Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Dec 12;213(13):2931-2947.
doi: 10.1084/jem.20160303. Epub 2016 Nov 29.

Interferon Regulatory Factor 2 Protects Mice From Lethal Viral Neuroinvasion

Affiliations
Free PMC article

Interferon Regulatory Factor 2 Protects Mice From Lethal Viral Neuroinvasion

Melody M H Li et al. J Exp Med. .
Free PMC article

Abstract

The host responds to virus infection by activating type I interferon (IFN) signaling leading to expression of IFN-stimulated genes (ISGs). Dysregulation of the IFN response results in inflammatory diseases and chronic infections. In this study, we demonstrate that IFN regulatory factor 2 (IRF2), an ISG and a negative regulator of IFN signaling, influences alphavirus neuroinvasion and pathogenesis. A Sindbis virus strain that in wild-type (WT) mice only causes disease when injected into the brain leads to lethal encephalitis in Irf2-/- mice after peripheral inoculation. Irf2-/- mice fail to control virus replication and recruit immune infiltrates into the brain. Reduced B cells and virus-specific IgG are observed in the Irf2-/- mouse brains despite the presence of peripheral neutralizing antibodies, suggesting a defect in B cell trafficking to the central nervous system (CNS). B cell-deficient μMT mice are significantly more susceptible to viral infection, yet WT B cells and serum are unable to rescue the Irf2-/- mice. Collectively, our data demonstrate that proper localization of B cells and local production of antibodies in the CNS are required for protection. The work advances our understanding of host mechanisms that affect viral neuroinvasion and their contribution to immunity against CNS infections.

Figures

Figure 1.
Figure 1.
IRF2 prevents lethal neuroinvasion of disparate neurotropic viruses. (A and B) Mortality (A) and disease progression (B) of mice infected i.p. with 13,000 PFU SVN were monitored for 14 d. For mortality, a total of 14–15 mice per genotype were infected, divided among seven independent experiments. For measuring clinical symptoms, 9–10 mice per genotype were scored each day. The p-value for the survival curve was determined by the log-rank test (P < 0.0001). (C) Viral titers in various tissues of mice infected i.p. with SVN were measured on days 0–7 p.i. Brain, serum, liver, and whole spleens from three to six mice per genotype per time point were harvested and/or homogenized for infection of BHK-21 cells in plaque assays. The dashed line indicates the plaque assay detection limit. Lines represent the mean. Two-way ANOVA test: brain, P < 0.0001; whole spleen, P < 0.001. Bonferroni posttests: *, P < 0.05; **, P < 0.01; ***, P < 0.001. (D) Mortality of mice infected intracranially (I.C.) with 13,000 PFU SVN was monitored for 7 d. A total of 7–10 mice per genotype were infected, divided among four independent experiments. The p-value for the survival curve was determined by the log-rank test. (E) Mortality of mice infected i.p. with 10,000 PFU VSV was monitored for 14 d. A total of seven mice per genotype were infected in four independent experiments. The p-value for the survival curve was determined by the log-rank test (P = 0.02).
Figure 2.
Figure 2.
Extensive neuronal death is observed only in the brains of Irf2−/− mice but not in those of WT mice with high viral titers. (A–C) Representative brain images of three SVN-infected mice from three different groups that demonstrate differences in neuronal morphology (A), perivascular cuffing (B), and neuronal cell death by TUNEL staining (C) are shown. Sections of brain from five SVN-infected WT littermate control mice (left; one animal), five moribund Irf2−/− mice (middle; one animal), and six WT mice with high CNS viral titers (right; one animal) were stained with H&E and examined for neuronal cell death (arrows), neuropil vacuolation (asterisks; A), and perivascular cuffing (arrowheads) by morphology (B). (C) TUNEL staining of slides further confirmed the presence of dead cells (brown; arrows). Bars: (A and C) 50 µm; (B) 100 µm. (D) SINV capsid staining of brain slides from the same mice including those with high CNS viral titers was quantified by ImageJ. A sagittal section of the entire brain near the midline was used for viral antigen staining. The p-value was determined by the unpaired, two-tailed Student’s t test. pos, positive.
Figure 3.
Figure 3.
Dysregulation of type I IFN signaling during SVN infection is not the cause of accelerated disease and death in Irf2−/− mice. (A) Serum IFN-α levels in WT and Irf2−/− mice on days 0–7 p.i. were measured by ELISA during SVN infection. Between three and six infected mice per genotype per time point were tested. Because 70% of the Irf2−/− mice succumbed to SVN infection between days 6 and 8 p.i., serum samples from less than three mice were tested on days 6 and 7 p.i. The dotted line indicates the detection limit of the ELISA (concentration of the lowest standard), and error bars represent SD. P-values were determined by the unpaired, two-tailed Student’s t test. *, P = 0.0167; **, P = 0.0073. (B) Protein concentrations of MCP-1 and IL-6 were measured by cytometric bead array. The right halves of the brains from three to four mice per genotype at baseline or on days 1, 3, and 5 p.i. were harvested and homogenized for measurement of inflammatory cytokines. Error bars represent SD. Two-way ANOVA test: MCP-1, P < 0.0001; IL-6, P < 0.05. Bonferroni posttests: ****, P < 0.0001. (C) Mice were infected i.p. with SVN and injected i.p. with Evans blue (EB) dye to measure BBB permeability on day 2, 3, or 5 p.i. Evans blue cannot cross the BBB unless there is a breach. Brains were harvested and homogenized 3 h after dye injection, and fluorescence present in the homogenates was measured spectrophotometrically. Between five and six mice per genotype served as uninfected controls or were infected, injected with the dye at various time points p.i., and harvested in a total of 11 independent experiments. Between two and four SVN-infected mice per genotype were not injected with the dye but instead harvested for measurement of background fluorescence. RU, relative units. (D) mRNA levels of the indicated ISGs in the brains of WT and Irf2−/− mice were measured by RT-qPCR. The right halves of the brains from three to four mice per genotype at baseline or on days 1, 3, and 5 p.i. were harvested and homogenized for RNA extraction. ISG mRNA levels present in the brains of the five WT mice with high CNS viral titers (Fig. 2 D) were also determined. Fold-changes normalized to baseline WT mice for all ISGs tested are shown in the table. ISG mRNA fold-changes with significant differences between the WT and Irf2−/− mice were determined using a two-way ANOVA test (IFITM3, P < 0.05; IFIT1, P < 0.05) and Bonferroni posttests (IFITM3 on day 5 p.i., P < 0.01; IFIT1 on day 5 p.i., P < 0.01; MX1 on day 5 p.i., P < 0.05; protein kinase R [PKR] on day 5 p.i., P < 0.01; IRF1 on day 5 p.i., P < 0.05). pos, positive. (E) Mortality of mice treated with 500 µg or 1 mg IFNAR1-blocking antibody or 500 µg isotype control antibody by i.p. route 1 d before or 2 d after i.p. infection with SVN. A total of 9–11 Irf2−/− and 7–10 WT mice were treated with IFNAR1-blocking antibody or isotype control and infected with SVN, divided among nine independent experiments. The p-values for survival curves were determined by the log-rank test. Only the significant differences are shown (isotype vs. IFNAR1 500 µg day −1 Irf2−/−, P = 0.0044; isotype vs. IFNAR1 500 µg 2 d p.i. Irf2−/−, P = 0.0436; isotype vs. IFNAR1 1 mg day −1 Irf2+/+, P = 0.0479).
Figure 4.
Figure 4.
Development and maturation of multiple immune cell subsets are compromised in Irf2−/− mice at baseline and upon SVN infection. (A) Bulk splenocytes from uninfected and SVN-infected WT and Irf2−/− mice on days 1, 2, and 5 p.i. were isolated and stained with antibodies specific for T cell (T), B cell (B), granulocyte (Gran), macrophage (Mac), DC, monocyte (Mono), and NK cell surface markers for flow cytometric analyses (see Fig. S1 for gating strategy). The graphs show frequencies of various immune cell populations as percentages of live splenocytes (black bars, WT; red bars, Irf2−/−). (B–E) The percentages of CD4+ and CD8+ T cells (B), CD8+ and CD11b+ cDCs (C), NK subsets with increasing maturation indicated by color intensity (D), and Ly6Chi and Ly6Clo monocytes (E) are shown. Black open triangles and red closed circles represent WT and Irf2−/− mice, respectively. Spleens from two to seven mice per genotype per time point were analyzed in a total of seven independent experiments. Error bars represent SD. Bonferroni posttests: *, P < 0.05; **, P < 0.01; ****, P < 0.0001.
Figure 5.
Figure 5.
Depletion of NK cells or MNPs does not significantly alter viral pathogenesis. (A) Mortality of WT mice treated i.p. with 300 µg NK1.1 antibody (Ab) or an isotype control 1 d before i.p. infection with SVN was monitored. A total of six to seven WT mice were treated and infected, divided between two independent experiments. (B) Mortality of WT and Irf2−/− mice treated i.p. with 200 µl clodronate or PBS liposomes 3 d before i.p. infection with SVN was monitored. A total of 9–12 mice per genotype were treated and infected, divided among eight independent experiments. The p-values for survival curves were determined by the log-rank test.
Figure 6.
Figure 6.
B cells and virus-specific IgG level are significantly reduced in Irf2−/− mouse brains. (A) Brain slides from five Irf2−/− moribund mice and six SVN-infected WT mice with high CNS viral titers from Fig. 2 D were stained with antibodies specific for CD3 (T cell), B220 (B cell), granzyme (cytotoxic T and NK cells), and Mac2 (macrophage). Immune cell infiltrates were semiquantitatively scored using a scale of zero to four as follows: 0, absent; 1, minimal; 2, mild; 3, moderate; and 4, marked. (B) Images that consistently sample eight different regions in the brains (Fig. S2) of WT and Irf2−/− mice were taken, and the B cell staining present in those images was quantified by ImageJ (measurement of number of positive pixels) or by eye (number of B cells). The p-values were determined by the unpaired, two-tailed Student’s t test (*, P < 0.05). pos, positive. (C) SINV-specific IgG and IgM antibodies in serum and brain homogenate samples of infected mice were measured by ELISA, as detailed in the ELISA section of Materials and methods. Mean OD value represents the level of virus-specific IgG or IgM antibody. Serum samples from three to six mice (except for n = 2 for Irf2−/− on day 7 p.i.) and brain homogenates from three mice per genotype per time point were tested for antibody titers. Error bars represent SD. Bonferroni posttests: *, P < 0.05. (D) PRNT was performed on serum samples of uninfected (n = 1 per genotype) and day 5 SVN-challenged WT and Irf2−/− mice (n = 4–5 per genotype). Infection with a constant concentration of virus incubated with serum samples from uninfected mice resulted in ∼200 plaques per well. The point at which the dilution curve crosses the dotted line represents the concentration of serum from SVN-infected mice required to reduce the number of plaques by 50%.
Figure 7.
Figure 7.
B cells are protective against lethal viral neuroinvasion. (A) Mortality of B cell–deficient μMT and age-matched WT C57BL/6J mice infected i.p. with 13,000 PFU SVN was monitored for 18 d. A total of 9–16 mice per genotype were infected, divided between two independent experiments. The p-value for the survival curve was determined by the log-rank test (P = 0.0182). (B) WT and Irf2−/− mice with B cells retroorbitally adoptively transferred from WT mice were infected i.p. with 13,000 PFU SVN the day after, and their mortality was monitored for 14 d. A total of six to eight mice per genotype were infected, divided between two independent experiments. The p-value for the survival curve was determined by the log-rank test (P = 0.001). (C) WT and Irf2−/− mice were infected i.p. with 13,000 PFU SVN, and 1 d later, serum from infected WT mice (day 7 p.i.) was adoptively transferred to the animals. Their mortality was monitored for 14 d. A total of 10 mice per genotype were infected, divided among five independent experiments. The p-value for the survival curve was determined by the log-rank test.

Similar articles

See all similar articles

Cited by 1 article

  • Genetic control of alphavirus pathogenesis.
    Baxter VK, Heise MT. Baxter VK, et al. Version 2. Mamm Genome. 2018 Aug;29(7-8):408-424. doi: 10.1007/s00335-018-9776-1. Epub 2018 Aug 27. Mamm Genome. 2018. PMID: 30151711 Free PMC article. Review.

References

    1. Auffray C., Fogg D.K., Narni-Mancinelli E., Senechal B., Trouillet C., Saederup N., Leemput J., Bigot K., Campisi L., Abitbol M., et al. 2009. CX3CR1+ CD115+ CD135+ common macrophage/DC precursors and the role of CX3CR1 in their response to inflammation. J. Exp. Med. 206:595–606. 10.1084/jem.20081385 - DOI - PMC - PubMed
    1. Baccala R., Welch M.J., Gonzalez-Quintial R., Walsh K.B., Teijaro J.R., Nguyen A., Ng C.T., Sullivan B.M., Zarpellon A., Ruggeri Z.M., et al. 2014. Type I interferon is a therapeutic target for virus-induced lethal vascular damage. Proc. Natl. Acad. Sci. USA. 111:8925–8930. 10.1073/pnas.1408148111 - DOI - PMC - PubMed
    1. Bick M.J., Carroll J.W., Gao G., Goff S.P., Rice C.M., and MacDonald M.R. 2003. Expression of the zinc-finger antiviral protein inhibits alphavirus replication. J. Virol. 77:11555–11562. 10.1128/JVI.77.21.11555-11562.2003 - DOI - PMC - PubMed
    1. Brennan K., and Bowie A.G. 2010. Activation of host pattern recognition receptors by viruses. Curr. Opin. Microbiol. 13:503–507. 10.1016/j.mib.2010.05.007 - DOI - PubMed
    1. Chiossone L., Chaix J., Fuseri N., Roth C., Vivier E., and Walzer T. 2009. Maturation of mouse NK cells is a 4-stage developmental program. Blood. 113:5488–5496. 10.1182/blood-2008-10-187179 - DOI - PubMed

Publication types

MeSH terms

Feedback