Japanese Encephalitis Virus NS4A Protein Interacts with PTEN-Induced Kinase 1 (PINK1) and Promotes Mitophagy in Infected Cells

Abstract

The nonstructural protein 4A (NS4A) of flaviviruses has been implicated as a "central organizer" of the membrane-bound replication complex during virus replication. However, its role in the host responses to virus infection is not understood. Using the yeast-two-hybrid library screen, we identified a multitude of host proteins interacting with the Japanese encephalitis virus (JEV) NS4A protein. Several of these interacting proteins are known to localize to the mitochondria. One of these proteins was PTEN-induced kinase 1 (PINK1), a serine/threonine-protein kinase known for its role in mitophagy. Here, we demonstrate the JEV-NS4A localization to the mitochondria and its interaction with PINK1 in Huh7 cells during JEV infection. The JEV-infected cells showed an enhanced mitophagy flux with a concomitant decline in the mitochondrial mass. We present data showing that JEV-NS4A alone was sufficient to induce mitophagy. Interference with mitochondrial fragmentation and mitophagy resulted in reduced virus propagation. Overall, our study provides the first evidence of mitochondrial quality control dysregulation during JEV infection, largely mediated by its NS4A protein. IMPORTANCE The JEV-infected mammalian cells show an enhanced mitophagy flux with a concomitant decline in the mitochondrial mass. We show that the NS4A protein of JEV localized to the mitochondria and interacted with PINK1 in Huh7 cells during infection with the virus and demonstrate that JEV-NS4A alone is sufficient to induce mitophagy. The study provides the first evidence of mitochondrial quality control dysregulation during JEV infection, largely mediated by its NS4A protein.

Keywords: flavivirus; mitochondria; protein interactome; yeast-two-hybrid.

FIG 1

The mouse JEV-NS4A interactome. (A) The Y2H screen of the mouse brain cDNA library with JEV-NS4A identified 75 interacting protein partners. Each node in the left panel represents an identified protein, and edges represent protein-protein associations not necessarily representing physical interactions. The pathway enrichment analysis (right panel) was conducted with the help of the STRING (version 10.5) bioinformatics tool at the confidence level of 0.15. A P value of <0.05 was used as the cutoff criterion. The red nodes in the interaction map (left panel) represent proteins from the mitochondrion cellular component, and the purple nodes denote the proteins enriched in Parkinson’s disease. (B) The subinteractome of the genes enriched for the Parkinson’s disease (PD) pathway (left panel). The pathway enrichment analysis (right panel) was conducted with the help of the STRING (version 10.5) bioinformatics tool at the confidence level of 0.15. A P value of <0.05 was used as the cutoff criterion. The red nodes (left panel) depict the proteins enriched in PD, while the purple nodes depict proteins enriched in Alzheimer’s disease.

FIG 2

Subcellular localization of JEV-NS4A to ER and mitochondria. (A) Huh7 cells were infected with JEV (MOI 1) and 48 h later stained with MitoTracker Red, fixed with 4% PFA, and immune-stained with JEV-NS4A and SERCA2 antibody (ER marker). These cells were then stained with respective Alexa fluor secondary antibodies, stained with DAPI for nuclei, and visualized under a confocal microscope (left panel). The colocalization is depicted as a line profile in the right panel. (B) Mock- and JEV-infected Huh7 cells harvested at 24 and 48 hpi were subjected to fractionation. We performed Western blotting on the cytosolic, mitochondrial, and microsomal fractions with JEV-NS4A antibody. GAPDH, VDAC, and calnexin were used as markers for cytosol, mitochondria, and ER, respectively.

FIG 3

Interaction of JEV-NS4A with PINK1 in silico. The molecular docking studies were undertaken followed by molecular dynamic simulations to predict the NS4A-PINK1 complex stability. (A) The best-docked pose in the form of the lowest docking energy is shown in the cartoon. The JEV NS4A is shown in blue, whereas PINK1 is shown in yellow. The dotted rectangle denotes the interaction zone in the complex. (B) Histogram of protein-protein docking score. The dotted rectangular box is the region selected for NS4A-PINK1 complex analysis. (C) The molecular dynamic simulation graph highlighting the stability of the NS4A-PINK1 trajectory is shown in green. PINK1 alone is shown in yellow, and JEV-NS4A alone is shown in blue.

FIG 4

Colocalization of JEV-NS4A with PINK1. (A) Huh7 cells were mock-infected or infected with JEV (MOI 1) and fixed at 48 hpi followed by immunostaining with antibodies against PINK1, JEV-NS4A, and respective Alexa fluor secondary antibodies. Images were taken using the Leica TCS SP8 confocal microscope. The inset has zoomed-in yellow spots depicting JEV-NS4A and PINK1 colocalization marked by the arrows. The quantitation of colocalization by Pearson’s correlation coefficient is depicted as a graph in the top right corner. A total of 21 region of interests (ROIs) each (1 cell equivalent to 1 ROI) were quantified. The data between the mock- and JEV-infected cells were compared by the Student’s t test (***, P < 0.001). (B) Huh7 cells were transfected with plasmids expressing JEV-NS4A tagged with HA (NS4A-HA) or the empty-HA vector and GFP-tagged PINK1 (PINK1-GFP). The cells were fixed 48 h posttransfection and immunostained with anti-HA antibody followed by the corresponding Alexa fluor secondary antibody. Images were captured using the Leica TCS SP8 confocal microscope. The quantitation of colocalization by Pearson’s correlation coefficient is depicted as a graph in the top right corner. A total of 18 ROIs each (1 cell equivalent to 1 ROI) were quantified. The data between the empty-HA and NS4-HA transfected cells were compared with the Student’s t test (***, P < 0.001).

FIG 5

Interaction of JEV-NS4A with PINK1 in mammalian cells. (A) HEK293T cells were transfected with the plasmid expressing GFP-tagged PINK1 and 24 h later were mock-infected or infected with JEV (MOI 3). Cell lysates were prepared 48 hpi and used for pulldown with GFP-trap beads. The immunoprecipitate (IP), flowthrough, and input (10% of lysate used for immunoprecipitation) were subjected to Western blotting. (B) Huh7 cells were mock-infected or infected with JEV (MOI 3) and fixed at 24 hpi. These cells were subjected to the Duolink proximity ligation assay (PLA). The cell nucleus was stained with DAPI. Imaging was done using the Leica TCS SP8 confocal microscope. Positive interaction between NS4A and PINK1 is detected as discrete fluorescent red spots (PLA signals). Representative confocal images are shown in the left panel. The right panel depicts the quantification of the number of PLA signals per cell in six different fields. Image analysis was done using ImageJ software. (C) Huh7 cells were cotransfected with plasmids expressing PINK1-GFP and NS4A-HA or empty-HA, fixed at 48 hpi, and subjected to PLA. Imaging was done using the Leica TCS SP8 confocal microscope. Representative confocal images are shown in the left panel. The right panel depicts the quantification of the number of PLA signals per cell in nine different fields. The Student’s t test was used to determine the statistical significance of the difference between the JEV-infected and mock-infected cells (***, P < 0.001).

FIG 6

JEV infection modulates PINK1 expression. Huh7 cells were mock-infected or infected with JEV (MOI 1), and the cells were harvested at 24 and 48 hpi. (A) Total RNA isolated from the cells was used to measure the PINK1 mRNA by qRT-PCR. GAPDH was used as the internal control. Levels of PINK1 mRNA in JEV-infected cells relative to those in the mock-infected cells are depicted. (B) The cell lysates were subjected to Western blotting to detect the full-length (FL) and cleaved (CL) forms of PINK1 (left panel). GAPDH was used as the loading control, and JEV-prM was used as the infection marker. CCCP-treated Huh7 cell lysates were used as a positive control for the full-length (FL) PINK1. The right panels show the densitometric analysis of full-length and cleaved PINK1 band intensities. (C) Three-week-old BALB/c mice (n = 4) were mock- or JEV-infected by the intraperitoneal mode of injection. The liver tissue was harvested 5 days later and prepared as lysate. The left panel shows the Western blotting of FL- and CL-PINK1 in JEV-infected mouse liver lysate. The JEV-prM was used as the infection marker, and actin was used as the internal loading control. The right panel shows quantification of FL- and CL-PINK1 band intensities of mice tissue by the Western blots. (D) HEK293T cells were transfected with plasmids expressing JEV-NS4A with HA tag (NS4A-HA) or the empty vector expressing only the HA tag (empty-HA). The cell lysates were prepared 48 h later and subjected to Western blotting (left panel). GAPDH was used as the loading control. The right panel shows the quantification of full-length and cleaved PINK1 band intensities of the Western blots. The Western blot data are representative of the three independent experiments whose data are plotted as bar graphs. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).

FIG 7

Mitochondrial dynamics in JEV-infected Huh7 cells. Huh7 cells were mock-infected or infected with JEV (MOI 1) and 48 h later stained with the MitoTracker dye before fixation with 4% PFA. Cells were immunostained for TOMM20 and JEV-E proteins. (A) Images taken using the Leica TCS SP8 confocal microscope are shown. The MitoTracker and TOMM20 images were merged to observe the mitochondria shape. (B) Quantification of the mitochondrial branch length and footprint in mock- and JEV-infected cells. (C) Cell lysates obtained from mock- and JEV-infected cells at 24 and 48 hpi were subjected to Western blotting for the mitochondrial fission and fusion proteins as indicated. (D) Quantification of MFN1, MFN2, OPA1, and DRP1 by band intensities, relative to GAPDH, in mock- and JEV-infected cells. The Western blots (in panel C) are representative of the three independent experiments whose data are plotted as bar graphs. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).

FIG 8

Mitochondrial and mitophagy-related proteins during JEV infection. (A) Huh7 cells were mock-infected or infected with JEV (MOI 1), and cells were harvested at 24 and 48 hpi. The cell lysates were subjected to Western blotting for the indicated proteins. GAPDH was used as the loading control, and JEV-prM was used as the infection marker. The Western blots are representative of the three independent experiments whose data are plotted as bar graphs. (B) Three-weeks-old BALB/c mice (n = 4) were infected with JEV or mock-infected by the intraperitoneal mode of injection. The liver tissues were harvested 5 days later and prepared as lysates. The left panels show the Western blotting of indicated proteins in JEV-infected mouse liver lysate. The JEV-prM was used as the infection marker, and actin was used as the internal loading control. The right panel shows quantification of protein band intensities of mouse tissue from the Western blots. (C) Mock- and JEV-infected (MOI 1) Huh7 cells were treated at 30 hpi with bafilomycin for 8 h. The cell lysates were subjected to Western blotting for the indicated proteins. GAPDH was used as the loading control, and JEV-prM or JEV-NS3 was used as the infection marker. The band intensities of COX2, COX4, and TOMM20 compared to GAPDH are shown at the bottom. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).

FIG 9

Mitophagy flux in JEV-infected Huh7 cells. (A) Huh7 cells were transfected with the mitophagy reporter plasmid mito-mRFP-EGFP and 24 h later infected with JEV (MOI 3). Huh7 cells treated with CCCP were used as a positive control for mitophagy. The cells were fixed 48 hpi and immunostained with the JEV-E antibody. Images were taken using the Leica TCS SP8 confocal microscope. The red and green fluorescence signals indicating the mRFP- and EGFP-expressing mitochondria, respectively, are seen. The mRFP (red channel) and EGFP (green channel) images were merged to observe mitochondria displaying only the red fluorescence. The images are representative of the three independent experiments whose data are plotted as bar graph showing the number of red mitochondria per cell. (B) Huh7 cells were cotransfected with the mitophagy reporter mito-mRFP-EGFP and JEV-NS4A-HA or empty-HA plasmids. The cells were fixed 48 h later, immunostained with JEV-NS4A antibody, and visualized under a Leica TCS SP8 confocal microscope. The red and green fluorescence signals indicating the mRFP- and EGFP-expressing mitochondria, respectively, are seen. The red (mRFP) and green (EGFP) channels were merged to detect mitochondria exhibiting only red fluorescence indicative of complete mitophagy. The images are representative of the three independent experiments whose data are plotted as bar graphs showing the number of red mitochondria per cell. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).

FIG 10

Role of mitochondrial fission in JEV life cycle. (A) Huh7 cells were treated with indicated siRNA, and 48 h later, the levels of DRP1 were determined by Western blotting. The siRNA-treated cells were infected with JEV (MOI 3) and 48 h later fixed with 4% PFA. Cells were immunostained for TOMM20 and JEV-E proteins. Images taken using the Leica TCS SP8 confocal microscope are shown. The mean mitochondrial branch length in siNT- and siDRP1-treated cells infected with JEV was quantified by Image J. The images are representative of the three independent experiments whose data are plotted as bar graphs. (B) The siRNA-transfected cells were infected with JEV (MOI 3) at 48 hpi, and 48 h later, the culture supernatant and cells were harvested. The total RNA isolated from the cells and culture supernatants were subjected to qRT-PCR to estimate the intracellular and extracellular JEV genome copy numbers. The virus titers in the culture supernatants were determined by the FFU assay. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).

FIG 11

Role of mitophagy in JEV life cycle: (A) Huh7 cells were treated with indicated siRNA, and 48 h later, the levels of PINK1 were determined by Western blotting. The siRNA-transfected cells were transfected 16 h later with the mitophagy reporter plasmid mito-mRFP-EGFP and infected with JEV (MOI 3) 16 h later. The cells were fixed 36 h later with 4% PFA and immunostained for JEV-E protein. Images taken using the Leica TCS SP8 confocal microscope are shown. The number of mitochondria displaying only the mRFP fluorescence was quantified by Image J. The images are representative of the three independent experiments whose data are plotted as bar graphs showing the number of red mitochondria per cell. (B) Forty-eight hours posttransfection, the siRNA-transfected cells were infected with JEV (MOI 3), and 48 h later, the culture supernatant and cells were harvested. The total RNA isolated from the cells and culture supernatants were subjected to qRT-PCR to estimate intracellular and extracellular JEV genome copy numbers. The virus titers in the culture supernatants were determined by the FFU assay. Statistical analysis was performed using the Student’s t test (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ns, not significant).