living dangerously: how helicobacter pylori survives in the ... - Nature

REVIEWS

LIVING DANGEROUSLY: HOW HELICOBACTER PYLORI SURVIVES IN THE HUMAN STOMACH Cesare Montecucco and Rino Rappuoli Helicobacter pylori was already present in the stomach of primitive humans as they left Africa and spread through the world. Today, it still chronically infects more than 50% of the human population, causing, in some cases, severe diseases such as peptic ulcers and stomach cancer. To succeed in these long-term associations, H. pylori has developed a unique set of virulence factors, which allow survival in a unique and hostile ecological niche — the human stomach.

GRAM-NEGATIVE BACTERIA

Bacteria whose cell walls do not retain a basic blue dye during the Gram-stain procedure. These cell walls are thin, consisting of a layer of lipopolysaccharide outside a peptidoglycan layer. ADENOCARCINOMA

Cancer originating from uncontrolled proliferation of epithelial cells of the ducts and acini of glandular organs. LYMPHOMA

Cancer originating from uncontrolled proliferation of lymphocytes.

Centro CNR Biomembrane e Dipartimento di Scienze Biomediche, Università di Padova, Via G. Colombo 3, 35121 Padova, Italy and Centro di Ricerche IRIS, Chiron S.p.A., Via Fiorentina, 1, 53100 Siena, Italy Correspondence to C.M. e.mail: [email protected]

Helicobacter pylori is a GRAM-NEGATIVE BACTERIUM, specialized in the colonization of the human stomach1, a unique ecological niche characterized by very acidic pH — a condition lethal for most microbes. H. pylori is so well adapted to this unfriendly environment that, after the first infection, which usually occurs early in life, it establishes a life-long chronic infection2. The selection of a niche with no competition and the ability to establish a chronic infection make H. pylori one of the most successful human bacterial parasites, which colonizes more than half of the human population3. The comparison of H. pylori isolated from patients of different ethnical origin and geographical locations indicates that their nucleotide sequences segregate similarly to those of humans. This suggests that this bacterium was already present in the stomach of humans when they left Africa to colonize the world, and co-evolved with them since then4. Hence, it is likely that primitive humans suffered from stomach aches, although the first written reports were made by ancient Greek doctors and gastric ulcers were first described5 in 1586. Most infected people are asymptomatic, with moderate inflammation detectable only by biopsy and histology. However, an important minority of them (15–20%) during their life develop severe gastroduodenal pathologies, including stomach and duodenal ulcers, ADENOCARCINOMAS and stomach LYMPHOMAS. The different outcomes of

the infection are believed to be substantially influenced by an excessive or inappropriate reaction of the host, by bacterial polymorphisms and by environmental factors6. The successful life-lasting colonization of the human stomach by H. pylori is achieved through a combination of factors, which address the different challenges presented by the harsh environment (FIG. 1). H. pylori synthesizes a urease to buffer the pH of its immediate surroundings within the stomach. Its helicoidal shape and the action of flagella allow it to cross the thick layer of MUCUS lining the stomach. H. pylori then binds to LEWIS ANTIGENS present on host gastric cells, and it secretes factors that attract and stimulate inflammatory cells, as well as the multifunctional toxin VacA. Last, the presence of the cag pathogenicity island, a 40-kb DNA that encodes a type IV secretion system, seems to be necessary for optimal fitness of the bacterium and the appearance of pathogenic traits (BOX 1). Buffering the pH

To colonize the gastroenteric tract, a bacterium has to survive in the acid lumen of the stomach7. H. pylori dedicates several genes to the biosynthesis of a cytosolic urease, a Ni2+-containing enzyme8,9, which hydrolyses urea into NH3 and CO2 (FIG. 2). The biosynthesis of urease is governed by a seven-gene cluster, including the genes encoding the UreA (26.5 kDa) and UreB (60.3 kDa) subunits of the urease and the accessory proteins that are

NATURE REVIEWS | MOLECUL AR CELL BIOLOGY

VOLUME 2 | JUNE 2001 | 4 5 7

© 2001 Macmillan Magazines Ltd

REVIEWS responsible for Ni2+ uptake and insertion into the active site of the apo-enzyme2,9. Urease is a dodecamer of six UreA and six UreB subunits, arranged as a double ring of 13-nm diameter10. The amount of urease produced by the bacterium varies with culture conditions and may reach as much as 10% of total bacterial protein11. Urea is taken up by H. pylori through a protongated channel12, and its hydrolysis by urease generates ammonia that buffers the cytosol and PERIPLASM, and creates a neutral layer around the bacterial surface (FIG. 2). In vitro, dead H. pylori cells release urease, which binds strongly to the surface of living cells13, and such a mode of ‘surface expression’ might also operate in

MUCUS

Slimy substance secreted by mucous cells of the mucosae that consists predominantly of mucins — highly glycosylated proteins of high molecular weight. The function of mucus is to protect the linings of body cavities.

Decreasing pH

H. pylori

Gastric lumen

Mucus layer Urease

VacA HPNAP

Pedestal CagA H+

Epithelial cells

IL-8

ROIs Neutrophil or monocyte

Neutrophil or monocyte

Blood vessel

Figure 1 | Schematic representation of the stomach mucosa colonized by Helicobacter pylori, showing the main virulence factors involved in colonization and disease. During infection, the bacterium enters the gastric lumen where the urease allows survival in the acidic environment (red indicates a strongly acidic environment, yellow indicates a mildly acidic one) by producing ammonia molecules that buffer cytosolic and periplasmic pH as well as the surface layer around the bacterium (light blue). The flagella propel the helicoidal bacterium into the mucus layer and allow it to reach the apical domain of gastric epithelial cells, to which it sticks using specialized adhesins. H. pylori then injects the cagA protein into the host cells by a type IV secretion system and releases other toxic factors such as H. pylori neutrophil-activating protein (HP-NAP) and VacA. VacA induces alterations of tight junctions and the formation of large vacuoles. Vacuoles are evident in cells in culture and in the stomach epithelial cells of human and mouse biopsies, although they are not apparent in gerbils. The neutrophilactivating protein HP-NAP crosses the epithelial lining and recruits neutrophils and monocytes, which extravasate and cause tissue damage by releasing reactive oxygen intermediates (ROIs). Injected Cag proteins cause alteration of the cytoskeleton, pedestal formation and signal the nucleus to release proinflammatory lymphokines, which amplify the inflammatory reaction with recruitment of lymphocytes and further induce the release of ROIs. The combined toxic activity of VacA and of ROIs leads to tissue damage that is enhanced by loosening of the protective mucus layer and acid permeation.

458

vivo as urease is found on the surface of H. pylori present in human biopsies14. The essential role of urease as a virulence factor is shown by the fact that urease-defective H. pylori mutants cannot colonize the stomach15. However, urea is toxic to the bacterium at neutral pH because an unfavourable alkaline environment is generated16. Therefore, the urea channel is regulated positively by protons, opening at acidic pH values to allow more urea in to buffer cytosolic and surface pH, and closing at neutral pH to avoid over-alkalinization. H. pylori urease might cause damage to the host cells through the production of ammonia, an agent known to be toxic by itself and/or in conjunction with NEUTROPHIL metabolites on cultured cells and on stomach tissue preparations17–19. Membrane-permeant weak bases cause various cell alterations, including swelling of intracellular acidic compartments, alterations of vesicular membrane transport, depression of protein synthesis and ATP production, and cell-cycle arrest. Ammonia can also react with intermediates released by the activity of neutrophil MYELOPEROXIDASE to form carcinogenic agents that might participate in the H. pylori-associated development of stomach adenocarcinoma18. However, ammonia is toxic at high doses, and the concentration dependence of the toxic effects varies considerably with cell type and culture conditions; therefore, it is not immediately evident how toxic the quantities of ammonia are that H. pylori produces in vivo. Moreover, circulation of extracellular fluids should lead to a rapid dilution of the produced ammonia with a marked concentration gradient from the bacterial surface to the medium that surrounds host cells. The role of urease in the pathogenesis of H. pyloriassociated diseases is not limited to colonization as ammonia produced by the urease enters the H. pylori nitrogen metabolism and is eventually incorporated into proteins20. Urease might also help to recruit neutrophils and MONOCYTES in the inflamed mucosa and to activate production of PROINFLAMMATORY CYTOKINES21. Moreover, urease is one of the main antigens recognized by the human immune response to H. pylori 22, although the extent and the nature of this immune response after infection are not fully clear23. Swimming through the mucus

Although well equipped to survive in strong acid, H. pylori is not an acidophile and needs to leave the lumen, also to avoid discharge in the intestine. With its polarsheathed flagella, H. pylori is a good swimmer and reaches the thick mucus layer that covers and protects the epithelial lining of the stomach mucosa. Here, propelled by its flagella, the helicoidal-shaped bacterium travels across the viscous mucus film like a screw into a cork; non-motile mutants cannot colonize the stomach24. The movement of H. pylori towards the stomach mucosa is guided by chemotactic factors, which include urea and bicarbonate ions24,25. The mucus film is produced by mucous cells that secrete granules containing highly glycosylated proteins26. The secreted glycoproteins, together with an www.nature.com/reviews/molcellbio

| JUNE 2001 | VOLUME 2


REVIEWS

LEWIS ANTIGEN

A system of soluble antigens in secretions and plasma that represents one of the serologically distinguishable human blood-group substances. Lewis specificities are carried on glycosphingolipids and glycoproteins. PERIPLASM

Space between the outer and the inner membranes of Gramnegative bacteria. NEUTROPHIL

A phagocytic cell of the myeloid lineage that has an important role in the inflammatory response, undergoing chemotaxis towards sites of infection or wounding. MYELOPEROXIDASE

Peroxidase from neutrophils that takes part in the bactericidal activity of these cells. The name originates from the first isolation from the blood of patients with myeloid leukaemia. MONOCYTE

Large leukocytes with a horseshoe-shaped nucleus. They derive from pluripotent stem cells and become phagocytic macrophages when they enter the tissues. PROINFLAMMATORY CYTOKINES

Secreted proteins with autocrine or paracrine action that regulate the inflammatory response. There are many types of cytokine, which elicit different cellular responses including control of cell proliferation and differentiation, regulation of immune responses and haematopoiesis. EPITHELIAL CELLS

Polarized cells that cover the outer surfaces of the body and line internal cavities or tubes (except blood vessels). OXYNTIC CELLS

Cells of the gastric mucosa that secrete hydrochloric acid. PERISTALSIS

A wave-like sequence of contraction and relaxation that passes along a tube-like structure, resulting in a net forward movement of the content. LECTIN

Agglutinins and other antibodylike proteins of nonimmune origin that bind sugars.

Box 1 | A speculative model of the evolution and transmission of Helicobacter pylori Clinical isolates of H. pylori are classified as cag+ or cag –, vacA+ or vacA–, baba+ or babA–, depending on whether they have functional genes coding for the cag PAI, VacA and BabA, respectively. Whereas most babA– isolates have a defective gene, and vacA– are actually defective in either gene or toxin expression, in most instances cag– organisms have a complete deletion of the 40 kb of DNA. In fact, the island is excised with low frequency from the chromosome of cag+ strains generating cag– strains. A model, which might explain the evolution and transmission of H. pylori, assumes that cag+, vacA+, babA+ H. pylori strains represent a ‘fitness peak’, and that inactivation of any of these functions decreases bacterial fitness (see the figure). As a consequence, during the many years the bacterium lives in the stomach of each individual, it continuously generates cag – or vacA– or babA–. The defective derivatives multiply, and in some instances may also outgrow wild-type (wt) bacteria, and therefore both cag+ and cag – can be isolated from the same individual. However, these defective derivatives cannot take over in the long term and therefore they represent dead branches of the evolutionary tree. In conclusion, only the wild-type bacteria are efficient in long-term colonization and in person-to-person transmission and therefore they are those that govern H. pylori evolution.

additional film of surface-active phospholipids located on the stomach lumen side27, form a continuous viscous gel layer, which covers and protects the EPITHELIAL CELLS lining the stomach mucosa. In fact, the mucus has defined permeability properties (FIG. 1) and acts as a semipermeable barrier that allows the flow of protons from the H+-releasing OXYNTIC CELLS to the stomach lumen, but not backwards. On the contrary, it is poorly permeable to bicarbonate anions28. The net result is that there is a large pH gradient from the stomach lumen (strongly acid) to the apical surface of stomach mucosa (only slightly acidic). Human biopsies show H. pylori in the mucus as well as strongly adherent to the apical membrane of epithelial cells. In these locations, H. pylori is well protected from discharge by PERISTALSIS and mucus shedding. The actual volume available to large solutes below the mucus film and above the cells is not known, but it is likely to be very small. As a consequence, large molecules released by H. pylori, even in low amounts, may reach high local concentrations. The supply of nutrients necessary for bacterial growth, including the essential ions Fe3+ and Ni2+, is restricted because of the low permeability of the mucus layer on one side, and because of the tightness of the polarized epithelial monolayer on the other side. It is likely that H. pylori, similarly to other gastrointestinal bacteria, releases hydrolytic enzymes that can alter the integrity of the mucous film, making it more permeable to bacterial nutrients, as well as to toxic substances and protons coming from the stomach lumen29.

Infection

wt cag+ vacA+ babA+

cag+ vacA+ babA+ cag – vacA+ babA+ cag+ vacA– babA+ cag+ vacA+ babA–

cag+ vacA+ babA+ wt

Infection of next generation

Establishing contact

H. pylori adheres strongly to gastric cells, and binding is probably mediated by several proteins and glycolipids. BabA, an outer membrane protein of H. pylori, binds to the Lewis B-type antigen of human cells30,31. Another H. pylori protein (HpaA, HP0410) binds sialylated glyconjugates32,33 and the genome of H. pylori harbours a gene (HP0492) encoding a closely related protein that could have a similar function34,35. Another H. pylori LECTIN binds the sialic residues of laminin36. Additional bacterial proteins37–39 and glycolipids40 promote the adhesion of H. pylori to the gastric epithelium of the host. Cell adhesion is an essential step in H. pylori colonization, but one that is very difficult to dissect at the molecular level. Because several copies of putative binding molecules are present on the H. pylori cell surface, and electron microscopy reveals extensive areas of adhesion to the host cells, it is difficult to distinguish the relative role of each type of adhesion. In fact, when many ligands are present on the same particle, even single weak binding interactions become relevant to the establishment of a very strong overall interaction, owing to their combined action. So, it is likely that H. pylori mutants defective in single genes will never be in the condition to reveal the real contribution of the single binding component because they will adhere with unaltered strength, owing to the presence of the other ligands33. H. pylori binding to the apical plasma membrane of the tightly sealed, polarized epithelial monolayers is not without consequences for the host cells. The contact with the bacterium is intimate and almost irreversible,


VOLUME 2 | JUNE 2001 | 4 5 9


REVIEWS a ENDOTHELIAL CELLS

b

Flattened cells that grow in a single layer and line blood vessels. LPS

(Lipopolysaccharide.) Components of the outer membrane of Gram-negative bacteria that are made of a lipid, a core oligosaccharide and an O-linked sugar side chain.

NH2 O

A type of chemotactic cytokine that acts primarily on haematopoietic cells in acute and inflammatory processes.

TISSUE FACTOR

A transmembrane glycoprotein that initiates blood coagulation.

NH2

CO2 + NH3

FIBRINOLYSIS

The proteolysis of fibrin by plasmin in blood clots.

Urea

C

CHEMOKINE

NH3 NH3

NH3

Figure 2 | Helicobacter pylori. a | Electron micrograph and b | schematic representation showing shape, polar flagella, urease, H+-gated urea channel and the production of ammonia, which neutralizes the yellow acidic environment and the cytosol and the immediate environment around the bacterium (light blue). Panel a courtesy of Adrian Lee, Jani O’Rourke and Lucinda Thompson. Reproduced with permission from REF. 101 © (2001) American Society for Microbiology.

and leads to a profound rearrangement of the plasma membrane below the zone of contact41,42. Little or no invasion of epithelial cells by H. pylori seems to take place. Rather, the plasma membrane changes shape and extends to contact a large portion of the bacterial surface, after an H. pylori-induced reorganization of the underlying actin cytoskeleton42. At least part of these changes are induced by bacterial proteins injected by H. pylori directly into the host cell (BOX 2; TABLE 1). Recruiting inflammatory cells

Neutrophils and mononuclear inflammatory cells infiltrate the H. pylori-infected stomach mucosa and the

degree of mucosal damage correlates with neutrophil infiltration1,2,43–45. H. pylori does not invade the epithelial layer, and its components and metabolites have to permeate the epithelial barrier to induce the chemotaxis and activation of inflammatory cells. The identification and dissection of the activities of several H. pylori molecules that might be implicated in such a phenomenon is under way. A 150-kDa protein oligomer composed of identical 15-kDa subunits — H. pylori neutrophil-activating protein, or HP-NAP — was shown to promote the adhesion of human neutrophils to ENDOTHELIAL CELLS and the production of reactive oxygen radicals46,47. Purified recombinant HP-NAP, free of LPS, is chemotactic for human neutrophils and monocytes, indicating that it might have a role in the accumulation of these cells at the site of infection48. HP-NAP is a powerful stimulant of the production of reactive oxygen radicals48 and acts through a cascade of intracellular activation events, which is completely prevented by pertussis toxin: it includes the increase of cytosolic Ca2+ and the phosphorylation of proteins, leading to the assembly of functional NADPH oxidase on the neutrophil plasma membrane48 (FIG. 3). These features indicate that HP-NAP acts in a CHEMOKINE-like fashion and that its receptor is a seven-transmembrane-spanning type of receptor. HPNAP also alters the coagulative-FIBRINOLYSIS balance of human monocytes, by increasing the expression of TISSUE FACTOR and of the inhibitor of fibrinolysis PAI-II (REF. 49). This effect occurs at tenfold lower concentrations than those inducing burst and chemotaxis, and it might hamper tissue repair, which requires fibrinolysis. The nucleotide sequence of the gene coding for HP-NAP is highly conserved among different H. pylori isolates, and orthologue genes are present in many bacteria. The atomic structure of HP-NAP reveals a four-helix bundle protein, which oligomerizes to a form a dodecamer with a central iron-containing cavity (G. Zanotti and C.M., unpublished observations). The function of HP-NAP is not known, but we would like to propose that it might be to induce a state of mild to moderate inflammation. This pro-inflammatory activity is further supported by a set of cytokines

Box 2 | Bacterial secretion systems The concept that pathogenic bacteria alter host cells through bacterial toxins, which may diffuse at a distance in the host, is well established. A more recent realization is that many Gram-negative bacteria elaborate alternative systems that directly inject bacterial molecules into adhering cells. A first group of such molecular syringes derives from duplication and re-elaboration through evolution of flagella (type III secretion systems). The second type of molecular syringe (type IV secretion systems) derives from the evolution of conjugative pili, and comprises the cag system (FIG. 5). Both flagella and pili consist of a basal body spanning the inner membrane, the periplasmic space and the outer membrane, and an external filamentous portion, which may extend at considerable distances out of the bacterial surface. This latter part is an extended helicoidal oligomer of small repeating units, with an axial cavity that can transport the monomers to be added to the tip of the growing filament, as well as proteins and other bacterial molecules into the cytosol of higher eukaryotes. A list of the bacteria known to produce type III and IV secretion systems is listed in TABLE 1. Despite their similar structural architecture, no apparent similarity is evident between the sequences of the protein components of the two families of secretion systems, whereas homology exists among members of the same family. In addition, type IV secretion systems are likely to have a larger inner diameter than type III secretion systems as they can transport protein oligomers (the pentameric pertussis toxin of Bordetella pertussis) and protein–DNA complexes (Ti plasmid T-DNA of Agrobacterium tumefaciens).

460

www.nature.com/reviews/molcellbio



REVIEWS HP-NAP

Table 1 | Bacterial secretion systems

PT-tox Ca2+

Bacterial species

NADPH oxidase Trimeric G protein

O2

α β γ

O2–

ROIs

Ca2+ Membrane

p47 phox PLC

p67 phox

Rac1– GTP

Ins(1,4,5)P3

Rac2– GTP Cytoskeleton

DAG PKC

Renaturable kinases

ER

+ +

Bordetella bronchiseptica

+

+

Bordetella pertussis

+

+

Brucella suis

+ +

Citrobacter rodentium

+

Escherichia coli

+

Enteropathogenic E. coli

+

Enterohaemorrhagic E. coli

+

Erwinia amylovora

+

Erwinia chrysanthemi

+

Erwinia herbicola pv. gysophila

+

Erwinia stewartii

+

Hafnia alveii

+

Helicobacter pylori Figure 3 | Scheme of the intracellular events involved in the HP-NAP activation of human leukocytes. Helicobacter pylori neutrophil-activating protein (HP-NAP) binds to a specific receptor, which is coupled to a pertussis toxin (PT-tox)-sensitive trimeric G protein. Binding triggers the entry of calcium through plasma-membrane channels and through the activation of channels located on the endoplasmic reticulum (ER), which are opened by inositol-1,4,5-trisphosphate Ins(1,4,5)P3, produced by activated phospholipase C (PLC). The activated HP-NAP receptor also stimulates a phosphatidylinositol 3-kinase (PI3K) activity, which is inhibited by wortmannin. The rise in the cytosolic Ca2+ concentration and the activity of PI3K leads to the activation of renaturable kinases that phosphorylate the cytosolic subunits of the NADPH oxidase of phagocytes (p47phox, p67phox and p40phox), causing their migration to the plasma membrane. Together with the Rac-1 and Rac-2 GTPases, they activate the oxidase activity with production of superoxide anions and reactive oxygen intermediates (ROIs). Adapted from REF. 48. (DAG, diacylglycerol; PKC, protein kinase C.)

(interleukin-8 (IL-8), GRO, epithelial-derived neutrophil-activating peptide 78, RANTES, metaphase chromosome protein 1) produced by the stomach mucosa infected by H. pylori 50,51. The tissue degradation that follows the recruitment and the activation of inflammatory cells would generally not be too harmful to the host, but would strongly promote the production and release of nutrients that are necessary for bacterial growth. Therefore, we speculate that HPNAP was originally an iron-binding/iron-regulated protein (as HP-NAP-like molecules might still be in other bacteria), and that HP-NAP has later evolved to function as a leukocyte activator. On the host side, human neutrophils might have ‘learned’ to recognize some of these highly conserved dodecameric ironbinding components as hallmarks of bacteria, similarly to LPS. Causing damage

Vacuolating cytotoxin A (VacA) is a 95-kDa protein of H. pylori that induces the formation of large cytoplasmic vacuoles in cultured cells52. Not only is it an important antigen in the human immune response to H. pylori 23, but also it is an important virulence factor of H. pylori,

Type IV

Agrobacterium tumefaciens

Chlamydia spp.

p40 phox

PI3K

Ca2+

Type III

Actinobacillus actinomycetemcomitans

+

+

Legionella pneumophila

+

Pseudomonas aeruginosa

+

Pseudomonas syringae

+

Ralstonia solanacearum

+

Rickettsia prowazekii

+

Rhizobium spp.

+

Salmonella enterica

+

Shigella spp.

+

Xantomonas spp.

+

Yersinia spp.

+

which confers a strong competitive advantage to wildtype strains with respect to VacA-defective mutants in the colonization of the stomach53. VacA is secreted in a two-step process involving an amino-terminal 33 amino-acid signal peptide, which directs secretion from the cytoplasm to the periplasm, and a 45-kDa carboxy-terminal autotransporter, which directs export across the outer membrane. The secreted protein is structured in two distinct parts. First, it has an amino-terminal 37-kDa region, predicted to be rich in β-pleated sheets. This region begins with a 32-residue hydrophobic segment with a propensity to insert into membranes54, and ends with a protease-sensitive segment. Indeed, part of the VacA toxin released into the medium is nicked at this point55. Second, the following 58-kDa part is predicted to consist of two domains, separated by a flexible segment of variable length: the first domain is highly conserved, whereas the second one is genetically considerably diverse56. Part of VacA is released in the medium, but 40–60% of the mature VacA remains associated to the outer membrane of H. pylori. Apart from straindependent sequence variations in the signal sequence of VacA, which influence the amount of VacA released,


VOLUME 2 | JUNE 2001 | 4 6 1


REVIEWS a

VacA b Urease

Urea

X–

NH3

Fe3+ Ni2+ X–

Endosome ADP ATP v-ATPase

Vacuole NH3

H+

+

NH4 H2O

Nucleus

TRANS-EPITHELIAL RESISTANCE

Electric resistance across epithelial sheets, measured across the apical–basolateral axis of the cell. TIGHT JUNCTION

A belt-like region of adhesion between adjacent epithelial or endothelial cells. Tight junctions regulate paracellular flux, and contribute to the maintenance of cell polarity by stopping molecules from diffusing within the plane of the membrane. APICAL SURFACE

Surface of an epithelial or endothelial cell that faces the lumen of a cavity or tube, or the outside of the organism.

462

Figure 4 | Current model of the cellular alterations induced by VacA cytotoxin. a | The toxin is an oligomer. b | It binds to the apical portion of epithelial cells and inserts into the plasma membrane, forming a hexameric anionselective channel of low conductance. These channels release bicarbonate and organic anions from the cell cytosol to support bacterial growth. The toxin channels are slowly endocytosed and eventually reach late endosomal compartments, increasing their permeability to anions with enhancement of the electrogenic vacuolar ATPase (vATPase) proton pump. In the presence of weak bases, including the ammonia generated by the Helicobacter pylori urease, osmotically active acidotropic ions will accumulate in the endosomes. This leads to water influx and vesicle swelling, an essential step in vacuole formation. By an as-yetunidentified mechanism, the VacA toxin alters tight junctions and increases the paracellular route of permeability providing iron, nickel and other nutrients, essential for H. pylori growth, from the underlying mucosa.

a second highly divergent segment is present in the second half of p58 (REF. 56), which is probably involved in target-cell interaction. The two alleles might have evolved to bind different receptors57. VacA is mainly released in the limited volume sandwiched between the mucus layer and the apical domain of stomach epithelial cells. So, the existence of a high-affinity VacA receptor on the apical domain is not strictly necessary. Indeed, a non-saturable number of low-affinity receptors were found for radioactive VacA on HeLa cells58, and different groups have reported different putative VacA receptors. Whatever the mode of binding, VacA inserts into the plasma

membrane of cells, where it forms anion-specific channels of low conductance59. The electrophysiological properties of these channels are similar to those of the hexameric channels that VacA forms in planar lipid bilayers after low-pH activation60,61. The secreted VacA toxin has a strong tendency to oligomerize into rosettes62 (FIG. 4). The oligomeric form of VacA has little vacuolating activity, but it is activated by short exposure to acid or alkaline media and, remarkably, it is not denatured at pH 1.5 and has an unusual resistance to pepsin digestion63,64. Low pH dissociates the VacA oligomer into monomers, exposing hydrophobic patches on the surface, which mediate its insertion into biological membranes65. Polarized epithelial monolayers develop TRANSEPITHELIAL RESISTANCE (TER) by sealing the cell borders of adjacent cells through TIGHT JUNCTIONS and other intercellular structures. The measurable TER correlates with the degree of cell sealing66. Acid-activated VacA added apically to epithelial monolayers grown on filters, induces a rapid drop of TER with a concomitant increase in paracellular permeability to small organic molecules and to Fe3+ and Ni2+ (REF. 67) (FIG. 4). H. pylori grown on the APICAL SURFACE of the epithelial cells causes the same increase in permeability, whereas VacA-defective H. pylori strains do not68. These results indicate that one function of VacA might be to increase the supply of essential nutrients — necessary for growth of H. pylori — from the underlying tissue. This TER effect of VacA could be mediated through specific interactions with the recently identified cytosolic protein VIP54 (REF. 69). VacA activity is blocked by cytochalasin D and by dominant-negative mutants of proteins involved in endocytosis58,70, indicating that plasma-membraneinserted VacA is endocytosed58 through an ATP- and actin-dependent mechanism. At least part of VacA is internalized through non-clathrin-coated structures and reaches early and then late endosomes71,72, where the VacA channel is expected to preserve its functions as it is active at low pH. VacA-induced vacuoles are acidic because their limiting membrane contains the vacuolar-type ATPase proton pump (v-ATPase), the operation of which is essential for vacuole formation73. Accordingly, vacuoles are promoted by the accumulation of membrane-permeable weak bases that are trapped by protonation in their lumen71 (FIG. 4). These vacuoles contain membrane protein markers of late endosomes and lysosomes, and their formation requires active Rab7, a small GTPase involved in controlling membrane transport through late endosomal compartments74. This indicates that vacuole biogenesis is a specific process taking place at the level of late endosomes, and not a generalized swelling of flattened and tubular compartments due to cell damage. Negative-staining electron microscopy of biopsies of stomach mucosa from H. pylori-infected patients and of VacA-treated HeLa cells shows vacuoles largely devoid of the large array of internal membranes and multivesicular bodies characteristic of late endosomes/lysosomes71. Therefore, VacA induces an extensive rearrangement of the internal www.nature.com/reviews/molcellbio



REVIEWS

Figure 5 | The secretion system of Helicobacter pylori injects bacterial proteins into the cytosol of host cells. a | Immunofluorescence image showing H. pylori (red) injecting the CagA protein (green) into a host cell. b | Schematic drawing of the picture shown in a. c | Schematic representation of the type IV injection system composed of a pilus-like structure, which crosses the inner and outer membrane of the bacterium and, like a needle, inserts into the membrane of an enterocytic host cell to inject CagA, and possibly other bacterial proteins, into the cytosol. Panel a is adapted with permission from REF. 92 © (1999) National Academy of Sciences.

ANTIGEN-PRESENTING CELLS

Cells specialized in the generation of epitopes that are presented through major histocompatibility complex II (MHC II) to T lymphocytes. CD4+ T CELLS

T helper cells, which collaborate with antigen-presenting cells in the initiation of an immune response. CD8+ CYTOTOXIC TCELLS

T cytotoxic cells, which are directly responsible for killing cells that present peptides through MHC I. TRANSPOSABLE ELEMENTS

Also called transposons. Specific DNA sequences that are transferred as a unit from one replicated DNA sequence to another. HORIZONTAL TRANSFER

Transfer of DNA sequences from one bacterium to another. COMMENSAL

Either of two species that live in close association with benefit to one partner but with little or no effect on the other partner.

organization of these compartments. In late endosomes/lysosomes, the action of the vATPase proton pump builds up an electrochemical proton gradient that progressively depresses its further activity. The necessary counter ion is normally provided by a Cl– channel present on the same membrane, which is essential for acidification75. We suggest that the anion-selective channel activity of VacA strongly promotes the proton pumping activity of the v-ATPase, leading to an enhanced accumulation of protons. This, in turn, will lead to an accumulation of ammonia and any other weak bases present in the medium. Even a limited uptake of osmotic species is expected to be sufficient to cause a significant increase in osmotic pressure given the limited membrane-free space of late endosomal compartments. Accordingly, anion-channel inhibitors also inhibit vacuolation of HeLa cells exposed to VacA76. In addition, VacA-transfected cells vacuolate, and such an activity requires the whole amino-terminal domain plus a contiguous region of the carboxy-terminal domain77,78. Cytosolexpressed VacA has a wide choice of intracellular membranes to bind to, in addition to late endosomes. In particular, the overexpressed amino-terminal domain of VacA associates with mitochondria and, alone or in cooperation with endogenous proteins, it promotes cytochrome c release, leading to cell death by apoptosis79. However, VacA is not apoptotic when added to the extracellular medium, most probably because in this case the localization of the VacA channel is restricted to the plasma membrane and to endocytic membranes. VacA-induced vacuolization has several consequences for cellular physiology that might contribute

to pathogenesis and to H. pylori survival. It causes a marked decrease of the proteolytic activity in the endocytic pathway, including antigen proteolysis in the antigen-processing compartment of ANTIGEN PRESENTING CELLS (APC) that is needed to generate peptide epitopes80. Consequently, VacA inhibits the stimulation of T-cell clones, specific for epitopes generated in the antigen-processing compartment80. The relevance of this effect in vivo remains to be established but there is evidence that H. pylori does depress the local immune response because VacA-specific CD4+ T CELLS are found at a low frequency in the stomach mucosa of H. pylori-infected subjects81 and persistent H. pylori infection down-modulates specific CD8+ CYTOTOXIC T 82 CELL response, prolonging viral infection . The inhibition of local antigen processing by VacA could be part of a strategy of survival for H. pylori that could significantly contribute to its chronic infection of the human stomach. Vacuolization of late endosomal compartments also brings about an extensive alteration of protein transport from the trans-Golgi network (TGN) to late endosomes83. Lysosomal acid hydrolases are made in the endoplasmic reticulum as pre-pro-enzymes and reach late endosomes and lysosomes, where the preand pro-segments are removed and the hydrolase domains are activated. In VacA-intoxicated cells, prepro-acid hydrolases are instead released into the extracellular medium83. If the same transport alteration takes place in the epithelial lining of the stomach, the secreted hydrolases can be converted in the apical domain into active enzymes that can loosen the meshwork and thickness of the mucin film with consequent increased permeability to ions and nutrients to support H. pylori growth. Reaching optimal fitness

H. pylori isolates are classified as cag + and cag –, depending on the presence of a chromosome pathogenicity island (PAI) of 40 kb, containing ~30 genes, termed cag PAI84. PAIs are characterized by a different nucleotide composition with respect to the host bacterial genome; they are flanked by TRANSPOSABLE ELEMENTS and are usually acquired by HORIZONTAL TRANSFER85. Pathogenicity islands seem to increase the fitness of bacteria in a given environment by providing them with environment-specific functions. The cag PAI provides H. pylori with at least two unique properties: an increased transmission probability and the transformation of what would be an almost COMMENSAL into a potential pathogen (BOX 1). The role of the cag PAI in pathogenesis is shown by the fact that in Mongolian gerbils cag – bacteria cause only mild asymptomatic inflammation of the stomach, whereas cag + organisms cause severe inflammation, gastric ulcers and gastric tumours86 (TABLE 2). These data are consistent with many epidemiological studies reporting that, in humans, severe gastric diseases are always associated with infection by cag+ strains87–89. A major pathogenetic event contributed by the cag PAI is the induction of host cells to release pro-inflammatory chemokines50,51.


VOLUME 2 | JUNE 2001 | 4 6 3


REVIEWS

Table 2 | Diseases associated with Helicobacter pylori* –

Strain

Wild type

vacA

cag

Mild gastritis

100

100

100

Severe gastritis

100

93

0

Gastric ulcer

96

68

0

Intestinal metaplasia

61

54

0

Carcinoid

4

18

0

Adenocarcinoma

4

4

0

–

* Values show the frequency (%) of animals with disease when infected with isogenic strains of Helicobacter pylori. The data, which are obtained from experimental infections of Mongolian gerbils86 correlate well with epidemiological observations, showing that the presence of the cag pathogenicity island correlates with increased virulence and disease severity in humans87–89.

Injecting the host

About a dozen genes of the cag PAI code for the building blocks of a secretion apparatus — a type IV secretion system — that delivers bacterial proteins directly into the cytosol of host cells90 (BOX 2). Cytotoxin-associated gene A (cagA), a 128–145-kDa protein that becomes phosphorylated in the cytosol of host cells91–94, is the only protein known so far to be injected by H. pylori into host cells95. This is the first evidence of a functional secretion apparatus in H. pylori, previously inferred only from sequence comparisons4. It is likely that other H. pylori proteins are delivered in a similar way, as the intact type IV secretion system, but not CagA, is required to induce the secretion of IL-8 from gastric epithelial cells50, although it cannot be excluded that the mere insertion of the H. pylori type IV needle into the host plasma membrane is sufficient to activate endogenous signals (FIG. 5). The factors responsible for this effect have not been identified, but the activation of the nuclear transcription factors NF-κB and AP1, and of the cytosolic phospholipase A2 after the stimulation of the mitogen-activated protein kinase cascade and the p21-activated kinase has been documented96,97. CagA injected by H. pylori is tyrosine-phosphorylated and may bind Src homology region 2 domain (SH2)-containing proteins, forming a signalling complex that promotes the reorganization of cortical actin, and, possibly, of some membrane components of the apical membrane98. The signalling events triggered by phosphorylated CagA are still to be unravelled. However, the actin polymerization and the formation of pedestals formed shortly after bacterial adhesion are believed to be due to CagA injection.

1.

2.

464

Warren, J. R. & Marshall, B. J. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 1, 1273–1275 (1983). The beginning of a Copernican revolution in gastroenterology, which has led to a complete change of perspective and therapeutic approach with a tremendous improvement of human health. Achtman, M. & Suerbaum, S. (eds) Helicobacter pylori: Molecular and Cellular Biology (Horizon Scientific, Norfolk, 2001). The most updated multi-author book covering the molecular and tissue mechanisms of action of H. pylori virulence factors.

3.

4.

5. 6.

7.

An important but unexplored observation about CagA is its high immunogenicity, as judged by the prevalence of antibodies against CagA and by the frequency of CagA-specific T cells in the gastric mucosa of H. pylori-infected patients23. It is possible that CagA is handled by host cells as a cytosolic viral protein and enters the MHC class I antigen presentation pathway. Fighting the enemy

Urease, CagA, VacA and HP-NAP, in addition to being important determinants of H. pylori stomach colonization and pathogenicity, are also major antigens in the human immune response to H. pylori infection23. During the past ten years, these antigens have been used in many animal studies and each of them was found to induce an immune response that could protect animals from infection (prophylactic vaccines) and eradicate an already existing infection (therapeutic vaccines)23.Vaccines to be tested in humans were then developed from one or more of the above antigens expressed in Escherichia coli. Both oral or systemic immunization were considered. To render the vaccine effective, oral immunization requires the use of mucosal adjuvants, such as the cholera toxin or the E. coli heat labile enterotoxins (LT). However, these are used only in animal models because of their toxicity. Recently, several non-toxic derivatives, which can be safely used in humans, were developed, with LTK63 and LTR72 being the most promising ones99. An intriguing question that derives from animal experiments concerns the possible mechanism of protection of H. pylori vaccines. In γ-globulin-deficient mice, immune protection was achieved in the total absence of antibodies and was mediated by CD4+ T cells100. If the continuing clinical studies for the human vaccine are successful, we might soon witness the elimination of a pathogen that has been living in humans for more than 150,000 years. Links DATABASE LINKS urease | HP-NAP | IL-8 | GRO | epithelial-derived neutrophil-activating protein 78 | RANTES | metaphase chromosome protein 1 | Rab7 | cytochrome c | cagA | NF-κB | AP1 | phospholipase A2 | mitogen-activated protein kinase | p21-activated kinase | MHC class I FURTHER INFORMATION The complete sequences of two strains of H. pylori (J99 and 26695) | A yeast two-hybrid protein–protein interaction map | A two-dimensional electrophoresis map of H. pylori 26696

Parsonnet, J. The incidence of Helicobacter pylori infection. Aliment. Pharmacol. Ther. 9, suppl. 2, 45–51 (1995). Covacci, A., Telford, J. L., Del Giudice, G., Parsonnet, J. & Rappuoli, R. Helicobacter pylori virulence and genetic geography. Science 284, 1328–1333 (1999). Donati, M. De Medica Historia Mirabili (Mantua, 1586). Go, M. F. What are the host factors that place an individual at risk for Helicobacter pylori-associated disease? Gastroenterol. 113, S15–S20 (1997). Marshall, B. J. & Langton, S. R. Urea hydrolysis in patients with Campylobacter pyloridis infection. Lancet 26, 965–966 (1986).

8.

Hu, L.-T. & Mobley, H. L. T. Purification and N-terminal analysis of urease from Helicobacter pylori. Infect. Immun. 58, 992–998 (1990). 9. Labigne, A., Cussac, V. & Courcoux, P. Shuttle cloning and nucleotide sequences of Helicobacter pylori genes responsible for urease activity. J. Bacteriol. 173, 1920–1931 (1991). 10. Jabri, E., Carr, M. B., Hausinger, R. P. & Karplus, P. A. The crystal structure of urease from Klebsiella aerogenes. Science 268, 998–1004 (1995). 11. Bauerfeind, P., Garner, R., Dunn, B. E. & Mobley, H. L. Synthesis and activity of Helicobacter pylori urease and catalase at low pH. Gut 40, 25–30 (1997).




REVIEWS 12. Weeks, D. L., Eskandari, S., Scott, D. R. & Sachs, G. A H+-gated urea channel: the link between Helicobacter pylori urease and gastric colonization. Science 287, 482–485 (2000). A remarkable paper on how this bacterium handles the vital urea molecule. 13. Phadnis, S. H. et al. Surface localization of Helicobacter pylori urease and a heat shock protein homolog requires bacterial autolysis. Infect. Immun. 64, 905–912 (1996). 14. Dunn, B. E. et al. Localization of Helicobacter pylori urease and heat shock protein in human gastric biopsies. Infect. Immun. 65, 1181–1188 (1997). 15. Eaton, K. A., Brooks, C. L., Morgan, D. R. & Krakowka, S. Essential role of urease in pathogenesis of gastritis induced by Helicobacter pylori in gnotobiotic piglets. Infect. Immun. 59, 2470–2475 (1991). 16. Clyne, M., Labigne, A. & Drumm, B. Helicobacter pylori requires an acidic environment to survive in the presence of urea. Infect. Immun. 63, 1669–1673 (1995). 17. Megraud, F., Neman-Simha, V. & Brugman, D. Further evidence of the toxic effect of ammonia produced by Helicobacter pylori urease on human epithelial cells. Infect. Immun. 60, 1858–1863 (1992). 18. Suzuki, M. et al. Helicobacter pylori-associated ammonia production enhances neutrophil-dependent gastric mucosal cell injury. Am. J. Physiol. 263, G719–G725 (1992). 19. Sommi, P. et al. Significance of ammonia in the genesis of gastric epithelial lesions induced by Helicobacter pylori: an in vitro study with different bacterial strains and urea concentrations. Digestion 57, 299–304 (1996). 20. Williams, C. L., Preston, T., Hossack, M., Slater, C. & McColl, K. E. Helicobacter pylori utilizes urea for amino acid synthesis. FEMS Immunol. Med. Microbiol. 13, 87–94 (1996). 21. Harris, P. R., Mobley, H. L., Perez-Perez, G. I., Blaser, M. J. & Smith, P. D. Helicobacter pylori urease is a potent stimulus of mononuclear phagocyte activation and inflammatory cytokine production. Gastroenterol. 111, 419–425 (1996). 22. Ferrero, R. L., Thiberge, J.-M., Huerre, M. & Labigne, A. Recombinant antigens prepared from the urease subunits of Helicobacter pylori spp: evidence of protection in a mouse model of gastric infection. Infect. Immun. 62, 4981–4989 (1994). 23. Del Giudice, G., Covacci, A., Telford, J. L. Montecucco, C. & Rappuoli, R. The design of vaccines against Helicobacter pylori and their development. Annu. Rev. Immunol. 19, 523–563 (2001). A thorough and updated analysis of the current status of the effort to produce an anti H. pylori vaccine. 24. Josenhans, C. & Suerbaum, S. in Helicobacter pylori: Molecular and Cellular Biology (eds Achtman, M. & Suerbaum, S) 171–184 (Horizon Scientific, Norfolk, UK, 2001). 25. Yoshiyama, H., Nakamura, H., Kimoto, M., Okita, K. & Nakazawa, T. Chemotaxis and motility of Helicobacter pylori in a viscous environment. J. Gastroenterol. 34, suppl. 11, 18–23 (1999). 26. Slomiany, B. L. & Slomiany, A. Role of mucus in gastric mucosal protection. J. Physiol. Pharmacol. 42, 147–161 (1991). 27. Lichtenberger, L. M. The hydrophobic barrier properties of gastrointestinal mucus. Annu. Rev. Physiol. 57, 565–583 (1995). 28. Bhaskar, K. R. et al. Viscous fingering of HCl through gastric mucin. Nature 360, 458–461 (1992). 29. Slomiany, B. L., Murty, V. L., Piotrowski, J. & Slomiany, A. Gastroprotective agents in mucosal defense against Helicobacter pylori. Gen. Pharmacol. 25, 833–841 (1994). 30. Boren, T., Falk, P., Roth, L. A., Larson, G. & Normark, S. Attachment of Helicobacter pylori to human gastric epithelium mediated by blood group antigens. Science 262, 1892–1895 (1993). 31. Ilver, D. et al. Helicobacter pylori adhesin binding fucosylated histo-blood group antigens revealed by retagging. Science 279, 373–377 (1998). 32. Evans, D. G., Karjalainen, T. K., Evans, D. J., Graham, D. Y. & Lee, C.-H. Cloning, nucleotide sequence, and expression of a gene encoding an adhesion subunit protein of Helicobacter pylori. J. Bacteriol. 175, 674–683 (1993). 33. Jones, A. C. et al. A flagellar sheath protein of Helicobacter pylori is identical to HpaA, a putative Nacetylneuraminyllactose-binding hemagglutinin, but is not an adhesin for AGFS cells. J. Bacteriol. 179, 5643–5647 (1997). 34. Tomb, J. F. et al. The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388,

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

539–547 (1997). Although H. pylori is the most recent bacterial pathogen discovered, its genome was one of the first to be sequenced. The genome sequence provides a fundamental step in the understanding of H. pylori. Alm, R. A. et al. Genomic-sequence comparison of two unrelated isolates of the human gastric pathogen Helicobacter pylori. Nature 397, 176–180 (1999). Valkonen, K., Wadstrom, T. & Moran, A. P. Identification of the N-acetylneuraminyllactose-specific laminin binding protein. Infect. Immun. 65, 916–923 (1993). Lingwood, C. A. et al. Receptor affinity purification of a lipid-binding adhesin from Helicobacter pylori. Infect. Immun. 61, 2474–2478 (1993) Namavar, F., Sparrius, M., Veerman, E. C. I., Appelmelk, B. J. & Van Vandenbroucke-Grauls, C. M. Neutrophilactivating protein mediates adhesion of Helicobacter pylori to sulphated carbohydrates on high-molecular weight salivary mucin. Infect. Immun. 66, 444–447 (1997). Odenbreit, S. et al. Genetic and functional characterization of the alpAB gene locus essential for the adhesion of Helicobacter pylori to human gastric tissue. Mol. Microbiol. 31, 1537–1548 (1999). Edwards, N. J. et al. Lewis X structures in the O antigen side-chain promote adhesion of Helicobacter pylori to the gastric epithelium. Mol. Microbiol. 35, 1530–1539 (2000). Smoot, D. T. et al. Adherence of Helicobacter pylori to cultured human gastric epithelial cells. Infect. Immun. 61, 350–355 (1993). Segal, E. D., Falkow, S. & Tompkins, L. S. Helicobacter pylori attachment to gastric cells induces cytoskeletal rearrangements and tyrosine phosphorylation of host cell proteins. Proc. Natl Acad. Sci. USA 93, 1259–1264 (1996). Marshall, B. J., Armstrong, J. A., McGeche, D. B. & Glancy, R. J. Attempt to fulfil Koch’s postulates for pyloric Campylobacter. Med. J. Austr. 142, 436–439 (1985). A classic of medical microbiology. Fiocca, R. et al. Epithelial cytotoxicity, immune responses, and inflammatory components of Helicobacter pylori gastritis. Scand. J. Gastroenterol. 205, 11–21 (1994). Dixon, M. F., Genta, R. M., Yardley, J. H. & Correa, P. Classification and grading of gastritits. The updated Sidney System. Am. J. Surg. Pathol. 20, 1161–1181 (1996). Yoshida, N. et al. Mechanisms involved in Helicobacter pylori-induced inflammation. Gastroenterology 105, 1431–1440 (1993). Evans, D. J. Jr et al. Characterization of a Helicobacter pylori neutrophil-activating protein. Infect. Immun. 63, 2213–2220 (1995). Satin, B. et al. The neutrophil activating protein (HP-NAP) of Helicobacter pylori is a protective antigen and a major virulence factor J. Exp. Med. 191, 1467–1476 (2000). An accurate characterization of the mode of activation and chemotaxis induced by HP-NAP, including the first demonstration that HP-NAP is a major antigen. Montemurro, P. et al. HP-NAP of Helicobacter pylori alters the coagulation-fibrinolysis balance of human blood mononuclear cells by stimulating the expression of Tissue Factor and PAI-2. J. Infect. Dis. 183, 1055–1062 (2001). Crabtree, J. E. et al. Induction of interleukin-8 secretion from gastric epithelial cells by cagA negative isogenic mutant of Helicobacter pylori. J. Clin. Pathol. 48, 967–969 (1995). Shimoyama, T., Everett, S. M., Dixon, M. F., Axon, A. T. R. & Crabtree, J. E. Chemokine mRNA expression in gastric mucosa is associated with Helicobacter pylori positivity and severity of gastritis. J. Clin. Pathol. 51, 765–770 (1998). Cover, T. L. & Blaser, M. J. Purification and characterization of the vacuolating toxin from Helicobacter pylori. J. Biol. Chem. 267, 10570–10575 (1992). The original description of the VacA toxin. Salama, N., Otto, G., Tompkins, L. & Falkow, S. Vacuolating cytotoxin of Helicobacter pylori plays a role during colonization in a mouse model of infection. Infect. Immun. 69, 730–736 (2001). An almost textbook example of insightful study of bacterial ecology. McClain, M. S., Cao, P. & Cover, T. L. Amino-terminal hydrophobic region of Helicobacter pylori vacuolating cytotoxin (VacA) mediates transmembrane protein dimerization. Infect. Immun. 69, 1181–1184 (2001). Telford, J. L. et al. Gene structure of Helicobacter pylori cytotoxin and evidence of a key role in gastric disease. J. Exp. Med. 179, 1653–1658 (1994). Atherton, J. C. et al. Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori. Association of specific VacA


57.

58.

59.

60.

61.

62.

63.

64.

65.

66. 67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

types with cytotoxin production and peptic ulceration. J. Biol. Chem. 270, 17771–17777 (1995). Pagliaccia, C. et al. The m2 form of the Helicobacter pylori cytotoxin has cell-type-specific vacuolating activity. Proc. Natl Acad. Sci. USA 95, 10212–10217 (1998). McClain, M. S., Schraw, W., Ricci, V., Boquet, P. & Cover, T. L. Acid activation of Helicobacter pylori vacuolating cytotoxin (VacA) results in toxin internalization by eukaryotic cells. Mol. Microbiol. 37, 433–442 (2000). Szabo, I. et al. Formation of anion-selective channels in the cell plasma membrane by the toxin VacA of Helicobacter pylori is required for its biological activity. EMBO J. 18, 5517–5527 (1999). Tombola, F. et al. Helicobacter pylori vacuolating toxin forms anion-selective channels in planar lipid bilayers. Biophys. J. 76, 1401–1409 (1999). Iwamoto, H., Czajkowsky, D. M., Cover, T. L., Szabo, G. & Shao, Z. VacA from Helicobacter pylori: a hexameric chloride channel. FEBS Lett. 450, 101–104 (1999). Lupetti, P. et al. Oligomeric and subunit structure of the Helicobacter pylori vacuolating cytotoxin. J. Cell. Biol. 133, 801–807 (1996). de Bernard, M. et al. Low pH activates the vacuolating toxin of Helicobacter pylori, which becomes acid and pepsin resistant. J. Biol. Chem. 270, 23937–23940 (1995). Yahiro, K. et al. Activation of Helicobacter pylori VacA toxin by alkaline or acid conditions increases its binding to a 250-kDa receptor protein-tyrosine phosphatase beta. J. Biol. Chem. 274, 36693–36699 (1999). Cover, T. L., Hanson, P. I. &. Heuser, J. E. Acid-induced dissociation of VacA, the Helicobacter pylori vacuolating toxin, reveals its pattern of assembly. J. Cell Biol. 138, 759–769 (1997). Eaton, S. & Simons, K. Apical, basal, and lateral cues for epithelial polarization. Cell 82, 5–8 (1995). Papini, E. et al. Selective increase of the permeability of polarized epithelial cell monolayers by Helicobacter pylori vacuolating toxin. J. Clin. Invest. 102, 813–820 (1998). Pelicic, V. et al. Helicobacter pylori VacA cytotoxin associated with bacteria increases epithelial permeability independently of its vacuolating activity. Microbiology 145, 2043–2050 (1999). de Bernard, M., Moschioni, M., Napolitani, C., Rappuoli, R. & Montecucco, C. The VacA toxin of Helicobacter pylori identifies a new intermediate filament interacting protein. EMBO J. 19, 48–56 (2000). Hotchin, N. A., Cover, T. L. & Akhtar, N. Cell vacuolation induced by the VacA cytotoxin of Helicobacter pylori is regulated by the Rac1 GTPase. J. Biol. Chem. 275, 14009–14012 (1999). Ricci, V. et al. Helicobacter pylori vacuolating toxin accumulates within the endosomal-vacuolar compartment of cultured gastric cells and potentiates the vacuolating activity of ammonia. J. Pathol. 183, 453–459 (1997). Molinari, M. et al. Vacuoles induced by Helicobacter pylori toxin contain both late endosomal and lysosomal markers. J. Biol. Chem. 272, 25339–25344 (1997). Papini, E. et al. The vacuolar ATPase proton pump is present on intracellular vacuoles induced by Helicobacter pylori. J. Med. Microbiol. 44, 1–6 (1996). Papini, E. et al. The small GTP binding protein rab7 is essential for cellular vacuolation induced by Helicobacter pylori cytotoxin. EMBO J. 16, 15–24 (1997). The demonstration that VacA induced cell vacuolization is a specific process requiring the enzymic activity of a specific Rab protein involved in the control of membrane transport at the late endosomal level. Gunther, W., Luchow, A., Cluzeaud, F., Vandewalle, A. & Jentsch, T. ClC-5, the chloride channel mutated in Dent’s disease, colocalizes with the proton pump in endocytotically active kidney cells. Proc. Natl Acad. Sci. USA 95, 8075–8080 (1998). Identification of ClC-5 as the Cl– channel isoform that limits the rate of endosomal acidification by providing the counter-anion necessary for electric balance. Tombola, F. et al. Inhibition of the vacuolating and anion activities of the VacA toxin of Helicobacter pylori. FEBS Lett. 460, 221–225 (1999). De Bernard, M. et al. Helicobacter pylori toxin VacA induces vacuole formation by acting in the cell cytosol. Mol. Microbiol. 26, 665–674 (1997). Ye, D., Willhite, D. C. & Blanke, S. R. Identification of the minimal intracellular vacuolating domain of the Helicobacter pylori vacuolating toxin. J. Biol. Chem. 274, 9277–9282 (1999). Galmiche, A. et al. The N-terminal 34 kDa fragment of Helicobacter pylori vacuolating cytotoxin targets mitochondria and induces cytochrome c release. EMBO J.

VOLUME 2 | JUNE 2001 | 4 6 5


REVIEWS 19, 6361–6370 (2000). 80. Molinari, M. et al. Selective inhibition of Li-dependent antigen presentation by Helicobacter pylori toxin VacA. J. Exp. Med. 187, 135–140 (1998). 81. D’Elios, M. M. et al. Th1 effector specific for Helicobacter pylori in the gastric antrum of patients with peptic ulcer disease. J. Immunol. 158, 962–967. 82. Shirai, M., Arichi, T., Nakazawa, T. & Berzofsky, J. A. Persistent infection by Helicobacter pylori downmodulates virus-specific CD8+ cytotoxic T cell response and prolongs viral infection. J. Infect. Dis. 177, 72–80 (1998). 83. Satin, B. et al. Vacuolating toxin of Helicobacter pylori inhibits maturation of procathepsin D and degradation of epidermal growth factor in HeLa cells through a partial neutralization of acidic intracellular compartments. J. Biol. Chem. 272, 25022–25028 (1997). 84. Censini, S. et al. Cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and diseaseassociated virulence factors. Proc. Natl Acad. Sci. USA 93, 14648–14653 (1996). The demonstration that virulent H. pylori contain a 40-kb foreign DNA (called pathogenicity island or PAI) coding for a type IV secretion system represented a milestone in the understanding of H. pylori pathogenesis and provided an explanation for the presence of the cagA gene only in a subset of strains. Also see reference 95. 85. Hacker, J. & Kaper, J. B. Pathogenicity islands and the evolution of microbes. Annu. Rev. Microbiol. 54, 641–679 (2000). 86. Ogura, K. et al. Virulence factors of Helicobacter pylori responsible for gastric diseases in Mongolian gerbil. J. Exp. Med. 192, 1601–1610 (2000). 87. Perez-Perez, G. I. et al. The role of CagA status in gastric and extragastric complications of Helicobacter pylori.

466

J. Physiol. Pharmacol. 50, 833–885 (1999). 88. Parsonnet, J. et al. Risk for gastric cancer in people with CagA positive or CagA negative Helicobacter pylori strains. Gut 40, 297–301 (1997). 89. Webb, P. M., Crabtree, J. E. & Forman, D. Gastric cancer, cytotoxin-associated gene A-positive Helicobacter pylori and serum pepsinogens: an international study, the Eurogast Study Group. Gastroenterology 116, 269–276 (1999). 90. Christie, P. J. & Vogel, J. P. Bacterial type IV secretion: conjugation systems adapted to effector molecules to host cells. Trends Microbiol. 8, 354–360 (2000). 91. Asahi, M. et al. Helicobacter pylori CagA protein delivered into the gastric epithelial cells can be tyrosine phosphorylated. J. Exp. Med. 191, 593–602 (2000). 92. Segal, E. D., Cha, J., Lo, J., Falkow, S. & Tompkins, L. S. Altered states: involvement of phosphorylated CagA in the induction of host cellular growth changes by Helicobacter pylori. Proc. Natl Acad. Sci. USA 96, 14559–14564 (1999). 93. Stein, M., Rappuoli, R. & Covacci, A. Tyrosine phosphorylation of Helicobacter pylori CagA after cagdriven host translocation. Proc. Natl Acad. Sci. USA 97, 1263–1268 (2000). 94. Odenbrait, S. et al. Translocation of Helicobacter pylori CagA into gastric epithelial cells by type IV secretion. Science 287, 1497–1500 (2000). References 91–94 each independently reported this unique and novel finding in bacterial pathogenesis. A bacterial protein (CagA) is introduced by a specialized secretion system (type IV secretion) into the host cell, which tyrosine phosphorylates the bacterial protein and initiates a cascade of events, which ultimately led to tissue damage and decrease. 95. Covacci, A. et al. Molecular characterization of the 128kDa immunodominant antigen of Helicobacter pylori

associated with cytotoxicity and duodenal ulcer. Proc. Natl Acad. Sci. USA 90, 5791–5795 (1993). 96. Naumann, M. et al. Activation of activator protein 1 and stress response kinases in epithelial cells colonized by Helicobacter pylori encoding the cag pathogenicity island. J. Biol. Chem. 274, 31655–31662 (1999). 97. Pomorski, T., Meyer, T. F. & Naumann, M. Helicobacter pylori-induced prostaglandin E2 synthesis involves activation of phospholipase A2 in epithelial cells. J. Biol. Chem. 276, 804–810 (2001). 98. Covacci, A. & Rappuoli, R. Tyrosine-phosphorylated bacterial proteins: Trojan horses for the host cells. J. Exp. Med. 191, 587–592 (2000). 99. Rappuoli, R. et al. Structure and mucosal adjuvanticity of cholera and Escherichia coli heat-labile enterotoxins. Immunol. Today 20, 493–500 (1999). 100. Ermark, T. H. et al. Immunization of mice with urease vaccine affords protection against Helicobacter pylori infection in the absence of antibodies and is mediated by MHC class II–restricted responses. J. Exp. Med. 188, 2277–2288 (1998). 101. Solnick, J. V. & Schauer, D. B. Emergence of diverse Helicobacter species in the pathogenesis of gastric and enterohepatic diseases. Clin. Microbiol. Rev. 14, 59–97 (2001).

Acknowledgements We apologize to the authors of the many relevant papers that we could not quote owing to space limitations. We thank A. Covacci, M. de Bernard, G. Del Giudice, W. Dundon, E. Papini, J. L. Telford and M. Zoratti for many discussions and for critical reading of the manuscript, and to G. Corsi for the artwork. Work carried out in the authors’ laboratories was supported by European Community grants, by the CNR–MURST 5% Project, by MURST 40% Projects on Inflammation and by the Armenise–Harvard Medical School Foundation.


