Identification of plasma hsa_circ_0008673 expression as a potential biomarker and tumor regulator of breast cancer

Abstract Objective Cell‐free circular RNAs (circRNAs) are stable and abundantly exist in body fluids. In this study, we aimed to investigate plasma cell‐free circRNAs as a novel class of biomarkers for the diagnosis of breast cancer (BC). Methods Differentially expressed circRNAs from 6 normal and 6 BC plasma samples were detected by microarray. Hsa_circ_0008673 was then screened and validated in the plasma of 102 normal and 378 BC samples. A receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value. The correlations between hsa_circ_0008673 expression and demographic characteristics, tumor features, and prognosis were analyzed. The effects of hsa_circ_0008673 on BC cell proliferation and metastasis were also measured. Results Of the top ten up‐regulated (hsa_circ_0008673, hsa_circ_0008500, hsa_circ_0005260, hsa_circ_0003423, hsa_circ_0119881, hsa_circ_0000987, hsa_circ_0007386, hsa_circ_0000091, hsa_circ_0016601, and hsa_circ_0008549) and top ten down‐regulated (hsa_circ_0000826, hsa_circ_0072697, hsa_circ_0004587, hsa_circ_0000471, hsa_circ_0007786, hsa_circ_0001417, hsa_circ_0005982, hsa_circ_0001566, hsa_circ_0003823, and hsa_circ_0003823) circRNAs from microarray, hsa_circ_0008673 was the most significantly up‐regulated circRNA in BC, and represented a good diagnostic value. Hsa_circ_0008673 was remarkably down‐regulated after breast mastectomy. Hsa_circ_0008673 expression was associated with larger tumor size, distant metastasis, positive estrogen receptor (ER) status, and positive progesterone receptor (PR) status. Additionally, hsa_circ_0008673 could serve as a prognostic predicator of overall survival (OS) and disease‐specific survival (DSS). Cell assays proved that hsa_circ_0008673 knockdown contributed to inhibition of tumor cell proliferation and migration. Conclusion Plasma cell‐free hsa_circ_0008673 was up‐regulated in BC, which was associated with poorer prognosis and promoted tumor proliferation and metastasis. Hsa_circ_0008673 is a promising biomarker for tumor diagnosis and prognostic assessment of BC patients.


| INTRODUC TI ON
Breast cancer is responsible for 14% of total cancer-related deaths and is the leading cause of cancer in women. 1  Although early diagnosis, radical surgery, neoadjuvant/adjuvant therapy, and targeted drug applications have contributed to substantial improvements in the survival rate of breast cancer patients with curative intent, the long-term mortality rate and high recurrence rate remain urgent clinical problems. 3 As a result, it is essential to investigate novel therapeutic targets that can increase the recovery rate and promote long-term survival of breast cancer.
Circular RNAs (circRNAs) are class of RNAs formed by back-splicing events as loops, and are found in all types of organisms. 4,5 They differ from long noncoding RNA (lncRNA) and microRNAs (miRNAs) in that they do not have the 5′ and 3′ end structures but represent covalently closed cyclic structures. 5 With newly developed technologies of high-throughput sequencing and computational approaches, particularly RNA sequencing, up to 30 000 circRNAs have been identified. 6 CircRNAs are widely involved in the regulation of human physiology and pathology by three main mechanisms including function as a miRNA sponge, as a protein-binding molecule, and as a template for translation into polypeptides. 7,8 It is reasonable to hypothesize that dysregulation of circRNAs may influence the progress of various diseases, including cancer. 9 Zhang et al have reported that most circRNAs express tissue or developmental stage specificity, and circRNAs are involved in the regulation of a variety of biological activities in cancers. 10 It is important that the unique construction of circRNAs makes them insensitive to ribonucleases, which indicates that cir-cRNAs can exist in tissues and serum, and as a result could serve as biomarkers for human cancer. 11 Recently, studies have reported that circRNAs could act as promising biomarkers in numerous types of cancers. 12, 13 Li et al have shown that hsa_circ_0000729 is a potential prognostic biomarker in lung adenocarcinoma. 12 Hsa_circ_0001445 could regulate the proliferation and migration of hepatocellular cancer and serve as a diagnostic biomarker. 13 CircRNAs have also been proved to play important roles in breast cancer. For instance, Li et al have reported that circular RNA VRK1 is correlated with favorable prognosis, and could inhibit cell proliferation but promote apoptosis in breast cancer. 14 CircKDM4C could suppress tumor progression and attenuates doxorubicin resistance by regulating miR-548p/PBLD axis in breast cancer. 15 CircRNA hsa_circ_0000519 has been reported to be critical in the pathogenesis of breast cancer and may serve as a future therapeutic biomarker. 16 However, the functions and mechanisms of circRNAs in breast cancer remain uncertain.
In this study, a novel circRNA hsa_circ_0008673 was firstly screened and analyzed in both normal and breast cancer plasma samples. We found hsa_circ_0008673 was significantly up-regulated in breast cancer patients and positively correlated with larger tumor size, distant metastasis, positive ER status, and positive PR status. We further found that hsa_circ_0008673 was also a promising marker for distinguishing breast cancer patients from normal volunteers compared with CA153 and CEA. Moreover, the OS and DSS of patients with high expression of hsa_circ_0008673 were significantly shorter compared with the low-expression group. Cell assay further proved that knockdown of hsa_circ_0008673 resulted in inhibiting the proliferation and metastasis of breast cancer cells.
In conclusion, plasma cell-free hsa_circ_0008673 is a promising biomarker for tumor diagnosis and prognostic assessment of breast cancer patients.

| Microarray analysis
CircRNA microarray analysis was performed using Human CircRNA Array v2.1 (CapitalBio Technology). Total RNA was quantified using biomarker, breast cancer, hsa_circ_0008673, prognosis analyze the acquired array images. Quantile normalization and subsequent data processing was performed using the R software limma package. Differentially expressed circRNAs were identified through fold change filtering. Hierarchical clustering was performed to show the distinguishable circRNA expression pattern among samples.

| RNA extraction and qRT-PCR
Total RNA was extracted from breast cancer plasma samples and normal samples using the TRIzol reagent (Invitrogen) according to the manufacturer's protocol. Total RNA from each specimen was quantified, and quality assurance was conducted using a NanoDrop ND2000 spectrophotometer (NanoDrop). Reaction mixture (20 µL) containing 1 µg total RNA was reverse-transcribed to cDNA using T47D were cultured in RPMI (Invitrogen) supplemented with 10% FBS with and without insulin. All cells were cultured in a humidified atmosphere containing 5% CO2 at 37°C.

| Establishment of stable hsa_circ_0008673 knockdown cell lines
Vectors that stable knockdown (pLKO.1) hsa_circ_0008673 were obtained from Vigene Biosciences and were transfected into MDA-MB-231 and MCF-7 using a Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific, Inc) according to the provided protocol.
Cells were then cultured with puromycin to filter the stable cell lines.

| Cell cycle assay
Cells were collected, washed twice with 1X PBS, and fixed in 70% ethanol at −20°C. After 24 hours of fixation, cells were incubated with RNase A (Takara Bio, Inc) at 100 µg/mL in 1X PBS for 30 minutes at 37°C. Cells were then stained with propidium iodide (PI; BD Biosciences) at 50 µg/mL for 30 minutes at room temperature.
Subsequently, cells were analyzed for DNA content using a BD FACSCalibur™ flow cytometer (BD Biosciences).

| Migration and invasion assay
Migration and invasion assays were performed as described previously using the Transwell system (Corning Costar). 17 In the migration assay, 700 µL of medium with 20% FBS was added to the lower well of each chamber and 1 × 10 5 cells suspended in serum-free medium were added to the upper inserts. After incubation for the indicated time, the total number of cells adhering to the lower surface of the membrane was quantified in six representative fields. The invasion assay was performed in the same way as the migration assay except that the membrane was coated with Matrigel (BD Biosciences).

| Statistical analysis
Statistical differences between independent groups were calculated by Student's t test or one-way ANOVA test. AUC values, sensitivity, and specificity for cell-free hsa_circ_104056 were quantified by using receiver operating characteristic (ROC) analysis to assess its diagnostic efficiency in differentiating patients with breast cancer from healthy controls. The optimal cutoff thresholds for diagnosis were obtained by Youden index. Correlation was determined by the Spearman test. Patients were equally divided into two groups based on their relative expression, OS and DSS curves were evaluated by the Kaplan-Meier method, and the survival differences of patients in subgroups were estimated by the log-rank test. Factors predicting survival were determined by univariate and multivariate Cox's proportional hazard regression. All graphing and statistical analyses were carried out using SPSS 17.0. P < .05 (two-tailed test) was regarded as significant difference, and data were exhibited as mean ± standard deviation.

| Hsa_circ_0008673 was up-regulated in breast cancer plasma samples
Plasma samples from 6 healthy controls and 6 breast cancer patients were sent for microarray. As shown in Figure 1A, a total of 54 circRNAs were up-regulated and 94 circRNAs were down-reg- hsa_circ_0005982, hsa_circ_0001566, hsa_circ_0003823, and hsa_circ_0003823) circRNAs was described in Table S1. Of the significantly changed circRNAs, hsa_circ_0008673 was filtered out because it was the most remarkable up-regulated circRNAs, and the expression level in microarray is shown in Figure 1B. Moreover, the relative expression between groups was evaluated by qRT-PCR, which was in accordance with the result of microarray ( Figure 1C). shown in Figure 1D, hsa_circ_0008673 was remarkably increased in breast cancer patients, which indicates that hsa_circ_0008673 is a potential breast cancer-associated circRNA.

| Baseline characteristics and hsa_circ_0008673 expression of the 378 breast cancer patients
The baseline characteristics of the 378 breast cancer patients are shown in Table 1

| Diagnostic efficacy of plasma cell-free hsa_ circ_0008673
Since the expression of hsa_circ_0008673 was increased in breast cancer samples and was correlated with malignant biological behaviors, we further examined the diagnostic efficacy of plasma cell-free hsa_circ_0008673. As shown in Figure

| Association of plasma cell-free hsa_ circ_0008673 expression with the prognosis of breast cancer patients
To further evaluate the clinical significance of cell-free hsa_ circ_0008673, the association between hsa_circ_0008673 and OS (overall survival) and DSS (disease-specific survival) was analyzed.
As shown in Figure 4, the Kaplan-Meier survival curves manifested that breast cancer patients with high plasma cell-free hsa_ circ_0008673 expression had significantly worse OS ( Figure 4A) and DSS ( Figure 4B) than those with low hsa_circ_0008673 expression, indicating that cell-free hsa_circ_0008673 was a potential biomarker for predicting the prognosis of breast cancer.

| Factors affecting the prognoses of breast cancer patients
Univariate Cox's analysis displayed that hsa_circ_0008673 (high vs low) was associated with poorer prognoses for both OS (HR = 1.900,  (Table 3).

| Knockdown of hsa_circ_0008673 suppressed the proliferation and metastasis of breast cancer cells
To further investigate the mechanism of hsa_circ_0008673 in breast cancer, hsa_circ_0008673 relative expressions in breast cancer cell lines and normal breast epithelial cell line were detected by qPCR. As shown in Figure 5A, we found that hsa_circ_0008673 was up-regulated in breast cancer cell lines and down-regulated in

| D ISCUSS I ON
The identification of molecular markers in body fluids (eg, sera and urine), which can be used as noninvasive diagnostic, prognostic, and surveillance markers in cancer management, is one of the most ambitious challenges in oncologic research. 19 To date, the increased interest on noninvasive biomarkers, allowed by using novel methodologies (such as next-generation sequencing, single-cell sequencing approaches, and digital PCR), has greatly improved the translational potential of researches into clinical application. 20,21 Recently, circR-NAs have been considered as greatly potential biomarkers in many kinds of tumors due to their stability in body fluids, 11  In our study, we firstly found hsa_circ_0008673 was remarkably up-regulated in 6 breast cancer samples compared with 6 normal controls based on microarray and qRT-PCR ( Figure 1A-C), which was further verified in 378 breast cancer and 102 normal samples (P ˂ .001) by quantitative qRT-PCR assays ( Figure 1D).
Clinicopathological features showed that up-expression of hsa_ circ_0008673 level was positively associated with larger tumor size, distant metastasis, positive ER status, and positive PR status, indicating hsa_circ_0008673 might contribute to malignant behaviors of breast cancer ( Figure 2). Furthermore, we evaluated whether hsa_circ_0008673 could serve as a valuable biomarker for the diagnosis of breast cancer. As shown in Figure 3A, hsa_circ_0008673 could distinguish breast cancer patients from normal controls, which proved its diagnostic value. Indeed, the AUC and the specificity of hsa_circ_0008673 were much higher than those of the conventional tumor markers such as CA153 and CEA. Further, qRT-PCR showed that the level of hsa_circ_0008673 in plasma might reflect the expression in tumor ( Figure 3C). Based on the Kaplan-Meier method, we showed that hsa_circ_0008673 could predict the prognosis of both OS and DSS in breast cancer patients, in which high expression of hsa_circ_0008673 predicted poorer prognosis (Figure 4).
Also in this study, we took further steps to ectopically silence hsa_circ_0008673 in breast cancer cells and then evaluate the functional role of hsa_circ_0008673 in cancer malignant behaviors.
Through several in vitro assays, we found that hsa_circ_0008673 inhibition significantly suppressed both tumor cell proliferation via inducing G1 phase cell cycle arrest and metastatic abilities ( Figure 5).
These data suggest that hsa_circ_0008673 could influence the malignant behaviors of breast cancer cells.
In conclusion, our research found a novel biomarker, which was positively correlated with the malignant clinical characteristic of patients, and could distinguish breast cancer from normal control and predict the prognosis of both OS and DSS. Moreover, we identified that hsa_circ_0008673 was an oncogene, which contributed to the proliferation and metastasis of breast cancer cells. This study sheds lights on understanding the mechanisms of disease-associated cir-cRNAs and improving the diagnosis and prevention of circRNA-associated diseases.

ACK N OWLED G M ENTS
This research did not receive grants from any funding agency in the public, commercial, or not-for-profit sectors.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.