Recent whooping cough (pertussis) outbreaks in many countries highlight the crucial need for a better understanding of the pathogenesis of Bordetella pertussis infection of the respiratory tract. The baboon is a recently described preclinical model for the study of B. pertussis infection and may be ideal for the evaluation of new pertussis vaccines. However, many pathophysiological aspects, including bacterial localization and interactions, have yet to be described in this model. Here, we used a baboon model of infection with a fluorescent GFP-expressing B. pertussis strain, derived from European clinical isolate B1917. Juvenile baboons were used to evaluate susceptibility to infection and transmission. Non-invasive in vivo imaging procedures, using probe-based confocal endomicroscopy coupled with bronchoscopy, were developed to track fluorescent bacterial localization and cellular interactions with host cells in the lower respiratory tract of infected animals. All B1917-GFP-challenged animals developed classical pertussis symptoms, including paroxysmal cough, nasopharyngeal colonization, and leukocytosis. In vivo co-localization with antigen presenting cells and progressive bacterial colonization of the lower airways were also assessed by imaging during the first weeks of infection. Our results demonstrate that in vivo imaging can be used to assess bacterial colonization and to point out interactions in a baboon model of pertussis.
Whooping cough, or pertussis, is mainly caused by Bordetella pertussis infection of the airways. There were 24.1 million estimated pertussis cases in young children and 160,700 associated deaths world-wide in 20141. The main symptoms observed are a prolonged characteristic paroxysmal cough, sometimes coupled with mild fever and leukocytosis. This disease affects mostly children and may be life-threatening in infants, in whom hyper-leukocytosis may lead to pulmonary hypertension and respiratory failure. High vaccine coverage globally with diphtheria-tetanus-pertussis vaccines2 (86% in 2016) has failed to control the disease, which is now re-emerging in several developed countries, including the USA, the UK, and Australia3. The reasons for the resurgence of pertussis are still a matter of debate and may include better detection due to enhanced surveillance and diagnostic tools, strain adaptation to escape vaccine-induced immunity, waning immunity, and asymptomatic carriage and transmission of the causative agent3.
A better understanding of the pathophysiology of whooping cough is required for the development of new pertussis control strategies. A non-human primate (NHP) model to study whooping cough has become available in recent years4. In contrast to other animal models, young baboons (seven to nine months old) display the considerable advantage of developing all clinical symptoms of pertussis disease following B. pertussis infection, including paroxysmal cough, nasopharyngeal colonization, and leukocytosis4. Moreover, this model has also allowed the study of the effects of pertussis vaccination in NHPs and a comparison of pre-existing commercial and candidate pertussis vaccines5,6,7,8. Apart from the study of infection in human, which has its limitations, this model appears to be the most appropriate to study bacterial localization and bacteria-host interactions during infection. Previous studies using murine9 and swine10 models of B. pertussis infection provided some evidence of bacterial presence in the nasopharynx and pulmonary bronchi. However, these models do not reproduce the full spectrum of pertussis symptoms and these observations were mainly established after post-mortem analysis. It is therefore difficult to make conclusions concerning bacterial colonization and physiologically relevant interactions during the course of the disease. Here, we aimed to evaluate the kinetics of B. pertussis colonization and to detect its interactions with antigen-presenting cells (APCs) in the lower respiratory tract of baboons by in vivo imaging with probe-based confocal laser endomicroscopy (pCLE) coupled with bronchoscopy. We thus developed and evaluated a juvenile baboon model of infection and transmission with fluorescent B. pertussis B1917-GFP, derived from the European clinical isolate B191711 genetically close to D420 US strain previously used in baboon models. Older baboons than the ones used in previously published model, but still juveniles and susceptible to B. pertussis infection, were chosen at first to implement this imaging technique as they better support bronchoscopy. This in vivo imaging procedure in B. pertussis-infected baboons will provide new insights concerning bacterial localization and co-localization with host cells to improve our understanding of pertussis pathogenesis.
Generation and characterization of B. pertussis B1917-GFP
The pBBPG vector is a pBBR1MCS12 plasmid derivative carrying the GFP-encoded gene under the control of the strong and constitutive B. pertussis BPSM porin gene promoter (Ppor) (Fig. 1A). The B1917 fluorescent derivative clone was analyzed by PCR to verify the genuity from the parental European B1917 strain; the sequences showed the fluorescent derivative to contain the ptxA1, prn-2, and fim3-2 alleles, like the B1917 and D420 strains.
Fluorescence of the bacteria freshly grown on chloramphenicol-containing BG plates was assessed by confocal microscopy. B. pertussis B1917-GFP showed a bright and intense signal after excitation with a 488-nm laser (Fig. 1B).
Previous studies indicated that transformation of B. pertussis with a GFP-encoding plasmid did not affect bacterial virulence factor production13. We compared the growth of B1917 and B1917-GFP in liquid culture to validate the use of B1917-GFP for in vivo infection studies. The growth rate of the GFP-producing bacteria was non-significantly (p = 0.09) lower by 0.22 + /−0.11 OD units during the exponential phase (Fig. 1C).
Infection of baboons
Young baboons have previously been shown to be a relevant model for B. pertussis infection and transmission4,14. Three animals (39–41 months) were exposed to 108 to 1010 CFU of B1917-GFP by intranasal and intra-tracheal routes to investigate the susceptibility of baboons to infection. All animals developed nasopharyngeal colonization and the clinical pertussis symptoms previously described, including leukocytosis and paroxysmal cough (Fig. 2). Nasopharyngeal colonization by the bacteria was detected from day 2 post-inoculation with a maximum of 1.0 × 107 CFU/mL at day 9 (Fig. 2A). We also observed the characteristic paroxysmal cough for all animals with a peak of 23.6 ± 10 coughs per hour during the second week of infection (Fig. 2B). The level of circulating white blood cells also progressively increased (Fig. 2C) with a peak of 14,9 ± 5 × 103 cells/µL (Fig. 2E, p = 0,04) two weeks post-challenge. Among these white blood cells we detected an increase trend of circulating lymphocytes (Fig. 2D,F) but without reaching statistical significance (p = 0,08). Overall, these results show that the B1917-GFP strain can colonize the upper respiratory tract of baboons and induce classical pertussis symptoms during infection. We obtained similar results when we compared exposure of 39-month-old baboons to B. pertussis D420 and B1917 (Supplementary Fig. 1).
B. pertussis transmission to naïve co-housed baboons
Airborne and direct transmission of B. pertussis between baboons has been previously demonstrated for young animals14. Here we performed two duplicate experiments to assess direct B1917-GFP transmission between juvenile animals. In each experiment, a 39-month-old naïve baboon (identified as K921A and K937B) was co-housed in the same cage unit with another animal of the same age challenged with B. pertussis B1917-GFP. Both co-housed naïve animals were colonized by B. pertussis in the nasopharynx with diverse kinetics. One animal, K921A, was briefly colonized for one week after 9 days of co-housing with the infected animal (Fig. 3A). The second baboon, K937B, was colonized from D21 post co-housing, but the bacteria were recovered from the nasopharynx for more than three weeks (Fig. 3A). Hence, some fluorescent bacteria were detected in the suspensions of nasopharyngeal swabs at D9 for K921A and D20 for K937B by confocal microscopy, confirming colonization of both baboons by the B1917-GFP strain carried by the co-housed infected baboon (Fig. 3C,D). Only K937B showed an increase of circulating WBCs from the beginning of nasopharyngeal colonization (Fig. 3B). These data indicate that juvenile baboons can transmit B1917-GFP from one animal to another.
Ex vivo imaging of baboon airways using pCLE
Ex vivo imaging was performed using baboon tissues to establish pCLE technique in airways. APC detection was assessed using this technique as they have been shown to play a key role in the early immune response against B. pertussis15,16,17,18. APCs were selectively labeled and detected in the lower respiratory tract, from the trachea (Fig. 4A,B) to the alveoli (Fig. 4C,D), with anti-HLA-DR-AF647 (Fig. 4A,C), whereas there was no detectable signal with the IgG2a-AF647 control (Fig. 4B,D). APC labeling specificity was confirmed by confocal visualization of frozen tissue slices (Supplementary Fig. 2). Overall, these results show that APCs can be selectively labeled and tracked in the lower respiratory tract of baboons by pCLE imaging.
In vivo imaging of B. pertussis colonization in infected baboons
We reproducibly performed in vivo bronchoscopy on 39-month-old baboons down to the secondary bronchi without affecting oximetry or cardiac rate. In vivo bronchoscopy coupled with pCLE imaging was performed on naïve and B. pertussis B1917-GFP-infected baboons (Fig. 5A–G). Previously published data on human airway pCLE imaging reported a significant signal in the green channel due to endogenous auto-fluorescence of elastin fibers, located behind the epithelial barrier19. Surprisingly, there was almost no auto-fluorescence in the green channel at the tracheal and bronchial surface in the naïve animals prior to challenge (Fig. 5H).
We detected co-localized GFP and HLA-DR-AF647 signals for all infected animals at D2 post-challenge (representative images for one animal in Fig. 5A–F), confirming the ex vivo co-culture findings (Fig. 6A). The GFP signal was detectable in the trachea of all B. pertussis B1917-GFP-infected animals from D2 (Fig. 5A,C) and increased up to D14 (Fig. 5D,F), indicating progressive bacterial colonization of the lower respiratory tract during the early phase of infection (Fig. 5H). However, there was no GFP signal after D21 in challenged or non-challenged co-housed animals, despite B. pertussis-positive nasopharyngeal cultures (Fig. 3), most likely due to the progressive loss of the bacterial fluorescence over time (Supplementary Fig. 3).
We also detected B. pertussis bacteria post mortem by IHF in the trachea and bronchi of infected baboons five weeks post-inoculation (Fig. 6B), whereas no signal was observed with the isotype control (Fig. 6C) nor in naïve animals (Fig. 6D).
Recent pertussis outbreaks in many countries with high vaccination coverage have highlighted the crucial need to improve current vaccination strategies. This may be aided by better knowledge concerning the pathogenicity of pertussis disease. In vitro and preclinical animal studies, mostly in mice, have revealed diverse aspects of pertussis pathogenicity, immunomodulatory effects of bacterial virulence factors and protective immune responses to B. pertussis infection16,17,18,20,21,22,23. However, these models do not reproduce the full spectrum of pertussis symptoms, most notably the cough. The recently developed baboon model of whooping cough has made it possible to study the modalities of pertussis transmission, providing strong support for the asymptomatic carriage hypothesis, as well as evaluating the efficiency of pertussis vaccines. A better understanding of how B. pertussis interacts with this host and a comparison with other models or in vitro data will help to build a strong basis for future vaccine development.
Here, we developed and applied in vivo imaging technologies to study bacterial localization and to evaluate bacterial-host cell interactions in B. pertussis-infected baboons. We have successfully constructed a GFP-producing B. pertussis strain from clinical isolate B1917, belonging to the ptxP3-ptxA1 clade, as a representative of currently circulating strains of B. pertussis11. Infection of juvenile baboons with this fluorescent strain induced classical pertussis symptoms. The kinetics of bacterial colonization in the upper airways was comparable to previous published data for younger baboons infected with the B. pertussis D420 strain24. We also observed characteristic paroxysmal cough, especially in the first two weeks of infection, but with residual cough for up to one month. The intensities of the general symptoms appear to be milder in B1917-GFP-infected 39-month-old baboons than in young juveniles challenged with B. pertussis D4204,5,6. This observation is consistent with the reduced and more variable clinical illness observed in adult humans relative to infants25. An investigation of the influence of strain or GFP synthesis on clinical pertussis symptoms in comparison with the infection profile of 39-month-old baboons revealed no major differences in the clinical symptoms induced by inoculated with B1917-GFP, B1917 or D420 strains, indicating that GFP production did not significantly affect clinical symptoms. We used the GFP derivative strain to investigate natural B. pertussis transmission to naïve co-housed baboons and confirmed previous published data from younger animals14. However, the efficiency of transmission to naïve baboons and the intensity of associated clinical symptoms may vary due to social interactions and inter-individual variability. Nevertheless, effective clinical infection of the challenged animals validated B1917-GFP as an appropriate strain for the study of pertussis pathogenesis by imaging. This allowed us to investigate bacterial-host cell interactions using ex vivo and in vivo imaging technologies.
We used this baboon infection model to investigate whether cell populations of interest, such as APCs, and bacteria could be labeled and detected in vivo during a B. pertussis infection by pCLE imaging. The imaging technologies used here are well described for the study of human airways, but have not yet been extensively used in preclinical models without invasive surgery or euthanasia26. Unexpectedly, only low auto-fluorescence was observed in baboons relative to human airways, where elastin fibers induce strong background fluorescence. This difference may be explained by species specificities26 or the young age of the animals used here. We were therefore able to specifically detect progressive bacterial colonization of the trachea in infected animals during the first two weeks of infection by pCLE, as well as co-localization of the pathogen with APCs. This provides in vivo confirmation of the strong interactions previously described between B. pertussis and macrophages and dendritic cells of the airways27 but so far, precise bacterial localization, extra or intra-cellular, still need to be confirmed. APC labeling by topical application of fluorescent anti-HLA-DR antibodies has been successfully implemented in the tracheal and bronchial mucosa in baboons, but had to be repeated every week for optimal detection, as described for a skin model28. This is the first description of non-invasive in vivo bacterial detection in airways by optical imaging in NHP. A study in rats has described lung colonization by Aspergillus fumigatus using external detection by pCLE, but this required invasive surgery and was performed only at day 5 post-challenge26,29. The spatial resolution of pCLE probes used in this study, with a lateral resolution of 3.5 μm, prevented us from detecting single bacteria, but we could detect aggregates or biofilms as already demonstrated by confocal microscopy in mice9. Another limitation to bacterial infection imaging in our study was the gradual loss of plasmid-mediated fluorescence in the absence of selective pressure. This led to the progressive loss of fluorescence in vivo, which prevented us from following bacterial colonization by imaging for more than two weeks. However, long-term B. pertussis colonization may be studied in baboons using our imaging technologies with a B. pertussis strain that strongly and stably expresses a chromosomally inserted gfp gene. This pathophysiological imaging study in juvenile baboons demonstrates that it is possible to follow bacterial colonization of the lower respiratory tract and explore host-pathogen interactions during infection with minimal invasiveness. Our results open many new opportunities to characterize colonization by B. pertussis and the cellular populations involved in the local in vivo response to B. pertussis infection in detail. This should lead to an improved understanding of the pathogenesis of whooping cough in primates.
Materials and Methods
All in vivo studies were performed on juvenile Papio Anubis (39–41 months). Baboons were obtained from the Station de Primatologie (CNRS, Rousset-sur-Arc, France) and housed in the facilities of the Infectious Disease Models and Innovative Therapies (IDMIT) center at the “Commissariat à l’Energie Atomique et aux Energies Alternatives” (CEA, Fontenay-aux-Roses, France). The animals were used in accordance with French national regulations (CEA Authorization Number A92-032-02). The CEA complies with the Standards for Humane Care and Use of Laboratory Animals of the Office for Laboratory Animal Welfare (OLAW, USA) under OLAW Assurance number #A5826-01. The use of NHPs at the CEA complies with the recommendations of European Directive 2010/63 (recommendation N°9). These experiments were approved by the ethics committee number 44 of the French Ministry of Research (statement number A16-041). Experimental procedures (animal handling, bacterial inoculation, and blood sampling) were conducted after animal sedation with ketamine chlorydrate (Rhône-Mérieux, France, 10 mg/kg) or Zoletil®100 (Virbac, France, 5 mg/kg). Tissues were collected at necropsy: animals were sedated with ketamine and then euthanized by intravenous injection of 180 mg/kg sodium pentobarbital.
Bacterial strains and growth media
B. pertussis clinical isolates B1917 and D42024 were provided by Frits Mooi and came from the RIVM collection (Bilthoven, The Netherlands). The B1917 strain was transformed with pBBPG30, encoding green fluorescent protein (GFP) (Fig. 1A). Bacterial fluorescence was verified on slices using the Nikon-A1R confocal laser-scanning microscope attached to an inverted ECLIPSE-Ti (Nikon, Japan).
B. pertussis strains were grown in Bordet-Gengou agar plates containing 10% (v/v) sheep blood (BG plates, BD) for 72 h at 37 °C. For growth studies, bacteria were suspended in modified Stainer-Scholte medium supplemented with cyclodextrin (1 g/L) and casamino acids (5 g/L) without any antibiotics31 at an OD600nm of 0.15. Liquid cultures were incubated at 37 °C and 320 rpm in disposable 15-mL flasks. Optical density OD600nm was measured twice a day to evaluate bacterial growth. Cultures were performed in four replicates.
Preparation of inoculum and infection of baboons
B. pertussis colonies grown on BG plates for 72 h at 37 °C were resuspended in sterile phosphate-buffered saline (PBS) at an OD600nm of 0.9, which corresponds to a density of 108 to 109 CFU/mL. Baboons were infected by intranasal and intra-tracheal routes as previously described4,32. The inoculum concentration was validated by serial dilutions and plating on BG plates and colony counting after seven days of incubation at 37 °C. Automatic colony counting was performed using a Scan®300 colony counter (Interscience, France).
Bacterial colonization and monitoring of infection
Bacterial content in the nasopharynx was measured twice weekly using flocked swabs in each naris of animals. Samples were then collected in liquid Amies medium (COPAN, USA). This solution (50 μL) was plated in duplicate at various dilutions on BG plates. The number of colony-forming units (CFU) per plate was estimated by automatic counting after seven days of incubation. Whole blood was evaluated for the number of circulating white blood cells and lymphocytes by a complete blood count. Baboons were euthanized six weeks after challenge.
Quantification of coughing
Animal cages were monitored with a video-recording system. Cough occurrences for each animal were assessed daily during a 6-h period (6:00–9:00 a.m.; 5:00–8:00 p.m.) using Windows Movie Maker software.
Mouse anti-human HLA-DR antibody that targets APCs (BD, clone L243) and its isotype control (mouse IgG2a (BD)) were labeled with AlexaFluor647 using the Microscale Protein Labeling Kit (Invitrogen). Antibody and fluorophore concentrations were measured after coupling by nanodrop spectrophotometry.
Ex vivo imaging of the baboon respiratory tract
Lower respiratory tract explants were obtained from euthanized baboons. Immediately after euthanasia, lung lobes were isolated and tracheal rings cut and preserved in complete medium (DMEM containing 10% SVF, NEAA (1X concentration), 1 mM sodium pyruvate, and 0.01 M HEPES) at 37 °C. A solution of anti-HLA-DR or IgG2a mouse antibody labeled with AlexaFluor647 (200 µL, 100 µg/mL) was dropped into the bronchus of a lobe and the sample incubated for 2 h at 37 °C. Tracheal rings were submerged in the same antibody solution (200 µL, 100 µg/mL) for 2 h at 37 °C. Then, a solution of acriflavine (Sigma Aldrich; 0.05 mg/mL; 1 mL) was topically administrated to the main bronchus of each lobe and tracheal rings submerged in the same solution for 5 min before imaging. Such acriflavine RNA staining allows detection of all cellular structures at 488 nm. No rinsing was performed to avoid modification of the alveolar structure and cell composition due to the lavage. Probe-based Confocal Laser Endomicroscopy (pCLE) was performed using the Cellvizio DualBand® device (MaunaKea Technologies, France) with S-1500 or miniZ probes. For tracheal ring imaging, probes were deposited directly on the tracheal epithelial surface. For lung lobe imaging, probes were inserted into the main bronchus and directed down to the alveolar structures. For ex vivo co-culture experiments with B1917-GFP, acriflavine was replaced by the bacterial inoculum (1 mL, 108 CFU) and samples were incubated 48 h at 37 °C.
In vivo pCLE imaging of the baboon respiratory tract
Xylocaine (5%; Aspen Pharma, France) was topically sprayed into the back of the mouth of anaesthetized baboons. Bronchoscopy was performed using a Karl Storz Special fibroscope. Anti-HLA-DR-AF647 antibodies (200 µL, 100 µg/mL) were first topically dropped at the top of the trachea and into the left main bronchus via the working channel of the endoscope to label the APCs. After 2 h of incubation, pCLE with bronchoscopy was performed as previously described19. Briefly, the optical miniZ probe was inserted into the working channel of the bronchoscope and imaging sequences performed from the top of the trachea down to the secondary bronchi, especially in the area where the antibodies were dropped. APCs were detected at 660 nm and B. pertussis B1917-GFP at 488 nm. Signal background in these two channels was assessed before baboon challenge, in the absence of fluorescent bacteria and antibodies. In vivo imaging sequences were performed once a week for four weeks starting seven days prior to animal infection with B. pertussis.
Tissues (lungs, bronchi, and trachea) were first fixed with 4% PFA for 24 h at 4 °C. Samples were then dehydrated in 30% sucrose for 24 h, embedded in OCT, and finally frozen in isopentane and liquid nitrogen. B. pertussis was detected in 10-µm tissue sections using anti-B. pertussis-LOS-A (ThermoFisher, clone D26E) and goat anti-mouse IgG3-AlexaFluor594 (Life Technologies) antibodies. APCs were detected using anti-human-HLA-DR and goat anti-mouse IgG2a-AlexaFluor488 (Life Technologies) antibodies. Cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Image acquisition was performed using a SP8 confocal microscope (Leica, Germany).
Films obtained using pCLE were first selected using IC-Viewer software and then exported to ImageJ (National Institute of Mental Health, USA). Intensity thresholds were fixed to minimize the background to signal ratio. Bacterial colonization of the lower airways was assessed using GFP area quantification. Briefly, the total area of GFP-positive signal was calculated for each imaging time-point using Visiopharm® Software (Visiopharm, Denmark) in 30 images randomly extracted from the pCLE acquisition. No signal size restrictions were applied due to the small size of the bacteria. For post-mortem tissue analysis, acquisitions performed on the SP8 confocal microscope were analyzed using ImageJ.
Data are reported as the means ± standard errors for replicate experiments. For bacterial growth experiments and colonization assessment by pCLE, unpaired t-tests were conducted to detect statistical differences (p < 0.05). Paired t-tests were also conducted to compare levels of leukocyte and lymphocyte counts before challenge and at the peak of infection (GraphPad Prism software).
All data generated or analyzed during this study are available from the corresponding author on reasonable request.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yeung, K. H. T., Duclos, P., Nelson, E. A. S. & Hutubessy, R. C. W. An update of the global burden of pertussis in children younger than 5 years: a modelling study. Lancet Infect. Dis. 17, 974–980 (2017).
WHO | Diphtheria-tetanus-pertussis (DTP3) immunization coverage. WHO Available at: http://www.who.int/gho/immunization/dtp3/en/. (Accessed: 22nd January 2018).
Locht, C. Live pertussis vaccines: will they protect against carriage and spread of pertussis? Clin. Microbiol. Infect. 22(Supplement 5), S96–S102 (2016).
Warfel, J. M., Beren, J., Kelly, V. K., Lee, G. & Merkel, T. J. Nonhuman Primate Model of Pertussis. Infect. Immun. 80, 1530–1536 (2012).
Warfel, J. M., Papin, J. F., Wolf, R. F., Zimmerman, L. I. & Merkel, T. J. Maternal and Neonatal Vaccination Protects Newborn Baboons From Pertussis Infection. J. Infect. Dis. 210, 604–610 (2014).
Warfel, J. M., Zimmerman, L. I. & Merkel, T. J. Acellular pertussis vaccines protect against disease but fail to prevent infection and transmission in a nonhuman primate model. Proc. Natl. Acad. Sci. USA 111, 787–792 (2014).
Warfel, J. M., Zimmerman, L. I. & Merkel, T. J. Comparison of Three Whole-Cell Pertussis Vaccines in the Baboon Model of Pertussis. Clin. Vaccine Immunol. CVI 23, 47–54 (2015).
Locht, C. et al. Live Attenuated Pertussis Vaccine BPZE1 Protects Baboons Against Bordetella pertussis Disease and Infection. J. Infect. Dis. 216, 117–124 (2017).
Serra, D. O. et al. FHA-Mediated Cell-Substrate and Cell-Cell Adhesions Are Critical for Bordetella pertussis Biofilm Formation on Abiotic Surfaces and in the Mouse Nose and the Trachea. PLOS ONE 6, e28811 (2011).
Elahi, S. et al. Infection of Newborn Piglets with Bordetella pertussis: a New Model for Pertussis. Infect. Immun. 73, 3636–3645 (2005).
Bart, M. J. et al. Complete Genome Sequences of Bordetella pertussis Isolates B1917 and B1920, Representing Two Predominant Global Lineages. Genome Announc. 2 (2014).
Kovach, M. E., Phillips, R. W., Elzer, P. H., Roop, R. M. & Peterson, K. M. pBBR1MCS: a broad-host-range cloning vector. BioTechniques 16, 800–802 (1994).
Weingart, C. L. et al. Fluorescent labels influence phagocytosis of Bordetella pertussis by human neutrophils. Infect. Immun. 67, 4264–4267 (1999).
Warfel, J. M., Beren, J. & Merkel, T. J. Airborne transmission of Bordetella pertussis. J. Infect. Dis. 206, 902–906 (2012).
Brummelman, J., Wilk, M. M., Han, W. G. H., van Els, C. A. C. M. & Mills, K. H. G. Roads to the development of improved pertussis vaccines paved by immunology. Pathog. Dis. 73, ftv067 (2015).
Locht, C., Coutte, L. & Mielcarek, N. The ins and outs of pertussis toxin. FEBS J. 278, 4668–4682 (2011).
Masin, J., Osicka, R., Bumba, L. & Sebo, P. Bordetella adenylate cyclase toxin: a unique combination of a pore-forming moiety with a cell-invading adenylate cyclase enzyme. Pathog. Dis. 73, ftv075 (2015).
Higgs, R., Higgins, S. C., Ross, P. J. & Mills, K. H. G. Immunity to the respiratory pathogen Bordetella pertussis. Mucosal Immunol. 5, 485–500 (2012).
Thiberville, L. et al. In Vivo Imaging of the Bronchial Wall Microstructure Using Fibered Confocal Fluorescence Microscopy. Am. J. Respir. Crit. Care Med. 175, 22–31 (2007).
Raeven, R. H. M. et al. Immunological Signatures after Bordetella pertussis Infection Demonstrate Importance of Pulmonary Innate Immune Cells. PLOS ONE 11, e0164027 (2016).
Raeven, R. H. M. et al. Molecular Signatures of the Evolving Immune Response in Mice following a Bordetella pertussis Infection. PLOS ONE 9, e104548 (2014).
Cerny, O., Anderson, K. E., Stephens, L. R., Hawkins, P. T. & Sebo, P. cAMP Signaling of Adenylate Cyclase Toxin Blocks the Oxidative Burst of Neutrophils through Epac-Mediated Inhibition of Phospholipase C Activity. J. Immunol. 198, 1285–1296 (2017).
Kamanova, J. et al. Adenylate Cyclase Toxin Subverts Phagocyte Function by RhoA Inhibition and Unproductive Ruffling. J. Immunol. 181, 5587–5597 (2008).
Boinett, C. J. et al. Complete Genome Sequence of Bordetella pertussis D420. Genome Announc. 3 (2015).
Cherry, J. D. Epidemiological, clinical, and laboratory aspects of pertussis in adults. Clin. Infect. Dis. Off. Publ. Infect. Dis. Soc. Am. 28(Suppl 2), S112–117 (1999).
Morisse, H. et al. In vivo and in situ imaging of experimental invasive pulmonary aspergillosis using fibered confocal fluorescence microscopy. Med. Mycol. 50, 386–395 (2012).
Lamberti, Y., Gorgojo, J., Massillo, C. & Rodriguez, M. E. Bordetella pertussis entry into respiratory epithelial cells and intracellular survival. Pathog. Dis. 69, 194–204 (2013).
Salabert, N. et al. Intradermal injection of an anti-Langerin-HIVGag fusion vaccine targets epidermal Langerhans cells in nonhuman primates and can be tracked in vivo. Eur. J. Immunol. 46, 689–700 (2016).
Morisse, H. et al. In vivo molecular microimaging of pulmonary aspergillosis. Med. Mycol. 51, 352–360 (2013).
Hibrand-Saint Oyant, L. et al. Role of Bordetella bronchiseptica adenylate cyclase in nasal colonization and in development of local and systemic immune responses in piglets. Vet. Res. 36, 63–77 (2005).
Imaizumi, A., Suzuki, Y., Ono, S., Sato, H. & Sato, Y. Effect of heptakis (2,6-O-dimethyl) beta-cyclodextrin on the production of pertussis toxin by Bordetella pertussis. Infect. Immun. 41, 1138–1143 (1983).
Merkel, T. J. & Halperin, S. A. Nonhuman primate and human challenge models of pertussis. J. Infect. Dis. 209(Suppl 1), S20–23 (2014).
We warmly thank the IDMIT infrastructure staff for excellent technical assistance, especially Laetitia Bossevot, Naya Sylla and Sebastien Langlois. The IDMIT facility is supported by the French government “Programme d’Investissements d’Avenir” (PIA), under grants ANR-11-INBS-0008 and ANR-10-EQPX-02-01. This work was supported by PERISCOPE. PERISCOPE has received funding from the Innovative Medicines Initiative 2 Joint Undertaking, under grant agreement no. 115910. This Joint Undertaking received support from the European Union’s Horizon 2020 research and innovation Program and European Federation of Pharmaceutical Industries and Associations (EFPIA) and Bill and Melinda Gates Foundation (BMGF).