The attachment of Helicobacter pylori to the human gastric mucosa is a complex process involving several speci®c structures recognised by the cell surface receptors. Sialylated multivalent high mol. wt glycoproteins have been shown to inhibit H. pylori sialic acidspeci®c haemagglutination. This study explored whether sialylated glycoconjugates from bovine milk could inhibit an experimental H. pylori infection in a mouse model. BALB=cA mice (6±8 weeks old) were inoculated with a mouse-passaged H. pylori strain 317p. Four weeks after infection the mice were given lactoferrin (iron-free LF or 20% iron-saturated LF) or bovine milk fat globule membrane fractions (MFGM or defatted MFGM) orally (400 mg=kg body weight) once daily for 10 days and then killed to examine for bacterial colonisation and gastritis. Mice treated with iron-free LF, 20% iron-saturated LF, MFGM or defatted MFGM showed 30%, 10%, 20% or 20% healing rates, respectively, when compared with the H. pylori-infected control. Gastric colonisation by H. pylori was remarkably decreased in all mice treated with bovine milk glycoconjugates and the in¯ammation score was also signi®cantly lower in treated mice than in infected control animals. The fact that there was no signi®cant difference between iron-free LF and iron-saturated LF or MFGM and defatted MFGM suggested that iron is not crucial for inhibition of H. pylori by lactoferrin and that the lipid part of MFGM is not important for anti-H. pylori activity. In conclusion, bovine milk glycoconjugates showed potencies to inhibit H. pylori infection in this mouse model and, therefore, could be considered as candidates for non-antibiotic strategies against H. pylori infection in man.

Helicobacter pylori is a human pathogen associated with gastritis and peptic ulcer. Adhesion properties of H. pylori to various structures have been described in the literature, including evidence for sialic acid-binding. To study the specificity and frequency of sialic acid-binding, fourteen H. pylori strains were investigated using haemagglutination with derivatized erythrocytes carrying sialic acids only on defined glycans and using haemagglutination inhibition assays. From these studies H. pylori strains can be grouped into sialic acid-dependent and sialic acid-independent classes. The sialic acid-dependent strains require alpha-2,3-linked sialic acid for haemagglutination. The potential roles of sialic acid-dependent adhesions for H. pylori-related infections are discussed.

Unlike Helicobacter felis and other Helicobacter species of animal origin, Helicobacter pylori colonizes the lower gastric mucin layer of the stomach and adheres to human gastric epithelial cells. It is still an open question if H. pylori can interact with specific glycoconjugates in the gastric mucin layer. It is possible that colonization of the oral cavity is a first step of a complex infectious process. Most likely resting or slow growing cells of H. pylori interact with Lewis blood group substances in the gastric mucin layer and on the epithelium. This initial colonization is probably followed by binding to specific cell surface glycoconjugates (glycoproteins and glycolipids such as GM3) and specific sialylated or highly sulphated molecules such as cell surface sulphatides and heparan sulphate. H. pylori may also bind to specific phospholipid molecules such as phosphatidylethanolamine on the gastric cells. The adhesion process of certain strains can stimulate 'close' cell adhesion including pedestal formation similar to the phenomenon typical for a special class of enterovirulent Escherichia coli called attaching effacing E. coli. After gastric cell destruction by ammonia and H. pylori toxins (such as the vacuolating toxin) H. pylori may colonize the extracellular matrix (ECM). This phenomenon seems to include binding of cell surface sialic acid specific haemagglutinin to one ECM component, i.e. laminin. It is also likely that H. pylori may use similar events to penetrate intercellular junctions of gastric epithelial cells. These adhesion‐ penetration phenomena also involve coating of the microbe with host proteins to escape the host immune system and initiate a chronic lifelong infection process.

Plant and animal lectins with various carbohydrate specificities were used to type 35 Irish clinical isolates of Helicobacter pylori and the type strain NCTC 11637 in a microtiter plate assay. Initially, a panel of eight lectins with the indicated primary specificities were used: Anguilla anguilla (AAA),Lotus tetragonolobus (Lotus A), and Ulex europaeus I (UEA I), specific for α-l-fucose;Solanum tuberosum (STA) and Triticum vulgaris(WGA), specific for β-N-acetylglucosamine; Glycine max (SBA), specific for β-N-acetylgalactosamine;Erythrina cristagali (ECA), specific for β-galactose and β-N-acetylgalactosamine; and Lens culinaris(LCA), specific for α-mannose and α-glucose. Three of the lectins (SBA, STA, and LCA) were not useful in aiding in strain discrimination. An optimized panel of five lectins (AAA, ECA, Lotus A, UEA I, and WGA) grouped all 36 strains tested into eight lectin reaction patterns. For optimal typing, pretreatment by washing bacteria with a low-pH buffer to allow protein release, followed by proteolytic degradation to eliminate autoagglutination, was used. Lectin types of treated samples were stable and reproducible. No strain proved to be untypeable by this system. Electrophoretic and immunoblotting analyses of lipopolysaccharides (LPSs) indicated that the lectins interact primarily, but not solely, with the O side chain of H. pylori LPS.

Lactobacillus isolates from healthy Estonian and Swedish children were characterised by a lectin typing technique; 56 isolates from six species (L. acidophilus, L. paracasei, L. plantarum, L. fermentum, L. brevis and L. buchneri) were tested. The typing system was based on an agglutination assay with a panel of six commercially available lectins, which were chosen on the basis of their carbohydrate speci®cities. The isolates were also subjected to proteolytic degradation before lectin typing to decrease auto-agglutination of whole cells in the assay. The 56 isolates were divided into 15 different lectin types by their lectin agglutination patterns. Proteolytic treatment reduced auto-agglutination for the majority of species, apart from L. acidophilus, which remained predominantly autoagglutinating (eight of nine strains). The system produced stable and reproducible results under standardised culture conditions. Lactobacilli are important bacteria for use as probiotics and this system may supplement current molecular typing techniques and may help in identi®cation of strains that could be useful in this role.

No abstract

Effects on aqueous extracts of medicinal plants on ten Helicobacter pylori strains were studied by the salt aggregation test to determine the possibility to modulate their cell surface hydrophobicity and by an agar diffusion assay for detection of antimicrobial activity. It was established that aqueous extracts of bearberry and cowberry leaves enhance cell aggregation of all H. pylori strains tested by the salt aggregation test, and the extract of bearberry possessed a remarkable bacteriostatic activity. Pure tannic acid showed a result similar to that of bearberry and cowberry extracts which contained a large amount of tannins. In contrast, extracts of wild camomile and pineapple-weed, which blocked aggregation of H. pylori, contained small amounts of tannins and did not reveal any antimicrobial activity. Tannic acid seems to be the component of bearberry and cowberry aqueous extracts with the highest activity to decrease cell surface hydrophobicity as well as in antibacterial activity against H. pylori.

The heparin-binding properties of six different species of coagulase-negative staphylococci were examined by a particle agglutination assay. Heparin (mol. wt 4000-6000), mildly treated with sodium periodate, was covalently coupled to amino-modified latex beads (0.72 p m diameter). The particle agglutination assay was validated by comparing results with the adhesion (percentage binding of adherent cells) of coagulase-negative staphylococcal strains to heparinised microtitration plates. Of 38 different coagulasenegative staphylococcal strains tested, 30 showed agglutination reactivity with heparincoated latex beads. Strains of different coagulase-negative staphylococcal species agglutinated heparin-coated latex beads to various extents (e.g., cells of Staphylococcus haemolyticus strains reacted more strongly than cells of S. epidermidis strains). The agglutination reaction was significantly inhibited by fucoidan, suramin, &carrageenan and other sulphated compounds, but not by non-sulphated carbohydrate polymers such as hyaluronic acid. Agglutination of staphylococcal cells with heparin-coated latex beads was completely blocked by a cell-surface extract. These results suggest that structures responsible for heparin binding are exposed on the cell surface.