circ-UBR5 in hypoxia-induced exosomes may mediate lung adenocarcinoma metastasis via the targeting of HNRNPR

Introduction

Lung cancer is the most prevalent malignancy globally and remains the leading cause of cancer-related mortality, accounting for 1.8 million deaths in 2022 (1). Non-small cell lung cancer (NSCLC) is the most common subtype, and lung adenocarcinoma (LUAD) accounts for approximately 40% of all NSCLC cases (2). LUAD is characterized by high aggressiveness, resistance to therapy, and early metastatic potential; even after radical resection combined with adjuvant therapy, nearly 50% of patients still experience disease recurrence and distant metastasis (3). Consequently, LUAD is associated with poor prognosis, and the underlying pathological mechanisms driving its progression and metastasis require further investigation.

The tumor microenvironment (TME) plays a pivotal role in tumor cell growth, invasion, and distant metastasis (4-6), with hypoxia being a hallmark feature of the TME. Solid tumors frequently contain hypoxic regions due to the high oxygen consumption of rapidly proliferating tumor cells and infiltrating immune cells, as well as the structural and functional abnormalities in the tumor vasculature that lead to insufficient oxygen delivery (7). Tumor hypoxia not only enhances cancer cell resistance to chemotherapy and radiotherapy but also promotes an invasive phenotype, which ultimately contributes to a poor prognosis (8,9). For lung cancer specifically, there is a growing body of evidence indicating that hypoxia accelerates tumor growth and facilitates metastatic dissemination (10,11). Notably, oxygen availability within tumors is spatially heterogeneous, with oxygen concentration levels decreasing as the distance from functional blood vessels increases, resulting in intratumoral regions with relative normoxia and hypoxia (12). Hypoxic tumor cells can further secrete soluble factors or release extracellular vesicles to transmit metastasis-associated signals to adjacent normoxic cancer cells or stromal cells, thereby driving tumor progression and metastatic spread (13).

In recent years, the role of exosomes in tumor progression has garnered increased attention (14,15). As a major subclass of extracellular vehicles (EVs) that encompass a variety biomolecules (including proteins, lipids, and nucleic acids), exosomes serve as key mediators of intercellular communication in the TME (16-18). Tumor-derived exosomes have been shown to remodel the TME, promote metastatic colonization, and induce premetastatic niche formation (19), with non-coding RNAs (ncRNAs) being the critical functional cargo in this process (20). Circular RNAs (circRNAs), a class of endogenous ncRNAs characterized by covalently closed-loop structures, are widely expressed in eukaryotic cells (21). Owing to their resistance to exonuclease-mediated degradation, circRNAs are highly enriched in exosomes and act as stable mediators of intercellular communication within tumors, regulating processes such as immune evasion, angiogenesis, drug resistance, cell proliferation, and metastasis (22,23). For instance, cisplatin-resistant NSCLC cells secrete exosomes containing circVMP1, which activates the miR-524-5p-METTL3/SOX2 signaling axis in recipient sensitive cells to promote NSCLC progression and chemoresistance (24). This phenomenon underscores the importance of circRNA-mediated intracellular communication in lung cancer. Moreover, hypoxic tumor-derived exosomal circRNAs have been reported to mediate the crosstalk between tumor cells and tumor-associated macrophages, thereby facilitating lung cancer progression and metastasis (25,26). However, whether exosomal circRNAs derived from hypoxic LUAD cells can directly act on adjacent normoxic LUAD cells to promote metastatic phenotypes remains unclear.

Metastasis is a complex, multistep biological process involving chromosomal instability, dysregulated protein-protein interactions, and extensive TME remodeling, driven by multiple overlapping molecular mechanisms (27-29). Ribonucleoproteins (RNPs) are key regulators of RNA metabolism, participating in RNA processing, intracellular transport, and functional modulation. Heterogeneous nuclear ribonucleoproteins (hnRNPs) are a family of RNPs that bind to nascent RNA transcripts and regulate aspects of (pre-)messenger RNA (mRNA) processing (e.g., splicing, polyadenylation, and nuclear export) (30,31), thereby influencing the gene expression, patterns, and metabolic pathways in various cancers, including cholangiocarcinoma, colorectal cancer, hepatocellular carcinoma, and lung cancer (32-36). Heterogeneous nuclear ribonucleoprotein R (hnRNPR), a member of the hnRNP family, has been reported to mediate epithelial-mesenchymal transition in hepatocellular carcinoma by binding to UPF3B pre-mRNA via its RNA recognition motif 2 domain. Additionally, HNRNPR promotes the dephosphorylation of LATS1 and nuclear translocation of YAP1, leading to the suppression of the Hippo/YAP signaling pathway and enhanced tumor invasion and migration (37). Furthermore, HNRNPR has been implicated in the progression and metastasis of gastric cancer and neuroblastoma (38,39), but its expression pattern, functional role, and regulatory mechanisms in LUAD have not been characterized.

In this study, we explored the role of circ-UBR5 in LUAD. Specifically, we investigated the intracellular communication between hypoxic and normoxic LUAD cells, with a specific focus on the role of hypoxia-induced exosomal circRNAs in modulating LUAD progression. We found that hypoxic LUAD cells secreted exosomes enriched in circ-UBR5 and that the delivery of this exosomal circ-UBR5 to recipient normoxic LUAD cells upregulated HNRNPR expression, thereby promoting LUAD metastasis. Furthermore, in pathological LUAD samples, circ-UBR5 expression was positively correlated with HNRNPR expression, with the expression of both these molecules being associated with advanced disease stage and poor prognosis. Collectively, these findings suggest that the exosomal circ-UBR5-HNRNPR axis may serve as a potential therapeutic target for inhibiting LUAD metastasis and improving clinical outcomes. We present this article in accordance with the MDAR reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1413/rc).

Methods

Cell culture

Human lung cancer cell lines A549 and PC-9 were purchased from the Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China). A549 cells were cultured in F-12K medium (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), and PC-9 cells were maintained in RPMI 1640 medium (Gibco). Both basal media were supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin-streptomycin. Cells were incubated in a humidified atmosphere at 37 ℃ with a 5% CO₂ atmosphere. For hypoxia experiments, cells were cultured under 1% oxygen conditions.

Isolation of exosomes from medium and plasma

Exosomes were isolated through differential centrifugation. Briefly, cell culture supernatants or plasma samples were first centrifuged at 300 ×g and then at 2,000 ×g to remove cells and cellular debris. The resulting supernatant was centrifuged at 10,000 ×g to eliminate shedding vesicles and other larger-sized vesicles. Finally, the supernatant was ultracentrifuged at 100,000 ×g for 70 minutes. All centrifugation steps were performed at 4 ℃. The exosome pellet was harvested and then resuspended in phosphate-buffered saline (PBS) for subsequent experiments.

Transmission electron microscopy (TEM) assay

For TEM analysis, exosome pellets were resuspended in 2.5% glutaraldehyde and fixed overnight at 4 ℃. Samples were then postfixed in 1% osmium tetroxide at room temperature (RT), embedded in 10% gelatin, and refixed in glutaraldehyde at 4 ℃. The gelatin-embedded samples were cut into small blocks (<1 mm³). Following dehydration through a graded ethanol series (30%, 50%, 70%, 90%, 95%, and 100%, with three changes in each step), samples were infiltrated with propylene oxide and gradually embedded in Quetol-812 epoxy resin (Nisshin EM, Japan). Ultrathin sections (100 nm) were prepared with a UC6 ultramicrotome (Leica, Wetzlar, Germany) stained with uranyl acetate and lead citrate for 5 min each at RT, and examined under a Tecnai T20 transmission electron microscope (FEI Company, Thermo Fisher Scientific) operating at 120 kV.

PKH26 staining for exosomes

Exosome membranes were labeled with PKH26 red fluorescent cell linker kits (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s protocol. Briefly, exosomes were resuspended in 100 µL of diluent C. A dye solution (4×10⁻⁶ M) was prepared through the addition of 0.4 µL of PKH26 ethanolic dye solution to 100 µL of diluent C. The exosome suspension (100 µL) was mixed with the dye solution (100 µL) through gentle pipetting. After 1–5 min of incubation with periodic mixing, the staining reaction was terminated by the addition of 200 µL of serum followed by 1 minute of incubation. Stained exosomes were washed twice with 1× PBS and resuspended in fresh PBS in sterile conical polypropylene tubes.

Establishment of stable cell lines

Lentiviral vectors encoding circ-UBR5 overexpression, HNRNPR overexpression, HNRNPR short hairpin RNA (shRNA) for knockdown, and their corresponding control vectors were purchased from Genechem (Shanghai, China). For lentivirus production, the target lentiviral vectors were cotransfected with packaging plasmids psPASX2 and pMD2.G at the recommended ratio into 293T cells (at 90% confluence) via Lenti-Pro Transfection Reagent (Shanghai Genechem, China). Viral supernatants were collected 48 hours after transfection and added to A549 or PC-9 cells. After 8 hours of incubation with the virus-containing medium, the medium was replaced with fresh medium containing 10% FBS. Stable cell lines were established through the selection of infected cells with puromycin starting 72 hours after infection to enrich for stably transduced cells.

RNA isolation and quantitative reverse transcription polymerase chain reaction

Total RNA was extracted from cultured cells and tissues with TRIzol reagent (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s protocols. Complementary DNA (cDNA) was reverse-transcribed from 500 ng of total RNA with the PrimeScript Reverse Transcriptase Reagent Kit (Takara Bio, Kusatsu, Japan). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed with SYBR Premix Ex TaqII (Takara Bio) on a Light Cycler 480 II real-time PCR system (Roche, Basel, Switzerland). GAPDH served as the internal reference gene, and the relative gene expression levels were calculated with the 2^−ΔΔCt method. Primers were designed and synthesized by Sangon Biotech (Shanghai, China), with the following sequences: 5'-AGGCAATGAATCAGCAGACAA-3' (has_circ_0085207/circ-UBR5, forward primer), 5'-GCAAGCCACAAGTTCGACACT-3' (has_circ_0085207/circ-UBR5, reverse primer); 5'-ATGTTCTTGACAGAGTGTCAGGTTC-3' (hsa_circ_0053754, forward primer), 5'-CTGGGGTTCGTTCACAATCTC-3' (hsa_circ_0053754, reverse primer); 5'-TCATTCACTTGAGGAGTGTCTGGTA-3' (hsa_circ_0005276, forward primer), 5'-CCTGGATACCATTTAGCATGTTGTT-3' (hsa_circ_0005276, reverse primer); 5'-ACGCCATCGAATCCGGAA-3' (hsa_circ_0000416, forward primer), and 5'-GGTCTAGAAACCAAATGTGAAGATG-3' (hsa_circ_0000416, reverse primer).

Cell apoptosis flow cytometry

A549 and PC-9 were seeded in six-well plates at an appropriate initial density and incubated under hypoxic (1% O₂) or normoxic (21% O₂) conditions. After 12 hours, cells were harvested and stained with Annexin V and propidium iodide (PI) (Beyotime Biotechnology, Shanghai, China) for 15 min in the dark. Stained cells were analyzed with a flow cytometer (Beckman Coulter, Brea, CA, USA). The percentage of early apoptotic cells was defined as the proportion of Annexin V-positive cells relative to the total cell proportion.

Cell proliferation assay

Cellular proliferation was evaluated with Cell Counting Kit-8 (CCK-8; Beyotime Biotechnology). Approximately 3×10³ cells were seeded in 96-well plates and incubated for 24, 48, 72, and 96 hours in hypoxic or normoxic conditions. Ten microliters of CCK-8 solution was added to each well, and the plates were incubated at 37 ℃ for 1 hour. Absorbance at 450 nm was measured with an ELX-800 spectrometer reader (Bio-Tek Instruments, Winooski, VT, USA).

Western blot analysis

Protein expression was assessed via Western blotting, with GAPDH used as the internal reference. Cells and tissues were lysed in radioimmunoprecipitation (RIPA) buffer (Solarbio, Beijing, China) supplemented with fresh protease inhibitor cocktail (Roche). Total proteins were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes (MilliporeSigma, Burlington, MA, USA). Membranes were blocked with 5% bovine serum albumin (BSA) at RT for 1 hour and then incubated overnight at 4 ℃ with the following primary antibodies: anti-CD63 (1:1,000; BD Bioscience, Franklin Lakes, NJ, USA), anti-CD9 (1:1,000; MilliporeSigma), anti-CD81 (1:1,800; Medix Biochemica, Espoo, Finland), anti-HNRNPR (1:500; Proteintech, Rosemont, IL, USA), and anti-GAPDH (1:3,000; Proteintech). After washing, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at RT for 1 hour. Protein bands were visualized with an enhanced chemiluminescence (ECL) kit (MilliporeSigma) according to the manufacturer’s protocol.

ncRNA sequencing and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis

Total RNA was extracted from serum and plasma samples with a RNeasy Plus Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. RNA integrity was assessed with a 2100 Bioanalyzer system (Agilent, Santa Clara, CA, USA). The extracted RNA was amplified and labeled with Cy3 with a Low Input Quick Amp Labeling Kit (Agilent), and the labeled RNA was purified via the RNeasy Mini Kit (Qiagen). For hybridization, 1.65 µg of Cy3-labeled RNA was applied to each microarray slide using the Gene Expression Hybridization Kit (Agilent) and incubated at 65 ℃ for 17 hours. After washing with the Gene Expression Wash Buffer Kit (Agilent), the slides were scanned using the SureScan DX microarray scanner (Agilent) at a 3-µm resolution. Raw data were extracted using Feature Extraction Software version 10.7 (Agilent) and subsequently processed for downstream analyses. KEGG pathway enrichment analysis of coexpressed genes was performed using the R package “clusterProfiler”.

Bioinformatics analysis and selection of candidate circRNAs in lung cancer

Raw microarray data were normalized using the quantile method implemented in the “limma” package in R. CircRNA expression profiles were extracted for subsequent analysis. Differentially expressed circRNAs between lung cancer patients and control samples were identified using “limma”, with an adjusted P value <0.05 and |log₂fold change (FC)| above the predefined threshold. Differentially expressed circRNAs were ranked by log₂FC and P values, and the top candidates were selected for further validation. Based on expression abundance, consistency across samples, and relevance to lung cancer-related pathways, circ-UBR5 was identified as one of the most significantly upregulated circRNAs and selected for subsequent experimental validation.

Tissue samples and tissue microarray

Formalin-fixed paraffin-embedded (FFPE) LUAD tissue samples were collected from patients who underwent surgical resection at Shanghai Chest Hospital (Shanghai, China) between January 2014 and January 2020. A tissue microarray containing 70 LUAD samples was constructed with 0.6-mm-diameter core being punched from the FFPE blocks, including both tumor tissues and adjacent normal tissues. All samples were reviewed by experienced pathologists and handled in accordance with the Declaration of Helsinki and its subsequent amendments. All related procedures were conducted with approval from the Ethics Committee of Shanghai Chest Hospital (No. KS22018) and informed consent was taken from all the patients.

In situ hybridization and immunofluorescence

In situ hybridization was performed with a commercial kit (Beyotime). Cells were fixed in 4% paraformaldehyde for 10 min and washed with PBS. After digestion with 10 µg/mL proteinase K at RT for 10 min, samples were treated with 0.5-M HCl and acetylated with a freshly prepared acetylation solution. Fluorescent RNA probes were synthesized based on target sequences. Prehybridization was performed with hybridization buffer containing yeast RNA. Denatured probe-containing hybridization solution was applied to samples and incubated overnight in a humidified chamber. Posthybridization, stringent washes were performed with preheated washing buffers under light protection. For tissue microarray sections, immunofluorescence staining was performed via incubation with anti-HNRNPR antibody (1:500; Proteintech) and then with fluorescent secondary antibody. Nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI). Images were acquired with a fluorescence microscope. Fluorescence intensities of circ-UBR5 and HNRNPR were quantified with ImageJ software (US National Institutes of Health, Bethesda, MD, USA). The median fluorescence intensity was used as the cutoff: values above the median were defined as high expression and those below as low expression.

Cell migration assay

Cell migration capacity was evaluated via wound-healing assay. A549 and PC-9 cells were seeded in six-well plates. After 24 hours, a 20-µL pipette tip was used to create two linear cell-free regions in each well. Medium containing 2% FBS was added, and cells were incubated under standard conditions. Images were captured at 0, 12, 24, 48, and 72 hours using inverted light microscope to dynamically monitor gap closure for assessment of wound closure. The migration area is calculated by the ratio of the scratched area to the field of view area by ImageJ.

Cell invasion assay

Cell invasion was analyzed using Transwell chambers (Corning, Corning, NY, USA) in a 24-well plate. The lower chamber was filled with 600 µL of medium containing 10% FBS. The upper chamber was coated with precooled Matrigel (Corning), and 5×10⁴ cells in 100 µL of serum-free medium were seeded into the upper chamber. After 24 hours of incubation, noninvading cells on the upper surface of the membrane were removed. Invading cells on the lower surface were fixed in 4% formaldehyde, stained with 0.1% crystal violet, air-dried, and imaged under a light microscope. Cells were counted in three randomly selected fields per membrane, and the average number of invading cells was calculated by ImageJ.

RNA pulldown assay and mass spectrometry

A549 and PC-9 were washed with ice-cold PBS and lysed in 500 µL of lysis buffer [20 mM of Tris-HCl at pH 7.0, 150 mM of NaCl, 0.5% NP-40, and 5 mM of ethylenediaminetetraacetic acid (EDTA)] supplemented with 1 mM of dithiothreitol (DTT), 1 mM of phenylmethylsulfonyl fluoride (PMSF), and 0.4 U/µL of RNase inhibitor. Lysates were incubated with 3 µg of biotinylated DNA oligonucleotide probes targeting endogenous or ectopically expressed circ-UBR5 at 4 ℃ for 2 hours. Subsequently, 50 µL of Dynabeads MyOne streptavidin C1 magnetic beads (Invitrogen) were added to each reaction and incubated at 4 ℃ for another 2 hours. The beads were washed three times with lysis buffer, and bound proteins were analyzed via Western blotting. For mass spectrometry, pulled-down proteins were separated by 12% SDS-PAGE and stained with Coomassie brilliant blue (Tiangen Biotech, Beijing, China). Gel bands were excised and subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) for protein identification.

The probe sequences were as follows: circ-UBR5-biotin, UUCGACACUCAAAAAUAUCUUUGAUUGGUGUCAUCAUCA; and circ-NC-biotin, UAUCACGUAGCCGUUGCAUUUGCCGUAGCCCUGUGGGCC.

RNA immunoprecipitation (RIP)

A549 and PC-9 cells were washed in ice-cold PBS and lysed in lysis buffer (20 mM of Tris-HCl at pH 7.0, 150 mM of NaCl, 0.5% NP-40, 5 mM EDTA) supplemented with 1 mM of DTT, 1 mM of PMSF, and 0.4 U/µL of RNase inhibitor. Lysates were incubated with 5 µg of primary antibody at 4 ℃ for 2 hours, which was followed by the addition of 50 µL of Protein A/G PLUS-Agarose (Santa Cruz Biotechnology, Dalla, TX, USA) and incubation at 4 ℃ for 4 hours. The beads were washed with PBS, and bound RNA was extracted with TRIzol reagent (Invitrogen). Isolated RNA was subject to qRT-PCR to detect the target sequences of specific primers.

Statistical analysis

Data processing and statistical analyses were performed with GraphPad Prism 9 (Dotmatics, Boston, MA, USA) and R version 4.2.1 software. Results are presented as the mean ± standard deviation (SD) from at least three independent experiments, each performed in triplicate. Continuous variables are expressed as the mean ± SD or median with the interquartile range (IQR). Statistical significance was determined via the Student t-test, Wilcoxon rank-sum test, Mann-Whitney test, and one-way analysis of variance as appropriate. Pearson correlation analysis and simple linear regression were used to determine the coefficient of determination (R²). Survival differences were analyzed via Kaplan-Meier analysis with the log-rank test. Categorical variables were compared with the chi-squared test. Differences were considered statistically significant at P<0.05.

Results

Hypoxia influenced tumor heterogeneity and drove exosome-mediated lung cancer metastasis

We observed that LUAD cells (A549 and PC-9) exhibited significantly increased proliferation (Figure S1A) and reduced basal apoptosis under hypoxic conditions (1% O₂) (Figure S1B), consistent with previous reports (40,41). Given that vascular abnormalities in solid tumors often result in heterogeneous oxygen distribution, we further investigated the influence of hypoxic LUAD cells on their normoxic counterparts. Conditioned media collected after 72 hours of hypoxic culture markedly enhanced the proliferation, migration, and invasion of normoxic LUAD cells compared with normoxic conditioned media (Figure 1A-1C), suggesting that hypoxic LUAD cells may secrete soluble factors that promote the metastatic potential of normoxic cells.

Figure 1 Hypoxic LUAD cells promoted proliferation and motility of normoxic LUAD cells via exosomes. (A) Cell viability and proliferation of normoxic LUAD cells treated with NM or HM were measured with the CCK-8 assay. (B) Transwell invasion assay was performed to assess the invasive capacity of LUAD cells treated with NM or HM. Invaded cells were stained with crystal violet for cell number counting. (C) Scratch wound-healing assay showing the migration ability of LUAD cells under different conditioned media treatments, observed using an inverted light microscope and calculating the migration area. (D) Western blot analysis of exosome markers in conditioned media from normoxic and hypoxic LUAD cells. (E) Exosomes were isolated from NM and HM, and their morphology was examined by TEM and NTA. (F) Uptake of PKH26-labeled exosomes by normoxic LUAD cells observed under fluorescence microscopy. (G) Cell viability and proliferation of LUAD cells treated with N- or H-exo cells as assessed by CCK-8 assay. (H) Transwell invasion assay of LUAD cells following treatment with N- or H-exo. Invaded cells were stained with crystal violet for cell number counting. (I) Scratch wound-healing assay showing the migration of normaxic LUAD cells treated with N- or H-exo at 0 and 72 h, observed using an inverted light microscope and calculating the migration area. *, P<0.05; **, P<0.01. CCK-8, Cell Counting Kit-8; DAPI, 4’,6-diamidino-2-phenylindole; H-exo, hypoxic exosomes; HM, hypoxic media; LUAD, lung adenocarcinoma; N-exo, normoxic exosomes; NM, normoxic media; NTA, nanoparticle tracking analysis; OD, optical density; TEM, transmission electron microscopy.

Since EVs are critical mediators of intercellular communication, we isolated exosomes from the conditioned media of LUAD cells cultured under hypoxic or normoxic conditions (Figure 1D). TEM and nanoparticle tracking analysis confirmed the presence of exosomes and revealed distinct morphological differences between the two groups (Figure 1E). PKH26 labeling demonstrated that these exosomes could be internalized by LUAD cells, with some even localized in the nucleus (Figure 1F). When normoxic LUAD cells were cocultured with exosomes derived from hypoxic or normoxic cells, those treated with hypoxia-derived exosomes displayed significantly enhanced proliferation, migration, and invasion (Figure 1G-1I). Importantly, these effects were abrogated by the exosome secretion inhibitor GW4869 (Figure S2). Collectively, these findings indicate that hypoxic LUAD cells can promote metastatic behavior in normoxic LUAD cells through exosome-mediated signaling.

Hypoxic exosome-derived circ-UBR5 acted as a key prometastatic effector

ncRNAs play crucial biological roles in tumor initiation and progression and have emerged as promising therapeutic targets (42). Therefore, we focused on the potential role of certain ncRNA signals in exosomes in LUAD metastasis. Through ncRNA sequencing, we examined the exosomal RNA from culture media under hypoxic and normoxic conditions and identified 342 upregulated genes and 585 downregulated genes (Figure 2A). KEGG pathway analysis revealed that these differentially expressed genes were mainly enriched in pathways closely related to lung cancer metastasis, including NSCLC, focal adhesion, and adherens junction pathways (Figure 2B). circRNAs are endogenous covalently closed ncRNAs with high stability and are enriched in exosomes. circRNAs have been reported to specifically bind micro RNAs (miRNAs), mRNAs, or proteins and to regulate cellular functions through, for example, acting as “molecular sponges”, and thus contributing prominently to tumorigenesis and progression (43). Based on our ncRNA-sequencing results, we identified candidate circRNAs including circ-HUWE1, circ-UBR5, circ-ITCH, circ-BIRC6, circ-UBE2K, circ-UBE3A, circ-XIAP, and circ-MDM2 and then designed primers for qRT-PCR validation in cell lines (Figure 2C). Among these circRNAs, circ-UBR5 was significantly upregulated in hypoxia-treated exosomes and was resistant to ribonuclease (RNase) digestion (Figure 2D), suggesting a specific role of circ-UBR5 in hypoxic exosomes.

Figure 2 Hypoxic exosomes promoted oncogenic phenotypes of normoxic LUAD cells via circ-UBR5. (A) Volcano plot of noncoding RNA-sequencing data of exosomes derived from hypoxic versus normoxic LUAD cells. Genes with |log₂(fold change)| ≥1 and P<0.05 were considered significantly differentially expressed. Red and blue dots represent up-regulated and down-regulated genes, respectively, while gray dots indicate non-significant genes. (B) KEGG pathway enrichment analysis showing pathways related to tumor metastasis. (C) qRT-PCR validation of circRNA expression in hypoxic LUAD cells. (D) RNase R treatment assay confirming the circular structure of circ-UBR5. (E) Overexpression of circ-UBR5 in A549 cells via lentiviral infection. (F) CCK-8 assay showing proliferation of A549 cells in the control and circ-UBR5-overexpression groups. (G) Transwell invasion assay of A549 cells after circ-UBR5 overexpression. Invaded cells were stained with crystal violet for cell number counting. (H) Scratch wound-healing assay showing migration of A549 cells at 0 and 72 h, observed using an inverted light microscope and calculating the migration area. (I) Knockdown of circ-UBR5 in A549 cells via shRNA. (J) CCK-8 assay assessing the proliferation of A549 cells under indicated treatments: N-exo, exosomes from normoxic LUAD cells; H-exo, exosomes from hypoxic LUAD cells; shNC + H-exo, scramble shRNA + H-exo; shcirc-UBR5 + H-exo, circ-UBR5 knockdown shRNA + H-exo. (K) Transwell invasion assay of A549 cells under the indicated treatments. Invaded cells were stained with crystal violet for cell number counting. (L) Wound-healing assay showing migration of A549 cells at 0 and 72 h under the indicated treatments, observed using an inverted light microscope and calculating the migration area. *, P<0.05; **, P<0.01; ns, not significant. CCK-8, Cell Counting Kit-8; H-exo, hypoxic exosomes; KEGG, Kyoto Encyclopedia of Genes and Genomes; LUAD, lung adenocarcinoma; N-exo, normoxic exosomes; OD, optical density; OE, overexpression; qRT-PCR, quantitative real-time polymerase chain reaction; shNC, short hairpin negative control; shRNA, short hairpin RNA.

To evaluate its biological function, we overexpressed circ-UBR5 in normoxic LUAD cell lines. Using lentiviral vectors, we successfully overexpressed circ-UBR5 in A549 cells (Figure 2E) and found that circ-UBR5 overexpression promoted cell proliferation (Figure 2F), migration, and invasion (Figure 2G,2H), mimicking the effects of hypoxic exosome treatment. Conversely, silencing circ-UBR5 with shRNA abolished the promotive effect of hypoxic exosomes on normoxic cells (Figure 2I-2L). These results were further confirmed in PC9 cells (Figure S3). Taken together, our findings indicate that circ-UBR5 is a key prometastatic effector in hypoxic exosomes in LUAD.

circ-UBR5 bound and activated HNRNPR to drive LUAD cell metastasis

To further investigate the downstream mechanism by which circ-UBR5 exerts its effects in normoxic LUAD cells, we performed RNA pulldown and then mass spectrometry analysis. The circ-UBR5 probe captured 56 specific proteins (Figure 3A,3B). Functional enrichment analysis indicated that these differential proteins were significantly enriched in pathways related to ribosome function and RNA splicing (Figure 3C). RNPs are functional complexes formed by RNA molecules and specific proteins through noncovalent interactions and are involved in all RNA-related cellular processes. hnRNPs mainly regulate nucleic acid metabolism and posttranscriptional gene expression by binding nucleic acids, thereby contributing to alternative splicing, mRNA stability, and transcriptional and translational regulation (44). Due to these functions, hnRNPs have attracted considerable attention in cancer research (31,36,45). HNRNPR, a member of the hnRNP family, has been identified as a critical oncogenic splicing factor (37). Western blot analysis in our study confirmed that circ-UBR5 and HNRNPR interact (Figure 3D). Moreover, overexpression of circ-UBR5 in LUAD cell lines induced upregulation of HNRNPR, suggesting that HNRNPR may act as a downstream protein specifically bound by circ-UBR5 (Figure 3E). These findings were further validated by RIP experiments (Figure 3F). Subcellular localization by immunofluorescence revealed that circ-UBR5 and HNRNPR colocalize in the nucleus, indicating potential nuclear interaction (Figure 3G). Functional rescue experiments demonstrated that circ-UBR5 overexpression in A549 cells upregulated HNRNPR protein levels and enhanced cell proliferation, migration, and invasion. Importantly, HNRNPR knockdown largely reversed the prometastatic effects mediated by circ-UBR5 overexpression (Figure 3H-3L). Parallel experiments in PC-9 cells further confirmed that this regulatory axis similarly promotes metastasis in LUAD cells (Figure S4). Collectively, these results indicate that the circ-UBR5-HNRNPR axis is a key molecular basis by which hypoxic exosomes promote and enhance the metastatic phenotype of normoxic LUAD cells.

Figure 3 circ-UBR5 promoted tumor progression via HNRNPR. (A,B) RNA pulldown assay (with biotin-labeled circ-UBR5 or negative control RNA) combined with mass spectrometry to identify circ-UBR5-specific interacting proteins. (C) KEGG pathway annotation of differential proteins. (D) Western blot validation of mass spectrometry results. (E) Overexpression of circ-UBR5 to confirm HNRNPR as a downstream target. (F) RIP assay validating the specific interaction between circ-UBR5 and HNRNPR. (G) In situ hybridization assay showing the subcellular localization of circ-UBR5 in LUAD cells. (H) Overexpression of HNRNPR in A549 cells via lentiviral infection. (I) Knockdown of HNRNPR in A549 cells via shRNA. (J) CCK-8 assay assessing proliferation of A549 cells under indicated treatments (Ctrl, OEcirc-UBR5, OEcirc-UBR5 + shNC, OEcirc-UBR5 + sh-HNRNPR, and OE-HNRNPR). (K) Transwell invasion assay of A549 cells under the indicated treatments. Invaded cells were stained with crystal violet for cell number counting. (L) Wound-healing assay showing migration of A549 cells under the indicated treatments, observed using an inverted light microscope and calculating the migration area. *, P<0.05; **, P<0.01; ns, not significant. COVID-19, coronavirus disease 2019; DAPI, 4’,6-diamidino-2-phenylindole; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HNRNPR, heterogeneous nuclear ribonucleoprotein R; IP, immunoprecipitation; KEGG, Kyoto Encyclopedia of Genes and Genomes; LUAD, lung adenocarcinoma; OD, optical density; OE, overexpression; RIP, RNA immunoprecipitation; shNC, short hairpin negative control; shRNA, short hairpin RNA.

High expression of circ-UBR5 and HNRNPR correlated with LUAD progression, metastasis, and poor prognosis

In cell lines, hypoxic circ-UBR5-containing exosomes were found to promote LUAD progression and metastasis via HNRNPR. To further investigate the clinical significance of circ-UBR5 and HNRNPR, we collected pathological tissue samples from 70 patients with LUAD and performed immunofluorescence staining on paraffin-embedded sections and further analyzed their corresponding clinical and pathological data. Based on fluorescence intensity, the 70 cases were divided into circ-UBR5-low expression and UBR5-high expression groups and into HNRNPR-low expression and HNRNPR-high expression groups. Compared with their respective low-expression groups, the circ-UBR5- and HNRNPR-high expression groups exhibited significantly higher rates of lymph node metastasis and more advanced tumor stage; meanwhile, the UBR5-high expression group had a significantly higher tumor grade and a higher rate of distant metastasis (Figure 4A and Tables 1,2).

Figure 4 Clinical validation of circ-UBR5 and HNRNPR in LUAD and their association with metastasis and prognosis. (A) Representative multiplex immunofluorescence images of LUAD tissue samples showing high or low expression levels of circ-UBR5 and HNRNPR (the left panels show images at a true magnification of 50×, with a scale bar of 200 µm; the right panels show detailed views at a further 2.8× magnification relative to the left images). (B) Correlation and linear regression analysis of circ-UBR5 and HNRNPR expression in tumor tissues. (C) Comparison of circ-UBR5 expression between LUAD tumor tissues and adjacent normal tissues (N=70). (D) Differential expression of HNRNPR between LUAD tumor tissues and adjacent normal tissues. (E) Comparison of circ-UBR5 expression in early-stage (stage I) and advanced-stage (stage II–IV) LUAD tumor tissues. (F) Comparison of HNRNPR expression in early-stage (stage I) and advanced-stage (stage II–IV) LUAD tumor tissues. (G) circ-UBR5 expression in LUAD tumor tissues with or without lymph node metastasis. (H) HNRNPR expression in LUAD tumor tissues with or without lymph node metastasis. (I) Kaplan-Meier analysis of overall survival in patients with LUAD comparing the high- and low-circ-UBR5-expression groups. (J) Kaplan-Meier overall survival analysis comparing the high- and low-HNRNPR expression groups. **, P<0.01. DAPI, 4’,6-diamidino-2-phenylindole; HNRNPR, heterogeneous nuclear ribonucleoprotein R; LUAD, lung adenocarcinoma.

Pearson correlation analysis and linear regression revealed a positive correlation between circ-UBR5 and HNRNPR expression in tumor tissues (R²=0.4628; Figure 4B). Both molecules were expressed at significantly higher levels in tumor tissues than in adjacent nontumor tissues (Figure 4C,4D). These findings are consistent with our in vitro experiments, suggesting that circ-UBR5 may exert its biological effects in vivo by upregulating HNRNPR. The expression of circ-UBR5, HNRNPR, and their interaction were closely associated with LUAD biological characteristics. Further analysis of circ-UBR5 and HNRNPR and their association with tumor stage indicated that the expression of both was lower in early-stage (stage I) LUAD and higher in advanced-stage (stage II–IV) LUAD, suggesting that high expression of circ-UBR5 and HNRNPR is associated with tumor progression and metastasis (Figure 4E,4F). Additionally, the tumor tissue expression of circ-UBR5 and HNRNPR was significantly higher in patients with lymph node metastasis than in those without it (Figure 4G,4H).

Survival analysis with Kaplan-Meier curves demonstrated that overall survival was significantly shorter in the circ-UBR5- and HNRNPR-high expression groups than in the low expression groups. Specifically, the 5-year survival rates for the circ-UBR5-high and -low expression groups were 20.0% and 54.3%, respectively, while those for the HNRNPR-high and -low expression groups were 17.1% and 57.1%, respectively (Figure 4I,4J).

Taken together, these results indicate that HNRNPR is a downstream target of circ-UBR5 in LUAD and that the high expression of circ-UBR5 and HNRNPR plays a critical role in LUAD progression and metastasis, correlating with poor prognosis.

Discussion

Cancer metastasis remains a major determinant of poor prognosis in patients with LUAD, and the hypoxic TME imposes selective pressures on tumor growth regions (46,47). Intratumoral oxygen tension is highly heterogeneous, and hypoxic tumor cells exhibit enhanced antioxidant capacity and metastatic potential (48). Evidence from research on colorectal cancer suggests that hypoxic tumor cells can deliver signaling molecules to normoxic cells via exosomes, thereby reshaping the tumor niche and promoting metastasis (49). However, the molecular mechanisms underlying intercellular communication across oxygen gradients in aggressive lung cancer remain largely unexplored. In this study, we identified circ-UBR5 as a key molecule enriched in exosomes derived from hypoxic LUAD cells, which promotes metastasis by modulating the expression of the RNA-binding protein HNRNPR in normoxic tumor cells.

We demonstrated that circ-UBR5 is secreted by hypoxic tumor cells and internalized into normoxic LUAD cells via exosomes. Following uptake, circ-UBR5 translocates to the nucleus and binds to HNRNPR, upregulating its expression and driving an invasive phenotype. HNRNPR, an RNA-binding protein, is known to regulate posttranscriptional gene expression and protein synthesis; it can stabilize ASCL1 via m6A-dependent mechanisms to promote neuroblastoma progression or drive hepatocellular carcinoma metastasis through mRNA splicing of UPF3B (39). Our study is the first to reveal the role of the circ-UBR5-HNRNPR axis in hypoxia-induced lung cancer metastasis, although the precise posttranscriptional regulatory mechanisms mediated by HNRNPR in this context should be further investigated. Moreover, we found that both circ-UBR5 and HNRNPR were significantly upregulated in samples from patients with advanced-stage LUAD, with their expression levels being closely correlated with tumor stage and patient survival. These observations suggest that circ-UBR5 could serve not only as a predictive biomarker for LUAD metastasis and prognosis but also as a potential indicator for disease monitoring through liquid biopsy. Notably, the circular structure of circRNAs confers high stability both intracellularly and within exosomes, making them reliable mediators of intercellular communication (50). Our study indicates that circ-UBR5 can be efficiently transported across cells under varying oxygen conditions via exosomes, supporting its potential as a therapeutic target for inhibiting NSCLC metastasis.

Despite these compelling findings, several limitations should be noted. First, the detailed molecular mechanisms by which circ-UBR5 regulates HNRNPR, particularly the posttranscriptional modifications driving hypoxia-induced metastasis, require further elucidation. Second, mechanistic validation of the circ-UBR5-HNRNPR axis was performed primarily in A549 and PC9 LUAD cell lines, and the in vivo models and clinical sample size were relatively limited. Future studies using larger, multicenter patient cohorts, as well as in vivo animal models or patient-derived organoid systems, are warranted to comprehensively evaluate the clinical relevance of circ-UBR5 and its potential as a therapeutic target. In parallel, further exploration of potential therapeutic targets should be considered, including drug repurposing strategies based on GLP-1 therapy (51).

In summary, our study established a newly discovered mechanism in which hypoxic LUAD cells deliver circ-UBR5 to normoxic LUAD cells via exosomes, thereby promoting metastasis through HNRNPR. circ-UBR5, being a stable exosomal molecule capable of intercellular transfer, holds promise as both a biomarker for predicting LUAD metastasis and patient prognosis and as a potential therapeutic target in the development of antimetastatic treatments.

Conclusions

This study clarified a intercellular communication mechanism in which hypoxic LUAD cells promote the metastatic progression of normoxic tumor cells through exosomal circ-UBR5. We demonstrated that circ-UBR5 is specifically enriched in hypoxic exosomes and that upon internalization by recipient normoxic cells, binds to and upregulates HNRNPR, thereby driving aggressive malignant phenotypes. The significant correlation between high circ-UBR5 or HNRNPR expression and advanced disease stage, metastasis, and poor patient prognosis underscores the clinical relevance of this axis. These findings not only deepen our understanding of hypoxia-mediated metastasis but also position the exosomal circ-UBR5-HNRNPR pathway as a promising therapeutic target and a potential source of prognostic biomarkers for LUAD. Future work should focus on translating these insights into targeted antimetastatic strategies.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1413/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1413/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1413/prf

Funding: This work was supported by the National Natural Science Foundation of China (No. 82073191), National Natural Science Foundation of China Projects for International Exchanges Scheme (No. 12411530129), Computational Biology Program of Science and Technology Commission of Shanghai Municipality (STCSM) (No. 24JS2840300), Shanghai Health Commission Emerging Interdisciplinary Research Program (No. 2022JC010), and Beijing Xisike Clinical Oncology Research Foundation of Early-Stage NSCLC Research Grant [No. Y-2024AZ(NSCLC)MS-0211].

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1413/coif). X.C. reports that this work was funded by the National Natural Science Foundation of China (No. 82073191), National Natural Science Foundation of China Projects for International Exchanges Scheme (No. 12411530129), Computational Biology Program of Science and Technology Commission of Shanghai Municipality (STCSM) (No. 24JS2840300), Shanghai Health Commission Emerging Interdisciplinary Research Program (No. 2022JC010), and Beijing Xisike Clinical Oncology Research Foundation of Early-Stage NSCLC Research Grant [No. Y-2024AZ(NSCLC)MS-0211]. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Shanghai Chest Hospital Ethics Committee (No. KS22018) and informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Leiter A, Veluswamy RR, Wisnivesky JP. The global burden of lung cancer: current status and future trends. Nat Rev Clin Oncol 2023;20:624-39. [Crossref] [PubMed]
3. Jones GD, Brandt WS, Shen R, et al. A Genomic-Pathologic Annotated Risk Model to Predict Recurrence in Early-Stage Lung Adenocarcinoma. JAMA Surg 2021;156:e205601. [Crossref] [PubMed]
4. Elhanani O, Ben-Uri R, Keren L. Spatial profiling technologies illuminate the tumor microenvironment. Cancer Cell 2023;41:404-20. [Crossref] [PubMed]
5. Sokolov D, Sharda N, Banerjee A, et al. Differential Signaling Pathways in Medulloblastoma: Nano-biomedicine Targeting Non-coding Epigenetics to Improve Current and Future Therapeutics. Curr Pharm Des 2024;30:31-47. [Crossref] [PubMed]
6. Eldash S, Sanad EF, Nada D, et al. The Intergenic Type LncRNA (LINC RNA) Faces in Cancer with In Silico Scope and a Directed Lens to LINC00511: A Step toward ncRNA Precision. Noncoding RNA 2023;9:58. [Crossref] [PubMed]
7. Rankin EB, Giaccia AJ. Hypoxic control of metastasis. Science 2016;352:175-80. [Crossref] [PubMed]
8. Vaupel P. Prognostic potential of the pre-therapeutic tumor oxygenation status. Adv Exp Med Biol 2009;645:241-6. [Crossref] [PubMed]
9. Vaupel P, Höckel M, Mayer A. Detection and characterization of tumor hypoxia using pO2 histography. Antioxid Redox Signal 2007;9:1221-35. [Crossref] [PubMed]
10. Wang T, Zhu G, Wang B, et al. Activation of Hypoxia Inducible Factor-1 Alpha-Mediated DNA Methylation Enzymes (DNMT3a and TET2) Under Hypoxic Conditions Regulates S100A6 Transcription to Promote Lung Cancer Cell Growth and Metastasis. Antioxid Redox Signal 2024;41:138-51. [Crossref] [PubMed]
11. Koyasu S, Horita S, Saito K, et al. ZBTB2 links p53 deficiency to HIF-1-mediated hypoxia signaling to promote cancer aggressiveness. EMBO Rep 2023;24:e54042. [Crossref] [PubMed]
12. Sorg BS, Hardee ME, Agarwal N, et al. Spectral imaging facilitates visualization and measurements of unstable and abnormal microvascular oxygen transport in tumors. J Biomed Opt 2008;13:014026. [Crossref] [PubMed]
13. Gilkes DM, Semenza GL, Wirtz D. Hypoxia and the extracellular matrix: drivers of tumour metastasis. Nat Rev Cancer 2014;14:430-9. [Crossref] [PubMed]
14. Atta H, Alzahaby N, Hamdy NM, et al. New trends in synthetic drugs and natural products targeting 20S proteasomes in cancers. Bioorg Chem 2023;133:106427. [Crossref] [PubMed]
15. Hamdy NM, Noureldein MH, Gouhar SA, et al. Diagnostic and therapeutic potential of exosomal circRNAs in cancer: decoding the circular code toward precision medicine. Cancer Cell Int 2025;25:432. [Crossref] [PubMed]
16. Kalluri R, LeBleu VS. The biology, function, and biomedical applications of exosomes. Science 2020;367:eaau6977. [Crossref] [PubMed]
17. Wortzel I, Dror S, Kenific CM, et al. Exosome-Mediated Metastasis: Communication from a Distance. Dev Cell 2019;49:347-60. [Crossref] [PubMed]
18. Hamdy NM, Sanad EF, Kassab SE, et al. Treatment of non-small cell lung cancer using chem-bioinformatics-driven engineering of exosomal cargo-vehicle for telmisartan and pioglitazone targeted-delivery. Sci Rep 2025;15:25166. [Crossref] [PubMed]
19. Lobb RJ, Lima LG, Möller A. Exosomes: Key mediators of metastasis and pre-metastatic niche formation. Semin Cell Dev Biol 2017;67:3-10. [Crossref] [PubMed]
20. Xu Z, Chen Y, Ma L, et al. Role of exosomal non-coding RNAs from tumor cells and tumor-associated macrophages in the tumor microenvironment. Mol Ther 2022;30:3133-54. [Crossref] [PubMed]
21. Kristensen LS, Andersen MS, Stagsted LVW, et al. The biogenesis, biology and characterization of circular RNAs. Nat Rev Genet 2019;20:675-91. [Crossref] [PubMed]
22. Zhang F, Jiang J, Qian H, et al. Exosomal circRNA: emerging insights into cancer progression and clinical application potential. J Hematol Oncol 2023;16:67. [Crossref] [PubMed]
23. Youness RA, Hassan HA, Abaza T, et al. A Comprehensive Insight and In Silico Analysis of CircRNAs in Hepatocellular Carcinoma: A Step toward ncRNA-Based Precision Medicine. Cells 2024;13:1245. [Crossref] [PubMed]
24. Xie H, Yao J, Wang Y, et al. Exosome-transmitted circVMP1 facilitates the progression and cisplatin resistance of non-small cell lung cancer by targeting miR-524-5p-METTL3/SOX2 axis. Drug Deliv 2022;29:1257-71. [Crossref] [PubMed]
25. Wang D, Wang S, Jin M, et al. Hypoxic Exosomal circPLEKHM1-Mediated Crosstalk between Tumor Cells and Macrophages Drives Lung Cancer Metastasis. Adv Sci (Weinh) 2024;11:e2309857. [Crossref] [PubMed]
26. Zhu S, Liao B. Mechanism of hypoxia-induced exosome circ_0051799 regulating the progression of lung adenocarcinoma. Biol Chem 2024;405:143-60. [Crossref] [PubMed]
27. Gao C, Su Y, Koeman J, et al. Chromosome instability drives phenotypic switching to metastasis. Proc Natl Acad Sci U S A 2016;113:14793-8. [Crossref] [PubMed]
28. Valastyan S, Weinberg RA. Tumor metastasis: molecular insights and evolving paradigms. Cell 2011;147:275-92. [Crossref] [PubMed]
29. Zhang Y, Zhang Y, Song J, et al. Targeting the “tumor microenvironment”: RNA-binding proteins in the spotlight in colorectal cancer therapy. Int Immunopharmacol 2024;131:111876. [Crossref] [PubMed]
30. Dreyfuss G, Kim VN, Kataoka N. Messenger-RNA-binding proteins and the messages they carry. Nat Rev Mol Cell Biol 2002;3:195-205. [Crossref] [PubMed]
31. Kędzierska H, Piekiełko-Witkowska A. Splicing factors of SR and hnRNP families as regulators of apoptosis in cancer. Cancer Lett 2017;396:53-65. [Crossref] [PubMed]
32. Phoomak C, Park D, Silsirivanit A, et al. O-GlcNAc-induced nuclear translocation of hnRNP-K is associated with progression and metastasis of cholangiocarcinoma. Mol Oncol 2019;13:338-57. [Crossref] [PubMed]
33. Jin Z, Liang F, Yang J, et al. hnRNP I regulates neonatal immune adaptation and prevents colitis and colorectal cancer. PLoS Genet 2017;13:e1006672. [Crossref] [PubMed]
34. Zhou ZJ, Dai Z, Zhou SL, et al. HNRNPAB induces epithelial-mesenchymal transition and promotes metastasis of hepatocellular carcinoma by transcriptionally activating SNAIL. Cancer Res 2014;74:2750-62. [Crossref] [PubMed]
35. Tauler J, Zudaire E, Liu H, et al. hnRNP A2/B1 modulates epithelial-mesenchymal transition in lung cancer cell lines. Cancer Res 2010;70:7137-47. [Crossref] [PubMed]
36. Ryu HG, Jung Y, Lee N, et al. HNRNP A1 Promotes Lung Cancer Cell Proliferation by Modulating VRK1 Translation. Int J Mol Sci 2021;22:5506. [Crossref] [PubMed]
37. Wang H, Qian D, Wang J, et al. HnRNPR-mediated UPF3B mRNA splicing drives hepatocellular carcinoma metastasis. J Adv Res 2025;68:257-70. [Crossref] [PubMed]
38. Chen EB, Qin X, Peng K, et al. HnRNPR-CCNB1/CENPF axis contributes to gastric cancer proliferation and metastasis. Aging (Albany NY) 2019;11:7473-91. [Crossref] [PubMed]
39. Hu T, Zeng C, Song Z, et al. HNRNPA2B1 and HNRNPR stabilize ASCL1 in an m6A-dependent manner to promote neuroblastoma progression. Biochim Biophys Acta Mol Basis Dis 2024;1870:167050. [Crossref] [PubMed]
40. Wang Y, Liu B, Min Q, et al. Spatial transcriptomics delineates molecular features and cellular plasticity in lung adenocarcinoma progression. Cell Discov 2023;9:96. [Crossref] [PubMed]
41. Li L, Yang L, Fan Z, et al. Hypoxia-induced GBE1 expression promotes tumor progression through metabolic reprogramming in lung adenocarcinoma. Signal Transduct Target Ther 2020;5:54. [Crossref] [PubMed]
42. Toden S, Zumwalt TJ, Goel A. Non-coding RNAs and potential therapeutic targeting in cancer. Biochim Biophys Acta Rev Cancer 2021;1875:188491. [Crossref] [PubMed]
43. Misir S, Wu N, Yang BB. Specific expression and functions of circular RNAs. Cell Death Differ 2022;29:481-91. [Crossref] [PubMed]
44. Geuens T, Bouhy D, Timmerman V. The hnRNP family: insights into their role in health and disease. Hum Genet 2016;135:851-67. [Crossref] [PubMed]
45. Zhu L, Tian Q, Gao H, et al. PROX1 promotes breast cancer invasion and metastasis through WNT/β-catenin pathway via interacting with hnRNPK. Int J Biol Sci 2022;18:2032-46. [Crossref] [PubMed]
46. Maley CC, Aktipis A, Graham TA, et al. Classifying the evolutionary and ecological features of neoplasms. Nat Rev Cancer 2017;17:605-19. [Crossref] [PubMed]
47. Hamdy NM, Zaki MB, Abdelmaksoud NM, et al. Comprehensive insights and In silico analysis into the emerging role of LincRNAs in lung diseases pathogenesis; a step toward ncRNA precision. Funct Integr Genomics 2025;25:34. [Crossref] [PubMed]
48. Godet I, Shin YJ, Ju JA, et al. Fate-mapping post-hypoxic tumor cells reveals a ROS-resistant phenotype that promotes metastasis. Nat Commun 2019;10:4862. [Crossref] [PubMed]
49. Yang H, Zhang H, Yang Y, et al. Hypoxia induced exosomal circRNA promotes metastasis of Colorectal Cancer via targeting GEF-H1/RhoA axis. Theranostics 2020;10:8211-26. [Crossref] [PubMed]
50. Li J, Zhou W, Wang H, et al. Exosomal circular RNAs in tumor microenvironment: An emphasis on signaling pathways and clinical opportunities. MedComm (2020) 2024;5:e70019.
51. Mostafa AM, Hamdy NM, Abdel-Rahman SZ, et al. Effect of vildagliptin and pravastatin combination on cholesterol efflux in adipocytes. IUBMB Life 2016;68:535-43. [Crossref] [PubMed]

(English Language Editor: J. Gray)