Venezuelan equine encephalitis virus capsid protein inhibits nuclear import in Mammalian but not in mosquito cells

Abstract

Venezuelan equine encephalitis virus (VEEV) represents a continuous public health threat in the United States. It has the ability to cause fatal disease in humans and in horses and other domestic animals. We recently demonstrated that replicating VEEV interferes with cellular transcription and uses this phenomenon as a means of downregulating a cellular antiviral response. VEEV capsid protein was found to play a critical role in this process, and its approximately 35-amino-acid-long peptide, fused with green fluorescent protein, functioned as efficiently as did the entire capsid. We detected a significant fraction of VEEV capsid associated with nuclear envelope, which suggested that this protein might regulate nucleocytoplasmic trafficking. In this study, we demonstrate that VEEV capsid and its N-terminal sequence efficiently inhibit multiple receptor-mediated nuclear import pathways but have no effect on the passive diffusion of small proteins. The capsid protein of the Old World alphavirus Sindbis virus and the VEEV capsid, with a previously defined frameshift mutation, were found to have no detectable effect on nuclear import. Importantly, the VEEV capsid did not noticeably interfere with nuclear import in mosquito cells, and this might play a critical role in the ability of the virus to develop a persistent, life-long infection in mosquito vectors. These findings demonstrate a new aspect of VEEV-host cell interactions, and the results of this study are likely applicable to other New World alphaviruses, such as eastern and western equine encephalitis viruses.

FIG. 1.

Downregulation of the importin-α/β-mediated nuclear import in cells infected with VEEV TC-83. BHK-21 cells were either infected with packaged VEErep/4xTomato (A) or VEErep/4xTomato-3xNLS (B) replicons at an MOI of ∼20 i.u./cell or coinfected with VEErep/4xTomato-3xNLS replicon and VEEV TC-83 at the same MOIs (C). Distribution of 4xTomato and 4xTomato-3xNLS proteins was evaluated at 8 h postinfection (see Materials and Methods for details). (Subpanels a) Distribution of the 4xTomato and 4xTomato-3xNLS proteins; (b) cell nuclei stained with Sytox green (Invitrogen); (c) overlay of the two images. Schematic representations of used VEErep/4xTomato and VEErep/4xTomato-3xNLS replicons and the VEEV TC-83 genome are shown. Bars correspond to 20 μm.

FIG. 2.

Expression of C_VEE affects nuclear import of the proteins with different NLSs. (A) BHK-21 cells were coinfected with packaged VEErep/4xTomato-3xNLS and VEErep/C_VEE/GFP (see Materials and Methods for details). (a) Distribution of the 4xTomato-3xNLS protein; (b) distribution of GFP; (c) overlay of the two images. (B) BHK-21 cells were coinfected with VEErep/4xTomato-M9 and VEErep/C_VEE/GFP replicons. (a) Distribution of the 4xTomato-M9 protein; (b) distribution of GFP; (c) overlay of the two images. (C) BHK-21 cells were coinfected with VEErep/4xTomato-H2b and VEErep/C_VEE/GFP replicons. (a) Distribution of the 4xTomato-H2b protein; (b) distribution of GFP; (c) overlay of the two images. (D) Distribution of fluorescent proteins in BHK-21 cells infected with VEErep/C_VEE/GFP, VEErep/4xTomato-M9, or VEErep/4xTomato-H2b. Cells infected with VEErep/4xTomato-3xNLS are presented in Fig. 1. All of the images were acquired at 8 h postinfection. Bars correspond to 20 μm. The schematic representation of the replicons is shown for each panel.

FIG. 3.

BHK-21 cells (5 × 10⁵) in six-well Costar plates were infected with the indicated replicons or viruses at an MOI of 20 i.u. or PFU per cell, respectively. Proteins were pulse-labeled with [³⁵S]methionine at 8 h postinfection and analyzed on a sodium dodecyl sulfate-10% polyacrylamide gel. The gel was dried and autoradiographed. Positions of the expressed proteins of interest are indicated on the right side of the gel; molecular weight markers (in thousands) are shown on the left.

FIG. 4.

Analysis of the 4xTomato-3xNLS distribution in cells expressing either C_SIN or different variants of C_VEE, fused with GFP. (A) BHK-21 cells were coinfected with packaged VEErep/4xTomato-3xNLS and VEErep/C_SIN/GFP (see Materials and Methods for details). (a) Distribution of the 4xTomato-3xNLS protein; (b) distribution of GFP; (c) overlay of the two images. (B) BHK-21 cells were coinfected with VEErep/4xTomato-3xNLS and VEErep/C_VEE1-68-GFP replicons. (a) Distribution of the 4xTomato-3xNLS protein; (b) distribution of C_VEE1-68-GFP; (c) overlay of the two images. (C) BHK-21 cells were coinfected with VEErep/4xTomato-3xNLS and VEErep/C_VEEfrsh-GFP replicons. (a) Distribution of the 4xTomato-3xNLS protein; (b) distribution of C_VEEfrsh-GFP; (c) overlay of the two images. (D) BHK-21 cells were coinfected with VEErep/4xTomato-3xNLS and VEErep/C_VEE-GFP replicons. (a) Distribution of the 4xTomato-3xNLS protein; (b) distribution of C_VEE-GFP; (c) overlay of the two images. All of the images were acquired at 8 h postinfection. Bars correspond to 20 μm. The schematic representation of the replicons is shown for each panel.

FIG. 5.

Distribution of VEEV nsP2 in the cells infected with VEEV TC-83. The nsP2 distribution in BHK-21 (A), NIH 3T3 (B), and HEK293 (C) cells is shown. Staining was performed at 8 h postinfection (see Materials and Methods for details). (D) Staining of the mock-infected cells with VEEV nsP2-specific antibodies. (Subpanels a) Staining with VEEV nsP2-specific antibodies; (b) nuclear staining with Sytox orange; (c) overlays of the images. Bars correspond to 20 μm.

FIG. 6.

SINV nsP2 distribution depends on the capsid protein, encoded by viral genome. Intracellular distribution of SINV nsP2 during SINV Toto 1101 replication (A) or replication of SIN/VEEV recombinant virus in BHK-21 cells (B). (C) Staining of the mock-infected BHK-21 cells. (Subpanels a) Staining with the SINV nsP2-specific antibodies; (b) nuclear staining with Sytox orange; (c) overlays of the images. Staining was performed at 8 h postinfection (see Materials and Methods for details). Images were acquired at 8 h postinfection. Bars correspond to 20 μm. In the schematic representations of viral genomes, SINV-specific sequences are indicated by open boxes and VEEV sequences are indicated by filled boxes.

FIG. 7.

VEEV nsP2-GFP fusion protein distribution in the cells infected with VEEV/nsP2GFP. BHK-21 cells were infected with the chimeric virus at an MOI of 20 PFU/cell, and nsP2/GFP distribution was evaluated at 8 h postinfection with a confocal microscope. (a) Distribution of VEEV nsP2/GFP. (b) Overlay of nsP2/GFP and nuclear staining with Sytox orange. Bars correspond to 20 μm. The schematic representation of the viral genome is shown.

FIG. 8.

Distribution of VEEV nsP2 in the cells infected by the constructs encoding no C_VEE. (A) Distribution of VEEV nsP2 in the cells infected with VEErep/Cherry. (a) Staining with VEEV nsP2-specific antibodies; (b) distribution of Cherry; (c) overlay of the two images. (B) Distribution of VEEV nsP2-HA in the cells infected with the SINrep2V/nsP2VEE-HA replicon. (a) Staining with anti-HA antibodies; (b) nuclear staining with Sytox orange; (c) overlay of the two images. All of the images were acquired at 8 h postinfection. Bars correspond to 20 μm. The schematic representation of the replicons is shown for each panel.

FIG. 9.

C_VEE does not interfere with nuclear import in mosquito cells. C₇10 cells were transfected by in vitro-synthesized VEErep/4xTomato-3xNLS and VEErep/C_VEE-GFP replicon RNAs (A) and VEErep/4xTomato-3xNLS and VEErep/C_VEE/GFP RNAs (B). (Subpanels a) distribution of 4xTomato-3xNLS protein; (b) distribution of C_VEE-GFP and GFP in panels A and B, respectively; (c) overlay of the two images. All of the images were acquired at 24 h posttransfection. Bars correspond to 20 μm. The schematic representation of the replicons is shown for each panel.