Double-Edged Role of Interleukin 17A in Streptococcus pneumoniae Pathogenesis During Influenza Virus Coinfection

Abstract

Background

We sought to determine the role of host interleukin 17A (IL-17A) response against colonizing Streptococcus pneumoniae, and its transition to a pathogen during coinfection with an influenza virus, influenza A H1N1 A/Puerto Rico/8/1934 (PR8).

Method

Wild-type (WT) C57BL/6 mice were intranasally inoculated with S. pneumoniae serotype 6A to establish colonization and later infected with the influenza strain, PR8, resulting in invasive S. pneumoniae disease. The role of the IL-17A response in colonization and coinfection was investigated in WT, RoRγt^−/− and RAG1^−/− mice with antibody-mediated depletion of IL-17A (WT) and CD90 cells (RAG1^−/−).

Results

RAG1^−/− mice did not clear colonization and IL-17A neutralization impaired 6A clearance in WT mice. RoRγt^−/− mice also had reduced clearance. S. pneumoniae–PR8 coinfection elicited a robust IL-17A response in the nasopharynx; IL-17A neutralization reduced S. pneumoniae invasive disease. RoRγt^−/− mice also had reduced S. pneumoniae disease in a coinfection model. Depletion of CD90⁺ cells suppressed the IL-17A response and reduced S. pneumoniae invasion in RAG1^−/− mice.

Conclusion

Our data show that although IL-17A reduces S. pneumoniae colonization, coinfection with influenza virus elicits a robust innate IL-17A response that promotes inflammation and S. pneumoniae disease in the nasopharynx.

Influenza virus infection in S. pneumoniae colonized mice elicits a robust IL-17A response, leading to neutrophilia and tissue-damage in the nasopharynx. Suppression of IL-17A response reduces S. pneumoniae pathogenesis during an influenza co-infection.

Streptococcus pneumoniae causes a variety of infections in humans, including otitis media, sinusitis, pneumonia, and life-threatening invasive pneumococcal diseases, such as sepsis and meningitis [1–4]. Nasopharyngeal (NP) colonization is a precursor for S. pneumoniae disease, and coinfection with influenza virus is a significant risk factor for the development of S. pneumoniae disease [2, 5, 6]. Clinically, >50% of young children and a substantial proportion of adults are colonized by S. pneumoniae and other otopathogens in the NP, and there is a significant correlation between the development of S. pneumoniae diseases and viral upper respiratory tract coinfections [1, 2]. Influenza vaccination reduces S. pneumoniae diseases in humans, highlighting the impact of viral control on the containment of S. pneumoniae diseases [3]. However, the lack of an efficacious influenza vaccine necessitates the development of an alternative therapeutic approach to mitigate influenza-dependent S. pneumoniae diseases.

S. pneumoniae colonization is an immunizing event during which antigen-specific antibody and CD4⁺ T-cell responses develop [4–6]. In murine colonization models, the T-helper (Th) 17 response confers protection against primary S. pneumoniae colonization, and both Th17 and antibody responses protect against subsequent colonization episodes [7–9]. Th17 cells promote recruitment of neutrophils and enhance phagocytic function of resident myeloid cells in the NP, leading to the gradual clearance of S. pneumoniae colonization [10]. However, there is a lack of data on the role of the interleukin 17 (IL-17) response in the transition of S. pneumoniae colonization to disease or invasion during coinfection with influenza virus in the NP. Airway coinfection is characterized by complex interactions between coinfecting pathogens and the host, leading to hyperinflammation, tissue damage, and the disruption of physical barriers [11]. The IL-17 response is implicated in influenza-mediated airway tissue pathology [12, 13], but its effect on S. pneumoniae pathogenesis in the setting of S. pneumoniae colonization and acute inflammation caused by influenza is unknown.

Using a murine NP coinfection model, we investigated the role of influenza coinfection and IL-17 response in the transition from S. pneumoniae colonization to disease. In this model, influenza virus was introduced in S. pneumoniae–colonized mice. This resulted in transition of S. pneumoniae colonization to invasive disease, mimicking natural S. pneumoniae pathogenesis in the setting of influenza coinfection [11, 14–16]. Using interleukin 17A (IL-17A) neutralization (wild-type [WT]) and IL-17–deficient RoRγt^−/− mice, we determined the role of host IL-17 response in S. pneumoniae colonization and the development of S. pneumoniae disease during an influenza coinfection. We also developed a coinfection model in RAG1^−/− mice to determine the role of IL-17–producing innate cell/s in S. pneumoniae disease. Our data provide direct evidence for the involvement of the influenza-induced IL-17A response in the promotion of S. pneumoniae disease in the NP.

MATERIALS AND METHODS

Animals

WT C57BL/6, RAG1^−/−, and RoRγt^−/− mice were purchased from Jackson Laboratory. Mice were bred in house and used with the approval and in accordance with the guidelines of the University of North Dakota Animal Care and Use Committee. Equal proportions of age-matched (6–8-week-old) male and female mice were included in the study.

Strains and Reagents

S. pneumoniae serotype 6A strain BG7322 was provided by Rochester General Hospital Research Institute and was originally obtained from Sanofi Pasteur [17, 18]. Mouse-adapted influenza A H1N1 A/Puerto Rico/8/1934 (PR8) virus was purchased from Charles River, and a median tissue culture infective dose (TCID₅₀) assay was performed to determine the influenza virus infection inoculum for the experiments.

Mouse Infection Models

To develop S. pneumoniae colonization, 6–8 week-old C57BL/6 mice were anesthetized with isoflurane and intranasally inoculated with 1 × 10⁶ colony-forming units (CFUs) of S. pneumoniae in 10 µL of phosphate-buffered saline (PBS). At days 6 and 14 after colonization, the mice were euthanized, and NP lavage fluid was collected from the nostrils using 200 µL of PBS, as described elsewhere [19]. For coinfection, mice were intranasally inoculated with S. pneumoniae (1 × 10⁶ CFUs in 10 µL of PBS) and 24 hours later, with PR8 (100× TCID₅₀ in 10 µL of PBS). Six days after coinfection, mice were bled and euthanized with carbon dioxide_. NP lavage fluid and lungs were aseptically collected. The NP lavage fluid, homogenized lungs, and blood were serially diluted and plated on blood agar, and colonies were enumerated the next day. The total CFUs were then calculated based on the volume of lavage fluid, lung homogenate, and blood plated. In separate experiments, coinfected mice were observed for mortality rates.

Flow Cytometry

NP lavage fluid from S. pneumoniae, PR8 and coinfected mice was pooled, and stained to detect neutrophils (CD11b⁺ Ly6G⁺), as described elsewhere [15]. To determine the expression of CD44, NP-associated lymphoid tissues (NALTs) were aseptically extracted from mock-infected (PBS) and colonized mice, and a single cell suspension was prepared. NALT cells were stained with anti-CD3, anti-CD4, and anti-CD44 antibodies. To measure intracellular IL-17A, splenocytes from mock-infected and colonized mice were stimulated for 2 hours with heat-killed S. pneumoniae (6A) in 5% carbon dioxide. Brefeldin A was added at 10 μg/mL, and stimulated cells were further incubated for 4 hours (total stimulation 6 hours). Cells were stained with anti-CD3, anti-CD4 for surface staining. Subsequently, the cells were washed, treated with Cytofix/Cytoperm (BD Biosciences) according to the manufacturer’s instructions, and stained intracellularly for IL-17A, as described elsewhere [20]. An LSR II flow cytometer was used to acquire 100 000 cells, and the data were analyzed using FlowJo software (version 10) (Tree Star).

Cytokine and Chemokine Analysis

NP lavage fluid from mock-infected and infected mice was collected, pooled and preserved at −80°C until cytokine testing was performed. A multiplex cytometric bead array was performed to detect cytokines and chemokines in the NP lavage fluid, per the manufacturer’s instructions (LEGENDplex; Biolegend). The data are expressed as picograms per milliliter.

Lung Histology

Lung sections were prepared, processed and stained by the Histology Core, Department of Biomedical Sciences, University of North Dakota. Whole lungs were perfused and fixed with 10% formalin overnight. Tissues were subsequently transferred into ethanol until embedded in paraffin and sectioned. Each lung specimen was stained with hematoxylin-eosin (HE). Histological images were acquired using an Axiovert 200 inverted microscope (Zeiss).

IL-17A Neutralization (WT) and CD90 (Thy1.2) Cell Depletion (RAG1^−/−)

Purified anti–IL-17A antibody (eBiosciences) was used to neutralize IL-17A in colonized or coinfected mice. Mice were administered (intraperitoneally) 200 µg of anti-IL17A antibody every alternate day after establishment of colonization or coinfection. Control mice received an equal amount of isotype control antibody. RAG1^−/− mice were administered (intraperitoneally) 200 µg of anti-CD90 antibody every alternate day after establishment of coinfection. Control mice received an equal amount of isotype control antibody.

Quantitative Reverse-Transcription Polymerase Chain Reaction

Total RNA was extracted from NP lavage fluid collected with RLT lysis buffer (Qiagen) containing 0.001% 2-mercaptoethanol (Qiagen). Complementary DNA was synthesized using a SensiFast complementary DNA synthesis kit (Bioline), and quantitative reverse-transcription polymerase chain reaction was carried out using PowerUp SYBR green master mix (Life Technologies) on a Bio-Rad CFx384 real-time polymerase chain reaction system. The comparative cycle threshold (Ct) (2^−ΔΔCt) method was used to analyze changes in gene expression and normalized with housekeeping genes (RPL27 and actin). The primers used are listed in Supplementary Table 1.

Statistical Analysis

Statistical significance among multiple groups was determined with 1-way analysis of variance, followed by Tukey post hoc test for comparisons between 2 groups. Data involving the 2 groups were compared using the Mann-Whitney U test, and survival data were analyzed using the Mantel-Cox log-rank test. Differences were considered statistically significant at P < .05.

RESULTS

Adaptive Immunity and IL-17A Are Indispensable for Clearance of Primary S. pneumoniae Colonization in a Murine Model

Prior investigations have shown the protective role of IL-17A in murine S. pneumoniae colonization models [8, 10]. To determine the role of IL-17A in our colonization model, we used WT C57BL/6, RAG1^−/−, and RoRγt^−/− mouse strains. Antibody-mediated neutralization was performed to deplete IL-17A in WT colonized mice (Figure 1A). We used days 6 and 14 after colonization to capture the primary adaptive response; thus, days 6 and 14 represent innate (preadaptive) and adaptive response in our model. First, to determine the role of adaptive immunity, we colonized RAG1^−/− mice with S. pneumoniae serotype 6A. There was no difference in S. pneumoniae colonization density between days 6 and 14 (Figure 1B).

Figure 1.

A, Role of adaptive immunity and interleukin 17A (IL-17A) in Streptococcus pneumoniae colonization. C57BL/6 mice 6–8 weeks old (wild-type [WT]), RAG1^−/− and RoRγt^−/-) were intranasally inoculated with ST 6A. WT mice received isotype control (WT) or anti-IL-17A antibody WT (IL-17A⁻) on days 0, 2, and 4 after colonization. B, WT and RAG1^−/− mice were intranasally inoculated with 10 µL of phosphate-buffered saline (PBS) containing 1 × 10⁴ colony-forming units (CFUs) of ST 6A (with 1 × 10⁶ CFUs of ST 6A in 10 µL of PBS). C, WT and RoRγt^−/− mice were intranasally inoculated with 10 µL of PBS containing 1 × 10⁶ CFUs of ST 6A. IL-17A was neutralized in WT mice by injecting 200 µg of anti-IL-17A antibody. At 6 and 14 days after colonization, the mice were euthanized, and S. pneumoniae bacterial burdens in nasopharyngeal (NP) lavage fluid were determined. Data represent the results of 2 or 3 independent experiments (4–5 mice per group), are presented as medians, and were analyzed using 1-way analysis of variance, followed by Tukey test for pairwise comparison. *P < .01; †P < .001.

Next, we neutralized IL-17A in WT mice colonized with 1 × 10⁶ CFUs of S. pneumoniae serotype 6A and examined colonization in these mice and RoRγt^−/− mice, which are deficient in IL-17 responses. The outcome of colonization was unchanged after IL-17A neutralization on day 6 after colonization in all groups (WT, IL-17A–neutralized WT, and RoRγt^−/−) (Figure 1C). However, neutralization of IL-17A impaired bacterial clearance on day 14; IL-17A–neutralized WT mice had a significantly higher S. pneumoniae colonization density than WT mice that received an isotype antibody (Figure 1C). RoRγt^−/− mice also had a higher colonization density than WT mice, and no difference was observed between WT (IL-17A⁻) and RoRγt^−/− mice (Figure 1C). These data show that the preadaptive (<7 days) host response to colonization does not involve the IL-17 response. The effect of IL-17 induced clearance of colonization was observed only 14 days after colonization.

S. pneumoniae Colonization Develops a CD4⁺ T-Cell Activation Phenotype and an Antigen-Specific Th17 Response

We compared the CD4⁺ T-cell activation and the memory Th17 responses in WT mice on days 6 and 14 after colonization. NALTs and spleens were used to analyze CD4⁺ T-cell phenotype, and Th17 responses, respectively. CD4⁺ T-cell numbers were marginally higher on day 14 after colonization (Figure 2A and 2B), and CD4⁺ T-cell CD44 expression was significantly higher (Figure 2A and 2C). On day 14, splenocytes stimulated with S. pneumoniae ex vivo for IL-17A intracellular staining revealed a small frequency of IL-17A–expressing CD4⁺ T cells (Figure 2D). We did not detect CD4⁺ T-cell activation (NALTs) or IL-17A–expressing CD4⁺ T cells (splenocytes) in S. pneumoniae–colonized mice 6 days after colonization (data not shown). We also did not detect IL-17A protein in NP lavage fluid from colonized mice either 6 or 14 days after colonization (data not shown).

Figure 2.

Streptococcus pneumoniae colonization develops a T-helper 17 response. C57BL/6 mice 6–8 weeks old (wild-type [WT]) were intranasally inoculated with 1 × 10⁶ colony-forming units (CFUs) of ST 6A in 10 µL of phosphate-buffered saline. At 6 and 14 days after colonization, the mock-infected and colonized mice were euthanized, and single cells were prepared from their nasopharynx-associated lymphoid tissues (NALTs) and spleens. A, Representative flow plots of CD44 and CD62L expression in mock-infected and colonized mice gated through CD3⁺/CD4⁺ cells. B, Quantitative numbers of CD4⁺ T cells in NALTs of mock-infected and colonized mice. C, Quantitative numbers of CD44 and CD62L expressing CD4⁺ T cells in NALTs of mock-infected and colonized mice. D, Splenocytes from WT mock-infected and colonized mice were stimulated ex vivo for 6 hours with S. pneumoniae in the presence of brefeldin A, and interleukin 17A (IL-17A) was stained intracellularly. Data were analyzed using the Mann-Whitney test. *P < .05.

Influenza and S. pneumoniae–Influenza Coinfection Elicits a Robust IL-17A Response in the NP

We modified the previously described intranasal inocula of S. pneumoniae and PR8 influenza virus [15] to develop an S. pneumoniae–influenza coinfection model of invasive S. pneumoniae pathogenesis (Figure 3A). Intranasal inocula of serotype 6A (10⁶ CFUs) and PR8 (100× TCID₅₀) were required to cause bacteremia and 70%–80% mortality rates in a coinfection setting (Figure 3A). Neither S. pneumoniae nor PR8 caused death or disease in a monoinfection setting (Figure 3A). The introduction of PR8 in S. pneumoniae–colonized mice led to a significant increase (>1.5 log) in S. pneumoniae bacterial density in the NP, causing S. pneumoniae invasion (lung, bloodstream) and death (Figure 3A). Unlike with S. pneumoniae colonization, we detected robust expression of IL-17A protein in the NP lavage fluid from PR8 and coinfected mice (Figure 3B). Compared with S. pneumoniae–colonized mice, coinfected mice had higher expression of IL-17 receptor A (interleukin 17RA) on NP epithelial cells (Figure 3B). When compared with S. pneumoniae–colonized mice, we observed robust recruitment of neutrophils in PR8-infected mice, and a further increase in neutrophils in NP lavage fluid from coinfected mice (Figure 3C). The lungs of mice monoinfected with S. pneumoniae or PR8 mice showed mild to moderate bronchointerstitial infiltration of immune cells, and the lungs of coinfected mice showed a synergistic increase in immune cell infiltration and inflammation (HE staining) (Figure 3D).

Figure 3.

Expression of interleukin 17A (IL-17A) and neutrophil recruitment during Streptococcus pneumoniae–influenza coinfection. A, C57BL/6 mice 6–8-weeks old (wild-type) were intranasally inoculated with 10 µL of ST 6A. After 24 hours, they were intranasally inoculated with 10 µL of influenza A H1N1 A/Puerto Rico/8/1934 (PR8) (100× median tissue culture infective dose). Five days later, the mice were euthanized, and S. pneumoniae bacterial burdens (colony-forming units [CFUs]) in the nasopharynx (NP), lungs, and blood were ascertained. Data represent the results of 3 independent experiments (4–7 mice per group). Each dot represents a mouse, and error bars represent medians (bacterial burden). Data were analyzed using Mann-Whitney tests, and survival data compared using Mantel-Cox test. B, IL-17A levels in NP lavage fluid from mice colonized asymptomatically (S. pneumoniae) and PR8 infected or coinfected (S. pneumoniae + PR8). Data represent pooled NP lavage fluid from 2 independent experiments (5–7 mice per group). Data were analyzed using 1-way analysis of variance (ANOVA), followed by pair-based comparisons by Tukey test. Nasal mucosal cells were extracted from colonized and coinfected mice, and the expression of interleukin 17RA (IL-17RA) was determined by flow cytometry. Representative data are presented as histogram of IL-17RA⁺ after gating on CD45⁻ Epcam⁺ cells. Bar graph represents data from 2 independent experiments (3–5 mice per group); data were analyzed using Mann-Whitney test. C, Phenotyping of NP lavage fluid for neutrophils from mice colonized asymptomatically (S. pneumoniae) and PR8 infected or coinfected (S. pneumoniae + PR8). Representative flow data are expressed as density plots. Bar graph represents data from the pooled NP lavage fluid of 2 independent experiments (4–7 mice per group); data were analyzed using 1-way ANOVA, followed by pair-based comparisons by Tukey test. D, Representative hematoxylin-eosin staining of lung sections derived from mock-infected controls and mice infected with S. pneumoniae, PR8, or both (3–4 mice per group). *P < .05; †P < .01; ‡P < .001.

Neutralization of IL-17A Dampens NP Inflammatory Response and Mitigates S. pneumoniae Pathogenesis in a Coinfection Model

It is known that IL-17A promotes lung disease during influenza virus infection [12, 13]. Coinfected mice exhibited robust expression of IL-17A in NP lavage fluid (Figure 3B). To determine the role of IL-17A in our coinfection model of S. pneumoniae pathogenesis, we neutralized IL-17A (Figure 4A) and used IL-17–deficient RoRγt^−/− mice. Compared with coinfected WT mice (isotype control antibody), IL-17A–neutralized WT mice had significantly lower NP, lung, and blood S. pneumoniae CFU counts (Figure 4B). Coinfected IL-17A–neutralized WT mice also had higher survival rates than coinfected control WT mice (Figure 4C). Coinfected RoRγt^−/− mice also had significantly lower NP, lung, and blood CFU counts and higher survival rates than WT mice (Figure 4B and 4C). There was no difference in the NP, lung, and blood CFU counts and survival rates between IL-17A–neutralized WT and RoRγt^−/− coinfected mice (Figure 4B and 4C). The lungs of coinfected WT mice showed more severe infiltration of immune cells and inflammation than IL-17A–neutralized (WT) and RoRγt^−/− mice (HE staining) (Figure 4D).

Figure 4.

Interleukin 17 (IL-17) blockade and Streptococcus pneumoniae pathogenesis in an S. pneumoniae–influenza coinfection model. a. C57BL/6 mice 6–8 weeks old (wild-type [WT] and RoRγt^−/−) were intranasally inoculated with 10 µL of ST 6A. Twenty-four hours later, they were inoculated intranasally (10 µL) with influenza A H1N1 A/Puerto Rico/8/1934 (PR8) virus (100× median tissue culture infective dose). Interleukin 17A (IL-17A) was neutralized in WT mice (WT IL-17A⁻) by injecting 200 µg of anti-IL17A antibody. Control coinfected WT mice received an equal amount of isotype control antibody. Five days after coinfection, the mice were euthanized, and S. pneumoniae bacterial burdens in the nasopharynx (NP), lungs, and blood were ascertained. B, S. pneumoniae bacterial burden in NP lavage fluid, lungs, and blood of WT, WT (IL-17A⁻), and RoRγt^−/− coinfected mice. Data represent the results of 3 independent experiments (3–6 mice per group). Each dot represents a mouse, and error bars represent medians (bacterial burden). Data were analyzed using 1-way analysis of variance, followed by Tukey test for pairwise comparison. C, Survival of coinfected WT, WT (IL-17A⁻), and RoRγt^−/− mice. Survival data were compared using the Mantel-Cox test. D, Hematoxylin-eosin representative staining of lung sections derived from coinfected WT, WT (IL-17A⁻), and RoRγt^−/− mice (3–4 mice per group). *P < .05; †P < .01; ‡P < .001.

Hyperinflammation and tissue damage promote S. pneumoniae dissemination from the NP to lung and blood tissues, which is manifested as S. pneumoniae disease [21, 22]. Because the IL-17A–neutralized mice had reduced S. pneumoniae invasiveness and lethality in our coinfection model, we hypothesized that PR8-elicited IL-17A contributed to inflammation and disease in the NP, leading to S. pneumoniae dissemination and invasion. IL-17A–neutralized WT and RoRγt^−/− mice had significantly lower levels of proinflammatory cytokine interleukin 1α and 6, tumor necrosis factor α, and IL-17A (Figure 5A) and total protein (Figure 5B) in NP lavage fluid. Compared with WT coinfected mice, IL-17A–neutralized (WT) and RoRγt^−/− mice also had reduced levels of neutrophil recruitment factor KC and neutrophils in NP lavage fluid (Figure 5A, 5C). Epithelial tight junction (TJ) proteins constitute an essential structural component of barrier integrity leading to the prevention of bacterial invasion to the sterile tissues [23]. Therefore, we determined the expression of TJ proteins ZO-1 and occludin in IL-17–proficient and L-17–deficient conditions. IL-17A–neutralized (WT) and RoRγt^−/− coinfected mice had an increased expression of epithelial TJ proteins, ZO-1 and occludin (Figure 5D). Lower NP inflammation and improved epithelial barrier response correlated with reduced S. pneumoniae invasiveness and higher survival rates in WT IL-17A⁻ and RoRγt^−/− coinfected mice.

Figure 5.

Interleukin 17 (IL-17) blockade and Streptococcus pneumoniae pathogenesis in an S. pneumoniae–influenza coinfection model. A, C57BL/6 mice 6–8 weeks old (wild-type [WT] and RoRγt^−/−) were intranasally inoculated (10 µL) with ST 6A (1 × 10⁶ colony-forming units [CFUs]). Twenty-four hours later, they were inoculated intranasally (10 µL) with influenza A H1N1 A/Puerto Rico/8/1934 (PR8) virus (100× median tissue culture infective dose). Interleukin 17A (IL-17A) was neutralized in WT mice (WT IL-17A⁻) by injecting 200 µg of anti-IL-17A antibody. Control coinfected WT mice received an equal amount of isotype control antibody. Five days after coinfection, the mice were euthanized, and the first retrograde nasopharyngeal (NP) lavage fluid sample was collected (with phosphate-buffered saline). Cytokine levels in NP lavage fluid from coinfected mice (WT, WT IL-17A⁻, RoRγt^−/−) were determined by means of cytometric multiplex bead array. Data represent pooled NP lavage fluid from 2 independent experiments (5 per group). B, Total protein in pooled NP lavage fluid from WT, WT IL-17A⁻ and RoRγt^−/− coinfected mice. C, Neutrophil levels in NP lavage fluid from coinfected mice (WT, WT IL-17A⁻, and RoRγt^−/−). Data are expressed as representative contour plots and bar graphs with the percentage of neutrophils. Bar graphs represent data from pooled NP lavage fluid of 2 independent experiments (4–7 mice per group). D, The second retrograde NP lavage fluid sample from coinfected mice was collected in cell lysing RLT buffer (Qiagen). The messenger RNA (mRNA) levels of epithelial tight junction proteins ZO-1 and occludin were determined with quantitative reverse-transcription polymerase chain reaction. Data were analyzed using 1-way analysis of variance, followed by pair-based comparisons with the Tukey test. *P < .05; †P < .01; ‡P < .001. Abbreviations: IL-1α, interleukin 1α; IL-6, interleukin 6; TNF-α, tumor necrosis factor α.

Depletion of CD90⁺ (Thy1.2) Cells Reduced S. pneumoniae Pathogenesis in RAG1^−/− Coinfection Model

Because the coinfection window is just 5 days in our model, we determined whether the source of the IL-17 response was innate immune cell/s in lymphocyte-deficient RAG1^−/− mice with or without CD90⁺ cell depletion (Figure 6A). CD90 is expressed on the IL-17–producing innate cells, such as innate lymphoid cells and a subset of natural killer cells [24, 25]. Consistent with the results in WT mice, IL-17A expression was detected in NP lavage fluid from RAG1^−/− mice coinfected with S. pneumoniae and PR8 influenza. CD90⁺ cell depletion resulted in lower NP, lung, and blood CFU counts compared with CD90⁺ cell–sufficient mice (Figure 6B). Coinfected CD90 depleted RAG1^−/− mice also had higher survival than isotype control–treated RAG1^−/− coinfected mice (Figure 6B). Depletion of CD90⁺ cells led to reduced NP levels of IL-17A, KC (Figure 6C), and neutrophils (Figure 6D), which was correlated with reduced S. pneumoniae dissemination.

Figure 6.

CD90 depletion and coinfection in RAG1^−/− mice. A, RAG1^−/− mice 6–8 weeks old were intranasally inoculated (10 µL) with ST 6A (1 × 10⁴ colony-forming units [CFUs]). Twenty-four hours later, they were inoculated intranasally (10 µL) with influenza A H1N1 A/Puerto Rico/8/1934 (PR8) virus (100× median tissue culture infective dose). CD90 cells were depleted by injecting (intraperitoneally) 200 µg of anti-CD90.2 (Thy1.2) antibody every alternate day after establishment of coinfection. Control coinfected RAG1^−/− mice received an equal amount of isotype control antibody. Four days after coinfection, the mice were bled and euthanized, and nasopharyngeal (NP) lavage fluid (phosphate-buffered saline) and lungs were collected. B, Streptococcus pneumoniae bacterial burden in NP lavage fluid, lungs, and blood of RAG1^−/− (isotype vs CD90-depleted) mice. Data represent the results from 2 independent experiments (5–6 mice per group), are presented as medians (bacterial burden), and were analyzed using the Mann-Whitney test. Survival data in RAG1^−/− coinfected (isotype vs CD90-depleted) mice were compared using the Mantel-Cox test. C, Levels of interleukin 17A (IL-17A) and keratinocyte chemoattractant in NP lavage fluid from RAG1^−/− coinfected (isotype vs CD90-depleted) mice. Data were analyzed using the Mann-Whitney test. D, Levels of neutrophils in the NP lavage fluid from RAG1^−/− coinfected (isotype vs CD90 depleted) mice. Data are expressed as representative contour plots flow panels. Bar graphs represent data from the pooled NP lavage fluid from 2 independent experiments (5–6 mice per group), and data were analyzed using the Mann-Whitney test. *P < .05; †P < .01; ‡P < .001.

DISCUSSION

We investigated the effect of influenza coinfection and host IL-17 response on S. pneumoniae colonization. Our data show that S. pneumoniae–colonized mice that are infected with influenza exhibit more NP inflammation, bacterial dissemination, and reduced survival. Because a significant proportion of humans are colonized by S. pneumoniae, our model mimics events that are likely to occur in humans who acquire influenza infection. Using S. pneumoniae colonization models in WT, RAG1^−/−, and RoRγt^−/− mice, we show that early S. pneumoniae colonization is maintained without inflammation and the absence of IL-17 expression in the NP. A Th17 response then leads to gradual clearance of S. pneumoniae colonization in the NP. However, influenza coinfection in S. pneumoniae–colonized mice elicited a robust IL-17A response in the NP that was associated with increased S. pneumoniae growth and invasion. Blockade of IL-17 response dampened NP inflammation and neutrophilia, leading to reduced S. pneumoniae invasiveness and a reduced mortality rate in coinfected mice. The data for RoRγt^−/− coinfected mice mirrored with that of the IL-17A neutralized WT mice. In addition, data from RAG1^−/− mice showed that IL-17A is derived from CD90⁺ expressing innate cells in our model.

Prior reports have shown that IL-17 response reduces S. pneumoniae colonization in murine colonization models [8, 10, 26]. However, the role that the innate IL-17 response plays in colonization has not been shown previously. Because a number of immune cells can produce IL-17A [27–29], we determined the role of the innate and adaptive IL-17 response in our models. We used a different set of mouse strains than used in prior investigations [8, 10]. We obtained data on days 6 and 14 after colonization to examine the effect of innate (preadaptive) and adaptive immune responses. Colonization was not cleared in RAG1^−/− mice, and there was no difference in colonization densities between days 6 and 14. WT and RAG1^−/− mice had similar colonization densities on day 6, but a significant difference was observed between the 2 groups on day 14. These data show that adaptive immunity is indispensable for inducing clearance of primary S. pneumoniae colonization in mice.

To determine the role of the innate and adaptive IL-17 (Th17) response in colonization, we neutralized IL-17A in WT mice. Deficiency of IL-17A response (IL-17A⁻ WT and RoRγt^−/−) did not affect colonization on day 6 (preadaptive phase); however, IL-17–deficient mice (IL-17A⁻ WT and RoRγt^−/−) had significantly higher bacteria in the NP than WT mice on day 14. IL-17A–mediated clearance of colonization on day 14 was correlated with higher levels of memory CD4⁺ T cells (CD4⁺ CD44⁺) in NALTs and the detection of S. pneumoniae–specific Th17 cells in the spleen, though we did not detect IL-17A protein in NP lavage fluid. We did not detect S. pneumoniae–specific Th17 cells or higher CD44 expression in the mice colonized for 6 days (data not shown). Thus, our data show IL-17A does not participate in S. pneumoniae clearance during the first week of colonization, whereas antigen-specific Th17 cells induce clearance between 6 and 14 days after colonization.

IL-17 receptor signaling has shown to contribute to influenza-mediated pathology in the lung [12, 13, 30, 31]. The role of influenza in the IL-17 response has not been studied in the context of S. pneumoniae infections in the NP. Unlike in S. pneumoniae colonization, the IL-17A response was robust at the protein level in NP lavage fluid from influenza or coinfected mice. The IL-17A level in coinfected mice was associated with higher neutrophil counts in the NP and S. pneumoniae invasiveness. Neutralization of IL-17A in coinfected mice led to a >1-log reduction of S. pneumoniae bacteria in the NP, with reduced invasiveness and improved survival rate.

Coinfected RoRγt^−/− mice also had reduced invasiveness and a higher survival rate, mirroring results in IL-17A–neutralized WT mice. In addition, coinfected RoRγt^−/− and IL-17A–neutralized (WT) mice had lower pulmonary inflammation seen with HE staining than WT coinfected mice. Hyperinflammation and tissue damage in the NP promote S. pneumoniae bacterial growth, transmission, and invasion to sterile tissues [14, 21, 22, 32, 33]. In our coinfection model, neutralization of IL-17A led to suppression of overall inflammatory response in the NP, characterized by the reduced levels of key proinflammatory cytokines, such as interleukin 1α, interleukin 6, tumor necrosis factor α, and IL-17A, and neutrophils. Epithelial TJ proteins are central components of airway barrier function, leading to the prevention of S. pneumoniae bacterial invasion. Reduced S. pneumoniae disease in WT IL-17A⁻ and RoRγt^−/− mice were correlated with increased expression of TJ proteins, ZO-1 and occludin, in NP epithelial cells. Our findings are consistent with prior reports that hyperregulated and dysregulated inflammation leads to damage, bacterial outgrowth, and invasion [34–38].

In the Staphylococcus aureus acute pneumonia model, CD4⁺ T cells contribute to lung disease, and mice lacking T-cell populations exhibit improved outcomes in the lung pathology and clearance of acute lung infection [39]. Kudva et al [40] showed that influenza virus inhibits Th17-mediated host defense against S. aureus bacterial pneumonia in mice. Our coinfection model is different from those explained above, because we used low-volume intranasal inocula to establish coinfection in the NP, resulting in bacterial dissemination/invasion to the lungs and bloodstream. In this context, to the best of our knowledge, our data are the first to show the pathological role of IL-17A as a contributor of influenza acute inflammation, leading to initiation of S. pneumoniae disease in the NP.

Because the coinfection window in our model is only 5 days, we posited that the IL-17A response was derived from innate immune cells. To identify the innate cell type that was the source of IL-17A, we used lymphocyte-deficient RAG1^−/− mice and depleted their CD90⁺ cells. In RAG1^−/− mice, CD90 is primarily expressed by innate lymphoid cells and a subset of natural killer cells [41, 42]. Depletion of CD90⁺ cells led to a significant reduction in IL-17A and neutrophil levels in the NP of RAG1^−/− coinfected mice, which correlated with reduced S. pneumoniae invasiveness and improved survival. These data show that in both lymphocyte-sufficient and lymphocyte-deficient mice, IL-17A is a significant determinant of S. pneumoniae invasiveness during influenza coinfection, and suppression of IL-17A response rescued mice from severe S. pneumoniae pathogenesis during influenza coinfection (Figure 7). In addition, our data indicate that CD90⁺ cells represent the primary source of IL-17A in RAG1^−/− mice.

Figure 7.

Schematic representation of interleukin 17 (IL-17)–mediated inflammatory response during Streptococcus pneumoniae–influenza coinfection. IL-17 deficiency reduces neutrophilia and inflammatory burden in the nasopharynx (NP), leading to reduced S. pneumoniae invasiveness and disease. Abbreviations: IL-17A and IL-17RA, interleukin 17A and 17RA.

There is evidence that both S. pneumoniae and influenza vaccinations reduce S. pneumoniae diseases in vaccinated individuals [3]. However, replacement of capsular serotypes has occurred [43, 44], thus requiring expansion of the valency of current S. pneumoniae vaccine formulations. Therefore, despite continued vaccination programs, S. pneumoniae infections still occur. Similarly, the lack of an efficacious influenza vaccine, along with the failure to implement influenza vaccination universally, continues to pose challenges vis-à-vis S. pneumoniae diseases promoted by seasonal influenza infections [45]. Therefore, there is a need to identify inflammatory targets that could be used to modulate pulmonary inflammation and thus prevent airway S. pneumoniae diseases. Our findings provide direct evidence of the double-edged role of IL-17A as a bridge between a protective immune response against S. pneumoniae colonization and as a contributor to hyperinflammation and S. pneumoniae disease during an influenza coinfection. However, we recognize that more work is needed to determine whether our findings apply to other S. pneumoniae serotypes and influenza strains and/or whether control of the IL-17A response may hold therapeutic promise. These questions are under investigation in our laboratory.

SUPPLEMENTARY DATA

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. All authors have read and approved the final version of the manuscript.

Financial support. This work was supported by the National Institutes of Health (NIH) (grant P20GM113123 to M. N. K. and NIH Center of Biomedical Research Excellence (Host-Pathogen) grant 5P20GM113123 and IDeA Network of Biomedical research Excellence grant 5P20GM103442 to the Flow Cytometry core facility, University of North Dakota).

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Presented in part: American Association of Immunologist meeting, Austin, Texas, May 4–8, 2018.