Double-stranded RNA orbivirus disrupts the DNA-sensing cGAS-sting axis to prevent type I IFN induction

Abstract

Cyclic GMP-AMP synthase (cGAS) is a DNA sensing cellular receptor that induces IFN-I transcription in response to pathogen and host derived cytosolic DNA and can limit the replication of some RNA viruses. Some viruses have nonetheless evolved mechanisms to antagonize cGAS sensing. In this study, we evaluated the interaction between Bluetongue virus (BTV), the prototypical dsRNA virus of the Orbivirus genus and the Sedoreoviridae family, and cGAS. We found mitochondrial damage and DNA accumulation in the cytoplasm of infected cells. In addition, we show that BTV infection blocks DNA-induced IFN-I transcription and that virus infection prevents DNA sensing by inducing cGAS and STING degradation. We identify BTV-NS3 as the viral protein responsible for cGAS degradation, showing that NS3 physically interacts with cGAS and induces its degradation through an autophagy-dependent mechanism. Taken together, these findings identify for the first time a mechanism by which a dsRNA virus interferes with a DNA sensing pathway to evade the innate immune response.

Keywords: Autophagy; Bluetongue; Interferon; RNA virus; STING; cGAS.

Fig. 1

BTV infection disrupts mitochondrial membrane potential and induces cytosolic DNA accumulation. (A). Mitochondrial membrane potential assay. ST cells were mock-treated, infected at different MOIs of BTV-8 (0.1, 1 and 5) or inactivated BTV (iBTV). Twenty-four hours post-infection, cells were treated with JC-1, a cationic dye that aggregates in energized mitochondria yielding a red to orange colored emission (590 ± 17.5 nm) whereas when mitochondrial membrane is compromised, it predominantly remains in the cytosol as a monomer that emits green fluorescence (530 ± 15 nm). (FCCP) treatment (100µM for 1 h) was used as a depolarization control. Fluorescence was analyzed by flow cytometry and data of three independent experiments (n = 3) are represented as the mean ± SD of JC-1 Ratio of aggregates/monomers. (B). A portion of the cells from A were fixed and stained with anti-BTV-VP7 antibody and then analyzed by flow cytometry in order to verify BTV infection and geometric mean fluorescence intensity (MFI) of VP7 + cells is represented. (C-D). Analysis of mitochondrial distribution and morphology and detection of cytoplasmic DNA by immunofluorescence. Vero cells were mock-treated or infected with BTV-8 (MOI 5) and 0, 2, 8, 16–24 h post-infection (hpi) cells were fixed and stained for BTV-8 VP2, and (C) mitochondrial marker TOM20, or (D) ssDNA. Nuclei were stained with DAPI. Scale bar = 20 μm. TOM20 or ssDNA marker fluorescence at the different times post-infection was measured by quantifying the mean fluorescence and the area (µm²) of at least 40 cells (n ≥ 40) for each condition. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test, comparing each condition to mock condition. Statistical significance is represented as: NS = not significant; ** = p < 0.01; *** = p < 0.001

Fig. 2

Bluetongue virus interferes with DNA induced type-I interferon production. Sheep thymus cells (ST) were treated with mock, inactivated BTV (iBTV) or infected with BTV-8 (MOI 5). 6 h post-infection, cells were infected with (A-B) MVA (MOI 2) or (C-D) transfected with E.coli DNA (1 µg / 2 × 10⁵ cells) for 12 h. (A-D). Quantification of Ifnα and Isg15 transcripts by qPCR. Cell lysates were collected for RNA isolation and RT-qPCR quantification of Ifnα and Isg15 transcripts. Data of three independent experiments (n = 3) are represented as relative expression using the 2^−∆∆Ct method. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test. Statistical significance is represented as: NS = not significant; ** = p < 0.01; *** = p < 0.001. (E-F). Quantification of MVA DNA, by qPCR by amplifying a 281pb fragment from MVA genome, and BTV RNA by RT-qPCR, by amplifying a 165pb fragment of the segment 5 of BTV. Cell lysates were collected after removing supernatant and washing cell pellet with PBS. Nucleic acids were isolated from the lysates to quantify (E) MVA genomic DNA by qPCR and (F) BTV genomic RNA by RT-qPCR. Data are represented by means ± SD. Dotted line in panel F indicates RT-qPCR detection limit for BTV RNA

Fig. 3

Bluetongue virus interferes with cGAS pathway by degrading cGAS through the NS3 protein. (A). THP-1 cells were infected with BTV-8 (MOI 1). 24- and 48-hours post-infection, endogenous cGAS expression was analyzed by SDS-PAGE and immunoblotting with cGAS and β-actin antibodies (A). (C-D). THP-1 cells were infected with BTV-8 (MOI 0.1) and one hour later, medium was replaced with fresh medium containing cGAS inhibitor G140 (20 µM) or mock-treated. 16 h post-treatment, cells were collected to determine viral titers by analyzing percentage of viral replication by plaque assay (C) or qPCR (D). Data in (C) and (D) represent the mean ± SD of at least three independent experiments performed in duplicates(E). HEK-293T cells were transfected with cGAS and STING encoding plasmids for 16 h and then treated with mock or infected with BTV-8 (MOI 0.1, 1 or 5). 24 h post-infection, cell lysates were collected and protein expression was visualized via SDS-PAGE and immunoblotting with cGAS, STING and β-actin antibodies. (B, F). A portion of the cells from A and E were fixed and stained with anti-BTV-VP7 antibody and then analyzed by flow cytometry to verify BTV infection. Geometric mean fluorescence intensity (MFI) of VP7 + cells is represented. (G, H). HEK-293T cells were transfected with plasmids encoding cGAS, STING and increasing concentrations of plasmids encoding BTV NS3-FLAG (G) or NS4-FLAG (H) (250, 500 and 1000ng). 24 h post-transfection cell lysates were analyzed by SDS-PAGE and immunoblotting with cGAS, STING, FLAG-tag and β-actin antibodies. Densitometry in analysis of cGAS and STING relative to β-actin is expressed in A, E, G and H as a percentage using ImageJ software. Representative images of at least 3 individual experiments. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test. Statistical significance is represented as: * = p < 0.05; ** = p < 0.01; *** = p < 0.001

Fig. 4

NS3-defective BTV does not degrade cGAS but induces cytosolic DNA accumulation. (A, B). HEK-293T cells were transfected with a plasmid encoding cGAS and 16 h later mock-infected or infected with reverse genetic BTV-8 with a defective NS3 encoding segment (BTV‐8 ΔNS3) or with reverse genetic BTV‐8 control (rgBTV-8). 24 and 48 h post-infection, a portion of the cells were lysed in order to analyze protein expression by SDS-PAGE followed by immunoblotting with cGAS, and β-actin antibodies (A), while the other portion of the cells were fixed and stained with anti-BTV-VP7 antibody and analyzed by flow cytometry to confirm BTV infection (B). Geometric mean fluorescence intensity (MFI) of VP7 + cells is represented in B. Densitometry in analysis of cGAS relative to β-actin is expressed in A using ImageJ software. Representative images of at least 3 individual experiments. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test. Statistical significance is represented as: * = p < 0.05; ** = p < 0.01; *** = p < 0.001. (C). Detection of cytoplasmic DNA accumulation during NS3-defective BTV infection by inmunofluorescence. Vero cells were mock-treated or infected with reverse genetic BTV-8 with a defective NS3 encoding segment (BTV‐8 ΔNS3) or with reverse genetic BTV‐8 control (rgBTV-8) (MOI5), and 24 h post-infection cells were fixed and stained for BTV-8 VP2 and ssDNA marker. Nuclei were stained with DAPI. Scale bar = 20 μm. These data indicate that cytosolic DNA accumulation is independent of NS3 expression by BTV

Fig. 5

BTV NS3 protein interacts with cGAS. BTV-NS3 and cGAS co-immunoprecipitation assay. HEK-293T cells were co-transfected with (A) a plasmid encoding NS3-FLAG and increasing concentrations of a plasmid encoding cGAS-HA or with (B) cGAS-HA plasmid and increasing concentrations of NS3-FLAG plasmid. 24 h post-transfection, a fraction of the cell lysate was collected as whole cell extract (WCE) and the other fraction was incubated with (A) anti-FLAG or (B) anti‐HA affinity gel antibody beads to perform immunoprecipitation (IP) assays. Protein interactions were visualized via SDS-PAGE followed by immunoblotting

Fig. 6

BTV NS3 protein induces cGAS autophagic degradation. (A – B). HEK-293T cells were transfected with plasmids encoding cGAS and BTV-NS3 an empty vector as control. 24 h post-transfection, cells were mock-treated or treated with proteasome inhibitors MG-132 (10 µM) and lactacystin (20 µM) or autophagy inhibitors 3-MA (5 mM) and BAF-A1 (100 nM) (A) or with increasing concentrations of proteasome inhibitor MG-132 (0.2, 2 or 20 µM) or increasing concentrations of autophagy inhibitor BAF-A1 (1, 10 or 100 nM) (B). 6 h post-treatment, cell lysates were analyzed by SDS-PAGE and immunoblotting with cGAS, FLAG-tag and β-actin antibodies. (C). Autophagy inhibition assay. HEK-293T cells were transfected with control (CTL) siRNA or with ATG7 siRNA for 48 h and transfected with plasmids encoding cGAS or BTV-NS3-FLAG for 24 h. Cell lysates were analyzed by SDS-PAGE and immunoblotting using indicated antibodies. Densitometry in analysis of cGAS or STAT2 relative to β-actin is expressed in (A-D) as a percentage using ImageJ software. Representative images of at least 3 individual experiments. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test. Statistical significance is represented as: * = p < 0.05; ** = p < 0.01; *** = p < 0.001

Fig. 7

BTV NS3 ubiquitination is required for efficient cGAS degradation and inhibition of IFN-I transcription. (A). NS3 non-ubiquitinated mutant is less efficient in cGAS degradation. HEK-293T cells were transfected with plasmids encoding cGAS and BTV wild type NS3 or non-ubiquitinated NS3 mutant (K13,15/R-NS3). 24 h post-transfection, protein expression was visualized by SDS-PAGE and immunoblotting with cGAS, FLAG-tag and β-actin antibodies. Densitometry in analysis of cGAS relative to β-actin is expressed as a percentage using ImageJ software. (B). Evaluation of type-I interferon production of cells expressing BTV NS3 protein. 293T–IFN-β–FFLuc cells were transfected with an empty vector or plasmids encoding cGAS and STING and an empty vector, wild type NS3, or non-ubiquitinated NS3 mutant (K13,15/R-NS3). 24 h post-transfection, cell lysates were collected for luminescence quantification. Data representative of three independent experiments (n = 3) are represented means ± SD of fold induction over the empty vector condition. Statistical analysis was carried out with one-way ANOVA with Tukey´s multiple comparisons test. Statistical significance is represented as: NS = not significant; * = p < 0.05; *** = p < 0.001

Fig. 8

Schematic representation of cGAS degradation during BTV infection. Virus infection leads to a decrease in mitochondrial membrane potential, alterations in mitochondria morphology and accumulation of cytosolic DNA, potentially activating cGAS-STING pathway to induce IFN transcription to combat the infection. However, BTV is capable of degrading cGAS through an autophagy-dependent process mediated by BTV-NS3 protein, which binds to cGAS and induces its degradation as a mechanism to antagonize the IFN response