COMMENT Vi e w p o i n t Understanding Bacillus anthracis pathogenesis Philip C. Hanna and John A.W. Ireland A nthrax has been both a scourge and an historically important model in understanding infectious diseases and phagocyte functions. It was the first experimental model for developing central postulates of infectious disease and the first to outline the role of macrophages in cellular immunity1–3. The disease is initiated by introduction of Bacillus anthracis endospores into the body4,5. An early report showed that endospores have a high affinity for the regional macrophages and are efficiently and rapidly phagocytosed in vivo6. In this guinea pig inhalation model, spore-containing macrophages detach from the lung and are carried toward the regional lymph nodes in the media stinum. It is during this period that germination occurs. There is no overt pulmonary infection during inhalation anthrax and the lungs remain clear of vegetative bacilli6. A recent study by Guidi-Rontani et al.7 using elegant, modern microscopic technologies shows these events at high resolution and in delightful detail. After spore germination and local multiplication within the macrophage (intracellular events), vegetative bacteria kill the macrophage and are released into the bloodstream, where they live as extracellular pathogens and reach high numbers (up to 108 bacteria per ml)5. B. anthracis grows rapidly as a Gram-positive rod containing two large virulence plasmids. The vegetative bacilli respond to host signals of physiological body temperature and CO2 levels in order to transcriptionally activate capsule and toxin genes8. The action of the lethal toxin, a zinc-metalloprotease, on other macrophages in the body releases proinflammatory cytokines re- sponsible for the sudden and fatal shock9–11. The first overt sign of the disease is often death itself, which can occur as early as 1–7 days post exposure5. Endospore germination The events occurring during the initial moments when bacterial pathogens first encounter the host are critical for successful establishment of infectious loci (Box 1). Establishment greatly influences both the severity and the ultimate outcome of the disease. Currently, the events leading to successful establishment are not completely understood for any bacterial pathogen. Inherent in the B. anthracis infectious cycle is an unique opportunity to develop a new model for detailed examination of these crucial events. The anthrax infectious contagion is the dormant endospore. Anthrax infections occur only when endospores enter the body from the external environment. There are no well-documented cases of vegetative anthrax bacilli occurring in a natural system and no natural examples of liveanimal-to-live-animal transmission of anthrax. The endospore has no measurable metabolism and, like endospores from other species, has little or no water, no ATP production, no macromolecular synthesis and no active enzymology. It is stable for decades, perhaps centuries. This implies a necessity for nearly synchronous de novo expression of genes vital for the initiation of infection, as P.C. Hanna* and J.A.W. Ireland are in the Depts of Microbiology and Immunology, Duke University Medical Center, PO Box 3020, Durham, NC 27710, USA. *tel: 11 919 681 6702, fax: 11 919 684 8735, e-mail: hanna@abacus.mc.duke.edu well as genes for vegetative growth. Host-specific germination represents a process that is both relevant to the in vivo situation and that can be exploited as a model to study immediate-early events in bacterial establishment. Until last month, only a handful of publications concerning hostinduced germination of B. anthracis endospores existed, and our best insights into this process came from the 1957 study described earlier in which whole guinea pigs were formaldehyde-fixed by perfusion at various times postchallenge followed by pathological examination6. Now, using immunofluorescence staining, confocal scanning laser microscopy and image cytometry analysis, Guidi-Rontani et al. have provided us with high-resolution images of germinating anthrax spores in murine alveolar macrophages7. This study clearly depicts endospore germination occurring in macrophage phagosomal compartments (Fig. 1) and shows that the anthrax toxin genes are expressed early after germination within the macrophages. Host-specific germination signals Research on the germination processes of the nonpathogens Bacillus subtilis and Bacillus megaterium supports the notion that specific chemical germinants in the microenvironment bind to and trigger a receptor in the endospore. These germinants are small molecules whose identities vary between species and, in some cases, between strains12,13. The most common, and wellresearched, Bacillus sp. germinant is L-alanine12,14. Binding of the germinant to its receptor causes loss of endospore refractility, swelling of the cortex, loss of 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. PII: S0966-842X(99)01507-3 TRENDS IN MICROBIOLOGY 180 VOL. 7 NO. 5 MAY 1999 COMMENT Anthrax challenges Recent world events have raised the specter of anthrax as a weapon of choice for bioterrorists. Documentation of the accidental release of aerosolized anthrax spores from a military microbiology facility in Sverdlovsk (in the former USSR), resulting in the death of many people, has added scientific credence to this threat15. Epidemic outbursts of inhalation anthrax can only be the result of a massive penetration of endospores into the atmosphere by human design5. Last summer, President Clinton announced that biowarfare countermeasures represent a national priority in the USA. The current FDA-approved anthrax vaccine is toxoid based, and its effectiveness to those exposed to aerosolized anthrax endospores (inhalation anthrax) has been questioned16,17. Most current research into new vaccines is also toxoid based16. Clearly, there is a need for new candidate antigens for vaccine development, especially those that act prior to expression of anthrax toxin in the body. Therapeutic drugs that inactivate B. anthracis targets vital for early steps in the infectious cycle would also represent a major supplement to traditional antibiotic intervention, especially in the case of infection with multidrug-resistant strains or TRENDS IN Box 1. Proposed stages of systemic anthrax infectious cycle Immediate-early (min–hours) Entry → phagocytosis into macrophage → endospore germination Early (min–hours) New metabolism → outgrowth → macrophage survival → multiplication and escape Middle (hours) Extracellular growth → expression of toxin and other virulence factors Late (days) Exotoxin → macrophage → cytokines → host symptoms and death Spread Nutrient depletion → sporulation → return to environment Trends in Microbiology resistance to damaging chemical and physical challenges, release of spore components and initiation of metabolism12. The genes for germination proteins are preformed in the spore (transcriptionally expressed during sporulation), and antibiotics that inhibit nucleic acid or protein synthesis do not block the germination process12. Triggering of the L-alanine receptor by its ligand is believed to activate its endogenous proteolytic activity, which then converts the proenzyme of a germinationspecific cortex-lytic enzyme (GSLE) to an active form. The active GSLE then, in an as yet undefined manner, allows cortex hydrolysis, uptake of water and all other downstream events12. The specific chemical germinants that are recognized by B. anthracis in the host remain to be discovered. Fig. 1. Laser scanning confocal micrographs showing phagocytosis of Bacillus anthracis in Balb/c alveolar macrophages. (a)–(c) Germinated spores studied 5 h after infection with an antibacillus serum and a fluorophore-conjugated secondary antibody. Scale bar 5 2 mm. (a) Germinated spores (pseudocolor green). (b) F-actin was detected with phalloidin (pseudocolor red). (c) In double overlays, germinated spores appear yellow as a result of the overlap of the green and red colors. (d)–(e) Alveolar macrophages 3 h after inhalation of spores. Cells were fixed, and double immunofluorescence staining was performed to detect germinated spores. Scale bar 5 5 mm. (d) Lysosomal compartments appear green. (e) Germinated spores (red) inside a phagolysosomal compartment appear yellow as a result of double overlays. Adapted with permission from Ref. 7. MICROBIOLOGY 181 VOL. 7 NO. 5 MAY 1999 COMMENT strains that have been engineered with additional toxins18. New knowledge about the germination systems of anthrax might provide important new antigens and potential targets for drug intervention. Acknowledgements Supported in part by the NIH (AI08649 and AI40644), the ACS (IRG158K) and funds from the Duke University Medical Center. References 1 Pasteur, L. (1881) C.R. Acad. Sci. Agric. Bulg. 92, 429–435 2 Koch, R. (1881) Mitt. Kaiserliche Gesundheitsamte 1, 174–206 3 Metchnikoff, E. (1905) Immunity to Infective Diseases, Cambridge University Press 4 Turnbull, P. (1986) Abstr. Hyg. Trop. Med. 61, R1–R13 5 Hanna, P. (1998) Curr. Top. Microbiol. Immunol. 225, 13–35 6 Ross, J.M. (1957) J. Pathol. Bacteriol. 73, 485–494 7 Guidi-Rontani, C. et al. (1999) Mol. Microbiol. 31, 9–17 8 Koehler, T.M., Dai, Z. and KaufmanYarbary, M. (1994) J. Bacteriol. 176, 586–595 9 Hammond, S. and Hanna, P. (1998) Infect. Immun. 66, 2374–2378 10 Duesbery, N. et al. (1998) Science 280, 734–736 11 Hanna, P., Acosta, D. and Collier, J. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 10198–10201 12 Moir, A. et al. (1994) J. Appl. Bacteriol. Symp. Suppl. 76, 9S–16S 13 Foster, S.J. and Johnstone, K. (1990) Mol. Microbiol. 4, 137–141 14 Hills, G.M. (1949) Biochem. J. 45, 363–370 15 Meselson, M. et al. (1994) Science 266, 1202–1208 16 Turnbull, P. (1992) Vaccine 9, 533–539 17 Hanna, P. (1998) Science 280, 1671 18 Pomerantsev, A. et al. (1997) Vaccine 15, 1846–1850 Evolution of Helicobacter pylori: the role of recombination Sebastian Suerbaum and Mark Achtman H elicobacter pylori is one of the most common pathogenic bacteria infecting humans. It is estimated that more than half of the human population is chronically colonized with this bacterium, which invariably causes a chronic inflammatory response of the stomach known as type B gastritis. Type B gastritis is asymptomatic in most patients but leads to ulceration of the stomach or the duodenum in ~10% of those infected. H. pylori infection is also a risk factor for malignant gastric diseases, including carcinoma and lymphoma of the mucosa-associated lymphoid tissue (MALT). H. pylori has long been known to exhibit considerable genetic diversity. Molecular typing methods, such as restriction patterns of genomic DNA, random amplified polymorphic DNA (RAPD), pulsed-field gel electrophoresis or nucleotide sequencing of single genes, show that almost all H. pylori isolates from unrelated patients are unique and no clonal relationships have been ob- served1–4. The origin of this diversity has been attributed to several mechanisms, including an elevated mutation rate, impaired DNA repair mechanisms and frequent recombination (for an extensive review, see Ref. 5). Recent molecular genetic, population genetic and genomic studies have quantified the extent of H. pylori diversity and elucidated some of the underlying mechanisms. Recombination during mixed infection with multiple H. pylori strains In a recent article in Mol. Microbiol.6, Dangeruta Kersulyte, Henrikas Chalkauskas and Douglas Berg presented a detailed analysis S. Suerbaum* is in the Ruhr-Universität Bochum, Abteilung für Medizinische Mikrobiologie, D-44780 Bochum, Germany; M. Achtman is in the Max-Planck-Institut für Molekulare Genetik, D-14195 Berlin, Germany. *tel: 149 234 7007887, fax: 149 234 7094197, e-mail: sebastian.suerbaum@ruhr-uni-bochum.de of six strains of H. pylori that were isolated from a single Lithuanian patient. RAPD PCR and restriction fragment length polymorphism (RFLP) analysis showed that the six strains belong to two different types: type A (one isolate) and type B (five isolates). The RAPD patterns of the type B isolates were similar but exhibited subtle differences, suggesting that genetic variants of the original strain had arisen during chronic infection. Kersulyte et al.6 further demonstrated that recombination had occurred between strains of RAPD types A and B on at least six occasions, thus accounting for the observed heterogeneity. Of particular significance are their observations regarding the cag pathogenicity island (PAI). The cag PAI is a ~37-kb chromosomal segment that is present in most H. pylori strains from patients with duodenal ulcers or MALT lymphoma7. The type A strain and two of the type B strains are cag negative, whereas the three other type B strains possess the cag PAI. Intra- 0966-842X/99/$ - see front matter © 1999 Elsevier Science. All rights reserved. PII: S0966-842X(99)01505-X TRENDS IN MICROBIOLOGY 182 VOL. 7 NO. 5 MAY 1999