Severe acute respiratory syndrome coronavirus ORF6 antagonizes STAT1 function by sequestering nuclear import factors on the rough endoplasmic reticulum/Golgi membrane

Abstract

The host innate immune response is an important deterrent of severe viral infection in humans and animals. Nuclear import factors function as key gatekeepers that regulate the transport of innate immune regulatory cargo to the nucleus of cells to activate the antiviral response. Using severe acute respiratory syndrome coronavirus (SARS-CoV) as a model, we demonstrate that SARS-COV ORF6 protein is localized to the endoplasmic reticulum (ER)/Golgi membrane in infected cells, where it binds to and disrupts nuclear import complex formation by tethering karyopherin alpha 2 and karyopherin beta 1 to the membrane. Retention of import factors at the ER/Golgi membrane leads to a loss of STAT1 transport into the nucleus in response to interferon signaling, thus blocking the expression of STAT1-activated genes that establish an antiviral state. We mapped the region of ORF6, which binds karyopherin alpha 2, to the C terminus of ORF6 and show that mutations in the C terminus no longer bind karyopherin alpha 2 or block the nuclear import of STAT1. We also show that N-terminal deletions of karyopherin alpha 2 that no longer bind to karyopherin beta 1 still retain ORF6 binding activity but no longer block STAT1 nuclear import. Recombinant SARS-CoV lacking ORF6 did not tether karyopherin alpha 2 to the ER/Golgi membrane and allowed the import of the STAT1 complex into the nucleus. We discuss the likely implications of these data on SARS-CoV replication and pathogenesis.

FIG. 1.

SARS-CoV encodes an IFN antagonist. (A) MA104 cells were infected with SARS and SeV or mock infected for 12 h. Extracted RNA was analyzed by RT-PCR for the induction of IFN-β mRNA. RT-PCR-amplified GAPDH transcripts are shown as a loading control. (B) Media from icSARS-, Urbani virus-, and SeV-infected MA104 cells were analyzed for secreted type I IFN across a time course of infection. An IFN bioassay (as described in Materials and Methods) was used to analyze the amount of type I IFN secreted from infected cells. The dotted line is the minimal level of detection for the assay.

FIG. 2.

STAT1 localization during SARS infection. STAT1/GFP plasmid was transfected into Vero cells, and its localization was assayed after SARS infection. At 24 h posttransfection, cells were infected with either icSARS (top) or icSARSΔORF6 (middle) at an MOI of 3 for 12 h. Before fixation in 4% PFA, cells were treated with 100 IU of IFN-β for 1 h. The cells in the bottom panels were cotransfected with STAT1/GFP and HA-ORF6 prior to infection with icSARSΔORF6. Cells were then labeled with anti-SARS spike antibody to visualize the SARS-infected cells or anti-HA antibody (to visualize the transfected cells) and an Alexa 546-conjugated secondary antibody. Cells were visualized by using a confocal microscope.

FIG. 3.

ORF6 affects localization of KPNA2. (A) 293 cells were transfected with Flag-tagged karyopherin plasmids. At 24 h posttransfection, half of the cells were treated with 100 IU of IFN-β for 30 min, and protein was harvested as described in Materials and Methods. Proteins were then immunoprecipitated (IP) with anti-Flag (αFlag) M2 antibodies and run on an SDS-PAGE gel before being transferred for Western blotting. Blots were probed with STAT1, STAT2, or Flag antibodies. + indicates treatment with 100 IU/ml of IFN-β. (B) Immunoprecipitations were performed on 293 cells cotransfected with each Flag-tagged karyopherin and HA-tagged ORF6. Lysates were prepared and immunoprecipitated as described in Materials and Methods. The top panel was probed with anti-HA antibody to visualize ORF6, and the bottom panel was probed with anti-KPNB1. A1, KPNA1; A2, KPNA2; A3, KPNA3; A4, KPNA4. (C) Vero cells were transfected with Flag-tagged KPNA plasmids. At 24 h posttransfection, cells were fixed and labeled with anti-Flag antibody. (D) Vero cells were cotransfected with Flag-tagged KPNA plasmids and HA-ORF6 plasmid. At 24 h posttransfection, cells were fixed and labeled with anti-Flag antibody and anti-HA antibody. Anti-Flag antibody was visualized with Alexa Fluor 488-conjugated secondary antibody, and anti-HA antibody was visualized with Alexa Fluor 546-conjugated secondary antibody. Note the overlapping localization of ORF6 and KPNA2.

FIG. 4.

Interaction between KPNA2 and ORF6 in vivo. The top panel shows a schematic of the split YFP assay. The cargo is fused to each half of YFP. When the cargo interacts, the YFP halves are brought together to re-form and fluoresce. Vero cells were transfected with YFP plasmids, and 24 h after transfection, YFP fluorescence was visualized on a confocal microscope using a YFP filter. (A) YFP-N/leucine zipper (LZ) and YFP-C/leucine zipper (positive controls). (B) YFP-N/KPNA2. (C) YFP-C/ORF6. (D) YFP-N/leucine zipper plus YFP-C/ORF6. (E) YFP-N/KPNA2 and YFP-C/ORF6. (F) Karyopherin localization during SARS infection. Vero cells were transfected with Flag-tagged KPNA1 or KPNA2 24 h prior to infection with either icSARS or icSARSΔORF6. Cells were fixed in 4% PFA prior to antibody staining. SARS mouse anti-N antibody (αN) was used to localize SARS-infected cells, and rabbit anti-Flag antibody was used to localize karyopherins. Mouse anti-N antibody was visualized with Alexa Fluor 488-conjugated secondary antibody, and anti-Flag antibody was visualized with Alexa Fluor 546-conjugated secondary antibody.

FIG. 5.

ORF6 interacts with KPNB1. (A) ORF6 colocalization with KPNB1. Vero cells were transfected with Flag-tagged KPNB1 and HA-ORF6. At 24 h posttransfection, cells were fixed and stained. Cells were incubated with mouse anti-HA (αHA) and rabbit anti-Flag antibodies. Anti-Flag antibody was visualized with Alexa Fluor 546-conjugated secondary antibody, and anti-HA antibody was visualized with Alexa Fluor 488-conjugated secondary antibody. Coverslips were visualized with a Zeiss confocal microscope. (B) 293 cells were transfected with HA-tagged ORF6 and Flag-tagged KPNA1, KPNA2, KPNA3, or KPNA4. Cells were lysed 24 h posttransfection and immunoprecipitated (IP) with anti-HA antibody. Proteins were separated on a 4 to 12% SDS PAGE gel and analyzed by Western blot (WB) probing with anti-HA, anti-Flag, and anti-KPNB1 antibody.

FIG. 6.

KPNA2 interactions with ORF6 and KPNB1. (A) 293 cells were transfected with either Flag-tagged KPNA1, KPNA2, or KPNA2ΔN in combination with HA-tagged ORF6. At 24 h after transfection, cells were treated with 100 IU/ml IFN-β for 30 min, and lysates were collected. Proteins were immunoprecipitated (IP) with anti-Flag antibodies as described in Materials and Methods. Immunoprecipitated extracts were then separated on a 4 to 12% SDS-PAGE gel and analyzed by Western blot (WB) probed with anti-HA, anti-KPNB1, anti-STAT1, and anti-Flag antibodies. Whole-cell extracts (WCL) were also blotted for ORF6 transfection with anti-HA antibodies. (B and C) Vero cells were transfected with STAT1/GFP and either Flag-tagged KPNA2 or KPNA2ΔN. At 24 h posttransfection, cells were treated with 100 IU/ml IFN-β for 60 min before fixation. Proteins were visualized with anti-Flag (αFLAG) antibodies and for the presence of GFP using confocal microscopy. (D and E) Vero cells were transfected as described above (B) except for the addition of HA-tagged ORF6. Cells were stained with anti-Flag and anti-HA antibodies and visualized using confocal microscopy for Flag (Alexa Fluor 546)-, HA (Alexa Fluor 633)-, and GFP-tagged proteins.

FIG. 7.

C-terminal mutations of ORF6 affect STAT1 localization. (A) Schematic of ORF6 alanine mutants expressed in B to D. Numbers correspond to amino acids in ORF6. The black box indicates the amino acids that are changed to alanine in the protein (B) Each alanine mutant was fused to a C-terminal GFP and expressed in Vero cells. Localization is shown via a GFP filter on a confocal microscope. (C) Each HA-tagged alanine mutant was cotransfected with STAT1/GFP. At 24 h posttransfection, cells were treated with 100 IU/ml of IFN-β for 60 min before fixation in 4% PFA. STAT1/GFP was visualized on a confocal microscope.

FIG. 8.

C-terminal mutations of ORF6 affect karyopherin localization. (A) HA-tagged alanine scanning mutants of ORF6 (as described in the text) were cotransfected with either Flag-tagged KPNA1 or KPNA2. At 24 h posttransfection, cells were fixed and labeled with anti-HA and anti-Flag antibodies. Cells were visualized using a confocal microscope for Alexa Fluor 488 and Alexa Fluor 546 fluorescence.

FIG. 9.

Model for ORF6 function. Upon IFN-α or -β stimulation of the IFN receptor on the surface of a cell, the STAT1:STAT2:IRF9 complex is formed. The NLS formed from STAT1 and STAT2 is recognized by KPNA1 (Kα1) for import into the nucleus. SARS-CoV-infected cells that are ER/Golgi membrane localized (shown here only on the ER membrane for clarity) bind to KPNA2, which recruits KPNB1 to the membrane complex as well. This creates a concentration-dependent competition between the ORF6:KPNA2 and ISGF3:KPNA1 complexes for the free unbound KPNB1 in the cytoplasm. This depletion of free KPNB1 in the cytoplasm produces a block in import of the ISGF3 complex in response to both IFN-β and IFN-γ and leaves KPNB1 bound to KPNA2 and ORF6 on the ER/Golgi membrane.