cGAS-mediated stabilization of IFI16 promotes innate signaling during herpes simplex virus infection

Abstract

Interferon γ-inducible protein 16 (IFI16) and cGMP-AMP synthase (cGAS) have both been proposed to detect herpesviral DNA directly in herpes simplex virus (HSV)-infected cells and initiate interferon regulatory factor-3 signaling, but it has been unclear how two DNA sensors could both be required for this response. We therefore investigated their relative roles in human foreskin fibroblasts (HFFs) infected with HSV or transfected with plasmid DNA. siRNA depletion studies showed that both are required for the production of IFN in infected HFFs. We found that cGAS shows low production of cGMP-AMP in infected cells, but instead cGAS is partially nuclear in normal human fibroblasts and keratinocytes, interacts with IFI16 in fibroblasts, and promotes the stability of IFI16. IFI16 is associated with viral DNA and targets to viral genome complexes, consistent with it interacting directly with viral DNA. Our results demonstrate that IFI16 and cGAS cooperate in a novel way to sense nuclear herpesviral DNA and initiate innate signaling.

Keywords: DNA sensing; innate immunity; protein–protein interactions; virus–host interactions.

Fig. 1.

Expression and localization of IFI16 and cGAS in human cells. (A) Western blot of IFI16 and cGAS in normal and immortalized cell lines. Lysates from equal numbers of cells were loaded on a 4–12% SDS/PAGE gel and probed for cGAS and IFI16 protein expression. (B) Nuclear/cytoplasmic fractionation of human cells. GAPDH and H3 were evaluated to confirm fractionation efficiency. The asterisk marks the cGAS specific band lost during siRNA depletion. (C) Basal levels of cGAS and STING in human cells. RNA was harvested from cells and analyzed by qRT-PCR. Transcripts were normalized to 18S rRNA. (D) Induction of IFNβ RNA in response to HSV-1 infection. Cells were infected with HSV-1 d109 virus at an MOI of 10 and RNA was harvested 6 hpi. IFNβ RNAs were normalized to 18S rRNAs and then relative to mock-infected samples.

Fig. 2.

Effects of depletion of STING, IFI16, or cGAS on innate immune responses to HSV-1 infection in normal human fibroblasts. (A) Efficiency of siRNA-mediated reduction of cellular gene expression at the RNA (Left) and protein (Right) level. RNA and protein were harvested from siRNA-transfected cells at 48 hpt. STING, IFI16, and cGAS transcripts were quantified by qRT-PCR and normalized to 18S rRNA. Whole-cell lysates were analyzed by Western blotting to determine the steady state protein levels of IFI16, STING, and cGAS. GAPDH was examined as a loading control. (B and C) Cytokine and IFN expression in HSV-1 d109 (B) or HSV-1 7134 (C) virus-infected human fibroblasts. siRNA-transfected cells were infected with the indicated HSV-1 recombinant viruses at an MOI of 10, and total RNA was harvested at 6 hpi. ISG54, IFNβ, and IL-6 expression were measured by qRT-PCR, normalized to 18S rRNA, and plotted as a fold-induction over mock-infected cells. *P < 0.05, **P < 0.01.

Fig. 3.

Effects of cGAS on IFI16 protein stability. (A) HFF cells were treated with control or cGAS siRNAs for 48 h and then incubated in medium containing 100 μg/mL CHX or control medium for 0, 1, 2, 3, and 4 h. Total cellular lysates were harvested, and IFI16 protein levels were examined by Western blotting. (B) IFI16 and cGAS levels were quantified by ImageJ. Protein levels were set to 1 at time 0 h for each condition. IFI16 protein levels in sicGAS treated cells were initially 50% of the sicontrol-treated cells. (C) siRNA-depleted HFF cells were incubated with 1 μM MG132 or control medium for 8 h. Total cellular lysates were harvested, and IFI16 protein levels were examined by Western blotting. *P < 0.05, **P < 0.01.

Fig. 4.

IFI16 and cGAS interact during HSV-1 infection in HFF cells. (A) Western blotting analyses of HFF cells lysates (Input), and immunopurified (Elution) fractions demonstrating the isolation of endogenous IFI16 in uninfected (Mock) and cells infected with WT HSV-1 virus at an MOI of 10 for 3 h. (B) Confirmation of amino acid sequences of endogenous cGAS peptides identified in the IFI16 immunopurifications. Representative collision-induced dissociation MS/MS spectra are shown. (C) Representative extracted ion chromatograms of cGAS peptides, detected by LC-MS data and visualized using Skyline software. The peaks illustrate the detection of cGAS peptides in either IFI16 (blue) or IgG (red) isolations in uninfected HFF cells or HFF cells infected with WT HSV-1.

Fig. 5.

cGAMP activity in plasmid DNA-transfected or HSV-infected HFF cells. (A) cGAMP production in response to DNA transfection or HSV-1 infection. HFFs, NOKs, and L929 cells were transfected with 200 ng of plasmid DNA or infected with HSV-1 d109 virus at an MOI of 1 for 24 h. Extracts of the transfected or infected cells were prepared, DNase- and heat-treated, and incubated with permeabilized HFF cells for 6 h. Reporter cells were treated with 200 nM cGAMP as a positive control. IFNβ RNA induction was analyzed by qRT-PCR and normalized to 18S rRNA. (B) Cellular gene expression in response to transfected or viral DNA. Cells were infected and transfected as in A. Total RNA was harvested at 8 and 24 h poststimulation and analyzed by qPCR. (C) 293T cells were transfected with 1 μg total DNA: pEV or pcGAS for 24 h. Permeabilized HFF cells were incubated for 30 min with DNase- and heat-treated cell extracts, harvested at 6 h, and IFNβ induction was analyzed by qRT-PCR. Reporter cells were treated with 200 nM cGAMP as a positive control. IFNβ transcripts were normalized to 18S rRNA. The data represent three independent experiments. (D) HFF cells were transfected with 200 ng of plasmid DNA (Left) or infected with HSV-1 d109 (MOI = 3) (Right). Cells were harvested at the indicated times, and cellular proteins were examined by Western blotting.

Fig. 6.

Localization of IFI16 and cGAS in HSV-infected cells. (A) HFF cells were mock-infected or infected with HSV-1 7134 virus (MOI = 10). Cells were fixed at 24 hpi, and cells on the edge of viral plaques were imaged by IFI16 and ICP8 staining. (B) Human foreskin fibroblasts were mock-infected or infected with WT HSV-1 at an MOI of 100 in the presence or absence of 100 μg/mL CHX. Cells were fixed and stained at 2.5 hpi for IFI16 and cGAS. (Magnification: 63×.)

Fig. 7.

Contribution of cellular signaling components to innate immune responses to transfected DNA. (A) Kinetics of innate transcriptional responses to transfected DNA. HFF cells were transfected with 200 ng of plasmid DNA, and total RNA was harvested and analyzed by qRT-PCR at the indicated time points. Values were normalized to cellular GAPDH and are plotted as a fold-induction over a DNA-negative Lipofectamine LTX control. (B and C) Kinetics of innate transcriptional responses in cGAS- or IFI16-depleted cells transfected with plasmid DNA (200 ng) (B) or infected with HSV-1 d109 (MOI = 3) (C). (D) cGAMP activity. siRNA-depleted cells were transfected as in B, and cGAMP was harvested at 24 hpt and assayed on reporter HFF cells. ISG54 and IFNβ mRNA induction were measured in reporter cells at 6 h posttreatment by qRT-PCR. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 8.

Model of the distinct roles of IFI16 and cGAS in sensing nuclear herpesviral or transfected DNA. Following HSV-1 infection (Left), IFI16 senses HSV-1 viral DNA in the nucleus and promotes downstream signaling to activate IFNβ expression. cGAS indirectly activates signaling by promoting IFI16 stability. During DNA transfection (Right), cGAS initially senses transfected DNA (1) followed by IFI16-dependent sensing (2).