ABSTRACT

INTRODUCTION

Pseudomonas aeruginosa (PA) is one of the leading causes of hospital-acquired infections globally and the most common pathogen associated with respiratory chronic illnesses, such as cystic fibrosis (CF) (1,–3). Overuse of antimicrobial agents during the past decades has resulted in the emergence of PA-resistant strains, particularly to β-lactams, carbapenems, and aminoglycosides antibiotics (4). Multidrug-resistant (MDR) strains of PA are even harder to eradicate and cause higher mortality rates (5, 6). As a consequence, the World Health Organization (WHO) has included MDR strains of PA in a list of pathogens requiring urgent attention for the development of new therapeutic strategies (7).

Approaches that can enhance and rescue antibiotic efficacy are of significant interest and are increasingly being studied (8). In this context, host-directed molecules in combination with antibiotics are a promising strategy for preventing infections and restoring antibiotic effectiveness. Flagellin, a structural component of the flagellum of Gram-negative bacteria, has emerged as a promising host-directed immunomodulatory drug (9). In particular, the recombinant form of flagellin (flagellin_174-400), which lacks its antigenic part, has been described for its immuno-stimulatory capacity and antimicrobial properties against intestinal infections caused by Enterococcus faecium, Clostridium difficile, and Escherichia coli (10,–12) and against respiratory infections caused by Streptococcus pneumoniae (13).

In the airways, the mechanism of action of flagellin_174-400 (hereafter only designated as flagellin) is well-described. Flagellin binds to toll-like receptor 5 (TLR5), a pattern recognition receptor (PRR) present in epithelial cells (ECs), alveolar macrophages (AM), conventional dendritic cells (cDCs), and lymphocytes (14). This interaction strongly activates the transcription factor nuclear factor-kappa B (NFκB), mitogen-activated protein kinases (MAPK), and interferon (IFN)-regulatory factor (IRF) pathways, resulting in the transcription of genes encoding several immune mediators (15). This includes chemokines such as CXCL1, CXCL2, CXCL5, CCL2, and CCL20; cytokines such as interleukin-6 (IL-6), granulocyte-macrophage colony-stimulating factor (GM-CSF), and interleukin-1 beta (IL-1β) as well as mucins; and several antimicrobial peptides such as β-defensins (16,–19).

It has been demonstrated that flagellin triggers a short-lived response (peaking 2 h after administration and decaying 2–3 h after the peak) due to strong feedback regulatory mechanisms (20). This controlled response is crucial because it allows the inflammation to resolve toward homeostasis. Interestingly, the same study also shows that a second stimulus with flagellin (either with soluble protein or PA infection) leads to a less intense proinflammatory state, suggesting that a single prophylactic flagellin instillation can reduce the inflammatory response to PA (20).

Previous studies have also shown that nasal administration of flagellin has a synergistic effect with the antibiotic amoxicillin in controlling antibiotic-resistant pneumococcal infections (13). Other data have suggested that prophylactic nasal administration of flagellin can decrease the bacterial load of pigs infected with the P. aeruginosa (PAK) strain (20). However, flagellin has never been tested in combination with an antibiotic to fight PA and was never evaluated against a MDR strain of PA.

Here, using an in vivo mouse model, we show that although prophylactic flagellin alone attenuates infection caused by a MDR strain of PA (PA^MDR), its combination with the antibiotic gentamicin (GNT) leads to a strong decrease in bacterial load in the lung and a significant reduction in cell infiltration and inflammatory cytokines. Moreover, mice receiving flagellin in combination with GNT showed a 100% survival rate, a result that GNT alone was unable to provide against PA^MDR.

MATERIALS AND METHODS

Animal care and handling

C57Bl/6 female mice were purchased from the Centre d'Elevage R. Janvier (Le Genest Saint-Isle) and were used at about 8 weeks of age. All procedures were in accordance with the European animal welfare regulations. All mice were housed under specific pathogen-free conditions at the PST “Animaleries” animal facility (Université de Tours) and had access to food and water ad libitum. All experiments complied with the French government’s ethical and animal experiment regulations (APAFIS#201604071220401 .V2-4885 and 2016111512369894 V3 - 7590).

Mice were treated with 10 µg of recombinant flagellin (FLAMOD) 24 h before infection with 5 × 10⁵ or 8 × 10⁵ CFU/mouse of PA^MDR by the nasal route (Fig. 1A). One hour post-infection, mice were treated by intraperitoneal injection (i.p.) with 15 mg/kg of gentamicin purchased from Sigma Aldrich (Sant-Quentin Fallavier). The choice of gentamicin dose was based on previous experiments in vivo, showing that a single dose of 15 mg/kg i.p. was the minimum dose required to decrease PA^MDR bacterial load (in mice infected with 10⁵ CFU/mouse). Sixteen hours after challenge with PA^MDR, mice were sacrificed, and airways were washed four times with 0.5 mL of phosphate buffered saline (PBS) for bronchioalveolar lavage (BAL) collection. After centrifugation at 400 × g for 5 min, BAL fluids were stored at −80°C for subsequent measurement of inflammatory mediators, and pellets were recovered in PBS 2% fetal bovine serum (FBS). The pellet was resuspended and filtered, and erythrocytes were discarded using a red blood cell lysis buffer. Leukocytes were counted and analyzed by flow cytometry. After BAL recovery, the lungs were perfused with 10 mL PBS injected into the heart. The lungs were photographed for inflammation assessment based on lung morphology and color. The right lungs were homogenized with 1 mL of PBS using the Gentle MACSTM Octo Dissociator (Miltenyi Biotech), and 100 µL of lung suspension was plated for bacterial count.

Open in a separate window

Fig 1

Timeline for the treatment and infection of mice and bacterial load 16 h post-infection. (A) Tweenty-four hours before infection mice in group “untreated” (n = 7) and “GNT” (n = 6) received NaPi buffer and mice in groups “flagellin” (n = 8) and “flagellin + GNT” (n = 9) received FliC (10 µg), resuspended in NaPi buffer, by the nasal route. At T0, all animals (with the exception of controls in “healthy” group) were intranasally infected with 5 × 10⁵ CFU of a MDR strain of P. aeruginosa. One hour after infection, mice in group GNT and flagellin + GNT received GNT 15 mg/Kg by intraperitoneal injection. Healthy mice were used as control (group “healthy”). All mice were sacrificed 16 h p.i.. (B) Colony forming units (CFU)/lung lobe were counted in all infected mice. Each symbol represents an individual animal in two independent experiments. Differences between groups were analyzed with Mann-Whitney test. Data are reported as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

RESULTS

Preventive intranasal administration of flagellin decreases proinflammatory cytokines and cell infiltration but increases neutrophils and dendritic cells in the BAL of PA^MDR-infected mice

Using the same experimental design as described above, we examined the inflammatory status of mice. We analyzed the macroscopic aspect of their lungs (Fig. 2A), immune cell populations recruitment (Fig. 2B), and the secretion of proinflammatory cytokines in BAL (Fig. 2C), at 16 h post-infection, for all experimental groups.

Open in a separate window

Fig 2

Macroscopic aspect of lungs, immune cell population recruitment, and the secretion of proinflammatory cytokines in BAL. Tweenty-four hours before infection with 5 × 10⁵ CFU of a MDR P. aeruginosa strain, mice were treated or not with flagellin (10 µg) intranasally. Then, 1 h post-infection, mice were treated intraperitoneally or not with Gentamicin (GNT, at 15 mg/Kg body weight). All mice were sacrificed 16 h p.i.. Lungs were perfused for visual assessment of inflammation (A), and BAL fluids were collected to determine the number of immune and inflammatory cells by flow cytometry (B), and the levels of GM-CSF, TNF-α, IL-1β, and IL-6 (C). All data are represented as the mean ± SEM and are cumulative of two independent experiments. The number of mice per group is indicated by symbols [gray (n = 2), purple (n = 7), green (n = 6), red (n = 8), and orange (n = 8)]. Statistical analysis was performed using the Mann-Whitney test. *P < 0.05, **P < 0.01, ***P < 0.001, .****P < 0.0001

Figure 2A shows the difference between representative lungs of a mouse from each group. As expected, untreated mice infected with PA^MDR showed a higher extent of gross pathology and hemorrhage. Interestingly, mice only treated with GNT experienced similar macroscopic lung pathology than did the untreated mice. Although animals pre-treated with flagellin before infection exhibited a clear reduction in hemorrhage, the reduction was even more reduced in animals that received the dual therapy of flagellin and GNT.

The total number of cells and leukocytes (CD45+) in the BAL (Fig. 2B.1 and 2B.2, respectively) was analyzed by flow cytometry using the gating strategy described in Fig. S2. We observed that animals treated with flagellin, with or without GNT, presented significantly fewer cells in the BAL at 16 h post-infection, compared with infected non-treated or those only treated with GNT (P < 0.05 for group flagellin vs group untreated and P < 0.001 for group flagellin + GNT vs group GNT). Interestingly, among the cells present in the BAL, significantly more were CD45 +immune cells for the animals that received preventive flagellin (P < 0.01 for group flagellin vs group untreated and P < 0.001 for group flagellin + GNT vs group GNT).

According to the study of Fougeron et al. (23), the number of neutrophils and total DCs in the lungs of mice remain elevated for approximately 72 h after being challenged nasally with flagellin. In this context, we investigated both immune cell populations (neutrophils identified as CD45⁺CD11b⁺Ly6G⁺ and total DCs as CD45⁺CD11c⁺SiglecF^-MHCII⁺) in the BAL (Fig. 2B.3 and 2B.4, respectively). As expected, P. aeruginosa infection induced neutrophil recruitment to the airways. The mucosal delivery of flagellin (in combination or not with GNT) induced an even higher increase in the number of neutrophils (P < 0.01 for group flagellin vs group untreated and P < 0.001 for group flagellin + GNT vs group GNT). Flagellin instillation also caused an increase in total DCs (P < 0.01 for group flagellin vs group untreated and P < 0.001 for group flagellin + GNT vs group GNT). Regarding AM (identified as CD45⁺CD11c⁺SiglecF⁺; Fig. 2B.5), infection with PA^MDR decreased the number of this cell population in BAL (a 5-fold reduction compared with the non-infected group) and treatment with the antibiotic GNT was not enough to revert such drop. Pre-treatment with flagellin (in combination or not with GNT) preserved the level of AM (P < 0.01 for group flagellin vs group untreated and P < 0.001 for group flagellin + GNT vs group GNT).

Sixteen hours after challenge with PA^MDR, infected non-treated animals presented high levels of IL-6, tumor necrosis factor alpha (TNF-α), GM-CSF, and IL-1β 16 h post-infection (Fig. 2C). Although GNT-treated mice also presented high proinflammatory cytokines, mice that received flagellin (in combination or not with GNT) showed a significant reduction for all tested cytokines (P < 0.01 or P < 0.001). Thus, flagellin pretreatment reduced the expression of proinflammatory cytokines triggered by PA^MDR infection. The combination of flagellin and GNT fully restores the level of pro-inflammatory mediators released from the lungs of PA^MDR -infected mice to a homeostatic basal level.

Combination of prophylactic flagellin and therapeutic gentamicin protects mice against a lethal dose of a MDR strain of P. aeruginosa

We evaluated the effectiveness of the flagellin + GNT combination treatment in protecting mice against a lethal PA^MDR infection. To this end, we monitored mice survival over 60 h post-bacterial infection (Fig. 3A). As anticipated for a MDR strain of PA, most infected non-treated mice succumbed to the infection within 24 h (Fig. 3B). Similarly, the mice only treated with GNT showed similar susceptibility to bacterial infection. However, prophylactic flagellin treatment alone was sufficient to significantly protect the mice against the PA infection (80% survival for group “flagellin” vs 20% survival for group “GNT”, P < 0.001). Moreover, the combination of prophylactic flagellin with therapeutic GNT resulted in full protection compared with non-treated animals (Fig. 3B).

Open in a separate window

Fig 3

Timeline for treatment and infection of mice and survival study. (A) Tweenty-four hours before infection, C57Bl/6 mice received flagellin (10 µg) [red (n = 5) and orange (n = 7) groups] resuspended in NaPi buffer or only NaPi buffer [purple (n = 7) and green (n = 7) groups] by the nasal route. At T0, all animals were intranasally infected with 8 × 10⁵ CFU of a MDR strain of P. aeruginosa. One hour after infection, mice received GNT 15 mg/kg by intraperitoneal injection (green and orange groups) or PBS (purple and red groups). (B) Animal survival was monitored daily. All data are represented are cumulative of two independent experiments. Statistical analysis was performed using the log-rank (Mantel-Cox) test. *P < 0.05, **P < 0.01, ***P < 0.001, .****P < 0.0001

DISCUSSION

In the present study, we used an in vivo mouse model to investigate the effect of prophylactic nasal administration of flagellin in combination, or not, with the antibiotic GNT against a MDR strain of P. aeruginosa (PA^MDR). We demonstrated that (i) a single prophylactic respiratory administration of flagellin was sufficient to decrease bacterial load in the lungs, and when combined with GNT 1 h post-infection, this reduction was even more accentuated, which suggests a synergistic effect between the two treatments; (ii) pre-treatment with flagellin, in combination or not with GNT, reduced inflammation but increased the number of neutrophils, total DCs, and maintained AM levels in the BAL, 16 h post-infection.; and (iii) combination of prophylactic flagellin and GNT 1 h post-infection led to 100% survival, contrary to the control groups.

Several studies have shown that nasal administration of flagellin can reduce bacterial load in an animal infection model (11, 17, 24, 25). For example, the study from Muñoz N., et al. demonstrated that respiratory treatment with flagellin leads to S. pneumoniae clearance in the lungs and promotes survival of infected mice. They attributed such results to increased neutrophil infiltration into the airways as a result of strong chemokine secretion by flagellin-stimulated ECs. Our findings are consistent with such studies. We observed that animals pre-treated with flagellin 24 h before infection, receiving or not GNT 1 h after challenge, presented significantly less PA^MDR bacterial load and higher neutrophil counts in the BAL, compared with infected non-treated mice or only treated with GNT. Neutrophils play a critical role in combating infections through their antimicrobial functions. Particularly in the context of P. aeruginosa, it has been demonstrated that early neutrophil influx to the respiratory airways is essential for the clearance of the pathogen (12, 26, 27). Accordingly, we hypothesize that the rise in neutrophil counts in flagellin-treated mice could play a role in the decrease of P. aeruginosa CFU in those same mice. The secretion of antimicrobial compounds by ECs (11, 17) may also contribute to the reduction of PA^MDR. To fully comprehend the effect of antimicrobial effectors in our model, additional experiments should be carried out.

While standalone flagellin was effective in clearing PA^MDR, our study showed that the combination of prophylactic flagellin and therapeutic GNT confers an additional advantage. Mice treated with this combination showed an even greater reduction in lung bacterial load compared with those treated with flagellin alone, suggesting a synergy between the antibiotic and flagellin. Since we did not find any significant difference in the number of immune cell populations, particularly neutrophils, between mice treated with flagellin alone and those treated with flagellin +GNT, the observed additional advantage in the combination group may be attributed to the direct antibiotic’s impact despite GNT’s limited efficacy against PA^MDR when used alone. We hypothesize that the prophylactic nasal administration of flagellin may have contributed not only to the recruitment of phagocytes to the site of infection but also to the secretion of multiple antimicrobial peptides, which in turn led to the elimination of, at least, some of the bacteria present in the respiratory tract. When GNT was administered, 1 h after infection, less viable bacteria in the flagellin-regulated lung mucosa enabled the antibiotic to efficiently clear most of the remaining bacteria.

Thus, our work is reminiscent of previous studies showing that a host-directed immunomodulator can contribute to the clearance of a pathogen, facilitating the action of an antibiotic (13, 28,–31). Secoeudesma sesquiterpenes lactone A (SESLA) is an anti-inflammatory molecule proven to modulate NFκB and PI3K/Akt signaling pathways (31). When in simultaneous combination with the antibiotic meropenem, it promotes bacterial clearance and survival in mice infected with carbapenem-resistant Klebsiella pneumoniae. Recently, a similar strategy was found to be efficacious in the context of intestinal infections with Salmonella Typhimurium (28). Indeed, the combination of systemic administration of ciprofloxacin and a TLR4 or TLR9 agonist was able to further reduce the number of antibiotic-tolerant Salmonella in the gut and the associated draining lymph nodes, compared with standalone treatments (28). Another study also showed that the simultaneous combination of flagellin and the antibiotic amoxicillin, administered 12 h post-infection, can provide additional protection by effectively eliminating a resistant strain of S. pneumoniae (13).

A host-directed immunomodulator, when paired with an antibiotic, should also promote a balanced inflammatory response and confer long-lasting immunity (32, 33). In our study, respiratory administration with flagellin 24 h before challenge, either alone or in combination with GNT, effectively reduced excessive local inflammation caused by PA^MDR infection. Mice receiving the combination of flagellin and GNT exhibited even lower tissue hemorrhage, a decrease in total cell infiltration in the BAL, and a return of proinflammatory cytokine levels to steady state by 16 h post-infection. These findings are in alignment with a recent investigation conducted by López-Gálvez et al. (20). Using airway epithelial cells and a pig model, their research demonstrated that flagellin serves as an immunostimulant molecule capable of eliciting a transient proinflammatory reaction. This proinflammatory stimulus plays a pivotal role in initiating an efficient host immune response toward a forthcoming pathogen. However, because flagellin’s proinflammatory stimulus is short-lived and the molecule itself is also rapidly degraded within the airway mucosa, a few hours after its administration, there is no longer inflammation driven by flagellin. They also show that particularly in the context of P. aeruginosa infection, a first prophylactic stimulus with flagellin is enough to attenuate P. aeruginosa-triggered inflammatory response. Although the detailed mechanism is still unclear, López-Gálvez et al. (20) suggest that epigenetic changes or trained immunity caused by flagellin (also present in P. aeruginosa flagellum) may also contribute to restraining inflammation upon P. aeruginosa challenge (34, 35). For example, the research of Bigot and colleagues demonstrated that bronchial epithelial cells pre-exposed with flagellin modify their inflammatory response to a second stimulus due to epigenetic regulation (35). The observed decreased inflammatory response could also be related to TLR5 signaling desensitization due to a decrease in receptor availability or tachyphylaxis, as observed when repeated doses of TLR ligands are administered to mice (36). In addition, it has been shown that flagellin treatment increased the concentration of TNFAIP3. This key negative regulator of the innate immune response can decrease the signalling of both TLR4 and TLR5 pathways, thereby modifying the ability of airway epithelial cells to respond to a second stimulation by P. aeruginosa (37, 38).

Our findings also demonstrate that preventive administration of flagellin can enhance the preservation of AM, which were diminished following P. aeruginosa infection. Such apparent reduction has also been previously reported in the context of influenza and SARS-Cov-2 infection (39, 40). Although we have yet to fully elucidate the precise mechanisms underlying this decrease in our PA^MDR infection model, our data strongly suggest that nasal flagellin administration effectively mitigates AM reduction. The advantages of preserving AM in the context of respiratory infections are vast (41). These cells are the first to encounter pathogens in the alveoli. After an antigen exposure, AM become activated and not only are able to internalize and kill pathogens but also produce several proinflammatory cytokines and chemokines that recruit neutrophils to the site of infection, helping with clearance. Thus, flagellin-dependent AM preservation may contribute to the reduction of P. aeruginosa infections.

It has been shown that total DCs, which include monocyte derived-DCs and conventional DCs, are increased in the airway mucosa after stimulus with flagellin. Although we did not discriminate DC subtypes in the current study, we also observed an increase in these immune cell populations in animals that were nasally challenged with flagellin, 24 h before infection (receiving or not GNT 1 h post-infection). These results may indicate possible long-lasting immunity driven by prophylactic nasal flagellin against a MDR strain of P. aeruginosa, although further studies would have to be conducted to confirm such a hypothesis.

Finally, the beneficial synergy between flagellin and gentamicin was evidenced in a survival study. The combination between flagellin and GNT protected all infected mice, which was not observed in any other group. The observed survival outcome may be attributed to the synergistic effects of flagellin and the restoration of GNT effectiveness, involving several mechanisms as follows: (i) flagellin enhances the influx of neutrophils into the lungs, (ii) these cells aid in bacterial clearance, (iii) the reduced bacterial burden increases the antibiotic-to-bacteria ratio, thereby restoring the efficacy of GNT, and (iv) flagellin’s prophylactic effect mitigates excessive inflammation in the lungs, ultimately contributing to improved host survival. Such results highlight the advantageous synergy between the two molecules and underline the importance of investigating such combination further. Moreover, our work extends the current knowledge about the beneficial effect of flagellin as host-directed immunomodulator to MDR strains of P. aeruginosa, which present a significant challenge for patients with chronic respiratory conditions and limited treatment options. Although more research is required to confirm the feasibility of prophylactic flagellin administration in patients, this study provides a basis for a novel dual therapy against P. aeruginosa in the context of increasing antimicrobial resistance.

ACKNOWLEDGMENTS

REFERENCES