Dengue virus serotype 2 blocks extracellular signal-regulated kinase and nuclear factor-κB activation to downregulate cytokine production

Abstract

Background: Dengue virus (DENV) infection is the most common mosquito-borne viral disease threatening human health around the world. Type I interferon (IFN) and cytokine production are crucial in the innate immune system. We previously reported that DENV serotype 2 (DENV-2) induced low levels of interferon regulatory factor 3 and NF-κB activation, thus leading to reduced production of IFN-β in the early phase of infection. Here, we determined whether DENV infection not only hampers type I IFN activation but also cytokine production triggered by Toll-like receptor (TLR) signaling.

Methodology/principal findings: We used quantitative RT-PCR and found that only low levels of IFN-β and inflammatory cytokines such as interleukin 10 (IL-10), IL-12 and tumor necrosis factor α (TNFα) mRNA were detected in DENV-2-infected bone-marrow-derived dendritic cells. Furthermore, DENV-2 infection repressed cytokine production triggered by TLR signaling. To elucidate the molecular mechanisms underlying this suppression event, we measured NF-κB activation by p65 nuclear translocation and luciferase reporter assay and found that NF-κB activation triggered by TLR ligands was blocked by DENV-2 infection. As well, extracellular signal-regulated kinase (ERK) activity was suppressed by DENV-2 infection.

Conclusions/significance: To downregulate the host innate immunity, DENV-2 by itself is a weak inducer of type I IFN and cytokines, furthermore DENV-2 can also block the TLR-triggered ERK-NF-κB activation and cytokine production.

Figure 1. Low levels of IFN-β and cytokine production in DENV-2–infected BMDCs.

Bone-marrow–derived dendritic cells (BMDCs) were mock-infected or infected with DENV-2 at a multiplicity of infection (MOI) of 5 or stimulated with the toll-like receptor (TLR) ligands polyI:C (100 µg/ml) or CpG (1 µg/ml), for various times. Quantitative real-time PCR (qPCR) analysis of mRNA levels of IFN-β (A), IL-10 (B), IL-12p40 (C), TNFα (D) and DENV-2 viral 5′ UTR RNA (E) normalized to the internal control HPRT. Data are mean ±SD of 3 determinations. Data for DENV-2-induced cytokine expression is magnified at the right of each panel. *p≤0.05 compared with mock infection. (F) DENV-2-infected BMDCs were immunofluorescently stained with anti-DENV-2 NS3 antibody (green fluorescence, a–f). DAPI staining indicates the location of cell nucleus (blue fluorescence, g–l).

Figure 2. DENV-2 infection blocks TLR-triggered IFN-β and cytokine production.

J774A.1 macrophages were mock-infected or infected with DENV-2 (MOI 5) for 24 h before TLR ligand stimulation. qPCR analysis of mRNA expression of IFN-β and IL-10 on cells coincubated with lipopolysaccharide (LPS; 100 ng/ml) (A) or polyI:C (100 µg/ml) (B) for the indicated times. (C) BMDCs were mock-infected or infected with DENV-2 (MOI 5) for 24 h before polyI:C (100 µg/ml) stimulation. qPCR analysis of mRNA expression of IFN-β and IL-10 was analyzed at the indicated times. Normalization was done with the expression level of internal control HPRT. Data are mean±SD of 3 determinations. *, p<0.01; **, p<0.005.

Figure 3. NF-κB activation triggered by LPS and polyI:C is suppressed by DENV-2 infection.

(A) Dual luciferase assay of reporter activity. A549 cells (1×10⁵) transfected with NF-κB-Luc reporter (0.7 µg) and pRL-TK (0.02 µg), were infected with DENV-2 for 24 h and then were stimulated with LPS or polyI:C (both 1 µg/ml) for various times. Data are mean±SD from 3 determinations. *, P≤0.001 (B) The cells lysates collected from cells as described in panel A were subjected for immunoblotting assay to detect the expression of DENV-2 NS3 and actin.

Figure 4. NF-κB p65 nuclear translocation triggered by LPS is blocked by DENV-2 infection.

(A) Vero cells were infected with DENV-2 for 32 h before stimulation with LPS (1 µg/ml). After 6 h of LPS treatment, the localization of NF-κB p65 was determined by immunostaining with anti-NF-κB p65 antibody (red fluorescence, panels a–d), and DENV-2 infection was determined by anti-DENV-2 NS3 antibody (green fluorescence, panels e–h). Representative cells from the same field are shown for each experimental group. (B) A549 cells were infected with DENV-2 for 24 h (panels a and d), or treated with polyI:C (2 µg) by transfection for 24 h prior to DENV-2 infection (panels b and e) or 24 h post DENV-2 infection (panels c and f). DENV-2 infection was monitored by IFA with anti-DENV-2 NS3 antibody (green fluorescence) and the cell nuclei were counter stained by DAPI (blue fluorescence).

Figure 5. TLR-mediated ERK activation is blocked by DENV-2 infection.

(A) A549 cells (5×10⁵) were infected with DENV-2 (MOI 5), cell lysates were harvested at the indicated hours post infection (p.i.), and underwent immunoblotting with antibodies (1: 1000 dilution) against phospho-ERK1/2 (p-ERK1/2), total ERK1/2 (t-ERk1/2), DENV-2 NS3, and actin as indicated. (B) Immunoblotting analysis of mock- and DENV-2-infected (MOI 5) A549 cells (1×10⁶) stimulated with polyI:C (1 µg/ml) or LPS (1 µg/ml) for 24 h. (C) The ratio of phospho-ERK to total ERK in panel B was quantified and calculated with BioSpectrum Image System software (UVP). The mock infection without stimulation is used as the normalization control.