, and HER2. The results of qPCR analysis of these four subtypes showed that hsa_circ_0001756 expression was significantly elevated in the luminal A, luminal B, and TN subtypes compared with health, but not the HER2-enriched subtype (Figure 7A). ROC analysis highlighted the superior diagnostic accuracy of hsa_circ_0001756 for the luminal B subtype (AUC =0.888, specificity =76.5%, and sensitivity =98%, Figure 7B-7D).
Figure 7.
Clinical application of hsa_circ_0001756 as biomarkers for the major subtypes of BC. (A) Serum expression levels of hsa_circ_0001756 for the luminal A, luminal B, HER2-enriched, and TN subtypes compared with health. (B) ROC of the luminal A and healthy control groups. (C) ROC of the luminal B and healthy control groups. (D) ROC of the TN and healthy control groups. **, P<0.01; ***, P<0.001; ****, P<0.0001. AUC, area under the curve; BC, breast cancer; HER2, human epidermal growth factor receptor 2; ROC, receiver operating characteristic; TN, triple negative; TNBC, triple-negative breast cancer.
Discussion
CircRNAs have garnered significant attention in the field of cancer research, which was recently reviewed by Conn et al. (14). With the rapid development of bioinformatics, the roles of many circRNAs have been gradually elucidated (15). Numerous studies have reported aberrant expression of circRNAs in tumor tissues and their influence on tumor cell growth, apoptosis, invasion, and metastasis, highlighting potential as novel biomarkers (8,16).
To identify dysregulated circRNAs in BC, the GEO circRNA microarray dataset was referenced to assess differential expression of BC tissues and adjacent non-tumor tissues. Surprisingly, hsa_circ_0001756 expression was significantly upregulated in BC tissues and correlated with tumor size and TNM stage. Recent studies have shown that circHIPK2 (hsa_circ_0001756) contributes to cell growth in CRC by enhancing TAZ translation (17). CircHIPK2 plays a key role in colon cancer through the miR-373-3p/RGMA/BMP signaling axis (18). However, the role of hsa_circ_0001756 in BC has not been reported.
The structure of circRNAs is distinct from that of linear mRNA counterparts. Divergent primers were designed specifically for the splicing site of hsa_circ_0001756. The specificity of the designed primers and identity of the amplified products were confirmed by agarose gel electrophoresis and qPCR analysis. A previous report has confirmed that circRNAs can resist digestion by RNase R (19). In the present study, RNase R digestion had relatively little effect on hsa_circ_0001756. In addition, since circRNAs are formed by reverse splicing of pre-mRNAs, gDNA contains no complementary sites for the primers. Agarose gel electrophoresis confirmed that hsa_circ_0001756 is amplified from cDNA, but not gDNA. Collectively, these findings established a foundation for further study of the role of hsa_circ_0001756 in BC.
The results of this study show that hsa_circ_0001756 is aberrantly expressed in different BC cells, and expression in BC tissues is significantly associated with TNM stage and tumor size. Moreover, low expression of hsa_circ_0001756 was related to decreased proliferation of BC cells. Furthermore, EMT, a fundamental process in tumor biology, plays a pivotal role in the invasive and metastatic capacities of tumor cells (20,21). An intriguing aspect of our findings is the cell context-dependent nature of hsa_circ_0001756’s function. While its KD significantly suppressed migration in both MDA-MB-231 and MCF-7 cells, the relative effect was more substantial in MCF-7 cells. Given that the KD efficiency was comparable between both cell lines (Figure S3), we propose that the differential effect may stem from the distinct molecular and phenotypic contexts of these cells. The highly aggressive, mesenchymal-type MDA-MB-231 cells possess a complex network of redundant pathways driving migration and invasion. Consequently, targeting one component, such as the hsa_circ_0001756/miR-584-5p axis, may yield a less pronounced relative impact due to compensatory mechanisms. On the other hand, the epithelial-type MCF-7 cells, with their generally lower migratory capacity and fewer redundant pathways, appear more vulnerable to the disruption of this specific regulatory circuit. This observation underscores the importance of the cellular background in determining the functional outcome of a non-coding RNA. In vitro assays revealed that KD of hsa_circ_0001756 increased E-cadherin expression and inhibited expression of N-cadherin and vimentin. In vivo, downregulation of hsa_circ_0001756 inhibited the tumorigenesis of BC cells in nude mice. Collectively, these results demonstrated that hsa_circ_0001756 promoted the progression of BC.
To investigate the potential mechanism of hsa_circ_0001756 in BC, RNA was isolated from the nuclear and cytoplasmic fractions of BC cells. The results proved that hsa_circ_0001756 was distributed in both the cytoplasm and nucleus of BC cells and that it therefore had the potential to function as a miRNA sponge. Bioinformatics and dual-luciferase reporter assays found that miR-584-5p was the main target of hsa_circ_0001756. Previous studies have reported that circPITX1 regulates the malignancy of human glioblastoma via the miR-584-5p/KPNB1 signaling pathway (22), and circSMYD4 modulates the viability of liver cancer cells by adsorbing miR-584-5p (23). However, the impact of hsa_circ_0001756 on the malignant potential of BC cells and the possible involvement of miR-584-5p have not been explored.
Therefore, in vitro experiments were conducted to investigate the impact of miR-584-5p on BC cells. The results showed that overexpression of miR-584-5p significantly reduced the proliferation, migration, and invasion of BC cells. Western blot analysis also demonstrated that miR-584-5p mimics impeded EMT of MDA-MB-231 cells. Rescue studies indicated that the regulatory effects of hsa_circ_0001756 in BC cells were partially reversed by the inhibition of miR-584-5p. While the precise mechanistic details, such as AGO2 dependency, were not explored herein, our functional evidence-including direct binding confirmed by dual-luciferase assay and a compelling rescue of phenotype-strongly supports the conclusion that hsa_circ_0001756 functions as a decoy for miR-584-5p. Future studies utilizing techniques like AGO2-RIP could provide further molecular insights.
TRAF6 is an adaptor protein that activates downstream pathways involving both TNFR and the IL-1R/TLR families. TRAF6 participates in a variety of biological processes, including cancer progression and cell autophagy (24). For example, ubiquitination of GPX4 by TRAF6 facilitates its recognition by p62 and subsequent degradation via selective autophagy (25). In addition, downregulation of TRAF6 increased pyroptosis and inhibited autophagy of BV2 cells via the miR-146a-5p/TRAF6 axis (26). According to bioinformatics analysis and confirmatory experiments, TRAF6 was selected as a target gene of miR-584-5p. KD of TRAF6 attenuated growth and mobility of BC cells. Rescue studies confirmed that the biological function of miR-584-5p was dependent on targeting TRAF6. All of these indicated that the hsa_circ_0001756/miR-584-5p/TRAF6 signaling axis was associated with the development of BC.
Several studies have confirmed the association of circRNAs with the pathological type, TNM stage, and lymph node metastasis of BC (27), indicating the potential of circRNAs as novel biomarkers (28). In the present study, serum levels of hsa_circ_0001756 were relatively increased in BC patients and strongly associated with tumor size, patient age, metastasis status, and TNM stage, suggesting it as a novel biomarker for the diagnosis, and prognosis of BC. ROC analysis revealed that hsa_circ_0001756 (AUC =0.827, 95% CI: 0.743–0.910) demonstrated a significantly higher diagnostic accuracy than CEA (AUC =0.646, 95% CI: 0.524–0.768; DeLong’s test P=0.02) and a comparable AUC to CA15-3 (AUC =0.766, 95% CI: 0.662–0.871; P=0.39). Furthermore, the combination of hsa_circ_0001756, CA15-3, and CEA yielded a combined model with an AUC of 0.897 (95% CI: 0.772–0.944), which was significantly superior to any single biomarker (all P<0.05). The robustness of the combined model was supported by internal validation using bootstrap resampling (n=1,000), showing a mean AUC of 0.906. In addition, low expression of hsa_circ_0001756 in the serum of BC patients after surgery implied its potential utility as an indicator for monitoring surgical efficacy. Moreover, hsa_circ_0001756 showed diagnostic potential across various BC subtypes with the exception of the HER2-enriched type. Given these findings, hsa_circ_0001756 may be a promising biomarker for BC. Next, our study will focus on the levels of expression of hsa_circ_0001756 in the tumor microenvironment and tumor-immersed immune cells, as well as the saliva samples, to confirm the diagnostic value in BC. While our data demonstrate that hsa_circ_0001756 functions as a sponge for miR-584-5p, particularly through loss-of-function and rescue experiments, this study has certain limitations. First, the evidence would be further strengthened by gain-of-function approaches, such as hsa_circ_0001756 overexpression combined with miR-584-5p mimic rescue. Second, the use of a single siRNA for hsa_circ_0001756 KD, while effective, prompts caution; future studies employing additional distinct siRNAs or in vivo models would help to conclusively rule out off-target effects and reinforce the specificity of our findings. Third, the diagnostic performance of the biomarkers was evaluated in a single-center cohort. The absence of an independent, external validation cohort necessitates caution in generalizing these findings, and this remains an important objective for future research. Despite these limitations, the consistent data from multiple experimental approaches support our proposed model.
Conclusions
These findings suggest that hsa_circ_0001756 may contribute to the carcinogenesis of BC via the hsa_circ_0001756/miR-584-5p/TRAF6 signaling axis. Consequently, serum hsa_circ_0001756 shows potential as a biomarker worthy of further investigation for its possible role in the screening, diagnosis, and prognosis of BC.
Supplementary
The article’s supplementary files as
Acknowledgments
None.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Review Board of The Second Affiliated Hospital of Soochow University (project No. JD-LK-2021-115-01), and written informed consent was obtained from all participants. The protocol of the animal study was approved by the Institutional Animal Care and Use Committee of The Second Affiliated Hospital of Soochow University (approval No. JD-LK-2021-115-01) and in compliance with the Chinese guidelines for the care and use of animals.
Footnotes
Reporting Checklist: The authors have completed the MDAR and ARRIVE reporting checklists. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1610/rc
Funding: This work was supported by the National Natural Science Foundation of China (No. 81702078); The Natural Science Foundation of Jiangsu Province (No. BK20170356); Gusu Talent Program (No. GSWS2023041); “National Tutor System” Training Program for Health Youth Key Talents in Suzhou (No. Qngg2023008); The Second Affiliated Hospital of Soochow University Pre-Doctoral Research Fund (No. SDFEYBS2218); Discipline Construction Support Project Advantageous Disciplines Support Project (No. XKTJ-XK202403-2).
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1610/coif). B.W. reports funding from The Second Affiliated Hospital of Soochow University Pre-Doctoral Research Fund (No. SDFEYBS2218). H.Y. reports funding from the National Natural Science Foundation of China (No. 81702078); The Natural Science Foundation of Jiangsu Province (No. BK20170356); Gusu Talent Program (No. GSWS2023041); “National Tutor System” Training Program for Health Youth Key Talents in Suzhou (No. Qngg2023008); and Discipline Construction Support Project Advantageous Disciplines Support Project (No. XKTJ-XK202403-2). The other authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1610/dss
References
- 1.Ferlay J, Colombet M, Soerjomataram I, et al. Cancer statistics for the year 2020: An overview. Int J Cancer 2021. [Epub ahead of print]. doi: . 10.1002/ijc.33588 [DOI] [PubMed] [Google Scholar]
- 2.Salzman J, Gawad C, Wang PL, et al. Circular RNAs are the predominant transcript isoform from hundreds of human genes in diverse cell types. PLoS One 2012;7:e30733. 10.1371/journal.pone.0030733 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hansen TB, Jensen TI, Clausen BH, et al. Natural RNA circles function as efficient microRNA sponges. Nature 2013;495:384-8. 10.1038/nature11993 [DOI] [PubMed] [Google Scholar]
- 4.Zhang Y, Zhang XO, Chen T, et al. Circular intronic long noncoding RNAs. Mol Cell 2013;51:792-806. 10.1016/j.molcel.2013.08.017 [DOI] [PubMed] [Google Scholar]
- 5.Zhang Y, Jiang J, Zhang J, et al. CircDIDO1 inhibits gastric cancer progression by encoding a novel DIDO1-529aa protein and regulating PRDX2 protein stability. Mol Cancer 2021;20:101. 10.1186/s12943-021-01390-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Xu X, Zhang J, Tian Y, et al. CircRNA inhibits DNA damage repair by interacting with host gene. Mol Cancer 2020;19:128. 10.1186/s12943-020-01246-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zhang C, Tian C, Zhu R, et al. CircSATB1 Promotes Colorectal Cancer Liver Metastasis through Facilitating FKBP8 Degradation via RNF25-Mediated Ubiquitination. Adv Sci (Weinh) 2025;12:e2406962. 10.1002/advs.202406962 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhang XY, Li SS, Gu YR, et al. CircPIAS1 promotes hepatocellular carcinoma progression by inhibiting ferroptosis via the miR-455-3p/NUPR1/FTH1 axis. Mol Cancer 2024;23:113. 10.1186/s12943-024-02030-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Huang R, Zhang Y, Han B, et al. Circular RNA HIPK2 regulates astrocyte activation via cooperation of autophagy and ER stress by targeting MIR124-2HG. Autophagy 2017;13:1722-41. 10.1080/15548627.2017.1356975 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang G, Han B, Shen L, et al. Silencing of circular RNA HIPK2 in neural stem cells enhances functional recovery following ischaemic stroke. EBioMedicine 2020;52:102660. 10.1016/j.ebiom.2020.102660 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ji J, Li C, Wang J, et al. Hsa_circ_0001756 promotes ovarian cancer progression through regulating IGF2BP2-mediated RAB5A expression and the EGFR/MAPK signaling pathway. Cell Cycle 2022;21:685-96. 10.1080/15384101.2021.2010166 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang H, Wang B, Wang T, et al. Toll-like receptor 4 prompts human breast cancer cells invasiveness via lipopolysaccharide stimulation and is overexpressed in patients with lymph node metastasis. PLoS One 2014;9:e109980. 10.1371/journal.pone.0109980 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Harbeck N. Breast cancer is a systemic disease optimally treated by a multidisciplinary team. Nat Rev Dis Primers 2020;6:30. 10.1038/s41572-020-0167-z [DOI] [PubMed] [Google Scholar]
- 14.Conn VM, Chinnaiyan AM, Conn SJ. Circular RNA in cancer. Nat Rev Cancer 2024;24:597-613. 10.1038/s41568-024-00721-7 [DOI] [PubMed] [Google Scholar]
- 15.Hansen TB. Improved circRNA Identification by Combining Prediction Algorithms. Front Cell Dev Biol 2018;6:20. 10.3389/fcell.2018.00020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Guo Y, Huang Q, Heng Y, et al. Circular RNAs in cancer. MedComm (2020) 2025;6:e70079. [DOI] [PMC free article] [PubMed]
- 17.Zeng X, Tang J, Zhang Q, et al. CircHIPK2 Contributes Cell Growth in Intestinal Epithelial of Colitis and Colorectal Cancer through Promoting TAZ Translation. Adv Sci (Weinh) 2024;11:e2401588. 10.1002/advs.202401588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lun J, Zhang Y, Yu M, et al. Circular RNA circHIPK2 inhibits colon cancer cells through miR-373-3p/RGMA axis. Cancer Lett 2024;593:216957. 10.1016/j.canlet.2024.216957 [DOI] [PubMed] [Google Scholar]
- 19.Széliová D, Müller S, Zanghellini J. Costs of ribosomal RNA stabilization affect ribosome composition at maximum growth rate. Commun Biol 2024;7:196. 10.1038/s42003-024-05815-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.de Farias Morais HG, de Morais EF, Carlan LM, et al. Epithelial-mesenchymal transition modulates lower lip carcinogenesis and promotes cancer progression. Arch Oral Biol 2022;136:105387. 10.1016/j.archoralbio.2022.105387 [DOI] [PubMed] [Google Scholar]
- 21.Bo H, Zhang S, Gao L, et al. Upregulation of Wnt5a promotes epithelial-to-mesenchymal transition and metastasis of pancreatic cancer cells. BMC Cancer 2013;13:496. 10.1186/1471-2407-13-496 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cao Y, Wang F, Chen Y, et al. CircPITX1 Regulates Proliferation, Angiogenesis, Migration, Invasion, and Cell Cycle of Human Glioblastoma Cells by Targeting miR-584-5p/KPNB1 Axis. J Mol Neurosci 2021;71:1683-95. 10.1007/s12031-021-01820-y [DOI] [PubMed] [Google Scholar]
- 23.Zhang Y, Wang H, Li C, et al. CircSMYD4 regulates proliferation, migration and apoptosis of hepatocellular carcinoma cells by sponging miR-584-5p. Cancer Cell Int 2020;20:556. 10.1186/s12935-020-01648-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kim MJ, Min Y, Jeong SK, et al. USP15 negatively regulates lung cancer progression through the TRAF6-BECN1 signaling axis for autophagy induction. Cell Death Dis 2022;13:348. 10.1038/s41419-022-04808-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gong R, Wan X, Jiang S, et al. GPX4-AUTAC induces ferroptosis in breast cancer by promoting the selective autophagic degradation of GPX4 mediated by TRAF6-p62. Cell Death Differ 2025;32:2022-37. 10.1038/s41418-025-01528-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hua T, Yang M, Song H, et al. Huc-MSCs-derived exosomes attenuate inflammatory pain by regulating microglia pyroptosis and autophagy via the miR-146a-5p/TRAF6 axis. J Nanobiotechnology 2022;20:324. 10.1186/s12951-022-01522-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wang F, Wang X, Li J, et al. CircNOL10 suppresses breast cancer progression by sponging miR-767-5p to regulate SOCS2/JAK/STAT signaling. J Biomed Sci 2021;28:4. 10.1186/s12929-020-00697-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xu Y, Gao Z, Sun X, et al. The role of circular RNA during the urological cancer metastasis: exploring regulatory mechanisms and potential therapeutic targets. Cancer Metastasis Rev 2024;43:1055-74. 10.1007/s10555-024-10182-x [DOI] [PubMed] [Google Scholar]