Figure 1.
Vital role for G-CSF produced in the lung of infected mice.
(A) G-CSF measured in the sera of mice 3 days after infection with SeV or with influenza virus ×31 (IAV). (B) Percent weight change of G-CSF+/+ and G-CSF−/− mice infected with SeV (n = 6), influenza ×31 (n = 7) or mock treated controls. P>0.6 two-way ANOVA for infected groups. (C) Survival of G-CSF+/+ and G-CSF−/− mice monitored after infection with SeV (n = 6, p = 0.0009, logrank test) or influenza ×31 (n = 7, p = 0.012, logrank test). (D) SeV titer in the lungs of G-CSF+/+ and G-CSF−/− mice at the indicated time points (n = 8). (E) Viral titers in the lungs of G-CSF−/− mice infected with SeV and treated with daily injections of H2O or rhG-CSF four times before infection (Before) or continued treatment throughout the course of the experiment (Daily). ND: non-detected, NM: non-measured. Asterisks indicate p values. Data presented is representative of more than two independent experiments.
Figure 2.
G-CSF regulates the lung inflammatory response during virus infection.
(A–B) Cytokine secretion in the lungs of G-CSF−/− and G-CSF+/+ mice infected with SeV was measured by multiplex ELISA (n = 4).
Figure 3.
Adaptive immunity in the absence of G-CSF.
(A) SeV specific antibodies in lung homogenate of G-CSF+/+ and G-CSF−/− mice measured by ELISA 7 days after infection (n = 4). Experiment shown is representative of two independent experiments. IgG2a p>0.15, IgG3 p>0.07, IgG2b p<0.006, total IgG p<0.03 between infected G-CSF+/+ and infected G-CSF−/− for all dilutions. (B) Mock and SeV infected G-CSF+/+ or G-CSF−/− mice were injected i.v. 7 days after infection with CFSE-labeled peptide-pulsed splenocytes. Splenocytes labeled with a high dose of CFSE were pulsed with SeV NP peptide and those labeled with a low dose of CFSE were pulsed with the irrelevant influenza virus (PR8) NP peptide. Empty histograms show CFSE staining on mock-infected animals 24 h after injection of labeled/pulsed splenocytes. Filled histograms correspond to CFSE staining on SeV infected mice 24 h after injection of labeled/pulsed splenocytes (n = 4 per group). Experiment shown is representative of two independent experiments. (C) Survival of G-CSF+/+ and G-CSF−/− mice infected with a sublethal dose of influenza virus ×31 or mock treated and challenged at day 19 with a normally lethal dose of influenza virus PR8. Survival was monitored daily (n = 14/16 mice in ×31 challenged groups, n = 5 in mock treated controls). Experiment shown is representative of two independent experiments.
Figure 4.
G-CSF regulates the cellular infiltrate during virus infection.
(A) Cell numbers in the lungs of SeV infected mice calculated at the indicated time points from total cell counts and flow cytometry analysis. Populations were identified with the following markers: neutrophils (CD11b+Ly6c+Ly6G+), monocytes (CD11b+Ly6C+Ly6G−), pDCs (CD11b−B220+mPDCA+), CD4+ T cells (CD3+CD4+CD8−), CD8+ T cells (CD3+CD8+CD4−), and NK cells (CD11b+NK1.1+). Pre-gated on PI−CD45+ cells. Error bars represent standard deviation from the mean. Asterisks indicate p values (n = 3). (B) Ly6G+ granulocytes in the lung, BM and blood of G-CSF−/− and G-CSF+/+ mice mock treated or infected with SeV for 4 days. Pre-gated on PI−CD45+CD11b+ cells. Results are representative of more than two independent experiments.
Figure 5.
G-CSF deficiency reduces lung infiltrate during virus infection.
(A) Total cell counts in cytospins made with bronchoalveolar lavage (BAL) 8 days after infection with SeV. (B) Representative pictures of H&E-stained lung sections from naive and SeV infected G-CSF+/+ and G-CSF−/− C57BL/6 mice.
Figure 6.
Activated Ly6G+ granulocytes protect from mortality during virus infection.
(A) Cytokine gene expression in Ly6G+ enriched bone marrow cells. (B) Uptake of fluorescent latex beads by neutrophils (PI−CD45+Ly6G+CD11b+) in the lung of GCSF+/+ and G-CSF−/− mice four days after infection with SeV (n = 4). (C) Representative flow cytometry analysis of neutrophils (CD11b+Ly6cint) and monocytes (CD11b+Ly6chi) in the blood of mice injected with a single dose of anti-Ly6G antibody (1A8) or an isotype control. Neutrophil depleted WT C57BL6 mice were infected with SeV and monitored for (D) weight loss (p>0.9 two way ANOVA for infected groups) and (E) survival (n = 10 for isotype treated, SeV infected and n = 20 for Ly6G depleted and SeV infected. Logrank test, p<0.0001). Experiments shown are representative of two or three independent experiments.