Interferon alpha/beta and interleukin 12 responses to viral infections: pathways regulating dendritic cell cytokine expression in vivo

Abstract

Interferon (IFN)-alpha/beta and interleukin (IL)-12 are cytokines critical in defense against viruses, but their cellular sources and mechanisms of regulation for in vivo expression remain poorly characterized. The studies presented here identified a novel subset of dendritic cells (DCs) as major producers of the cytokines during murine cytomegalovirus (MCMV) but not lymphocytic choriomeningitis virus (LCMV) infections. These DCs differed from those activated by Toxoplasma antigen but were related to plasmacytoid cells, as assessed by their CD8alpha(+)Ly6G/C(+)CD11b(-) phenotype. Another DC subset (CD8alpha(2)Ly6G/C(-)CD11b(+)) also contributed to IL-12 production in MCMV-infected immunocompetent mice, modestly. However, it dramatically increased IL-12 expression in the absence of IFN-alpha/beta functions. Conversely, IFN-alpha/beta production was greatly reduced under these conditions. Thus, a cross-regulation of DC subset cytokine responses was defined, whereby secretion of type I IFNs by CD8alpha(+) DCs resulted in responses limiting IL-12 expression by CD11b(+) DCs but enhancing overall IFN-alpha/beta production. Taken together, these data indicate that CD8alpha(+)Ly6G/C(+)CD11b(-) DCs play important roles in limiting viral replication and regulating immune responses, through cytokine production, in some but not all viral infections. They also illustrate the plasticity of cellular sources for innate cytokines in vivo and provide new insights into the roles of IFNs in shaping immune responses to viruses.

Figures

Figure 1.

IFN-α/β production by DCs during MCMV versus LCMV infections. Mice were infected intraperitoneally for 1.5 d with 104 PFUs MCMV, or for 2 d with 2 × 104 PFUs LCMV, or vehicle treated. The contribution of DCs to IFN-α/β production was then evaluated. (A) IFN-α/β (top graph, bioassay) and IFN-α (bottom graph, ELISA) titers in 24 h CM from CD11c-enriched (CD11c+), unfractionated (total), and CD11c-depleted (CD11c−) splenic leukocyte populations isolated from 129 mice. The data shown are representative of nine independent experiments for MCMV and three experiments for LCMV. The percentages of CD11c+ cells were generally <2% in CD11c− fractions and ranged from 3–7% in unfractionated splenic leukocytes versus 85–95% in enriched DC populations. BLD indicates below limit of detection (i.e., <1,500 pg/ml IFN-α or <8 U/ml IFN-α/β). Similar results were obtained in C57BL/6 mice (three experiments for MCMV and one for LCMV). (B) Analyses of the expression of IFN-α/β mRNA by splenic leukocytes freshly isolated from day 1.5 MCMV–infected or vehicle-treated 129 mice and fractionated for CD11c expression. (C) Levels of IFN-α/β production in 24 h CM from total CD11c-enriched (CD11c+) versus CD11c+CD8α+-enriched (CD8α+) cell populations from day 1.5 MCMV–infected 129 RAG-2M mice. Similar results were obtained with CD8α+ cells enriched from E26 mice (data not shown).

Figure 2.

Impact of in vivo depletion of Ly6G/C+ cells on IFN-α/β production during MCMV versus LCMV infections. (A) Flow cytometry analysis of the percentages of CD8α+Ly6G/C+ cells within splenic DCs from mice uninfected, infected by MCMV or by LCMV, and untreated or injected with anti-Ly6G/C versus control antibody. (B) Titration of IFN-α (ELISA, black bars) and IFN-α/β (bioassay, gray bars) in the sera from the same mice. The sera of the mice injected with the antibodies but uninfected were negative for IFN-α/β (data not shown). The data shown are representative of three independent experiments in 129 mice for MCMV and one for LCMV. Similar results were obtained in C57BL6 mice (2 and 1 experiments, respectively).

Figure 3.

IL-12 production by DCs during MCMV infection. The samples used and number of experiments performed are identical to those of Fig. 1 A and C. (A) IL-12p40 and p70 titers in 24 h CM from splenic leukocytes sorted for CD11c expression. (B) IL-12p40 titers in 24 h CM from CD11c-enriched (CD11c+) versus CD11c+CD8α+-enriched (CD8α+) cell populations from day 1.5 MCMV–infected 129 RAG-2M mice. Limits of detection were 20 pg/ml for IL-12p40 and 7 pg/ml for IL-12p70. Similar results were obtained with CD8α+ cells enriched from E26 mice and with purified CD8α+Ly6G/C+ DCs from immunocompetent 129 mice (data not shown).

Figure 4.

Comparative phenotypic analysis of the DC-producing IL-12 in response to STAg versus MCMV. The top dot plots show intracellular staining for IL-12p40 in CD11c-enriched cell populations from splenic leukocytes of 129 mice injected with vehicle or with STAg, or infected with MCMV. Note that IL-12 is expressed only in CD11c+ cells. Histogram plots show the stainings with antibodies specific for various membrane markers (gray area) versus isotype control (black line) within the IL-12+ DC gate. The dot plots show the relative distribution of various pairs of membrane markers within IL-12+ DC gate. Percentages of cells within IL-12 gates, histograms markers, or dot-plot quadrants are shown on the graphs. The data shown are representative of three independent experiments for MCMV infection.

Figure 5.

Impact of IFN-α/β functions on IL-12 production by DCs in vivo during MCMV infection. IFN-α/βR+/+, IFN-α/βR−/−, STAT-1+/+, STAT-1−/−, untreated, control antibody-treated, or anti-Ly6G/C-treated 129 mice were infected intraperitoneally for 1.5 d with 104 PFUs MCMV. The percentages and phenotypes of the DCs expressing IL-12p40 were then evaluated by flow cytometry. (A) Percentages of IL-12+ cells within DCs (CD8α+CD11b− DCs: gray bars; CD8α−CD11b+ DCs: black bars; and CD8α−CD11b− DCs: white bars). (B) MFI for IL-12 staining in CD8α+CD11b− DCs (gray bars) versus CD8α−CD11b+ DCs (black bars). The data shown are representative of three independent experiments for IFN-α/βR−/−. Similar results were obtained in STAT-1−/− mice on the C57BL6 genetic background (data not shown). CD8α and CD11b were consistently found to have mutually exclusive expression on IL-12+ DCs in the experimental systems described here.

Figure 6.

Localization of MCMV antigen, IFN-α/β, and IL-12 in the spleen. Immunohistochemistry or immunofluorescence were performed on spleen sections from uninfected or day 1.5–infected IFN-α/βR+/+ and IFN-α/βR−/− mice. (A) Staining for MCMV antigen by immunofluorescence (scale bars represents 1 mm for the larger photographs and 100 μm for the inserts). (B) IL-12 and (C) IFN-α/β staining by immunohistochemistry (scale bars represents 100 μm for the larger photographs and 50 μm for the insert).