Umeå universitets logga

Helicobacter pylori (H. pylori) is a human-specific gastric pathogen which is responsible for a spectrum of diseases ranging from superficial gastritis to gastric and duodenal ulceration, and which is also highly associated with gastric cancer. The pathogenesis of severe gastric disorders caused by H. pylori is multifactorial and involves complex interactions between the microbe and the gastric mucosa. H. pylori expresses several adhesion proteins. These molecules have important roles in the establishment of persistent infection and chronic inflammation, which cause tissue damage.

The aim of this thesis was to study the attachment of this bacterium to human gastric epithelium, mediated by blood group antigens in both health and disease. One of the bestcharacterized H. pylori adhesins is the histo-blood group antigen binding adhesin (BabA), which binds specifically to the Lewis b antigen (Leb) in the gastric mucosa.

A protective mucus layer lines the stomach. The mucosal glycosylation patterns (GPs) vary between different cell lineages, different locations along the gastrointestinal (GI) tract and different developmental stages. In addition, GPs undergo changes during malignant transformation. MUC5AC is a mucin molecule produced by the surface epithelium. Three distinctly different types of human gastrointestinal tissue were studied by bacterial adherence analysis in situ. MUC5AC is the most important carrier of Leb and the new results demonstrate that it forms major receptors for H. pylori adherence.

By analysing an H. pylori babA-deletion mutant, a novel adhesin-receptor binding mode was found. Surprisingly, the mutant bound efficiently to both human gastric mucosa and to gastric mucosa of Leb transgenic mice. The sialylated and fucosylated blood group antigen, sialyl-dimeric-Lewis x (sdiLex), was structurally identified as the new receptor. A positive correlation was found between adherence of H. pylori to sialyl-Lewis x (sLex) and elevated levels of inflammation response in the human gastric mucosa. These results were supported by detailed analysis of sialylated and fucosylated blood group antigen glycosylation patterns and, in addition, in situ bacterial adherence to gastric mucosa of experimentally challenged Rhesus monkey. The cognate sialic acid-binding adhesin (SabA) was purified by the retagging technique, and the corresponding sabA-gene was identified.

H. pylori lipopolysaccharide (LPS) contains various Lewis blood group antigens such as Lewis x (Lex) and Lewis y (Ley). Additional bacterial adherence modes, which are independent of the BabA and/or SabA adhesins, could possibly be mediated by Lex interactions. Adherence of a clinical isolate and its corresponding Lex mutant to human gastric mucosa with various gastric pathologies was studied in situ. The results suggest that H. pylori LPS plays a distinct but minor role in promotion of bacterial adhesion.

Taken together, the results suggest mechanisms for continuous selection of H. pylori strains, involving capacity to adapt to changes in the local environment such as shifts in cell differentiation and associated glycosylation patterns. Adherence of H. pylori is dependent on both the BabA and the SabA adhesin. Multi-step dependent attachment mechanisms may direct the microbes to distinct ecological niches during persistent infections, driving the chronic inflammation processes further toward the development of peptic ulcer disease and/or malignant transformation.

Key words: H. pylori, BabA, adhesin, Lewis b, MUC5AC, sialyl-dimeric-Lewis x, chronic inflammation, SabA, Lewis x, LPS.

Being among the most prevalent of persistent infectious agents in humans worldwide, Helicobacter pylori induces chronic inflammation (gastritis), which may progress to peptic ulceration and stomach cancer. The ability to adhere to the gastric mucosa is considered to be both a colonization and virulence property of H. pylori. For adherence, H. pylori expresses surface-located attachment proteins (adhesins) that bind to specific receptors in the gastric mucosa. The best characterized H. pylori adhesin-receptor interaction is that between the blood group antigen binding adhesin (BabA) and the fucosylated blood group antigens, which are glycans highly expressed in the gastric mucosa.

Our recent results have changed the view of the blood group antigen-specific binding mode of H. pylori. We have tested clinical isolates of H. pylori from human populations worldwide for their ability to bind to ABO blood group antigens. The results revealed that more than 95% of isolates from Sweden, Germany, Spain, Japan and Alaska that bind fucosylated blood group antigens, bind both the Lewis b antigen (Leb) (of blood group O) and the blood group A-related antigen A-Lewis b, i.e. they exhibit a generalist type of binding mode. In contrast, the majority of strains (62%) from South American Amerindians bound best to Leb, i.e. they exhibit a specialist blood group “O antigen” binding mode. This specialization in binding coincides with the unique predominance of blood group O in the South American Amerindian populations. Furthermore, we also showed that H. pylori could switch from specialist to generalist binding modes by chromosomal integration of foreign babA gene fragments.

A mutant strain lacking the babA gene turned out to adhere to inflamed gastric epithelium, despite the fact that it did not bind Leb. We identified the receptor to which the mutant binds to as the sialyl-dimeric-Lewis x antigen (sdiLex) and found its expression to be associated with persistent H. pylori infection and chronic inflammation, both in humans and Rhesus monkeys. The cognate sialic acid binding adhesin (SabA) was identified by our ReTagging technique. Deletion of sabA caused loss of H. pylori binding to sialylated glycans, and screening of single colony isolates revealed a high frequency of spontaneous on⇒off phase variation in sLex binding.

Using erythrocytes as a model for sialyl dependent cell adhesion, we could show that SabA is the sought-after H. pylori sialyl-dependent hemagglutinin. Swedish clinical H. pylori isolates were analyzed for sialyl-dependent hemagglutination (sia-HA), and the sia-HA titers were found to be highly correlated to the levels of sLex binding. Clinical isolates were shown to exhibit several distinct binding modes for sialylated glycans, which suggest that SabA exhibit polymorphism in binding. We also found that SabA binds to sialylated glycans on neutrophil surfaces by mechanisms involving “selectin mimicry”, and that SabA plays an important role in nonopsonic activation of neutrophils.

In the human stomach, H. pylori is exposed to selective pressures such as immune and inflammatory responses, and this is reflected by changes in mucosal glycosylation patterns. The high mutation and recombination rates of H. pylori in combination with bio selection will continuously generate clones that are adapted to changes in individual gastric mucosa. Such adaptive selection contributes to the remarkable diversity in binding modes and to the extraordinary chronicity of H. pylori infections worldwide.

The gastric pathogen Helicobacter pylori infects more than half of the population worldwide. H. pylori manage to establish persistent infection, which would be life-long if not treated. In order to establish such an infection, this pathogen has to deal with the host immune system. H. pylori has certain characteristics which make the bacteria less announced to the host immune system. Additionally, for remaining in the harsh and acidic environment of the stomach with peristaltic movements and a high frequency of turnover of epithelial cells, H. pylori has developed different binding modes to structures present both in the mucus and on the surface of gastric cells and also to extracellular matrix proteins. Evidently, adhesion has a determinant role for a successful colonization by H. pylori. It has been shown that a small fraction of the H. pylori infection is in intimate contact and attached to the host epithelium. Despite its small proportion, this group maintains the persistency of infection. As there is no suitable in vitro system to mimic the human stomach for studies of H. pylori infection, we have developed the In Vitro Explant Culture technique (IVEC). By using this model we could show that H. pylori use the Lewis b blood group antigen to bind to the host gastric mucosa, during experimental conditions most similar to the in vivo situation. Furthermore, we could show that the host tissue responses to the bacterial attachment by expression of Interleukin 8 (IL- ), which will guide the inflammatory processes. Interestingly, by inhibition of bacterial adhesion through receptor competition i.e., by use of soluble Lewis b antigen, IL-8 production was hampered in the IVEC system, which further validates the presence of a tight relation between bacterial adhesion and induction of host immune responses. One of the inflammation signaling cursors in vivo is the upregulated sialylated Lewis x (sLex) antigen, an inflammation associated carbohydrate structure well established as a binding site for the selectin family of adhesion molecules. We could show that during chronic gastric inflammation, which is actually caused by the persistent H. pylori infection, the bacterial cells adapt their binding mode, and preferentially bind to sLex, which will provide an even more intimate contact with the host cells. This interaction is mediated by SabA, the H. pylori adhesin for sialylated oligosaccharides/glycoconjugates. By employing red blood cells as a model we could further demonstrate that SabA is identical to the “established” H. pylori hemagglutinin. We could also show that SabA binds to sialylated glycolipids (gangliosides) rather than glycoproteins on cell surfaces. Our result also revealed that SabA also binds to and activates human neutrophils. Such effect was unrelated to BabA and the H. pylori Neutrophil Activating Protein (HP- AP), which were not directly involved in the activation of neutrophils. Furthermore, phagocytosis of bacteria by neutrophils was demonstrated to be mainly dependent on presence of SabA. Interestingly, HP-NAP showed a possible role in guiding the bacterial adhesion during conditions of limited sialylation, i.e. equivalent to mild gastritis, when the tissue would be less inflamed and sialylated. In conclusion, H. pylori adhesion causes host tissue inflammation, then the bacteria will adapt to the new condition and bind to epithelial cells in a tighter mode by synergistic activities of BabA and SabA. Additionally, SabA bind to and activate human neutrophils, which will exacerbate inflammation responses and cause damage to host tissue. Thus, BabA and SabA are potential candidates to be targeted for therapeutic strategies against H. pylori and gastric disease.