Interleukin-1 Signaling Prevents Alveolar Macrophage Depletion during Influenza and Streptococcus pneumoniae Coinfection

Abstract

Influenza and bacterial coinfection is a significant cause of hospitalization and death in humans during influenza epidemics and pandemics. However, the fundamental protective and pathogenic mechanisms involved in this complex virus-host-bacterium interaction remain incompletely understood. In this study, we have developed mild to lethal influenza and pneumococcal coinfection models for comparative analyses of disease pathogenesis. Specifically, WT and interleukin-1 receptor type 1-deficient (Il1r1^−/−) mice were infected with influenza virus, and then super-challenged with noninvasive pneumococcal serotype 14 (Spn14) or 19A (Spn19A). The coinfections were followed by comparative analyses of inflammatory responses and animal protection. We found that resident alveolar macrophages are efficient in clearance of both pneumococcal serotypes in the absence of influenza infection; on the other hand, they are essential for airway control of Spn14 infection but not Spn19A. In agreement, TNF-α and neutrophils play a compensatory protective role in secondary bacterial infection associated with Spn19A; however, the essential requirement for AM-mediated clearance significantly enhances the virulence of Spn14 during post-influenza pneumococcal infection. Furthermore, we show that although IL-1 signaling is not required for host defense against pneumococcal infection alone, it is essential for sustaining antibacterial immunity during post-influenza pneumococcal infection, as evidenced by significantly aggravated bacterial burden and animal mortality in Il1r1^−/− mice. Mechanistically, we show that through preventing AM depletion, inflammatory cytokine IL-1 signaling is critically involved in host resistance to influenza and pneumococcal coinfection.

Introduction

Polymicrobial infection is an important but poorly understood clinical problem. Particularly, influenza and pneumococcal coinfection is a significant cause of morbidity and mortality during influenza epidemics and pandemics (1–4). Early studies of this virus-bacterium synergy suggest that influenza virus induces epithelial damage to promote bacterial adherence and systemic invasion (1, 2, 5, 6). Conversely, recent investigations have been more focused on influenza-induced defects in innate antibacterial immunity (7–15). Nonetheless, the fundamental protective and pathogenic immune mechanisms involved in the complex virus-host-bacterium interaction remain incompletely understood.

Recent evidence suggests that suppression of innate immune responses is responsible for influenza-induced susceptibility to pneumococcal infection. Conversely, excessive inflammation is considered to be the key contributor to the morbidity and mortality associated with influenza infection (16). Cytokines are essential regulators of inflammatory responses. Of particular interest, IL-1 signaling is known to exacerbate acute lung immunopathology but increase animal survival during influenza infection (17); on the other hand, it exerts a redundant but protective role in immune defense against invasive pneumococcal infection (18). Nonetheless, its role in influenza/S. pneumoniae synergy is still unclear. In this study, we focused on the effect of IL-1 signaling during influenza and pneumococcal coinfection, with a goal to understand the protective and detrimental contributions of inflammatory responses to the disease outcome.

Epidemiology study has demonstrated that influenza is associated with the greatest increases in pneumococcal disease caused by serotypes with lower invasive potential (19). In the absence of influenza infection, both S. pneumoniae serotypes 19A (Spn19A) and 14 (Spn14) are considered to be noninvasive strains in mice (20, 21). These low-virulent strains allowed us to test the hypothesis that influenza infection enables otherwise noninvasive S. pneumoniae infection to cause severe to fatal pneumonia. Given the prevalence of Spn19A and Spn14 carriages in humans (22, 23), we consider that our current models of influenza and pneumococcal coinfection are more clinically relevant.

Distinct from previous investigations (24–26), we have developed mild to lethal coinfection models for comparative analyses of disease progression in this study. Specifically, WT and interleukin-1 receptor type 1-deficient (Il1r1^−/−) mice were infected with influenza virus X31 (H3N2) or PR8 (H1N1), and then super-challenged with pneumococcal serotype 19A or 14. Such an approach not only allowed us to address the multifactorial nature of coinfection, but also to separate upstream causes of defective antibacterial immunity from downstream inflammatory responses to bacterial outgrowth. For the first time, our study demonstrated that through preventing alveolar macrophage (AM) depletion, inflammatory cytokine IL-1 signaling is critically involved in host resistance to influenza and pneumococcal coinfection.

Materials and Methods

Murine model of viral and bacterial infection

Specific pathogen-free, C57BL/6 WT, Il1r1^−/− and Casp1^−/− mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and bred at University of Nebraska Medical Center following IACUC guidelines. All animal experiments were approved by University of Nebraska Medical Center, and all experiments were carried out in accordance with University of Nebraska Medical Center Assurance of Compliance with PHS Policy on humane Care and Use of Laboratory Animals, which is on file with the Office of Protection from Research Risks, NIH.

Viral challenge was performed with a sublethal dose of X31 (~5×10³ PFU/mouse) or PR8 (~10 PFU/mouse) administered i.n. to anesthetized mice in 50 µl of sterile PBS. Titers of virus stocks and viral levels in the bronchoalveolar lavage fluids (BALF) and lungs of infected mice were determined by plaque assays on MDCK cell monolayers. Of note, PR8 is a mouse-adapted H1N1 influenza A strain with high-pathogenicity, and X31 is an H3N2 influenza A strain that only causes mild illness in mice (27).

To induce bacterial pneumonia, anesthetized mice were inoculated i.n. or i.p. with 50 µl of PBS containing 2×10⁵ CFU of ATCC serotype 19A strain BAA-475 (multi-drug resistant) or 2×10⁴ CFU of serotype 14 strain TJO983 unless otherwise specified (20). Bacterial burdens in the BALF and lungs were measured by sacrificing infected mice at the indicated time points, and plating serial 10-fold dilutions of each sample onto blood agar plates. Although pneumococcal serotype 19A is associated with a high level of resistance to multiple antibiotics, with a LD50 of >10⁶ CFU after intranasal (i.n) or intraperitoneal (i.p) infection, serotypes 19A and 14 are both considered to be noninvasive strains in normal mice (20, 21).

Bronchoalveolar lavage (BAL) cell analysis

BALF samples were collected by making an incision in the trachea and lavaging the lung twice with 0.8 ml PBS, pH 7.4. Total leukocyte counts were determined using a hemacytometer.

For flow cytometric analysis, BALF cells were incubated with 2.4G2 mAb against FcγRII/III, and stained with APC-conjugated anti-CD11c (BioLegend), PE-Cy7-conjugated anti-CD11b (BD Biosciences), FITC-conjugated anti-Ly6G (clone 1A8, BD Biosciences), PerCp-Cy5.5- conjugated anti-Ly6C (eBiosciences), and PE-conjugated anti-Siglec-F (BioLegend) mAbs for myeloid cell analysis. In some experiments, cells were stained with PE-conjugated anti-active caspase 3 using a commercially available kit from BD Biosciences before surface marker staining. To differentiate early-stage apoptotic cells from the late-stage apoptotic and necrotic cells, BALF cells were stained with Fixable Viability Dye (FVD) eFluor® 780 and Annexin V PerCP-eFluor® 710 (eBiosciences), using BUV395-conjugated anti-CD11b (BD Biosciences), FITC- conjugated anti-Ly6C (BD Biosciences), BV421-conjugated anti-Ly6G (clone 1A8), PE-conjugated anti-Siglec-F and APC-conjugated anti-CD11c mAb for cell surface markers. The stained cells were analyzed on a BD LSRII-green using BD FACSDiva and FlowJo software analysis.

Determination of cytokine/chemokine production by ELISA

BALF were harvested and assayed for TNF-α, IL-1β, IL-6, IFN-γ, KC and MIP-2 by ELISA using commercially available kits from BD Biosciences and R&D Systems (Minneapolis, MN).

Neutrophil depletion

Neutrophils were depleted using anti-Gr-1 mAb RB6-8C5 (BioXCell). Specifically, starting at one day before bacterial infection, mice were injected i.p. with anti-Gr-1 mAb RB6-8C5 (0.1 mg/day) to deplete neutrophils or with rat IgG as a control. The efficiency of neutrophil depletion in bacterial-infected mice was confirmed by flow cytometry.

Alveolar macrophage depletion

Depletion of alveolar macrophages in mice was achieved by i.n. instillation of 100 µl of clodronate-containing liposomes 48 h before infection. Another group of mice was instilled with PBS-containing liposomes as a control for possible liposome effects (12).

Quantification of live bacteria associated with phagocytes

Seven days after inoculation of X31 influenza virus, mice were i.n. infected with Spn14 and 24 h later, BALF cells were collected and sorted using FACSAria (BD Biosciences). Cell populations in the airway were classified using these surface markers: AM (CD11c^hiCD11b^low), inflammatory monocytes (CD11b⁺Ly6C^hi), and neutrophils (CD11b⁺Ly6G⁺). Cytospins of cells were prepared and Diff-Quick stained. The sorted cells were also washed in PBS, lysed in sterile water, and then plated on blood agar plates. The number of CFU was expressed per 100 cells for each myeloid cell subsets.

In vivo TNF-α and IFN-γ neutralization

For TNF-α neutralization, mice were inoculated i.p. with 160 µg/day of MP6-XT22 rat anti-murine TNF-α mAb beginning at the time of Spn19A infection. Other groups of mice were treated with rat IgG as controls. For IFN-γ neutralization, mice were inoculated i.p. with 500 µg of XMG1.2 rat anti-murine IFN-γ mAb on days 5, 6 and 7 following influenza infection. Other groups of mice were treated with rat IgG as controls.

Statistics

Significant differences between experimental groups were determined using a two-tailed Student t-test (to compare two samples), an ANOVA analysis followed by Tukey's multiple comparisons test (to compare multiple samples) or Mann Whitney test (nonparametric test) in GraphPad Prism 6 (La Jolla, CA). Survival analyses were performed using the log-rank test. For all analyses, a P value <0.05 was considered to be significant.

Results

IL-1 signaling improves protection against lethal influenza and pneumococcal pneumonia

To determine the protective or pathogenic role of IL-1 receptor (IL-1R) signaling during influenza/S. pneumoniae coinfection, C57BL/6 WT and Il1r1^−/− mice were infected either with influenza virus X31 or high-virulent PR8, and seven days later super-challenged with S. pneumoniae serotype 19A (Spn19A) or 14 (Spn14). Note that WT and Il1r1^−/− mice infected with either influenza X31 or PR8 showed transient weight loss (~10% of their original body weight) but no animals succumbed to influenza infection alone, whereas mice infected with pneumococcal Spn19A or Spn14 alone did not lose body weight nor did they display signs of clinical disease. Conversely, as expected, influenza and pneumococcal coinfection significantly increased animal mortality (Fig. 1). The distinct coinfection outcomes further suggest that disease severity is mutually determined by all three components in the interaction, namely, viral virulence (PR8>X31) (Fig. 1C–D), bacterial virulence (Spn14>Spn19A) (Fig. 1B, 1D), and host susceptibility (Il1r1^−/− >WT) (Fig. 1B–C). Importantly, these mild to lethal influenza and pneumococcal coinfection models allowed us to dissect the protective and pathogenic immune mechanisms involved in the complex virus-host-bacterium interaction.

Figure 1. IL-1 signaling improves protection against lethal influenza and pneumococcal pneumonia.

Survival of C57BL/6 WT and Il1r1^−/− mice after super-challenge with Spn19A (A, B) or Spn14 (C, D) on day 7 after X31 (A, C) or PR8 (B, D) influenza infection. ***, P< 0.001, log-rank test. Data shown are representative of at least two independent experiments.

IL-1 signaling enhances pulmonary bacterial clearance during influenza and pneumococcal coinfection

To determine how IL-1 signaling contributes to protection against secondary pneumococcal infection, we first compared antibacterial defense in WT and Il1r1^−/− mice with or without prior influenza infection. In the absence of X31 infection, bacterial burdens were comparable in WT and Il1r1^−/− mice (Fig. 2A–B), suggesting that IL-1 signaling is dispensable for acute clearance of Spn19A and Spn14 in healthy mice. In contrast, under coinfection conditions, Il1r1^−/− mice exhibited significantly increased bacterial burdens compared with corresponding WT controls (Fig. 2A–B), indicating that IL-1 signaling enhances pulmonary bacterial clearance during influenza and pneumococcal coinfection.

Figure 2. IL-1 signaling is required for bacterial clearance during influenza and pneumococcal coinfection.

C57BL/6 WT and Il1r1^−/− mice were infected with influenza X31 and seven days later super-challenged with Spn19A or Spn14. At 24 h lungs were analyzed for (A) Spn19A burdens (mean±SD, n=5), (B) Spn14 burdens, and (C) viral titers (mean±SE, 8~14 mice/group). P<0.05, two-way ANOVA; *P< 0.05, *P< 0.01, ***P< 0.001, Tukey's multiple comparisons test. Data shown are representative of at least two independent experiments.

To determine the possible impact of IL-1 signaling on antiviral immunity, we examined viral titers in X31-infected WT and Il1r1^−/− lungs. The results indicate that IL-1 signaling plays a limited role during X31 infection alone or X31/Spn19A coinfection, but facilitates viral clearance in WT mice during X31/Spn14 coinfection, probably due to a stimulatory effect of Spn14 super-infection on antiviral response (Fig. 2C). Of interest, this temporary “beneficial” effect of bacterial super-infection on viral clearance has been observed in our previous studies (16). Nonetheless, viral burdens in Il1r1^−/− mice were unaffected by bacterial super-infection, suggesting that differential viral control is not the primary cause of 100% mortality observed in Il1r1^−/− mice after X31/Spn14 coinfection.

IL-1 signaling prevents depletion of alveolar macrophages (AMs) during influenza and pneumococcal coinfection

Given the clear difference in the ability of Spn19A and Spn14 to replicate and cause lethal pneumonia in influenza-infected mice (Figs. 1&2), we determined the immune factors responsible for their clearance from the lower airways (Fig. 3). Clodronate-depletion of AMs led to a significant increase in bacterial burdens at 24 h after infection with Spn14 or Spn19A (Fig. 3A–C). Interestingly, antibody-mediated depletion of neutrophils had no significant effect on bacterial control (Fig. 3D–E), and at an inoculum of 2×10⁵ CFU/mouse, the majority of bacteria were eliminated from lungs within 24 h after infection with Spn14 or Spn19 (Fig. 3E). These findings indicate that in the absence of influenza infection, AMs are essential for efficient clearance of Spn19A and Spn14, with neutrophils playing an insignificant role. Of particular interest, after AM depletion, bacterial outgrowth, i.e., an increase in 24 h lung bacterial counts compared with the inoculum, was detected only in mice infected with Spn14 (Fig. 3B) but not Spn19A (Fig. 3C). Together, these results suggest that while AMs are efficient in clearance of both serotypes of pneumococci, they are essential for lung control of Spn14 but not Spn19A.

Figure 3. Alveolar macrophages are more important for lung clearance of Spn14 than Spn19A.

(A) Numbers (mean±SD, n=4) of airway alveolar macrophages (AM) and neutrophils (PMN), and (B–C) lung bacterial burdens at 24 h after infection of WT mice with Spn14 (A–B) or Spn19A (C). Lip-control, PBS-liposome treated mice; Lip-Clodronate, clodronate-liposome treated mice. In (B–C), the CFU are represented as percentages of the bacterial titer in the inoculum. (D) Numbers (mean±SD) of airway AMs and neutrophils, and (E) lung bacterial burdens (mean±SD) at 24 h after infection of rat IgG or α-PMN antibody-treated WT mice (n=4) with 2×10⁵ CFU of Spn19A or Spn14. In (A–B), P<0.05, one-way ANOVA; *P< 0.05, ***P< 0.001, Tukey's multiple comparisons test. In (C–D), *P< 0.05, ***P< 0.001, t test. Data shown are representative of two independent experiments.

Of particular note, prior X31 infection promoted Spn19A outgrowth in Il1r1^−/− but not WT mice (Fig. 2A). We wanted to know whether this is due to the regulatory effect of IL-1 on airway phagocyte recruitment. Thus, the percentage and number of CD11c⁺ AMs, CD11b⁺Ly6G⁻ monocytes, and CD11b⁺Ly6G⁺ neutrophils in WT and Il1r1^−/− airways was quantified by flow cytometry analysis following X31 and/or Spn19A infection (Fig. 4). Compared with Spn19A infection alone, the percentage (Fig. 4A) and number (Fig. 4B) of neutrophils were similarly increased in WT and Il1r1^−/− airways following X31/Spn19A coinfection. In contrast, the percentage (Fig. 4A) and number (Fig. 4B) of AMs were significantly decreased in coinfected Il1r1^−/− airways compared with corresponding WT controls (Fig. 4B). Considering the critical contribution of AMs to pneumococcal clearance, these results suggest that after influenza infection, IL-1 signaling is required for maintaining alveolar macrophages for effective antibacterial defense. Even so, only few Il1r1^−/− mice died from X31/Spn19A coinfection (Fig. 1A). These results further indicate that while IL-1 signaling is required for maintaining AMs for efficient bacterial control, it is not essential for animal survival from X31/Spn19A coinfection.

Figure 4. IL-1 signaling prevents AM depletion during X31/Spn19A coinfection.

C57BL/6 WT and Il1r1^−/− mice were infected with influenza X31 and seven days later super-challenged with Spn19A or PBS control. At 24 h BALF (4~5 mice/group) were analyzed for (A) airway myeloid cell profiles (mean±SD), and (B) numbers (mean±SD) of AMs, inflammatory monocytes (MO), and neutrophils (PMN) after X31 and/or Spn19A infection. *P< 0.05, **P< 0.01, ***P< 0.001, t test. Data shown are representative of two independent experiments.

TNF-α protects Il1r1^−/− mice against an otherwise lethal X31/Spn19A coinfection

The observations above suggest that alternative immune mechanisms exist to contain and clear bacteria in Il1r1^−/− mice during X31/Spn19A coinfection. Analysis of inflammatory cytokine profiles during X31/Spn19A coinfection revealed significantly increased TNF-α production in Il1r1^−/− mice compared with WT animals (Fig. 5A). To determine whether the increased TNF-α production is protective or detrimental during coinfection, we examined the impact of TNF-α on lung bacterial clearance and animal survival in Il1r1^−/− mice after X31/Spn19A coinfection. Administration of anti-TNF-α antibodies resulted in 10-fold increases in lung bacterial burden (Fig. 5B) and rapid death (Fig. 5C) of Il1r1^−/− mice compared with rat IgG-treated controls. These results indicate that intact TNF-α response is essential to rescue Il1r1^−/− mice from an otherwise lethal X31/Spn19A coinfection.

Figure 5. TNF-α protects Il1r1^−/− mice against X31/Spn19A coinfection.

C57BL/6 WT and Il1r1^−/− mice were infected with influenza X31 and seven days later super-challenged with Spn19A. (A) At 24 h BALF (7~9 mice/group) were analyzed for inflammatory cytokine levels (mean±SE) after X31 and/or Spn19A infection. (B) Lung bacterial burdens at 24 h, and (C) survival of α-TNF-α antibody-treated Il1r1^−/− mice following X31/Spn19A coinfection. In (A–B), *P< 0.05, **P< 0.01, ***P< 0.001, t test; in (C), ***, P< 0.001, log-rank test. Data shown in (A–B) are representative of two independent experiments. Data shown in (C) were combined from two independent experiments.

A compensatory role of neutrophils in protection against influenza and Spn19A coinfection

TNF-α is a known stimulus for neutrophil activities during pneumococcal infection (28). Considering that AMs are less important for lung clearance of Spn19A than Spn14 (Fig. 3B–C), we hypothesized that the survival of Il1r1^−/− mice from X31/Spn19A coinfection is due to the compensatory role of neutrophils in bacterial clearance. Compared with rat IgG-treated controls, anti-Gr-1 mAb treatment significantly reduced the relative number of CD11b⁺Ly6G⁺ neutrophils in WT and Il1r1^−/− mice during X31/Spn19A coinfection (Fig. 6A). Importantly, antibody-mediated neutrophil depletion significantly increased lung bacterial burdens (Fig. 6B) and thereby diminished the survival of Il1r1^−/− mice (Fig. 6C) after X31/Spn19A coinfection. These findings indicate that when AM numbers are reduced in Il1r1^−/− mice, neutrophils become essential for bacterial clearance and animal survival from X31/Spn19A coinfection.

Figure 6. Neutrophils are required to protect Il1r1^−/− but not WT mice against lethal X31/Spn19A coinfection.

C57BL/6 WT and Il1r1^−/− mice were infected with influenza X31 and seven days later super-challenged with Spn19A. (A) At 24 h BALF were analyzed for the relative numbers (mean±SD) of AMs, inflammatory monocytes, and neutrophils in α-PMN antibody-treated WT and Il1r1^−/− mice (5 mice/group). Control mice were treated with rat IgG. (B) Lung bacterial burdens (mean±SD, 5 mice/group) at 24 h, and (C) survival of α-PMN antibody-treated WT and Il1r1^−/− mice following X31/Spn19A coinfection. Control mice were treated with rat IgG. In (B), P<0.001, two-way ANOVA; **P< 0.01, ***P< 0.001, t test; in (C), ***, P< 0.001, log-rank test. Data shown in (A–B) are representative of two independent experiments. Data shown in (C) were combined from two independent experiments.

Of particular note, neutrophil depletion following X31/Spn19A coinfection led to a similar increase of bacterial burdens in WT and Il1r1^−/− lungs (Fig. 6B). Nonetheless, neutrophil-depleted WT mice still exhibited significantly improved bacterial control even compared with Il1r1^−/− animals received rat IgG treatment (Fig. 6B). As a result, all WT mice survived X31/Spn19A coinfection with or without neutrophil depletion (Fig. 6C). These data suggest that IL-1 signaling primarily enhances host resistance by facilitating PMN-independent clearance of Spn19A, as a further indication for its critical role in preventing AM depletion following coinfection.

IL-1 signaling prevents death of AMs and inflammatory cells during X31/Spn14 coinfection

In contrast to nonfatal X31/Spn19A coinfection, IL-1 signaling was required for animal survival from X31/Spn14 coinfection (Fig. 1C). The lethal X31/Spn14 coinfection in Il1r1^−/− mice was associated with 10-fold further increases in bacterial burdens compared with corresponding WT controls (Fig. 2B). In agreement with that, the relative numbers of AMs in Il1r1^−/− mice were significantly decreased compared with corresponding WT controls (Fig. 7A). Considering that AMs are essential for control of Spn14 infection (Fig. 3B), these results suggest that IL-1 signaling prevents AM depletion and thereby improves bacterial clearance and animal survival during X31/Spn14 coinfection.

Figure 7. IL-1 signaling prevents death of AMs and inflammatory cells during X31/Spn14 coinfection.

C57BL/6 WT and Il1r1^−/− mice were infected with influenza X31 and seven days later super-challenged with Spn14. (A) At 24 h BALF were analyzed for myeloid cell profiles. Numbers indicate the percentages (mean±SD) of 5 mice/group. (B–C) AMs (gated on CD11c⁺Siglec-F⁺) from naïve, 4 and 24 h coinfected WT and Il1r1^−/− mice were analyzed by flow cytometry after staining with Fixable Viability Dye (FVD), and with or without (-Annexin) Annexin V. The numbers in representative FACS plots indicate mean percentages (B), and data in (C) are presented as mean±SD (n=4). (D) Inflammatory monocytes and neutrophils from 4 and 24 h coinfected WT and Il1r1^−/− mice were analyzed by flow cytometry after staining with FVD, and with or without (-Annexin) Annexin V. The numbers in representative FACS plots indicate percentages (mean±SD, n=4). *P< 0.05, t test. Data shown are representative of at least two independent experiments.

In agreement with reduced AM numbers, Il1r1^−/− mice exhibited a significant increase in the percentage of late apoptotic/necrotic (Annexin^+/−FVD^hi) AMs at 24 h after Spn14 super-infection (Fig. 7B–C). The percentage of late apoptotic/necrotic inflammatory monocytes and neutrophils also significantly increased in coinfected Il1r1^−/− mice as compared with WT controls. However, active caspase-3 was barely detectable in these lung myeloid cells from WT or Il1r1^−/− mice (Supplemental Fig. S1), probably due to the efficient removal of apoptotic cells (i.e., efferocytosis process) in vivo.

BALF cells were next isolated to examine phagocytic activities of AMs during coinfection. Unlike S. aureus, S. pneumoniae was not easily visible inside AMs by Diff-Quick staining analysis, despite heightened lung bacterial loads during coinfection (Fig. 8A). In fact, only a small proportion of AMs were found to be associated with live bacteria (Fig. 8B). These results imply that the increased cell death in Il1r1^−/− mice is not a direct result of intracellular bacterial replication within AMs and inflammatory cells after X31/Spn14 coinfection.

Figure 8. Influenza infection promotes extracellular growth of Spn14 in the airway.

(A) Diff-Quick stained AMs, and (B) numbers of live bacteria associated with CD11c⁺ (AM), CD11b⁺Ly6C⁺ (MO), and CD11b⁺Ly6G⁺ (PMN) BALF cells sorted from WT and Il1r1^−/− airways at 24 h after X31/Spn14 coinfection. In (A), AMs isolated from S. aureus-infected WT mice were used as positive controls to show intracellular bacteria. In (B), the number of CFU was expressed per 10⁴ cells for each sorted myeloid cell subset, data shown were combined from three independent experiments. **P< 0.01, paired t test.

The protective role IL-1 signaling is Caspase-1-independent during influenza and pneumococcal coinfection

Our data above demonstrated that the susceptibility to secondary pneumococcal pneumonia significantly increases in the absence of IL-1 signaling. IL-1R1 is a receptor for IL-1 family cytokines including IL-1β (29). Caspase-1 is the protease responsible for the processing and activation of IL-1β during inflammasome activation (30). However, in contrast to Il1r1^−/− mice, influenza-infected caspase-1-deficient (Casp1^−/−) mice did not exhibit further increased susceptibility to influenza and pneumococcal coinfection compared with WT controls (Fig. 9). These results suggest that the protective role of IL-1 signaling during influenza and bacterial coinfection does not require caspase-1-dependent interleukin-1β secretion.

Figure 9. Caspase-1-deficient (Casp1^−/−) mice are not different from WT mice in susceptibility to influenza and pneumococcal coinfection.

Survival of C57BL/6 WT, Casp1^−/− and Il1r1^−/− mice after super-challenge with Spn14 on day seven after X31 (A) or PR8 (B) influenza infection. **, P< 0.01, log-rank test. Data shown are representative of at least two independent experiments.

Discussion

Using comparative mouse models of influenza and S. pneumoniae coinfection, here we evaluated the relative contribution of three central components to disease pathogenesis, namely, viral virulence, bacterial virulence, and host resistance. Our findings suggest that during host response to secondary pneumococcal infection, IL-1 receptor signaling prevents AM depletion, thereby reducing host susceptibility to both Spn14 and Spn19A super-infection. On the other hand, although neutrophils play a compensatory protective role in coinfections associated with Spn19A, the essential requirement for AMs for bacterial control significantly increases the virulence of Spn14 compared with Spn19A. Importantly, our results suggest that together, the combined requirement for AMs in bacterial control and for IL-1 signaling in preventing AM depletion leads to lethal X31/Spn14 coinfection in Il1r1^−/− mice.

In agreement, a mathematical model of influenza and pneumococcal coinfection predicts that bacterial outgrowth is primarily due to AM dysfunction (31). Furthermore, it has been shown that influenza infection of BALB/c mice depletes AMs and thereby leads to increased susceptibility to pneumococcal super-infection (24). Conversely, multiple previous studies have proposed suppression of neutrophil recruitment as the key pathogenic mechanism of influenza-induced susceptibility to secondary bacterial infection. In some studies, this suppression has been attributed to activation of the type I IFN signaling pathway during influenza infection (11, 25, 32), while others have shown its association with suppressed TLR sensing and proinflammatory cytokine responses (13, 26). Even so, it has been shown that neutrophil depletion did not increase bacterial burdens in mice challenged with pneumococci 6 days after influenza infection (33). In some settings, neutrophilic inflammation may even adversely affect clinical outcomes (34, 35).

Unlike previous investigations, we have developed mild to lethal influenza and pneumococcal coinfection models for comparative analyses of protective and pathogenic mechanisms in this study. In our models, neutrophils are not essential for acute bacterial clearance during pneumococcal infection alone. However, neutrophils do play a compensatory protective role in Spn19A-associated coinfections. For example, neutrophil depletion resulted in rapid death of WT animals after an otherwise nonfatal PR8/Spn19A coinfection (Supplemental Fig. S2). Likewise, neutrophils were required for bacterial clearance and survival of Il1r1^−/− mice following X31/Spn19A coinfection. Of particular note, during Spn19A or Spn14 infection alone, bacterial clearance occurred with minimal inflammatory cytokine and leukocyte responses, and was independent of IL-1 signaling. However, under severe coinfection conditions, bacterial outgrowth elicited an intense inflammatory response in the lung. This pattern of inflammatory response to pneumococcal infection, with or without prior influenza infection, is just the opposite to that of invasive bacterial strains used in prior studies (36–38).

Although the relationships among capsular structure, macrophage phagocytosis, and mouse virulence in S. pneumoniae still remain unclear, it has been shown in Klebsiella pneumoniae that expression of a capsule which is recognized by the mannose receptor markedly affects the interaction with macrophages and blood clearance (39). Given that the class A scavenger receptor macrophage receptor with collagenous structure (MARCO) and scavenger receptor SR-AI/II are primarily responsible for pneumococcal uptake by AMs (40, 41), it is likely the susceptibility of each pneumococcal serotype to AM-dependent killing is determined its ability to interact with these scavenger receptors. Nonetheless, our results indicate that both pneumococcal serotypes Spn14 than Spn19A are highly susceptible to AM-mediated clearance, even though AMs are more important for lung clearance of Spn14 than Spn19A.

During X31/Spn19A coinfection, we did not detect a significant effect of influenza infection on lung bacterial containment in WT mice. Conversely, Il1r1^−/− mice exhibited significant increases in bacterial load but no reduction in survival rate. The defective bacterial control in Il1r1^−/− mice was associated with significantly reduced AM numbers but increased proinflammatory cytokine response. It has been shown that during pneumococcal infection alone, there are no differences in neutrophil response in the absence of IL-1 or TNF-α alone, but triple-mutant mice deficient in TNFR1, TNFR2, and IL-1R1 are defective in neutrophil recruitment (18, 42). This compensatory interaction between IL-1 and TNF-α was also observed during protection against X31/Spn19A coinfection, as evidenced by heightened bacterial burden and rapid death of Il1r1^−/− mice after TNF-α neutralization or neutrophil depletion.

In addition, the reported findings in Il1r1^−/− mice suggest that during invasive pneumococcal infection alone, IL-1 occupies a role in the pulmonary immune defense that is less prominent than that of TNF-α (18, 43). However, it is clear from the diminished survival of Il1r1^−/− mice that elevated TNF-α production is incapable of compensating for the function of IL-1 during X31/Spn14 coinfection (Supplemental Fig. S3A). These observations support the concept that IL-1 may play a more important role in immune defense against post-influenza pneumococcal infection, by preventing early death of AMs and neutrophils. Importantly, although treatment with IL-1 receptor blocking agents such as anakinra is associated with reduced mortality in certain sepsis patients (44), our data shown here suggest that blockade of IL-1 signaling may not necessarily be beneficial in all clinical infections.

Caspase-1 is the protease responsible for the processing and activation of IL-1β during inflammasome activation (30). However, in contrast to Il1r1^−/− mice, influenza-infected Casp1^−/− mice did not exhibit further increased susceptibility to pneumococcal infection compared with WT controls. These results suggest that the protective role of IL-1R signaling does not require caspase-1-dependent interleukin-1β secretion. In agreement with our finding, it has been shown that IL-1α but not IL-1β is required for inducing proliferation of alveolar macrophages during granuloma formation (45). Nonetheless, although we show that IL-1R signaling prevents depletion of AMs that are essential for airway clearance of S. pneumoniae, future studies are necessary to exclude the possibility that IL-1R signaling in stromal cells, especially lung epithelium, also contributes to the protective role of IL-1 during influenza and S. pneumoniae coinfection.

Of note, the lethal X31/Spn14 coinfection in Il1r1^−/− mice was associated with significantly increased IFN-γ production compared with WT animals (Supplemental Fig. S3A). IFN-γ neutralization not only reduced lung bacterial loads but also increased relative numbers of AMs in Il1r1^−/− mice, suggesting a contradictory impact of IL-1 and IFN-γ on AM-mediated antibacterial immunity (Supplemental Fig. S3B–C). Considering our previous finding that IFN-γ impairs phagocytic uptake of pneumococci by AMs (12), we suggest that the increased IFN-γ production contributes to the lethal X31/Spn14 coinfection in Il1r1^−/− mice. Similarly, the lethal PR8/Spn14 coinfection in WT mice was associated with significantly further increased IFN-γ production and lung bacterial burden as compared with X31/Spn14 coinfection (Supplemental Fig. S4)

In summary, our data shown in this study indicate that influenza infection provides opportunities for otherwise nonvirulent S. pneumoniae to cause lethal pneumonia. The disease outcome is mutually determined by viral, bacterial and host factors. Importantly, our findings indicate that lethal influenza and pneumococcal coinfection is primarily due to inhibition of AM antibacterial activity; and through preventing AM depletion, inflammatory cytokine IL-1 signaling is critically involved in host resistance to influenza and pneumococcal coinfection.