Diagnosis of Helicobacter Heilmannii sensu stricto infections in dogs, cats and humans

Helicobacter (H.) pylori is the most prevalent Helicobacter species colonizing the gastric
mucosa of humans and has been associated with gastritis, peptic ulcer disease and gastric
cancer. This slightly-curved, Gram-negative and microaerophilic gastric bacterium is highly
adapted to the human host and has succeeded to colonize more than 50% of the human
population worldwide. Besides H. pylori, other spiral-shaped helicobacters, naturally
colonizing the stomach of animals, have also been detected in the stomach of humans. These
non-H. pylori Helicobacter (NHPH) species have been associated with gastritis, gastric and
duodenal ulcers and low-grade mucosa-associated lymphoid tissue (MALT) lymphoma in
humans. Moreover, the risk of developing MALT lymphoma is considered to be higher after
infection with NHPH than with H. pylori. NHPH represents a group of closely related but
distinct Helicobacter species that are found in different animal species, such as H. felis, H.
salomonis, H. bizzozeronii, H. heilmannii sensu stricto (s.s.), H. cynogastricus and H.
baculiformis in cats and dogs and H. suis in pigs. NHPH bacteria have an extremely fastidious
nature, which so far has resulted in a limited number of in vitro isolates available worldwide.
In 2011, our research group was the first to successfully isolate H. heilmannii s.s. from the
feline stomach and cultivate this species in vitro. The general aims of this PhD research were
to obtain better insights into the pathogenesis of human gastric disease associated with H.
heilmannii s.s. infection, to determine the possible virulence-associated factors of this
microorganism, and to compare its outer membrane protein repertoire with other NHPH.
In chapter one, we used an in vivo Mongolian gerbil model to study the bacterium-host
interactions of 9 different feline H. heilmannii s.s. strains, identified on the basis of their 16S
rRNA and ureAB genes. At 9 weeks after experimental infection, the colonization levels in the
stomach, the intensity of the induced gastritis, the rate of gastric epithelial cell alterations and
the expression level of the peptide hormone gastrin were determined. In addition, the immune
response following H. heilmannii s.s. infection was characterized by measuring the expression
levels of various inflammatory cytokines in the stomach.
We showed the induction of an antrum-dominant chronic active gastritis with formation of
lymphocytic aggregates after infection with 7 out of 9 H. heilmannii s.s. strains. In addition, a
high number of proliferating epithelial cells was seen in the antrum of the gerbil stomach. After
infection with the 2 other strains, no explicit antral inflammation and no increased gastric
epithelial cell proliferation were observed. In all H. heilmannii s.s.-infected gerbils, only limited
signs of inflammation were detected in the fundus of the stomach, and the epithelial cell
proliferation rate was not elevated in this stomach region.
With quantitative PCR, a high-level antral colonization was revealed for 4 H. heilmannii s.s.
strains, while colonization of 4 other strains was more restricted and one strain was not detected
in the stomach at 9 weeks post infection. In general, the colonization capacity in the fundus was
lower than in the antrum for all strains tested and the lowest number of bacteria was detected
in the duodenum. A reduced expression of the gastric hydrogen potassium (H+ /K+ ) ATPase,
functioning as a proton pump in gastric acid-secreting parietal cells, was detected in the antrum
of the stomach after infection with 3 highly colonizing H. heilmannii s.s. strains. This indicates
a possible loss of parietal cells and a reduced gastric acid secretion. However, no significant
change in H+ /K+ ATPase expression was seen in the fundus, where the majority acid-secreting
parietal cells are located. Two highly colonizing strains caused an increased expression of the
peptide hormone gastrin in the fundus of the stomach, which may lead to hypergastrinemia.
Although the gastrin-producing G-cells are mainly located in the antrum of the stomach, no
antral upregulation of gastrin expression was detected.
All H. heilmannii s.s. strains that induced a chronic active gastritis, caused an upregulation of
IL-1U+03B2 in the antrum. This pro-inflammatory cytokine is a potent inhibitor of gastric acid
secretion and plays a role in the acute phase of inflammation. Remarkably, there was no
upregulation of typical Th1 or Th2 cytokines in the stomach of the gerbils at 9 weeks after H.
heilmannii s.s. infection.
In conclusion, our experimental infection study in Mongolian gerbils indicates diversity in
bacterium-host interactions and virulence between 9 different H. heilmannii s.s. strains. Based
on the differences in colonization capacity and the level of antral inflammation, 5 strains were
shown to be highly virulent. The 4 other strains were low virulent. Since the Mongolian gerbil
model is considered to be a good model for human Helicobacter-induced pathology, this strain
variation is most probably also relevant for human infections with this microorganism and
might be important for infections in the natural hosts of H. heilmannii s.s. as well.
In chapter 2, we investigated if the differences in colonization capacity and virulence properties
between the 9 H. heilmannii strains were related to the presence/absence of specific virulenceassociated genes. Therefore, the genomes of the strains were sequenced to investigate their
phylogenetic relationships and to define their gene content and diversity. In addition, several in
vitro binding assays were carried out to determine possible strain differences in the capacity for
adhesion to the gastric mucosa.
Via phylogenetic analysis and the use of several bio-informatics tools, we showed that the 4
isolates with a lower virulence in Mongolian gerbils actually do not belong to the H. heilmannii
species, but to a new, closely-related species which we named H. ailurogastricus. This new
Helicobacter species cannot be distinguished from H. heilmannii by means of its 16S rRNA
and ureAB gene sequences. In addition, we showed that these feline Helicobacter species are
morphologically very similar, with the exception that H. heilmannii has more spiral turns and
bipolar flagella than H. ailurogastricus. Also the biochemical characteristics of H.
ailurogastricus were very similar to those of H. heilmannii, with the only difference that this
new species presents alkaline phosphatase activity, which is absent in H. heilmannii.
We found that several homologs of H. pylori virulence-associated factors are present in H.
heilmannii but absent in H. ailurogastricus. These include the ulcer-associated protein IceA1;
the HrgA protein, which is part of a DNA restriction-modification system in virulent H. pylori
strains; and a jhp0562-like glycosyltransferase that is involved in immune evasion and peptic
ulcer development. Possibly, these factors contribute to the more virulent character of H.
heilmannii compared to H. ailurogastricus.
Similar to H. pylori, our analysis revealed a large outer membrane protein (OMP) repertoire in
both H. heilmannii and H. ailurogastricus. However, both feline species shared only a few
homologs of the H. pylori Hop, Hor, and Hom proteins and lacked all H. pylori adhesins
described so far. Interestingly, H. heilmannii harbored 6 putative OMPs that are absent in H.
ailurogastricus. These OMPs might play a role in the colonization process of H. heilmannii.
Furthermore, we identified a homologous gene of the H. pylori VacA-like autotransporter,
which enhances the capacity to colonize the stomach, in both H. heilmannii and H.
ailurogastricus. In H. pylori, the passenger domain of the VacA-like autotransporter, which
confers the effector function of this protein, contains three VacA2 regions. The passenger
domains of the VacA-like autotransporters of H. heilmannii also contained three VacA2
regions, whereas in H. ailurogastricus, this domain contained four VacA2 regions. Since H.
ailurogastricus has a lower in vivo colonization ability than H. heilmannii, the size of the
passenger domain might play a role in gastric colonization.
The differences in in vivo colonization capacity between H. heilmannii and H. ailurogastricus
were also reflected in their in vitro capacity to adhere to the gastric mucosa. We noted a trend
towards higher binding of H. heilmannii to human gastric mucins than H. ailurogastricus. H.
heilmannii also had a higher ability to bind human- and mouse-derived gastric epithelial cells.
Additionally, H. heilmannii bound mainly to the glandular cells of both the antrum and the
corpus of the stomach, whereas H. ailurogastricus had a higher binding capacity to the surface
epithelial cells lining the gastric mucosa.
In conclusion, we described a new feline gastric species H. ailurogastricus, which is closely
related to H. heilmannii. H. ailurogastricus lacks several homologs encoding H. pylori
virulence and colonization factors and has a lower capacity for binding to gastric epithelial cells
in vitro. This may explain why its virulence is lower than that of H. heilmannii.
The outer membrane of H. pylori is equipped with a large set of OMPs and several of these
proteins are involved in colonization of the human gastric mucosa. In contrast, it is largely
unknown which OMPs play a role in the colonization process of H. heilmannii, H.
ailurogastricus and other gastric NHPH. These species lack all important H. pylori Hop
adhesins, indicating that other OMPs are used for adhesion to the gastric mucosa.
In chapter 3 of this thesis, we characterized the OMP repertoire of gastric NHPH by using
phylogenetic analyses and we identified the OMP families that are possibly involved in their
colonization and virulence properties.
Similar to H. pylori, we found that gastric NHPH harbor proportionally more OMPs than other
Gram-negative bacteria, which might be the result from an adaptation to the harsh gastric
environment.
Several well-conserved OMP families, also present in Campylobacter, E. coli and enterohepatic
helicobacters, and with a possible function in colonization or bacterial virulence, were identified
in gastric NHPH. These include TonB-dependent receptors with a function in iron uptake, an
outer membrane factor involved in antimicrobial resistance and an outer membrane
phospholipase that plays a role in colonization. A porin Imp/OstA, involved in antimicrobial
resistance, was well-conserved among the Campylobacter and Helicobacter genera.
Interestingly, several OMP families were detected only in Helicobacter and primarily in gastric
species. These included SfpA/LpxR that functions in immune evasion, two Helicobacterspecific outer membrane porin families with probable functions in adhesion, and a
Helicobacter-specific VacA-like cytotoxin (characterized in chapter 2) with a role in
colonization capacity. A significant part of all OMPs of gastric NHPH clustered into the two
Helicobacter-specific porin families. The first porin family contained the largest number of
OMPs, including the H. pylori Hop adhesins that are absent in NHPH, except for H. acinonychis
and H. cetorum. The H. pylori Hor, Hom and other Hop OMPs also belong to this porin family,
and a few homologs of these proteins were present in other gastric NHPH. A last subgroup
contained putative OMPs that were specific for canine, feline and porcine NHPH. The finding
of a unique porin repertoire, together with the absence of the H. pylori Hop adhesins, indicates
that gastric NHPH use other OMPs to colonize the gastric mucosa and that their pathogenesis
is different from H. pylori.
The second Helicobacter-specific porin family was composed of the 8 H. pylori Hof OMPs that
were highly conserved among gastric NHPH. In contrast to H. pylori, the hof genes of NHPH
were located in a large unique locus. The function of most of these Hof proteins are currently
unknown with only HofE and HofF being involved in gastric colonization.
Interestingly, various species-specific OMP families were found in canine and feline NHPH,
that are absent in other Helicobacter species. The function and the importance of these putative
proteins remain to be elucidated.
In conclusion, we showed that H. heilmannii s.s., isolated from the feline stomach, is capable
of inducing gastric disease in Mongolian gerbils, an in vivo model for human gastric pathology.
In addition, we described a new feline species H. ailurogastricus that is closely related to H.
heilmannii. By using both in vitro and in vivo studies, we showed that H. ailurogastricus is less
virulent than H. heilmannii. We identified several virulence-associated factors in H. heilmannii
that might be related to the difference in pathogenicity between these two feline Helicobacter
species. Finally, we characterized the OMP repertoire of all gastric NHPH, which will be of
great value for future in vitro and in vivo colonization studies.