Fig 1.
DENV infection impairs mitochondrial dynamics.
(A and B) Diagram (A) and the results (B) of mitochondria intermixing assay. Mitochondria of A549 cells were labeled by transfection with mitoYFP or mitoCherry to distinguish the origin of each mitochondrion. Cells were mock or infected with DENV serotype 2 (multiplicity of infection [moi] 10) for 24 h before HVJ-E-mediated cell fusion. The fused cell hybrids with or without DENV infection were magnified and analyzed for green and red fluorescence intensity by use of ZEN lite 2011 (Carl Zeiss MicroImaging GmbH). Green: mitoYFP; red: mitoCherry; magenta: DENV NS3. (C to I) Time course study of DENV infection in human A549 cells with moi 10 for the indicated hours. Culture supernatants were harvested for virus titration (C) and release of lactic dehydrogenase (LDH) (D). Cell lysates were harvested for western blot analysis with the indicated antibodies (E) and RT-qPCR (F to I) for the indicated genes. Data are mean ± SD (n = 3 per group). The data for LDH release was compared by one way ANOVA and Bonferroni multiple-comparison test with use of Prism 5 (GraphPad; La Jolla, CA, USA). ns, no significance; *, p<0.05. Nocodazole treatment (NOC; 100 ng/ml for 16 h) served as positive control to induce Drp1 phosphorylation at S616 residue. RQ, relative quantification. (J) STAT1-/- mice were inoculated with DENV (serotype 2, strain NGC-N) by an intraperitoneal plus intracerebral route. The peripheral blood mononuclear cells (PBMC) were isolated on day 0, 1, and 2 as indicated (n = 2 for each time point) and then sampled for western blot analysis with the indicated antibodies.
Fig 2.
DENV infection suppresses MFN1-mediated mitochondrial dynamics.
(A) Anti-HA and MitoTracker staining of intracellular localization of ectopic-expressed MFN1 and mitochondrial morphology in A549+on/HA-MFN1 cells treated with or without Dox (1 μg/ml, 24 h). (B) High-power analysis of MFN1-overexpressing A549 cells as described in A. The magnified area is marked by square. (C) A549+on/HA-MFN1 cells were infected with DENV serotype 2 (moi 10) for 36 h and then stained with MitoTracker for 30 min before Dox induction. The mitochondrial dynamics of mock- and DENV-infected cells were photographed side by side every 30 sec continuously after Dox treatment. A representative cell for each condition is shown. (D) A549+on/HA-MFN1 cells were infected, stained, and treated with Dox as in panel C. Cells were fixed after 6 h of Dox treatment and then mitochondrial morphologic features were quantified by analyzing immunofluorescent images (one set of random selected fields in each group is shown at the right) from 49 individual fields by high-content analysis (ImageXpress Micro Imaging XL System, Molecular Devices). Data at left are mean ± SD from 49 individual fields of each sample.
Fig 3.
Effects of MFN1 or MFN2 overexpression on DENV infection.
(A) Western blot analysis of A549+on/HA-MFN1 cells induced with Dox (1 μg/ml) or not for 18 h, then infected with DENV serotype 2 for 24 or 48 h by the indicated moi. The relative ratios of band intensity were quantified by ImageJ. p.i. (h): hours post infection. (B) DENV plaque-forming assay of culture supernatants from A549+on/HA-MFN1 cells cultured with or without Dox (1 μg/ml) for 18 h, then infected with DENV (moi 0.1 or 10) as indicated. Data in panels B-D and F-G are mean ± SD (n = 3 per group) and were compared by two-tailed Student’s t test. (C) RT-qPCR analysis of IFNβ mRNA expression at indicated time point in DENV-infected (moi 10) A549+on/HA-MFN1 cells with or without an 18 h-Dox-pretreatment. (D) Analysis of antiviral activity against dSinF-Luc/2A virus in culture media from A549+on/HA-MFN1 cells with DENV infection (moi 5, 48 h) with or without 18-h Dox pretreatment. (E) Western blot analysis of A549+on/HA-MFN2 cells induced with Dox (1 μg/ml, 18 h) or not, then infected with DENV (moi 10) for the indicated time. (F and G) Analysis of the culture supernatants derived from panel E for DENV titer (F) and LDH release (G). (H) Flow cytometry of A549+on/HA-MFN2 cells double stained with JC-1 and annexin V. Cells were infected with DENV (moi 5) for 48 h with or without 18-h Dox pre-treatment. Decreased red/green fluorescence ratio of JC-1 represents disrupted MMP.
Fig 4.
DENV cleaves human MFN1 and MFN2.
(A) Alignment of amino acid sequences surrounding the MFN1 and MFN2 cleavage sites of DENV protease. The wild-type (WT) and mutated MFNs used in this study and the relative position of each domain are illustrated. (B) Western blot analysis of the cotransfection of C-terminal V5-tagged human MFN1 or MFN2 with Flag-tagged WT or mutated viral protease of DENV or JEV in A549 cells for 24 h. WT: wild-type; S135A: protease-dead mutant. (C) Coimmunoprecipitation analysis of A549 cells cotransfected with V5-tagged MFN1 or MFN2 with Flag-tagged NS2B3(S135A) of DENV or JEV for 24 h. (D) Coimmunoprecipitation analysis of A549 cells expressing S135A mutated DENV NS2B3 transfected with V5-tagged WT or mutated MFN1 or MFN2 for 24 h. (E) Western blot analysis of Flag-tagged DENV protease NS2B3 cotransfected with the indicated (WT or mutant) C-terminal V5-tagged MFNs constructs in A549 cells in the absence (lane 1–5) or presence (lane 6–10) of pan-caspase inhibitor (zVAD; 100 μM) for 24 h. Filled arrow: full-length; open arrow: cleaved product; star: non-specific band.
Fig 5.
Suppression of MFN-triggered mitochondrial hyperfusion can be attributed to cleavage by DENV protease.
Confocal microscopy of A549 cells cotransfected with Flag-tagged DENV protease and the indicated MFN constructs for 24 h. Arrows indicate the cells expressing both Flag-tagged DENV protease and HA-tagged MFN. Green: anti-HA; magenta: anti-Flag; red: MitoTracker; blue: DAPI.
Fig 6.
The cleaved MFN fragments were further degraded by host proteasome machinery.
(A) Western blot analysis of A549+on/MFN1 (lane 1–6) and /MFN2 (lane 7–12) cells with Dox treatment (all lanes) and DENV infection (lane 3–6 and 9–12). The cells were infected with DENV serotype 2 (moi 5) for 24 h, then incubated in culture media containing Dox (1 μg/ml) with (lanes 5–6 and 11–12) or without MG132 (0.5 μM) for 16 h. (B) Western blot analysis of four different serotypes DENV-infected A549 cells incubated with MG132-containing medium. (C) Western blot analysis of endogenous MFN1 and 2 cleavage in A549 stable cells overexpressing WT or protease-dead mutant (S135A) of DENV NS2B3 with or without 18-h treatment with MG132 (0.5 μM). Filled arrow: full-length; open arrow: cleaved product; star: non-specific band. The longer exposure images for endogenous MFN1 and 2 signals are also shown in panels B and C.
Fig 7.
DENV protease governs mitochondrial morphology.
(A) Confocal microscopy of A549 stable cells overexpressing DENV NS2B3 wild type (WT) or protease-dead mutant (S135A). Mitochondria were labeled by stably expressing mitoYFP and the magnified area is marked by a rectangle. Green: mitoYFP; magenta: DENV NS3; blue: DAPI. (B and C) Diagram (B) and the results (C) of mitochondria intermixing assay in cell hybrid harboring WT or mutated DENV protease. Mitochondria of A549 cells stably expressing WT or S135A-mutated DENV protease were labeled by transfection with mitoYFP or mitoCherry. Cells were fused by HVJ-E-mediated cell fusion, and the cell hybrids with or without DENV protease activity were magnified and analyzed for green and red fluorescence intensity by use of ZEN lite 2011 (Carl Zeiss MicroImaging GmbH). Green: mitoYFP; red: mitoCherry; magenta: DENV NS3.
Fig 8.
Silencing MFN1 or MFN2 reveals their distinct roles in DENV infection.
(A and B) RT-qPCR (A) and western blot (B) validation of A549 cells stably expressing shRNA targeting LacZ, MFN1 or MFN2. Data in panels A and C-F are mean ± SD (n = 3 per group) and were compared by two-tailed Student’s t test. Quantification in panel B is the relative ratio of the indicated protein to actin expression. (C) LDH release assay of shLacZ-, shMFN1- or shMFN2-expressing A549 cells infected with DENV serotype 2 (moi 10) for the indicated time. (D) DENV plaque-forming assay of shLacZ-, shMFN1- or shMFN2-expressing A549 cells infected with DENV (moi 0.1) for 24 or 48 h. (E) Analysis of antiviral activity in culture media from indicated cells with DENV infection (moi 5, 48 h) against dSinF-Luc/2A virus. (F) RT-qPCR analysis of IFNβ mRNA level in DENV infected (moi 10) A549-shLacZ, -shMFN1, and -shMFN2 cells for the indicated time. (G) Western blot analysis of DENV infected (moi 10) A549 cells expressing shLacZ, shMFN1 or shMFN2. The relative ratios of band intensity were quantified as indicated. (H) Flow cytometry with JC-1 staining of shLacZ-, shMFN1 or shMFN2-expressing A549 cells infected with DENV (moi 5) for 30 h. Decreased red/green fluorescence ratio of JC-1 represents disrupted MMP. CCCP, carbonyl cyanide m-chlorophenyl hydrazone, an ionophore that disrupts MMP. (I) Plaque forming assay of vesicular stomatitis virus (VSV) by using A549 cells expressing shLacZ, shMFN1 or shMFN2 as indicated. pfu, plaque-forming unit.
Fig 9.
MFN1 and MFN2 manipulate MAVS-triggered signaling differently.
(A and B) Western blot (A) and RT-qPCR (B) analysis of A549+on/myc-MAVS cells with stably expressing shRNA targeting the indicated genes after 48 h of Dox treatment. RT-qPCR data are mean ± SD (n = 3 per group) and were compared by two-tailed Student’s t test. (C) Dual-luciferase activity analysis of A549 cells cotransfected with the indicated MFNs (cloned in pcDNA3.1 vector; 0.5 μg), stimulator or control (Flag-MAVS/pcDNA3 or Flag-GFP/pcDNA3; 0.4 μg), IRF3/pCR3.1 (0.3 μg), p125-Luc (0.2 μg) and pRL-TK (0.1 μg) for 24 h. The firefly luciferase activity (p125-Luc) was normalized to that of renilla luciferase (pRL-TK) and the relative luciferase activities are shown. Data are mean ± SD (n = 3 per group) and were compared by two-tailed Student’s t test.
Fig 10.
Blocking mitochondrial fusion attenuates RLR signaling and facilitates DENV infection.
(A) Mitochondrial morphology analysis of A549 stable cell line overexpressing control GFP or V5-tagged MFN1-T109A. Morphology and subcellular localization of mitochondria and MFN1-T109A were revealed by MitoTracker and anti-V5 staining, respectively. (B) Western blot analysis of GFP- and MFN1-T109A-expressing A549 cells upon DENV infection. Cells were infected with DENV serotype 2 (moi 10) and harvested at indicated time. (C) MMP measurement of vector control (Ctrl) or MFN1-T109A-expressing cells after DENV infection. Cells were infected with DENV (moi 5) for 36 h and harvested for flow cytometry analysis with JC-1 staining. (D to G) Time course study of DENV infection in GFP- and MFN1-T109A-A549 cells. Samples were harvested for the quantification of intracellular IFNβ mRNA (D) and DENV RNA (E) by RT-qPCR, and for the measurement of LDH release (F) and cell viability by trypan blue exclusion (G). Data are mean ± SD (n = 3 per group) and were compared by two-tailed Student’s t test.
Fig 11.
DENV governs MFN-mediated signaling by protein cleavage.
MFN1 and MFN2 are involved in the initiation step of mitochondrial fusion by tethering mitochondria together. DENV infection suppresses mitochondrial fusion and mitochondrial dynamics by cleaving MFNs via viral protease NS2B3. Because MFN1 enhances host antiviral signaling and MFN2 maintains mitochondrial membrane potential during DENV infection, the cleavage of both MFNs by DENV protease would attenuate interferon production and increases cell death of DENV-infected cells.