Human airway epithelium controls Pseudomonas aeruginosa infection via inducible nitric oxide synthase

Abstract

Introduction: Airway epithelial cells play a central role in the innate immune response to invading bacteria, yet adequate human infection models are lacking.

Methods: We utilized mucociliary-differentiated human airway organoids with direct access to the apical side of epithelial cells to model the initial phase of Pseudomonas aeruginosa respiratory tract infection.

Results: Immunofluorescence of infected organoids revealed that Pseudomonas aeruginosa invades the epithelial barrier and subsequently proliferates within the epithelial space. RNA sequencing analysis demonstrated that Pseudomonas infection stimulated innate antimicrobial immune responses, but specifically enhanced the expression of genes of the nitric oxide metabolic pathway. We demonstrated that activation of inducible nitric oxide synthase (iNOS) in airway organoids exposed bacteria to nitrosative stress, effectively inhibiting intra-epithelial pathogen proliferation. Pharmacological inhibition of iNOS resulted in expansion of bacterial proliferation whereas a NO producing drug reduced bacterial numbers. iNOS expression was mainly localized to ciliated epithelial cells of infected airway organoids, which was confirmed in primary human lung tissue during Pseudomonas pneumonia.

Discussion: Our findings highlight the critical role of epithelial-derived iNOS in host defence against Pseudomonas aeruginosa infection. Furthermore, we describe a human tissue model that accurately mimics the airway epithelium, providing a valuable framework for systemically studying host-pathogen interactions in respiratory infections.

Keywords: Pseudomonas aeruginosa; airway epithelia; airway organoids; iNOS; innate immunity.

Figure 1

Pseudomonas aeruginosa infects differentiated human airway organoids and resides in the intracellular space. (A) Immune fluorescence imaging of human airway organoids in basal-out configuration (left panel) and apical-out configuration (middle panel). Representative epifluorescence images of organoids stained for actin (red) and nuclei (blue), mounted on slides. Terminally differentiated organoid (right panel). Representative confocal image, stained for apically located cilia (red) and nuclei (blue). (B) Schematic representation of the organoid infection model used herein. (C) Confocal immune fluorescence imaging revealing intra-organoid bacteria (PA-GFP, green rods) in infected organoids after 3h of active infection and removal of extracellular bacteria. (D) Quantification of intracellular bacteria during the gentamicin-protected course of infection. Infected organoids were lysed at indicated time intervals, and lysates were plated onto LB-agar plates for CFU quantification. Data from three independent experiments are shown as a scatter plot with mean ± SD. *** denotes p < 0.001 for ANOVA with post hoc statistical testing.

Figure 2

Pro-inflammatory signaling shapes organoid responses to infection. (A) Volcano plot showing differentially expressed genes of infected human airway organoids in comparison to uninfected controls after 4h of gentamicin-protected infection. (B) Gene-set analysis with over-representation test (ORA) showing Gene ontology biological processes. (C) Differential IL-6 (left panel) and IL-8 (right panel) mRNA expression in infected organoids. Organoids were harvested at indicated time intervals of gentamicin-protected infection. Data is shown as mean ± SD of a triplicate experiment. (D) Levels of the inflammatory cytokines IL-6 (left panel) and IL-8 (right panel) in supernatants of infected organoids. Supernatants were collected at indicated time intervals of gentamicin-protected infection. Data is shown as mean ± SD of a triplicate experiment. (E) Heat map graphical representation of differentially expressed genes in infected versus uninfected organoids. The genes (X axis) are derived from Gene ontology nitric oxide metabolic process (GO:0046209) set of genes. The gene expression is indicated by z-score. ** denotes p < 0.01, *** denotes p < 0.001 for ANOVA with post-hoc statistical testing.

Figure 3

Human airway organoids induce iNOS in response to infection with Pseudomonas aeruginosa. (A) Graphical representation of infection-associated induced genes as nodes. Pathogen-detection-associated genes are shown as ellipses, intracellular inflammatory signalling genes are shown as squares, and antimicrobial effectors are depicted as diamonds. The colour of the nodes indicates the log2-fold change from the differential expression analysis of infected human airway organoids compared to uninfected controls (B) Differential NOS2 mRNA expression in infected organoids. Organoids were harvested at indicated time intervals of gentamicin-protected infection. Data is shown as mean ± SD of three independent experiments. (C) Differential NOS2 mRNA expression at the 4h time interval in infected organoids and organoids exposed to heat-inactivated (HI) bacteria or treated with sterile-filtrated bacterial supernatant (SN). Organoids were harvested at indicated time intervals of gentamicin protected infection. Data is shown as mean ± SD of three independent experiments. (D) Western blot of the MAPK subunit p38, phosphorylated p38, and iNOS protein of infected organoids and uninfected controls harvested directly after the active infection phase (0h-time interval), shown in duplicates. (E) Densitometric quantification of the Western blots targets iNOS and phosphorylated p38. * denotes p < 0.05 for statistical testing with a two-sided unpaired t-test, *** denotes p < 0.001 for ANOVA with post-hoc statistical testing.

Figure 4

iNOS is induced in ciliated cells of infected airway organoids and human bronchial epithelia during Pseudomonas aeruginosa pneumonia. (A) Immune fluorescence imaging of differentiated human airway organoids, infected with PA14 (lower panels). Representative epifluorescence images of organoids stained for iNOS (yellow), ciliated cells (red) and nuclei (blue). Co-localization of iNOS and ciliated cells is depicted in white (right panel). (B) Histological analysis of human lung samples stained for iNOS protein obtained from a non-infected lung (left panel) and a lung with confirmed Pseudomonas aeruginosa pneumonia (right panel).

Figure 5

Human airway organoids expose Pseudomonas aeruginosa to nitrosative stress. (A) Graphical illustration of the experimental setup: PA were either incubated for 3h in antibiotic-free AO medium alone or in the presence of AOs. After this, mRNA was extracted. (B) Differential mRNA expression of bacterial genes associated with denitrification, in PA versus PA exposed to human AOs. Data is shown as mean ± SD of three independent experiments. ns denotes not significant, ** denotes p < 0.01 for a two-sided, unpaired t-test.

Figure 6

Human airway organoids control intracellular Pseudomonas aeruginosa growth via iNOS. (A) Graphical illustration of the experimental setup: Infected AOs were incubated with or without the specific iNOS inhibitor L-NIL during the 16h gentamicin-protected infection phase. Afterwards, mRNA was extracted. (B) Differential mRNA expression of bacterial genes associated with denitrification, in PA infecting AOs versus PA infection of AOs in the presence of the iNOS inhibitor L-NIL after 16h of gentamicin protected infection. Data is shown as mean ± SD of three independent experiments. (C) Quantification of intracellular bacteria after 24h of gentamicin-protected infection of AOs treated with 50µM iNOS inhibitor L-NIL, 100µM NO-donor NOC-18 or left untreated. Infected organoids were lysed, and lysates were plated onto LB-agar plates for CFU quantification. Data from three independent experiments performed in triplicates are shown as a scatter plot with mean ± SD, normalized to infection control (PA). * denotes p < 0.05 for unpaired t-test, *** denotes p < 0.001 for ANOVA with post-hoc statistical testing.