In Vivo Germination of Bacillus anthracis Spores During Murine Cutaneous Infection

MAJOR ARTICLE In Vivo Germination of Bacillus anthracis Spores During Murine Cutaneous Infection Jean-Philippe Corre,1,2,3,a Alejandro Piris-Gimenez,1,3,a Marie Moya-Nilges,1,2,3 Gregory Jouvion,4 Agnes Fouet,1,3,5,6,7 Ian J. Glomski,1,3,12 Michèle Mock,1,2,3 Jean-Claude Sirard,1,3,8,9,10,11 and Pierre L. Goossens1,2,3 1 Toxines et Pathogénie Bactériennes, 2Pathogénie des Toxi-Infections Bactériennes, 3Centre National de la Recherche Scientifique, URA 2172, and Histopathologie Humaine et Modèles Animaux, Institut Pasteur, 5INSERM U-1016, Institut Cochin, Paris, 6CNRS UMR-8104, 7Université ParisDescartes Sorbonne Paris Cité, Paris, and 8Centre d’Infection et d’Immunité de Lille, Institut Pasteur de Lille, 9Université Lille Nord de France, 10Centre National de la Recherche Scientifique, UMR 8204, and 11Institut National de la Santé et de la Recherche Médicale, U1019, Lille, France; and 12 Department of Microbiology, Immunology, and Cancer Biology, University of Virginia, Charlottesville 4 Keywords. Bacillus anthracis; spore; germination; cutaneous infection; mice. Cutaneous anthrax is the most frequent form of the natural disease in humans. It has a lower mortality than inhalational and gastrointestinal forms [1–3], suggesting the existence of control mechanisms in cutaneous tissue. Resistance to cutaneous anthrax infection in some animal species has been correlated with the extent of local inflammatory reaction [4]. As anthrax occurs as an acute disease, innate immunity is the first line of defense that may control Bacillus Received 10 April 2012; accepted 28 August 2012; electronically published 12 November 2012. a J.-P. C. and A. P.-G. contributed equally to this study. Correspondence: Pierre L. Goossens, MD, PhD, Pathogénie des Toxi-Infections Bactériennes, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris cedex 15, France ( ). The Journal of Infectious Diseases 2013;207:450–7 © The Author 2012. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: . DOI: 10.1093/infdis/jis686 450 • JID 2013:207 (1 February) • Corre et al anthracis infection at an early stage in unimmunized hosts. Secretory type II-A phospholipase A2, an effector of the innate immune response, is highly bactericidal in vitro and in vivo against B. anthracis [5, 6]. Interferon γ (IFN-γ) helps cells and mice to resist anthrax infection [7–9]. Natural killer (NK) cells are the prime innate immune cells secreting IFN-γ at the onset of an infection; B. anthracis spores activate NK cells to secrete high levels of IFN-γ and to play a role in controlling the initial steps of infection [10]. Anthrax is a combination of toxemia and rapidly spreading infection that evolves into septicemia. The general perception is that B. anthracis spores are transported from the original site of inoculation to draining lymph nodes, where the spores germinate [11] and give rise to the toxin-producing encapsulated bacilli [2]. The lethal and edema toxins alter host cell signaling and modulate the host immune response [12, 13]. The poly-gamma-D-glutamate capsule [14, 15] is Background. Germination is a key step for successful Bacillus anthracis colonization and systemic dissemination. Few data are available on spore germination in vivo, and the necessity of spore and host cell interactions to initiate germination is unclear. Methods. To investigate the early interactions between B. anthracis spores and cutaneous tissue, spores were inoculated in an intraperitoneal cell-free device in guinea pigs or into the pinna of mice. Germination and bacterial growth were analyzed through colony-forming unit enumeration and electron microscopy. Results. In the guinea pig model, germination occurred in vivo in the absence of cell contact. Similarly, in the mouse ear, germination started within 15 minutes after inoculation, and germinating spores were found in the absence of surrounding cells. Germination was not observed in macrophage-rich draining lymph nodes, liver, and spleen. Edema and lethal toxin production were not required for germination, as a toxin-deficient strain was as effective as a Sterne-like strain. B. anthracis growth was locally controlled for 6 hours. Conclusions. Spore germination involving no cell interactions can occur in vivo, suggesting that diffusible germinants or other signals appear sufficient. Different host tissues display drastic differences in germinationtriggering capacity. Initial control of bacterial growth suggests a therapeutic means to exploit host innate defenses to hinder B. anthracis colonization. METHODS B. anthracis Strains and Mice Female outbred Swiss OF1 mice (weight, 22–24 g) and Hartley guinea pigs (weight, 200–250 g) were obtained from Charles River (L’Arbresle, France). Animals were housed in the animal facilities of Institut Pasteur, licensed by the French Ministry of Agriculture and in compliance with European regulations. All animal experiments were conducted according to European Union guidelines (http://ec.europa.eu/environment/chemicals/ lab_animals/home_en.html) and were approved by the animal care and use committee at the Institut Pasteur. The B. anthracis strains used were the encapsulated nontoxinogenic ΔpagA 9602P strain (median lethal dose, <25 spores subcutaneously) [29, 30], the 9602P-lux strain [28], the Sterne 7702 strain, and the Sterne-equivalent 9602R ( pXO1+, pXO2–) strain [31]. Infection Experiments Suspensions of purified spores in phosphate-buffered saline (PBS; 10 µL) were injected into the ear [28]. The inoculum size (mean inoculum [ ± SD], 4.03 ± 0.16 log10 CFU; n = 14) was verified retrospectively by plating 10-fold serial dilutions on brain heart infusion (BHI) agar plates. In vivo germination in the tissues was addressed through CFU counting. To accurately determine the extent of germination in the infected tissues and avoid germination triggering due to mechanical dissociation, for each experiment 2 groups of mice were inoculated. At various times after inoculation, the infected ear, the draining lymph node, the liver, the spleen, and blood were recovered. For one group of infected animals, tissues were immediately homogenized, and bacteria were counted to determine total CFU (ie, spores and bacilli). Tissues from the other group were heated immediately for 30 minutes at 65°C then homogenized before CFU determination—this reveals the number of heat-resistant CFU (ie, spores; hereafter termed “spore CFU”). The spore CFU is a measure of the spore load in the organs; any vegetative cells and spores having undergone germination that are present in the tissue are killed by the heat treatment, as confirmed by preliminary tests after infection with vegetative cells (data not shown). Groups of at least 6 animals were used to calculate the mean ± standard error of the mean (SEM) of the CFU in the 2 groups (nonheated and heated). Statistical comparisons of the CFU for the (...truncated)