Cyclooxygenase-2 facilitates dengue virus replication and serves as a potential target for developing antiviral agents

Abstract

Cyclooxygenase-2 (COX-2) is one of the important mediators of inflammation in response to viral infection, and it contributes to viral replication, for example, cytomegalovirus or hepatitis C virus replication. The role of COX-2 in dengue virus (DENV) replication remains unclear. In the present study, we observed an increased level of COX-2 in patients with dengue fever compared with healthy donors. Consistent with the clinical data, an elevated level of COX-2 expression was also observed in DENV-infected ICR suckling mice. Using cell-based experiments, we revealed that DENV-2 infection significantly induced COX-2 expression and prostaglandin E₂ (PGE₂) production in human hepatoma Huh-7 cells. The exogenous expression of COX-2 or PGE₂ treatment dose-dependently enhanced DENV-2 replication. In contrast, COX-2 gene silencing and catalytic inhibition sufficiently suppressed DENV-2 replication. In an ICR suckling mouse model, we identified that the COX-2 inhibitor NS398 protected mice from succumbing to life-threatening DENV-2 infection. By using COX-2 promoter-based analysis and specific inhibitors against signaling molecules, we identified that NF-κB and MAPK/JNK are critical factors for DENV-2-induced COX-2 expression and viral replication. Altogether, our results reveal that COX-2 is an important factor for DENV replication and can serve as a potential target for developing therapeutic agents against DENV infection.

Figure 1. DENV induces COX-2 expression and PGE₂ production in DF patients, DENV-infected mice, and human hepatoma cells.

(A and B) Elevated COX-2 expression and PGE₂ levels in the blood of dengue fever patients. COX-2 mRNA and PGE₂ levels in blood samples from 13 clinical DF patients and 6 healthy donors were determined by RT-qPCR or ELISA, respectively. (C) The induced COX-2 expression in DENV-2-infected ICR suckling mice. Six-day-old suckling mice were injected with 2.5 × 10⁵ pfu of DENV-2 or heat-inactivated DENV-2 (iDENV) by intracerebral injection. Each group comprised six suckling mice (n = 6). Six days after inoculation, COX-2 mRNA levels of mouse brain tissues were determined by RT-qPCR. DENV-2 time-dependently induced (D) COX-2 protein expression, (E) COX-2 RNA replication, and (F) PGE₂ production. Huh-7 cells were infected with DENV-2 at an MOI of 0.1, and the cell lysate and cellular RNA were extracted at the indicated time points (24, 48, and 72 hpi). Western blotting was performed with anti-COX-2, anti-NS2B, and anti-GAPDH antibodies. Relative RNA levels of DENV-2 and COX-2 were determined by RT-qPCR following the normalization of cellular gapdh mRNA levels. Supernatants were collected at the indicated time points and subjected to a PGE₂ ELISA assay. All data from cell-based experiments are indicative of at least three independent experiments, with each measurement carried out in triplicate. Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 2. COX-2 overexpression and PGE₂ treatment increase DENV-2 replication.

COX-2 overexpression induced DENV-2 replication in (A) DENV-2 replicon cells and (B and C) the DENV infection system. Huh-7-D2-FLuc-SGR-Neo DENV replicon reporter cells were transfected with pcDNA4/Myc or pcDNA4-COX-2-Myc at the indicated concentrations. After 3 days of incubation, the cell lysates were subjected to a luciferase activity assay. Huh-7 cells were transfected with pcDNA4/Myc or pcDNA4-COX-2-Myc at the indicated concentrations, and the transfected cells were infected with DENV-2 at an MOI of 1. After 3 days of incubation, the cell lysates and cellular RNA were subjected to western blotting and RT-qPCR. (D) COX-2 overexpression increased DENV-2 propagation. The transfected Huh-7 cells were infected by DENV-2 at a MOI of 1 for 3 days. Supernatants were collected and subjected to a viral plaque assay. PGE₂ treatment induced DENV-2 replication in (E) viral replicon cells and (F and G) the DENV infection system. Huh-7-D2-FLuc-SGR-Neo DENV replicon reporter cells were treated with PGE₂ at the indicated concentrations for 3 days and the cell lysates were subjected to a luciferase activity assay. Huh-7 cells were infected with DENV-2 at an MOI of 1, and the infected cells were treated with PGE₂ at the indicated concentrations for 3 days. Western blotting was performed with anti-NS2B, anti-Myc, and anti-GAPDH antibodies. Relative RNA levels of DENV-2 was determined by RT-qPCR following the normalization of cellular gapdh mRNA levels. (H) PGE₂ treatment induced DENV-2 propagation. Huh-7 cells were infected with DENV-2 at an MOI of 1 and then treated with PGE₂. Supernatants were collected and subjected to a viral plaque assay. (I) PGE₂ treatment induced DENV-2 NS5 polymerase activity. Huh-7 cells were cotransfected with p(+)RLuc-(−)DV-UTRΔC-Fluc reporter template (0.5 μg) and pcDNA-NS5-Myc expression plasmid (0.5 μg), and the transfected cells were treated with PGE₂ at the indicated concentrations for 3 days. The cell lysates were subjected to a Dual-Glo Luciferase Assay. All data were indicative of at least three independent experiments, with each measurement carried out in triplicate. Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 3. Reduction of COX-2 expression and catalytic activity by shRNA or NS398 reduce DENV replication.

COX-2 shRNA reduced DENV replication in (A) DENV replicon cells and (B to D) a DENV infection system. Huh-7-D2-FLuc-SGR-Neo DENV replicon cells were transfected with GFP or COX-2 shRNA at the indicated concentrations for 3 days and the cell lysates were subjected to a luciferase activity assay and western blotting. Huh-7 cells were transfected with COX-2 shRNA at the indicated concentrations, and the transfected cells were infected with DENV-2 at an MOI of 1. After 3 days of treatment, the cell lysate, cellular RNA and supernatants were analyzed by western blotting, RT-qPCR or plaque assay, respectively. NS398 reduced DENV replication in (E) DENV replicon cells and (F to H) a DENV infection system. Huh-7-D2-FLuc-SGR-Neo DENV replicon cells were treated with NS398 at different concentrations (0, 5, 10, 20, and 40 μM) for 3 days, and the cell lysates were subjected to a luciferase activity assay and western blotting. Huh-7 cells were infected with DENV-2 at an MOI of 1 and then treated with NS398 at different concentrations (0, 5, 10, 20, and 40 μM) for 3 days. Western blotting was performed with anti-COX-2, anti-NS2B, and anti-GAPDH antibodies. The relative RNA level of DENV-2 was determined by RT-qPCR following normalization to the cellular gapdh mRNA level. All data are indicative of at least three independent experiments, with each measurement performed in triplicate. Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 4. COX-2 expression is required for viral replication.

(A to C) Exogenous COX-2 expression restored DENV-2 protein synthesis, RNA replication and viral propagation in COX-2 shRNA-transfected cells. Huh-7 cells were cotransfected with COX-2 shRNA (1.0 μg) and pCMV-COX-2-Myc (0.25, 0.5, and 1.0 μg), followed by DENV-2 infection at an MOI of 1. (D to F) Exogenous COX-2 expression restored DENV-2 protein synthesis, RNA replication and viral propagation in NS398-treated cells. Huh-7 cells were transfected with pcDNA4/Myc (0.5 μg) or pCMV-COX-2-Myc (0.25, 0.5, and 1.0 μg), followed by DENV-2 infection at an MOI of 1. The cells were treated with DMSO or 40 μM NS398 for 3 days. Western blotting was performed with anti-NS2B, anti-Myc, and anti-GAPDH antibodies. The relative RNA level of DENV-2 was determined by RT-qPCR following normalization to the cellular gapdh mRNA level. All data are indicative of at least three independent experiments, with each measurement performed in triplicate. Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 5. NS398 protects ICR suckling mice from life-threatening DENV-2 infection.

Six-day-old suckling mice were injected with 2.5 × 10⁵ pfu of DENV-2 or heat-inactivated DENV-2 (iDENV), as a negative control, by intracerebral injection. Then, NS398 was injected into DENV-2-infected mice at a dose of 1 and 5 mg/kg by intracerebral injection at 1, 3, and 5 dpi. Following inoculation, (A) clinical score, (B) mouse body weights, and (C) survival rate were recorded daily. The symptoms of the clinical score are shown as follows: 0 for no illness symptoms, 1 for ruffled fur and anorexia, 3 for paralysis, 4 for lethargy, and 5 for moribund. (D) Body weight and (E) survival rate of the DENV-2-infected mice were recorded daily until 11 dpi. Each group comprised 12 suckling mice (n = 12). Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 6. DENV-2 elevates COX-2 promoter activation through mediation of NF-κB and C/EBP binding elements.

(A) NF-κB and C/EBP were critical binding elements for DENV-2-induced COX-2 promoter activity. Huh-7 cells were cotransfected with 0.2 μg of pCMV-Renilla-Luc and COX-2 promoter-linked firefly luciferase reporter vectors containing various deletions of transcriptional factor binding elements, including WT, ΔNF-κB, and ΔNF-κB/C/EBP. The transfected cells were infected by DENV-2 at an MOI of 1, and the cell lysates were subjected to a Dual-Glo Luciferase Assay at 2 dpi. (B) DENV-2 infection induced the activation of NF-κB and C/EBP binding elements. Huh-7 cells were cotransfected with 0.2 μg of pCMV-Renilla-Luc and pNF-κB-Luc, pC/EBP-Luc, pCRE-Luc, or pAP-1-Luc. Then, the transfected cells were infected with DENV-2 at an MOI of 1. The cell lysates were subjected to a Dual-Glo Luciferase Assay at 2 dpi. All data are indicative of at least three independent experiments, with each measurement carried out in triplicate. Error bars are expressed as mean ± SD of three independent experiments; *P < 0.05.

Figure 7. NF-κB and MAPK/JNK-mediated C/EBP are responsible for DENV-2-induced COX-2 expression and viral replication.

(A) DENV-2 induced activation of the NF-κB signaling pathway. Huh-7 cells were infected by DENV-2 at an MOI of 1 and the cell lysates were extracted at the indicated time points. Western blotting was performed and the relative blot intensities were quantified by densitometry scanning. (B) CAPE significantly suppressed DENV-2-induced COX-2 expression. Huh-7 cells were pretreated with DMSO or CAPE (20 μM) for 2 h and then infected with DENV-2 at an MOI of 1. The cell lysates were analyzed at 2 dpi by western blotting. (C and D) CAPE dose-dependently suppressed DENV-2 protein synthesis and RNA replication. Huh-7 cells were infected by DENV-2 and treated with CAPE at different concentrations (0, 5, 10, and 20 μM). After 3 days of treatment, the cell lysates and cellular RNA were analyzed by western blotting or RT-qPCR, respectively. (E) DENV-2 induced the activation of the MAPK/JNK pathway. Huh-7 cells were infected by DENV-2, and the cell lysates were extracted at the indicated time points. Western blotting was performed and the relative blot intensities were quantified by densitometry scanning. (F) SP600125 significantly suppressed DENV-2-induced COX-2 expression. Huh-7 cells were pretreated with DMSO or SP600125 (20 μM) for 2 h, and then, the cells were infected by DENV-2. After 2 days of treatment, cell lysates were subjected to western blotting. (G and H) SP600125 dose-dependently suppressed DENV-2 protein synthesis and RNA replication. Huh-7 cells were infected by DENV-2, and the cells were treated with DMSO or SP600125 at different concentrations (0, 5, 10, and 20 μM). After 3 days of treatment, the cell lysates and cellular RNA were analyzed by western blotting or RT-qPCR, respectively. GAPDH served as a loading control in western blotting. The relative RNA level of DENV-2 was determine by RT-qPCR following normalization to the cellular gapdh mRNA level. All data are indicative of at least three independent experiments, with each measurement carried out in triplicate. Error bars are expressed as the mean ± SD of three independent experiments; *P < 0.05.

Figure 8. Proposed model to illustrate the mechanism of increased COX-2 expression and PGE₂ production during DENV infection.

DENV infection induces COX-2 expression through activation of the NF-κB and MAPK/JNK-mediated C/EBP signaling pathways. Phosphorylated NF-κB and activated C/EBP are translocated into the nucleus and bind to the COX-2 promoter region. COX-2 induction and elevated PGE₂ production by DENV infection lead to enhanced activity of the viral polymerase and viral propagation.