Alternative cGAS signaling promotes herpes simplex encephalitis

Abstract

During infection, foreign DNA is sensed by cyclic GMP-AMP synthase (cGAS) leading to the production of cGAMP, STING-dependent type I interferon and proinflammatory cytokine expression, and autophagy. To prevent a response to self-DNA, cGAS activity is tightly regulated. Dysregulation of cGAS underpins interferonopathies, such as Aicardi-Goutières syndrome, as well as Lupus and neurodegenerative diseases like Parkinson's disease. Thus, cGAS and its product cGAMP are therapeutic targets. However, if cGAS functions independently of cGAMP signaling is undefined. Here, we identified an alternative signaling pathway that cGAS engages independent of cGAMP synthesis. We demonstrate that alternative cGAS signaling promotes hyperexpression of CXCL1 and enhanced neutrophil recruitment that facilitates viral dissemination during herpes simplex encephalitis. Our study reports of an alternative cGAS response independent of cGAMP, highlighting a previously uncharacterized scaffold function for cGAS.

Keywords: HSV; cGAS; infection; mouse model; type I IFN.

Fig. 1.

cGAS^E211A/D213A macrophages have impaired HSV-1 responses. (A–D) BMDMs from WT, cGAS^−/−, and cGAS^E211A/D213A mice were mock infected with PBS (Mock) or infected with the indicated multiplicities of infection (MOI) of HSV-1 or with Sendai virus (SV, MOI = 10) and the levels of the cytokines IFNβ (A), TNF-α (B), CXCL10 (C), and CXCL9 (D) were measured by ELISA. (E–J) Graphs showing the relative mRNA levels of Ifnβ (E), Cxcl10 (F), Ccl5 (G), Cxcl9 (H), Il-6 (I), and Isg15 (J) compared to the housekeeping gene, TBP, as assessed by qRT-PCR in WT, cGAS^−/−, and cGAS^E211A/D213A mice that were mock infected (Mock) or infected with HSV-1 (MOI = 10) for the indicated timepoints. (K) Western blot showing the total protein levels and phosphorylated protein levels for IRF3, TBK1, and STAT1, as well as total cGAS in WT, cGAS^−/−, and cGAS^E211A/D213A BMDMs infected with HSV-1 for the indicated time points (hours postinfection). β-Actin is shown as a loading control. The blot is representative of three experiments. A multiple comparison analysis was performed using a two-way ANOVA; **P < 0.01, ***P < 0.001.

Fig. 2.

cGAS^E211A/D213A mice are hypersusceptible to neuroinvasive HSV-1 infection. (A–D) WT, cGAS^−/−, and cGAS^E211A/D213A mice were mock infected with PBS (Mock) or infected with HSV-1 (2 × 10⁵ plaque-forming units [PFUs]) and monitored daily for weight change (A), survival (B), hydrocephalus (C), and neurological score (D) for up to 8 d (n =9 to 13 per group). The dotted line in A indicates baseline weight. (E) HSV-1 PFUs were measured in the brain tissue of WT, cGAS^−/−, and cGAS^E211A/D213A mice infected with HSV-1 (2 × 10⁵ PFUs) at day 5 postinfection. For comparisons of two groups, a two-tailed Student’s t test was performed. A multiple comparison analysis of Cgas^−/−, and cGAS^E211A/D213A was performed using a two-way ANOVA; ****P < 0.0001.

Fig. 3.

cGAS^E211A/D213A promotes hyperproduction of HSV-1-induced CXCL1 production. (A) Heatmap illustrating the differentially expressed genes in BMDMs from WT, cGAS^−/−, and cGAS^E211A/D213A mice transfected with dsDNA (100 ng) at the indicated time points. (B and C) WT, cGAS^−/−, and cGAS^E211A/D213A BMDMs were uninfected, transfected with HSV DNA (B), or infected with HSV-1 (MOI = 10) (C), and the level of Cxcl1 mRNA was assessed by qRT-PCR. (D–E) Protein levels of CXCL1 measured by ELISA in WT, cGAS^−/−, and cGAS^E211A/D213A mice that were mock infected with PBS (Mock) or infected with the indicated MOIs of HSV-1 or SV (SV, MOI = 1 for 24 h (D) or LPS (100ng/ml) for 24 h. (F) Cxcl1 mRNA expression assessed by qRT-PCR performed on BMDMs that were uninfected or infected with HSV-1 (MOI = 10) and either untreated or treated with a STING inhibitor (C176). (G and H) BMDMs from WT, Sting^−/−, and Ifnar^−/− mice were infected with HSV-1 (MOI = 1) for the indicated time points and assessed for CXCL1 mRNA levels by qRT-PCR (G) and Cxcl1 protein levels by ELISA (H). (I) Cxcl1 mRNA levels assessed by qRT-PCR in WT, cGAS^−/−, cGAS^E211A/D213A, Sting^−/−, and Sting^−/−cGAS^E211A/D213A BMDMs. For comparisons of two groups, a two-tailed Student’s t test was performed. A multiple comparison analysis was performed using a two-way ANOVA.

Fig. 4.

HSV-1-induced CXCL1 expression is dependent on TLR2 and Ox Phos. (A) WT and Cgas^E211A/D213A BMDMs were pretreated with a TLR2 inhibitory peptide (TLR2p) and assessed by qRT-PCR for Cxcl1 mRNA. (B) WT and MyD88^−/− BMDMs were mock infected with PBS (Mock) or infected with HSV-1 (MOI = 10) and Cxcl1 mRNA was measured at the indicated time points by qRT-PCR. (C and D) WT BMDMs were mock infected with PBS (Mock) or infected with HSV-1 (MOI = 10) and then treated with DMSO (Vehicle) or the TAK inhibitor (5z-7) and the Cxcl1 mRNA levels were assessed by qRT-PCR (C) and the CXCL1 protein levels were assessed by ELISA (D). (E) WT, cGAS^−/−, and cGAS^E211A/D213A BMDMs were mock infected with PBS (Mock) or infected with HSV-1 (MOI = 1) and untreated (Vehicle) or treated with the TAK inhibitor (5-z7) and Cxcl1 mRNA levels were assessed by qRT-PCR. (F) WT BMDMs were mock infected with PBS (Mock) or infected with HSV-1 (MOI = 10) and untreated or treated with a glycolysis inhibitor (2-DG) and Cxcl1 mRNA levels were assessed by qRT-PCR. All qRT-PCR results are shown relative TBP. For comparisons of two groups, a two-tailed Student’s t test was performed. A multiple comparison analysis was performed using a two-way ANOVA.

Fig. 5.

Modulation of CXCL1 levels regulates the severity of HSE. (A and B) WT, Cgas^−/−, and cGAS^E211A/D213A mice were mock infected with PBS (Mock) or infected with HSV-1 intraperitoneally and Cxcl1 protein levels in the serum were measured by ELISA (A) and Ly6G+ neutrophils were assessed by FACS (B). (C)WT, cGAS^−/−, and cGAS^E211A/D213A mice were infected with HSV-1 (2 × 10⁵ PFUs) via the ocular route and Cxcl1 mRNA levels were measured by qPCR. (D–F) cGAS^E211A/D213A mice were infected as in (C) and treated with an anti-Cxcl1 antibody or an isotype control (IgG) and weight change (D), hydrocephalus (E), and survival (F) measurements were taken at the indicated time points. (G) Survival rate in WT mice infected with the McKrae HSV-1 strain to induce HSE and treated with vehicle or a glycolysis inhibitor (2-DG).