Fig 1.
STING deficiency leads to increased morbidity and mortality during WNV infection in vivo.
(A) Increased mortality in STING-/- mice. n = 22 per strain; Mantel-Cox analysis, p = 0.05*; p = 0.005**, p = 0.0005***. (B-C) Body weight loss (B) and clinical scores (C) are more pronounced in STING-/- mice, indicating increased morbidity during WNV infection. n = 22 per strain; two-way ANOVA, Bonferroni posttest; p = 0.05*; p = 0.005**, p = 0.0005***. (D) Schematic of infection and harvest time points. (E) Schematic of clinical signs and predicted pathology associated with WNV damage of the CNS. Anatomical model and clinical associations modified from previously described studies [85–89]. (F) Clinical signs observed during WNV infection of WT and STING-/- mice. R/D: ruffled/decreased; Ab: abdominal; R/T: reflex/tone. n = 10 per strain. (G) Body weight loss in (top) Terminal (T) vs (bottom) Survivor (S) cohorts. n = 22 per strain; two-way ANOVA, Bonferroni posttest; p = 0.05*; p = 0.005**, p = 0.0005***. (H) Clinical score analysis in (top) Terminal (T) vs (bottom) Survivor (S) populations. n = 22 per strain; two-way ANOVA, Bonferroni posttest; p = 0.05*; p = 0.005**, p = 0.0005***. (I-J) Pathological damage observed in the brains (I) spinal cord (J) by H&E staining in WT and STING-/- Terminal (T) and Survivor (S) mice. n = 3–9 per condition; students t-test (unpaired); p = 0.05*; p = 0.005**, p = 0.0005***.
Fig 2.
STING is not required for viral control of WNV in neurons in vitro or in the CNS during intracranial infection in vivo.
(A) WNV viral load in macro-dissected brain sections (cortex, sub-cortex, cerebellum and brain-stem) and spinal cord of WT and STING-/- infected mice, D4 and D8 post infection. n = 6–10 per strain per time-point. Graphed as stacking points. Limit of detection indicated by dashed line. Unpaired students t-test; p = 0.05*; p = 0.005**. (B) WNV IHC in the brains of mock infected and WNV infected WT and STING-/- Terminal and Survivor cohort. Mock tissues are unremarkable with non-specific staining of capillaries (arrows). Terminal mice have punctate staining near foci of gliosis (WT, circle) or neuronal degeneration (WT and STING-/-, arrows). No discernable specific signal for WNV antigen was observed in either WT or STING-/- Survivors, despite observable gliosis in STING-/- (circle). All panels, original magnification 200X. (C) TUNEL IHC stains of representative WT and STING-/- Survivor (18 dpi) mice. Brown stain indicates neuronal death. (D). Single and multistep virologic analysis of primary cortical neurons from WT and STING-/- mice. Pooled samples of 3 embryos per genotype. (E) Titer in mice infected with WNV via intracranial inoculation D4 pi. n = 6 WT and n = 5–6 STING-/-. Students t-test, p = 0.05*; p = 0.005**.
Fig 3.
STING deficiency leads to increased innate immune signaling during WNV infection.
(A-B) (A) WNV detection in BMDM by RT-qPCR, (B) innate immune response gene expression in WNV-infected BMDM over an infection time course. Bone marrow was harvested and differentiated into BMDM with mMCSF for 7 days. Cells were infected and harvested at the indicated time-points. Mock infected cells harvested at 12hpi. n = 3 infectious replicates. Results were reproducibly significant in multiple studies. Calculated as linear fold change over WT Mock. Unpaired students t-test; p = 0.05*; p = 0.005**, p = 0.0005***. (C) Splenic WNV titers at D4 and D8 post infection (PFU/g detected by plaque assay). n = 3–4. Graphed as stacking points. Limit of detection indicated by dashed line. Unpaired students t-test; p = 0.05*. (D) In vivo innate immune profile in splenic and CNS tissues. RT-qPCR detection of innate immune genes in the spleen, spinal cord and brain regions (brain stem, cerebellum, sub-cortex). Columns indicate individual mice; rows the different tissues. Calculated as log (WNV) or linear (immune) fold change over GAPDH in WT Mock. Dark red indicates values outside of (above) set scale.
Fig 4.
STING is not activated during WNV infection.
(A) Immunofluorescence STING re-localization assay to assess STING activation. HFF cells were treated with PBS (Mock), transfected with ctDNA (ISD) for 3hr infected with WNV for 24h (MOI = 1). Cells were stained for endogenous STING and dsRNA. The data are representative for two independent experiments. (B-C) Western blot to detect activation of endogenous STING during WNV infection. (B) HFF were infected with WNV MOI = 1 and harvested at the indicated times post inoculation (hpi). (C) HFF cells were infected with WNV (MOI = 1, MOI = 10) and lysed at 24 and 48 hpi. In parallel, HFFs were transfected with ctDNA and harvested at the indicated times (B-C). The data are representative of three replicate experiments.
Fig 5.
STING is required to program the adaptive immune response during WNV infection.
(A) Luminex analysis of cytokines and chemokines from serum in mock (PBS) and WNV infected mice at D4 and D8 post infection. n = 3–8 per condition. Dark red indicates values outside of (above) set scale. (B) CD4+ and CD8+ T cell frequency and absolute number in D8 post infection spleens (top). Flow schematic (bottom left) and CD4/CD8 ratio (bottom right). n = 4–8 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). (C) Characterization of D8 pi splenic CD8+ T cell sub-populations (CXCR3+, Ki67+, CD44+ and NS4b+). Left: frequency; Right: Absolute number. n = 4–8 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). (D) Characterization of D8 pi splenic CD4+ T cell sub-populations (CXCR3+, Ki67+, CD44+, CD73+ and CTLA-4+). Left: frequency; Right: Absolute number. n = 4–8 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). (E) Splenic FoxP3- and FoxP3+ CD4+ T cells. Top: frequency; Middle; absolute number; Bottom: gating scheme. n = 4–8 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). (F-G) Analysis of FoxP3+ (F) and FoxP3- (G) subpopulations (CXCR3+, Ki67+, CD44+, CD73+ and CTLA-4+). Left: frequency; Right: Absolute number. n = 4–8 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). B-F: WT, wild type; ST-/-, STING-/-.
Fig 6.
STING-/- have a defective adaptive immune response in the CNS during WNV infection.
(A) Cellular infiltrate in the brain in Terminal and Survivor mice by pathological review of H&E. n = 3–5. (B) CD3 IHC of WNV infected brains. Boxed regions, higher magnification, lower panel. T cells stain brown; hematoxylin counterstain. (C-H) Flow analysis of brain lymphocytes. Left: frequency; Right: Absolute number. N = 3–5 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***). (I) CD4/CD8 ratio in the brain. (J) Flow gating scheme. (K) FoxP3+CD25+CD4 T cells in brains. Left: frequency; Right: Absolute number. N = 3–5 per condition. Unpaired students t-test; p = 0.05*; p = 0.005*; p = 0.0005***).
Fig 7.
STING is essential in limiting neuropathology and WNV in the CNS.
(A-B) Regional analysis of pathology in WT and STING-/- brains (A) and spinal cords (B) in Terminal (mice meeting euthanasia criteria) and Surviving (harvested at study endpoint) mice. n = 3–5 per condition. (C-D) Representative H&E of pathological lesions and cellular infiltration in Brain (C) and Spinal Cord (D). (C) Mock tissues are unremarkable. Terminal mice have minimal lesions, observable in STING-/- as perivascular mononuclear cells and minimal gliosis. For both genotypes, Survivor mice have readable observable neuropathology which is severe in the STING-/- mice with intense gliosis, neuronal degeneration and death and perivascular cuffing with mononuclear cells. In the WT Survivors, there is a mild focus of gliosis and neuronal degeneration. All panels, original magnification 200X. (D) Mock tissues are unremarkable. Terminal mice have minimal lesions, observable as perivascular mononuclear cells. For both genotypes, Survivor mice have readable observable infiltration of mononuclear cells and neuronal degeneration. All panels, original magnification 200X. (E) M1 and M2 gene expression analysis by RT-qPCR. Calculated as relative fold change. n = 3–5 per condition.
Table 1.
Histological scoring.
Table 2.
RT-qPCR Primers.