Background
Prostate cancer (PCa) is the second most frequent cancer and the fifth-leading cause of cancer-related death in men [
1]. Many therapeutic strategies have offered the opportunity for the cure of PCa, including radical prostatectomy or radiation therapy. Unfortunately, most PCa patients are often diagnosed in an advanced stage in which radical prostatectomy cannot be performed leading to poor prognosis. The 5-year relative survival rate for localized PCa is close to 100%, compared to 30% for advanced metastatic prostate cancer (mPCa) [
2]. Therefore, exploring tumor-related biomarkers and understanding the molecular events of PCa will be useful for early diagnosis and efficient treatments.
Circular RNAs (circRNAs) are a group of endogenous non-coding RNA molecules that are linked head to tail by reverse splicing to form a covalently closed loop structure [
3]. Growing number of studies have revealed that circRNAs play critical roles in regulating tumor proliferation and metastasis of multiple human malignancies [
4]. They may act as miRNA decoys [
5], RNA-binding protein sponges and protein scaffolds [
6], transcriptional regulators [
7], or templates for protein translations [
8]. Due to their high stability, broad expression and tissue specificity, circRNAs are emerging as promising biomarkers and therapeutic targets for cancers. Several circRNAs, such as circSMARCA5 [
9], hsa_circ_0003258 [
10] and circPFKP [
11] have been found to impact the development and progression of PCa. However, the specific roles and mechanism of circRNAs in PCa progression are largely unknown.
The tumor microenvironment (TME) is a highly heterogeneous ecosystem that typically contains a collection of tumor cell populations, immune cells, and tissue-specific resident and recruited stromal cell types [
12]. Tumor-associated macrophages (TAMs) are an important component of the TME and displays two major phenotypes, M1 and M2. It is known that M1 macrophages can be polarized by lipopolysaccharide (LPS) and interferon-γ (IFN-γ), whereas M2 macrophages are polarized in the presence of IL-4 and IL-13 [
13,
14]. Increased TAMs infiltration is associated with advanced disease and poor overall survival in breast cancer [
15], pancreatic cancer [
16] and bladder cancer [
17]. Therefore, targeting TAMs infiltration could be a promising target for cancer therapy.
Here, we conducted circRNA sequencing in the plasma of PCa patients to identify circRNAs that are involved in PCa progression. For the first time, we reported a novel circRNA, named circSMARCC1 (circBase ID: hsa_circ_0001296), which is highly expressed in PCa tissue samples and cell lines. Then, gain- and loss-of-function experiments were conducted to reveal the biological roles of circSMARCC1 in cell growth, invasion and metastasis both in vitro and in vivo. Mechanistically, we found that circSMARCC1 increases CC-chemokine ligand 20 (CCL20) by sponging miR-1322 and activates the PI3K-Akt pathway to promote growth and epithelial mesenchymal transformation (EMT) of PCa cells. More importantly, we found that circSMARCC1 promotes recruitment of macrophage and M2 polarization, which in turn facilitates the progression of PCa. Our findings indicate that circSMARCC1 could be a promising therapeutic target for PCa.
Materials and methods
Patients and clinical samples
The plasma of PCa patients and BPH individuals used in this study were obtained from patients who were hospitalized at Nanfang Hospital of Southern Medical University (Guangzhou, China). Formalin-fixed paraffin-embedded PCa specimens were acquired from patients that underwent radical prostatectomy at Nanfang hospital, and patients’ clinical information was obtained by reviewing the follow-ups of their electronic medical records. All experimental procedures were approved by the Medical Ethics Committee of Nanfang Hospital of Southern Medical University (NFEC-2022-083). The selection criteria for PCa patients are as follows. Inclusion criteria: 1) Pathologically confirmed prostate adenocarcinoma; 2) New cases without any preoperative treatment; 3) Patients signed informed consent. Exclusion criteria: 1) The pathological examination type was neuroendocrine prostate cancer (NEPC)/small cell prostate cancer; 2) Combined with other tumors; 3) Patients with other acquired, congenital immunodeficiency disease, severe liver, kidney or other systemic diseases, or a history of organ transplantation; 4) Patients who received preoperative chemotherapy or radiotherapy before surgery. The median age of the enrolled patients was 65.5 years, and the average age was 66 (range: 48–81 years). Clinical TNM staging and Gleason scores of patients were based on the American Joint Committee on Cancer Eighth Edition (2017) and the 2016 World Health Organization classification of genitourinary tumors.
Cell culture and treatment
The human monocyte THP-1cells, human embryonic kidney HEK-293 T cells, normal prostate epithelial RWPE-1 cells and PCa cells (PC-3, DU145, 22Rv1, C4–2 and LNCaP) were obtained from the Cell Bank of the Chinese Academy of Sciences. Cells were grown in RPMI-1640 medium (Gibco, United States) supplemented with 10% FBS (Gibco, United States). All cells were maintained at 37 °C with 5% CO
2. Three small interfering RNAs (siRNAs) targeting circSMARCC1 (si-circ # 01, 02, 03) and miR-1322 inhibitors or mimics were purchased from RiboBio Company (Guangzhou, China). Cell lines stably overexpressed circSMARCC1(lv- circSMARCC1) or knocked down circSMARCC1(sh-circSMARCC1) were established using lentivirus vectors (GeneChem Bio-Medical Biotechnology, Shanghai, China), and the transfected cells were selected in puromycin (2 g/ml) for 1 week. All the target sequences are shown in Supplementary Table S
1–
2.
CircRNA microarrays
The circRNAs from the plasma of PCa patients and control individuals for microarray analysis were based on the previous protocols [
18]. Arraystar Human circRNA Array v2 (Kangcheng Biotech, Shanghai, China) was applied to the analysis of the circRNA microarray. Sample preparation was performed according to the Arraystar standard protocols, as described previously [
19]. Differentially expressed circRNAs were identified via fold change filtering. We defined the statistical criteria for selecting differentially expressed circRNAs using fold change > 1.5 with
p < 0.05.
Total RNA was extracted from tissues or cell lines using TRIzol reagent (Takara, Dalian, China). RNAs from the nucleus and cytoplasm of PCa cells were separated by Nuclear/Cytoplasmic Isolation Reagent (Thermo Fisher Scientific, United States) following the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was applied for further detection. For RNase R treatment, cells were treated with 2 mg of total RNA for 10, 20, and 30 min at 37 °C with 3 U/g of RNase R (Epicentre Biotechnologies, Madison, WI, United States). In addition, total RNA from PCa cells was treated with 1 μg/ml actinomycin D (Cell Signaling Technology, Beverly, MA, United States) against new RNA synthesis for 0, 4, 8, 12, and 24 h. RNA was reversely transcribed into cDNAs with the PrimeScript RT reagent Kit (Takara, Dalian, China) according to the manufacturer’s instructions. SYBR Green PCR Master Mix (Takara, Dalian, China) and the Applied Bio-systems 7500 Fast Real-Time RCR System (Applied Biosystems, United States) were used for RT-qPCR analysis. Each measurement was performed in triplicate and the results were standardized against the internal control GAPDH. The relative expression of target genes was calculated using the 2
-△△Ct method. All the primers used are shown in Supplementary Table S
3.
Fluorescence in situ hybridization (FISH)
Cy3-labeled circSMARCC1 (RiboBio, Guangzhou, China) and Alexa 488-labeled miR-1322 probes (FOCOFISH, Guangzhou, China) were used to observe the co-localization of circSMARCC1 and miR-1322 in PCa tissues and cells. The FISH experiment was conducted using a Fluorescent in Situ Hybridization Kit (No. C10910, RiboBio, Guangzhou, China), according to the official guidelines. Cell nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI, Beyotime, China). The images were photographed under the fluorescence microscope (LSM 880 with Airyscan, Carl Zeiss, Germany).
Cell proliferation, scratch assay, migration and invasion assays
CCK-8, EDU, colony formation, scratch assay, transwell migration, and invasion assays were performed as previously reported [
20]. Specifically, for cell proliferation assays, stably transfected PCa cells were treated with the conditioned medium (CM) from TAMs (co-cultured with PCa cells), human recombinant CCL20 protein (rh-CCL20, 20 ng/ml, #0511102, Peprotech) and neutralizing antibody to CCL20 (anti-CCL20, 5 μg/ml; ab9829, Abcam). The cell proliferative rate was assessed using the Cell Counting Kit-8 (CK-04, Dojindo). For tumor cell migration assays, the stable cells lines treated with or without 5 μg/ml CCL20 neutralizing antibody were seeded in the upper chambers; the lower chambers were filled with medium containing 10% FBS with or without 20 ng/ml CCL20 recombinant protein. In addition, the CM of TAMs was placed in the lower chamber and used as an attractor for PCa cell migration and invasion experiments.
Macrophage generation, macrophage migration, and co-culture assay
We induced M2 macrophage generation by sequential stimulation as follows: THP-1 cells were treated with 100 ng/ml phorbol-12-myristate-13-acetate (PMA) (Beyotime, Shanghai, China) for 24 h for differentiation into adhered THP-1 macrophages (THP-1-Mø). Then, THP-1-Mø were incubated with 20 ng/ml IL-4 (AF-200-04, PeproTech) and 20 ng/ml IL-13 (AF-200-13, PeproTech) for 48 h to obtain M2 polarization (THP-1-M2). For macrophage migration, the migration assays were performed using 6.5 mm transwell plates with 5.0 μm pore size inserts. The CM of stably transfected PCa cells with or without CCL20 recombinant protein was served as an elicitor in the lower chamber of the 24-well plate, and THP-1-M2 were added to the upper transwell insert (#09717050, Corning). To explore the potential mechanisms by which CCL20 acts, TAMs were incubated with CCR6-neutralizing antibody (anti-CCR6, #MHH-160-F(E), Creative Biolabs) before migration assays were performed. After 48 h incubation, the cells adhering to the lower filter surface were fixed with 4% paraformaldehyde for 10 min and stained with Giemsa (Baso Diagnostics, Inc., Zhu Hai, China) for transwell migration assays. For co-culture experiments, PCa cells with stably overexpressed or knocked down circSMARCC1 were inoculated into the upper insert and then transferred to 6-well plates pre-inoculated with THP-1-Mø. After 48 h, macrophages were collected for the experiments.
Cell-cycle analysis and flow cytometry
For cell cycle analysis, human PCa cells were digested using 0.25% Trypsin/EDTA solution and fixed with ice-cold 70% ethanol at 4 °C overnight, then stained with propidium iodide (PI) (keygentec, Nanjing, China) and measured by flow cytometry (FACS Calibur, Becton Dickinson). The THP-1 macrophages were collected, washed, and incubated for 30 min at 4 °C with florescence-conjugated antibodies. To facilitate intracellular staining, cells were fixed and permeabilized with a fixation/permeabilization solution kit (BD Cytofix/Cytoperm) and 1% paraformaldehyde (PFA). For evaluating the M2 polarization of macrophages, PE-Cyanine7 Monoclonal anti-human CD68 (eBioscienc, Invitrogen) and PE Monoclonal anti-human CD163 (eBioscienc, Invitrogen) were used. The results were analyzed using the FlowJo 10.7 software program. All assays were repeated three times.
Luciferase reporter assay
The sequences of circSMARCC1 or CCL20 3′-untranslated region containing the wild-type (Wt) or mutant (Mut) binding site of hsa-miR-1322 were designed and packaged into pEZX-MT06 vector (GeneCopoeia, Guangzhou, China). HEK-293 T cells were co-transfected with the corresponding plasmids and miR-1322 mimics/ miR-nc or miR-1322 inhibitors/inh-nc with Lipofectamine 3000 (Invitrogen, United States). The relative luciferase activity was measured utilizing Luc-PairTM Duo-Luciferase HS Assay Kit (GeneCopoeia, China). Each group was confirmed in triplicate.
RNA pulldown assay
To pull down the miRNA by circRNA, biotinylated-circSMARCC1 probe (5′-CATCTTCCTCATCACAGCAC-3′) was synthesized by RiboBio (Guangzhou, China), and the oligo probe (5′-TATGTTGTTGATTTGCTGGC-3′) was used as a control. CircRNA pulldown assay was carried out using PierceTM Magnetic RNA-Protein Pull-Down Kit (No: 20164, Thermo Fisher Scientific, United States). All procedures followed the manufacturer’s instructions. Then the final RNA was extracted by TRIzol (Invitrogen, United States) and analyzed by RT qPCR.
Western blot analysis
The RIPA lysis buffer containing protease inhibitors (# KGP250, KeyGEN BioTECH, Nanjing, China) was used to extract PCa cell protein following the standard protocol. Then, equal amounts of proteins in the cell lysates were separated by SDS/PAGE gels (4–12%, Bio-Rad) and electronically transferred onto polyvinylidene fluoride (PVDF, Millipore) membranes. The membranes were then blocked with 5% bovine serum albumin (BSA) and incubated overnight at 4 °C with the following specific primary antibodies: rabbit CCL20 antibody (#ab9829, Abcam) and rabbit CCR6 antibody (#ab110641, Abcam); EMT Antibody Sampler Kit (#9782), rabbit CD68(#97778) and CD163(#93498), rabbit CDK2 antibody (#18048), rabbit P27 Kip1 antibody (#3686), rabbit p21 Waf1/Cip1 antibody (#2947), rabbit Akt antibody (#4691), rabbit phospho-AktSer473 antibody (#4060), and rabbit phospho-AktThr308 antibody (#13038) were purchased from Cell Signaling Technology except for mouse β-actin antibody (#60008–1-Ig, Proteintech Group). Subsequently, horseradish peroxidase (HRP)-conjugated secondary antibody was used to incubate the samples for 1 h at room temperature. The bands were visualized using the enhanced chemiluminescence (ECL) detection system (Pierce Biotechnology, Rockford, IL, United States).
Immunohistochemistry (IHC) and immunofluorescence (IF)
Assays were performed as previously reported [
20]. In brief, for the IF experiment, cells were incubated with primary antibodies against CCL20 (1:200, #ab9829) (Abcam) at 4 °C overnight, then incubated with the fluorescent secondary antibody Alexa Fluor 594-conjugated goat anti-mouse IgG (Cell Signaling Technology) and imaged using a fluorescence microscope (DM5000B, Leica). For IHC, the paraffin sections were incubated with antibodies against CCL20 (1:200, #ab9829) (Abcam). Intensity scores were recorded as: 0 (no staining), 1 (weakly staining, light yellow), 2 (moderately staining, yellowish brown), and 3 (strongly staining, brown). In addition, we also detected the expression of ki67 (1:1000, #9449), CD68 (1:200, #97778), CD163 (1:400, #93498), CD206 (1:200, #24595) and CD31 (1:100, #77699) (Cell Signaling Technology) in PCa tissue or xenograft tissue. Images were observed under an Olympus multifunction microscope (Olympus BX51, Tokyo, Japan). All evaluations were performed by three independent senior pathologists using the same microscope.
Enzyme-linked immunosorbent assay (ELISA)
The concentration of CCL20 was detected with a commercial ELISA kit (Human MIP-3 alpha ELISA, RayBiotech) according to the manufacturer’s instructions. The concentration of cytokines in serum or cell lysates was quantitated by comparison of the ELISA data with a standard curve obtained with known concentrations of cytokines.
RNA-seq processing
RNA sequencing and sequence quality control of the DU145-vector and DU145-lv-circSMARCC1 cells were performed using the BGISEQ platform. The human genome reference was established from UCSC version GRCh38/hg38 chromosomes 1–22, X, Y, and mitochondrial DNA. Additional analysis, including a heatmap, gene set enrichment analysis (GSEA) was completed using the BGI Dr. Tom system. The Kyoto Encyclopedia of Genes [KEGG] was used for gene annotation of sequencing data.
Animal models
Xenograft models were created through injection of 5 × 106 DU145-vector, DU145-lv-circSMARCC1 cells (n = 7 per group), on the axillae of BALB/c male mice (4–5 weeks). The mice were obtained from the Animal Center of Southern Medical University, Guangzhou, China. All experimental animal procedures were authorized by the Nanfang Hospital Animal Ethics Committee of Southern Medical University (NFYY-2020-0132). The mice were raised under Specific Pathogen Free (SPF) conditions. Tumor size was measured every 5 days and the volume was calculated using the formula: volume = (length × width2)/2. To assess metastasis, 5 × 106 cells in 100 μL of PBS were injected via the tail veins of nude mice (n = 7 per group). After 7 weeks, the mice were anesthetized, and D-luciferin (#D-Luciferin, Apexbio) was injected intraperitoneally. The IVIS imaging system (Caliper Life Sciences) was used to visualize the luciferase signal. IHC and hematoxylin-eosin (H&E) staining were used to evaluate the characteristics of xenograft tumors and lung metastasis.
Statistical analysis
Data were analyzed using the student’s t test or non-parametric Mann–Whitney U Test and one-way analysis of variance (ANOVA) in GraphPad Prism version 8. The clinicopathological parameters in PCa cases were analyzed using Pearson’s chi-square test or Fisher’s exact chi-square test. The receiver operating characteristic (ROC) curve and Kaplan–Meier survival analyses were used to estimate the diagnostic and prognostic value. P < 0.05 was considered statistically significant. Data were presented as the means ± standard error of mean.
Discussion
The presence of circRNA in the cytoplasm of eukaryotic cells was first discovered by Hsu and Coca-Prados in 1979 [
30]. For decades afterwards, circRNAs were thought to be the product of shearing errors. With the development of RNA-seq technology and bioinformatics, a large number of circRNAs have entered the scientific community. The extensive expression and disease regulation mechanisms of circRNAs have made them functional biomarkers and therapeutic targets for a variety of diseases. However, the role of circRNA in the progression of PCa is still not well studied.
Here, we investigated circRNA expression profiles in plasma samples from four pairs of mPCa patients and control patients using circRNA sequencing. We focused on a significantly differentially expressed novel circRNA, named circSMARCC1, which was significantly up-regulated in PCa and correlated significantly with clinical Gleason scores and T stage. Furthermore, a series of in vitro and in vivo experiments demonstrated that circSMARCC1 promoted PCa cell proliferation, migration and invasion, while knockdown of circSMARCC1 expression had the opposite effect.
Many circRNAs contain potential miRNA response elements, which suggests that circRNAs can act as miRNA sponges, forming a circRNA-miRNA-mRNA axis to exert their biological roles [
31]. For instance, circASAP1 acts as a competing endogenous RNA (ceRNA), and binds to miRNAs as miRNA sponges in cells, which promotes hepatocellular carcinoma cell proliferation and invasion [
32]. Hsa_circ_0000326 acts as a miR-338-3p sponge to facilitate lung adenocarcinoma progression [
33]. In this study, we found that circSMARCC1 is highly expressed in PCa, especially in the cytoplasm of tumor cells. Therefore, we speculated that circSMARCC1 may also act as a miRNA sponge in PCa. Through bioinformatics analysis, luciferase assay and miRNA pulldown assays, we confirmed that circSMARCC1 serves as a sponge for miR-1322. Previous studies have reported that miR-1322 acts as tumor suppressors in some types of cancers, including hepatocellular carcinoma cells [
34] and lung adenocarcinoma [
35]. Consistent with that, our results confirmed that miR-1322 function as a tumor suppressor in PCa, which mediates the counteracts the oncogenic roles of circSMARCC1 on PCa proliferation and metastasis.
MiRNAs have been proved to play crucial roles in tumor cell differentiation, growth, and metastasis [
36], which alleviating the inhibitory effects of their target genes [
37]. In the present study, mRNA expression profiling, bioinformatics analysis and dual luciferase reporter gene assays revealed that CCL20 might act as direct targets of miR-1322. Further validation experiments demonstrated that circSMARCC1 may serve as an endogenous miRNA sponge to inhibit the expression of miR-1322 by binding to miR-1322 in PCa cells, resulting in alleviating the inhibitory effect of miR-1322 on CCL20 and ultimately promotes PCa cell proliferation and metastasis.
CCL20 (also known as macrophage inflammatory protein-3α, MIP-3α) is a pro-inflammatory chemokine. The CCL20 expression has been reported to be up-regulated in many cancers, such as hepatocellular carcinoma [
38] and pancreatic cancer [
39]. CCR6 is a specific receptor for CCL20, and CCR6 expression has also been shown on tumor cells. It was reported that tumor cells can stimulate their own proliferation and migration via CCL20-CCR6 in an autocrine manner [
40‐
42]. Previous studies found that CCL20 can be a novel predictive marker for taxanes response, and blockade of CCL20 or its downstream pathway might reverse the taxanes resistance in breast cancer patients [
43]. Whereas the specific role of CCL20 in PCa progression remains uncertain. Here, our study found that overexpression of circSMARCC1 increased the expression of CCL20, without changing the expression of its receptor CCR6 in PCa cells. Then, we confirmed that the up-regulation of CCL20 significantly accelerated the proliferation and metastasis of PCa cells, and that CCL20-neutralizing antibody reversed the above pro-tumorigenic effects. Therefore, our data indicates that circSMARCC1 sponges miR-1322 and up-regulates CCL20 to promote PCa progression in an autocrine manner.
Macrophages make up 30 to 50% of solid tumor-infiltrating immune cells, and they represent an important component of immunotherapy. In most cancers, high TAMs density is associated with poorer patient prognosis and treatment resistance. Macrophage depletion studies have shown great success in limiting tumor growth and metastatic spread and in restoring responsiveness to chemotherapy [
44,
45]. An alternative to targeting TAM is to inhibit their recruitment to the primary tumor. CCL2 is a chemokine that regulates monocyte and macrophage migration, and the CCL2/CCR2 axis has been shown to have multiple pro-tumor effects, ranging from mediating tumor growth and angiogenesis to recruiting and usurping host stromal cells to support tumor progression [
46]. In a phase 1 trial, the administration of a CCR2 inhibitor (PF-04136309) was well tolerated and showed promising anti-tumor activity in patients with advanced pancreatic cancer [
47]. CXCL12 is also a chemokine that promotes migration of macrophages across the endothelial barrier into the tumor environment [
48]. For this reason, inhibition of chemokines and their receptors signaling is a promising strategy for regulating macrophage infiltration and preventing tumor progression. In PCa, several studies reported that more TAM infiltration in the TME promoted PCa cell proliferation and migration and was associated with PSA failure or PCa progression after hormonal therapy [
49,
50]. In fact, increased infiltration of TAMs often predicts poor prognosis in patients with mPCa [
51,
52]. On the one hand, it has been reported that TAM-secreted CCL5 can promote PCa cell migration, invasion, and epithelial-mesenchymal transition (EMT) by activating β-catenin/STAT3 signaling [
53]. On the other hand, high levels of cytokines and chemokines, including TGF-β, IL10 and CCL2 are secreted by PCa, which may contribute to the recruitment of various immunosuppressive cells (myeloid-derived suppressor cells and regulatory T cells) and tumor-promoting TAMs [
54,
55], thereby promoting treatment tolerance and immune evasion by inhibiting CD4
+ helper T (Th1) and CD8
+ cytotoxic T cells [
56]. Overall, these data suggest that TAMs play a dual role as “tumor promoter” and “immunosuppressant”, and targeting TAMs may represent a potential therapeutic strategy for mPCa.
Some noncoding RNAs, including circRNAs, have been shown to have strong effects on TAM in a variety of tumors, such as circASAP1 [
32], circPPM1F [
57] has-circ-0005567 [
58]. However, whether circRNAs mediate TAM in PCa has not been elucidated. Our data show that circSMARCC1 is involved in TAM infiltration and polarization through the CCL20-CCR6 signaling pathway. The evidence is as follows: First, PCa and mouse xenograft tissues with high expression of circSMARCC1 exhibited increased numbers of M2 macrophages (CD68
+, CD163
+ and CD206
+) and upregulation of CCL20 expression. Second, conditioned medium (CM) from PCa cells overexpressing circSMARCC1 or the addition of CCL20 recombinant protein alone promoted the recruitment and polarization of M2 macrophages, which could be reversed by CCR6 neutralizing antibodies. Overall, these results reveal a critical role for circSMARCC1 in mediating macrophage infiltration and M2 polarization in PCa. In cancers with M2 macrophage-mediated TME, the CCL20-CCR6 axis may be a promising therapeutic target.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.