Severe Fever with Thrombocytopenia Syndrome Virus NSs Interacts with TRIM21 To Activate the p62-Keap1-Nrf2 Pathway

Abstract

Nuclear factor erythroid 2-related factor 2 (Nrf2) dissociates from its inhibitor, Keap1, upon stress signals and subsequently induces an antioxidant response that critically controls the viral life cycle and pathogenesis. Besides intracellular Fc receptor function, tripartite motif 21 (TRIM21) E3 ligase plays an essential role in the p62-Keap1-Nrf2 axis pathway for redox homeostasis. Specifically, TRIM21-mediated p62 ubiquitination abrogates p62 oligomerization and sequestration activity and negatively regulates the Keap1-Nrf2-mediated antioxidant response. A number of viruses target the Nrf2-mediated antioxidant response to generate an optimal environment for their life cycle. Here we report that a nonstructural protein (NSs) of severe fever with thrombocytopenia syndrome virus (SFTSV) interacts with and inhibits TRIM21 to activate the Nrf2 antioxidant signal pathway. Mass spectrometry identified TRIM21 to be a binding protein for NSs. NSs bound to the carboxyl-terminal SPRY subdomain of TRIM21, enhancing p62 stability and oligomerization. This facilitated p62-mediated Keap1 sequestration and ultimately increased Nrf2-mediated transcriptional activation of antioxidant genes, including those for heme oxygenase 1, NAD(P)H quinone oxidoreductase 1, and CD36. Mutational analysis found that the NSs-A46 mutant, which no longer interacted with TRIM21, was unable to increase Nrf2-mediated transcriptional activation. Functionally, the NS wild type (WT), but not the NSs-A46 mutant, increased the surface expression of the CD36 scavenger receptor, resulting in an increase in phagocytosis and lipid uptake. A combination of reverse genetics and assays with Ifnar^-/- mouse models revealed that while the SFTSV-A46 mutant replicated similarly to wild-type SFTSV (SFTSV-WT), it showed weaker pathogenic activity than SFTSV-WT. These data suggest that the activation of the p62-Keap1-Nrf2 antioxidant response induced by the NSs-TRIM21 interaction contributes to the development of an optimal environment for the SFTSV life cycle and efficient pathogenesis.IMPORTANCE Tick-borne diseases have become a growing threat to public health. SFTSV, listed by the World Health Organization as a prioritized pathogen, is an emerging phlebovirus, and fatality rates among those infected with this virus are high. Infected Haemaphysalis longicornis ticks are the major source of human SFTSV infection. In particular, the recent spread of this tick to over 12 states in the United States has increased the potential for outbreaks of this disease beyond Far East Asia. Due to the lack of therapies and vaccines against SFTSV infection, there is a pressing need to understand SFTSV pathogenesis. As the Nrf2-mediated antioxidant response affects viral life cycles, a number of viruses deregulate Nrf2 pathways. Here we demonstrate that the SFTSV NSs inhibits the TRIM21 function to upregulate the p62-Keap1-Nrf2 antioxidant pathway for efficient viral pathogenesis. This study not only demonstrates the critical role of SFTSV NSs in viral pathogenesis but also suggests potential future therapeutic approaches to treat SFTSV-infected patients.

Keywords: Nrf2 pathway; SFTSV; TRIM21; nonstructural protein.

FIG 1

NSs interacts with TRIM21. (A) HEK293T cells were transfected with NSs-GST and TRIM21-V5, and whole-cell extracts (WCEs) were pulled down by glutathione beads, followed by immunoblotting with the indicated antibody. (B) HEK293T cells were transfected with NSs-3×Flag and TRIM21-V5, and WCEs were applied to SiMPull analysis. (Left) Three representative images. (Right) Molecular numbers, in which the bar graphs indicate the average number of fluorophores per image. Error bars represent the SD of the mean across >20 images. The results of three independent experiments are represented. (C) RAW 264.7 cells were infected with SFTSV-NSs-GFP, a recombinant virus expressing GFP-tagged NSs, for 24 h and subjected to immunoprecipitation (IP) with anti-GFP antibody to pull down the GFP-NSs complex, followed by immunoblotting with anti-TRIM21 antibody to detect endogenous TRIM21. (D) HeLa cells were transfected with NSs-3×Flag-GFP and TRIM21-V5. The cells were fixed and stained with primary mouse anti-V5 antibody and with secondary Alexa Fluor 568-conjugated anti-mouse IgG antibody for confocal microscopy. Hoechst staining was used for the nucleus. The microscope images represent those from three independent experiments.

FIG 2

SFTSV NSs specifically binds to and colocalizes with TRIM21. (A) HEK293T cells were transfected with TRIM21-V5 and NSs-3×Flag of three viruses (SFTSV, HRTV, and UUKV), and WCEs were immunoprecipitated by an anti-V5 antibody, followed by immunoblotting with the indicated antibody. (B) HeLa cells were transfected with TRIM21-V5 and NSs-3×Flag, as described in the legend to panel A. Cells were fixed and stained with primary antibodies (mouse anti-V5 and rabbit anti-Flag) and with secondary antibodies (Alexa Fluor 488-conjugated anti-mouse IgG and Alexa Fluor 568-conjugated anti-rabbit IgG) for confocal microscopy. Hoechst staining was used for the nucleus. The microscope images represent those from three independent experiments.

FIG 3

SFTSV NSs inhibits the TRIM21-p62 interaction. (A and B) Mapping of TRIM21 for NSs binding. (A) HEK293T cells were transfected with NSs-3×Flag and the individual domain (RBCC [R, ring; B, B box; CC, coiled-coil] and PRY/SPRY) of TRIM21-V5, and WCEs were immunoprecipitated with an anti-V5 antibody, followed by immunoblotting with the indicated antibody. (B) HEK293T cells were transfected with NSs-GST and the individual domain (PRY/SPRY, PRY, or SPRY) of TRIM21-V5, and WCEs were pulled down by glutathione beads, followed by immunoblotting with the indicated antibody. (C) A model of the molecular action of NSs in TRIM21 function. Ub, ubiquitin. (D) HEK293T cells were transfected with increasing amount of NSs-3×Flag, TRIM21-V5, and p62-Myc, and WCEs were immunoprecipitated by anti-V5 antibody, followed by immunoblotting with the indicated antibody. (E) HeLa cells were transfected with NSs-3×Flag-GFP, TRIM21-V5, and p62-Myc. The cells were fixed and stained with primary antibodies (rabbit anti-V5 or mouse anti-Myc) and with secondary antibodies (Alexa Fluor 568-conjugated anti-rabbit IgG or Alexa Fluor 350-conjugated mouse IgG) for confocal microscopy. No nucleus staining was performed. (F) HeLa cells were transfected with NSs-3×Flag-GFP and p62-Myc. The cells were fixed and stained with primary antibody (mouse anti-Myc) and with secondary antibody (Alexa Fluor 568-conjugated anti-mouse IgG) for confocal microscopy. Hoechst staining was used for the nucleus. The microscope images represent those from three independent experiments. (G) HEK293T cells were transfected with NSs-V5 and Keap1-Flag, and WCEs were immunoprecipitated with an anti-Flag antibody, followed by immunoblotting with the indicated antibody.

FIG 4

Identification of the NSs-A46 mutant. (A) Amino acid sequence of NSs for alanine-scanning mutations. The NSs-A46 mutation is indicated with red. (B) HeLa cells were transfected with TRIM21-V5 and GFP-fused NSs-WT-3×Flag or individual alanine-scanning NSs mutant-3×Flag. The cells were fixed and stained with primary antibody (mouse anti-Myc) and with secondary antibody (Alexa Fluor 568-conjugated anti-mouse IgG) for confocal microscopy. Hoechst staining was used for the nucleus. The image at the top right represents an enlarged image of the area inside the red dashed-line box in the NSs-A46 panel.

FIG 5

TRIM21-binding-deficient NSs mutant. (A) Schematic diagram of the NSs-A46 mutant carrying alanine substitutions at K₂₂₆KTDG₂₃₀. (B) HEK293T cells were transfected with NSs-WT-3×Flag, NSs-A46-3×Flag, and TRIM21-V5, and WCEs were immunoprecipitated with an anti-V5 antibody, followed by immunoblotting with the indicated antibody. (C) HEK293T cells were transfected with pEBG-GST, GST-NSs-WT, and GST-NSs-A46, and WCEs were pulled down by glutathione beads, followed by immunoblotting with an anti-TRIM21 antibody. (D) HEK293T cells were transfected with GST-NSs-WT, GST-NSs-A46, TRIM21-V5, and p62-Myc, and WCEs were immunoprecipitated with an anti-V5 antibody, followed by immunoblotting with the indicated antibody. (E) HeLa cells were transfected with NSs-A46-3×Flag-GFP, TRIM21-V5, and p62-Myc. The cells were fixed and stained with primary antibodies (rabbit anti-V5 or mouse anti-Myc) and with secondary antibodies (Alexa Fluor 568-conjugated anti-rabbit IgG or Alexa Fluor 350-conjugated mouse IgG) for confocal microscopy. No nucleus staining was performed. (F) HEK293T cells were transfected with NSs-WT-3×Flag, NSs-A46-3×Flag, p62-HA, p62-Myc, and TRIM21-V5, and WCEs were immunoprecipitated with an anti-Myc antibody, followed by immunoblotting with the indicated antibody.

FIG 6

NSs mutants for interacting with three different host factors. (A to C) (Top) HEK293T cells were transfected with TRIM21-V5 and NSs-GST for four NSs (NSs-WT, NSs-A46, NSs-PA, and NSs-KR) (A), ABIN2-V5 and NSs-3×Flag for three NSs (NSs-WT, NSs-A46, and NSs-KR) (B), and TBK1-Flag with NSs-V5 for three NSs (NSs-WT, NSs-A46, and NSs-PA) (C). (Bottom) WCEs were pulled down by glutathione beads (A) or immunoprecipitated by anti-Flag antibody (B) or anti-V5 antibody (C), followed by immunoblotting with the indicated antibody.

FIG 7

NSs activates the Nrf2 pathway and induces Nrf2-mediated gene expression. (A and B) (Left) RAW 264.7 cells stably expressing the vector, NSs-WT, or NSs-A46 were harvested, and WCEs (A) and the cytoplasmic/nuclear fractions (B) were analyzed by immunoblotting for the indicated proteins. (Right) The relative levels of Nrf2 in RAW 264.7 cells compared to the levels of β-actin for WCEs (A) and p84 for the nucleus (B) were calculated. (C and D) The mRNA levels of the Nrf2 target genes (Hmox1 and Nqo1) (C) and Nrf2 (D) in RAW 264.7 cells expressing the vector, NSs-WT, or NSs-A46 were measured by qRT-PCR. (E) The Nqo1 mRNA level in RAW 264.7 cells treated with dimethyl sulfoxide (DMSO) and an Nrf2 inhibitor (INH; 10 μM) for 24 h was measured by qRT-PCR.

FIG 8

NSs induces CD36 expression via the Nrf2 pathway and increases lipid uptake. (A) The level of Cd36 mRNA in RAW 264.7 cells expressing the vector, NSs-WT, or NSs-A46 was measured by qRT-PCR. (B) CD36 surface expression by RAW 264.7 cells expressing the vector, NSs-WT, or NSs-A46 was measured by flow cytometry. The surface expression of FITC fluorescence on cells was determined using FlowJo software (left), and the relative expression levels are provided (right). FACS, fluorescence-activated cell sorting. (C) The level of Cd36 mRNA in RAW 264.7 cells treated with DMSO or an Nrf2 inhibitor (INH, 10 μM) for 24 h was measured by qRT-PCR. (D) RAW 264.7 cells were treated with rabbit IgG-FITC complexed with latex beads at a final dilution of 1:500. (Right) The cells were washed/harvested, and flow cytometry was applied to measure the internalized rabbit IgG-FITC-complexed latex beads. (Left) The percentage of cells that phagocytosed the beads was determined using FlowJo software. (E) The organic phase of RAW 264.7 cells was extracted to measure the amount of free or total (free and esterified) cholesterol. A colorimetric assay was used to measure intracellular cholesterol levels at an absorbance of 570 nm, based on cholesterol standards. (F) RAW 264.7 cells were plated and treated with fluorescently tagged cholesterol at 20 μg/ml for 48 h. The cells were washed, fixed with paraformaldehyde, and applied to a microplate reader with filter sets designed for FITC and GFP. The amount of cholesterol taken up was described at a relative level.

FIG 9

SFTSV-A46 mutant virus generation, plaque formation, and replication. (A) Sequence analysis of the NSs-A46 mutation of SFTSV-A46. (Top) The alanine substitution mutations of positive-strand viral mRNA are indicated in red. (Bottom) Sequencing results. (B) Plaque formation by SFTSV-WT and SFTSV-A46. The crystal violet-stained images are representative of those from three independent experiments. (C) The replication of SFTSV-WT and SFTSV-A46 in Vero E6 cells was examined by plaque assay (left) and by RT-PCR to measure the viral copy number (M segment) (right).

FIG 10

SFTSV-A46 pathogenesis in mouse models. (A and B) Ifnar^−/− mice were intramuscularly infected with 10² PFU of SFTSV-WT (n = 7) or SFTSV-A46 (n = 11) and were then monitored (A) and weighed (B) for 7 days. (C) Cd36 and Hmox1 mRNA levels in the spleens were measured by qRT-PCR at 4 days after infection with 10² PFU of SFTSV-WT or SFTSV-A46. (D) The viral copy number (M segment) in spleens from Ifnar^−/− mice infected with SFTSV-WT or SFTSV-A46 was measured by qPCR.