Flu Virus Attenuates Memory Clearance of Pneumococcus via IFN-γ-Dependent Th17 and Independent Antibody Mechanisms

Abstract

Bacterial coinfection is a major cause of influenza-associated mortality. Most people have experienced infections with bacterial pathogens commonly associated with influenza A virus (IAV) coinfection before IAV exposure; however, bacterial clearance through the immunological memory response (IMR) in coinfected patients is inefficient, suggesting that the IMR to bacteria is impaired during IAV infection. Adoptive transfer of CD4⁺ T cells from mice that had experienced bacterial infection into IAV-infected mice revealed that memory protection against bacteria was weakened in the latter. Additionally, memory Th17 cell responses were impaired due to an IFN-γ-dependent reduction in Th17 cell proliferation and delayed migration of CD4⁺ T cells into the lungs. A bacterium-specific antibody-mediated memory response was also substantially reduced in coinfected mice, independently of IFN-γ. These findings provide additional perspectives on the pathogenesis of coinfection and suggest additional strategies for the treatment of defective antibacterial immunity and the design of bacterial vaccines against coinfection.

Keywords: Immunology; Microbiology; Virology.

Graphical abstract

Figure 1

Memory-Mediated Bacterial Clearance Was Impaired in Coinfected Mice which Showed a Reduced Th17 Cell Response to Secondary Sp Infection (A) Mice were intratracheally (i.t.) inoculated with Streptococcus pneumoniae (Sp) and then challenged with Sp 5 weeks later. The mice were euthanized 1, 3, or 5 days after challenge for the determination of the numbers of colony-forming units (CFUs) in the lungs. (B) Schematic illustration of the experimental design. Mice were inoculated (i.t.) with Sp. Four weeks later, the mice were infected (i.t.) with the influenza A virus PR8 strain and challenged with Sp 7 days after infection. Five days after challenge, the mice were euthanized, and samples were taken for analyses. (C) The numbers of CFUs in the lungs were determined 5 days after challenge (n = 4–9). (D) Body weight was measured once daily (n = 4). (E) Mice were infected as in (B) but were challenged with a high dose of Sp following PR8 infection. The mortality rate was recorded daily (n = 10). Mice were infected, challenged, and euthanized as described in (B). (F) The proportion of IL-17⁺ cells among pulmonary CD4⁺ T cell population was determined by flow cytometry (n = 6–8). (G) IL-17 concentration in lung homogenates was determined by ELISA (n = 6–8). Data are represented as mean ± SEM of 2–3 independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (A and C) Two-tailed unpaired Mann-Whitney U nonparametric t test; (E) log rank test; (F and G) one-way ANOVA, followed by Tukey's multiple comparisons test.

Figure 2

The Severity of Coinfection Was Associated with an Increased IFN-γ Response to IAV Infection (A) IFN-γ concentration in lung homogenates after infection with the influenza A virus PR8 strain was determined by ELISA (n = 4). (B) The proportion of CD4⁻ NK1.1⁺ cells, CD3⁺ CD4⁺ cells, or CD3⁺ CD8⁺ cells among IFN-γ⁺ cell population in the lungs was determined by flow cytometry 7 days after PR8 infection (n = 3). Mice were infected, challenged, and euthanized as described in Figure 1B. (C) The numbers of colony-forming units (CFUs) in the lungs were counted (n = 9–11). (D) Body weight was measured as described in Figure 1D (n = 6). (E) Mice were infected and challenged as described in Figure 1E. The mortality rate was recorded daily (n = 20). Data are represented as mean ± SEM of 2–3 independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001. (A) One-way ANOVA, followed by Tukey's multiple comparisons test; (C) 2-tailed unpaired Mann-Whitney U nonparametric t test; (E) log rank test.

Figure 3

IFN-γ Deficiency Rescued the Response of Memory Th17 Cells to the Bacteria in Coinfected Mice (A and B) (A) Splenocytes from different groups of mice were cultured with or without heat-killed Streptococcus pneumoniae (HK-Sp) or (B) costimulated with the indicated concentration of recombinant mouse IFN-γ for 7 days (n = 6). IL-17 concentration in the culture supernatants was measured by ELISA. Mice were infected, challenged, and euthanized as shown in Figure 1B. (C and D) (C) The number of pulmonary IL-17⁺ CD4⁺ T cells and (D) the proportion of IL-17⁺ cells among CD4⁺ T cell population in the lungs were determined by flow cytometry (n = 3–8). (E) IL-17 concentration in lung homogenates was measured by ELISA (n = 6–8). (F) Schematic illustration of the experimental design for recombinant mouse IFN-γ treatment and infection in the mice. (G and H) (G) The number of pulmonary IL-17⁺ CD4⁺ T cells and (H) the proportion of IL-17⁺ cells among CD4⁺ T cell population in the lungs were detected by flow cytometry. (I) IL-17 concentration in lung homogenates was measured by ELISA. (J) Ifng^−/− mice were infected and challenged as described in Figure 1B and were intraperitoneally injected with neutralizing antibody directed against IL-17 (isotype antibody was used as a control) every other day after infection with the influenza A virus PR8 strain. The numbers of colony-forming units (CFUs) in the lungs were determined 5 days after challenge (n = 3–4). Data are represented as mean ± SEM of 2–3 independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (A-E and G-I) One-way analysis of variance (ANOVA) A, followed by Tukey's multiple comparisons test; (J) 2-tailed unpaired Mann-Whitney U nonparametric t test.

Figure 4

The Activation of DCs and IL-23 Produced by DCs Were Both Inhibited in IAV-Infected Mice Independently of IFN-γ Mice were infected and challenged as described in Figure 1B. Two days after Streptococcus pneumoniae (Sp) challenge, the mice were euthanized, and lung samples were taken for dendritic cell (DC) analyses (n = 4). (A–C) (A) The number of CD11c⁺ cells in the lungs and the mean fluorescence intensity (MFI) for (B) MHC class Ⅱ and (C) CD86 on the surface of DCs were determined by flow cytometry. (D and E) (D) IL-23 and (E) IL-12p70 concentration in lung homogenates was determined by ELISA. (F) Mice were infected, challenged, and euthanized as described in Figure 1B. IL-23 concentration in lung homogenates was determined by ELISA (n = 9–11). Data are represented as mean ± SEM of 2–3 independent experiments. ns, not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (A-F) One-way ANOVA, followed by Tukey's multiple comparisons test.

Figure 5

The Proliferation of Th17 Cells in Response to Secondary Sp Infection Was Suppressed in IAV-Infected Mice (A) Schematic illustration of adoptive transfer and mouse infection. The hilar lymph nodes (HLNs) of the recipients were removed for flow cytometric analyses. (B–G) (B) The number of donor CD4⁺ T cells and (C) the proportion of donor CD4⁺ T cells among the total CD4⁺ T cell population; (D) the proportion of fast-proliferating cells among the donor CD4⁺ T cell population and (E) the number of fast-proliferating cells; and (F) the proportion of RORγt⁺ cells among the donor CD4⁺ T cell population and (G) the number of RORγt⁺ CD4⁺ donor T cells were determined by flow cytometry. Data are represented as mean ± SEM of 2–3 independent experiments. ∗p < 0.05, ∗∗∗∗p < 0.0001. (B–G) One-way ANOVA, followed by Tukey's multiple comparisons test.

Figure 6

The Trafficking of Th17 Cells to the Lungs Was Delayed in Coinfected Mice in Response to Secondary Sp Infection Mice were infected, challenged, and euthanized as described in Figure 1B. (A) The relative expression of CCL17 in lung cells was determined by qPCR (n = 6–8). (B) Immunohistochemical staining of lung sections using an anti-CCL17 antibody. Scale bar represents 200 μm (left). The positive rate of CCL17 expression cells was assessed (right). The adoptive transfer was performed as described in Figure 5A. The lungs of recipients were removed for flow cytometric analyses (n = 3–4). (C–F) (C) The number of donor CD4⁺ T cells and (D) the proportion of donor CD4⁺ T cells among the total CD4⁺ T cell population and (E) the number of RORγt⁺ CD4⁺ donor T cells and (F) the proportion of RORγt⁺ cells among the donor CD4⁺ T cell population were determined by flow cytometry. Data are represented as mean ± SEM of 2–3 independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (A–F) One-way ANOVA, followed by Tukey's multiple comparisons test.

Figure 7

Memory Antibody Responses to Sp Were Reduced in Coinfected Mice Independently of IFN-γ Mice were infected, challenged, and euthanized as described in Figure 1B. Streptococcus pneumoniae (Sp)-specific IgG concentrations in serum were determined by ELISA (n = 6–12). Data are represented as mean ± SEM of 3 independent experiments. ns, not significant, ∗∗∗p < 0.001. One-way ANOVA, followed by Tukey's multiple comparisons test.