An exploratory study of Helicobacter suis control strategies

Helicobacter suis is a Gram-negative, spiral-shaped bacterium that colonizes the stomach of the majority of slaughter pigs worldwide. An infection with this microorganism has been associated with erosive and ulcerative lesions in the non-glandular part of the porcine stomach and with chronic gastritis. A reduction in daily weight gain in experimentally infected pigs has been described, emphasizing the importance of H. suis infections for the pig industry. Furthermore, it is the most prevalent non-H. pylori Helicobacter species colonizing the stomach of humans suffering from gastric disease. In humans, H. suis has been associated with gastritis, peptic ulcers and mucosa-associated lymphoid tissue lymphomas of the stomach. Pigs are considered to be an important source of human infections. Either contact with pigs or the consumption of raw or undercooked contaminated pork are proposed as source of infection for humans. H. pylori uses a variety of factors to persistently colonize the gastric mucosa and to induce gastric pathologies in humans. For a long time, the lack of pure H. suis in vitro isolates hampered the progress of research on this bacterium. At the start of this thesis, virtually nothing was known about the presence of genes involved in gastric pathogenicity of H. suis and in its specific adaptation to the gastric environment. Our research group was the first to successfully isolate H. suis in vitro in 2008. This created new opportunities to perform the present PhD studies which have the general aim to obtain better insights into the pathogenesis of H. suis infections and to develop control strategies against this bacterium. Effective measures to control H. suis infections may not only decrease the number of pigs suffering from gastric disease, but are also important from a public health point of view. The prevalence of H. suis infections in adult pigs is very high, indicating that the immune response after natural infection does not result in eradication of the bacterium from the stomach. Vaccination with whole-cell lysate was shown to protect against a subsequent H. suis infection in a mouse model. However, it was not known which bacterial proteins are important for induction of protective immunity against H. suis infections. In addition, at the start of this thesis, no data were available on the antimicrobial susceptibility pattern of H. suis. In Chapter 1, we describe the genome sequence of two H. suis strains: H. suis strain 1 (type strain = LMG 23995T) and 5, both isolated from the gastric mucosa of swine. After performing a draft pyrosequencing assay and assemblage, the genome was annotated by cross-mapping with three well-investigated H. pylori strains. As some virulence factors may differ between both species, ab initio annotations of the H. suis genome were performed as well. Comparison with the H. pylori genome revealed that all genes described to be essential for gastric colonization are present in the H. suis genome. Homologs of genes encoding the pro-inflammatory H. pylori neutrophil-activating protein (NapA), the apoptosis-inducing γ-glutamyl transpeptidase (GGT), and some type IV secretion systems were also identified in H. suis. H. suis also possesses several presumptive virulence-associated genes, including homologs for mviN, the H. pylori flavodoxin gene (fldA) and the H. pylori high-temperature requirement A gene (htrA). Sequences coding for outer membrane proteins involved in adhesion to gastric cells, such as H. pylori adhesin A (HpaA), HorB and some Hof members were present in H. suis as well. On the other hand, H. suis lacks homologs of several other H. pylori adhesion factors, including genes coding for the blood group antigen-binding adhesin BabA, the sialic acid-binding adhesin SabA and the adherence-associated lipoproteins AlpA and AlpB. It was concluded that although genes coding for some important virulence factors in H. pylori, such as the cytotoxin-associated protein A (CagA) and the vacuolating cytotoxin (VacA), were not detected in the H. suis genome, homologs of other genes associated with colonization and virulence of H. pylori and other bacteria were present. Previous studies in mice showed that H. suis infection does not result in protective immunity, whereas immunization with H. suis whole-cell lysate (lysate) protects against a subsequent experimental infection. In Chapter 2, we described for the first time immunoproteomics of H. suis. Two-dimensional gel electrophoresis of total H. suis proteins was performed followed by immunoblotting with pooled sera from H. suis-infected mice or mice immunized with lysate. Little reactivity against H. suis proteins was observed in post-infection sera. Sera from lysate-immunized mice, however, showed immunoreactivity against a total of 19 protein spots which were identified using mass spectrometry (LC-MS/MS). The H. suis urease subunit B (UreB) showed most pronounced reactivity against sera from immunized mice and was not detected by sera from infected mice. Other identified proteins included H. suis chaperonin GroEL, urease subunit A, flagellin A and elongation factor G. Based on this analysis, the immunoreactive UreB was selected for further in vivo testing. As a control we included H. suis NapA, which has been previously described as a possible virulence factor but was not recognized by sera of mice immunized with whole-cell lysate, nor by sera from H. suis-infected mice. In a subsequent study, the protective efficacy of intranasal vaccination of BALB/c mice with the H. suis UreB and NapA, both recombinantly expressed in E. coli (rUreB and rNapA, respectively), was compared with that of H. suis lysate. Cholera toxin was used as adjuvant. We found that immunization with the rUreB and lysate induced a significant reduction of gastric H. suis colonization compared to non-vaccinated and H. suis-infected controls. Immunization with rNapA had no significant protective effect. Reduction in gastric bacterial load was correlated with increased mRNA expression levels of IL-17, IFN-γ, and antigen specific serum IgG. It was concluded that rUreB could be a promising vaccine candidate for the use in vaccines against H. suis infections. Also, we suggested that probably a combination of local Th1 and Th17 responses, complemented by antibody responses play a role in the protective immunity against H. suis infections. In Chapter 3, it was evaluated whether inclusion of an additional antigen could improve the protective efficacy of subunit vaccination in a mouse model. Mice were intranasally immunized with rUreB, H. suis GGT, recombinantly expressed in E. coli (rGGT) or a combination of both proteins, administered simultaneously or sequentially. Control groups consisted of non-immunized and non-challenged mice (negative controls), sham-immunized and H. suis-challenged mice (sham-immunized controls), and finally, H. suis lysate-immunized and H. suis challenged mice. Cholera toxin was used as mucosal adjuvant. All immunizations induced a significant reduction of gastric H. suis colonization, which was least pronounced in the groups immunized with rGGT and rUreB alone. Consecutive immunization with rGGT followed by rUreB and immunization with the bivalent vaccine improved the protective efficacy compared to immunization with single proteins, with a complete clearance of infection observed in 50% of the animals. Immunization with whole-cell lysate induced a similar reduction of gastric bacterial colonization compared to rGGT and rUreB in combinations. Gastric lesions, however, were less pronounced in mice immunized with combinations of rUreB and rGGT compared to mice immunized with whole-cell lysate. In conclusion, vaccination with a combination of rGGT and rUreB protected mice against a subsequent H. suis infection and was not associated with severe post-vaccination gastric inflammation, indicating that it may be a promising method for control of H. suis infections. In vitro antimicrobial susceptibility of H. suis can not be determined using standard assays, as this agent only grows in a biphasic medium with an acidic pH. Therefore, in Chapter 4, we developed a combined agar and broth dilution method to analyse the activity of nine antimicrobial agents (ampicillin, ceftiofur, clarithromycin, enrofloxacin, gentamicin, lincomycin, metronidazole, tetracycline, and tylosine) against nine H. suis isolates. After 48h microaerobic incubation, minimal inhibitory concentrations (MICs) were determined by software-assisted calculation of bacterial growth. Only for enrofloxacin a clear bimodal distribution of MICs was demonstrated, indicating acquired resistance in one strain, which showed an AGT → AGG (Ser → Arg) substitution at codon 99 of gyrA. For the other antimicrobial agents a monomodal distribution of MIC values was observed, indicating absence of acquired resistance. For ampicillin, MICs were higher than those described for other gastric Helicobacter species, possibly indicating reduced susceptibility of H. suis to this antibiotic. For 7 isolates, MICs of metronidazole were also high. The significance of these findings for treatment of H. suis infections is not clear. It was concluded that the assay developed in this study is suitable for determination of the antimicrobial susceptibility of H. suis isolates, although activity of acid sensitive antimicrobial agents may be higher than predicted from MIC endpoints. As the results of this study are based on a small number of strains, additional tests are needed to determine if these results reflect the susceptibility of the H. suis population. The present PhD studies confirm that H. suis plays a role in gastric pathology, as its genome contains genes known to be essential for gastric colonization as well as genes encoding proteins that probably play a role in the induction of gastric lesions. Our results represent the first steps in the development of an effective subunit vaccine against H. suis infections. Combined vaccination with the H. suis UreB and GGT is suggested as a promising vaccine strategy. Finally, we developed an in vitro assay to evaluate the intrinsic and acquired resistance of H. suis isolates, and showed that resistance to fluoroquinolones may occur in H. suis field strains