Antibiotic-Induced Dysbiosis of the Gut Microbiota Shifts Host Tryptophan Metabolism and Increases the Susceptibility of Mice to Pulmonary Infection With Pseudomonas aeruginosa

Abstract

Pseudomonas aeruginosa is an opportunistic bacterium that mainly infects those who have previously been treated with antibiotics. We hypothesised that antibiotic treatment disrupts tryptophan metabolism, leading to increased susceptibility to P. aeruginosa infection. Our results showed that mice receiving antibiotics exhibited intestinal dysbiosis with alterations in host tryptophan metabolism, a higher mortality rate and a higher bacterial load compared to eubiotic mice. In the lungs of the dysbiotic mice, there was an increase in IDO1 (Indoleamine 2,3-dioxygenase 1) activity and an accumulation of kynurenine after infection, and IDO1^-/- mice were resistant to infection after induction of dysbiosis. Importantly, dysbiosis led to increased expression and activation of AHR (Aryl Hydrocarbon Receptor) in an IDO1-dependent manner. Blocking AHR activation in dysbiotic mice resulted in a lower bacterial load. Our data showed that increased AHR activation by kynurenine was associated with decreased phagocytosis of P. aeruginosa by macrophages and neutrophils. In conclusion, our results indicate that dysbiosis resulting from prolonged antimicrobial treatment alters tryptophan metabolism, leading to activation of the IDO1-AHR axis and increasing susceptibility to P. aeruginosa infection. Furthermore, these data suggest that IDO1 or AHR are potential host targets for the prevention of opportunistic infections in patients undergoing antimicrobial therapy.

Keywords: AHR and Pseudomonas aeruginosa; IDO1; antibiotics; host‐targeted therapy; inflammation; neutrophils.

FIGURE 1

Treatment with an antibiotic cocktail alters host tryptophan metabolism and increases mice susceptibility to P.aeruginosa infection. C57/BL6 mice were treated with an antibiotic cocktail in drinking water for 14 days. At the end of the protocol, the mice were euthanized, and serum was collected to determine the concentration of tryptophan (A) and kynurenine (B) by UPLC‐MS. One group of mice was following to determine survival for 7 days (C). In addition, mice were euthanized 6 or 24 h after infection, and BAL and lungs were collected for subsequent analyses: Bacterial load in BAL (D) and lungs (E). In A and B, statistical analyses were performed with the one‐way test ANOVA followed by the Newman–Keuls post‐test. *p < 0.05 versus NI; ^# p < 0.05 versus H₂O 24 h. Experimental N: 5–8. In C, results are expressed as percentage of survival after infection. Experimental N = 6–8. Statistical analysis comparing survival curves was performed with the log‐rank test (Mantel‐Cox). In D and E, the results of 3 independent experiments with experimental N: 5–8 were pooled. Statistical analysis was performed with a one‐way test ANOVA followed by a Newman–Keuls post‐test. *p < 0.05 versus NI ^# p < 0.05 versus H₂O 24 h.

FIGURE 2

Treatment with an antibiotic cocktail impacts neutrophil influx after infection with Pseudomonas aeruginosa . C57/BL6 mice were treated with the antibiotic cocktail for 14 days. At the end of the protocol, mice were intranasally infected with 10⁷ CFU of the PAO1 strain. At 6 and 24 h after infection, mice were euthanized and BAL and lungs were collected for subsequent analyses: MPO activity assay in lung homogenates (A); percentage (B) and the total number of neutrophils (B) by flow cytometry; total leukocytes (C), neutrophil (E), and alveolar macrophage (F) counts in BAL. Production of ROS by neutrophils was determined in alveolar lavage (G, H) by flow cytometry. In A and B, results from three independent experiments were pooled with N: 5–8. In C and D, one representative experiment from three independent experiments was shown (N: 5). In A–E, statistical analysis was performed with a one‐way test ANOVA followed by a Newman–Keuls post‐test. *p < 0.05 versus N.I. ^# p < 0.05 versus H₂O 24 h. In F, statistical analysis was performed with Student's t‐test. *p < 0.05 versus H₂O. Experiment N: 4.

FIGURE 3

IDO1‐mediated kynurenine production impairs clearance of P. aeruginosa during dysbiosis. C57/BL6 and IDO1^−/− mice were treated with the antibiotic cocktail for 14 days. At the end of the protocol, the mice were intranasally infected with 10⁷ CFU of the PAO1 strain. Twenty‐four hours after infection, the mice were euthanized, and the lungs were harvested for quantification of Trp metabolites by UPLC‐MS, and the ratio of Kyn to Trp was calculated (A and B). Twenty‐four hours after protocol infection, another group of mice were euthanized and BAL and the lungs were examined for bacterial load in BAL (C) and the lungs (D). The total number of leukocytes (E), neutrophil granulocytes in BAL (F), and the production of ROS by neutrophil granulocytes in BAL (G) were analysed. C57BL/6 mice received 25 mg/kg kynurenine or vehicle intraperitoneally 1 h before intranasal infection with 10⁷ CFU of the PAO1 strain. After 24 h, BAL and lungs were harvested to determine bacterial load (H and I) and total infiltrated leukocyte and neutrophil count at BAL (J and L). In A, results are expressed as relative increase over the mean of the group water NI. Statistical analysis was performed with the two‐way test ANOVA and Sidak multiple comparison test. *p < 0.05 versus respective NI group, #p < 0.05 versus respective WT group and +p < 0.05 versus respective water‐treated group. In E and F, statistical analysis was performed using the two‐way test ANOVA and the Sidak multiple comparison test. In C and D *p < 0.05 versus H₂O and #p < 0.05 versus WT ATB. In E *p < 0.05 versus NI #p < 0.05 versus WT 24 h. In F *p < 0.05 versus H₂O. In G–J, statistical analysis was performed with a one‐way test ANOVA followed by a Newman–Keuls post‐test. In F and G #p < 0.05 versus vehicle. In H and I *p < 0.05 versus NI, #p < 0.05 versus vehicle treated infected. Experimental N: 4–6.

FIGURE 4

AHR upregulation and activation mediate the increased susceptibility of dysbiotic mice to P. aeruginosa infection. C57/BL6 mice were treated with the antibiotic cocktail for 14 days. At the end of the protocol, mice were intranasally infected with 10⁷ CFU of the PAO1 strain. Twenty‐four hours after the infection, mice were euthanized and BAL was harvested for flow cytometric analysis to determine the percentage of neutrophils expressing AHR (A), the MFI of AHR staining in neutrophils (B and C) and lungs were collected to assess Cyp1a2 expression (D) and MFI of AHR expression in neutrophils (E‐F). A group of C57/BL6 mice treated with an antibiotic cocktail received 10 mg/kg of the AHR antagonist CH223191 or vehicle intraperitoneally 1 h before infection with 10⁷ CFU of the PAO1 strain. After 24 h, mice were euthanized, and BAL and lungs were harvested for quantification of bacterial load (G and H) and total neutrophil count in the BAL (I). In A–D, statistical analysis was performed with the one‐way ANOVA test followed by the Newman–Keuls post‐test. *p < 0.05 versus respective NI; #p < 0.05 versus H₂O 24 h. Experimental N: 3–6. In E, statistical analysis was performed with the two‐way ANOVA test followed by Sidak's for multiple comparisons test. #p < 0.05 versus H₂O 24 h; &p < 0.05 versus WT ATB 24 h. In G–I, statistical analysis was performed with the two‐way ANOVA test followed by Sidak's multiple comparisons test. *p < 0.05 versus respective N.I.; #p < 0.05 versus H₂O 24 h. Experimental N: 5–7.

FIGURE 5

IDO1‐mediated increased kynurenine production disrupts phagocytosis of P. aeruginosa in an AHR dependent manner. C57/BL6 and IDO1^−/− mice were treated with the antibiotic cocktail for 14 days. At the end of the protocol, mice were intranasally infected with 10⁷ CFU of the PAO1 strain. After 2 h after infection, mice were euthanized, and BAL was collected for determination of phagocytosis index (A and B). Bone marrow cells from WT and IDO1^−/− mice were harvested and differentiated into macrophages for 7 days. Phagocytosis and killing assays were then performed to determine the CFU of phagocytosed P. aeruginosa (C), and the percentage of killing (D). BMDM cells from WT mice and human peripheral blood neutrophils were treated with kynurenine at concentrations of 5 or 50 μM for 1 h. After 1 h of treatment, phagocytosis and killing assays were performed to determine the CFU of phagocytosed P. aeruginosa (E and G) and the percentage of killing (F and H). BMDM cells from WT mice were treated with 10uM CH223191 for 1 h and following by treatment with kynurenine at a concentration of 50 μM for 1 h. After the treatments, phagocytosis (I) and killing (J) assays were performed to determine the CFU of phagocytosed P. aeruginosa . For in vivo evaluation, C57BL/6 mice received 10 mg/kg of the AHR antagonist CH223191 or vehicle, and after 1 h of the treatment, mice received 25 mg/kg kynurenine or vehicle. Mice were infected intranasally with 10⁷ CFU of the PAO1 strain, and 2 h after infection, mice were euthanized and BAL collected for phagocytosis index (K) determination. In A, C and G statistical analyses were performed using Student's t‐test. In B, E, F and K statistical analysis was performed using the one‐way ANOVA test followed by the Newman–Keuls post‐test. In G and H statistical analysis was performed in a paired way. In A *p < 0.05 versus H₂O. In B #p < 0.05 versus WT H₂O; &p < 0.05 versus WT ATB. In C and D *p < 0.05 versus WT. In E–K *p < 0.05 versus vehicle and #p < 0.05 versus kynurenine 50 μM. Experimental N: 4–13.