Richard L. Ferrero

Vitamin B6 Is Required for Full Motility and Virulence in Helicobacter pylori

ABSTRACT Despite recent advances in our understanding of how Helicobacter pylori causes disease, the factors that allow this pathogen to persist in the stomach have not yet been fully characterized. To identify new virulence factors in H. pylori, we generated low-infectivity variants of a mouse-colonizing H. pylori strain using the classical technique of in vitro attenuation. The resulting variants and their highly infectious progenitor bacteria were then analyzed by global gene expression profiling. The gene expression levels of five open reading frames (ORFs) were significantly reduced in low-infectivity variants, with the most significant changes observed for ORFs HP1583 and HP1582. These ORFs were annotated as encoding homologs of the Escherichia coli vitamin B6 biosynthesis enzymes PdxA and PdxJ. Functional complementation studies with E. coli confirmed H. pylori PdxA and PdxJ to be bona fide homologs of vitamin B6 biosynthesis enzymes. Importantly, H. pylori PdxA was required for optimal growth in vitro and was shown to be essential for chronic colonization in mice. In addition to having a well-known metabolic role, vitamin B6 is necessary for the synthesis of glycosylated flagella and for flagellum-based motility in H. pylori. Thus, for the first time, we identify vitamin B6 biosynthesis enzymes as novel virulence factors in bacteria. Interestingly, pdxA and pdxJ orthologs are present in a number of human pathogens, but not in mammalian cells. We therefore propose that PdxA/J enzymes may represent ideal candidates for therapeutic targets against bacterial pathogens. IMPORTANCE Approximately half of the world’s population is infected with H. pylori, yet how H. pylori bacteria establish chronic infections in human hosts remains elusive. From gene array studies, we identified two genes as representing potentially novel colonization factors for H. pylori. These genes encoded enzymes involved in the synthesis of vitamin B6, an important molecule for many metabolic reactions in living organisms. Little is currently known regarding vitamin B6 biosynthesis in human pathogens. We showed that mutant H. pylori bacteria lacking an enzyme involved in de novo vitamin B6 biosynthesis, PdxA, were unable to synthesize motility appendages (flagella) and were unable to establish chronic colonization in mice. Thus, this work identifies vitamin B6 biosynthesis enzymes as novel virulence factors for bacterial pathogens. Interestingly, a number of human pathogens, but not their mammalian hosts, possess these genes, which suggests that Pdx enzymes may represent ideal candidates for new therapeutic targets. Approximately half of the world’s population is infected with H. pylori, yet how H. pylori bacteria establish chronic infections in human hosts remains elusive. From gene array studies, we identified two genes as representing potentially novel colonization factors for H. pylori. These genes encoded enzymes involved in the synthesis of vitamin B6, an important molecule for many metabolic reactions in living organisms. Little is currently known regarding vitamin B6 biosynthesis in human pathogens. We showed that mutant H. pylori bacteria lacking an enzyme involved in de novo vitamin B6 biosynthesis, PdxA, were unable to synthesize motility appendages (flagella) and were unable to establish chronic colonization in mice. Thus, this work identifies vitamin B6 biosynthesis enzymes as novel virulence factors for bacterial pathogens. Interestingly, a number of human pathogens, but not their mammalian hosts, possess these genes, which suggests that Pdx enzymes may represent ideal candidates for new therapeutic targets.

Bacterial membrane vesicles deliver peptidoglycan to NOD1 in epithelial cells

Gram‐negative bacterial peptidoglycan is specifically recognized by the host intracellular sensor NOD1, resulting in the generation of innate immune responses. Although epithelial cells are normally refractory to external stimulation with peptidoglycan, these cells have been shown to respond in a NOD1‐dependent manner to Gram‐negative pathogens that can either invade or secrete factors into host cells. In the present work, we report that Gram‐negative bacteria can deliver peptidoglycan to cytosolic NOD1 in host cells via a novel mechanism involving outer membrane vesicles (OMVs). We purified OMVs from the Gram‐negative mucosal pathogens: Helicobacter pylori, Pseudomonas aeruginosa and Neisseria gonorrhoea and demonstrated that these peptidoglycan containing OMVs upregulated NF‐κB and NOD1‐dependent responses in vitro. These OMVs entered epithelial cells through lipid rafts thereby inducing NOD1‐dependent responses in vitro. Moreover, OMVs delivered intragastrically to mice‐induced innate and adaptive immune responses via a NOD1‐dependent but TLR‐independent mechanism. Collectively, our findings identify OMVs as a generalized mechanism whereby Gram‐negative bacteria deliver peptidoglycan to cytosolic NOD1. We propose that OMVs released by bacteria in vivo may promote inflammation and pathology in infected hosts.

Essential role of Helicobacter pylori gamma-glutamyltranspeptidase for the colonization of the gastric mucosa of mice.

Constitutive expression of γ‐glutamyltranspeptidase (GGT) activity is common to all Helicobacter pylori strains, and is used as a marker for identifying H. pylori isolates. Helicobacter pylori GGT was purified from sonicated extracts of H. pylori strain 85P by anion exchange chromatography. The N‐terminal amino acid sequences of two of the generated endoproteolysed peptides were determined, allowing the cloning and sequencing of the corresponding gene from a genomic H. pylori library. The H. pylori ggt gene consists of a 1681 basepair (bp) open reading frame encoding a protein with a signal sequence and a calculated molecular mass of 61 kDa. Escherichia coli clones harbouring the H. pylori ggt gene exhibited GGT activity at 37°C, in contrast to E. coli host cells (MC1061, HB101), which were GGT negative at 37°C. GGT activity was found to be constitutively expressed by similar genes in Helicobacter felis, Helicobacter canis, Helicobacter bilis, Helicobacter hepaticus and Helicobacter mustelae. Western immunoblots using rabbit antibodies raised against a His‐tagged‐GGT recombinant protein demonstrated that H. pylori GGT is synthesized in both H. pylori and E. coli as a pro‐GGT that is processed into a large and a small subunit. Deletion of a 700 bp fragment within the GGT‐encoding gene of a mouse‐adapted H. pylori strain (SS1) resulted in mutants that were GGT negative yet grew normally in vitro. These mutants, however, were unable to colonize the gastric mucosa of mice when orally administered alone or together (co‐infection) with the parental strain. These results demonstrate that H. pylori GGT activity has an essential role for the establishment of the infection in the mouse model, demonstrating for the first time a physiological role for a bacterial GGT enzyme.

Helicobacter pylori hspA‐hspB heat‐shock gene cluster: nucleotide sequence, expression, putative function and immunogenicity

All Helicobacter pylori isolates synthesize a 54 kDa immunodominant protein that was reported to be associated with the nickel‐dependent urease of H. pylori. This protein was recently recognized as a homologue of the heat‐shock protein of the GroEL class. The gene encoding the GroEL‐like protein of H. pylori (HspB) was cloned (plLL689) and was shown to belong to a bicistronic operon including the hspA and hspB genes. In Escherichia coli. the constitutive expression of the hspA and hspB genes was initiated from a promoter located within an IS5 insertion element that mapped upstream to the two open reading frames (ORFs). IS5 was absent from the H. pylori genome, and was thus acquired during the cosmid cloning process. hspA and hspB encoded polypeptides of 118 and 545 amino acid residues, corresponding to calculated molecular masses of 13.0 and 58.2 kDa, respectively. Amino acid sequence comparison studies revealed that, although H. pylori HspA and HspB proteins were highly similar to their bacterial homologues, the H. pylori HspA featured a striking motif at the C‐terminus. This unique motif consists of a series of cysteine and histidine residues resembling a nickel‐binding domain, which is not present in any of the other bacterial GroES homologues so far characterized. When the plLL689 recombinant plasmid was introduced together with the H. pylori urease gene cluster (plLL763) into an E. coli host strain, an increase of urease activity was observed. This suggested a close interaction between the HspA and HspB proteins and the urease enzyme, and a possible role for HspA in ihe chelation of nickel ions. The genes encoding each of the HspA and HspB polypeptides were cloned, expressed independently as proteins fused to the maltose‐binding protein (WIBP) and purified in large scale. The MBP‐HspA and MBP‐HspB fusion proteins were shown to retain their antigenic properties. Both HspA and HspB represent antigens that are specifically recognized by the sera from H. pylori‐infected patients. Whereas HspB was known to be immunogenic in humans, this is the first demonstration that HspA per se is also immunogenic as proteins fused to the maltose‐binding protein (WIBP) and purified in large scale. The MBP‐HspA and WlBP‐HspB fusion proteins were shown to retain their antigenic properties. Both HspA and HspB represent antigens that are specifically recognized by the sera from H, py/or/‐infected patients. Whereas HspB was known to be immunogenic in humans, this is the first demonstration that HspA per se is also immunogenic.

Immune modulation by bacterial outer membrane vesicles.

Gram-negative bacteria shed extracellular outer membrane vesicles (OMVs) during their normal growth both in vitro and in vivo. OMVs are spherical, bilayered membrane nanostructures that contain many components found within the parent bacterium. Until recently, OMVs were dismissed as a by-product of bacterial growth; however, findings within the past decade have revealed that both pathogenic and commensal bacteria can use OMVs to manipulate the host immune response. In this Review, we describe the mechanisms through which OMVs induce host pathology or immune tolerance, and we discuss the development of OMVs as innovative nanotechnologies.

Helicobacter pylori Induces MAPK Phosphorylation and AP-1 Activation via a NOD1-Dependent Mechanism

Helicobacter pylori rapidly activates MAPKs and transcription factors, NF-κB and AP-1, in gastric epithelial cells following host attachment. Activation of these signal transducers is largely dependent on the cag pathogenicity island (cagPAI)-encoded Type IV Secretion System. H. pylori was shown to translocate peptidoglycan through the Type IV Secretion System, which is recognized by the pathogen recognition molecule, NOD1, thus resulting in NF-κB activation. The mechanisms of H. pylori-induced MAPK and AP-1 activation, however, are less well defined and therefore, we assessed the contribution of NOD1 to their activation. For this, we used gastric epithelial cell lines, stably expressing siRNA to either NOD1 or a control gene. In siNOD1-expressing cells stimulated with cagPAI+ H. pylori, we observed significant reductions in p38 and ERK phosphorylation (p < 0.05), whereas the levels of Jnk phosphorylation remained unchanged. Consistent with a previous report, however, we were able to demonstrate NOD1-dependent Jnk phosphorylation by the invasive pathogen Shigella flexneri, highlighting pathogen-specific host responses to infection. We also show that NOD1 was essential for H. pylori induction of not only NF-κB, but also AP-1 activation, implying that NOD1 induces robust proinflammatory responses, in an attempt to rapidly control infection. Pharmacological inhibition of p38 and ERK activity significantly reduced IL-8 production in response to H. pylori, further emphasizing the importance of MAPKs in innate immune responses to the pathogen. Thus, for the first time we have shown the important role for NOD1 in MAPK and AP-1 activation in response to cagPAI+ H. pylori.

Reduced activation of inflammatory responses in host cells by mouse‐adapted Helicobacter pylori isolates

Helicobacter pylori strains that harbour the Cag pathogenicity island (Cag PAI) induce interleukin (IL)‐8 secretion in gastric epithelial cells, via the activation of NF‐κB, and are associated with severe inflammation in humans. To investigate the influence of Cag PAI‐mediated inflammatory responses on H. pylori adaptation to mice, a selection of H. pylori clinical isolates (n= 12) was cag PAI genotyped and tested in co‐culture assays with AGS gastric epithelial cells, and in mouse colonization studies. Six isolates were shown to harbour a complete cag PAI and to induce NF‐κB activation and IL‐8 secretion in AGS cells. Of the eight isolates that spontaneously colonized mice, six had a cag PAI– genotype and did not induce pro‐inflammatory responses in these cells. Mouse‐to‐mouse passage of the two cag PAI+‐colonizing strains yielded host‐adapted variants that infected mice with bacterial loads 100‐fold higher than those of the respective parental strains (P= 0.001). These mouse‐adapted variants were affected in their capacity to induce pro‐inflammatory responses in host cells, yet no changes in cag PAI gene content were detected between the strains by DNA microarray analysis. This work provides evidence for in vivo selection of H. pylori bacteria with a reduced capacity to induce inflammatory responses and suggests that such bacteria are better adapted to colonize mice.

Local immunoglobulin G antibodies in the stomach may contribute to immunity against Helicobacter infection in mice

BACKGROUND & AIMS Orogastric immunization of mice with Helicobacter antigens, together with a mucosal adjuvant (cholera toxin), has been shown to confer immunity in the Helicobacter felis infection model. The aim of the study was to investigate the humoral immune responses associated with immunity and to compare these with responses in H. felis-infected mice. METHODS Enzyme-linked immunoassays were used to characterize the antibody-secreting cells and antibodies present at mucosal and systemic sites in mice. Animals were immunized orally with either whole-cell H. felis sonicates or Helicobacter pylon urease or heat-shock proteins. RESULTS Infection of mice with H. felis preferentially induced the recruitment of plasma cells committed to immunoglobulin (Ig) A synthesis in salivary gland and gastric tissues. Antigen-specific IgA was the major antibody class detected in mucosal secretions recovered from these tissues. In contrast, immunization of mice against H. felis infection induced the proliferation of large numbers of IgG-secreting cells, as well as the synthesis of local IgG antibodies, in the gastric mucosa of the animals. Protection against H. felis infection occurred in the absence of gastric IgA responses in sonicate-immunized mice. CONCLUSIONS It is proposed that locally synthesized specific IgG antibodies contribute to immunity against gastric Helicobacter infection.

The innate immune molecule, NOD1, regulates direct killing of Helicobacter pylori by antimicrobial peptides

The cytosolic innate immune molecule, NOD1, recognizes peptidoglycan (PG) delivered to epithelial cells via the Helicobacter pylori cag pathogenicity island (cagPAI), and has been implicated in host defence against cagPAI+H. pylori bacteria. To further clarify the role of NOD1 in host defence, we investigated NOD1‐dependent regulation of human β‐defensins (DEFBs) in two epithelial cell lines. Our findings identify that NOD1 activation, via either cagPAI+ bacteria or internalized PG, was required for DEFB4 and DEFB103 expression in HEK293 cells. To investigate cell type‐specific induction of DEFB4 and DEFB103, we generated stable NOD1‘knockdown’ (KD) and control AGS cells. Reporter gene assay and RT‐PCR analyses revealed that only DEFB4 was induced in an NOD1‐/cagPAI‐dependent fashion in AGS cells. Moreover, culture supernatants from AGS control, but not AGS NOD1 KD cells, stimulated with cagPAI+H. pylori, significantly reduced H. pylori bacterial numbers. siRNA studies confirmed that human β‐defensin 2 (hBD‐2), but not hBD‐3, contributes to the antimicrobial activity of AGS cell supernatants against H. pylori. This study demonstrates, for the first time, the involvement of NOD1 and hBD‐2 in direct killing of H. pylori bacteria by epithelial cells and confirms the importance of NOD1 in host defence mechanisms against cagPAI+H. pylori infection.

The mouse colonizing Helicobacter pylori strain SS1 may lack a functional cag pathogenicity island.

The SS1 strain of Helicobacter pylori was first described in 1997 [1], and has since then been extensively used in experimental infections of mice. The strain colonises the gastric mucosa at a high density and retains colonising ability after subculture for 20 passages [1]. The strain is cagA positive as determined by PCR [1], and generally assumed to be a good strain for analysis of the function of the cag pathogenicity island ( cag PAI) in experimental models. It is well documented that epithelial cell signalling responses induced by cag PAI positive strains of H. pylori , which are characterised by NFκ B activation and IL-8 induction, are dependent on multiple genes throughout the cag PAI [2–5]. It has been reported that SS1 failed to induce IL-8 secretion in gastric epithelial cells [6], and this was recently confirmed by Eaton et al. [7], and in two of the authors’ laboratories (JEC and RLF, unpublished data). Only one study to date has demonstrated that the SS1 strain induces IL-8 secretion in gastric epithelial cells [8]. We have previously demonstrated that inactivation of ORF7 (HP0521) reduced IL-8 transcription in gastric epithelial cells to the level observed with a total cag PAI deletion mutant [4]. Although polar effects on downstream genes can not be completely excluded, these data indicate that ORF7 is essential for epithelial cell signalling. Recently whole-genome microarray analysis demonstrated that SS1 lacks ORF7 (HP0521) in the left half of the cag PAI [9]. The absence of ORF7 in the SS1 isolate may therefore explain the observations on the lack of a functional cag PAI of the SS1 isolates tested in four of five independent laboratories around the world. Genomic comparisons have revealed considerable diversity in the ORF7 genes of H. pylori strains 26695 and J99 [10]. These genes were reported to be composed of 240 and 657 base pairs, respectively. The relatively short length of ORF7 in H. pylori 26695, however, can be partially attributed to the presence of a frame-shift in the published sequence, resulting in a gene with greater similarity to the 348-base pair homologue of H. pylori NCTC11638 [3]. Given that the published ORF7 sequence of H. pylori 26695 was used in the DNA microarray analyses of the SS1 strain [9], classical sequence analysis is required to confirm the absence of the ORF7 gene in the SS1 strain. Moreover, similar analyses on the various H. pylori SS1 isolates throughout the world would contribute to a clarification of the role of ORF7 in cag PAI function. These findings illustrate the importance of genome information for the interpretation of data emanating from DNA microarray analysis of H. pylori isolates. There is a marked diversity in host gene expression in gastric epithelial cells following exposure to cag PAI positive and cagPAI negative H. pylori strains [11]. Studies in animal models with isogenic cagPAI mutants and parental strains are critical to investigate the in vivo relevance of the cag PAI in pathogenesis. Recently, successful studies on the in vivo function of the cag PAI have been undertaken in the Mongolian gerbil model [12,13]. Given the high mutation rate of H. pylori during repetitive culture [8,14], and the large number of cagPAI genes required for induction of epithelial cell signalling responses [2–5], it is important that bioactivity of parental strains be examined before construction of isogenic mutants and animal experimentation. The H. pylori SS1 strain has proven to be an excellent strain for use in the mouse colonisation model, and has permitted in vivo studies on a variety of bacterial functions. Nevertheless, as suggested by recent studies, the likely absence of a functional cag PAI in this strain suggests that it is not suitable for determining the functional importance of the cag PAI in animal models. Recent studies suggest changes in cag genotype or in cag PAI related functions can occur during passage in mice [15,16]. Whilst the significance of such host induced adaptive changes is unclear, these data do indicate that a functional cag PAI may not be required for mouse colonisation. Additional evidence for this is in the recently reported isolation of a new mouse colonizing H. pylori strain (SS2000): from all 125 clinical H. pylori isolates tested, strain SS2000 persisted at highest numbers in mice, and in spite of the total absence of a cag PAI it clearly induced inflammation in both C57BL/6 and BALB/c mice [17]. Taken together this indicates that the mouse may not be the optimal species for investigating the functional importance of the cag PAI.