Helicobacter pylori adhesin binding group antigen

Abstract

A novel blood group antigen binding (BAB) adhesin protein was isolated and purified, whereby said protein or fractions thereof bind specifically to Helicobacter pylori fucosylated blood group antibodies. Said adhesin has a molecular weight of about 73.5 kDa and the N-terminal sequence for the adhesin and the corresponding DNA show homologies between different strains of H. pylori. Said adhesin and/or DNA is useful for diagnose and therapy and/or profylax directed against H. pylori induced infections, e.g. gastritis and acid peptic disease.

Description

FIELD OF THE INVENTION

The present invention relates to materials and methods for prevention, treatment and diagnosing of infections caused by Helicobacter pylori. More specifically the present invention relates to polypeptides and antibodies useful in vaccines for the treatment and prevention of pathologic infections caused by Helicobacter pylori strains. The present invention specifically relates to a bacterial blood group antigen binding adhesin (BAB-adhesin). The present invention further relates to polynucleotides useful for the recombinant production of said polypeptides and for use in immunisation therapies. In addition, it relates to polypeptides, antibodies, and polynucleotides used for the detection of said bacteria.

The present invention further relates to new immunoglobulins, which exhibit specific activity to a blood group binding adhesin, expressed by Helicobacter pylori, methods for the production of said immunoglobulins, their isolation and use. The present invention further relates to the treatment and prevention of H. pylori induced infections in the gastrointestinal tract.

BACKGROUND OF THE INVENTION

Helicobacter pylori is a causative agent for acid peptic disease and the presence of this organism is highly correlated to the development of gastric adenocarcinoma. Bacterial adherence to the human gastric epithelial lining was recently shown to be mediated by fucosylated blood group antigens.

Recent research has focused on the role of Helicobacter pylori in the development of ulcers in the gastric mucosa. Recent findings show a strong connection between H. pylori and chronic, active gastritis and gastric ulcers. Furthermore, there appears to be a strong correlation between ventricular cancer and gastric ulcers. Traditional treatment of gastric ulcers has involved gastric resection, the administration of bismuth compositions, the administration of H 2 -blockers and the administration of pH-buffering agents, to mention a few examples.

More recently, various forms of treatment have been supplemented with the administration of antibiotics. One problem with presently known treatments is the risk for treatment failure. Furthermore, not only do microbes develop antibiotic resistance, the antibiotics administered often upset the natural balance of benign microbes, colonising the intestinal tract. This leads to diarrhoea and other signs of intestinal discomfort, in addition to destabilising the benign flora in the intestines. Other treatments, e.g. H 2 -blockers often require life-long medication to prevent the recurrence of disease.

The foregoing, together with the fact that H. pylori is very widely spread among humans—according to a conservative estimate approximately 60% of all adult humans in the industrialised countries are infected—makes the diagnosing, prevention and treatment of H. pylori infections an urgent task.

Further, the fact that developing countries frequently lack the resources for conventional treatment of gastric ulcers further underlines the importance of finding new ways of treatment and prevention of H. pylori induced infections. It is obvious, for many reasons, that disease prevention with vaccines is a preferable mode. A vaccine would provide an easily administered and economical prophylactic regimen against H. pylori infections. An effective vaccine against H. pylori is nevertheless presently lacking.

STATE OF THE ART

H. pylori colonises the human gastric mucosa, in an equilibrium between adherence to the epithelial surface mucous cells and the mucous layer lining the gastric epithelium. Once infected, bacteria seems to colonise for a lifetime. Attachment to the epithelial lining protects the bacteria from the anti-microbial effects of the acidic gastric juice of the stomach lumen, as well as from physical forces such as peristalsis. For survival in this hostile ecological niche, H. pylori has developed a battery of virulence factors; such as production of the enzyme urease that buffers the micro-environment around the bacteria and the polar flagellae to ensure high motility, a prerequisite in an ecological niche where the turnover of the mucous layer is in the range of hours. A subset of H. pylori strains produces the vacuolating cytotoxin, VacA, and the cytotoxin associated antigen CagA.

Attachment is essential for colonisation of the epithelial lining and bacteria express surface associated adhesion molecules that recognise specific carbohydrate or protein receptors on the cell surfaces or mucous lining. The specificity in this interaction in combination with the genetically regulated receptor distribution results in a restricted range of cell lineages and tissues available for colonisation. Several putative receptor structures have been described for H. pylori, such as the hemagglutinin-sialic acid, sulphated glycoconjugates and sulphatides. Recently, the fucosylated blood group antigens H-1 and Lewis b were described (Borén et al., Science, 262, 18921993), mediating specific adherence of H. pylori to human and rhesus monkey gastric surface mucous cells in situ. The H-1 and Lewis b antigens are part of the blood group antigens that define blood group O in the ABO system.

Surface-exposed proteins are often constituents of the outer membrane. The outer membrane has a structural role and acts as a selective barrier, determining what enters the cell and what molecules are secreted. One class of outer membrane proteins are called porins, and create hydrophilic pores through the outer membrane where specific metabolites, such as sugar molecules, can cross. Recently the finding of a number of outer membrane proteins in H. pylori, was reported, which proteins were suggested to constitute a family of porin proteins.

The BAB adhesin has previously been identified and shown to be localised on the bacterial surface of H. pylori (SE 9602287-6). The blood group binding activity was shown to be pH dependent and the present inventors present evidence that the binding affinity to the Lewis b receptor reveals a high equilibrium constant. For the purification of the BAB adhesin, a crosslinker-labelled receptor conjugate was used in order to mediate specific transfer of biotin to the adhesins on the bacterial surface. Thereafter the biotin-labelled adhesin could be extracted by streptavidin coated magnetic beads. Determination of the amino terminal amino acid sequence of the purified BAB adhesin exhibit homologies to outer membrane proteins of H. pylori porins.

Intensive research has been directed to the immunological treatment and prevention of H. pylori induced infections. EP 0 484 148 (Ando & Nakamura) describes a method for treating and/or preventing upper gastrointestinal disease in mammals, said method comprising orally administering to a patient in need thereof an effective amount of a pharmaceutical composition comprising anti- Helicobacter pylori polyclonal immunoglobulins and a pharmaceutically acceptable carrier. Said description further dwells on the combination of said treatment in combination with the administration of antibiotics. As the method of producing said polyclonal antibodies, EP 0 484 148 describes the isolation and purification of anti- H. pylori immunoglobulins from the sera and milk of mammals. H. pylori itself was not found in the stomachs of cows, goats, sheep, swine or horses, according to EP 0 484 148, but it was assumed that these animal species have colonizing microorganisms with antigenic determinants similar to those of H. pylori because they have immunoglobulins which cross-react to strains of H. pylori found in humans. Preferably, according to EP 0 484 148, large mammals, e.g. pregnant cows, are immunized with whole cells of H. pylori and the immunoglobulins subsequently extracted from the milk or colostrum. In the immunization experiments, NCTC Strain 11362 and clinical isolate H. pylori No. 153 were used to trigger the production of immunoglobulins. On the other hand, NCTC Strain 11637 was used for analysing purposes. Immunization is claimed to yield an anti- H. pylori titer in the milk of such magnitude, that daily doses of 0.01-0.1 g/day immunoglobulin composition, are sufficient for successful therapy. The claimed interval of 0.01-0.1 g/day is however not supported by the experiments presented by Ando & Nakamura and so low doses have hitherto not proven efficient in clinical tests. The doses actually used in example 5 and 7 are in the order of magnitude of 1 g/day, i.e. 10-fold the upper limit of the given interval. Furthermore, it is very unlikely, that unspecific immunoglobulin mixtures as those manufactured by Ando & Nakamura, would be effective in claimed doses as similar doses are ineffective against other gastrointestinal pathogens. The simultaneous administration of antibiotics, extensively discussed in the description, underlines the insufficiency of the disclosed immunoglobulins.

EP 0 469 359 (Cordle & Schaller) likewise describes the immunization of mammals, preferably pregnant cows, with formalin killed H. pylori bacteria (ATCC Strain 26695). Anti- H. pylori polyclonal antibodies were isolated and purified from the milk and finally fed to piglets, in amounts of about 0.5 g immunoglobulins, three times daily. The results were assessed by determination of the number of biopsy specimens, which were positive for Gram-negative bacteria after the trial. Gram-negative bacteria was found in 78% of the piglets fed a non-immune nutrient but only (Sic!) in 35% of the piglets fed a nutrient containing so called specific anti- H. pylori antibodies.

Anti- H. pylori polyclonal antibodies, effective to cause aggregation of H. pylori, have thus been administered orally as a regimen in the treatment and prevention of H. pylori induced infections in the gastrointestinal tract. Nevertheless, as also noted in EP 0 484 148 A1, it is still not clear, how many antigenic determinants are present on the surface of H. pylori. The occurrence of a wide variety of H. pylori strains, makes questionable the practical efficiency of any polyclonal immunological therapy according to the state of the art. Immunization using whole bacteria will always trigger a highly polyclonal immunresponse with a low level of antibodies against a given antigenic determinant. This is underlined e.g. by the results presented by Cordle & Schaller, where, although the number of Helicobacter positive biopsies were reduced, complete cure was not obtained through the treatment according to their invention.

It is notable, that the dose of immunoglobulin needed for oral prophylaxis or therapy has not yet been clearly defined. In a normal human adult, approximately 5 g IgA is produced and secreted at mucosal surfaces each day. Obviously, doses of this magnitude are economically and practically unfeasible for large-scale therapy or prophylaxis. In studies on the effect of oral immunoglobulin on rotavirus infection, daily doses in the interval of 600 to 9000 mg have been tried in clinical tests. Successful intervention has also been reported when treating H. pylori and cryptosporidial infections with daily administrations of 3 to 15 g immunoglobulin from immunized cows (Hammarström et al., Immunol Rev, 139 (1994) 43-70). Generally speaking, all studies hitherto point to the necessity of using high doses of immunoglobulins when trying to combat an ongoing infection. The need for more specific immunoglobuline preparations, allowing the use of smaller doses, is thus an urgent one.

To maximize the potency of an immunological regimen for the treatment and prevention of H. pylori, it is of great importance to find a specific conserved antigenic determinant, which plays a central role for the pathogenicity of H. pylori. Using such an antigenic determinant would make it possible to produce highly specific and therapeutically efficient novel polyclonal and/or monoclonal immunoglobulin preparations.

SUMMARY OF THE INVENTION

The above problem of providing specific, cost-efficient and therapeutically superior immunoglobulin preparations for the treatment and prevention of H. pylori has now been solved through the composition and methods according to the attached patent claims. The present inventors have now surprisingly shown, that highly specific and therapeutically efficient polyclonal and/or monoclonal immunoglobulin preparations can be provided through the immunization of an animal with an adhesin protein, specific for H. pylori. The invention will now be described in closer detail with reference to the attached, non-limiting figures and examples.

One objective of the present invention was to further purify and characterize the H. pylori blood group antigen binding (BAB) adhesin to make possible the development of methods and materials for specific and selective diagnosing and treatment of H. pylori induced infections and related diseases and the development of said methods and materials. A further and equally important objective was to determine the DNA sequences of the genes involved in the expression of this protein. These objectives were fulfilled through the protein, the DNA and the methods and materials specified herein. The DNA sequences are attached as SEQ ID NOS: 1 and 2, disclosing the babA and babB sequences, respectively. The full protein sequences are disclosed in SEQ ID NOS: 3 and 4.

DESCRIPTION OF THE FIGURES

FIG. 1A ) illustrates the bacterial binding to soluble blood group antigens. H. pylori strains were incubated with 125 I-labeled blood group antigen glycoconjugates and bound 125 I-activity was measured (Note the absence of blood group antigen binding shown for strains MO19 and 26695),

FIG. 1B ) illustrates an receptor displacement assay. Strain CCUG 17875 was first incubated with 10 ng 125 I-labeled Le b antigen glycoconjugate and the complex was then challenged (1 h) with an excess of unlabeled Le b or Le a glycoconjugate, before the 125 I-activity in the bacterial pellet was measured. Concentrations of the unlabeled glycoconjugate ranged from 50 ng to 8 μg and

FIG. 1C ) shows the results of a Scatchard analysis of the H. pylori -Le b antigen interaction. Bacterial binding to the Le b glycoconjugate was titrated to an affinity constant (Ka) value of 8×10 −10 M −1 (13).

FIGS. 2 A-D: Upper panel: Prevalence of the BabA adhesin in the bacterial population. Cells of strain CCUG 17875 were incubated with biotinylated Le b FIG. 2 (A) or Le b FIG. 2 (B) glycoconjugate. Bound biotinylated Lewis-conjugate was detected with FITC-labeled streptavidin (green fluorescence) and bacteria were counterstained with propidium iodine (red fluorescence). Lower panel: Localisation of the BabA adhesin. For electron microscopy (15) cells of strain CCUG 17875 were incubated with biotinylated Le b FIG. 2 (C) or Le a FIG. 2 (D).

FIGS. 3A-C shows the characterization of the molecular weight of the BabA adhesin by the use of receptor overlay analysis FIGS. 3 (A, B) and receptor activity directed affinity tagging of BabA FIG. 3 (C).

FIGS. 4A-C shows receptor activity directed affinity tagging and protein purification of the BabA adhesin.

FIG. 5 shows the translated amino acid sequences for the babA and babB genes, corresponding to the N-terminal domain of the BabA (SEQ ID NOS: 7 and 8) adhesin.

FIG. 6 shows the procentual inhibition of H. pylori binding to 125 I-labeled Lewis b antigen for different preparations as a function of the antibody titre.

FIG. 7 shows a Western blot detection of the BabA adhesin by the different antibody preparations.

FIG. 8 shows four Western blot analyses of H. pylori proteins by the different antibody preparations.

DESCRIPTION OF THE INVENTION

The blood group antigen binding adhesin, BabA, has now been biochemically characterized and purified by a novel technique, receptor Activity Directed Affinity Tagging (Retagging). Two genes, babA and babB were found to code for two different but very similar proteins. The present invention thus comprises a novel blood group antigen binding adhesin. The DNA sequences are disclosed in SEQ ID NO: 1 (babA) and SEQ ID NO: 2 (babB) The protein sequences are disclosed in SEQ ID NOS: 3 and 4. The invention also includes any pharmaceutical composition comprising said adhesin protein and/or fractions thereof. Examples of such pharmaceutical compositions are for example medicaments for the prevention or treatment of Helicobacter pylori induced gastritis, gastric and duodenal ulcers and gastric adenocarcinoma. Optionally said pharmaceutical composition additionally encompasses pharmaceutically acceptable excipients.

Further, the present invention comprises the BAB-adhesin gene or genes for expression of an adhesin protein according to the invention. Said invention also comprises a novel method for the isolation and purification of said adhesin. The disclosed genes are contemplated to function as a cassette system, the organism alternating between these to avoid immunity in the host. It is very likely, that homologies of the disclosed sequences exist and additionally supplement said cassette function in other strains of H. pylori. Also, genes corresponding to a homology of the first 40 amino acids or genes, corresponding to a homology of the last, about 300 amino acids, can function to this effect. It is further highly likely, that Helicobacter pylori is able to switch between several genes, similar to the disclosed genes, in a so-called cassette system.

The invention additionally comprises monospecific antisera produced using the novel adhesin protein and/or fractions thereof. Said monospecific antisera is preferably produced according to any suitable, conventional method for producing monospecific antisera in vitro or in vivo, e.g. by inoculating a suitable animal. Such methods are familiar to a person skilled in the art. Antibodies raised in a suitable animal or in the patient to be treated, can subsequently be administered locally, e.g. orally to the patient.

The invention further comprises the use of said monospecific antisera for the manufacturing of a test kit for quantitative or qualitative determinations of adhesin protein or fractions thereof in cells, tissues or body fluids.

The invention further comprises the use of said adhesin protein or corresponding DNA for use in therapy or immunisation and/or in the manufacture of compositions for said uses. The invention specifically encompasses the use of said DNA for immunisation therapy and for the manufacture for compositions for such therapy. Preferably, in an immunisation therapy where said composition is administered orally to a patient, the adhesin protein, fractions thereof or said DNA is administered in combination with a pharmaceutically suitable immunostimulating agent. Examples of such agents include, but are not limited to the following: cholera toxin and/or derivatives thereof, heat labile toxins, such as E. coli toxin and similar agents. The composition according to the present invention can further include conventional and pharmaceutically acceptable adjuvants, familiar to a person skilled in the art of immunisation therapy. Preferably, in an immunisation therapy using the inventive DNA or fractions thereof, said DNA is preferably administered intramuscularly, whereby said DNA is incorporated in suitable plasmide carriers. An additional gene or genes encoding a suitable immunostimulating agent can preferably be incorporated in the same plasmide.

Said immunisation therapies are not restricted to the above-described routes of administration, but can naturally be adapted to any one of the following routes of administration: oral, nasal, subcutaneous and intramuscular. Especially the oral and nasal methods of administration are promising, in particular for large-scale immunisations.

The present inventors have surprisingly shown, that highly specific and therapeutically efficient polyclonal and/or monoclonal immunoglobulin preparations can be provided through the immunisation of an animal with an adhesin protein or fractions thereof, specific for H. pylori. When considering immunisation against H. pylori, it is worth noting that the infection is known to be lifelong despite a vigorous immune response in the gastric mucosa. An increased local production of IgA in the mucosa is not necessarily enough and the administration of monospecific antibodies directed against a central virulens factor, such as the adhesin according to the present invention, may constitute a more effective approach.

The term “immunisation” refers here to a method for inducing a continuous high level of antibody and/or cellular immunresponse. The term “animal” here preferentially denotes any member of the subphylum Vertebrata, a division that includes all animals, including mammals, which are characterized by a segmented bony or cartilaginous spinal column. All vertebrates have a functional immune system and respond to antigens by producing antibodies. The term “protein” is used here to denote a naturally occurring polypeptide and the term “polypeptide” is used here in its widest meaning, i.e. any amino acid polymer (dipeptide or longer) linked through peptide bonds. Accordingly the term “polypeptide” comprises proteins, oligopeptides, protein fragments, analogues, muteins, fusion proteins and the like. The term “antibody” as used in this context includes an antibody belonging to any of the immunological classes, such as immunoglobulins A, D, E, G or M. Of particular interest are nevertheless immunoglobulin A (IgA) since this is the principle immunoglobulin produced by the secretory system of warm-blooded animals. However, in cow colostrum, the main antibody class is IgG 1.

Borén et al. have recently isolated and characterized a Lewis b binding protein with a molecular weight of about 73500 Da (See the priority applications SE 9602287-6 and SE 9701014-4, which are referred to in their entirety). This adhesin protein is thought to be a conserved structure and specific for pathogenic strains of H. pylori. Said protein is specific for at least one of the H. pylori strains included in the following group: CCUG 17875, NCTC 11637, A5, P466, G109, G56, Ba 185, Ba 99, 931 and 932.

This adhesin protein or immunologically effective fractions thereof are characterized in that the following amino acid sequence (SEQ ID NO:5) included:

EDDGFYTSVGYQIGEAAQMV

or homologues thereof.

The following DNA sequence (SEQ ID NO:6) or homologues thereof is included in DNA for expression of said adhesin protein or fractions thereof:

5′-GAAGACGACGGCTTTTACACAAGCGTAGGCTATCAAATCGGT GAAGCCGCTCAAATGGTA-3′

According to one embodiment of the invention, a pregnant mammal, preferably a cow or another suitable domestic animal, is immunised with said Lewis b binding adhesin protein or fractions thereof. The adhesin protein or fractions thereof is/are preferably injected intramuscularly or subcutaneously in the chosen animal, optionally together with suitable adjuvants. Examples of such adjuvants include, but are not limited to immunostimulating agents such as cholera toxin and/or derivatives thereof, heat labile toxins, such as E. coli toxin and similar, conventional agents, such as classical adjuvants including mineral and vegetable oils. Subsequent to the regimen of immunization, comprising a necessary amount of doses, including so called booster-doses, over a time span allowing for optimal immunoglobulin expression, milk or sera is collected from said animal. Preferably the cow colostrum, which is specially high in immunoglobulins, is collected. The specific immunoglobulin fraction according to the present invention is then separated and purified in a conventional manner, e.g. including separation of fats, protein precipitation and concentration by ultrafiltration.

According to another embodiment of the invention, a bird, preferably a chicken or another suitable domestic bird, is immunized with said Lewis b binding adhesin protein or fractions thereof. The adhesin protein or fractions thereof is preferably injected intramuscularly or subcutaneously in the chosen bird, optionally together with suitable adjuvants. Examples of such adjuvants include, but are not limited to immunostimulating agents such as cholera toxin and/or derivatives thereof, heat labile toxins, such as E. coli toxin and similar, conventional agents, such as classical adjuvants including mineral and vegetable oils. Subsequent to the regimen of inmmunization, comprising a necessary amount of doses, including so called booster-doses, over a time span allowing for optimal immunoglobulin expression, sera or eggs is/are collected from said animal. Preferably the egg yolk, which is specially high in immunoglobulins, is collected. The specific immunoglobulin fraction according to the present invention is then separated and purified in a conventional manner, e.g. including protein precipitation and ultrafiltration. Alternatively, the egg yolk being of high nutritional value in addition to containing a high titer of specific antibodies according to the present invention, can be administered as such.

According to a preferred embodiment of the present invention, monoclonal immunoglobulin is produced by establishing transgenic animals. Said transgenic animals can be chosen from the following group of species: mammals, e.g. cow, goat and rabbit, and birds: e.g. chicken, duck, turkey. The mammal most preferably used is cow and the most preferable bird is chicken. Further developments of transgenic animals such as mice and rats could also offer new possibilities. The choice of animal is naturally governed by availability and local adaptation.

According to one embodiment, a stock of transgenic animals according to the present invention, adapted to the local conditions, are kept locally, e.g. in villages in developing countries to function as local units for the production of immunoglobulins for oral administration. For example transgenic cows, goats or chicken are suitable for this purpose and preferably chicken are used. Consumption of the milk or preferably the eggs, produced by the transgenic animals, can help to eradicate presently very difficult infectious diseases, e.g. diseases caused by H. pylori.

According to yet another embodiment of the present invention, monoclonal antibodies can be produced using the hybridoma method. The hybridoma method is well known to a skilled worker in the field of biochemistry and it is described e.g. in Galfre, G. And Milstein, C., Preparation of monoclonal antibodies: strategies and procedures (Methods in Enzymology, 73:3-46, 1981). A suitable host animal is immunized with the Lewis b binding adhesin protein or fractions thereof. When the immunization is accomplished, the animal is sacrificed, spleen cells collected and fused with cells from a neoplastic cell line, preferably myeloma cells. By choosing the growth conditions, the successfully fused hybridoma cells can be selected. The monoclonal antibodies produced by the hybridoma cell line can then be administered orally in a regimen for treatment and/or prevention of H. pylori infections.

Preferably the polyclonal and/or monoclonal antibodies are purified prior to administration and, more preferably, admixed with pharmaceutically suitable carriers and/or adjuvants. Examples of suitable carriers are saline, pharmaceutically acceptable fats, oils, carbohydrates and proteins. The carrier or carriers is/are preferably chosen so that the solubility and absorption of the immunoglobulin in the mucus layer lining the stomach is enhanced. Using suitable adjuvants the stability, therapeutic efficiency and nutritional value of the composition can be improved. To improve stability under storage, the immnunoglobulin composition can be lyophilized. Regardless of the exact preparation and formulation, it is of central importance to avoid denaturating the immunoglobulins.

The higher specificity, exhibited by the immunoglobulin preparation of polyclonal and/or monoclonal antibodies according to the invention, makes it possible use substantially lower doses compared to those presently used, thus lowering the cost and improving the availability of the treatment. The use of specific, monoclonal antibodies can make it possible to further lower the doses. The doses are in all cases a function of the antibody titer of the preparation. A high titer naturally allows the use of lower doses.

According to one embodiment of the invention, an immunoglobulin preparation is manufactured as follows: an animal is immunized with a Lewis b binding adhesin protein or fractions thereof, expressed by Helicobacter pylori, the immunoglobulin fraction is isolated from a excretion of said animal and subsequently purified. The purified immunoglobulin composition is admixed with suitable carriers and adjuvants to form a immunoglobulin preparation for the prevention or treatment of H. pylori infections. In cases where the antibody titer is sufficiently high and the other constituents of the immunoglobulin composition isolated from the animal are harmless, for example in the case of colostrum from immunized cows or egg yolk from immunized chicken, there is always the option of administering the colostrum or egg yolk to the patient without any further treatment of the colostrum or egg yolk.

The immunoglobulin composition according to the invention is preferably administered orally to the patient, in the smallest therapeutically or prophylactically effective dose. Presently conceived are doses in the interval of 0.1 to 1000 mg/day, preferably in the interval of 0.1 to 100 mg/day. The chosen doses naturally depend on the antibody titer of the preparation in question. The exact doses and the regimen of administration can be chosen by the physician responsible for the patient, infected by Helicobacter pylori. Routine experimentation and later, with increasing experience of this method, empirical information will suffice to establish the required amount. Multiple dosages may be used, as needed, to provide the desired level of therapeutic or profylactic effect. The immunoglobulin preparations according to the present invention can also, being free from adverse side effects and imposing practically no danger of overdosing, be taken prophylactically or therapeutically by a person without medical supervision.

A therapeutical effect can be attained, except with the specific antibody according to the present invention, also with at least two Fab-fragments of said antibody. Said embodiment is also encompassed by the scope of the present invention.

According to yet another embodiment, avirulent microorganisms, preferably bacteria, are used as expression systems for the specific antibody according to the present invention. An “avirulent microorganism” in this context is a microorganism which has the ability to colonize and replicate in an infected individual, but which does not cause disease symptoms associated with virulent strains of the same species of microorganism. The definition inherent in the GRAS (Generally Regarded As Safe) concept can be applied here. A GRAS-organism is suitable for use according to the present invention, provided that the organism externalises the antibody or can be modified to this effect. The term “microorganism” as used herein includes bacteria, protozoa and unicellular fungi. Preferably, bacteria are used as expression systems, e.g. bacteria of the genus Lactobacillus, Streptococcus or Enterobacteriae. The above mentioned expression system can be utilised in vitro for the production of the specific antibody according to the present invention or, according to a further embodiment of the invention, the micro-organism constituting the expression system can be administered directly to the patient. The micro-organisms can be harvested and administered as such, but they are preferably mixed with a suitable carrier, mixed in a suitable foodstuff, lyophilised, encapsulated or treated in any other conventional way, used for the delivery of viable micro-organisms to the gastrointestinal tract.

According to yet another embodiment, avirulent microorganisms, preferably bacteria, are used as expression systems for the specific adhesin protein according to the present invention. An “avirulent microorganism” in this context is a microorganism which has the ability to colonize and replicate in an infected individual, but which does not cause disease symptoms associated with virulent strains of the same species of microorganism. The definition inherent in the GRAS (Generally Regarded As Safe) concept can be applied here. A GRAS-organism is suitable for use according to the present invention, provided that the organism externalises the adhesin protein or can be modified to this effect. The term “microorganism” as used herein includes bacteria, protozoa and unicellular fungi. Preferably, bacteria are used as expression systems, e.g. bacteria of the genus Lactobacillus, Streptococcus or Enterobacteriae. The above mentioned expression system can be utilised in vitro for the production of the specific adhesin according to the present invention or, according to a further embodiment of the invention, the micro-organism constituting the expression system can be administered directly to the patient. The micro-organisms can be harvested and administered as such, but they are preferably mixed with a suitable carrier, mixed in a suitable foodstuff, lyophilised, encapsulated or treated in any other conventional way, used for the delivery of viable micro-organisms to the gastrointestinal tract.

The exact doses and the regimen of administration of said micro-organisms can be chosen by the physician responsible for the patient, infected by Helicobacter pylori. Routine experimentation and later, with increasing experience of this method, empirical information will suffice to establish the required amount. Multiple dosages may be used, as needed, to provide the desired level of therapeutic or prophylactic effect. The avirluent micro-organism expressing the antibody or adhesin protein according to the present invention can also, being free from adverse side effects and imposing practically no danger of overdosing, be taken prophylactically or therapeutically by a person without medical supervision. A preferred carrier in this specific application is a foodstuff, e.g. a fermented product such as fermented cereal or dairy product.

The creation of previously mentioned expression systems and still earlier mentioned methods of creating hybridomas and transgenic animals can include steps involving recombinant DNA techniques. Recombinant DNA techniques are now sufficiently well known and widespread so as to be considered routine. In very general and broad terms, recombinant DNA techniques consist of transferring part of the genetic material of one organism into a second organism, so that the transferred genetic material becomes a permanent part of the genetic material of the organism to which it is transferred. Methods for achieving this are well known and the mere choice of specific methods for achieving the objectives, set out in the present description and claims, fall under the scope of the invention.

It is possible, that H. pylori alone or together with related slow-acting bacteria are involved in the genesis and aggravation of other chronic inflammatory diseases in the gastrointestinal tract. It is obvious for a skilled practitioner how to modify the present invention, within the scope of the claims, to gain utility in the treatment and/or prevention of such diseases. Examples of such diseases are ulcerative colitis, Crohn's disease, sarcoidosis, Wegener's granulomatosis and other vasculithic disorders, as well as various neoplasms, including carcinomas of the colon, pancreas and prostate.

EXAMPLES

H. pylori strain CCUG 17875 was obtained from CCUG, Göteborg, Sweden. Strain A5, a gastric ulcer isolate, from Astra Arcus, Södertälje, Sweden. Strains P466 and MO19 were described previously (Borén et. al, Science, 262, 1892(1993)). Strain 26695 came from Dr. K. A. Eaton, The Ohio State University and its genome was recently sequenced by TIGR, Rockville, Md., USA. The panel of 45 H. pylori clinical isolates came from the University Hospital in Uppsala, Sweden. Bacteria were grown at 37° C. in 10% CO2 and 5% O2 for 48 h.

All blood group antigen glycoconjugates used, i.e. semi-synthetic glycoproteins constructed by the conjugation of purified fucosylated oligosacharides to serum albumin were from IsoSep AB, Tullinge, Sweden. The RIA was performed according to Falk et al. (Meth. Enzymol., 236, 353, 1994) with some modifications; the H-1, Le b , Le a , H-2, Le x and Le y glycoconjugates were 125I-labeled by the Chloramine T method. 1 ml of bacteria (A600=OD 0.10) was incubated with 300 ng of 125I -labelled conjugate (i.e. an excess of receptors) for 30 min. in phosphate buffered saline (PBS), 0.5% albumin, 0.05% Tween-20 (BB-buffer). After centrifugation, 125I-activity in the bacterial pellet was measured by gamma scintillation counting.

In this study the present inventors' first biochemically characterized and identified the H. pylori blood group antigen binding adhesin, BabA. H. pylori strains were analysed for binding to soluble 125 I-labeled fucosylated blood group antigens (FIG. 1 A). Binding of these strains to the soluble blood group antigens correlate with adherence in situ. The prevalence of blood group antigen binding (BAB)-activity was assessed among 45 clinical H. pylori isolates and the majority of the isolates, 71%, express Le b antigen binding properties (data not shown). In contrast, none of the reference strains (FIG. 1 A), or strains from the panel of 45 clinical isolates, bind to the Le a , H-2, Le x , or Le y antigens. These results support our previous findings of high receptor specificity for the Le b and H-1 blood group antigens and demonstrate the high prevalence of BAB activity among clinical isolates.

Based on the presence or absence of virulence factors such as the Cytotoxin associated gene A (CagA) and the Vacuolating cytotoxin A (VacA), H. pylori strains are classified as type I or type II strains. H. pylori isolates from patients with duodenal ulcers most often express the VacA and the CagA-proteins, i.e. type-I strains. By definition, type II strains express neither markers. Twenty-one clinical isolates previously defined for expression of CagA and VacA were analysed for Le b antigen binding properties. Expression of CagA was found to correlate with bacterial binding to the Le b antigen (Table 1). The cagA gene belongs to a 40 kb pathogenicity island that encodes components of secretion and transport systems. These findings could indicate functional crosstalk between the cag pathogenicity island and the BabA adhesin gene, for the correct presentation of the BabA adhesin protein in the bacterial outer membrane.

To further characterize BabA, the present inventors determined the affinity constant (K a ) between BabA and the Le b antigen. Since K a -values are based on equilibrium conditions (13), the present inventors first analysed the interaction by performing receptor displacement analysis. H. pylori CCUG 17875 (positive for Le b binding, FIG. 1A ) was first incubated with 125 I-labeled Le b glycoconjugate. Then unlabeled Le b glycoconjugate was added in a dilution series. The unlabeled Le b conjugate displaced the bound 125 I-labeled Le b glycoconjugate efficiently (FIG. 1 B). The results demonstrate that the receptor-adhesin complex formed is in a true state of equilibrium. An equivalent excess of Le a glycoconjugate did not dissociate the Le b -BabA complex, verifying the high receptor specificity (FIG. 1 B). The K a -value for the Le b -BabA complex of strain CCUG 17875 was titrated with Le b glycoconjugate in concentrations from 10 ng to 260 ng/ml and determined to be of an high affinity close to 1×10 10 M −1 (FIG. IC). The number of Le b glycoconjugate molecules bound to BabA on the bacterial cell surface was calculated to be around 500 per cell. This number is similar to the number or fimbriae organelles on the surface of E. coli (14). However, for the BabA adhesin, the calculations are based on the assumption that the majority of bacterial cells in the experiment exhibit an equal number of adhesin molecules with Le b antigen binding properties.

TABLE 1
BAB activity among H. pylori Type I and Type II strains
	Type	Strain	BAB activity
	Type I	CCUG 17874	−
	CagA + , VacA +	G39	−
		G11	−
		G20	−
		G27	+
		G56	+
		G106	−
		G109	+
		932	+
		Ba185	+
		87A300	+
	Type Ia	931	+
	CagA + , VacA −	Ba99	+
		Ba179	+
		Ba194	+
	Type Ib	G12	−
	CagA − , VacA +
	Type Id	G104	−
	ΔcagA, VacA +	Tx30	−
	Type II	G21	−
	CagA − , VacA −	G50	−
		G198	−

To determine the prevalence of BabA in the bacterial population, strain CCUG 17875 was incubated with Le b or Le a antigens, and bacterial binding activity was visualised by confocal fluorescence microscopy ( FIG. 2 , upper panel). The analyses demonstrate the high prevalence of BabA binding activity in the bacterial population to the Le b antigen ( FIG. 2A , green staining) and the complete lack of binding to the Le a antigen ( FIG. 2B , red counter staining).

Next, the localisation and density of BabA on the bacterial cell surfaces was investigated by immunogold electron microscopy. The Le b antigen binding activity of the adhesin localised gold particles to the bacterial outer membrane (FIG. 2 C). Individual bacterial cells exhibit an equal number of gold particles (data not shown). When the Le b antigen was substituted with the Le a antigen (lacking receptor activity), no gold particles were detected (FIG. 2 D).

The molecular weight of BabA was characterized by receptor overlay analysis. A protein extract of strain CCUG 17875 was separated on SDS-PAGE and blotted to a membrane. The membrane was incubated with biotinylated Le b glycoconjugate, followed by detection with streptavidin and enhanced chemiluminescence. The BabA adhesin activity corresponds to a single 74 kDa band (FIG. 3 A). The 40 kDa band is presumably endogenous peroxidase activity since it stains independently of the Le b conjugate overlay (lane 3). BabA was very heat stable and could regain some activity after heating to 97° C. ( FIG. 3A , lane 2). The panel of strains exhibited the same molecular weight of BabA (FIG. 3 B).

To purify BabA, a novel technique was developed, Receptor Activity Directed Affinity Tagging (ReTagging). Multi-functional crosslinking agents with radiolabeled donating tags have been previously used for receptor-ligand characterization studies. However, the use of affinity donating tags, such as biotin residues presented on flexible spacer structures, adds a new dimension to the applicability of crosslinker technology. An affinity tag, biotin, is transferred to the adhesin protein by the receptor activity and is used for further identification and for affinity purification of the adhesin part of the interaction, by streptavidin ( FIGS. 4A , B).

A multi-functional crosslinking agent with a biotin donating handle was attached to the Le b glycoconjugate. The receptor activity of the Le b glycoconjugate subsequently directed the targeted biotin tagging of the BabA adhesin protein ( FIGS. 4A , B). After crosslinking, the bacterial protein from strains A5, P466, and CCUG 17875 were separated on SDS-PAGE. Immunodetection with streptavidin demonstrated a biotin tagged protein, with the molecular weight of 74 kDa ( FIG. 3C ) (28), These results support the estimates of the molecular weight from the previous overlay analyses (FIG. 3 B). Strain MO19 devoid of Le b antigen binding properties ( FIG. 3B ) (FIG. 1 A), was negative for binding also in this set of analyses (FIG. 3 C).

The high specificity in the ReTagging technique provided a method for purification of the adhesin protein. Strains CCUG 17875 and A5, that both express the BabA adhesin ( FIG. 1A ) were processed by the ReTagging technique using crosslinker labelled Le b receptor conjugate as the biotin donor. After crosslinking, bacteria were suspended in SDS sample buffer. Streptavidin coated magnetic beads were subsequently added to the solubilised proteins, and biotin tagged BabA was extracted (FIG. 4 C). The N-terminal 20 amino acid sequences of the BabA adhesins from strains CCUG 17875 (Australia) and A5 (Sweden) were found to be identical, indicating a biologically conserved protein (FIG. 5 ). Recently, a series of outer membrane proteins from H. pylori were characterized. These proteins, HopA-E, are homologous in their N-terminal sequences to BabA (17), possible indicating a motif for a common secretion mechanism. The biotin tagged BabA adhesin was purified more than 3000-fold from the cell extract, and the yield was calculated to 20%. However, based on data from the Scatchard plots, the level of available BabA adhesin would be about 5-times higher, i.e. approximately 1 mg adhesin/750 mg bacterial protein, which nevertheless could be the reason for the high signal to noise ratio (FIG. 3 B). The purification of BabA via the ReTagging technique indicates the potential of this technique for the purification of lectins in complex receptor-ligand interactions, such as the selectin family of cell adhesion molecules.

To clone the gene encoding BabA, the N-terminal 20 aa sequence was utilised for the construction of degenerate primers (18). Two sets of clones were identified which both encode two different but very similar proteins. Both genes code for proteins having almost identical N-terminal domains and identical C-terminal domains, complicating the identification of the functional BabA gene. (FIG. 5 ). To identify the corresponding gene, the BabA adhesin was purified in large scale by ReTagging. This provided enough protein for an extended amino terminal sequence. 41 amino acids were identified and these residues unambiguously discriminated between the two genes by the differences in aa-positions 28, 35, 37, 38 and 41 (FIG. 5 ). The gene encoding BabA was named babA and correspond to a basic protein with a pI of 9.4 and a molecular weight of 78 kDa, i.e. of slightly higher molecular weight than that predicted from the SDS PAGE analyses (FIG. 3 ). The other gene, babB, corresponds to a protein of a calculated molecular weight of 75.5 kDa. In contrast to babA, the babB gene contains a predicted translational initiation codon (FIG. 5 ). This could indicate the existence of a third bab gene in the genome or mechanisms for recombination activities. Interestingly, the bab-genes were also detected in strains lacking Lewis b binding properties (data not shown). Gene cassette systems have been shown to promote antigenic variation in Neisseria gonorrhoeae (19). Another possibility would be the presence of similar genes coding for adhesins with differences in receptor specificity/host tissue tropism (20). Gene inactivation experiments targeting the bab-genes could aid in understanding this complex gene organisation.

Immunisation experiments with adhesins from Bordetella pertussis (21) indicate the potential for outer membrane proteins to act as vaccine candidates (discussed in ref 22). In a mouse model for persistent H. pylori infection, oral immunisation with H. pylori antigens proved protective against H. pylori infection (10). However, results from animal models are difficult to evaluate for human specific pathogens, such as H. pylori and Polio virus. For Polio, an animal model has been achieved by expressing the virus receptor in transgenic mice (23). A similar strategy was taken for H. pylori. A transgenic mouse was constructed by the use of an a1,3/4-fucosyltransferase, driving the synthesis of the human specific Le b antigen in the gastrointestinal tract (24). The Lewis b mouse can be useful for the evaluation of the role of the BabA adhesin as a colonisation/virulence factor and in addition for the evaluation of BabA as a vaccine candidate against acid peptic disease and gastric adenocarcinoma.

In the present study the ReTagging technique was used for the purification of the adhesin part of the microbial receptor-ligand interaction. By the use of purified adhesin/lectin-protein, the ReTagging technique could, in addition, be used to further study the receptor part of the interaction. Identification of the biologically active receptor structure, carrying Le b oligosaccharides, would aid in the understanding of the mechanisms supporting the chronic H. pylori infection.

Inhibition of H. pylori binding to 125 I-labeled Lewis b antigen by preparations is presented graphically, as a function of antibody concentration (mg/ml) in FIG. 6 : 1 ml aliquots of H. pylori bacteria (A 600 =OD 0.10) were pre-incubated with dilution series of antibody preparations, in 0.01-10 mg/ml for 2 hours in phosphate buffered saline (PBS), 0.5% albumin, 0.05% Tween-20™. Then 500 ng of 125 I-labeled conjugate (i.e. an excess of receptor structure) was added and incubated for 30 minutes. After centrifugation, 125 I-activity in the bacterial pellet was measured by gamma scintillation counting. The Lewis b blood group antigen glycoconjugates used, i.e. semi-synthetic glycoproteins constructed by the conjugation of purified fucosylated oligosaccharides to serum albumin were from IsoSep AB, Tullinge, Sweden.

Western blot detection of the BabA adhesin by the different antibody preparations is presented in FIG. 7 : Molecular weight rainbow marker (2 μL) from Amersham, Buckinghamshire, England, was dissolved in SDS sample buffer (lane 1). Approx. 100 ng of purified BabA adhesin (approx. 74 kDa with degradation product of approx. 55 kDa) was dissolved in SDS sample buffer (lane 2). SDS solubilized protein extracts of strain CCUG 17875 were prepared by dissolving the bacterial pellet corresponding to 0.15 ml of bacteria (A 600 =OD 0.10) by SDS sample buffer (lane 3). The 3 protein samples were then boiled at 100° C. for 5 minutes. The proteins were separated on SDS-PAGE, and transferred to a PVDF-membrane for Western blot immuno analysis. Five sets of PVDF-membranes were prepared. The PVDF membranes were blocked/incubated overnight with 4% human sera/plasma, in phosphate buffered saline, from a patient with no H. pylori infection, i.e. with no serum antibodies against H. pylori. The membrane was then washed in phosphate buffered saline (PBS), 0.5% albumin, 0.05% Tween-20, followed by the addition of the antibody preparations. The sets of membranes were incubated with the following 5 antibody preparations; 1) pooled human sera from H. pylori infected patients, diluted 1:500. 2) Chicken antibodies (positive) 1 mg/ml diluted 1:100×, 3) Bovine I preparation of antibodies, 1 mg/ml diluted 1:100×. 4) Bovine II preparation of antibodies, 1 mg/ml diluted 1:100×. 5) Bovine III preparation of antibodies, 1 mg/ml diluted 1:100× (indicated in the figure). These antibodies were incubated with the membrane for 2 hours followed by extensive washings in phosphate buffered saline (PBS), 0.05% Tween-20, followed by the addition of secondary anti-human, anti-chicken, and anti-bovine antibodies labeled with HRP-peroxidase (from DAKO, Denmark), all diluted 1:2000×. Membranes were incubated for 1 hour, followed by extensive washings in phosphate buffered saline (PBS), 0.05% Tween-20. The membranes were developed with enhanced chemoluminescens (ECL) from Amersham. The results show, that the antigenic response against the adhesin is strongly enchanced in the bovine preparations. This finding is also supported by the inhibition data in FIG. 6 .

Western blot analyses of H. pylori proteins by the different antibody preparations are shown in FIG. 8. 2 clinical isolates (1-2) from Dr. Lars Engstrand, Department of Clinical Microbiology and Cancerepidemiology, University Hospital, Uppsala, Sweden and strain CCUG 17875 (3), from Culture Collection, University of Göteborg, Department of Clinical Bacteriology, Göteborg, Sweden, and strain 52 (4) from Prof. Torkel Wadström, Dept. Medical Microbiology, Lunds University, were prepared for SDS-PAGE electrophoresis. Bacterial pellets corresponding to 0.15 ml of bacteria (A 600 =OD 0.10) were dissolved in SDS sample buffer and heated to 100° C. for 5 minutes. The proteins were separated on SDS-PAGE, and transferred to PVDF-membranes for Western blot immuno analysis. The western blot analyses were as described above, i.e. the sets of membranes were incubated with the following 4 antibody preparations; 1) pooled human sera from H. pylori infected patients, diluted 1:500. 2) Chicken antibodies (positive) 1 mg/ml diluted 1:100×, 3) Bovine I preparation of antibodies, 1 mg/ml diluted 1:100×. 4) Bovine III preparation of antibodies, 1 mg/ml diluted 1:100× (indicated in the figure). These antibodies were incubated with the membrane for 2 hours followed by extensive washings in phosphate buffered saline (PBS), 0.05% Tween-20, followed by the addition of secondary anti-human, anti-chicken, and anti-bovine antibodies labeled with HRP-peroxidase (from DAKO, Denmark), all diluted 1:2000×. Membranes were incubated for 1 hour, followed by extensive washings in phosphate buffered saline (PBS), 0.05% Tween-20. The membranes were developed with enhanced chemoluminescens (ECL) from Amersham. The results show, that the chicken antibodies and the bovine preparations reacts nearly identically against all four strains, indicating conserved properties in strains of different geographical origin.

Although the invention has been described with regard to its preferred embodiments, which constitute the best mode presently known to the inventors, it should be understood that various changes and modifications as would be obvious to one having the ordinary skill in this art may be made without departing from the scope of the invention which is set forth in the claims appended hereto.

REFERENCES AND NOTES

1. J. R. Warren, Lancet, i, 1273, 1983, B. Marshall, Lancet, i, 1273, 1983.

2. A. Dubois, Emerging Infectious Diseases 1, 79 (1995).

3. M. J. Blaser, Sci. Amer. 2, 92 (1996).

4. 3. M. J. Blaser, Trends Microbiol. 7, 255 (1993), D. E. Kirschner and M. J. Blaser, J. Theor. Biol. 176, 281 (1995).

5. P. Falk, T. Borén, and S. Normark, Meth. Enzymol. 236, 353 (1994).

6. D. G. Evans, D. J. Evans Jr., J. J. Moulds and D. Y. Graham, Infect. Immun. 56, 2896 (1988), S. Hirmo, M. Utt, M. Ringner and T. Wadström, FEMS Immunol. and Med Microbiol. 10, 301 (1995), T. Saitoh, et al FEBS Lett. 282, 385 (1991).

7. T. Borén, P. Falk, K. A. Roth, G. Larson, and S. Normark, Science. 262, 1892 (1993), P. Falk, et al Proc. Natl. Acad. Sci. U.S.A. 90, 2035 (1993).

8. H. pylori strain CCUG 17875 was obtained from CCUG, Göteborg, Sweden. Strain A5, a gastric ulcer isolate, came from Astra Arcus, Södertälje, Sweden. Strains P466 and MO19 were described previously (7). Strain 26695 came from Dr. K A. Eaton, The Ohio State University, and its genome was recently sequenced by The Institute for Genomic Research (TIGR), Rockville, Md. (J.-F. Tomb, et al, abstract 3B: 059, IX International Workshop on Gastroduodenal Pathology and Helicobacter pylori, Copenhagen, Denmark, 1996). The panel of 45 H. pylori clinical isolates came from the University Hospital in Uppsala, Sweden. Bacteria were grown at 37(C. in 10% CO 2 and 5% O 2 for 48 h.

9. All blood group antigen glycoconjugates used, i.e. semi-synthetic glycoproteins constructed by the conjugation of purified fucosylated oligosaccharides to serum albumin (7, 25), were from IsoSep AB, Tullinge, Sweden. The RIA was performed according to ref. 26 with some modifications; The H-1, Le b , Le a , H-2, Le x , and Le y glycoconjugates were 125 I-labeled by the Chloramine T method. 1 ml of bacteria (A 600 =OD 0.10) was incubated with 300 ng of 125 I-labeled conjugate (i.e. an excess of receptors) for 30 min. in phosphate buffered saline (PBS), 0.5% albumin, 0.05% Tween-20 (BB-buffer). After centrifugation, 125 I-activity in the bacterial pellet was measured by gamma scintillation counting.

10. A. Covacci, et al, Proc. Natl. Acad. Sci. U.S.A. 90, 5791 (1993), M. Marchetti, et al, Science 267, 1655 (1995).

11. S. Censini et al, Proc. Natl. Acad. Sci. U.S.A. 93, 14648, (1996).

12. Z. Xiang, et al, Infect. Immun. 63, 94 (1995).

13. A. G. Scatchard, Ann. N. Y. Acad. Sci. 51, 600 (1949).

14. O. Mol, and B. Oudega, FEMS. Microbiol. Reviews, 19, 25 (1996).

15. Confocal microscopy was performed on a Nikon/Multiprobe 2001 instrument (Molecular Dynamics, Sunnyvale, Calif.). Electron microscopy was performed on a JEOL 100 CX instrument.

16. J. Brunner, Trends in Cell Biol. 6, 154 (1996), J. D. Bleil and P. M. Wassarman, Proc. Natl. Acad. Sci. U.S.A. 87, 5563, (1990).

17. M. M. Exner, P. Doig, T. J. Trust, and R. E. W. Hancock, Infect. Immun. 63, 1567 (1995), P. Doig, M. M. Exner, R. E. W. Hancock and T. J. Trust, J. Bacteriol. 177, 5447 (1995).

18. The BabA N-terminal sequence analysis was used to make degenerate oligonucleotides, which were used in PCR to obtain an amplified fragment from the chromosome of the baba gene. A 59 bp fragment was identified and used as probe for the screening of a low-copy plasmide (pACYC184) library of Sau3A partially digested chromosomal DNA from strain CCUG 17875.

19. P. Hagblom, E. Segal, E. Billyard, and M. So, Nature, 315, 156 (1985), R. Haas and T. F. Meyer, Cell, 44, 107 (1986).

20. A.-B. Jonsson, D. Ilver, P. Falk, J. Pepose, and S. Normark, Mol. Microbiol, 13, 403 (1994), N. Strömberg, P. G. Nyholm, I. Pascher, and S. Normark, Proc. Nail. Acad. Sci USA 88, 9340 (1991).

21. A. Kimura, K. T. Mountzouros, D. A. Relman, S. Falkow, J. L. Cowell, Infect. Immun. 58, 7 (1990).

22. T. Borén, and P. Falk, Sci. Amer., Sci. & Med. 4 (1994), L. S. Tompkins and S. Falkow, Science 267, 1621 (1995).

23. R. B. Ren, et al, Cell 63, 353 (1990).

24. P. G. Falk, L. Bry, J. Holgersson, and J. I. Gordon, Proc. Natl. Acad. Sci. U.S.A. 92, 1515 (1995).

25. P. D. Rye, Nature Biotechnology. 2, 155 (1996).

26. P. Falk, T. Borén, D. Haslam, M. G. Caparon, Meth. Cell Biol. 45, 161 (1994)

27. Cell extracts were prepared in SDS sample buffer without mercapto ethanol and heated at 37° C. or 97° C. for 10 min. before separation on SDS-PAGE. Proteins were blotted onto a PVDF membrane. The membrane was incubated with 1 μg/ml biotinylated Le b glycoconjugate or biotinylated albumin (negative control) overnight, labelled as described in ref 7. After washing in PBS/0.05% Tween-20, the biotinylated structures bound by the BabA band were probed by HRP-streptavidin and detected using ECL reagents (Amersham, Buckinghamshire, England).

28. The bacterial suspension was incubated with Le b glycoconjugate, to which the Sulfo-SBED crosslinker (Pierce, Rockville, Ill.) had been conjugated by the N-hydroxysuccinimide ester (NHS), according to the manufacturers specifications. The aryl azide crosslinker group was activated by UV irradiation (360 nm). Bacteria were washed with PBS pH 7.6, 0.05% Tween-20 and protease inhibitors (EDTA and benzamidine) under reducing conditions with 50 mM dithiothreitol (DTT). Bacterial proteins were separated on SDS-PAGE, and the biotin tagged BabA protein was detected by immunodetection (PVDF membrane/HRP-streptavidin and ECL) (FIG. 3 C).

29. Strains CCUG 17875 and A5 were first processed by crosslinking and DTT treatment, as above (28), followed by solubilisation in SDS sample buffer. The biotin tagged BabA protein was then extracted with streptavidin coated magnetic beads (Advanced Magnetics Inc., Cambridge, Mass.). The beads were boiled in SDS sample buffer, and bound proteins were eluted and alkylated. The protein preparation was further fractionated by preparative SDS-PAGE (Prep-Cell 491, BioRad, Hercules, Calif.). Fractions with the biotin tagged protein, i.e. the BabA fractions, were identified by immunodetection using streptavidin/ECL. The pooled BabA preparation was then separated on SDS-PAGE and transferred to PVDF membrane. The BabA band was excised and the BabA protein was N-terminally sequenced using a Procise™ 494 instrument (Applied Biosystems, Foster City, Calif.).

8 1 3340 DNA Helicobacter pylori 1tttcagtcaa gcccaaagct atgcgcaaaa cgcttatgct aaagagaatt tacaagcaca 60gccgtccaag tatcaaaaca gcgtgcctga aatcaatatt gatgaagaag aaatcccctt 120taagggttaa aattaaggag acattatgga aagaaaacgc tattcaaaac gctattgcaa 180atacactgaa gctaaaatca gctttattga ctataaagat ttggacatgc tcaagcacac 240gctatcagag cgctataaaa tcatgccaag gaggttgaca ggcaatagca aaaagtggca 300agagagggtg gaagttagcg atcaaaagag cccgccacat ggctttaatc ccctacattg 360tggataggaa aaaagtcgtg gatagccctt ttaaacagca ctgaattttt gattagggct 420aatagggggc atgcctttta atcttgttta atcttggctc tatttttgtt aaacatcggt 480tataaaagcg ttaaaagcac ttttaaaatc caattaaaag cgttcaaaag taacgcaaaa 540aatcaaaaaa atgacaaaat ttttaagaaa atgacaaaaa aaaaaaaaac gctttatgct 600ataatattcc aaatacattc taatgcaaat gcattctaat gcaaatgtat aatgaatgta 660tgaaatccct aatattcaat ccaatttaat ccaaaaagga gaaaaaacac atcctttcat 720taactttagg ctcgctttta gtttccactt tgagcgctga agacgacggc ttttacacaa 780gcgtaggcta tcaaatcggt gaagccgctc aaatggtaac aaacaccaaa ggcatccaag 840atctttcaga caactatgaa aacttgagca aacttttgac ccgatacagc accctaaaca 900cccttatcaa attgtccgct gatccgagcg cgattaacgc ggcacgtgaa aatctgggcg 960cgagcgcgaa gaacttgatc ggcgatacca aaaattcccc cgcctatcaa gccgtgcttt 1020tggcgatcaa tgcggcggta gggttttgga atgtcttagg ctatgctacg caatgcgggg 1080gtaacgctaa tggtcaagaa agcacctctt caaccaccat cttcaacaac gagccagggt 1140atcgatccac ttccatcact tgcagtttga acaggtataa gcctggatac tacggcccta 1200tgagcattga aaatttcaaa aagcttaacg aagcctatca aatcctccaa acggctttaa 1260ataaaggctt acccgcgctc aaagaaaaca acggaacggt cagtgtaacc tacacctaca 1320catgctcagg ggaagggaat gataactgct cgaaaaaagc cacaggtgta agtgaccaaa 1380atggcggaac caaaactaaa acccaaacca tagacggcaa aaccgtaacc accacgatca 1440gttcaaaagt cgttgatagt caggcaaaag gtaatacaac aagggtgtcc tacaccgaaa 1500tcactaacaa attagacggt gtgcctgata gcgctcaagc gctcttggcg caagcgagca 1560cgctcatcaa caccatcaac acggcatgcc cgtattttag tgtaactaat aaaagtggtg 1620gtccacagat ggaaccgact agagggaagt tgtgcggttt tacagaagaa atcagcgcga 1680tccaaaagat gatcacagac gcgcaagagc tggtcaatca aacgagcgtc attaacgagc 1740atgaacaatc aaccccggta ggcggtaata atggcaagcc tttcaaccct ttcacggacg 1800caagcttcgc tcaaggcatg ctcgctaacg ctagtgcgca agccaaaatg ctcaatctag 1860cccatcaagt ggggcaaacc attaaccctg acaatcttac cgggactttt aaaaattttg 1920ttacaggctt tttagccaca tgcaacaaca aatcaacagc tggcactagt ggcacacaag 1980gttcacctcc tggcacagta accactcaaa ctttcgcttc cggttgcgcc tatgtggagc 2040aaaccataac gaatctaaac aacagcatcg ctcattttgg cactcaagag cagcagatac 2100agcaagctga aaacatcgct gacactctag tgaatttcaa atctagatac agcgaattag 2160ggaatactta taacagcatc actactgcgc tctccaaagt ccctaacgcg caaagcttgc 2220aaaacgtggt gggaaaaaag aataacccct atagcccgca aggcatagaa accaattact 2280acttgaatca aaactcttac aaccaaatcc aaaccatcaa ccaagaatta gggcgtaacc 2340cctttaggaa agtgggcatc gtcagttctc aaaccaacaa tggtgccatg aatgggatcg 2400gtatccaggt gggctacaag caattctttg ggcaaaaaag gaaatggggt gcaagatact 2460acggcttttt tgattacaac catgcgttca ttaaatccag cttcttcaac tcggcttctg 2520acgtgtggac ttatggtttt ggagcggacg ctctttataa cttcatcaac gataaagcca 2580ccaatttctt aggcaaaaac aacaagcttt ctgtggggct ttttggcggg attgcgttag 2640cgggcacttc atggcttaat tctgaatacg tgaatttagc caccatgaat aacgtctata 2700acgctaaaat gaacgtggcg aacttccaat tcttattcaa catgggagtg aggatgaatt 2760tagccagatc caagaaaaaa ggcagcgatc atgcggctca gcatggcatt gagttagggc 2820ttaaaatccc caccattaac acgaactact attcctttat gggggctgaa ctcaaatacc 2880gcaggctcta tagcgtgtat ttgaattatg tgttcgctta ctaaaaacta aaaatccttt 2940gtggaactcc ctttttaagg ggtttctttt aaagccttta tttttttttg gaggggttta 3000atttttttga aacctttgtt tttgaattct ctttttaatg ggtttctttt ttgaactctt 3060tgttttgaac tccttttttt gaactccctt ttttaaaccc tttctttttt aaaattctct 3120tttttggggg gtttgatgaa aaatcctttt ttagcgtttt ggtattggtt agtggaaaac 3180ttgatactaa tttaagcgat agtttttaaa aagtgcttct ttaatatagg gggtttaagt 3240tggtgattaa aaggggggaa tggtttcaaa gcgcttccta tccctttaag aaaataaaat 3300aaaactttaa taaaatgagt tttacaacaa aatgagatcc 3340 2 2781 DNA Helicobacter pylori 2catttgatcg cattggattt caaagaaggg cgttttgtga aaggctttgg tcaagcttat 60gatattttag gcgacaaaat cgcttatgtt gggggtaaag gcaacccaca caatttcgct 120cacaagaaat aaactttctc acccataagg ggcaaacgcc cccaaaagag tgctttttaa 180agaggttaag gcaaaatcaa gctctttagt atttaatctt aaaaaatact aaaagccttt 240ttatgggcta acaccacaca aaaagcgtca aaatcaaaaa aatgacaaaa ttttccccaa 300atgacaaaaa aaaaaaaaaa cgattttatg ctatattaac gaaatcttgt gataagatct 360tattctttta aaagatttac ctaaccattt taatttcaag gagaaaacat gaaaaaaaac 420ccttttactc tctctctctc tctctcgttt ttgctccacg ctgaagacga cggcttttac 480acaagcgtag gctatcaaat cggtgaagcc gctcaaatgg taaccaacac caaaggcatc 540caacagcttt cagacaatta tgaaaagctg aacaatcttt tgaataatta cagcacccta 600aacaccctta tcaaattatc cgctgatccg agtgcgatta acgacgcaag ggataatcta 660ggctcaagtg ctaagaattt gcttgatgtt aaaaccaact ccccggccta tcaagccgtg 720cttttagcgt tgaatgcggc ggtggggttg tggcaagtta caagctacgc ttttactgct 780tgtggtcctg gcagtaacga gagcgcaaat ggaggtatcc aaacttttaa taatgtgcca 840ggacaaaaga cgacaaccat cacttgcaat tcgtattatc aaccaggaca tggtgggcct 900atatccactg caaactatgc aaaaatcaat caagcctatc aaatcattca aaaggctttg 960acagccaatg aagctaatgg agatggggtc cccgttttaa gcgacaccac tacaaaactt 1020gatttcacta ttcaaggaga caaaagaacg ggtggccgac caaatacacc taaaaagttc 1080ccatggagtg atgggaaata tattcacacc caatggattg acacaacacc acaatcaaca 1140gaaacaaaga tcaacacaga aaataacgct caagagcttt taaaacaagc gagcatcatt 1200atcactaccc taaatgaggc atgcccaaac ttccaaaatg gtggtagcgg ttattggcaa 1260gggataagcg gcaatgggac aatgtgtggg atgtttaaga atgaaatcag cgctatccaa 1320ggcatgatcg ctaacgcgca agaagctgtc gcgcaaagta aaatcgttag tgaaaatgcg 1380caaaatcaaa acaacttgga tactggaaaa ccattcaacc ctttcacgga cgctagcttc 1440gctcaaagca tgctcaaaaa cgctcaagcc caagcagaga ttttaaacca agccgaacaa 1500gtggtgaaaa actttgaaaa aatccctaaa aattttgtat cagactcttt aggggtgtgt 1560tatgaagagc aagggggtga gcgtaggggc accaatccag gtcaggttac ttctaacact 1620ttcgcttccg gttgcgccta tgtggagcaa accataacga atctaaacaa cagcatcgct 1680cattttggca ctcaagagca gcagatacag caagctgaaa acatcgctga cactctagtg 1740aatttcaaat ctagatacag cgaattaggg aatacttata acagcatcac tactgcgctc 1800tccaaagtcc ctaacgcgca aagcttgcaa aacgtggtgg gaaaaaagaa taacccctat 1860agcccgcaag gcatagaaac caattactac ttgaatcaaa actcttacaa ccaaatccaa 1920accatcaacc aagaattagg gcgtaacccc tttaggaaag tgggcatcgt cagttctcaa 1980accaacaatg gtgccatgaa tgggatcggt atccaggtgg gctacaagca attctttggg 2040caaaaaagga aatggggtgc aagatactac ggcttttttg attacaacca tgcgttcatt 2100aaatccagct tcttcaactc ggcttctgac gtgtggactt atggttttgg agcggacgct 2160ctttataact tcatcaacga taaagccacc aatttcttag gcaaaaacaa caagctttct 2220gtggggcttt ttggcgggat tgcgttagcg ggcacttcat ggcttaattc tgaatacgtg 2280aatttagcca ccatgaataa cgtctataac gctaaaatga acgtggcgaa cttccaattc 2340ttattcaaca tgggagtgag gatgaattta gccagatcca agaaaaaagg cagcgatcat 2400gcggctcagc atggcattga gttagggctt aaaatcccca ccattaacac gaactactat 2460tcctttatgg gggctgaact caaataccgc aggctctata gcgtgtattt gaattatgtg 2520ttcgcttact agaaactaaa aatcctttgt ggaactccct ttttaagggg tttcttttaa 2580agcctttatt tttttttgga ggggtttaat ttttttgaaa cctttgtttt tgaattctct 2640ttttaatggg tttctttttt gaactctttg ttttgaactc ctttttttga actccctttt 2700ttaaaccctt tcttttttaa aattctcttt tttggggggt ttgatgaaaa atcctttttt 2760agcgttttgg tattggttag t 2781 3 744 PRT Helicobacter pylori PEPTIDE (24)..(64) N-terminal domain of the BabA adhesin 3Ser Lys Lys Glu Lys Lys His Ile Leu Ser Leu Thr Leu Gly Ser Leu 1 5 10 15Leu Val Ser Thr Leu Ser Ala Glu Asp Asp Gly Phe Tyr Thr Ser Val 20 25 30Gly Tyr Gln Ile Gly Glu Ala Ala Gln Met Val Thr Asn Thr Lys Gly 35 40 45Ile Gln Asp Leu Ser Asp Asn Tyr Glu Asn Leu Ser Lys Leu Leu Thr 50 55 60Arg Tyr Ser Thr Leu Asn Thr Leu Ile Lys Leu Ser Ala Asp Pro Ser 65 70 75 80Ala Ile Asn Ala Ala Arg Glu Asn Leu Gly Ala Ser Ala Lys Asn Leu 85 90 95Ile Gly Asp Thr Lys Asn Ser Pro Ala Tyr Gln Ala Val Leu Leu Ala 100 105 110Ile Asn Ala Ala Val Gly Phe Trp Asn Val Leu Gly Tyr Ala Thr Gln 115 120 125Cys Gly Gly Asn Ala Asn Gly Gln Glu Ser Thr Ser Ser Thr Thr Ile 130 135 140Phe Asn Asn Glu Pro Gly Tyr Arg Ser Thr Ser Ile Thr Cys Ser Leu145 150 155 160Asn Arg Tyr Lys Pro Gly Tyr Tyr Gly Pro Met Ser Ile Glu Asn Phe 165 170 175Lys Lys Leu Asn Glu Ala Tyr Gln Ile Leu Gln Thr Ala Leu Asn Lys 180 185 190Gly Leu Pro Ala Leu Lys Glu Asn Asn Gly Thr Val Ser Val Thr Tyr 195 200 205Thr Tyr Thr Cys Ser Gly Glu Gly Asn Asp Asn Cys Ser Lys Lys Ala 210 215 220Thr Gly Val Ser Asp Gln Asn Gly Gly Thr Lys Thr Lys Thr Gln Thr225 230 235 240Ile Asp Gly Lys Thr Val Thr Thr Thr Ile Ser Ser Lys Val Val Asp 245 250 255Ser Gln Ala Lys Gly Asn Thr Thr Arg Val Ser Tyr Thr Glu Ile Thr 260 265 270Asn Lys Leu Asp Gly Val Pro Asp Ser Ala Gln Ala Leu Leu Ala Gln 275 280 285Ala Ser Thr Leu Ile Asn Thr Ile Asn Thr Ala Cys Pro Tyr Phe Ser 290 295 300Val Thr Asn Lys Ser Gly Gly Pro Gln Met Glu Pro Thr Arg Gly Lys305 310 315 320Leu Cys Gly Phe Thr Glu Glu Ile Ser Ala Ile Gln Lys Met Ile Thr 325 330 335Asp Ala Gln Glu Leu Val Asn Gln Thr Ser Val Ile Asn Glu His Glu 340 345 350Gln Ser Thr Pro Val Gly Gly Asn Asn Gly Lys Pro Phe Asn Pro Phe 355 360 365Thr Asp Ala Ser Phe Ala Gln Gly Met Leu Ala Asn Ala Ser Ala Gln 370 375 380Ala Lys Met Leu Asn Leu Ala His Gln Val Gly Gln Thr Ile Asn Pro385 390 395 400Asp Asn Leu Thr Gly Thr Phe Lys Asn Phe Val Thr Gly Phe Leu Ala 405 410 415Thr Cys Asn Asn Lys Ser Thr Ala Gly Thr Ser Gly Thr Gln Gly Ser 420 425 430Pro Pro Gly Thr Val Thr Thr Gln Thr Phe Ala Ser Gly Cys Ala Tyr 435 440 445Val Glu Gln Thr Ile Thr Asn Leu Asn Asn Ser Ile Ala His Phe Gly 450 455 460Thr Gln Glu Gln Gln Ile Gln Gln Ala Glu Asn Ile Ala Asp Thr Leu465 470 475 480Val Asn Phe Lys Ser Arg Tyr Ser Glu Leu Gly Asn Thr Tyr Asn Ser 485 490 495Ile Thr Thr Ala Leu Ser Lys Val Pro Asn Ala Gln Ser Leu Gln Asn 500 505 510Val Val Gly Lys Lys Asn Asn Pro Tyr Ser Pro Gln Gly Ile Glu Thr 515 520 525Asn Tyr Tyr Leu Asn Gln Asn Ser Tyr Asn Gln Ile Gln Thr Ile Asn 530 535 540Gln Glu Leu Gly Arg Asn Pro Phe Arg Lys Val Gly Ile Val Ser Ser545 550 555 560Gln Thr Asn Asn Gly Ala Met Asn Gly Ile Gly Ile Gln Val Gly Tyr 565 570 575Lys Gln Phe Phe Gly Gln Lys Arg Lys Trp Gly Ala Arg Tyr Tyr Gly 580 585 590Phe Phe Asp Tyr Asn His Ala Phe Ile Lys Ser Ser Phe Phe Asn Ser 595 600 605Ala Ser Asp Val Trp Thr Tyr Gly Phe Gly Ala Asp Ala Leu Tyr Asn 610 615 620Phe Ile Asn Asp Lys Ala Thr Asn Phe Leu Gly Lys Asn Asn Lys Leu625 630 635 640Ser Val Gly Leu Phe Gly Gly Ile Ala Leu Ala Gly Thr Ser Trp Leu 645 650 655Asn Ser Glu Tyr Val Asn Leu Ala Thr Met Asn Asn Val Tyr Asn Ala 660 665 670Lys Met Asn Val Ala Asn Phe Gln Phe Leu Phe Asn Met Gly Val Arg 675 680 685Met Asn Leu Ala Arg Ser Lys Lys Lys Gly Ser Asp His Ala Ala Gln 690 695 700His Gly Ile Glu Leu Gly Leu Lys Ile Pro Thr Ile Asn Thr Asn Tyr705 710 715 720Tyr Ser Phe Met Gly Ala Glu Leu Lys Tyr Arg Arg Leu Tyr Ser Val 725 730 735Tyr Leu Asn Tyr Val Phe Ala Tyr 740 4 707 PRT Helicobacter pylori PEPTIDE (46)..(59) Corresponding to the N-terminal domain of the BabA adhesin 4Met Lys Lys Asn Pro Phe Thr Leu Ser Leu Ser Leu Ser Phe Leu Leu 1 5 10 15His Ala Glu Asp Asp Gly Phe Tyr Thr Ser Val Gly Tyr Gln Ile Gly 20 25 30Glu Ala Ala Gln Met Val Thr Asn Thr Lys Gly Ile Gln Gln Leu Ser 35 40 45Asp Asn Tyr Glu Lys Leu Asn Asn Leu Leu Asn Asn Tyr Ser Thr Leu 50 55 60Asn Thr Leu Ile Lys Leu Ser Ala Asp Pro Ser Ala Ile Asn Asp Ala 65 70 75 80Arg Asp Asn Leu Gly Ser Ser Ala Lys Asn Leu Leu Asp Val Lys Thr 85 90 95Asn Ser Pro Ala Tyr Gln Ala Val Leu Leu Ala Leu Asn Ala Ala Val 100 105 110Gly Leu Trp Gln Val Thr Ser Tyr Ala Phe Thr Ala Cys Gly Pro Gly 115 120 125Ser Asn Glu Ser Ala Asn Gly Gly Ile Gln Thr Phe Asn Asn Val Pro 130 135 140Gly Gln Lys Thr Thr Thr Ile Thr Cys Asn Ser Tyr Tyr Gln Pro Gly145 150 155 160His Gly Gly Pro Ile Ser Thr Ala Asn Tyr Ala Lys Ile Asn Gln Ala 165 170 175Tyr Gln Ile Ile Gln Lys Ala Leu Thr Ala Asn Glu Ala Asn Gly Asp 180 185 190Gly Val Pro Val Leu Ser Asp Thr Thr Thr Lys Leu Asp Phe Thr Ile 195 200 205Gln Gly Asp Lys Arg Thr Gly Gly Arg Pro Asn Thr Pro Lys Lys Phe 210 215 220Pro Trp Ser Asp Gly Lys Tyr Ile His Thr Gln Trp Ile Asp Thr Thr225 230 235 240Pro Gln Ser Thr Glu Thr Lys Ile Asn Thr Glu Asn Asn Ala Gln Glu 245 250 255Leu Leu Lys Gln Ala Ser Ile Ile Ile Thr Thr Leu Asn Glu Ala Cys 260 265 270Pro Asn Phe Gln Asn Gly Gly Ser Gly Tyr Trp Gln Gly Ile Ser Gly 275 280 285Asn Gly Thr Met Cys Gly Met Phe Lys Asn Glu Ile Ser Ala Ile Gln 290 295 300Gly Met Ile Ala Asn Ala Gln Glu Ala Val Ala Gln Ser Lys Ile Val305 310 315 320Ser Glu Asn Ala Gln Asn Gln Asn Asn Leu Asp Thr Gly Lys Pro Phe 325 330 335Asn Pro Phe Thr Asp Ala Ser Phe Ala Gln Ser Met Leu Lys Asn Ala 340 345 350Gln Ala Gln Ala Glu Ile Leu Asn Gln Ala Glu Gln Val Val Lys Asn 355 360 365Phe Glu Lys Ile Pro Lys Asn Phe Val Ser Asp Ser Leu Gly Val Cys 370 375 380Tyr Glu Glu Gln Gly Gly Glu Arg Arg Gly Thr Asn Pro Gly Gln Val385 390 395 400Thr Ser Asn Thr Phe Ala Ser Gly Cys Ala Tyr Val Glu Gln Thr Ile 405 410 415Thr Asn Leu Asn Asn Ser Ile Ala His Phe Gly Thr Gln Glu Gln Gln 420 425 430Ile Gln Gln Ala Glu Asn Ile Ala Asp Thr Leu Val Asn Phe Lys Ser 435 440 445Arg Tyr Ser Glu Leu Gly Asn Thr Tyr Asn Ser Ile Thr Thr Ala Leu 450 455 460Ser Lys Val Pro Asn Ala Gln Ser Leu Gln Asn Val Val Gly Lys Lys465 470 475 480Asn Asn Pro Tyr Ser Pro Gln Gly Ile Glu Thr Asn Tyr Tyr Leu Asn 485 490 495Gln Asn Ser Tyr Asn Gln Ile Gln Thr Ile Asn Gln Glu Leu Gly Arg 500 505 510Asn Pro Phe Arg Lys Val Gly Ile Val Ser Ser Gln Thr Asn Asn Gly 515 520 525Ala Met Asn Gly Ile Gly Ile Gln Val Gly Tyr Lys Gln Phe Phe Gly 530 535 540Gln Lys Arg Lys Trp Gly Ala Arg Tyr Tyr Gly Phe Phe Asp Tyr Asn545 550 555 560His Ala Phe Ile Lys Ser Ser Phe Phe Asn Ser Ala Ser Asp Val Trp 565 570 575Thr Tyr Gly Phe Gly Ala Asp Ala Leu Tyr Asn Phe Ile Asn Asp Lys 580 585 590Ala Thr Asn Phe Leu Gly Lys Asn Asn Lys Leu Ser Val Gly Leu Phe 595 600 605Gly Gly Ile Ala Leu Ala Gly Thr Ser Trp Leu Asn Ser Glu Tyr Val 610 615 620Asn Leu Ala Thr Met Asn Asn Val Tyr Asn Ala Lys Met Asn Val Ala625 630 635 640Asn Phe Gln Phe Leu Phe Asn Met Gly Val Arg Met Asn Leu Ala Arg 645 650 655Ser Lys Lys Lys Gly Ser Asp His Ala Ala Gln His Gly Ile Glu Leu 660 665 670Gly Leu Lys Ile Pro Thr Ile Asn Thr Asn Tyr Tyr Ser Phe Met Gly 675 680 685Ala Glu Leu Lys Tyr Arg Arg Leu Tyr Ser Val Tyr Leu Asn Tyr Val 690 695 700Phe Ala Tyr705 5 20 PRT Helicobacter pylori 5Glu Asp Asp Gly Phe Tyr Thr Ser Val Gly Tyr Gln Ile Gly Glu Ala 1 5 10 15Ala Gln Met Val 20 6 60 DNA Helicobacter pylori 6gaagacgacg gcttttacac aagcgtaggc tatcaaatcg gtgaagccgc tcaaatggta 60 7 41 PRT Helicobacter pylori 7Glu Asp Asp Gly Phe Tyr Thr Ser Val Gly Tyr Gln Ile Gly Glu Ala 1 5 10 15Ala Gln Met Val Thr Asn Thr Lys Gly Ile Gln Asp Leu Ser Asp Asn 20 25 30Tyr Glu Asn Leu Ser Lys Leu Leu Thr 35 40 8 20 PRT Helicobacter pylori 8Glu Asp Asp Gly Phe Tyr Thr Ser Val Gly Tyr Gln Ile Gly Glu Ala 1 5 10 15Ala Gln Met Val 20

Claims

1. An isolated and purified bacterial blood group anitgen binding adhesin protein (BabA) from Helicobacter pylori species, wherein said protein binds specifically to fucosylated Lewisb type I and H-1 blood group anitgen-glycoconjugates and,wherein said protein contains less than 20% bacterial protein impurities, has a molecular weight in the interval of 73 to 75 kDa as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and is not a HopA, HopB, HopC, HopD, or HopE protein.

2. The adhesion protein of claim 1, wherein the SEQ ID NO:5EDDGFYTSVGYQIGEAAQMV is in an amino terminal position.

3. The adhesion protein of claim 1, wherein the protein has a molecular weight of about 73.5 kDa, as determined by SDS-PAGE.

4. The adhesion protein of claim 3, wherein said Helicobacter pylori species is Helicobacter pylori strain CCUG 17875.

5. An immunogenic composition comprising an adhesion protein according to any one of claim 1, 2, 3, or 4, optionally together with pharmaceutically acceptable excipients.