Hsa_circRNA_000166 accelerates breast cancer progression via the regulation of the miR-326/ELK1 and miR-330-5p/ELK1 axes

Abstract

Purposes

To probe the expression, clinical significance, roles, and molecular mechanisms of circRNA_000166 in breast cancer (BC).

Methods

Clinical tissue samples were gathered from 84 BC patients who underwent surgery at the Affiliated Hospital of Chengde Medical College. Clinical data were obtained from medical records and postoperative follow-up. Expression levels of circRNA_000166, miR-326, miR-330-5p, and ELK1 mRNA in BC tissues and cells were measured by qRT-PCR, and ELK1 protein levels were assessed by WB. Pearson’s correlation analysis evaluated the interrelationships between these RNAs in clinical samples. Luciferase reporter assays verified the interactions between miR-326/miR-330-5p and circRNA_000166, as well as between miR-326/miR-330-5p and ELK1. Cell proliferation, migration, and apoptosis were examined using CCK-8, colony formation, transwell, and flow cytometry assays, respectively.

Results

CircRNA_000166 was highly expressed in BC tissues and inversely correlated with miR-326/miR-330-5p levels but positively with ELK1 mRNA levels. ELK1 mRNA also inversely associated with miR-326/miR-330-5p levels in BC tissues. Importantly, our findings demonstrated that circRNA_000166 targets miR-326 and miR-330-5p, while ELK1 is the target of miR-326 and miR-330-5p in BC cells. CircRNA_000166 levels positively correlated with tumour size, TNM stage, histological grade, and lymph node metastasis, and negatively associated with postoperative progression-free survival (PFS) and overall survival (OS) in BC patients. CircRNA_000166 was also highly expressed in BC cells, and knockdown of circRNA_000166 reduced proliferation and migration, and increased apoptosis via miR-326/ELK1 and miR-330-5p/ELK1 pathways in vitro.

Conclusion

CircRNA_000166 enhances BC progression through miR-326/ELK1 and miR-330-5p/ELK1 pathways and shows potential as a biomarker for BC diagnosis and treatment.

1. Introduction

Breast cancer (BC) is caused by the unchecked proliferation of mammary epithelial cells due to various carcinogenic factors. Typically, early-stage BC lacks discernible symptoms. Consequently, advanced-stage BC represents a major global health threat, with a mortality rate exceeding 90% due to distant metastasis and multiple organ involvement [1,2]. The incidence of female BC has increased considerably in recent years. While BC was the second most common cancer among women worldwide in 2018, by 2020, it had become the most common. BC was also the fourth leading cause of cancer-related deaths globally in both 2018 and 2020 [3]. Unfortunately, estimates suggest that by 2050, the global incidence of BC could reach approximately 3.2 million/year [4]. China has also experienced a gradual increase in the morbidity of BC. In 2015, the incidence of BC in China was 0.3 million, which was the fifth highest among all cancers. However, by 2020, the incidence of BC increased to 0.42 million, making it the fourth most common malignancy in China. Consequently, the number of BC-related deaths in Chinese men and women increased from 0.07 million in 2015 to 0.12 million in 2020 [3]. Notably, BC exhibited the highest incidence among all malignancies in women and was the fourth leading contributor to cancer-related deaths in this population [3]. The rising prevalence of BC, both in China and worldwide, highlights the need for better BC treatment strategies. Unfortunately, although the therapeutic strategies for BC have greatly improved, the overall survival rate remains dismal owing to recurrence and metastasis [5]. One key challenge lies in our incomplete understanding of BC aetiology. Therefore, the molecular mechanisms underlying BC pathogenesis must be delineated in order to improve BC diagnosis and treatment.

Circular RNAs (circRNAs) are a novel type of endogenous non-coding RNAs (ncRNAs) formed by the splicing of precursor mRNA (pre-mRNA). Unlike linear RNAs, circRNAs lack a 3′ tail and a 5′ cap, forming covalently closed continuous loops, which make them resistant to degradation by RNA exonucleases. As a result, circRNAs are more stable and evolutionarily conserved than linear messenger RNAs (mRNAs). CircRNAs are abundant in eukaryotic cells and exhibit strong cell- and tissue-specific expression [6]. Numerous studies indicate that the expression of circRNA can be dysregulated in various types of cancer [7]. Its functions are associated with tumorigenesis, proliferation, progression, recurrence, metastasis, and drug resistance [8]. At the mechanistic level, circRNAs predominantly reside in the cytoplasm [9], acting as miRNA "sponges" to sequester miRNAs and relieve their inhibitory effects on target genes, thereby regulating gene expression at the post-transcriptional level and playing biological functions. Recent evidence suggests that circRNAs regulate mRNA translation indirectly by competing with miRNAs [6,10,11]. Recently, specific circRNAs have been identified as targets for the therapy of various human cancers, such as BC [12]. Thus, circRNAs can serve as oncogenes or tumour suppressors during tumour development and potentially function as biomarkers for cancer diagnosis and prognosis, as well as targets for cancer therapy [8,13].

In the Gene Expression Omnibus (GEO) database, hsa_circRNA_000166 (also known as hsa_circ_0000512, hsa_circ_0000514, circ-0000520, circ-001846, and circRPPH1; located at chr14: 20811282–20811436; 154 bp; encoded by the RPPH1 gene) is the most prominently elevated circRNA in BC tissues compared to mammary gland tissues, according to circRNA microarray-seq datasets GSE101123 and GSE101124 [14–19]. Subsequently, this circRNA was selected for further exploration in several studies. The results confirmed that hsa_circRNA_000166 was upregulated in BC tissues and cells [14–19]. Hsa_circRNA_000166 is mainly localized in the cytoplasm of BC cells [14,15]. Inhibition of hsa_circRNA_000166 suppressed BC cell proliferation, colony formation, migration, invasion, and glycolysis. It also induced cell cycle arrest and apoptosis while restraining tumour growth. Conversely, elevation of hsa_circRNA_000166 expression facilitated BC cell proliferation, colony formation, migration, invasion, metastasis, and angiogenesis, and promoted epithelial-mesenchymal transition (EMT). In vivo experiments further confirmed its role in facilitating tumour growth [14–19]. High expression of hsa_circRNA_000166 is associated with poor prognosis (shorter overall survival (OS)) [18,19]. Thus, hsa_circRNA_000166 may accelerate BC malignant progression and provide a promising therapeutic target for BC treatment [14,15,17]. However, the relationship between hsa_circRNA_000166 expression and clinical features – including age, menopause status, molecular subtype, tumour diameter, TNM staging, lymph node status, recurrence, and metastasis – in patients with BC yields inconsistent conclusions [14–16,18,19]. Furthermore, studies on the role and molecular mechanisms of hsa_circRNA_000166 in BC progression are primarily based on small sample sizes. Therefore, we aim to further investigate hsa_circRNA_000166 in BC.

Regarding the mechanisms of hsa_circRNA_000166 in BC, a previous study verified that miR-326 was a target miRNA of hsa_circRNA_000166 in colorectal cancer (CRC) through a dual-luciferase reporter assay [7]. However, whether hsa_circRNA_000166 also functions as a sponge for miR-326 to participate in BC progression is unknown. Moreover, by using the ENCORI bioinformatics database, we predicted binding sites between hsa_circRNA_000166 and miR-330-5p, indicating that miR-330-5p may be a new target microRNA of hsa_circRNA_000166. Numerous studies have confirmed that miR-326 is expressed at low levels in human BC tissues and cells and acts as an anti-oncogene [20–22]. Some reports have also indicated that miR-330-5p levels are downregulated in human BC cells [20] and that this miRNA serves as a tumour inhibitor in BC [23]. Meanwhile, e-twenty six like-1 (ELK1) has been recognized as a common target gene of miR-326 [24] and miR-330-5p [25] in BC and cervical cancer, respectively. Several studies have reported that ELK1 is upregulated in human cancers, including BC, and exerts a carcinogenic effect [24,26–28]. Therefore, we hypothesized that hsa_circRNA_000166 facilitates the progression of BC by targeting the miR-326/ELK1 and miR-330-5p/ELK1 signalling axes. Hsa_circRNA_000166 may be a valuable therapeutic target for BC treatment (see Supplementary Figure 1 (SFigure 1)).

Hence, this study focused on investigating the expression, biological role, clinical significance, and potential molecular mechanisms of hsa_circRNA_000166 in human BC to obtain novel insights into BC pathogenesis and guide potential diagnostic and treatment strategies for this malignancy.

To fulfil our investigation, we initially utilized qRT-PCR to evaluate circRNA_000166 expression in BC tissues and cells, and performed survival analysis and collected clinical data to interpret the clinical implications of circRNA_000166. Following this, we used bioinformatics tools and conducted luciferase reporter assays to predict and confirm the miR-326/ELK1 and miR-330-5p/ELK1 pathways as the downstream target signals of circRNA_000166 in BC. Next, qRT-PCR was employed to determine the mRNA expression levels of miR-326, miR-330-5p, and ELK1, while WB was used to assess the protein expression level of ELK1 in BC tissues. The interrelationships between circRNA_000166, miR-326/miR-330-5p, and ELK1 in BC tissues were examined using Pearson’s correlation analysis. To identify and characterize circRNA_000166 in BC, we assessed its stability in BC cells through RNase R and Actinomycin D treatment assays, and determined its subcellular localization by performing qRT-PCR and fluorescence in situ hybridization (FISH) analyses in vitro. Finally, we evaluated the effects of knocking down circRNA_000166 on the expression of its parent gene RPPH1, miR-326, miR-330-5p, and ELK1, and assessed the impact of miR-326/miR-330-5p knockdown on ELK1 expression in vitro. To determine whether circRNA_000166 affects BC cell proliferation, migration, and apoptosis via the regulation of miR-326/ELK1 and miR-330-5p/ELK1 pathways, we conducted a series of in vitro experiments, including CCK-8 and colony formation assays, transwell assays, and flow cytometry assays (see SFigure 2).

2. Materials and methods

2.1. Patients and clinical specimens

Eighty-four pairs of freshly resected BC tissues and normal paracancerous tissues were collected from BC patients who underwent surgical procedures at the Affiliated Hospital of Chengde Medical College between March 2022 and May 2023. The inclusion criteria were as follows: 1) newly diagnosed BC patients confirmed by clinicopathological diagnosis by two pathologists; 2) no prior treatment before surgery. The exclusion criteria were as follows: 1) presence of other concurrent malignant tumours; 2) impaired heart, liver, or kidney function; 3) incomplete clinical case data. The patients’ clinical data were obtained from their electronic medical records. All patients received postoperative follow-up. Freshly resected tissues were promptly immersed in liquid nitrogen and subsequently stored at −80 °C until further experiments.

2.2. Cell culture

All human cell lines – including BC cells (MCF-7, MDA-MB-231, MDA-MB-468, BCAP-37, and BT20) and normal breast epithelial cells (MCF‑10A) – were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) in 2022. Each cell line was passaged for less than 6 months (post-resuscitation) in our laboratory before experimentation. All cells were authenticated using short tandem repeat (STR) profiling in 2022. The cells were cultivated in RPMI-1640 medium (Procell, China, Product No. PM150110B) containing 10% foetal bovine serum (FBS) (Procell, China, Product No. 164210-50), 100 μg/mL streptomycin, and 100 U/mL penicillin (Beijing Nuowei Biotechnology Co., LTD, Product No. 03-031-1B) and placed in an incubator (SCO6WE-2, SHELLAB, USA) containing air with 5% CO₂ at 37 °C. Subsequently, experiments were conducted using the two BC cell lines that exhibited the highest relative expression of hsa_circRNA_000166.

2.3. Cell transfection

Small interfering (si) RNA-negative control (NC) (si-NC), siRNA-1/2/3 against circRNA_000166 (si-circRNA_000166-1/2/3), inhibitor-NC, miR-326-inhibitor, miR-330-5p-inhibitor, and si-ELK1 were provided by GenePharma (Shanghai, China). The sequences of the three siRNAs targeting circRNA_000166 were as follows: CCATATTGAATCACAGTGCGT (siRNA 1), GCTTGGAACAGACTCACGGC (siRNA 2), and GUGAGUUCCCAGAGA (siRNA 3). The sequence of si-NC was GUGAGUUGGGUCCUA. Before transfection, 1 × 10⁶/mL of BT20 or MDA-MB-468 cells (logarithmic growth stage) were added to 6-well plates. When the cells reached 70% confluency, they were transfected with the aforementioned siRNAs and inhibitor constructs at 37 °C using the Lipofectamine 2000 Transfection Reagent (Invitrogen, Product No. 11668-019), following the manufacturer’s guidelines. After 7 h, the medium was replaced with fresh complete culture medium, and the transfection efficiency was examined using quantitative real-time polymerase chain reaction (qRT-PCR) assays. Subsequent experiments were conducted in the transfected cells following 24 h of stable expression.

2.4. qRT-PCR

qRT-PCR experiments were conducted to measure circRNA_000166, miR-326, miR-330-5p, and ELK1 mRNA expression in BC tissues and cells, as well as RPPH1 expression in BT20 and MDA-MB-468 cells. Briefly, total RNA was extracted from BC tissues and cells using an RNA Extraction Kit (Promega, USA, Product No. LS1040), according to the manufacturer’s instructions. To obtain a template for qRT-PCR, cDNA was synthesized using an RT-PCR kit (Solarbio, China, Product No. T2240) and the miRcute Plus miRNA First-Strand cDNA Kit (TIANGEN, Beijing, China, Product No. KR211-01), following the manufacturer’s directions. The Power SYBR Green kit (Invitrogen, USA, Product No. 4402954) was employed to perform qRT-PCR on an ABI 7900HT instrument (Applied Biosystems, USA). The qRT-PCR reaction included an initial pre-denaturation step at 95 °C for 5 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 60 °C for 40 s, and extension at 72 °C for 30 s. All PCR assays were performed in triplicate for each sample, and the 2^−ΔΔCt method was used for relative quantification. GAPDH and U6 were utilized as internal controls for circRNA_000166/ELK1 and miR-326/miR-330-5p, respectively. The primer sequences used for qRT-PCR are shown in Table 1.

Table 1.

Primer sequences used for qRT-PCR.

Gene	Forward primer (5′‑3′)	Reverse primer (5′‑3′)
GAPDH	CCACATCGCTCAGACACCAT	CCAGGCGCCCAATACG
circRNA_000166	CCATATTGAATCACAGTGCGT	ACAGCGCAGTAAGGTGCTCG
ELK1	TCCCTGCTTCCTACGCATACA	GCTGCCACTGGATGGAAACT
U6	CGCTTCGGCAGCACATATAC	TTCACGAATTTGCGTGTCAT
miR‑326	GGCGCCCAGAUAAUGCG	CGTGCAGGGTCCGAGGTC
miR-330-5p	TCTCTGGGCCTGTGTCTTAG	CAGTGCGTGTCGTGGAGT
RPPH1	GTCACTCCACTCCCATGTCC	CAGCCATTGAACTCACTTCG

qRT-PCR, quantitative real-time-polymerase chain reaction; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; circ, circular; ELK1, e-twenty six like-1; miR, microRNA.

2.5. Western blotting (WB)

Briefly, RIPA lysis buffer (Beyotime, Shanghai, China, Product No. P0013B) was employed to extract total protein from BC tissues and cells. Subsequently, protein concentration was evaluated using the BCA protein assay kit (Thermo Fisher Scientific, Inc., China, Product No. NCI3227CH). Proteins were then separated using 10% SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% skim milk powder for 1 h and then incubated overnight at 4 °C with rabbit-derived antibodies against human ELK1 (1:500, Abcam, UK, Product No. ab32106) and GAPDH (1:5000, Proteintech, China, Product No. HRP-60004) (internal reference control). The membranes were washed and incubated with an HRP-conjugated goat anti-rabbit immunoglobulin G (IgG) antibody (1:2000, Abcam, UK, Product No. ab6721). Protein bands were visualized using an ECL kit (Amersham, UK, Product No. BB-3501), and images were captured using a Bio-Rad Image Analysis System (Bio-Rad, Hercules, CA, USA). The intensity of protein bands was determined using the Quantity One v4.6.2 software. Each WB assay was performed in triplicate.

2.6. RNase R and actinomycin D treatment

To analyze the characteristics of circRNA_000166, RNA was isolated from BT20 and MDA-MB-468 cells using TRIzol reagent (Tiangen Biotech, Beijing, China, Product No. DP424). Then, 2 μg of RNA was incubated with or without (Mock, negative control) 3 U/μg RNase R (Geneseed Biotech Co., Ltd., Guangzhou, China, Product No. R0301) at 37 °C for 10 min. Meanwhile, BT20 and MDA-MB-468 cells were exposed to 2 mg/mL Actinomycin D (Sigma-Aldrich, St. Louis, MO, USA, Product No. 50-76-0) for 0, 4, and 8 h before RNA extraction. Subsequently, circRNA_000166 and RPPH1 mRNA expression in these cells was detected using qRT-PCR, with GAPDH serving as the internal reference.

2.7. Isolation of nuclear and cytoplasmic fractions

Cytoplasmic and nuclear fractions were isolated from BT20 and MDA-MB-468 cells using an RNA Subcellular Isolation Kit (Active Motif, USA, Product No. 25501) and TRIzol reagent (TIANGEN, Beijing, China, Product No. DP424), following the manufacturer’s protocol. qRT-PCR assays were employed to measure the expression of circRNA_000166 and RPPH1 mRNA.

2.8. Fluorescence in situ hybridization (FISH) assay

A specific fluorescence probe targeting circRNA_000166 was synthesized at GenePharma (Shanghai, China) and employed for FISH analysis. The location of circRNA_000166 in BT20 and MDA-MB-468 cells was determined using a FISH kit (BersinBio, Guangzhou, China, Product No. Bes1001(S)), and images were captured using a confocal microscope (Leica Microsystems, Germany).

2.9. Cell counting kit-8 (CCK-8) proliferation assay

First, BT20 and MDA-MB-468 cells were transfected with si-NC, si-circRNA_000166-1/2, si-circRNA_000166-1 + miR-326-inhibitor, or si-circRNA_000166-1 + miR-326-inhibitor + si-ELK1 and cultured in 96-well plates (5 × 10³ cells/well). The cells were incubated for 0, 24, 48, and 72 h before treatment with 20 μL of CCK-8 reagent (HY-K0301, MedChem Express, USA, Product No. HY-K0301). After 2 h of incubation with the CCK-8 reagent at 37 °C, the absorbance of each well at 450 nm was recorded using an EnVision microplate reader (PerkinElmer, USA).

2.10. Colony formation test

Forty-eight hours after transfection, BT20 and MDA-MB-468 cells were seeded into six-well plates (1 × 10³ cells/well) and incubated at 37 °C for 2 weeks. During this period, the culture medium was changed every 3 days. To assess the colony formation ability of the cells, the visible colonies were fixed with 4% paraformaldehyde (Sigma-Aldrich, USA, Product No. 158127) for 15 min and subsequently stained with 0.1% crystal violet (Yeasen, Shanghai, China, Product No. 60505ES25) for the same duration. Thereafter, the colonies were photographed and counted using a light microscope (Nikon ECLIPSE Ts2, Japan).

2.11. Transwell migration assay

To evaluate the migration capacity of BT20 and MDA-MB-468 cells transfected with si-NC, si-circRNA_000166-1/2, si-circRNA_000166-1 + miR-326-inhibitor, si-circRNA_000166-1 + miR-326-inhibitor + si-ELK1, si-circRNA_000166-1 + miR-330-5p-inhibitor, or si-circRNA_000166-1 + miR-330-5p-inhibitor + si-ELK1, Transwell migration assays were conducted. Briefly, 200 μL of cell suspension (1 × 10⁴ BT20 or MDA-MB-468 cells in serum‑free DMEM) was added to the upper Transwell chamber (Costar; Corning, Inc.), which was not coated with Matrigel. Meanwhile, culture medium supplemented with 10% FBS was added to the lower chamber. BC cells were added to the upper chamber and incubated at 37 °C for 24 h. Subsequently, the cells on the upper surface of the Transwell membrane were gently removed using a cotton swab, and the cells that had migrated to the lower chamber were stained with 0.5% crystal violet for 10 min at room temperature. The migratory cells were examined under a light microscope at ×100 magnification across five random microscopic fields.

2.12. Flow cytometry analysis

The cells were processed as described above for the Transwell migration assay, and an Annexin V-FITC apoptosis detection kit (Solarbio, China, Product No. CA1020) was used to examine apoptosis. Cultures of transfected cells (1 × 105 cells/well) were trypsinized and subsequently washed with PBS (Solarbio, Beijing, China, Product No. P1010). The cells were incubated with Annexin V-FITC (5 μL) for 10 min. Subsequently, 5 μL PI (Solarbio, Beijing, China, Product No. C0080-1) was added for staining under dark conditions for 5 min. Apoptotic cells were identified using CytoFLEX flow cytometry (Beckman Coulter, USA) and analyzed with apoptosis analysis software (v7.6; FlowJo LLC).

2.13. Target prediction

The potential interactions among circRNA, miRNA, and target mRNA were explored. ENCORI (https://rna.sysu.edu.cn/encori/agoClipRNA.php?source=circRNA) was employed to predict the binding sites between hsa_circRNA_000166 and miR-326/miR-330-5p. Meanwhile, TargetScan (https://www.targetscan.org/vert_80/) was used to predict the prospective binding sites of miR-326/miR-330-5p to the 3′ UTR of ELK1.

2.14. Plasmid and luciferase reporter assay

In brief, we generated psiCHECK2-ELK 3′ UTR wild-type (wt) and cirRNA_000166 wt plasmids, as well as psiCHECK2-ELK 3′ UTR scrambled (mut) and cirRNA_000166 mut plasmids. This was achieved by inserting the full-length ELK1 3′ UTR – which contains the binding sequence for miR-326/miR-330-5p (wt or mut) – downstream of the firefly luciferase gene in psiCHECK2. These plasmids were then co-transfected into BC cells along with the negative control (NC), miR-326 mimic/miR-330-5p mimic, si-circRNA_000166, and control Renilla luciferase expression plasmid (phRL-TK) (GenePharma, Shanghai, China) using Lipofectamine 2000 (Invitrogen, USA, Product No. 11668). Following a 24-hour transfection period, the luciferase and Renilla signals were quantified using the dual-luciferase reporter assay system (Promega, USA) according to the manufacturer’s guidelines. The experiment was repeated three times.

2.15. Statistical analysis

All data produced during this study were analyzed using SPSS 16.0 software (IBM Corp., USA), and the results were represented as the mean ± SD or as number and percentage. Comparisons between BC and paracancerous normal tissues were made using a paired Student’s t-test. Meanwhile, comparisons between two groups of BC cells were made using an unpaired Student’s t-test. Quantitative data were compared across multiple groups using one-way ANOVA (LSD and S-N-K for equal variances assumed, or Dunnett’s T3 for equal variances not assumed). Pearson’s correlation coefficient was utilized to analyze the association between circRNA_000166 and miR-326/miR-330-5p/ELK1 mRNA expression, and between ELK1 mRNA and miR-326/miR-330-5p expression in BC tissues. Rank-sum and chi-square tests were used to analyze the correlation between the expression of circRNA_000166 and the clinicopathological characteristics of BC. Survival analysis (OS, PFS) of circRNA_000166 was conducted using the log-rank test (Kaplan–Meier method). Differences with p < 0.05 were considered statistically significant.

3. Results

3.1. CircRNA_000166 was upregulated in BC tissues and cells and was associated with poor postoperative survival. The miR-326/ELK1 and miR-330-5p/ELK1 axes were the target signalling pathways of circRNA_000166 in BC

Compared with normal paracancerous tissues, BC tissues expressed higher levels of circRNA_000166 (p = 0.000) (p < 0.001) (Figure 1A). Moreover, we determined the expression of circRNA_000166 expression in several human BC cell lines using qRT-PCR. We discovered that circRNA_000166 levels were higher in all types of BC cells (MCF-7, p = 0.017; MDA-MB-435, p = 0.017; MDA-MB-231, p = 0.018; MDA-MB-468, p = 0.015; BCAP, p = 0.016; BT20, p = 0.009) than in normal human cells (MCF‑10A) (p < 0.05/p < 0.01) (Figure 1B). However, the BT20 and MDA-MB-468 cell lines showed the highest relative circRNA_000166 expression and were thus chosen for further experiments. According to survival analysis, the circRNA_000166 high expression group showed shorter postoperative PFS (p = 0.0164) and OS (p = 0.0166) than the circRNA_000166 low expression group (p < 0.05) (Figure 1C,D). Through bioinformatic analysis using the ENCORI and TargetScan tools, we successfully predicted specific interaction sites between miR-326/miR-330-5p and circRNA_000166, and between miR-326/miR-330-5p and ELK1 (Figure 1E, G, I, and K). Subsequently, we conducted a series of dual-luciferase reporter gene assays to validate these interactions. Notably, the luciferase activity of hsa_circRNA_000166 wild-type (wt) and ELK1 wt plasmids was markedly suppressed by the introduction of the miR-326 mimic (BT20: hsa_circRNA_000166 wt, p = 0.003; ELK1 wt, p = 0.001; MDA-MB-468: hsa_circRNA_000166 wt, p = 0.001; ELK1 wt, p = 0.000) and miR-330-5p mimic (BT20: hsa_circRNA_000166 wt, p = 0.002; ELK1 wt, p = 0.001; MDA-MB-468: hsa_circRNA_000166 wt, p = 0.001; ELK1 wt, p = 0.000) (p < 0.01/p < 0.001). Conversely, in the hsa_circRNA_000166 scrambled (mut) and ELK1 mut group, no discernible change in luciferase activity was detected (miR-326 mimic (BT20: hsa_circRNA_000166 mut, p = 0.142; ELK1 mut, p = 0.208; MDA-MB-468: hsa_circRNA_000166 mut, p = 0.376; ELK1 mut, p = 0.206) and miR-330-5p mimic (BT20: hsa_circRNA_000166 mut, p = 0.098; ELK1 mut, p = 0.055; MDA-MB-468: hsa_circRNA_000166 mut, p = 0.146; ELK1 mut, p = 0.058) (p > 0.05) (Figure 1F, H, J, and L). These experimental findings provided robust evidence indicating that miR-326 and miR-330-5p are target miRNAs of circRNA_000166 in BC cells, and that ELK1 is a common target gene of both miR-326 and miR-330-5p.

Figure 1.

Expression and clinical significance of circRNA_000166 in BC and its downstream targeting pathway prediction. (A) The level of circRNA_000166 was upregulated in human BC tissues compared to paracancerous tissues. *** vs paracancer tissues, p < 0.001. (B) The expression of circRNA_000166 was higher in human BC cells than in normal breast epithelial cells. * vs MCF‑10A, p < 0.05; ** vs MCF‑10A, p < 0.01. (C, D) BC patients with high circRNA_000166 expression have shorter postoperative PFS and OS than those with low circRNA_000166 expression. (E, G) Targeted binding sites for circRNA_000166 on miR-326 and miR-330-5p were predicted using an online bioinformatics tool, ENCORI. (I, K) Binding sites for miR-326 and miR-330-5p on ELK1 were predicted using an online bioinformatics tool, TargetScan. (F, H, J, L) Relationships between hsa_circRNA_000166 and miR-326/miR-330-5p expression, as well as between miR-326/miR-330-5p and ELK1 expression, were validated using a dual-luciferase reporter assay. ** vs miR-NC group, p < 0.01; *** vs miR-NC group, p < 0.001.

3.2. The expression levels of miR-326, miR-330-5p, and ELK1, as well as their correlations with circRNA_000166, in tissue specimens from BC patients

Compared with normal paracancerous tissues, BC tissues expressed lower levels of miR-326 (p = 0.000) and miR-330-5p (p = 0.000) (p < 0.001) (Figure 2A,B). Both mRNA (p = 0.000) and protein (p = 0.000) levels of ELK1 were upregulated in BC tissues (p < 0.001) (Figure 2C,D). A negative correlation between circRNA_000166 and miR-326/miR-330-5p expression, as well as between ELK1 mRNA and miR-326/miR-330-5p expression, was detected in BC tissues. Meanwhile, a positive correlation between circRNA_000166 levels and ELK1 mRNA levels was also observed in BC tissues (p < 0.0001) (Figure 2E–I).

Figure 2.

The expression levels of miR-326, miR-330-5p, and ELK1 mRNA in human BC tissues and their correlations with circRNA_000166. (A, B) MiR-326 and miR-330-5p levels were downregulated in human BC tissues compared to paracancerous tissues. (C, D) ELK1 mRNA and protein levels were higher in BC tissues than in paracancerous tissues. (E–I) There was an inverse association between circRNA_000166/ELK1 and miR-326/miR-330-5p expression, and a positive correlation between circRNA_000166 and ELK1 expression in BC tissues. ***, p < 0.001.

3.3. CircRNA_000166 expression was positively connected with the severity of BC

In order to investigate the relationship between circRNA_000166 expression and the clinicopathological features of BC, BC patients were divided into high and low circRNA_000166 expression groups based on the median circRNA_000166 levels. CircRNA_000166 expression was not correlated with age, menopausal status, hormone receptor status (estrogen receptor [ER] and progesterone receptor [PR]), or human epidermal growth factor receptor 2 (HER2) status (p > 0.05). However, it exhibited a positive correlation with tumour size, clinical TNM stage, histologic grade, and lymph node metastasis (p < 0.05/p < 0.01) (Table 2).

Table 2.

Relations between hsa_circRNA_000166 expression and clinicopathologic characteristics in BC sufferers.

		Relative expression of hsa_circRNA_000166
Pathological indicators	Number of patients	High expression	Low expression	P value
All cases	84	42	42
Age (years)
≥40	47	25	22	0.510
<40	37	17	20
Menopause
Yes	33	18	15	0.503
No	51	24	27
ER status
Positive	27	12	15	0.582
Negative	57	29	28
PR status
Positive	31	14	17	0.991
Negative	53	24	29
HER2 status
Positive	48	27	21	0.752
Negative	36	19	17
Pathological tumour size (d/cm)
<2 cm	36	12	24	0.008
≥2 cm	48	30	18
TNM stage
I/II	41	16	25	0.044
III/IV	43	26	17
Histological grade
I, II	39	15	24	0.044
III	45	27	18
Lymph node metastasis
Yes	41	26	15	0.016
No	43	16	27

BC, breast cancer; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal receptor 2; d, diameter; TNM, T, primary tumour; N, regional lymph nodes; M, distant metastasis.

3.4. Identification and characterization of circRNA_000166 in BC

To assess the stability of circRNA_000166 in BC cells, RNase R digestion and an Actinomycin D assay were performed. It is known that hsa_circRNA_000166, which is also called circRPPH1, is encoded by the linear gene RPPH1. As shown in Figure 3A, compared with the linear RPPH1 mRNA and GAPDH mRNA (internal reference controls), circRNA_000166 exhibited higher stability after treatment with RNase R exonuclease. Specifically, no significant variation in circRNA_000166 expression was observed after RNase R treatment in BT20 (p = 0.099) and MDA-MB-468 (p = 0.101) cells (p > 0.05), but a significant decrease in the expression of linear RPPH1 (p = 0.000 for all) and GAPDH (p = 0.000 for all) mRNAs was detected (p < 0.001). Similarly, circRNA_000166 showed higher stability after treatment with Actinomycin D compared to linear RPPH1 mRNA. A significantly higher expression of circRNA_000166 (p < 0.001/p < 0.01) was observed in BT20 (at 4 h, 8 h, 12 h, and 24 h, all p = 0.000) and MDA-MB-468 cells (at 4 h, 8 h, and 12 h, all p = 0.000; at 24 h, p = 0.001) (Figure 3B). Cytoplasmic and nuclear RNA analysis showed that circRNA_000166 was preferentially localized in the cytoplasm in BT20 and MDA-MB-468 cells. Specifically, the expression of circRNA_000166 was significantly higher in the cytoplasm than in the nucleus (p = 0.000 for all, p < 0.001), while the expression of RPPH1 was similar across the two compartments (BT20, p = 0.101; MDA-MB-468, p = 0.275; p > 0.05) (Figure 3C). FISH analysis further validated that circRNA_000166 is primarily localized in the cytoplasm, as nearly all of the green fluorescence-marked circRNA_000166 was detected in the cytoplasm of BT20 and MDA-MB-468 cells (Figure 3D).

Figure 3.

Identification and characterization of circRNA_000166 in BC. (A) qRT-PCR for determining the expression of circRNA_000166 and linear RPPH1 in BT20 and MDA-MB-468 cells after RNase R treatment. *** vs Mock, p < 0.001. (B) The qRT-PCR assay for detecting the stability of circRNA_000166 and linear RPPH1 after Actinomycin D treatment in BT20 and MDA-MB-468 cells. ** vs circRNA_000166, p < 0.01; *** vs circRNA_000166, p < 0.001. (C, D) The qRT-PCR and FISH analyses of the localization of circRNA_000166 and linear RPPH1 in BT20 and MDA-MB-468 cells. *** vs cytoplasm, p < 0.001. Blue color: DAPI (nuclei); Green color: circRNA_000166. Scale bar = 20 μm.

3.5. Perturbation of circRNA_000166 inhibited cell proliferation and migration, and increased apoptosis in vitro

In order to further understand the roles of circRNA_000166 in BC, we conducted a battery of in vitro experiments. First, we successfully knocked down circRNA_000166 in BT20 (si-circRNA_000166-1, p = 0.005; si-circRNA_000166-2, p = 0.006; si-circRNA_000166-3, p = 0.023) and MDA-MB-468 (si-circRNA_000166-1, p = 0.007; si-circRNA_000166-2, p = 0.013; si-circRNA_000166-3, p = 0.045) cells via the transfection of si-circRNA_000166-1/2/3 (p < 0.05/p < 0.01) (Figure 4A). Notably, the downregulation of circRNA_000166 did not affect the expression of its parent gene RPPH1 (p > 0.05) (BT20: si-circRNA_000166-1, p = 0.087; si-circRNA_000166-2, p = 0.165; si-circRNA_000166-3, p = 0.437; MDA-MB-468: si-circRNA_000166-1, p = 0.066; si-circRNA_000166-2, p = 0.234; circRNA_000166-3, p = 0.817) (Figure 4B) but significantly reduced ELK1 protein expression (BT20: si-circRNA_000166-1, p = 0.001; si-circRNA_000166-2, p = 0.005; si-circRNA_000166-3, p = 0.02; MDA-MB-468: si-circRNA_000166-1, p = 0.001; si-circRNA_000166-2, p = 0.001; circRNA_000166-3, p = 0.004) in these cells (Figure 4C). The siRNA-1 and siRNA-2 enabled the greatest inhibition of circRNA_000166 and ELK1 expression and were thus used for subsequent tests. In vitro, functional experiments revealed that compared with BT20 (24 h: siRNA-1, p = 0.015; 48 h: siRNA-1, p = 0.007; siRNA-2, p = 0.030; 72 h: siRNA-1, p = 0.002; siRNA-2, p = 0.004) and MDA-MB-468 (24 h: siRNA-1, p = 0.023; 48 h: siRNA-1, p = 0.007; siRNA-2, p = 0.023; 72 h: siRNA-1, p = 0.002; siRNA-2, p = 0.003) cells transfected with siRNA-NC, those transfected with siRNA-1/RNA-2 showed weakened cell proliferation, with a lower 450-nm OD value (p < 0.05/p < 0.01) (Figure 4D). Moreover, these cells also formed fewer colonies (BT20: siRNA-1, p = 0.000; siRNA-2, p = 0.002; MDA-MB-468: siRNA-1, p = 0.000; siRNA-2, p = 0.006) (p < 0.01/p < 0.001) (Figure 4E). Additionally, they had a lower migration capacity, with fewer migratory cells (BT20: siRNA-1, p = 0.002; siRNA-2, p = 0.007; MDA-MB-468: siRNA-1, p = 0.002; siRNA-2, p = 0.004) (p < 0.01) (Figure 4F). Finally, BC cells transfected with siRNA-1/RNA-2 also had a higher percentage of apoptotic cells (BT20: siRNA-1, p = 0.000; siRNA-2, p = 0.000; MDA-MB-468: siRNA-1, p = 0.000; siRNA-2, p = 0.000) (p < 0.001) compared with those transfected with si-NC (Figure 4G). Thus, given its significant effect on circRNA_000166 expression and the behavioural characteristics of BT20 and MDA-MB-468 cells, si-circRNA_000166-1 was selected for subsequent mechanistic experiments.

Figure 4.

CircRNA_000166 was overexpressed in human BC cells, and its downregulation suppressed cell proliferation and migration and facilitated apoptosis in vitro. (A) CircRNA_000166 was successfully downregulated in human BC cells (BT20 and MDA-MB-468) via transfection with si-circRNA_000166. (B) The transfection of si-circRNA_000166 did not affect the expression of the parent gene, RPPH1, in BT20 and MDA-MB-468 cells. (C) CircRNA_000166 downregulation dramatically inhibited the ELK1 protein expression. (D-G) CircRNA_000166 downregulation markedly inhibited BC cell proliferation and migration, while promoting apoptosis. Scale bar = 50 μm. Each in vitro test was repeated three times. * vs si-NC, p < 0.05; ** vs si-NC, p < 0.01; *** vs si-NC, p < 0.001.

3.6. Suppression of circRNA_000166 inhibited cell proliferation and migration, and enhanced apoptosis through modulation of the miR-326/ELK1 signalling pathway

The downregulation of circRNA_000166 induced an increase in miR-326 expression (BT20: p = 0.004; MDA-MB-468: p = 0.009) (p < 0.01) (Figure 5A) and a decrease in ELK1 mRNA expression (BT20: p = 0.045; MDA-MB-468: p = 0.029) (p < 0.05) (Figure 5B) in BT20 and MDA-MB-468 cells. Additionally, we successfully perturbed miR-326 expression in BC cells by transfecting a miR-326-inhibitor (BT20: p = 0.007; MDA-MB-468: p = 0.004) (p < 0.01) (Figure 5C). qRT-RCR and WB assays revealed that miR-326 downregulation enhanced ELK1 expression at both the mRNA level (BT20: p = 0.047; MDA-MB-468: p = 0.035) (p < 0.05) (Figure 5D) and the protein level (BT20: p = 0.0052; MDA-MB-468: p = 0.0006) (p < 0.05/p < 0.01) (Figure 5E) in BC cells. Moreover, as shown in Figure 5F–H, the si-circRNA_000166-1 group showed a lower 450-nm OD value (BT20: 2d, p = 0.041; 3d, p = 0.041; MDA-MB-468: 2d, p = 0.017; 3d, p = 0.016) (p < 0.05) (Figure 5F) and fewer migratory cells (BT20: p = 0.000; MDA-MB-468: p = 0.019) (p < 0.05/0.001) (Figure 5G) than the si-NC group. Furthermore, this group also showed a higher apoptotic cell rate (BT20: p = 0.001; MDA-MB-468: p = 0.003) (p < 0.01) (Figure 5H). Meanwhile, antipodal changes were detected in the si-circRNA_000166-1 + miR-326-inhibitor group when compared to the si-circRNA_000166-1 group (Figure 5F: BT20: 2d, p = 0.009; 3d, p = 0.028; MDA-MB-468: 2d, p = 0.011; 3d, p = 0.013; Figure 5G: BT20: p = 0.012; MDA-MB-468: p = 0.014; Figure 5H: BT20: p = 0.005; MDA-MB-468: p = 0.000) (p < 0.05/p < 0.01/p < 0.001). Finally, the 450-nm OD value (BT20: 2d, p = 0.018; 3d, p = 0.043; MDA-MB-468: 2d, p = 0.047; 3d, p = 0.038) (p < 0.05) and migratory cell number (BT20: p = 0.003; MDA-MB-468: p = 0.005) (p < 0.01) were lower in the si-circRNA_000166-1 + miR-326-inhibitor + si-ELK1 group than in the si-circRNA_000166-1 + miR-326-inhibitor group, and the cell apoptosis rate (BT20: p = 0.000; MDA-MB-468: p = 0.020) (p < 0.05/p < 0.001) was higher.

Figure 5.

Knockdown of circRNA_000166 attenuated cell proliferation and migration, while facilitating apoptosis via modulation of the miR-326/ELK1 signalling pathway. (A, B) Suppression of circRNA_000166 significantly upregulated miR-326 expression and downregulated ELK1 expression in BT20 and MDA-MB-468 cells. * vs si-NC, p < 0.05; ** vs si-NC, p < 0.01. (C) MiR-326 was successfully downregulated in BT20 and MDA-MB-468 cells following transfection with miR-326 inhibitor. (D, E) Downregulation of miR-326 increased ELK1 mRNA and protein expression in BT20 and MDA-MB-468 cells. * vs inhibitor-NC, p < 0.05; ** vs inhibitor-NC, p < 0.01. (F-H) Inhibition of circRNA_000166 expression decreased BC cell proliferation and migration, and facilitated apoptosis via modulation of the miR-326/ELK1 signalling pathway. Scale bar = 50 μm. All in vitro experiments were repeated thrice. * vs si-NC, p < 0.05; ** vs si-NC, p < 0.01; *** vs si-NC, p < 0.001. # vs si-circRNA_000166-1, p < 0.05; ## vs si-circRNA_000166-1, p < 0.01; ### vs si-circRNA_000166-1, p < 0.001. $ vs si-circRNA_000166-1 + miR-326-inhibitor, p < 0.05; $$ vs si-circRNA_000166-1 + miR-326-inhibitor, p < 0.01. $$$ vs si-circRNA_000166-1 + miR-326-inhibitor, p < 0.001.

3.7. Suppression of circRNA_000166 attenuated cell proliferation and migration, and enhanced apoptosis via modulation of the miR-330-5p/ELK1 signalling pathway

To verify whether another pathway, the miR-330-5p/ELK1 pathway, could mediate the effects of circRNA_000166 on BC progression, experiments similar to those described above were performed; however, the CCK-8 proliferation assay (Figure 5F) was replaced by a colony formation assay. The results revealed that circRNA_000166 downregulation could induce an increase in miR-330-5p levels (p < 0.05) in BT20 (p = 0.011) and MDA-MB-468 (p = 0.022) cells (Figure 6A). MiR-330-5p was successfully knocked down in BT20 (p = 0.004) and MDA-MB-468 (p = 0.001) cells following transfection with a miR-330-5p-inhibitor (p < 0.01) (Figure 6B); as a result, both ELK1 mRNA (BT20: p = 0.040; MDA-MB-468: p = 0.041) (p < 0.05) (Figure 6C) and protein expression (BT20: p = 0.0085; MDA-MB-468: p = 0.0006) (p < 0.01) (Figure 6D) were found to be elevated in these miR-330-5p-knockdown cells. As shown in Figure 6E–G, knockdown of circRNA_000166 significantly inhibited BC cell clone formation (BT20: p = 0.022; MDA-MB-468: p = 0.003) and migration (BT20: p = 0.002; MDA-MB-468: p = 0.016), and promoted apoptosis (BT20: p = 0.000; MDA-MB-468: p = 0.000) compared to the si-NC group (p < 0.05/p < 0.01/p < 0.001). Moreover, the simultaneous transfection of si-circRNA_000166-1 and the miR-330-5p inhibitor partly attenuated the si-circRNA_000166-1-induced effects in BT20 and MDA-MB-468 cells, including the decrease in colony formation (BT20: p = 0.013; MDA-MB-468: p = 0.013), the reduction in migratory cells (BT20: p = 0.004; MDA-MB-468: p = 0.023), and the increase in apoptosis (BT20: p = 0.003; MDA-MB-468: p = 0.001) (p < 0.05/p < 0.01). Similarly, the co-transfection of si-circRNA_000166-1, miR-330-5p-inhibitor, and si-ELK1 partly reversed the changes induced by the co-transfection of si-circRNA_000166-1 and miR-330-5p-inhibitor in BT20 (Figure 6E: p = 0.017; Figure 6F: p = 0.011; Figure 6G: p = 0.008) and MDA-MB-468 cells (Figure 6E: p = 0.027; Figure 6F: p = 0.028; Figure 6G: p = 0.008) (p < 0.05/p < 0.01).

Figure 6.

Inhibition of circRNA_000166 attenuated BC cell proliferation and migration, and enhanced apoptosis by modulating the miR-330-5p/ELK1 signalling pathway. (A) Knockdown of circRNA_000166 increased miR-330-5p expression in human BC cells (BT20, MDA-MB-468). * vs si-NC, p < 0.05. (B) MiR-330-5p expression was successfully attenuated in BT20 and MDA-MB-468 cells after transfection with miR-330-5p inhibitor. (C, D) MiR-330-5p inhibition upregulated ELK1 expression at both the mRNA and protein levels in BT20 and MDA-MB-468 cells. * vs inhibitor-NC, p < 0.05; ** vs inhibitor-NC, p < 0.01. (E-G) circRNA_000166 downregulation attenuated BC cell proliferation and migration and accelerated apoptosis via the miR-330-5p/ELK1 signalling pathway. Scale bar = 50 μm. All in vitro experiments were performed in triplicate. * vs si-NC, p < 0.05; ** vs si-NC, p < 0.01; *** vs si-NC, p < 0.001. # vs si-circRNA_000166-1, p < 0.05; ## vs si-circRNA_000166-1, p < 0.01; $ vs si-circRNA_000166-1 + miR-330-5p-inhibitor, p < 0.05; $$ vs si-circRNA_000166-1 + miR-330-5p inhibitor, p < 0.01.

4. Discussion

BC is the most common malignancy among women globally, and its incidence continues to rise [3,29]. Although BC mortality has shown a decline in developed nations over the past few decades owing to substantial improvements in treatment strategies, it is still a major cause of cancer-related deaths because of its recurrence and metastasis [5,29]. Hence, understanding the molecular mechanisms underlying BC progression is crucial. In this study, we elucidated the oncogenic role of circRNA_000166 in BC, demonstrating that it mediates its effects through the miR-326/ELK1 and miR-330-5p/ELK1 signalling pathways. Therefore, our findings suggest that alterations in the circRNA_000166/miR-326/ELK1 and circRNA_000166/miR-330-5p/ELK1 signalling pathways may contribute to BC pathogenesis, and targeting these pathways could have diagnostic or therapeutic implications for BC.

CircRNAs are endogenous noncoding RNAs (ncRNAs). They are widely and stably expressed in multiple eukaryotes due to their covalently closed-loop structure, which lacks a 5′ end or 3′ poly-A tail. This structure distinguishes them from classical linear mRNAs and ncRNAs, protecting them from ribonuclease degradation [30–33]. Our findings showed that circRNA_000166 is more stable than linear RPPH1 mRNA. It is primarily localized in the cytoplasm of BT20 and MDA-MB-468 cells. Previous studies have also demonstrated that circRPPH1 (an alias for circRNA_000166) is more stable than its linear RNA counterpart and predominantly localizes in the cytoplasm of BC cell lines MCF-7 and MDA-MB-231 [14,15]. Accumulating evidence has shown that several circRNAs are abnormally expressed in cancers, including BC, and are involved in tumour progression. Thus, these circRNAs can act as biomarkers and key regulators of cancer development and progression [34–37]. For instance, hsa_circ_0014717 was discovered to be downregulated in gastric cancer (GC) tissues and in gastric juice from GC patients. It is considered a diagnostic biomarker for GC because its expression inversely correlates with tumour stage, distant metastasis, tissue carcinoembryonic antigen levels, and carbohydrate antigen 19-9 levels [38]. Meanwhile, circRNA-ciRS-7 showed increased expression in both pancreatic ductal adenocarcinoma (PDAC) tissues and cells. Notably, a study found that circRNA-ciRS-7 acts as an oncogene by promoting cell proliferation, invasion, and metastasis through miR-7/EGFR/STAT3 pathway inhibition. Additionally, its presence is correlated with lymph node metastasis and venous invasion in PDAC patients [39]. Similarly, the upregulation of circRNA_0006528 was observed in BC tissues. This circRNA was found to have an oncogenic effect on BC progression and was strongly linked to the TNM stage and poor prognosis in BC patients [40]. In the present study, we discovered that circRNA_000166 is highly expressed in human BC tissues and cells. Its expression is positively related to tumour size, TNM stage, histologic grade, and lymph node metastasis, while negatively associated with OS and PFS of BC patients. Moreover, our findings indicate that the knockdown of circRNA_000166 reduces the proliferation and colony formation of BC cells (BT20 and MDA-MB-468), inhibits cell migration, and induces tumour cell apoptosis. CircRNA microarray-seq datasets GSE101123 and GSE101124 from the GEO database showed that hsa_circRNA_000166 is the most significantly elevated circRNA in human BC tissues [14–19]. To date, a total of six studies have investigated the expression, function, mechanism, and clinical significance of hsa_circRNA_000166 in BC. These studies have consistently found upregulated hsa_circRNA_000166 expression in both BC tissues and cells [14–19]. High expression of circ_0000520 (an alias of circRNA_000166) was relative to a lower five-year survival rate in BC patients [18] and shorter survival time in patients with triple-negative breast cancer (TNBC) [19]. These findings strongly support our results. But, it’s worth noting that there are different findings about the correlations between circRNA_000166 expression and the clinical characteristics of BC patients. Zhou YH et al. and Zhou Y et al.’s study reported that circRPPH1/circ_0000520 expression was not associated with age and menopause status, but was associated with later TNM stage, more lymph node metastasis, and larger tumour diameter in TNBC patients [15,19], which are consistent with our results. However, there are inconsistent conclusions. For example, Huang YX et al.’s research found that circRPPH1 level was associated with age and molecular subtypes, but not with TNM stage, lymph node status, recurrence, and metastasis in BC patients [14]. Yang L et al. and Zang HL et al.’s study discovered that circRPPH1/circ_0000520 expression was not related to age, menopausal status, and tumour size, but was related to lymph node metastasis and TNM stage in BC patients [16,18]. Previous researches have validated that knockdown of circRPPH1 expression weakened the proliferation, migration, invasion, and EMT abilities of BC cells (MDA-MB-231 and MCF-7), while the upregulation of circRPPH1 promoted the proliferation, colony-forming, migration, invasion, angiogenesis, and EMT abilities of BC cells in vitro, and facilitated tumour growth in vivo experiments [14,15]. Moreover, silencing circRPPH1 represses the metastasis and glycolysis of BC cells (MCF-7 and MDA-MB-231) [16], and increases apoptosis [17–19] and cell cycle arrest [18,19] in BC cells (MCF-7, BT549, and MDA-MB-231). The increased expression and oncogenic role of circRNA_000166 identified in these studies are consistent with our findings in BC. Moreover, we also confirmed that circRNA_000166 expression is positively correlated with the severity of BC, indicating its clinical significance and feasibility as a diagnostic biomarker and therapeutic target for BC.

CircRNAs typically contribute to disease pathogenesis by targeting miRNAs via their 3′-UTRs or interacting with other molecules to regulate gene expression, either during or after transcription [41,42]. Meanwhile, miRNAs post-transcriptionally inhibit target genes by binding to the 3′-UTRs of the corresponding mRNAs [43]. Evidence from bioinformatic analysis and luciferase reporter assays shows that miR-326 is the target miRNA for circRNA_000166 [7] in colorectal cancer and that ELK1 is a common target of miR-326 and miR-330-5p in BC [24] and cervical cancer [25], respectively. In this study, we used bioinformatics tools such as ENCORI and TargetScan to predict the presence of binding sites between miR-326/miR-330-5p and circRNA_000166, as well as between miR-326/miR-330-5p and ELK1. Furthermore, we experimentally validated these predictions through luciferase reporter assays in BC cells, confirming that circRNA_000166 directly interacts with miR-326 and miR-330-5p, and that ELK1 serves as a common target gene for both miR-326 and miR-330-5p. Moreover, we employed indirect methods to identify the relationships between these molecules in BC and to understand the molecular mechanisms through which circRNA_000166 promotes BC progression. We found inverse correlations between the expression of circRNA_000166 and miR-326/miR-330-5p, as well as between the expression of ELK1 and miR-326/miR-330-5p, in clinical BC tissue specimens. Meanwhile, circRNA_000166 expression showed a positive correlation with ELK1 expression in tissues. CircRNA_000166 silencing could increase miR-326/miR-330-5p expression and decrease ELK1 at both mRNA and protein levels in vitro. Notably, knockdown of circRNA_000166 did not affect the expression of its parent gene, RPPH1. Moreover, the inhibition of miR-326/miR-330-5p increased ELK1 mRNA and protein levels in vitro. Interestingly, the knockdown of both circRNA_000166 and miR-326/miR-330-5p in BC cells partly attenuated the decrease in cell proliferation and migration and the increase in apoptosis observed after circRNA_000166 downregulation alone. Notably, the additional ELK1 inhibition also partly weakened the effects of combined circRNA_000166 and miR-326/miR-330-5p downregulation on cell proliferation, migration, and apoptosis in vitro. These results demonstrate that miR-326/miR-330-5p could potentially serve as targets of circRNA_000166 in BC, while ELK1 may be targeted by miR-326/miR-330-5p. CircRNA_000166 may expedite the progression of BC by specifically targeting the miR-326/ELK1 and miR-330-5p/ELK1 pathways. Several studies have proven that miR-326 is involved in BC and exhibits abnormally low levels in BC tissues and cells. The upregulation of miR-326 can suppress the malignant behaviours of BC cells, including declining abilities of cell to proliferate, form colonies, migrate, invade, and undergo EMT, as well as enhanced ability to induce cell cycle arrest and apoptosis in vitro, and impaired capacity for tumour growth in vivo [20–22,44]. In addition, miR-330-5p levels in human BC cells are often low, as demonstrated by several reports [20,45]. Meanwhile, research has shown that higher levels of miR-330-5p suppress proliferation, migration, invasion, EMT, and angiogenesis of BC cells, while promoting apoptosis and enhancing the cells’ responsiveness to treatment [23,45]. Several studies focusing on ELK1 have suggested that the expression of this gene is elevated in BC tissues [24], and its overexpression can promote the proliferation, colony formation, invasion, and cell cycle progression of BC cells [24,46,47], as well as alleviate apoptosis, and is associated with worse recurrence-free survival in TNBC patients [48]. These previous findings validate the results of our study, which demonstrated significant downregulation of miR-326 and miR-330-5p, alongside upregulation of ELK1 in BC tissues. Moreover, our in vitro experiments revealed that inhibiting miR-326/miR-330-5p enhances BC cell proliferation and migration while reducing apoptosis, whereas downregulating ELK1 has the opposite effect. Importantly, a prior study indicated that hsa_circRNA_000166, as the top up-regulated circRNA in CRC, plays a stimulative role on CRC cell proliferation and suppressive effect on apoptosis by targeting the miR-326/LASP1 axis [7]. In BC, targeting the ELK1, miRNA-326 significantly inhibited the proliferation, colony formation, cell cycle progression, angiogenesis, and invasion of MCF-7 cells [24]. Our findings provide the first evidence that hsa_circRNA_000166 acts as an oncogene in BC by targeting the miR-326/ELK1 axis, thereby promoting BC cell proliferation and migration and suppressing apoptosis. Furthermore, by targeting ELK1, miR-330-5p enhanced apoptosis and the expression of pro-apoptotic protein cleaved caspase-3 in renal tubular epithelial cells in ischemic acute kidney injury (AKI) [49], restrained cervical cancer cell proliferation and migration in vitro, and suppressed tumour growth in vivo [25]. In addition, it weakened human kidney-2 (HK2) cell viability, promoted cell apoptosis, increased the expression of pro-apoptotic protein cleaved caspase-3, and decreased the expression of anti-apoptotic protein Bcl-2 in sepsis‑induced AKI [50]. Our results provide initial evidence that hsa_circRNA_000166 functions as an oncogene in BC by targeting the miR-330-5p/ELK1 axis, resulting in enhanced BC cell proliferation and migration and inhibited apoptosis.

It is imperative to discuss the potential of hsa_circRNA_000166 as a druggable target and its significance for future therapeutic interventions. With the advancement of RNA sequencing (RNA-seq) technology and bioinformatics in the twenty first century, circRNA has been discovered to be widely present in eukaryotic cells and involved in the pathogenesis and progression of various diseases, including BC. Among these advancements, high-throughput sequencing and circRNA microarray technologies have led to the identification of an increasing number of circRNAs, which hold potential value in the diagnosis, treatment, and prognosis of BC. Compared to adjacent normal tissues, dysregulation of circRNA expression in BC tissues can play either oncogenic or tumour-suppressive roles in its occurrence and development. Mechanistically, circRNAs act as miRNA sponges or competitive endogenous RNAs (ceRNAs), regulating the proliferation, metastasis, and invasion of cancer, including BC. For instance, circKIF4A, which is associated with poor prognosis in TNBC, exerts its regulatory function in TNBC by sponging miR-375, and modulating the expression of KIF4A [51]. Previous studies identified hsa_circRNA_000166 as the most significantly upregulated circRNA in human BC tissues, based on circRNA microarray-seq datasets (GSE101123 and GSE101124) from the GEO database. Subsequent in vivo and in vitro experiments validated the high expression of hsa_circRNA_000166 in BC tissues and cells, indicating its involvement in regulating BC cell proliferation, migration, invasion, metastasis, cell cycle, apoptosis, and tumour growth through the circRNA-miRNA-mRNA network. Elevated expression of hsa_circRNA_000166 is associated with poor prognosis in BC patients, and is considered an effective diagnostic biomarker and therapeutic target in BC [14–19]. Our study further demonstrated the high expression of hsa_circRNA_000166 in BC tissues and cells, promoting BC proliferation and migration, inhibiting apoptosis through targeting the miR-326/ELK1 and miR-330-5p/ELK1 axes, and being associated with malignant progression and shorter postoperative survival in BC patients. This highlights the potential of hsa_circRNA_000166 as a diagnostic and therapeutic target for BC. Besides surgical treatment, chemotherapy, targeted therapy, and immunotherapy have made unprecedented advancements. Currently, tumour drug resistance is the primary factor limiting the efficacy of tumour therapy and postoperative recovery. The issue of drug resistance caused by the long-term application of chemotherapeutic agents remains a major obstacle in current tumour therapy [6]. Studies have shown that circRNAs play a key role in carcinogenesis, metastasis, and treatment resistance [51]. Meanwhile, tumour-derived or tumour-associated exosomes are crucial in regulating tumour drug resistance [6]. Exosomes, released by various cells including tumour cells, can stably exist in various body fluids. The stability and tissue specificity of circRNA in exosomes or body fluids make circRNA a reliable tumour biomarker for diagnosis or prediction, as well as a potential therapeutic target for various cancers, including BC. Additionally, the ncRNAs in tumour cell-released exosomes can promote tumour angiogenesis, immune escape, metastatic phenotypes, and mediate tumour cell drug resistance [6,51]. For instance, hsa_circ_0000520, which is downregulated in gastric cancer (GC) tissues and cells, is involved in Herceptin resistance in GC cells. The overexpression of hsa_circ_0000520 may reverse the resistance of GC cells to Herceptin treatment by repressing the PI3K-Akt signalling pathway [52]. Circ-MMP11 was highly expressed in lapatinib-resistant (LR) BC tissues and cells, and knockdown of circ-MMP11 increased lapatinib sensitivity by suppressing cell viability, colony formation, migration, and invasion, and boosting apoptosis in LR BC cells through downregulating ANLN expression by sponging miR-153-3p [53]. CircBACH1 was overexpressed in paclitaxel (PTX)-treated BC-derived exosomes and BC tissue. This overexpression promoted PTX resistance by increasing cell viability, stemness, migration, and angiogenesis in BC cells, achieving this through upregulating G3BP2 expression via sponging miR-217 [54]. Hu et al. reported that circ_UBE2D2 was upregulated in exosomes released from tamoxifen (TAM)-resistant BC (MCF-7) cells, and circ_UBE2D2 contributed to TAM resistance of the BC cells by sponging miR-200a-3p [55]. Moreover, miR-326, miR-330-5p, and ELK1 are involved in drug resistance in various cancers, including BC. For example, overexpression of miR-326 can overcome chemoresistance driven by fibronectin (FN1) (a central factor in chemoresistance) by inhibiting the FAK/Src pathway and enhancing in vivo chemotherapeutic efficacy, making it an attractive therapeutic approach to augment the response to TNBC chemotherapy [56]. Overexpression of miR-326 sensitized BC cells to VP-16 and doxorubicin by down-regulating MRP-1 (multidrug resistance-associated protein) expression, suggesting the involvement of miR-326 in MRP-1-mediated multidrug resistance and suggesting that miR-326 may be a potent agent for preventing and reversing multidrug resistance in tumour cells [57]. Overexpression of miR-330-5p, a known suppressor of tumorigenesis and chemoresistance, increases the sensitivity of non-small cell lung cancer (NSCLC) [58], epithelial ovarian cancer (EOC) [59], and cervical cancer (CC) [60] cells to cisplatin (DDP) both in vivo and in vitro. Up-regulation of ELK1 contributes to increased acquired resistance to doxorubicin (DXR) chemotherapy in osteosarcoma [61] and to gefitinib resistance in NSCLC [62]. Thus, hsa_circRNA_000166 may regulate chemotherapy resistance in BC by linking the miR-326/ELK1 and miR-330-5p/ELK1 axes. Notably, relatively novel treatment modalities have been developed in recent years. Genetic mutations that result in abnormal gene expression often increase the risk of various types of cancer, especially familial hereditary malignancies such as BC. Mutations in genes such as TP53, BRCA1, ERBB2 (HER2), and ESR1 are particularly notable. The development of targeted drugs against these specific gene mutations has driven advancements in targeted therapies. Concretely, the rising global prevalence of breast tumours with TP53 mutation has led to the development of various therapies targeting p53/TP53. A multitude of molecular drugs such as nutlins, MI series, RO5693, PRIMA-1, RITA, etc. have been created [63]. Targeting BRCA1 mutated breast tumours, PARP inhibitors (olaparib, veliparib, and niraparib), PPAR agonists (PPARγ agonist (efatutazone) and n-3 PUFA (DHA)), and aromatase inhibitors have brought encouraging results [64]. The introduction of HER2-targeted therapies for patients with ERBB2 (HER2) amplification or overexpression in BC has significantly improved oncologic outcomes. Over the past 20 years, the FDA has approved five HER2-targeted therapies: trastuzumab (for metastatic and adjuvant use), pertuzumab (for metastatic and adjuvant use), lapatinib (for metastatic use), ado-trastuzumab emtansine (for metastatic use), and neratinib (for adjuvant use). In addition, the number of new HER2-targeting therapies is rapidly increasing, including small molecule inhibitors, antibody-drug conjugates, and bispecific antibodies. These approved and emerging HER2-targeted therapies may show benefits for the growing number of BC patients [65]. Moreover, for patients with metastatic BC and ESR1 mutations, the commonly used targeted drugs are aromatase inhibitors, fulvestrant, and CDK4/6 inhibitors [66]. Interestingly, exon skipping is a promising therapeutic strategy that removes affected exons to produce truncated proteins with partial or full function, alleviating the phenotypes of certain diseases. Antisense circRNAs (AS-circRNAs) can efficiently mediate exon skipping in the minigene and endogenous transcripts, specifically targeting precursor mRNA splicing without off-target effects, making them a potential alternative method for regulating RNA splicing and a new tool for the treatment of genetic disorders, such as Duchenne muscular dystrophy (DMD) [67]. Therefore, we may be able to employ AS-circRNA targeting hsa_circRNA_000166 to treat hereditary BC. Recently, immunotherapy has shown increasing clinical benefits, and circRNAs can remodel the tumour microenvironment by regulating T cells, NK cells, and macrophages, which offers potential opportunities for cancer therapy [68]. Additionally, aberrant expression of circRNAs is present in almost all cancers and plays an important effect in tumour immune escape. For example, circFGFR4 plays a critical role in immune escape and resistance to anti-PD-1 immunotherapy by regulating the miR-185-5p/CXCR4 axis in TNBC, suggesting that circFGFR4 has great potential as a biomarker for predicting susceptibility to anti-PD-1 immunotherapy and as a target for TNBC immunotherapy [69]. CircRNAs have also been reported to be expressed aberrantly in immune disorders, such as rheumatoid arthritis (RA) (a chronic autoimmune disease), and are involved in the pathophysiology of RA [70]. Importantly, the stability of circRNAs because of their unique structure allows them to be used as novel vaccines. Vaccines are effective tools for immunotherapy, for example, in the prevention of infectious diseases and the treatment of cancer. CircRNA vaccines represent a potential new avenue in the vaccine era, and recently, several circRNA vaccines have been synthesized and tested in vitro and in vivo [71]. Thus, hsa_circRNA_000166 is a promising candidate for novel immunotherapy in the treatment of BC.

Nevertheless, there are some limitations to this study. First, while we relied on previous studies that included the expression results of circRNA_000166 from circRNA microarray-seq datasets GSE101123 and GSE101124 in the GEO database, we once again validated the clinically significant yet controversial circRNA_000166 in BC. However, we did not integrate RNA-seq technology to comprehensively evaluate all the pathways and genes influenced by circRNA_000166. Instead, we focused on pathways of interest based on prior studies and bioinformatics predictions for our research and validation. Second, due to limited funding, we could not conduct in vitro functional experiments involving circRNA_000166 overexpression and additional rescue experiments. These included studies on miR-326/miR-330-5p/ELK1 expression, cell cycle changes (flow cytometry), cell apoptosis (WB detection of candidate genes such as caspase 3), cell invasion (Transwell invasion), and migration (scratch healing assay). Additionally, in vivo validation through animal experiments to explore the functional role of hsa_circRNA_000166, such as its effects on tumour proliferation, was also not feasible for the same reason. This limitation prevented a comprehensive understanding of the mechanism of action. Finally, due to the unavailability of antibodies, we failed to perform WB analysis to determine the impact of si-circRNA_000166 transfection on the protein expression of the parent gene RPPH1. Nevertheless, we plan to address these limitations and conduct these key experiments to explore new mechanisms in our future studies.

5. Conclusion

In summary, we demonstrated that circRNA_000166 is upregulated in human BC tissues and cells. This upregulation is positively related to tumour size, TNM stage, histologic grade, and lymph node metastasis, and negatively associated with the postoperative survival of BC patients. Suppression of circRNA_000166 significantly decreases BC cell proliferation and migration and promotes apoptosis via the miR-326/ELK1 and miR-330-5p/ELK1 axes (SFigure 3). Hence, circRNA_000166 is a feasible target for BC diagnosis and treatment. The clinical utility of circRNA_000166 as a diagnostic marker and therapeutic target must be further validated in future investigations.

Funding Statement

No external funding sources supported this study.

Authors contributions

Li-Hui Ma: conceptualization and design (lead); project administration (lead); writing – review and editing (lead). Ming-Hui Wang: writing – original draft (lead). Han-Cheng Liu: formal analysis (lead); visualization (lead); writing – review and editing (equal). Zi-Hui Liu: data curation (equal); investigation (equal); methodology (equal); writing – review and editing (equal). Hong-Xu Zhang: data curation (equal); investigation (equal); methodology (equal); writing – review and editing (equal). All authors were in agreement over the final version of this article.

Ethics approval and consent to participate

The research protocol was approved by the Ethics Committee of the Affiliated Hospital of Chengde Medical College (Ethical Approval Number: CYFYLL2022100). The study was performed based on the ethical principles outlined in the Helsinki Declaration. Written and verbal informed consent was obtained from all study participants before analyzing tissue specimens and clinical data.

Patient consent for publication

Not applicable.

Disclosure statement

No potential conflict of interest was reported by the authors.

Availability of data and materials

All data generated or analyzed in this article can be obtained from the corresponding author upon reasonable request.