circCPA4 induces malignant behaviors of prostate cancer via miR-491-5p/SHOC2 feedback loop

Highlights • circCPA4 deficiency inhibits PC3 cell metastasis.• circCPA4 interacts with miR-491-5p in affecting the biological behavior of PC3 cells.• circCPA4 mediates and affects the biological behavior of PC3.


Introduction
Prostate Cancer (PC) is a common cancer type affecting men's health worldwide. 1PC originates from the gland epithelial cells of the prostate and is characterized by the uncontrolled growth and proliferation of cancer cells. 2 Currently, treatment options for PC include surgery, immunotherapy, hormone therapy, radiation therapy, and chemotherapy. 3,4espite advances in PC treatment, mortality and morbidity associated with the disease remain high, and new treatment strategies are urgently needed.
Upon reverse splicing, circRNAs are formed by covalently linking their downstream 3′end and upstream 5′end. 5circRNAs are expressed in large quantities in tissues and involve gene regulation, including transcription and post-transcriptional regulation. 6Dysregulation of circR-NAs in cancers, including PC, has been shown, indicating their potential as a diagnostic and therapeutic target. 7,80][11] However, its biological function in PC remains unclear.miRNAs are crucial in gene expression modification 12 and miRNAs can target specific mRNAs and regulate their expression, while circRNAs can act as a sponge for miRNAs and regulate their availability to targeted mRNAs. 13This means that miRNA and circRNA can interact with each other to form complex regulatory networks and participate in the regulation of various activities of PC. 14 However, whether circCPA4 regulates miRNA/mRNA networks and is involved in PC development remains unclear.
This study investigated the biological function of circCPA4 in PC and put forward a hypothesis that circCPA4 participates in the development process of PC by regulating downstream miRNA/mRNA networks.

Clinical samples
This study obtained approval by the Ethics Committee of The First School of Clinical Medicine of Southern Medical University.All patients presented informed consent.Thirty-nine pairs of PC tissue samples and normal tissue samples were obtained from The First School of Clinical Medicine of Southern Medical University and confirmed by the pathologist.Patients with preoperative pathological diagnosis of PC signed informed consent before being enrolled in the study.Patients were excluded if they had a second primary tumor, HIV or syphilis virus, severe liver, kidney or other systemic disease, or other malignant disease, and had received chemotherapy or radiation before surgery.After excision or biopsy, the tissue was rapidly frozen and stored at -80°C.

RT-qPCR
A NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific) measured the quality of RNA obtained from cells/tissues using a Trizol reagent (Life Technologies).PrimeScript RT Master Mix (RR036A; Takara) was employed to generate cDNA of circRNA and mRNA, whereas PrimeScript™RT Reagent Kit (RR037A; Takara) was purchased for miRNA reverse transcription.ChamQ SYBR qPCR Master Mix (Q311-02; Vazyme) and Roter Gene 3000 sequence detection system (Corbett Research, Australia) performed PCR.To calculate relative gene expression, 2 −ΔΔCt method was utilized, and GAPDH or U6 was considered endogenous control.The primer sequence is found in Table 1.

Actinomycin D treatment
PC3 cells were plated in 6-well plates (5×10 5 cells/well) and allowed to 2h culture.RNA stability was analyzed by PCR after actinomycin D was administered (2 μg/mL, Sigma) at a specific time point.RNAse R treatment 10 μg RNA from PC3 cells was mixed with RNAse R (3 U/g, Epicenter) at 37°C and after 30 min, circRNA and linear RNA were evaluated using PCR.

FISH assay
RiboBio (Guangzhou, China) provided Cy3-labeled circCPA4 and DIG-labeled lock-in nucleic acid miR-491-5p probes, as well as a FISH kit to obtain images.A1Si laser scanning confocal microscope (Nikon) and ModFit LT software were auxiliary for data analysis.

Colony formation test
Six-well plates were covered with RPMI-1640 containing 10 % FBS, in which cells were seeded at 2000 cells/well for 10d.After fixation and crystal violet staining, the colonies were imaged and counted.

Flow cytometry
PC3 cells were assayed by Annexin V-FITC apoptosis detection kit (Vazyme).Cells after rinsing with cold PBS (Sangon, China) were resuspended in a binding buffer and reacted darkly with 5 µL Annexin V-FITC and 5 µL PI for 15 min.Analyses of flow cytometry (Beckman Coulter, USA) were conducted to determine apoptosis rates.

Transwell tests
The Transwell chamber (Corning, USA) was equipped with an 8 μm pore size polycarbonate filter.Cells (5×104 ) in serum-free medium were coated on the upper cavity, in which Matrigel (BD356230, Corning) was only required for invasion assay.Meanwhile, 600 μL 10 % FBS was fulfilled with the lower cavity.After culture, cells transferred to the lower cavity were fixed in 95 % ethanol for 15 min, stained with 0.1 % crystal violet for 20 min, and counted under a microscope (Leica Microsystems) in 5 regions.

Immunoblot
Proteins were isolated from tissues and cells using RIPA lysis buffer (Thermo Fisher) and quantified using a BCA analysis kit (BioVision, USA).

RIP experiment
RIP assay was performed using the Magna RIP kit (Millipore).Cells were lysed in 100 μL RIP lysis buffer containing a mixture of RNAse inhibitors (Promega) and protease inhibitors (Roche).After DNase I (Roche) treatment, 900 μL RIP immunoprecipitation buffer was added to the lysate and reacted with antibody-conjugated magnetic beads (anti-AgO2 or anti-IgG).The beads after RIP washing buffer treatment were processed with protease K for 30 min before PCR analysis.

Tumor growth in vivo
All experimental mouse procedures were approved by the Animal Ethics Committee of The First School of Clinical Medicine of Southern Medical University, and the researchers minimized the suffering of the mice.Six eight-week-old male BALB/c nude mice were purchased from Hunan SJA Animal Experimental Co., Ltd.As for in vivo tumor metastasis studies, PC3 cells (5×10 6 cells) with stable circCPA4 knockdown or nonknockdown circCPA4 were inoculated subcutaneously into mouse armpits.During the following 30d, using a digital caliper, the size of tumors was measured periodically in the mice.Tumor volume (cm 3 ) = 0.5 × length × width 2 .After that, the excised tumor was weighed, fixed with 4 % paraformaldehyde, and paraffin-embedded for IHC staining. 15

Data analysis
GraphPad Prism 9.0 was utilized for data analysis.Data were presented as mean ± standard deviation.Two-tailed Student t-test and oneway ANOVA compared the data.Chi-Square assay assessed the association of circCPA4 with clinicopathologic features in PC patients; *p < 0.05 was statistically significant.

circCPA4 high expression in PC
circCPA4 expression pattern in PC was examined.PC tissues (Fig. 1A) as well as four PC cell lines (Fig. 1B) had higher circCPA4 levels than controls.To examine the ring structure of circCPA4, both actinomycin D and RNAse R treatments were conducted, indicating that the two methods caused no change in the RNA stability of circCPA4, but reduced that of linear RNA (Fig. 1 C and D).Also, circCPA4 could be amplified by outward-diverging primers in cDNA, but circCPA4 could not be amplified from gDNA (Fig. 1E).Abnormal expression of circCPA4 was highly correlated with TNM staging and distal metastasis in PC patients (Table 2).A poorer survival prognosis was observed for those with high circCPA4 expression (Fig. 1F).
circCPA4 deficiency inhibits PC3 cell metastasis circCPA4-targeted siRNA was transfected into PC3 cells to analyze the biological function of circCPA4.Transfection of si-circCPA4 reduced circCPA4 expression effectively (Fig. 2A).Colony formation and EdU tests noted that circCPA4 knockdown reduced cell proliferation capacity and EdU-positive rate (Fig. 2 B and C).Also, circCPA4 depletion inhibited protein expression of proliferating protein Ki-67 (Fig. 2F).Cellular apoptosis was evaluated, as manifested by increased apoptosis, and cleaved caspase-3 protein expression in circCPA4-knockdown cells (Fig. 2 D and F).Transwell assay presented that circCPA4 depletion reduced the number of invading and migrating cells (Fig. 2E) and suppressed N-cadherin and increased E-cadherin expressions (Fig. 2F).

circCPA4 knockdown inhibits PC tumor growth
Tumor growth was observed after circCPA4 interference in mice.Knocking down circCPA4 lowered tumor volume and weight (Fig. 7  A-C).IHC staining suggested that circCPA4 deficiency reduced SHOC2 and Ki-67 expressions in tumors (Fig. 7D).

Discussion
Abnormal expression of circRNA affects PC development. 16,17This study highlights that downregulating circCPA4 inhibited the metastasis of PC.The identification of miR-491-5p/SHOC2 axis as a downstream target of circCPA4 provides a potential pathway for developing targeted therapeutics for PC.
In fact, circCPA4 has been emphasized to correlate with TNM staging, lymph node metastasis, and tumor size in lung cancer patients. 10The study further confirmed that abnormally high expression of circCPA4 was associated with TNM staging and distal metastasis in PC patients.In addition, circCPA4 expression levels had prognostic values for overall survival in PC patients.This highlights the potential clinical role of circCPA4 as a diagnostic and prognostic biomarker for PC.Thus, a deeper understanding of the molecular mechanisms of circCPA4′s role in PC could provide new insights into disease pathogenesis.This study declared that circCPA4 was involved in regulating the proliferation, invasion, and migration, apoptosis, and EMT processes of PC3 cells.EMT is a key process of cancer invasion and metastasis. 18,19Since high circCPA4 is highly associated with remote metastasis in PC patients, it is suggested that circCPA4′s promotion of invasive migration and EMT processes will contribute to remote metastasis of PC.In addition, it has been declared that high expression of circCPA4 can block apoptosis and induce proliferation of cancer, 11,20 showing consistency with the results of this study.
Notably, this study further confirmed the ring structure of circCPA4 in PC3 cells.Compared with linear RNA, circRNA is more stable. 21,22herefore, circCPA4 has a more permanent role in regulating downstream miRNA expression and function, which will continue to influence the malignant development of PC. circCAP4 regulates the biological behavior of different cancers through miRNA/mRNA networks.These miRNA/mRNA networks include miR-214-3p/TGIF2 10 and miR-760/ MEF2D. 9This study identified a novel mechanism network for circCPA4 in PC. circCPA4 promoted SHOC2 expression through competitive adsorption of miR-491-5p, thus accelerating multiple biological behaviors of PC3 cells.
However, understanding of the molecular mechanism and biological function of circCPA4 is still limited, and further studies are needed.Secondly, circRNA and miRNA/mRNA targeting regulatory network is extremely complex, which needs to be improved in subsequent studies.In addition, how to detect circCPA4 quickly, effectively, and accurately in the clinic is also a problem that needs to be solved.
Together, the circCPA4/mi-491-5p/SHOC2 axis is involved in PC by regulating cell proliferation, invasion and migration, apoptosis, and EMT.Thus, circCPA4 may be a promising target for developing novel diagnostic and therapeutic strategies for PC.

Fig. 1 .
Fig. 1. circCPA4 high expression in PC. (A-B) Higher circCPA4 expression in PC tissues versus normal tissues and in PC cell lines versus RWPE.(C-D) circCPA4 had ring structure and high stability.(E) Negative relationship between circCPA4 and survival prognosis of PC patients.

Table 2
Correlation analysis of clinicopathological characteristics of patients with circCPA4 and PC.