key: cord-0287862-uwqld0pf
authors: Leon-Icaza, Stephen Adonai; Bagayoko, Salimata; Iakobachvili, Nino; Ferrand, Chloé; Aydogan, Talip; Bernard, Celia; Dafun, Angelique Sanchez; Murris-Espin, Marlène; Mazières, Julien; Bordignon, Pierre Jean; Mazères, Serge; Bernes-Lasserre, Pascale; Ramé, Victoria; Lagarde, Jean-Michel; Marcoux, Julien; Bousquet, Marie Pierre; Chalut, Christian; Guilhot, Christophe; Clevers, Hans; Peters, Peter J.; Molle, Virginie; Lugo-Villarino, Geanncarlo; Cam, Kaymeuang; Berry, Laurence; Meunier, Etienne; Cougoule, Céline
title: Cystic fibrosis patient-derived bronchial organoids unveil druggable pathways against Mycobacterium abscessus infection
date: 2022-01-03
journal: bioRxiv
DOI: 10.1101/2022.01.03.474765
sha: adfc4c5d87e86d8c552b9195aca3b06f7ddd95ca
doc_id: 287862
cord_uid: uwqld0pf

Mycobacterium abscessus (Mabs) drives life-shortening mortality in cystic fibrosis (CF) patients, primarily because of its resistance to chemotherapeutic agents. Both our knowledge on and models to investigate the host and bacterial determinants that drive Mabs pathology in CF patients remain rudimentary. Here, we evaluated whether the lung organoid technology from CF patients is appropriate for modelling Mabs infection and whether antioxidant treatment is a candidate therapeutic approach in the context of CF disease. We derived airway organoids (AOs) from lung biopsy of a CF patient and characterized these AO by assessing CF transmembrane conductance regulator (CFTR) function, mucus and reactive oxygen species (ROS) production, lipid peroxidation, and cell death. We microinjected smooth (S-) or rough (R-)Mabs in the lumen of AOs to evaluate its fitness, responses of AOs to infection, and treatment efficacy by colony forming unit assay, qPCR and microscopy. We show that CF patient-derived AOs exhibited low residual CFTR function, enhanced mucus accumulation, and increased oxidative stress, lipid peroxidation, and cell death at basal state. While in AOs, S Mabs formed biofilm, R Mabs formed cord serpentines and displayed a higher virulence. S and R Mabs replicated more efficiently in CF AOs than in AOs derived from healthy lung. Pharmacological activation of antioxidant pathways resulted in better control of Mabs growth. In conclusion, we have established CF patient-derived AOs as a suitable human system to decipher mechanisms of CF-enhanced respiratory infection by Mabs and confirmed antioxidant approaches as a potential host-directed strategy to improve Mabs infection control.

Cystic Fibrosis (CF) is due to mutations in the CF transmembrane conductance regulator (CFTR) gene (1) , which regulates ion transport, that impair lung mucociliary clearance and result in pathological triad hallmarks of CF, i.e., chronic airway mucus build-up, sustained inflammation, and microbe trapping leading to parenchyma epithelial cell destruction. The major reason CF patients succumbing to this disease is respiratory failure resulting from chronic lung infection (2) .

CF Patients have a greater risk of infection by Non-Tuberculous Mycobacteria (NTM), mainly by the most virulent and drug-resistant NTM Mycobacterium abscessus (Mabs) (3) (4). Mabs display two distinct morphotypes based on the presence or absence of glycopeptidolipids (GPL) in their cell wall (5) . The smooth (S) GPL-expressing variant forms biofilm and is associated with environmental isolates. The Rough (R) variant does not express GPL, forms cording and induces more aggressive and invasive pulmonary disease, particularly in CF patients (5) (6) (7) . In CF patients, Mabs colonization is initiated by the infection with the S variant that, over time, switches to the R morphotype by losing or down-regulating surface GPL (8, 9) . Although animal models like immunocompromised mice (10), zebrafish (11) (12) (13) and Xenopus laevis (14) contributed to a significant advance in the understanding of Mabs infection (15) , their tissue architecture and cell composition are different from that of humans and do not recapitulate the hallmarks of CF (16) (17) (18) . Models with anatomical and functional relevance to the human airway and displaying the diversity of natural CFTR gene mutations would complement other in vivo models.

Recent advances in stem cell biology have allowed the growth of human tissues in vitro that resemble organs in vivo (19) (20) (21) . Human airway organoids (AOs) are derived from adult stem cells present in lung tissues (22) , are self-organized 3D structures and share important characteristics with adult bronchiolar part of the human lung (22, 23) . We and others have adapted AOs for modelling infectious diseases with bacteria, such as mycobacteria (24) and Pseudomonas aeruginosa (25) , with viruses, such as RSV (22) and SARS-CoV-2 (26) (27) (28) , and with parasites (29) . Of particular interest, organoids derived from CF patients constitute a unique system to model natural mutations of CFTR and its dysfunctions, thus recapitulating critical aspects of CF in human (22, 30, 31) that are not achievable with other cellular or animal models.

Here, we report the generation and characterization of an AO line derived from a CF patient that recapitulates hallmarks of human CF disease, including exacerbated oxidative stress. Both Mabs S and R infect and replicate within these AOs and display their specific features, such as forming biofilm and cording, respectively. Moreover, enhanced reactive oxygen species (ROS) production by Mabs infection or the CF context favours Mabs growth, which is reversed by antioxidant treatments.

Detailed protocols are provided in the supplement. All patients participating in this study consented to scientific use of their material; patients can withdraw their consent at any time, leading to the prompt disposal of their tissue and any derived material.

Healthy adjacent tissue from two independent donors with lung cancer, and one lung biopsy of cystic fibrosis heterozygous patient (G542X/1811+1.6kbA-->G) were used to derive organoids as previously described with minor changes (22, 24) .

For the epithelium thickness measurement, organoids were cultured for five weeks before acquired bright field images. Assessing CFTR function was performed by forskolin-induced organoid swelling assay (5 μM, 2 hr at 37 °C) (Sigma-Aldrich, Burlington, MA, USA) as described (33) . Live imaging was performed to quantify mucus accumulation with Zinpyr-1 (10 μM, overnight at 37 °C) (Santa Cruz 

Mycobacterium abscessus sensu stricto strain CIP104536T (ATCC19977T) morphotype S and R were grown as previously described (34) . Before infection, AO were seeded in Matrigel (Fisher Scientific, Hampton, NH, USA) and depending on the indicated conditions, pretreated or no with 10μM of Resveratrol (Sigma-Aldrich) or Sulforaphane (Selleck Chemicals, Houston, TX, USA) for 1hr or 6hr respectively. Both antioxidants were maintained throughout the experiment and refreshed every two days. The infection day, the bacteria were prepared for microinjection as described (24, 35) and density was adjusted to OD600 = 0.1-0.4. Injected organoids were individually collected, washed in PBS 1x and embedded into fresh matrix. Injected organoids were cultured for 3-4 day if not stated otherwise.

Organoids were collected at day 4 post-infection or stimulation and processed as reported (24) .

Primer sequences are provided in supplementary Table E1 .

Statistical analyses were performed using Prism 8 and 5 (GraphPad Software). Data were compared by Mann-Whitney or unpaired T test and results reported as mean with SD. Data statistically significant was represented by *P<0.05; **P<0.01; ***P<0.001 and **** P<0.0001.

We adapted the published protocol to generate AOs using lung biopsies (22, 24) Altogether, these results showed that organoids derived from CF lung tissue exhibited not only CFTR dysfunction and exacerbated mucus accumulation but also an increased oxidative stress, therefore representing a suitable ex vivo model to investigate CF-driven respiratory infections.

To investigate whether AOs can be used for modelling Mabs infection in vitro, we infected healthy AOs with S-and R-Mabs variants as previously described (24) Interestingly, we observed that S bacteria formed aggregates in the lumen of the organoids whereas R variant formed the characteristic serpentine cords ( Figure 2B -C) observed in vitro and in vivo (34) .

AOs were analysed by SEM and then TEM ( Figure 2D ). As previously described (22) , the organoid epithelium is composed of ciliated and goblet cells ( Figure 2D, 1st row) . As expected, Mabs S bacilli occupied homogeneously the organoid lumen, formed chaotically scattered aggregates ( Figure 2D , 2nd row) (45) . Mabs S preferentially localised in close contact with the apical side of the epithelial cells ( Figure 2D, 4) , particularly in the presence of cilia ( Figure 2D The contribution of cell protective antioxidant pathways during mycobacterial infection remains poorly understood (49, 50) . We then determined the consequence of boosting antioxidant pathways for Mabs fitness and found that treatment with either resveratrol or the Nrf-2 agonist sulforaphane reduced both S and R variant growth in AOs (Figures 3E-H) .

Altogether, the results showed that, independent of the immune system, epithelial cells mounted an oxidative response upon Mabs infection, which contributed to Mabs growth in the airway microenvironment.

As oxidative stress is enhanced in CF AOs, we hypothesised that the CF context could favour Mabs growth. To test this hypothesis, we infected CF-AOs with S-or R-Mabs variant and quantified Mabs (Figures 4C-D, Sup Fig. 4B ). Next, we used CFTR inhibitors to support our results. (Supp Fig. 4C) . We treated H-AOs with CFTR inhibitors then infected them with S-and R-Mabs variants and measured bacterial load 4 days post-infection. Treatment of AOs with CFTR inhibitors enhanced Mabs proliferation compared with untreated organoids (Sup Fig. 4D-F) , confirming that alteration of CFTR function promoted Mabs growth.

As oxidative stress favoured Mabs growth and CF-AOs displayed increased ROS production, we next evaluated whether antioxidants inhibited Mabs growth. CF-AOs treated with sulforaphane expressed higher NQO-1, confirming Nrf-2 activation (SuppFigure 4G), exhibited mitigated oxidative environment ( Figure 4E, Sup Fig. 4G) , and reduced bacterial load ( Figures 4F-G) , indicating that CFdriven oxidative stress stimulated Mabs growth.

Altogether, these results showed that the CF lung environment favoured Mabs fitness, which could be mediated, at least in part, by exuberant oxidative stress due to CFTR dysfunction.

In this study, we evaluate the pertinence of CF-patient derived AOs to model CF-associated respiratory infection. We show that AOs derived from a CF patient recapitulate key features of CF The organoid technology has demonstrated their usefulness in developing potential therapies for treating CF (51, 52) . CF patient-derived organoids bear natural CFTR mutations, thus allowing to recapitulate ex vivo the spectrum of CFTR dysfunctions and CF disease severities (30, 53) . Extended to the airway, CF patient-derived AOs have been shown to display epithelium hyperplasia, luminal mucus accumulation and abrogated response to forskolin-induced swelling, thus recapitulating CFTR dysfunction and consequences on the airway homeostasis (22) . Here, we derived an AO line from a CF patient carrying class I mutations, which lead to the most severe manifestation of CF disease (54) . These CF AOs reproduce the expected epithelium hyperplasia, mucus accumulation and defective response to forskolin-induced swelling. Moreover, these CF AOs display enhanced oxidative stress and lipid peroxidation, as previously measured in CF patients (55, 56) or in vitro in CFTR mutated cell lines (57) , as well as enhanced cell death recapitulating CF-driven tissue damage. Therefore, AOs derived from CF patients also recapitulate the CF-driven imbalance in oxidant/antioxidant status observed in patients and, importantly, demonstrate that CFTR dysfunction in epithelial cells is sufficient to cause the oxidative status imbalance in the airway epithelium, independent of immune cells.

We and others have already applied the organoid technology to model host-pathogen interactions (22, 24, 25, 29, 58, 59) . Here, we have reproduced Mabs infection hallmarks in AOs. Specifically, we show that both S-and R-Mabs replicate in AOs, that S variants are surrounded by an extracellular substance resembling a biofilm, and that the R variant forms serpentine cords associated with higher virulence, in agreement with reported in vivo Mabs R hypervirulence compared to Mabs S (5-7).

Although Mabs R form characteristic cords in vitro and in vivo (34, 60) , visualizing bacterial biofilm remains a challenge, especially in ex vivo and in vivo settings. The formation of biofilm plays a crucial role in establishing an infection and at protecting bacilli from immune response and antimicrobial agents, leading to treatment failure (61, 62) . Therefore, the detection of biofilm in Mabs S-infected AOs opens new venues to not only testing potential anti-biofilm compounds but also further investigating how the CF lung context favours biofilm formation (61, 63) .

ROS production is a part of host antimicrobial defence but requires a fine-tuned balance to prevent tissue damage. Indeed, during Mabs infection, the production of ROS is essential to control infection, as knock-down of NOX2, expressed in immune cells, resulted in uncontrolled bacteria proliferation in zebrafish and mouse models (12, 64) . Here, we show that Mabs infection of AOs triggers ROS production through enhanced expression of host ROS production genes NOX1 and DUOX1 concomitantly to enhanced expression of the Nrf2-mediated antioxidant pathway. In alveolar macrophages, Nrf2 deficiency results in a better control of Mtb infection (49) , indicating that early activation of cell protective pathways impairs the control of mycobacteria. By contrast, an oxidative environment has been shown to favour Mabs growth in macrophages (66, 67) .

Moreover, activation of Nrf2 reduces the bacterial burden both in vitro and in vivo (47, 48, 50, 68) .

Finally, apart from its direct antimicrobial effect, the antioxidant N-acetyl cysteine has been shown to inhibit Mtb growth in macrophages and in infected mice (69) . Consistently, we show here that treatment of Mabs-infected AOs with resveratrol or sulforaphane results in a better control of bacteria growth. Because resveratrol and sulforaphane do not inhibit Mabs growth in vitro, our results indicate that ROS production by epithelial cells is sufficient to generate a permissive environment for Mabs proliferation. Even in the absence of immune cells and low bacteria internalization by epithelial cells (24) , AOs recapitulate the contribution of oxidative stress in bacteria fitness in the airway.

Finally, we also show that the enhanced ROS production and the resultant oxidative microenvironment of CF AOs favour the multiplication of Mabs associated with enhanced cell death, recapitulating the higher susceptibility of CF patients to NTM infection (71, 72) . Cumulative oxidative stress due to both the CF context and bacterial infection might result in a permissive environment for bacteria growth and the establishment of chronic infection in the lung of CF patients.

Interestingly, CF is associated with a defective Nrf2 expression, which contributes to the excessive oxidative stress and lung tissue damage, whereas CFTR modulators rescue Nrf2 function and therefore improve tissue oxidative status (40) , which might contribute to a better control of bacterial infection. Here, we show that genetic or pharmacological inhibition of CFTR causes enhanced oxidative stress and bacteria growth, further showing that proper CFTR function in epithelial cells is a part of the intrinsic airway redox balance and homeostasis. CF AOs treated with sulforaphane have reduced oxidative stress and better control of Mabs growth, indicating that activation of Nrf2 can constitute a therapeutic strategy to improve tissue redox homeostasis and better infection control. Further investigation using animal models are required to integrate ROS production by immune cells. Interestingly, treating R Mabs-infected zebrafishes with resveratrol improves fish survival and reduces bacterial load (48) .

A limitation of our study is that only one CF patient-derived organoid line has been used. This CF AO line bears class I mutations that lead to the most severe form of CF disease. As a complementary approach, we also observe better bacteria growth by using CFTR inhibitors to partially alter the CFTR function indicating that this phenotype can also be observed with partially impaired CFTR function as seen with other classes of CFTR mutations. Nevertheless, establishing CF AO lines carrying a wide range of CFTR mutations might be facilitated by emerging protocols (73) and would constitute precious tools to further model the full spectrum of CF disease severity, susceptibility to infection and potential of host-directed therapies targeting oxidative stress on tissue damage (25) and bacteria fitness.

In conclusion, we have established CF patient-derived AOs as a pertinent model of CF airway dysfunction and susceptibility to NTM infection. Moreover, we have identified the cell protective Nrf2 pathway as a potential therapeutic target to restore CF tissue redox homeostasis and improve the control of bacteria growth, opening promising venues to further decipher CF airway dysfunction and susceptibility to infection. 

Molecular basis of cystic fibrosis: from bench to bedside

The Changing Face of Cystic Fibrosis in the Era of Precision Medicine

Mycobacterium abscessus: Environmental bacterium turned clinical nightmare

Advances in the management of pulmonary disease due to Mycobacterium abscessus complex

Non-tuberculous mycobacteria and the rise of Mycobacterium abscessus

Mycobacterium abscessus smooth and rough morphotypes form antimicrobialtolerant biofilm phenotypes but are killed by acetic acid

General Overview of Nontuberculous Mycobacteria Opportunistic Pathogens: Mycobacterium avium and Mycobacterium abscessus

Molecular Epidemiology of Mycobacterium abscessus, with Focus on Cystic Fibrosis

Efficacy of bedaquiline, alone or in combination with imipenem, against Mycobacterium abscessus in C3HeB/FeJ mice

Vivo assessment of drug efficacy against Mycobacterium abscessus using the embryonic zebrafish test system

CFTR Protects against Mycobacterium abscessus Infection by Fine-Tuning Host Oxidative Defenses

Inhibition of the β-lactamase BlaMab by avibactam improves the in vitro and in vivo efficacy of imipenem against Mycobacterium abscessus

Developing tadpole xenopus laevis as a comparative animal model to study Mycobacterium abscessus pathogenicity

The Diverse Cellular and Animal Models to Decipher the Physiopathological Traits of Mycobacterium abscessus Infection

Animal and Cell Culture Models for Cystic Fibrosis: Which Model Is Right for Your Application?

Airway disease phenotypes in animal models of cystic fibrosis

Animal Models in the Pathophysiology of Cystic Fibrosis

Human Organoids: Tools for Understanding Biology and Treating Diseases

Human organoids: model systems for human biology and medicine

Modeling Development and Disease with Organoids

Long-term expanding human airway organoids for disease modeling

Humans in a dish: The potential of organoids in modeling immunity and infectious diseases

Mycobacteriahost interactions in human bronchiolar airway organoids

Host phospholipid peroxidation fuels ExoU-dependent cell necrosis and supports Pseudomonas aeruginosa-driven pathology

Spheroids and organoids as humanized 3D scaffold-free engineered tissues for SARS-CoV-2 viral infection and drug screening

Sars-cov-2 entry into human airway organoids is serine protease-mediated and facilitated by the multibasic cleavage site

Identification of SARS-CoV-2 inhibitors using lung and colonic organoids

Modelling Cryptosporidium infection in human small intestinal and lung organoids

Forskolin-induced swelling of intestinal organoids correlates with disease severity in adults with cystic fibrosis and homozygous F508del mutations

Characterizing responses to CFTR-modulating drugs using rectal organoids derived from subjects with cystic fibrosis

Organoids as model for infectious diseases: Culture of human and murine stomach organoids and microinjection of helicobacter pylori

Forskolin-induced Swelling in Intestinal Organoids: An In Vitro Assay for Assessing Drug Response in Cystic Fibrosis Patients

Mycobacterium abscessus cording prevents phagocytosis and promotes abscess formation

Tuberculosis is associated with expansion of a motile, permissive and immunomodulatory CD16(+) monocyte population via the IL-10/STAT3 axis

Cystic fibrosis

Dysfunctional Inflammation in Cystic Fibrosis Airways: From Mechanisms to Novel Therapeutic Approaches

Dysfunction of Nrf-2 in CF epithelia leads to excess intracellular H2O2 and inflammatory cytokine production

Oxidative stress and antioxidant therapy in cystic fibrosis

Clinically approved CFTR modulators rescue Nrf2 dysfunction in cystic fibrosis airway epithelia

An assessment of techniques to manipulate oxidative stress in animals

Lipid peroxidation and antioxidant defenses in cystic fibrosis patients

Reactive Oxygen Species-Induced Lipid Peroxidation in Apoptosis, Autophagy, and Ferroptosis

The Chemistry of Reactive Oxygen Species (ROS) Revisited: Outlining Their Role in Biological Macromolecules (DNA, Lipids and Proteins) and Induced Pathologies

Mycobacterium abscessus Glycopeptidolipid Prevents Respiratory Epithelial TLR2 Signaling as Measured by HβD2 Gene Expression and IL-8 Release

Advanced biofilm staining techniques for TEM and SEM in geomicrobiology: Implications for visualizing EPS architecture, mineral nucleation, and microfossil generation

Caspaseindependent apoptosis in infected macrophages triggered by sulforaphane via Nrf2/p38 signaling pathways

Sirtuin 3 is essential for host defense against Mycobacterium abscessus infection through regulation of mitochondrial homeostasis

Alveolar macrophages generate a noncanonical NRF2-driven transcriptional response to Mycobacterium tuberculosis in vivo

Nrf2 regulates granuloma formation and macrophage activation during mycobacterium avium infection via mediating nramp1 and ho-1 expressions

Comparison of Organoid Swelling and In Vivo Biomarkers of CFTR Function to Determine Effects of Lumacaftor-Ivacaftor in Patients with Cystic Fibrosis Homozygous for the F508del Mutation

CFTR function and clinical response to modulators parallel nasal epithelial organoid swelling

Stratifying infants with cystic fibrosis for disease severity using intestinal organoid swelling as a biomarker of CFTR function

Splice mutation 1811+1.6kbA>G causes severe cystic fibrosis with pancreatic insufficiency: report of 11 compound heterozygous and two homozygous patients

Oxygen free radicals and antioxidants in cystic fibrosis: the concept of an oxidant-antioxidant imbalance

Circulating biomarkers of antioxidant status and oxidative stress in people with cystic fibrosis: A systematic review and meta-analysis

Aberrant GSH reductase and NOX activities concur with defective CFTR to pro-oxidative imbalance in cystic fibrosis airways

An organoid-derived bronchioalveolar model for SARS-CoV-2 infection of human alveolar type II-like cells

Differentiated human airway organoids to assess infectivity of emerging influenza virus

The distinct fate of smooth and rough Mycobacterium abscessus variants inside macrophages

Biofilm formation in the lung contributes to virulence and drug tolerance of Mycobacterium tuberculosis

The role of bacterial biofilms in chronic infections

Mycobacterial infection induces higher interleukin-1β and dysregulated lung inflammation in mice with defective leukocyte NADPH oxidase

Induction of heme oxygenase-1 contributes to survival of Mycobacterium abscessus in human macrophages-like THP-1 cells

Mycobacterium abscessus infection leads to enhanced production of type 1 interferon and NLRP3 inflammasome activation in murine macrophages via mitochondrial oxidative stress

An oxidative environment promotes growth of Mycobacterium abscessus

Activation of Nrf2 signaling by oltipraz inhibits death of human macrophages with mycobacterium tuberculosis infection

N-acetyl-cysteine exhibits potent antimycobacterial activity in addition to its known anti-oxidative functions

Functions of NQO1 in cellular protection and CoQ10 metabolism and its potential role as a redox sensitive molecular switch

Nontuberculous Mycobacteria in Cystic Fibrosis

Comparing the harmful effects of nontuberculous mycobacteria and Gram negative bacteria on lung function in patients with cystic fibrosis

Modelling of primary ciliary dyskinesia using patient-derived airway organoids

B) Representative bright-field images (A) and quantification (B) of epithelium thickness in healthy

Data from three independent wells per donor. (C) Percentage of area increase of H-AO (n=13), H-AO pre-treated with CFTR inhibitors for 4 days (CFTR-Inh, n=13), and CF-AO (n=13) after 2hr stimulation with 5µM forskolin. Data from two independent wells per donor

The volcano plot showing the fold-change (x-axis) versus the significance (y-axis) of the proteins identified by LC-MS/MS in CF-AOs vs in H-AOs. The significance (non-adjusted p-value) and the fold-change are converted to −Log10(p-value) and Log2(fold-change), respectively. (G, H) Representative images (G) and MFI quantification (H) of basal mitochondrial ROS production (MitoSOX) in H-AO (n=6) and CF-AO (n=6)

Representative images (J) and MFI quantification (K) of peroxidized lipids (BODIPY) in H-AO (n=14) and CF-AO (n=14). (L) MFI quantification of the basal plasma membrane permeabilization (propidium iodide incorporation) in H-AO (n=6) and CF-AO (n=6)

Figure 2. Mabs infection in healthy airway organoids. (A) Kinetics of Mabs S and R growth in H-AO

Graph shows three pooled independent experiments. (B) Representative images of Mock (PBS) infected H-AO or H-AO infected with tdTomato-labelled Mabs S or R. (C) Light-sheet fluorescence microscopy of a XY plane at the z=120µm (left two images) or z=80µm (right two images) positions of a H-AO infected with Mabs S and R, respectively

Electron micrographs obtained with a FEI Quanta200 scanning electron microscope set up in backscattered mode. Resin blocks were sectioned and imaged at different magnifications showing normal H-AO organization and the different cell types typical of lung epithelium (top row), the biofilm formed by Mabs S on the luminal face of the epithelial cells (middle row), and the bacterial aggregates typical of the cording in the lumen of Mabs R infected H-AO (bottom row). Targeted ultrathin sections were made and observed by transmission electron microscopy

Representative images (E) and MFI quantification (F) of propidium iodide incorporation in Mock infected H-AOs (n=13) or H-AOs infected with Wasabi-labelled Mabs S (n=17) or R (n=15) for

The dotted lines delimit the organoids circumference. Graph represents means ± SD from three independent experiments, indicated by different symbols

001 by Mann-Whitney test. Yellow boxes in figure 2B and 2E indicate inset area and scale bars represent 10m

Mabs promote an oxidative environment in healthy airway organoids. (A) Heatmap depicting ROS-related genes in Mock-infected H-AO or H-AO infected with Mabs S or R for 4 days

Heatmap represents means from three pooled independent experiments

001 by unpaired T test. (B, C) Representative images (B) and MFI quantification (C) of mitochondrial ROS production (MitoSOX) in Mock-infected H-AOs (n=17) or H-AO

Wasabi-labelled Mabs S (n=13) or R (n=8) for 3 days. (D) MFI quantification of H2O2 production (H2DCFDA) in Mock-infected H-AOs (n=7) or H-AO

E-F) Bacterial load by CFU assay of H-AOs pre-treated with (+) or without (-) resveratrol for 1hr before infection with Mabs S (E) (n+= 7; n-=6) or R (F)

H) Bacterial load by CFU assay of H-AO pre-treated with (+) or without (-) sulforaphane for 6hr before infection with Mabs S (G) (n+= 8; n-=8) or R (H) (n+= 9; n-=9) for 4 days. Except otherwise stated, graphs represent means ± SD from at least two independent experiments, indicated by different symbols

A-B) Bacterial load by CFU assay of H-AO and CF-AO infected for 4 days with Mabs S (A) (n healthy=11; n cystic fibrosis=15) or R (B) (n healthy=10; n cystic fibrosis=13). (C, D) Representative images (C) and MFI quantification (D) of propidium iodide incorporation in Mock-infected H-AO and CF-AO

MFI quantification of mitochondrial ROS production (MitoSOX) in H-AO (n=10) and CF-AO (n+=12; n-=12) after 4 days of being treated with (+) or without (-) sulforaphane. (F-G) Bacterial load by CFU assay of CF-AO pre-treated with (+) or without (-)sulforaphane for 6 hr before infection with Mabs S (G)

Except otherwise stated, graphs represent means ± SD from at least two independent experiments, indicate them by different symbols

We thank Nicole Schieber (EMBL Heidelberg, Germany) for sharing with us the embedding protocol.We thank Veronique Richard and Franck Godiard from the "Microscopie Electronique et Analytique" service of the University of Montpellier for assistance in ultramicrotomy and TEM, respectively. We thank Bruno Payre from the "Centre de Microscopie Electronique pour la Biologie" of the University of Toulouse 3 for his assistance in SEM. This manuscript was edited at Life Science Editors.