Coxsackievirus B3 Inhibited Colorectal Cancer by Upregulating miR-214-3P and Promoting Ferroptosis

Abstract

Introduction

Colorectal cancer (CRC) is the third most common cancer worldwide and a significant public health threat with far-reaching societal implications. The currently available CRC therapeutic strategies have limitations, thus requiring the development of new strategies. Coxsackievirus B3 (CVB3) exhibits strong oncolytic activity in CRC, although its mechanism of action remains unclear. This study aimed to investigate whether the induction of ferroptosis is a promising treatment strategy for CRC and whether CVB3 could activate ferroptosis during infection.

Methods

In vitro and in vivo experiments were conducted to evaluate whether CVB3 infection activates the ferroptosis pathway by upregulating miR-214-3p to suppress glutathione peroxidase 4 (GPX4) expression. Dual-luciferase assays and rescue experiments were performed to confirm this regulatory mechanism. Clinical CRC tissues and colon cancer xenograft models were used to demonstrate the mediating role of the miR-214-3p/GPX4 axis in the interaction between viral replication and ferroptosis.

Results

CVB3 demonstrated oncolytic virus properties by selectively lysing tumor cells. The in vitro and in vivo experiments confirmed that CVB3 activates the ferroptosis pathway by upregulating miR-214-3p to suppress GPX4 expression, thereby promoting viral replication and tumor regression. Antagonizing miR-214-3p reversed this process.

Conclusion

miR-214-3p expression was upregulated during CVB3 infection of CRC tissues and cells, activating the ferroptosis pathway and promoting tumor cell death.

Introduction

Colorectal cancer (CRC) is the third most common cancer and second most common cause of cancer-related mortality worldwide. ¹ CRC is a significant global health burden because of cancer-related morbidity and mortality. ² Despite the development in screening, early detection, and therapy, the prognosis of patients with advanced or metastatic CRC remains unsatisfactory. Surgical resection is the most common treatment method for early CRC, whereas chemotherapy and targeted therapy are the main treatment methods for late and recurrent CRC.^3,4 Owing to the influence of size, quantity, location, and degree of involvement of liver metastases, tumors cannot be completely removed by surgical methods in over 70% of patients. ⁵ Although chemotherapy and radiotherapy have improved the 5-year survival rate, further improvement in CRC survival is difficult to achieve. ⁶ Oncolytic virotherapy has become a novel and highly regarded cancer treatment strategy over the recent years. ⁷ Oncolytic viruses (OVs) are used to stimulate the antitumor immune response of the host by infecting and killing cancer cells. OVs can improve overall survival rates in patients with different types of tumors and at different stages of progression, even in those with metastatic and untreatable cancers. ⁸ For patients with advanced cancer, OV therapy is considered a promising life-saving approach and can cause complete regression or remission. ⁹ Clinical trials have demonstrated the efficacy of various OVs against many malignant tumors including primary and metastatic cancers. ¹⁰ OVs recognize receptors on the tumor cell surface, specifically target and inhibit tumor growth signaling pathways within tumor cells, and change the tumor microenvironment to lyse tumor cells.^11,12

Coxsackieviruses (CVs) belong to the Picornaviridae family and Enterovirus genus, which are single-stranded RNA viruses. Primary infection with Group B Coxsackievirus (CVB) occurs in the intestine and is commonly observed in children and adolescents. ¹³ Most cellular infections manifest as acute cytotoxic infections. After acute infection with CVB, persistent nontoxic infection occurs, which provides a prerequisite for sustained targeted therapy of metastatic cancer lesions. ¹⁴ Coxsackievirus Group B type 3 (CVB3) exhibits tumor lytic activity in multiple types of cancer cells, such as pancreatic cancer, cervical carcinoma, and breast cancer cells. ¹⁵ Both CVB3 and engineered CVB3 influence CRC treatment. ¹⁶ However, the potential mechanism of CVB3-mediated colon cancer inhibition remains unclear.

CVB3 induces various forms of cell death, among which ferroptosis has become a popular research topic in recent years. Ferroptosis is a new type of cell death reported by Stockwell et al in 2012. It results from iron-dependent lipid peroxidation, differs from traditional apoptosis and necrosis, and is characterized by a series of cytological changes. ¹⁷ Cytological changes driven by ferroptosis include cell volume shrinkage and increased mitochondrial membrane density. ¹⁸ Ferroptosis is involved in various physical conditions and diseases including inflammation and cancers. The mechanism underlying ferroptosis induction consists of inhibition of the system X_c⁻ and direct inhibition of glutathione peroxidase 4 (GPX4) and P53‐SLC7A11P53 axis. ¹⁹ Ferroptosis is closely associated with the development of CRC, cervical carcinoma, and gastric carcinoma. ²⁰ Recent researches have indicated that therapies targeting ferroptosis are ideal and prospective treatment strategies for CRC. Moreover, CVB3 infection induces ferroptosis. ²¹ However, whether CVB3 kills colon cancer cells by inducing ferroptosis is unclear. This study aimed to investigate whether CVB3 activates the ferroptotic pathway by upregulating miR-214-3p, a newly identified upstream regulator of GPX4, and reveal, for the first time, the core role of the new regulatory axis, “miR-214-3p/GPX4,” in the oncolytic process. CVB3 directly targeted and inhibited GPX4 expression by upregulating miR-214-3P, thereby relieving the inhibitory effect of GPX4 on lipid peroxidation and ultimately exerting an antitumor effect by promoting ferroptosis. Thus, CVB3 may be applied in the future clinical treatment of CRC.

Materials and Methods

Cell Culture

HeLa, HEK293, HCT-116, SW480, SW620, and NCM460 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) in the presence of fetal bovine serum, penicillin (100 U/ml), and streptomycin (0.1 mg/mL). These cells were incubated at 37°C with 5% CO₂.

Viruses and Viral Infection

The CVB3 Woodruff strain (CVB3) was recovered from HeLa cells transfected with the pMKS1 plasmid, which was provided by the Scripps Research Institute and contains entire CVB3 genome. HeLa cells were used to amplify CVB3. The CVB3 titer was 1 × 10⁹ pfu/mL, which was assayed in plaque-forming units according to the manufacturer’s protocol.

Murine Tumor Models

Male BALB/c nude mice (BALB/c-nu/nu strain, 4 weeks old, 18-22 g, n = 5 per group; Charles River Laboratories, with pathogen-free status confirmed) were housed under specific pathogen-free conditions (including a 7-day acclimatization period) at 22 ± 1°C with 12-hour light/dark cycles. A control group receiving PBS/Matrigel-only (n = 5) was included. Mice were randomly allocated to groups and investigators were blinded during measurements and analysis. Xenografts were established by subcutaneous injection into the right flank of 1.5 × 10⁶ SW620 cells suspended in 200 μL of a 1:1 PBS/Matrigel mixture. Tumor size was measured weekly using a digital caliper and calculated as (length × width ² ) × 0.52. Data (mean ± SD) were analyzed via two-way ANOVA (using GraphPad Prism 7.0a). No animals were excluded from the analysis. Cage positions were randomized weekly. The reporting of this study conforms to ARRIVE 2.0 guidelines. ²²

Transfection

Nucleotides, including miR-214-3p mimic (miR-214-3p), miR-214-3p inhibitor (AMO-214-3p), and the corresponding controls (miR-NC and AMO-NC), were synthesized by RiboBio (Guangzhou, China). Nucleotides were transiently transfected into HEK293 cells using the X-treme GENE transfection reagent (Roche, Mannheim, Germany) according to the manufacturer’s instructions. At 70% confluence, the culture medium was replaced with serum-free medium, the transfection mixture was then added to the cells and incubated for 4 h. Subsequently, the transfection mixture was replaced with complete culture medium.

RT-qPCR

Total RNA was extracted from cells and tissues using the TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. A NanoDrop2000 spectrophotometer was used to detect the purity and concentration of the total RNA. cDNAs were synthesized by PrimeScript RT Enzyme Mix I (TaKaRa, Otsu, Shiga, Japan) with 1 μg of total RNA in a reverse transcription system. SYBR Premix Ex Taq II (TaKaRa) with 25 μL system was used to perform real-time quantitative PCR. The relative abundance of gene expression was calculated using the 2^−ΔΔCt method. The expression of miRNAs was normalized to U6 expression, whereas mRNA expression was normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression. The primers used for miRNA and mRNA quantification are provided as a Supplemental File.

Western Blotting

We used radioimmunoprecipitation assay lysis buffer containing phenylmethylsulfonyl fluoride to extract total protein. Protein samples from cells and tissues were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride (PVDF) membranes. The PVDF membranes were blocked with 5% non-fat milk in Tris-buffered saline solution containing Tween 20 buffer for 1 h at room temperature. Next, the PVDF membranes were incubated with the corresponding primary antibodies overnight at 4°C. After incubation with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature, membranes were imaged using a FluorChem M CCD camera. GAPDH was used as an internal control via ImageJ v1.53. We used the primary antibodies anti-GPX4 antibody (rabbit monoclonal, Abcam, USA, Cat# ab125066; 1:1000 dilution), anti-GAPDH (mouse monoclonal, ABclonal, Wuhan, China, Cat# AC033-50UL; 1:1000 dilution), anti-3D antibody (produced in-house, rabbit polyclonal; 1:20 000 dilution), and anti-ACSL4 antibody (rabbit monoclonal, Biodragon, Suzhou, China, Cat# RM8098-50UL; 1:2000 dilution).

Immunofluorescence Staining

Cells were fixed in 4% paraformaldehyde for 20 min at room temperature. After washing in phosphate-buffered saline (PBS), the samples were permeabilized with 0.1% Triton-X for 1 h. Goat serum was placed on samples to block for 2 h at 37°C. The samples were added with corresponding primary antibodies at 4°C overnight. They were then incubated with the secondary antibody for 1 h at room temperature. Next, DAPI was used to stain the nuclei for 20 min. Images were captured and recorded using a fluorescence microscope.

Trivalent Iron Ion Assay

We used a Prussian blue iron staining kit to detect trivalent iron ions according to the manufacturer’s instructions. The tissue samples were embedded in paraffin, deparaffinized, and rehydrated. The staining solution was mixed at a 1:1 ratio and dripped onto the tissue sections. After incubation for 20 min at 37°C, the sections were washed with distilled water and incubated again for 20 min at 37°C. After washing thrice, an enhanced working solution was added to each slide, and the samples were stored in a humidified chamber for 20 min at 37°C. After washing with 1× PBS 3 times, the counterstain solution was added to the tissue sections, which were then incubated for 5 min. The tissue sections were washed with distilled water, dehydrated in gradient ethanol solutions, and cleared in xylene. Finally, the sections were sealed with neutral gum.

Ferrous Iron Colorimetric Assay

To determine the ferrous ion (Fe²⁺) content in the samples, we utilized an Iron(II) Colorimetric Assay Kit (Solarbio, Beijing, China, Cat#: BC5415). Next, 0.1 g of tissue was homogenized in 1 mL of extraction buffer on ice, followed by centrifugation at 10 000× g for 10 min at 4°C. The supernatants were collected and transferred to a 96-well microplate. Absorbance was measured at 593 nm using a microplate reader.

MDA Content Assay

The MDA content was assayed using an MDA Content Assay Kit (Boxbio, Beijing, China). A lysis buffer was used to homogenize the tissue. The samples were centrifuged at 12 000× g for 10 min, and the supernatant was collected. The protein concentration was determined using the bicinchoninic acid method. The samples were mixed with thiobarbituric acid detection solution, and the absorbance was measured at 532 nm. The MDA content was calculated according to a standard curve.

Hematoxylin and Eosin (HE) Staining

After fixation with 4% paraformaldehyde, the collected tissues were embedded in paraffin. The tissues were cut into slices and deparaffinized in xylene. After rehydration in ethanol, the sections were stained with HE eosin and sealed with neutral balsam. Sections were imaged using an optical microscope.

Immunohistochemistry

Paraffin sections were incubated in the corresponding primary antibody overnight at 4°C, followed by a secondary antibody and then diaminobenzidine. Using a microscope, the sections were observed, and images were collected.

ROS Assay

After 6 h of CVB3 infection, the culture medium was discarded, and cells were incubated with a ROS detection working solution (containing 10 μM DCFH-DA) for 30 min in the dark. Subsequently, the cells were washed 3 times with serum-free DMEM to remove residual probes. Fluorescence images were captured using a fluorescence microscope at the excitation and emission wavelengths of 488 nm/525 nm. Fluorescence intensity was quantified using the ImageJ software and normalized to cell counts for relative ROS level analysis.

Statistical Analysis

All research data obtained from at least 3 independent experiments are presented as means ± standard deviations. Data were analyzed using the GraphPad Prism 7 software. Student’s t-test or one-way analysis of variance was performed.

Results

CVB3 Induced Cell Death in Colon Cancer Cell Lines

To determine the effect of CVB3 on cancer cells, cell lines derived from human cervical, lung, liver, kidney, pancreatic, and oral carcinoma were infected with CVB3. CVB3 exerted a significant cytotoxic effect on most human primary tumor cells, and most cells exhibited swelling and rounding, nuclear cytoplasmic fragmentation, and rapid death (Figure 1A). Western blotting analysis revealed that the CVB3 3D level in tumor cells was significantly increased. We then infected normal human cell lines, such as cell lines derived from the human myocardium, vascular endothelium, renal tubular epithelium, normal liver, and human oral epithelium with CVB3, and observed cytopathic effects (CPEs). CPE and CVB3 3D protein expression were not observed in most cells, except in HL-7702 cells (a human normal liver cell line). Thus, CVB3, an OV, demonstrated relatively low toxicity for tumor treatment (Figure 1B). Compared with HeLa cells, CVB3 did not infect the normal epithelial cell line NCM460; however, the other 3 colon cancer cell lines were susceptible to infection (Figure 1C). Therefore, CVB3 may be prone to infecting human colon cancer cells.

Figure 1.

CVB3 Induced Cell Death in Colon Cancer Cell Lines. (A) Observation of the Cytotoxic Effect of CVB3 Infection on Tumor Cells by Light Microscopy; Scale bar: 100 μm. Western Blotting Analysis and Relative Quantification of 3D Protein Expression. (B) Observation of the Effects of CVB3 Infection on Normal Cells by Light Microscopy; Scale bar: 100 μm. Western Blotting Analysis and Relative Quantification of 3D Protein Expression. (C) Observation of the Effect of CVB3 on Colon Cells by Light Microscopy; Scale bar: 100 μm. Western Blotting Analysis and Relative Quantification of 3D Protein Expression. Compared With the Control, *P < 0.05, **P < 0.01, ***P < 0.001. Each Experiment was Conducted With n = 3 Replicates. CVB3, Coxsackievirus B3; 3D, Three-Dimensional

CVB3 Activated Ferroptosis to Inhibit the Growth of Colon Cancer

To clarify the mechanism of CVB3-induced cell death, we used the keyword “CVB3” from the GeneCards database to search for 239 genes related to CVB3 and identified 1217 tumor suppressor genes at the pan-cancer level and 535 tumor suppressor genes in CRC from the TSGene 2.0 database (https://bioinfo.uth.edu/TSGene/index.html). Among the 239 CVB3-related genes, 44 were tumor suppressor genes at the pan-cancer level, and 11 were tumor suppressor genes in CRC. We conducted separate KEGG pathway enrichment analyses for the 2 gene sets. Tumor suppressor genes at the pan-cancer level were mainly enriched in microRNAs (miRNAs) involved in the cancer pathway, indicating that CVB3 primarily exerts its anticancer function through miRNAs. In addition, both gene sets were significantly enriched in the p53 signaling pathway, suggesting the close association of CVB3 with the tumor suppressor gene TP53 signaling pathway. Among the 239 CVB3-related genes, TP53 (p53 in the pathway diagram), TFRC (TFR1 in the pathway diagram), SLC11A2 (DMT1 in the pathway diagram), PCBP2, and MAP1LC3A (LC3 in the pathway diagram) participated in ferroptosis (Figure 2A). These genes are closely related to CVB3, which may activate ferroptosis by altering the expression of these 5 genes, thereby exerting lethal effects on tumor cells. We subsequently analyzed which CVB3-related genes are included in the CRC “microRNAs in the cancer pathway” and the location of TP53 in the pathway. The CRC sub-pathway contained the CVB3-related gene miR-126, and CVB3 may inhibit CRC through miR126. In addition, the p53 pathway, which contains the CVB3-related genes, miR-125, and miR-30 (Figure 2A), is present in the bladder cancer sub-pathway. Thus, CVB3 may regulate ferroptosis by affecting miRNAs.

Figure 2.

CVB3 Activates Ferroptosis to Inhibit Colon Cancer Growth. (A) Pathways Regulated by Tumor Suppressor Genes Common to CVB3. Pathways Regulated by Tumor Suppressor Genes Related Pathways to Common to CVB3 and CRC. Genes Related to CVB3 are Labeled in Pink. (B) HE and Immunohistochemistry for ROS, Fe³⁺, and GPX4; Scale bar: 50 μm. (C) Ferrous Ion Detection and MDA Testing. Western Blot Analysis and Relative GPX4 Protein Expression. (D) Images of the Effects of CVB3 Inhibition on Xenograft Tumor Growth in Nude Mice. (E) ROS Detection, Ferrous Ion Detection, and MDA Testing. (F) Western Blot and Relative Expression of CVB-3D and GPX4 Protein. Compared With the Control, *P < 0.05, **P < 0.01, ***P < 0.001. Each Experiment was Conducted With n = 5 Replicates. CVB, Coxsackievirus; ROS, Reactive Oxygen Species; HE, Hematoxylin and Eosin; MDA, Malondialdehyde; CRC, Colorectal Cancer; CVB3, Coxsackievirus B3

To ascertain the role of ferroptosis in CRC, we collected 10 colon cancer tissue samples and corresponding proximal normal intestinal epithelial tissues to detect evidence of ferroptosis; HE staining and immunohistochemistry revealed that compared with those in normal tissues, inflammatory lesions and the levels of reactive oxygen species (ROS), Fe³⁺, and GPX4 were increased in colon cancer tissues (Figure 2B). The malondialdehyde (MDA) and Fe²⁺ levels in CRC tissues were lower than those in normal tissues. Western blotting revealed that GPX4 levels were higher in CRC tissues than in normal tissues (Figure 2C). The iron-dependent cell death pathway was inhibited in human colon cancer tissues.

To determine whether CVB3 influences colon cancer xenograft tumors in an animal model, 4-week-old BALB/c nude mice were injected with 1.5 × 10⁶ SW620 cells subcutaneously to establish a xenograft model. After 14 days, the subcutaneous transplant tumor volume in the mice significantly increased, and CVB3 was injected into the transplanted tumor.

Compared with the control group, the CVB3 virus-infected group presented significantly inhibited tumor growth (Figure 2D); the ROS, Fe²⁺, and MDA levels increased, although the GPX4 level decreased (Figure 2E,F). CVB3 exhibited oncolytic activity by activating ferroptosis.

CVB3 Inhibited the Growth of CRC Cells by Regulating miR-214-3p to Activate Ferroptosis

To clarify the mechanism by which CVB3 activates ferroptosis, we predicted the expression of GPX4-related miRNAs in humans and mice and determined that 130 GPX4-related miRNAs were expressed in both humans and mice (Figure 3A). Subsequently, we predicted binding ability (minimum free energy [MFE] value), screened 130 miRNAs related to GPX4 expression in both humans and mice with MFE values of < −25 kcal/mol, and used the intersection of the 4 sections to screen 24 miRNAs related to GPX4 expression in both humans and mice with MFE values of < −25 kcal/mol. Finally, we compared the results of the feasibility analysis of these 24 miRNAs using the ENCORI and miRDB databases, and selected 6 miRNAs for further experiments (Figure 3A). To verify the screening results, CVB3 (multiplicity of infection = 10) was used to infect SW620 cells, which were assayed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) to measure miRNA expression. The abundance of miR-214-3P was significantly elevated in the CVB3 infection group compared with that in the control group, and it was the most significantly altered among the 6 miRNAs (Figure 3B). Since some studies have reported that CVB-3D protein targets and regulates miR-214-3P, we selected miR-214-3P as the main subject for further research.

Figure 3.

CVB3 Reduces the Growth of CRC Cells Through Regulating miR-214-3p To Activate Ferroptosis. (A) DRAW was Used to Predict miRNAs Bound to GPX4. GPX4-Related miRNAs are Expressed in Both Humans and mice. Three Databases Were Used to Predict miRNAs With Binding Sites for Human GPX4. (B) qPCR Analysis Revealed the Relative Levels of the Selected miRNAs. (C) Relative mRNA Abundance of GPX4 and miR-214-3p. (D) Validation of the Relationship Between miR-214-3p and GPX4 mRNA via a Dual-Luciferase Reporter Assay. (E) Western Blot Analysis and Relative Quantification of GPX4 Protein Expression. Compared With the Control, *P < 0.05, **P < 0.01, ***P < 0.001. Each Experiment was Conducted With n = 3 Replicates. CVB3, Coxsackievirus B3; CRC, Colorectal Cancer; qPCR, Quantitative Polymerase Chain Reaction; GPX4, Glutathione Peroxidase 4

To further determine the interaction between GPX4 and miR-214-3p, we used a luciferase reporter system and constructed 2 reporter plasmids based on miR-214-3p binding sites in the GPX4 coding sequence to generate wild-type and mutant GPX4 (WT-GPX4 and Mut-GPX4). MiR-214-3p mimics or AMO-214-3p were co-transfected with a luciferase reporter system (WT-GPX4 or Mut-GPX4) into SW620 cells. Luciferase activity was decreased in cells co-transfected with miR-214-3p, whereas AMO-214-3p reversed this inhibition in the wild-type luciferase reporter system. However, luciferase activity did not significantly differ among the groups using the mutant luciferase reporter system (Figure 3C). To further clarify the binding of miR-214-3p to GPX4, miR-214-3p or AMO-214-3p was transfected into SW620 cells. RT-qPCR was used to assay the mRNA levels of GPX4 and miR-214-3p, and western blotting analysis was performed to measure GPX4 protein expression. GPX4 mRNA and protein expression decreased in the group transfected with miR-214-3p, whereas co-transfection with AMO-214-3p increased GPX4 expression at both the mRNA and protein levels (Figure 3D,E). Thus, miR-214-3p could target GPX4.

Downregulating miR-214-3p Expression Inhibited Ferroptosis Activated by CVB3

To clarify the effect of miR-214-3p on CVB3-activated ferroptosis, we transfected AMO-214-3P into SW620 cells, followed by the addition of the ferroptosis inhibitors ferrostatin-1 (Fer-1) and deferoxamine (DFO) to the SW620 cell culture medium, respectively, and subsequent infection with CVB3. Compared with the noninfected control (NC) group, the CVB3-infected and CVB3 + AMO-NC groups showed significantly decreased cellular proliferation levels and GPX4 expression levels but increased levels of 3D, Acyl-CoA synthetase long-chain family member 4 (ACSL4), Fe²⁺, C11-BODIPY, and ROS. Compared with the CVB3 + AMO-NC group, the CVB3 + AMO-miR-214-3P, CVB3 + Fer-1, and CVB3 + DFO groups exhibited upregulated cellular proliferation levels and GPX4 levels, along with downregulated expression levels of 3D, ACSL4, Fe²⁺, C11-BODIPY, and ROS (Figure 4A,B). Nucleic acid and protein detection analyses showed that, compared with the NC group, the CVB3 and CVB3 + AMO-NC groups had significantly elevated expression levels of 3D, miR-214-3p, and ACSL4, whereas GPX4 expression displayed a downward trend. By contrast, compared to the CVB3 + AMO-NC group, the CVB3 + AMO-miR-214-3P, CVB3 + Fer-1, and CVB3 + DFO groups showed opposite trends for all indicators (Figure 4C,D). The downregulation of miR-214-3p reduced CVB3-mediated ferroptosis in CRC cells.

Figure 4.

Downregulating miR-214-3p Inhibited Ferroptosis Activated by CVB3. (A) Light Microscopy was Used to Observe Cell Pathology in Different Groups. Scale bar: 100 μm. (B) 3D, ACSL4, Fe²⁺, C11-BODIPY, ROS, and GPX4 Expressions Detected by Fluorescence Microscopy. Scale bar: 100 μm. (C) Relative mRNA Levels of 3D, miR-214-3p and GPX4. (D) 3D, ACSL4, and GPX4 Protein Levels Were Assayed by Western Blotting. Compared With the Control, *P < 0.05, **P < 0.01, ***P < 0.001. Each Experiment was Conducted With n = 3 Replicates. CBV3, Coxsackievirus B3; ROS, Reactive Oxygen Species; GPX4, Glutathione Peroxidase 4

Discussion

Previous studies have suggested oncolytic virotherapy as a novel strategy for cancer therapy and have clarified the effect of OVs in clinical trials for many cancers over the last 2 decades.^1,2 OVs infect, replicate, and spread in cancer cells, leading to tumor lysis and death. Moreover, OVs stimulate immune response of the host, which leads to cell death of cancers. ⁹ RNA viruses undergo a replication cycle and release progeny viruses, which are utilized in oncolytic virotherapy. ⁸ CVB3 is an OV applied to treat various tumors. CVB3 induces tumor destruction through direct cell lysis and innate immunity. ¹⁵

Intracellular accumulation of lipid peroxides leads to a form of programmed cell death called ferroptosis. ¹⁷ CRC cells are the malignant tumor cells that can absorb iron from the bloodstream and intestinal lumen. The deubiquitinase OTUD1 of iron responsive element binding protein 2 promotes the deubiquitination of TFR1 and enhances iron uptake and ROS production in CRC cells, thereby promoting the iron-mediated death of CRC and improving the antitumor ability of the host. ²³ Compared with adjacent normal intestinal cells, CRC cells have higher levels of iron ions and are more prone to ferroptosis. ²⁴ Therefore, ferroptosis induction is a promising therapy for treating CRC. In this study, CVB3 inhibited CRC in vitro and in vivo. In mice with CRC xenografts, the volume of tumors injected with CVB3 was significantly reduced, and the levels of ROS, Fe²⁺, and MDA were significantly increased, suggesting that CVB3 activates ferroptosis in human CRC. Moreover, the ROS, Fe³⁺ and GPX4 levels in CRC tissues increased significantly compared to those in normal intestinal tissues, indicating that ferroptosis is inhibited in human CRC.

MiRNAs play important regulatory roles in the initiation and development of CRC. Overexpressed miR-15a-3p binds to the 3′-UTR of GPX4 to inhibit GPX4 expression, which leads to increased cellular ROS and Fe²⁺ levels, MDA accumulation in vivo, and ferroptosis induction in CRC. ²⁵ CVB3 infection can induce changes in miRNA abundance, which increases CVB3 replication and accelerates pathogenicity. Using bioinformatic prediction and experimental validation, we determined that CVB3 regulated GPX4 expression through miR-214-3p. AMO-214-3p, ferrostatin-1, and DFO transfection prohibited the antitumor effect of CVB3, followed by the upregulation of GPX4 expression. Meanwhile, the 3D, ACSL4, Fe²⁺, C11-BODIPY, and ROS levels decreased, thus confirming that miR-214-3p downregulation can alleviate ferroptosis activated by CVB3. These results further validate that miR-214-3p, targeted by CVB3, may be used to treat CRC.

In contrast to the previously characterized CVB3–TFRC pathway, the miR-214-3p/GPX4 axis identified in this study represents a novel regulatory mechanism in ferroptosis. GPX4, a key negative regulator of ferroptosis, was directly suppressed by miR-214-3p. This mechanism not only elucidates the molecular basis of CVB3-induced ROS and MDA accumulation but may also offer a therapeutic strategy to circumvent chemotherapy resistance mediated by GPX4 overexpression. Furthermore, unlike the miR-15a-3p/GPX4 pathway reported by Liu et al, ²⁵ the regulation of miR-214-3p is CVB3-specific, suggesting that CVB3 potentiates its oncolytic efficacy through a distinct miRNA-driven reprogramming strategy.

In recent years, strategies for the application of OVs in CRC therapy have primarily focused on the dual mechanisms of direct oncolysis and immune activation. For example, the recombinant OV AdC6-htertΔE1A-ΔE3 induces tumor cell apoptosis by activating the p53-dependent pathway, ²⁶ whereas the IL-12-encoding oncolytic measles virus (OMV) effectively triggers colon cancer cell apoptosis via the IL-12/IFN-γ/TNF-α inflammatory response. ²⁷ However, these approaches have not fully explored the regulation of non-immunogenic cell death pathways, such as ferroptosis. To the best of our knowledge, the present study is the first to reveal that CVB3 could specifically activate ferroptosis through the miR-214-3p/GPX4 axis, a mechanism distinct from the traditional oncolytic viral modes of action. Compared to the p53 pathway dependency of AdC6-htertΔE1A-ΔE3 or the cytokine storm mediated by OMV, CVB3-induced ferroptosis offers unique advantages.GPX4 inhibition bypasses therapy resistance caused by overexpression of antiapoptotic proteins in CRC. In addition, lipid peroxide accumulation from ferroptosis may enhance antigen presentation by tumor-associated macrophages, potentially synergizing with immunotherapy.

Our study has some limitations. First, additional experiments using techniques, such as transgenic technology, are needed to further verify the mechanisms of CVB3-induced CRC cell death, immune microenvironment regulation, and antitumor immunity. Second, more clinical samples are needed to further validate the correlation between human CRC and ferroptosis and to explore the antitumor effects of CVB3 on other ferroptosis defense mechanisms. Finally, although this study revealed the mechanism by which CVB3 could activate ferroptosis through the miR-214-3p/GPX4 axis, the following issues remain. As CVB3 is a cardiotropic virus, its potential toxicity to normal tissues was unclear in the animal models of this study. Moreover, CVB3 may induce off-target effects in non-target tissues through viral receptors; however, this study did not investigate its effect on normal intestinal epithelial cells or other organs. Future studies should combine genetic engineering technologies to optimize the tumor-targeting specificity of CVB3 and reduce the off-target risks.

CVB3 may have a significant cytotoxic effect on CRC cells because it regulates miR-214-3p to activate ferroptosis, which could induce CRC cell death. This study offers novel concepts and a relevant basis for the use of CVB3 as an OV for CRC treatment.

Conclusion

miR-214-3p expression was upregulated during CVB3 infection of CRC tissues and cells, activating the ferroptosis pathway and promoting tumor cell death. The use of CVB3, an OV, may be a novel approach for CRC therapy.

Appendix.

Abbreviations

CRC

colorectal cancer

CVB3

Coxsackievirus Group B type 3

OV

oncolytic virus

CV

Coxsackievirus

CVB

Coxsackievirus

GPX4

glutathione peroxidase 4

CPE

cytopathic effect

ORCID iD

Rui Huang https://orcid.org/0000-0003-0884-4278

Ethical Consideration

The study protocol was reviewed and approved by the Medical Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (Harbin, Heilongjiang Province, China) (approval numbers: YJSKY2023-024 [Clinical Research Ethics Approval]; YJSDW2023-118 [Animal Experiment Ethics Approval]) on September 4, 2023.