Abstract Background Accumulating evidence demonstrates that an increased tumor-associated macrophage abundance is often associated with poor prognosis in colorectal cancer (CRC). The mechanism underlying the effect of tumor-derived exosomes on M2 macrophage polarization remains elusive. Results The novel circular RNA circPOLQ exhibited significantly higher expression in CRC tissues than in paired normal tissues. Higher circPOLQ expression was associated with poorer prognosis in patients with CRC. In vitro and in vivo experiments showed that tumor-derived exosomal circPOLQ did not directly regulate CRC cell development but promoted CRC metastatic nodule formation by enhancing M2 macrophage polarization. circPOLQ activated the interleukin-10/signal transducer and activator of transcription 3 axis by targeting miR-379–3 p to promote M2 macrophage polarization. Conclusion circPOLQ can enter macrophages via CRC cell-derived exosomes and promote CRC metastatic nodule formation by enhancing M2 macrophage polarization. These findings reveal a tumor-derived exosome-mediated tumor–macrophage interaction potentially affecting CRC metastatic nodule formation. Keywords: immunity, macrophages, tumor microenvironment, colorectal cancer __________________________________________________________________ WHAT IS ALREADY KNOWN ON THIS TOPIC. * The propensity for distant metastasis in colorectal cancer (CRC) significantly contributes to the adverse prognosis and diminished long-term survival rates among patients with CRC. Emerging evidence underscores the pivotal role of exosomes within the tumor microenvironment (TME) in facilitating intercellular communication by transferring critical biomolecules such as RNAs and proteins. Tumor-associated macrophages (TAMs) are known to actively internalize these exosomes, thereby adopting an immunosuppressive stance that fosters tumor metastasis. However, the underlying mechanisms remain elusive. Therefore, identifying metastasis-associated driver genes and viable therapeutic interventions to inhibit tumor metastasis holds paramount importance for enhancing treatment outcomes and prognoses in patients with CRC. WHAT THIS STUDY ADDS * This study demonstrates that circRNAPOLQ exhibits high expression levels in CRC tissues, correlating with adverse prognostic outcomes. Both in vivo and in vitro models corroborate that circRNAPOLQ is conveyed to macrophages via exosomes derived from CRC cells. On internalization by macrophages, exosomal circRNAPOLQ targets miR-379–3 p, leading to the activation of the interleukin-10/signal transducer and activator of transcription 3 axis, driving the polarization towards an M2 phenotype, and ultimately facilitating the metastasis of CRC. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY * We explore a novel mode of TME formation where exosome-mediated communication between CRC cells and TAMs promotes CRC metastatic nodule formation. Additionally, this study presents evidence that exosome-derived circRNAPOLQ serves as a novel potential prognostic and therapeutic marker for CRC metastatic nodule formation. Introduction Tumor metastasis is an important cause of cancer-related death. Approximately 15–25% of colorectal cancers (CRCs) exhibit metastasis at the initial diagnosis.[44]1 Therefore, exploring tumor metastasis-related driver genes, potential mechanisms, and effective therapeutic targets to block tumor metastasis is highly important for improving the treatment efficacy and prognosis of patients with tumor metastasis. The tumor microenvironment (TME) plays a crucial role in tumor growth and metastasis.[45]2 Research has demonstrated that cells within the TME actively participate in facilitating tumor metastasis.[46]3 4 Tumor-associated macrophages (TAMs), the predominant immune cells within the TME, release a spectrum of cellular chemokines and cytokines to augment the proliferation, invasion, and metastasis of tumor cells.[47]5 6 TAMs are predominantly polarized toward the M2 phenotype and exhibit immunosuppressive and tumor progression-promoting effects.[48]7–9 Hence, the potential strategies for antitumor immunotherapy involve targeting M2-like TAMs and depleting them within the TME or repolarizing M2-like TAMs toward the M1-like phenotype. This approach directly enhances their cytotoxicity and indirectly stimulates cytotoxic T cells to eradicate tumor cells. Exosomes are membrane-bound extracellular vesicles that have a diameter ranging from 30 to 150 nm.[49]10 Exosomes contain various biomolecules, such as proteins, DNA, and RNA. When exosomes are released into the extracellular environment, they are taken up by recipient cells, allowing their contents to be transferred to recipient cells. This process facilitates intercellular communication.[50]11 Mounting evidence suggests that exosomes play a significant role in TME and tumor metastasis. This is achieved through the transfer of signal peptides, non-coding RNAs, or DNA to adjacent cells or tissues.[51]12–14 Circular RNAs (circRNAs) are RNA molecules that are covalently closed and exhibit a wide range of roles, remarkable stability, and evolutionary conservation.[52]15 CircRNAs possess a distinct covalently closed loop structure and exhibit a unique tertiary structure. They are predominantly localized in the cytoplasm and play multifaceted roles in various cellular processes.[53]16 Several studies have reported that circRNAs are involved in tumorigenesis and tumor progression.[54]17 18 Exosomal circRNAs are taken up by nearby or distant cells, thereby regulating the functions of the recipient cells.[55]19 However, the specific interactions and mechanisms linking circRNAs and TAMs are unknown. CircRNAs are anticipated to emerge as novel tumor markers and potential targets for therapeutic intervention. These findings hold promise for opening new avenues for tumor diagnosis and targeted therapy. In this study, we identified a novel circRNA, circPOLQ, known as circRNA4234. We evaluated the expression of circPOLQ in CRC through a series of clinical sample analyses in vitro and in vivo. We assessed whether CRC cell-derived exosomes induce M2 polarization of macrophages through the secretion of circPOLQ and explored the role and underlying mechanisms of circPOLQ in TAMs. These findings will help to identify new prognostic markers and provide a theoretical basis for developing potential tumor-targeted drugs. Materials and methods Patient tissue specimens and follow-up Six pairs of matched CRC tissues and matched adjacent normal tissues were obtained from surgically resected patients at the First Affiliated Hospital of Zhengzhou University. Our previous study confirmed the reliability of this cohort.[56]20 Human CRC and adjacent normal mucosa samples were collected from patients with CRC who underwent surgical procedures at the First Affiliated Hospital of Zhengzhou University between January 2017 and September 2021. 28 CRC tissues were surgically removed from the patients. These patients did not receive any anticancer therapy. The clinical data of these tissues are shown in [57]online supplemental table S1. Supplementary data [58]jitc-2023-008491supp001.pdf^ (1.5MB, pdf) Cell culture The HCT116, LoVo, HT-29, SW620, FHC (normal human colon epithelium cell line), THP-1 (human monocytic cell line), and 293T (human embryonic kidney 293 cell line) cell lines were acquired from the Type Culture Collection, Chinese Academy of Sciences (Shanghai, China). The human CRC cell line SW480 was obtained from the Biotherapy Center of The First Affiliated Hospital of Zhengzhou University. HCT116, LoVo, HT29, SW480, SW620, FHC, and 293 T cells were cultured in high-glucose DMEM (Gibco, USA) supplemented with 10% fetal bovine serum (Bio Industries, Cromwell, Connecticut, USA) at 37°C and 5% CO[2]. THP-1 cells were cultured in RPMI-1640 (Gibco, USA) supplemented with 10% fetal bovine serum (Bio Industries, Cromwell, Connecticut, USA). To induce THP-1 to differentiate into macrophages, THP-1 cells (1×10^6) were treated with 100 ng/mL phorbol 12-myristate 13-acetate (PMA; Sigma, USA) for 48 hours. Cell transfection siRNAs targeting circPOLQ and interleukin 10 (IL-10), miR-379–3 p mimics, and inhibitors were designed and synthesized by RiboBio (Guangzhou, China). circPOLQ overexpression plasmids were constructed using GV689 as the vector, and IL-10 overexpression plasmids were constructed using GV657 as the vector by GeneChem (Shanghai, China). The sequences of the siRNAs, mimics, and inhibitors used are listed in [59]online supplemental table S2. siRNAs and miR-379–3 p mimics and inhibitors were transiently transfected into cells using Lipofectamine 3000 (Invitrogen, Thermo Fisher Scientific, Carlsbad, California, USA) following the manufacturer’s instructions. All the sequences used in this study are summarized in [60]online supplemental tables S2–S5. Construction of stable cell lines The lentiviral circPOLQ overexpression plasmid, along with the pSPAX2 and pMD2G plasmids, was transfected into 293 T cells for viral packaging by GeneChem (Shanghai, China). The concentrated virus was then subjected to lentiviral transduction of HCT116 cells using LipoFilter Reagent. The cells were cultured with puromycin (2 µg/mL) to establish stable cell lines. Nuclear and cytoplasmic extraction RNA was extracted from the nuclear and cytoplasmic fractions using the PARIS Kit (Invitrogen, Thermo Fisher Scientific, Carlsbad, California, USA) following the manufacturer’s instructions. Then, the expression of circPOLQ was detected by realtime fluorescence quantitative polymerase chain reaction (RT-qPCR). Isolation and analysis of exosomes For exosome isolation, supernatants collected from 3-day cell cultures were first centrifuged at 500×g for 10 min to remove any cellular contamination. Subsequently, the supernatant was centrifuged at 12,000×g for 20 min to eliminate any potential apoptotic bodies or large cell debris. Furthermore, the exosomes were enriched by centrifugation at 100,000×g for 70 min at 4°C, and the resulting pellet was then washed with 200 µL of phosphate-buffered saline (PBS). The number and morphology (cup-shaped) of the exosomes were examined using a NanoSight NS300 microscope (Malvern Instruments, UK) and a Philips CM120 BioTwin transmission electron microscope (FEI Company, USA). RNA isolation, reverse transcription, and qPCR Total RNA was extracted using RNAiso Plus reagent (Takara, Dalian, China). Then, the RNA was reverse transcribed into complementary DNA (cDNA) using the PrimeScript RT Master Mix Kit (Takara, Dalian, China), followed by RT-qPCR using Hieff qPCR SYBR Green Master Mix (Low Rox Plus) (Yeasen, Shanghai, China) according to the manufacturer’s instructions. [61]Online supplemental table S3 lists all the primers used. All the data were normalized to the GAPDH data after analysis. Identification of circRNAs CRC cells were cultured with 100 ng/mL actinomycin D (Merck, Darmstadt, Germany) for 0 hour, 4 hours, 8 hours, 12 hours, and 16 hours. Total RNA was extracted, and the expression of circPOLQ and POLQ was detected via qPCR. In addition, convergent and divergent primers were designed, synthesized, and verified on the cDNA and genomic DNA (gDNA) of HCT116 and SW620 cells, respectively. [62]Online supplemental table S4 lists the divergent and convergent primers used. Western blot analysis Total protein was extracted using the radio immunoprecipitation assay (RIPA) buffer (Solarbio, Beijing, China) supplemented with phenylmethanesulfonyl fluoride (PMSF) and a phosphatase inhibitor (CWBIO, China), and protein was quantified using a BCA kit (Bicinchoninic Acid Assay Kit, Beyotime, China). The proteins were then separated by 8% or 10% SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Massachusetts, USA). The PVDF membranes were blocked with tris buffered saline with tween-20 (TBST) containing 5% (bovine serum albumin) BSA (Solarbio, Beijing, China). The membranes were incubated overnight at 4°C with the following primary antibodies: anti-CD206 (EPR22489-7), anti-TSG101 ([63]Ab125011), anti-CD9 ([64]Ab236630), and anti-STAT3 (Ab68153) from Abcam (Massachusetts, USA), anti-phospho-STAT3 (Tyr705) (#AF3293) from Affinity (Jiangsu, China), anti-ZO-1 (No. 21773–1-AP), anti-ZEB1 (No. 21544–1-AP), anti-α-SMA (No. 80008–1-RR), and anti-E-cadherin (No. 20874–1-AP) from Proteintech (Wuhan, China). The membrane was incubated with the secondary antibody for 1 hour at room temperature. The bands were visualized using a Chemiluminescence Kit (Millipore, USA). RNA in situ hybridization and fluorescence in situ hybridization For in situ hybridization (ISH), tissue sections were deparaffinized, incubated with a prehybridization solution, and subsequently hybridized with the circPOLQ probe (Servicebio, Wuhan, China). Then, the tissue sections were visualized using 3,3’-diaminobenzidine (DAB). The H-score was determined by Wuhan Sevier Biotechnology for fluorescence in situ hybridization (FISH). CRC cells were fixed with 4% paraformaldehyde in PBS for 20 min at room temperature. Hybridization of the Cy3-circPOLQ probe (GenePharma, Shanghai, China) was performed using a FISH kit (RiboBio, Guangzhou, China) following the manufacturer’s instructions. Images were acquired using a confocal laser scanning microscope (Zeiss, Jena, Germany). The probe sequences can be found in [65]online supplemental table S5. Immunohistochemistry Paraffin-embedded specimens were sectioned at a thickness of 3 µm. The sections were dewaxed and deparaffined in xylene and rehydrated in a graded alcohol series. The antigen retrieval process was performed by incubating the sections in Tris-EDTA buffer for 30 min. Subsequently, the primary antibodies against CD206 (No. 60143–1-Ig, Proteintech, Wuhan, China), CD163 (No. 16646–1-AP, Proteintech, Wuhan, China), N-cadherin (No. 22018–1-AP, Proteintech, Wuhan, China), VIM (No. 10366–1-AP, Proteintech, Wuhan, China) and their respective secondary antibodies were used for staining. Counterstaining with hematoxylin was then performed, followed by dehydration and mounting. Images of each section were taken in three randomly selected three fields of view under a microscope. The image analysis software ImageJ and Image-Pro Plus were used for data analysis. Fluorescent labeling and tracking of CRC exosomes into macrophages Exosomes derived from the CRC cell lines HCT116 and SW620 were labeled with PKH67 (green PKH membrane mini labeling kit, Sigma, Germany). Then, the above-pretreated exosomes were coincubated with THP-1-derived macrophages in 24-well plates for 0 hour, 6 hours, or 12 hours. The internalization of exosomes was examined using an LSM780 confocal laser scanning microscope (Zeiss, Germany). The nuclei of THP-1-derived macrophages were stained with 4’,6-diamidino-2-phenylindole (Yeasen, Shanghai, China), and the cytoskeleton was stained with a microfilament protein (F-actin; GenePharma, Shanghai, China). RNA pull-down assay The biotinylated miR-379–3 p probe and negative control probe were designed and synthesized by GenePharma (Shanghai, China). The probe (30 µg) and 40 µL of streptomycin-coated magnetic beads were incubated for 30 min at room temperature. Cells were lysed in RNA binding protein immunoprecipitation assay (RIP) buffer supplemented with a mixture of ribonuclease inhibitors and protease inhibitors. Cell extracts and probe bead complexes were mixed and incubated for 4 hours at 4°C with rotation. After rinsing with RIP wash buffer, the plates were digested with proteinase K digestion buffer at 900 rpm for 30 min at 55°C. The final RNA was extracted and reverse transcribed with an RNA Clean & Concentrator−25 kit (Zymo Research, USA) and a RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Carlsbad, California, USA) for reverse transcription. The abundance of circPOLQ and IL-10 was then analyzed via qPCR. Luciferase reporter assay To generate the luciferase construct, the circPOLQ and IL-10 genes containing the miR-379–3 p binding site were synthesized by GeneChem (China) and inserted into the GV272 vector. Next, PMA-treated THP-1 cells were treated with IL-10-3′-UTR-Luc firefly luciferase constructs (wild-type or mutant) using Lipofectamine 3000 reagent (Invitrogen, USA). The Luc firefly luciferase construct (wild-type or mutant) was co-transfected with an miR-379–3 p mimic. Finally, 48 hours after transfection, the cell lysates were collected, and the firefly/Renilla luminescence was quantified using a Veritas 96-well microplate luminometer (Promega, USA) following the procedure of the Dual-Luciferase Reporter Assay Kit procedure (Biyuntian, China). Neutase activity was assessed using a substrate dispenser from Promega (USA). Renilla luciferase activity was normalized to firefly luciferase activity. Transwell assay Validation of CRC cell migration and invasion using the Transwell assay. This experiment was performed using a Transwell 24-well chamber (Corning, USA) with an 8.0 µm pore size polycarbonate membrane. For the invasion assay, additional diluted Matrigel (Corning, USA) was added to the chamber of a 24-well Transwell plate. To summarize, 5×10^4 tumor cells suspended in 200 µL of serum-free medium were seeded in the upper chamber, while the lower chamber contained 600 µL of medium supplemented with 10% fetal bovine serum (FBS). After 48 hours of incubation at 37°C in a 5% CO[2] humidified atmosphere, the migrated or invaded cells in the bottom chamber were fixed with paraformaldehyde and stained with 0.1% crystal violet. Three fields of view were randomly selected for imaging under a microscope. Photoshop software was used for statistical analysis. Flow cytometry analysis The cells were digested and collected by pancreatic enzymes, washed with PBS, and stained on ice with Fixable Viability Stain 700 (564997, BD Pharmingen, USA) for 10 min, followed by centrifugation in PBS to terminate the staining and incubation in a room temperature Fc blocker (422301, BioLegend, USA) for 15 min. The surface marker antibodies CD11b (301404, BioLegend, USA), CD206 (321105, BioLegend, USA), CD301 (354705, BioLegend, USA), and CD200R (329311, BioLegend, USA) were stained on ice in the dark for 30 min. Afterward, the cells were washed with PBS. The samples were analyzed using a DxFLEX flow cytometer (Beckman Coulter, USA) and the data were analyzed using FlowJo V.10.8.1. Animal models To investigate the functional role of circPOLQ in CRC-derived exosomes in vivo, we developed models of liver and lung metastasis using 6-week-old, pathogen-free, female NOD/SCID gamma (NSG) mice with severe immunodeficiency. These mice, procured from the Vital River Laboratory in Beijing, were maintained in a specific pathogen-free environment within our animal research facility. We isolated exosomes secreted by HCT116 cells that stably overexpressed circPOLQ and incubated them with mature macrophages induced by PMA for 48 hours. The SW480 cell line, stably transfected with luciferase, was subsequently co-cultured with the supernatant of the above macrophages. We harvested these SW480 cells by injecting 2×10^6 cells into the tail vein (100 µL) or the spleen (50 µL) of the mice. The survival duration of the mice was meticulously monitored. After 30 days, the mice were anesthetized and subjected to imaging using the IVIS Illumina system (Caliper Life Sciences, USA). Subsequently, the mice were euthanized, and nodules from the lungs and liver were extracted for paraffin embedding, H&E staining, and immunohistochemical analysis. An orthotopic CRC model was used.[66]21–23 In brief, HCT116 cells stably overexpressing circPOLQ (1×10^6) and THP-1-derived macrophages were first mixed in equal amounts.[67]24 After NSG mice were anesthetized, a 1.5 cm incision was made around the midline of the abdomen, and the cecum was exteriorized and kept moist with PBS. The mixed cells (50 µL) were injected into the cecal wall using a 32-G needle, and the injection site was covered with a cotton swab for 3 min to prevent cell leakage. The cecum was returned to the abdominal cavity, and the abdominal wall and skin were carefully sutured with attention to aseptic disinfection. Mice were regularly monitored for orthotopic tumor formation and distant metastasis using the IVIS Illumina system. Six weeks later, mice were euthanized, and in situ tumors located at the cecum, liver, and lung were collected for pathological analysis. Statistical analysis All the data were analyzed using Prism V.10.0 (GraphPad, San Diego, California, USA) and are presented as the mean±SEM. Χ^2 tests were conducted using SPSS Statistics V.21 (IBM, Chicago, Illinois, USA). Student’s t-test was used to evaluate significant differences between two independent groups. Pearson’s correlation coefficient (r) and two-tailed p values were used for correlation analysis. Survival curves were assessed using the log-rank (Mantel-Cox) test. A p value<0.05 was considered to indicate statistical significance. All experiments were replicated at least three times. All the authors had access to the study data and had reviewed and approved the final manuscript. Results Elevated circPOLQ expression is positively associated with poor prognosis in patients with metastatic CRC To identify the potential circRNAs associated with CRC metastasis, we initially conducted high-throughput sequencing on six sets of CRC tissues and their corresponding paracancerous normal tissues. Subsequently, we acquired the expression profiles of the dysregulated circRNAs. This expression profile was also applied in our previous study. TAMs are the most prevalent immune cells in the TME. Therefore, as an essential component, TAMs have been reported to strongly affect tumor progression. First, we identified immune-related circRNAs by correlation analysis and generated a heatmap. circPOLQ was among the top ranked candidates and was significantly upregulated in CRC tissues ([68]figure 1A–C). circRNAPOLQ is derived from chromosome 3:121203887–121217517 and consists of five adjacent exons in the POLQ gene ([69]online supplemental figure 1A). To validate the circular structure of circPOLQ, we designed convergent and divergent primers targeting circPOLQ and found that it could be amplified from cDNA but not from gDNA ([70]figure 1D and [71]online supplemental table S4). Additionally, actinomycin D treatment of HCT116 and SW620 cells less potently reduced the half-life of circPOLQ than did treatment with linear POLQ ([72]figure 1E and [73]online supplemental figure 1B). Taken together, these findings suggest that circPOLQ has a stable circRNA ring structure. Furthermore, we discovered by RNA nuclear/cytoplasmic fractionation and FISH revealed that circPOLQ was localized mainly in the cytoplasm in HCT116 and SW620 cells, with a low proportion of circPOLQ localized in the nucleus ([74]figure 1F,G and [75]online supplemental figure 1C). Figure 1. [76]Figure 1 [77]Open in a new tab High expression of circPOLQ is positively related to poor prognosis of patients with CRC. (A) Flow chart for screening circRNAs in six pairs of CRC tissues. (B) Heatmap of deregulated circRNAs associated with immune mRNAs. (C) Expression of circPOLQ in the sequencing data of six pairs of CRC tissues and corresponding normal tissues. (D) Agarose gel analysis of PCR products using circPOLQ divergent primer and convergent primer. (E) qPCR analysis of circPOLQ and POLQ expression after culturing HCT116 cells with actinomycin D at 0 hour, 4 hours, 8 hours, 12 hours, and 16 hours. (F) Nuclear and cytoplasmic fractionation analysis assessed circPOLQ expression in HCT116 cytoplasmic fractionation. U6 and GAPDH were used as internal references. (G)