Potential Dual Role of West Nile Virus NS2B in Orchestrating NS3 Enzymatic Activity in Viral Replication

Abstract

West Nile virus (WNV) nonstructural protein 3 (NS3) harbors the viral triphosphatase and helicase for viral RNA synthesis and, together with NS2B, constitutes the protease responsible for polyprotein processing. NS3 is a soluble protein, but it is localized to specialized compartments at the rough endoplasmic reticulum (RER), where its enzymatic functions are essential for virus replication. However, the mechanistic details behind the recruitment of NS3 from the cytoplasm to the RER have not yet been fully elucidated. In this study, we employed immunofluorescence and biochemical assays to demonstrate that NS3, when expressed individually and when cleaved from the viral polyprotein, is localized exclusively to the cytoplasm. Furthermore, NS3 appeared to be peripherally recruited to the RER and proteolytically active when NS2B was provided in trans. Thus, we provide evidence for a potential additional role for NS2B in not only serving as the cofactor for the NS3 protease, but also in recruiting NS3 from the cytoplasm to the RER for proper enzymatic activity. Results from our study suggest that targeting the interaction between NS2B and NS3 in disrupting the NS3 ER localization may be an attractive avenue for antiviral drug discovery.

Keywords: NS2B; NS3; West Nile Virus; endoplasmic reticulum; flavivirus; membrane structures.

Figure 1

West Nile virus (WNV) nonstructural protein 3 (NS3) and NS2B associate with the ER during infection. (A) HEK293T cells were infected with WNV_NY99 at a multiplicity of infection (MOI) of 1 and after 1 h, transfected with Sec61β-RFP, ER marker, (red; a,d), Arf1-RFP, a Golgi marker, (red; b,e), or tubulin-GFP, a cytoskeletal marker, (green; c,f). Cells were fixed at 24 h after infection and immunostained with an anti-WNV NS3 antibody (green; a,b, red; c) or with an anti-WNV NS2B antibody (green; d,e, red; f). Nuclear DNA was stained with 4,6-diamidino-2-phenylindole (DAPI). Slides were analyzed by confocal laser scanning microscopy. Confocal microscopy images were of optical slice thickness ~1 μm. Scale bar, 10 μm. The main panels represent merged images from three separate channels that are individually shown on the side panels. White arrows indicate colocalization between the green and red channels. The infection–transfection efficiency (ITE) indicating the percentage of cells that were both WNV-infected and transfected for each treatment is listed at the bottom of the panel. Images are representative of two independent infection experiments. (B) Pearson’s correlation coefficient (PCC) analysis of the colocalization between NS3 (gray bars) or NS2B (black bars) with cellular markers in WNV infected cells conducted using Coloc 2 (FIJI). PCC values greater than 0.5 (dotted line) indicate a high level of colocalization. Letters in parenthesis beneath each pair of markers correspond to its representative confocal image in Figure 1A. Error bars indicate mean ± standard error of the mean (SEM); n = 5–10 cells per group. (C) Cells were infected with WNV_NY99 at a MOI of 1 and harvested after 48 h. Cell fractions were separated using sucrose gradient ultracentrifugation and subjected to SDS-PAGE before immunoblotting. The viral proteins were detected using anti-WNV NS3 and NS2B antibodies. The cytoplasmic and endoplasmic reticulum (ER) fractions were observed using antibodies against tubulin and calnexin, respectively.

Figure 2

WNV NS3 is distributed throughout the cytoplasm, while NS2B remains in the ER in transfected cells. (A) NS3-GFP transfected (a–c) or NS3-GFP and IKKε-FLAG co-transfected (d), HEK293T cells were fixed at 24 h post-transfection and immunostained with antibodies against protein disulfide isomerase (PDI), an ER marker, (red; a), anti-GM130, a Golgi apparatus marker, (red; b), anti-tubulin, a cytoskeletal marker, (red; c), or anti-IKKε, a soluble cytoplasmic protein marker, (red; d). (B) NS2B-V5/His transfected (a–c) or NS2B-V5/His and IKKε-FLAG co-transfected (d) HEK293T cells were fixed 24 h after transfection and immunolabeled with rabbit anti-PDI (green; a), anti-giantin (green; b), anti-tubulin (green; c) or anti-IKKε (green; d) antibodies to visualize the ER, Golgi apparatus, cytoskeleton or the cytoplasmic protein kinase IKKε protein, respectively. Visualized viral and host proteins are listed on the top right corner of each panel. Nuclear DNA was labeled with DAPI. Confocal microscopy images were of optical slice thickness ~1 μm. Scale bar, 10 μm. The main image depicts the merged image from three channels separately shown on the side. White arrows indicate colocalization between the green and red channels. The transfection efficiency (TE) for cells transfected with one construct or the co-transfection efficiency (Co-TE) for cells co-expressing the viral protein and IKKε are listed at the bottom of each panel. Images are representative of two independent transfection experiments. (C) Colocalization analysis between viral proteins and host markers was conducted using Coloc 2 (FIJI). Pearson’s correlation coefficient (PCC) was calculated for each indicated pair of proteins in NS3-GFP (gray bars) or NS2B-V5 (black bars) transfected cells. PCC values larger than 0.5 (dotted line) indicate a high probability that pixels from the red and green channels overlap. Parenthesized letters link each PCC value to its corresponding confocal image in Figure 2A or Figure 2B. Error bars indicate mean ± SEM; n = 5–10 cells per group. (D) NS2B-GFP or NS3-V5 singly transfected HEK293T cells were lysed 48 h post-transfection and fractionated by sucrose gradient centrifugation. Cell fractions were analyzed by Western blot using mouse anti-GFP or rabbit anti-V5 antibodies to detect the NS2B and NS3 proteins, respectively. The cytoplasmic and ER fractions were visualized using anti-tubulin and anti-calnexin antibodies, respectively.

Figure 3

NS3 localizes to the ER when NS2B is provided in trans and in cis. (A) HEK293T cells were co-transfected with NS3-GFP and NS2B-V5 expressing plasmids (a–c). At 24 h post-transfection, cells were fixed and co-stained with anti-GFP and antibodies against PDI (red; a), giantin (red; b) or tubulin (red; c). HEK293T cells were also triple-transfected with NS3-GFP, NS2B-V5/His and IKKε-FLAG expressing plasmids (d) and triple stained with anti-GFP (green; d), anti-V5/His (red; d) and anti-IKKε (blue; d) antibodies. (B) HEK293T cells were transfected with NS2B-NS3-GFP (a–c) or co-transfected with NS2B-NS3-GFP and IKKε-FLAG expressing plasmids (d) and processed for immunofluorescence at 24 h after transfection using antibodies against PDI (red; a), giantin (red; b), tubulin (red; c) and IKKε (red; d). The detected viral and host proteins are listed on the top right corner of each panel. White arrows indicate colocalization between the GFP-tagged viral protein and the ER marker. Nuclear DNA was labeled with DAPI. Scale bar represents 10 μm. The main panels depict merged images with the side panels showing the individual channels. The transfection efficiency (TE), the co-transfection efficiency (Co-TE) and the triple-transfection efficiency (Triple-TE) for cells expressing one, two, or three gene constructs, respectively, are listed at the bottom of the panel. Images are representative of two independent transfection experiments. (C) Quantitation of colocalization between NS3 and cellular markers when NS2B is provided in trans (gray bars) or in cis (diagonally striped bars) was determined using the Pearson’s correlation coefficient (PCC). Letters in parenthesis underneath each pair of proteins indicate its corresponding confocal image in Figure 3A or Figure 3B. PCC values greater than 0.5 (dotted line) indicate a high level of colocalization. Error bars indicate mean ± SEM; n = 5–10 cells per group. (D) HEK293T cells expressing both NS3-V5 and NS2B-GFP with NS2B provided in trans or cells expressing NS2B-NS3-GFP with NS2B provided in cis were lysed 48 h post-transfection and subjected to sucrose density gradient ultracentrifugation. Aliquots of fractions collected from the top of the gradient were analyzed by Western blotting. Rabbit anti-tubulin and anti-calnexin antibodies were used to distinguish between the cytoplasmic and ER fractions, respectively, and antibodies against the fused tag (GFP or V5/His) were used to detect the viral protein.

Figure 4

Processing of NS3 from the viral polyprotein precursor does not contribute to its ER localization. (A) Schematic representation of the GFP-tagged NS2B-NS3-NS4A polyprotein. The GFP epitope is indicated by the green box. Construct is drawn to scale according to the number of amino acid residues. The viral protease cleavage sites at the NS2B-NS3 and NS3-NS4A junctions are marked by the black arrowheads. (B) NS2B-NS3-NS4A-GFP transfected HEK293T cells were lysed 48 h post-transfection and subjected to sucrose density gradient ultracentrifugation. Aliquots of fractions collected from the top of the gradient were analyzed by Western blotting. The cytoplasmic and ER fractions were distinguished using anti-tubulin and anti-calnexin antibodies, respectively. The unprocessed full-length NS2B-NS3-NS4A polyprotein and processed NS4A protein were detected using an anti-GFP antibody. Processed NS2B and NS3 proteins were visualized using anti-WNV NS2B and NS3 antibodies.

Figure 5

NS3 is a functional viral protease when NS2B is provided in trans. (A) Schematic diagram depicting the V5/His or GFP-tagged WNV NS constructs used in this proteolytic processing assay. Constructs are drawn to scale according to the number of amino acid residues. The V5 or GFP tags are indicated by red or green boxes, respectively. The viral protease cleavage site between NS4B and NS5 is marked by the black arrowhead. (B) HEK293T cells were transfected with various combinations of WNV NS constructs (listed at the top of the panel) and cell lysates were harvested 48 h post-transfection. Cleavage of NS4B-NS5-GFP by NS3-V5/His when NS2B was provided in trans (lanes 2–4) or in cis (lanes 7–8) was assessed by Western blotting using anti-GFP and anti-V5/His antibodies to detect the transfected viral proteins and cleaved products. β-actin served as an internal loading control. Molecular weights (kDa) are given on the left side of each panel.