Therapeutic Synergy Between Antibiotics and Pulmonary Toll-Like Receptor 5 Stimulation in Antibiotic-Sensitive or -Resistant Pneumonia

Abstract

Bacterial infections of the respiratory tract constitute a major cause of death worldwide. Given the constant rise in bacterial resistance to antibiotics, treatment failure is increasingly frequent. In this context, innovative therapeutic strategies are urgently needed. Stimulation of innate immune cells in the respiratory tract [via activation of Toll-like receptors (TLRs)] is an attractive approach for rapidly activating the body's immune defenses against a broad spectrum of microorganisms. Previous studies of the TLR5 agonist flagellin in animal models showed that standalone TLR stimulation does not result in the effective treatment of pneumococcal respiratory infection but does significantly improve the therapeutic outcome of concomitant antibiotic treatment. Here, we investigated the antibacterial interaction between antibiotic and intranasal flagellin in a mouse model of pneumococcal respiratory infection. Using various doses of orally administered amoxicillin or systemically administered cotrimoxazole, we found that the intranasal instillation of flagellin (a dose that promotes maximal lung pro-inflammatory responses) induces synergistic rather than additive antibacterial effects against antibiotic-susceptible pneumococcus. We next set up a model of infection with pneumococcus that is resistant to multiple antibiotics in the context of influenza superinfection. Remarkably, the combination of amoxicillin and flagellin effectively treated superinfection with the amoxicillin-resistant pneumococcus since the bacterial clearance was increased by more than 100-fold compared to standalone treatments. Our results also showed that, in response to flagellin, the lung tissue generated an innate immune response even though it had been damaged by the influenza virus and pneumococcal infections. In conclusion, we demonstrated that the selective boosting of lung innate immunity is a conceptually advantageous approach for improving the effectiveness of antibiotic treatment and fighting antibiotic-resistant bacteria.

Keywords: Streptococcus pneumoniae; Toll-like receptor 5; antibiotic; flagellin; pneumonia; resistance; superinfection.

Figure 1

The effect of the flagellin dose on the transcriptional response of immune system-related genes. BALB/c mice (n = 4 per group) were infected intranasally with 2 × 10⁶ Sp1 and treated 12 h later with the antibiotic amoxicillin (AMX; 5 μg, intragastric administration) combined with the intranasal administration of various doses of flagellin FliC_Δ174−400 (0.001, 0.1, 0.3, 1, 2.5, 10, and 25 μg in 30 μl of PBS) or vehicle only (PBS). Lungs were collected 2 h post-treatment, and RNA was extracted and reverse-transcribed. Gene expression was analyzed using quantitative PCR assays. The relative expression level for each gene is expressed against that of the reference genes Actb and B2m and the reference condition AMX+PBS (arbitrarily set to a value of 1). The data are quoted as the mean ± SEM.

Figure 2

Synergy between intranasal flagellin and antibiotics in the treatment of a pneumococcal lung infection. Swiss mice (n = 12–20) were infected intranasally with 4 × 10⁶ pneumococcus Sp1. The animals were treated 12 h later with the intragastric administration of amoxicillin (AMX; 5 or 40 μg) (A,B), the intraperitoneal injection of cotrimoxazole that is the combination of the two antibiotics sulfamethoxazole and trimethoprim (SXT; 1 or 4 mg) (C,D), and the intranasal administration of flagellin FliC_Δ174−400 (2.5 μg in 30 μl of PBS) or PBS only. Lungs were collected at 12 h post-treatment, homogenized, and plated with serial dilution onto blood agar plates. (A,C) Bacterial counts in the lungs of mice. Each symbol represents an individual animal. Colony-forming unit (CFU) counts for individual mice are shown. The solid line represents the median value, and the dashed line represents the detection threshold. Data from flagellin-treated mice were compared with those from PBS-treated mice in a Mann–Whitney test (*p < 0.05, **p < 0.01, and ***p < 0.001). (B,D) The treatments' effects on bacterial growth were quantified as the percentage of residual growth (% _growth) in treated mice (antibiotic + PBS or antibiotic + FliC_Δ174−400) vs. untreated mice (PBS). The predicted additive effect was calculated as % _{growth[antibiotic]} × % _{growth[flagellin]}. The values were plotted according to the dose of antibiotic.

Figure 3

A murine model of pneumonia due to antibiotic-resistant pneumococcus. (A–C) C57BL/6J mice (n = 5–8) were infected intranasally with 10⁶ or 10⁷ antibiotic-resistant pneumococcus Sp3 in 30 μl of PBS or with 50 PFUs of H3N2 virus in 30 μl of PBS followed 7 days later by intranasal administration of 10³ Sp3. (B) Body weight was monitored after Sp3 infection and expressed as a percentage of the initial weight. The data are quoted as the mean ± SEM. (C) Survival was monitored daily for 12 days. Data were compared in a log-rank test. ***p < 0.001. (D,E) C57BL/6J mice were infection intranasally with 50 PFUs of H3N2 virus in 30 μl of PBS followed 7 days later by intranasal administration of 10³ Sp3. (E) Bacterial counts in the lung and spleen of mice (n = 5). Tissues were collected at the indicated times post-Sp3 infection, and plated in serial dilutions on blood-agar plates. The values correspond to the median (range) CFU count. The dashed line represents the detection threshold.

Figure 4

Synergy between amoxicillin and intranasal administration of flagellin in the treatment of pneumonia with antibiotic-resistant pneumococcus. (A) C57BL/6J mice (n = 12–28) were infected intranasally first with 50 PFUs of H3N2 virus in 30 μl of PBS and then 7 days later with 10³ antibiotic-resistant pneumococcus Sp3 in 30 μl of PBS. Mice were treated 12 h after Sp3 infection via the intranasal administration of flagellin FliC_Δ174−400 (2.5 μg in 30 μl of PBS), the intragastric administration of amoxicillin (AMX; 100 or 350 μg), or combination of both. Lungs were collected 24 h post-infection, homogenized, and plated in serial dilutions onto blood agar plates to measure the bacterial load. For survival experiment, mice received a second dose of the same treatment at 36 h post-Sp3 infection. (B) Lung bacterial counts. Colony-forming unit (CFU) counts for individual mice are shown, and the solid line represents the median value. The dashed line represents the detection threshold. Data from flagellin-treated and control (PBS-treated) mice were compared in a Mann-Whitney test (**p < 0.01, and ***p < 0.001). (C) The treatments' effects on bacterial growth were quantified as the percentage of residual growth (%_growth) in treated mice (AMX+PBS or AMX+FliC_Δ174−400) vs. untreated mice (the PBS group). The predicted additive effect was calculated as %_growth[AMX] × %_{growth[flagellin]}. The values were plotted according to the dose of AMX. (D) Survival was monitored daily for 12 days. Data from the treated groups were compared with data from an untreated group in a log-rank test (**p < 0.01, and ***p < 0.001).

Figure 5

Lung innate immune response during flagellin treatment in post-flu superinfection with antibiotic-resistant pneumococcus. C57BL/6J mice (n = 4–6) were infected intranasally first with 50 PFUs of H3N2 virus in 30 μl of PBS and then 7 days later with 10³ antibiotic-resistant pneumococcus Sp3 in 30 μl of PBS. Mice were treated 12 h after Sp3 infection with the antibiotic amoxicillin (AMX; 100 μg, intragastric administration) and the intranasal administration of flagellin FliC_Δ174−400 (2.5 μg in 30 μl of PBS) or PBS only. (A) Lungs were collected 2 h after treatment, and homogenized. After RNA extraction, expression levels of selected genes were then analyzed using RT-qPCR assays. The relative expression level for each gene is expressed against that of the reference genes Actb and B2m and the reference condition AMX+PBS (arbitrarily set to a value of 1). The data represent the mean ± SEM. Lungs (B) and BAL fluids (C) were collected 6 h after treatment and cytokine and chemokine levels were measured by ELISA. Data from AMX+flagellin-treated and AMX+PBS-treated mice were compared in a Mann-Whitney test and are represented as individual values and mean. Lungs (D) and BALs (E) were collected 12 h after treatment. Lungs and BAL cell suspensions were stained using a mixture of antibodies specific for surface markers before flow cytometry analysis. Neutrophils were defined as CD45⁺CD11b⁺Ly6G⁺ cells after exclusion of dead cells and alveolar macrophages (CD45⁺SiglecF⁺CD11c⁺ cells) from the analysis. Numbers of neutrophils in the lung parenchyma (D) and BAL fluids (E) are shown for individual animal and the line represents the mean. Data from AMX+flagellin group were compared to those of AMX+PBS group in a Mann-Whitney test. Statistical significance is indicated as follows: *p < 0.05, and **p < 0.01.