WO2000004922A1 - Vaccines against escherichia coli o157 infection - Google Patents

Abstract

This invention relates to conjugates of the O-specific polysaccharide of E. coli O157 with a carrier, and compositions thereof, and to methods of using of these conjugates and/or compositions thereof for eliciting an immunogenic response in mammals, including responses which provide protection against, or reduce the severity of, bacterial infections. More particularly it relates to the use of polysaccharides containing the tetrasaccharide repeat unit: (→3)-α-D-GalpNAc-(1→2)-α-D-PerpNAc-(1→3)-α-L-Fucp-(1→4)-β-D-Glcp-(1→, and conjugates thereof, to induce serum antibodies having bactericidal (killing) activity against hemolytic-uremic syndrome (HUS) causing E. coli, in particular E. coli O157. The conjugates, and compositions thereof, are useful as vaccines to induce serum antibodies which have bactericidal or bacteriostatic activity against E. coli, in particular E. coli O157, and are useful to prevent and/or treat illnesses caused by E. coli O157. The invention further relates to the antibodies which immunoreact with the O-specific polysaccharide of E. coli O157 and/or the carrier, that are induced by these conjugates and/or compositions thereof. The invention also relates to methods and kits using one or more of the polysaccharides, conjugates or antibodies described above.

Description

Vaccines Against Escherichia coli O157 Infection

FIELD OF THE INVENTION

This invention relates to conjugates ofthe O-specifϊc polysaccharide of Shiga toxin-producing bacteria, such as E. coli 0157, with a carrier, and compositions thereof, and to methods of using of these conjugates and/or compositions thereof for eliciting an immunogenic response in mammals, including responses which provide protection against, or reduce the severity of, bacterial infections. More particularly it relates to the use of polysaccharides containing the tetrasaccharide repeat unit:

(→3)-α-Z⁾-G NAc-(l→2)^-I⁾-PerpNAc-(l→3)-α-i -Fuςp-(l→4)-P-Z⁾-Glςp-(l→), and conjugates thereof, to induce serum antibodies having bactericidal (killing) activity against E. coli, in particular E. coli 0157. The conjugates, and compositions thereof, are useful as vaccines to induce serum antibodies which have bactericidal or bacteriostatic activity against against E. coli, in particular E. coli 0157, and are useful to prevent and/or treat illnesses caused by E. coli 0157.

The invention further relates to the antibodies which immunoreact with the O-specific polysaccharide of E. coli 0157 and/or the carrier, that are induced by these conjugates and/or compositions thereof. The invention also relates to methods and kits for detection, identification, and/or diagnosis of E. coli 0157, using one or more ofthe polysaccharides, conjugates or antibodies described above.

BACKGROUND

The most successful of all carbohydrate pharmaceuticals so far have been the carbohydrate-based, antibacterial vaccines [1]. The basis of using carbohydrates as vaccine components is that the capsular polysaccharides and the O- specific polysaccharides on the surface of pathogenic bacteria are both protective antigens and essential virulence factors. The first saccharide-based vaccines contained capsular polysaccharides of Pneumococci: in the United States a 14-valent vaccine was licensed in 1978 followed by a 23-valent vaccine in 1983. Other capsular polysaccharides licensed for human use include a tetravalent meningococcal vaccine and the Vi polysaccharide of Salmonella typhi for typhoid fever. The inability of most polysaccharides to elicit protective levels of anti-carbohydrate antibodies in infants and adults with weakened immune systems could be overcome by their covalent attachment to proteins that conferred T-cell dependent properties [2]. This principle led to the construction of vaccines against Haemophilus influenzae b (Hib) [3] and in countries where these vaccines are routinely used, meningitis and other diseases caused by Hib have been virtually eliminated [4]. Extension ofthe conjugate technology to the O-specific polysaccharides of Gram-negative bacteria has provided a new generation of glycoconjugate vaccines that are undergoing various phases of clinical trials [5].

Escherichia coli O157:H7, an emerging infectious agent, was first recognized as a human pathogen in 1983 [6]. Diseases caused by this pathogen have subsequently been recognized worldwide [7]. Infection with E. coli 0157 causes a spectrum of illnesses with high morbidity and mortality, ranging from watery diarrhea to hemorrhagic colitis and the extraintestinal complication of hemolytic-uremic syndrome (HUS). HUS can lead to acute renal failure requiring dialysis, and in children and infants this complication has a considerable mortality. In some studies, E. coli 0157 was the most common cause of dysentery in patients seen in hospital clinics [8].

E. coli strains associated with HUS produce at least one toxin identical to the exotoxin of Shigella dysenteriae serotype 1, referred to herein as Shiga toxin 1 (Stxl). This toxin has been variously referred to in the literature as Vero cytotoxin 1 (VT1), Shiga-like toxin 1 (SLT-I), and Shiga toxin l(Stx-I or Stxl). In some cases a second toxin (variously referred to as VT2, SLT-II, Stx-II, or Stx2), structurally and functionally related to Stxl and having a cross-reactive A subunit, is also produced. Infection with Stx-producing organisms has been correlated with HUS, and E. coli O157:H7 is a common serotype that produces these toxins. However, strains of E. coli 0157 without Stx have been isolated from patients with hemorrhagic colitis.

The pathogenicity of E. coli 0157 has been compared to that of Shigella dysenteriae type 1 [9, 10]. Both E. coli 0157 and S. dysenteriae type 1 secrete almost identical exotoxins (Stxl or Stx2) and cause bloody diarrhea, with its complications, only in humans. Antibiotic treatment does not ameliorate the course of enteritis caused by E. coli 0157, and it may in fact increase the incidence of HUS caused by E. coli and S. dysenteriae type 1 [11,12]. Unlike S. dysenteriae type 1, which is confined to humans, E. coli O157:H7 lives in cattle and in other domesticated animals without causing symptoms. The feces of infected animals serve as a source of E. coli 0157 infection in humans, through contamination of drinking water and meat.

Most adults have low or nondetectable levels of serum antibodies to E. coli 0157 O-SP and to Shiga toxins. High levels of O-SP antibodies and low or nondetectable levels of antitoxin are regularly found following infection with E. coli 0157 and the subsequent complication HUS. It is not known whether immunity follows infection with this pathogen.

Although there is no consensus on the host factors that might confer immunity to E. coli 0157, the O-specific polysaccharide portion ofthe lipopolysaccharides ofthe similar genus Shigella have emerged as possible protective antigens [13,14]. These polysaccharides were shown to be essential for the virulence of Shigella, and it is now well-established that the protection is serotype specific. Since each serotype is characterized by a distinct O-specific polysaccharide, it is fair to say that protection against E. coli 0157 is also O-specific polysaccharide specific. The safety and immunogenicity of a protein conjugate ofthe O-specific polysaccharides of S. sonnei, S.flexneri 2a, and S. dysenteriae type 1 has been demonstrated in human volunteers, and preliminary clinical trials have established the efficacy of these vaccines [9, 15, 16, 17].

The immunogenicity of saccharides, alone or as protein conjugates, is related to several variables: 1) species and the age ofthe recipient; 2) molecular weight ofthe saccharide; 3) density ofthe saccharide on the protein; 4) configuration ofthe conjugate (single vs. multiple point attachment); and 5) the immunologic properties ofthe protein.

Because high molecular weight polysaccharides can induce the synthesis of antibodies from B-cells alone, they are described as T-independent antigens. Three properties of polysaccharides are associated with T-independence; 1) their repetitive polymeric nature, which results in one molecule having multiple identical epitopes; 2) a minimum molecular weight that is related to their ability to adhere to and cross-link membrane-bound IgM receptors, resulting in signal transduction and antibody synthesis; and 3) resistance to degradation by mammalian enzymes. Most capsular polysaccharides are of comparatively high molecular weight (>150 kD), and elicit antibodies in older children and in adults but not in infants and young children. O-SPs are of lower molecular weight (<100 kD), and may be considered to be haptens because they combine with antibody (are antigenic) but do not elicit antibody synthesis (are not immunogenic). The immunogenicity of O-SPs as conjugates may be explained by two factors: 1) the increase in molecular weight that allows the O-SP to adhere to a greater number of membrane-bound IgM and induce signal transduction to the B-cell; and 2) their protein component, which is catabolized by the O-SP stimulated B cell resulting in a peptide-histocompatibility II antigen signal to T cells.

Synthesis of conjugates for use as vaccines in humans has special considerations. LPS is not suitable for parenteral administration to humans because of toxicity mediated by the lipid A domain. Usually, O-SP is prepared by treatment of LPS with either acid or hydrazine in order to remove fatty acids from lipid A. The resultant products retain the core region and the O-SP with its heterogeneous range of molecular weights (M_r). Conjugates are prepared by schemes that bind the carrier to the O-SP at multiple sites along the O-SP, or attempt to activate one residue ofthe core region.

In the case of E. coli 0157, vaccine development has been hindered because there is little information about mechanisms of immunity [9], and there are no valid animal models for diseases caused by E. coli O157[10].

There have been some efforts to date to attempt to obtain effective vaccine compositions against E. coli. See, e.g., Cryz et al. (U.S. Patent 5,370,872), which describes the isolation of O-SP derived from LPS of 12 serotypes of E. coli and their covalent linkage to P. aeruginosa toxin A as a carrier protein [18]. The twelve monovalent conjugates were combined to form a polyvalent vaccine, which was described as being safe and immunogenic in both rabbits and humans when administered by injection. An antibody response to both the O-SP and toxin A moieties was reported, and protection of rabbits against E. coli sepsis was demonstrated upon passive immunization with the resulting IgG antibodies. However, neither bactericidal activity ofthe antibodies nor protection after vaccination with the conjugates was shown, and antibodies against E. coli strain 0157 and protection against E. coli 0157 infection are not mentioned.

Because anti-LPS or anti-O-SP antibody-mediated protection is likely to be serotype-specific, it is unlikely that the polyvalent vaccine described in US Patent 5,370,872 would induce a significant level of antibodies against E. coli 0157 O-SP or LPS. There remains a need, therefore, for compositions and methods of inducing a significant level of antibodies against E. coli 0157. There also remains a need compositions and methods for inducing antibodies which have bactericidal activity against E. coli 0157, and which also prevent or ameliorate HUS.

BRIEF DESCRIPTION OF THE INVENTION

It is an object ofthe invention to produce antigens based on the O- specific polysaccharide of Shiga toxin-producing bacteria, particularly E. coli 0157, conjugated with a carrier, and compositions thereof, and to methods of using of these conjugates and/or compositions thereof for eliciting an immunogenic response in mammals, including responses which provide protection against, or reduce the severity of, bacterial infections. More particularly, it is an object ofthe invention to provide conjugates having polysaccharides containing the tetrasaccharide repeat unit: (→3)-α-£⁾-Gak7NAc-(l→2)-α-Z)-PerpNAc-(l→3)-α-Z-Fucp-(l→4)-β-j9-Glc^-(l→), and compositions thereof, to induce serum antibodies having bactericidal (killing) activity against E. coli, in particular E. coli 0157. The conjugates, and compositions thereof, are useful as vaccines to induce serum antibodies which have bactericidal or bacteriostatic activity against against E. coli, in particular E. coli 0157, and are useful to prevent and/or treat illnesses caused by E. coli 0157.

It is yet another object ofthe present invention to provide conjugates of E. coli 0157 O-SP bound to the non-toxic B-subunit of Shiga toxin 1 (StxBl), or mutated non- toxic holotoxin of Shiga toxin 1 or Shiga toxin 2. These conjugates have the advantage of inducing both (1) serum IgG anti-O157-LPS with bactericidal activity, and (2) neutralizing antibodies to Shiga toxin 1 or Shiga toxin 2 (Stxl or Stx2)[l 9,20,21].

It is also an object ofthe invention to provide antibodies which immunoreact with the O-specific polysaccharide of E. coli 0157 and/or the carrier, that are induced by these conjugates and/or compositions thereof. Such antibodies may be isolated, or may be provided in the form of serum containing these antibodies.

It is also an object ofthe invention to provide a method for the treatment or prevention of E. coli 0157 infection in a mammal, by administration of compositions containing the antibodies ofthe invention, or serum containing the antibodies ofthe invention.

The invention also provides methods and kits for identifying, detecting, and/or diagnosing E. coli 0157 infection or colonization using the antibodies which immunoreact with the O-specific polysaccharide of E. coli. The invention also relates to methods and kits for identifying, detecting and/or diagnosing the presence of Shiga toxins 1 or 2.

DETAILED DESCRIPTION OF THE INVENTION

The invention provides conjugates of an E. coli 0157 O-specific polysaccharide covalently bound, either directly or through a linker, to a carrier, and compositions thereof. The present invention also encompasses mixtures of such conjugates and compositions thereof. In a preferred embodiment, the carrier is the non-toxic B subunit of Shiga toxin 1 or 2 (StxBl, StxB2), or a non-toxic mutant of Stxl or Stx2 holotoxin. In yet another preferred embodiment, the particular E. coli 0157-Stx conjugate is part of a composition containing O-SP-carrier conjugates from other E. coli strains that commonly cause HUS, to form a multivalent vaccine for broad coverage against HUS. Hyperimmune plasma containing both anti-LPS and neutralizing antibodies to Stxs are expected to provide protective and therapeutic effects in at-risk individuals and in patients during outbreaks.

The invention also provides methods of using these conjugates or compositions thereof to induce in mammals, in particular, humans, the production of antibodies which immunoreact with the O-specific polysaccharide of E. coli 0157. In the preferred embodiment, antibodies which immunoreact with Shiga toxin 1 or Shiga toxin 2 are also produced. The antibodies which immunoreact with the O-specific polysaccharide of E. coli 0157 are useful for the identification, detection, and/or diagnosis of E. coli 0157 colonization and/or infection. Antibodies which have bactericidal or bacteriostatic activity against E. coli 0157 are useful to prevent and/or treat illnesses caused by E. coli 0157. Antibodies which immunoreact with Shiga toxins 1 and 2 are useful to neutralize Shiga toxins 1 and 2, and either decrease the incidence and/or severity of hemolytic-uremic syndrome, or prevent the increase of its incidence and/or severity, in established infections.

Pharmaceutical compositions of this invention are capable, upon injection into a human of an amount containing 25 μg of E. coli 0157 O-specific polysaccharide, of inducing in the serum bactericidal activity against E. coli 0157, such that the serum kills, in the presence of complement, 50% or more of E. coli 0157 at a serum dilution of 1300: 1 or more. Preferred compositions can induce serum bactericidal activity against E. coli 0157 such that the serum kills 50% or more of E. coli 0157 at a serum dilution of 32,000:1 or more, and the most preferred compositions can induce serum bactericidal activity against E. coli 0157 such that the serum kills 50% or more of E. coli 0157 at a serum dilution of 64,000: 1 or more. The O-SP conjugate vaccines of this invention are designed to induce serum IgG antibodies that will inactivate an inoculum of E. coli 0157 at the entrance ofthe jejunum before an infection is established.

The invention also provides a saccharide-based vaccine, which is intended for active immunization for prevention of E. coli 0157 infection, and for preparation of immune antibodies as a therapy, preferably for established infections. The vaccines of this invention are designed to confer specific preventative immunity against infection with E. coli 0157, and to induce antibodies specific to E. coli 0157 O-SP and LPS. The E. coli 0157 vaccine is composed of non-toxic bacterial components, suitable for infants, children of all ages, and adults.

The conjugates of this invention, and/or compositions thereof, as well as the antibodies thereto, will be useful in increasing resistance to, preventing, ameliorating, and/or treating E. coli 0157 infection in humans, and in reducing or preventing E. coli 0157 colonization in humans. This invention also provides compositions, including but not limited to, mammalian serum, plasma, and immunoglobulin fractions, which contain antibodies which are immunoreactive with E. coli 0157 O-SP, and which preferably also contain antibodies which are immunoreactive with Shiga toxins 1 or 2, in particular with the B subunit of Shiga toxins 1 or 2. These compositions, in the presence of complement, have bacteriostatic or bactericidal activity against E. coli 0157. These antibodies and antibody compositions are useful to prevent, treat, or ameliorate infection and disease caused by the microorganism. The invention also provides such antibodies in isolated form.

High titer anti-0157 sera, or antibodies isolated therefrom, could be used for therapeutic treatment for patients with E. coli 0157 infection or hemolytic- uremic syndrome (HUS). Antibodies elicited by the O-SP conjugates of this invention may be used for the treatment of established E. coli 0157 infections, and are also useful in providing passive protection to an individual exposed to E. coli 0157.

The present invention also provides diagnostic tests and/or kits for E. coli 0157 infection and/or colonization, using the conjugates and/or antibodies ofthe present invention, or compositions thereof.

The present invention also provides an improved method for synthesizing an O-SP peptide conjugate, particularly the E. coli 0157 O-SP conjugated to the B subunit of Shiga toxin 1 or 2 (Stxl or Stx2), or to a mutant, nontoxic Stxl or Stx2 holotoxin.

A number of primary uses for the conjugates of this invention are envisioned. The E. coli LPS-protein conjugates of this invention, and the antibodies they induce, are expected to be useful for several purposes, including but not limited to:

1) a vaccine for high-risk groups (children under 5 and senior citizens);

2) high-titered globulin for plasmapheresis, for prophylaxis and treatment of E. coli 0157-infected patients; and

3) diagnostic reagents for detecting and/or identifying E. coli 0157.

The invention is intended to be included in the routine immunization schedule of infants and children, and in individuals at risk for E. coli 0157 infection. It is also planned to be used for intervention in epidemics caused by E. coli 0157. Additionally, it is may be used as a component of a multivalent vaccine for E. coli 0157 and other enteric pathogens, useful for example for the routine immunization of infants. The invention is also intended to prepare antibodies with bacteriostatic bactericidal activity toward E. coli 0157, for therapy of established infection. The invention is also intended to provide a diagnostic test for E. coli 0157 infection and/or colonization. Definitions

Gdλp = galactosaminopyranosyl; Peip = perosaminopyranosyl; Fucp = fucopyranosyl; G cp = glucopyranosyl.

As used herein, the term "O-SP" when used alone refers generically to O-specific polysaccharide, whether produced by acidolysis or hydrazinolysis of lipopolysaccharide. When used in designating conjugates, however (e.g. O-SP-rEPA, DeA-LPS-rEPA, etc.) these products are differentiated by use ofthe term "O-SP" for O-specific polysaccharide produced by acidolysis, and the term "DeA-LPS" for O- specific polysaccharide produced by hydrazinolysis.

As used herein, the terms "immunoreact" and "immunoreactivity" refer to specific binding between an antigen or antigenic determinant-containing molecule and a molecule having an antibody combining site, such as a whole antibody molecule or a portion thereof.

As used herein, the term "antibody" refers to immunoglobulin molecules and immuno logically active portions of immunoglobulin molecules. Exemplary antibody molecules are intact immunoglobulin molecules, substantially intact immunoglobulin molecules and portions of an immunoglobulin molecule, including those portions known in the art as Fab, Fab', F(ab')₂ and F(v), as well as chimeric antibody molecules. Polymeric carriers

Carriers are chosen to increase the immunogenicity ofthe polysaccharide and/or to raise antibodies against the carrier which are medically beneficial. Carriers that fulfill these criteria are described in the art [22, 23, 24, 25]. A polymeric carrier can be a natural or a synthetic material containing one or more functional groups, for example primary and/or secondary amino groups, azido groups,^' or carboxyl groups. The carrier can be water soluble or insoluble.

Water soluble peptide carriers are preferred, and include but are not limited to natural or synthetic polypeptides or proteins, such as bovine serum albumin, and bacterial or viral proteins or non-toxic mutants or polypeptide fragments thereof, e.g., tetanus toxin or toxoid, diphtheria toxin or toxoid, Pseudomonas aeruginosa exotoxin or toxoid, recombinant Pseudomonas aeruginosa exoprotein A, pertussis toxin or toxoid, Clostridium perfringens and Clostridium welchii exotoxins or toxoids, mutant non-toxic Shiga toxin holotoxin, Shiga toxins 1 and 2, the B subunit of Shiga toxins 1 and 2, and hepatitis B surface antigen and core antigen.

Examples of water insoluble carriers include, but are not limited to, aminoalkyl SEPHAROSE, e. g., aminopropyl or aminohexyl SEPHAROSE (Pharmacia Inc., Piscataway, NJ), aminopropyl glass, and the like. Other carriers may be used when an amino or carboxyl group is added, for example through covalent linkage with a linker molecule. Methods for attaching polymeric carriers

Methods for binding a polysaccharide to a protein are well known in the art. For example, a polysaccharide containing at least one carboxyl group, through carbodiimide condensation, may be thiolated with cystamine, or aminated with adipic dihydrazide, diaminoesters, ethylenediamine and the like. Groups which can be introduced by such known methods include thiols, hydrazides, amines and carboxylic acids. Thiolated and aminated intermediates are stable, and may be freeze dried and stored cold. Thiolated intermediates may be covalently linked to a polymeric carrier containing a sulfhydryl group, such as a 2-pyridyldithio group. Aminated intermediates may be covalently linked to a polymeric carrier containing a carboxyl group through carbodiimide condensation. See for example reference [26], where 3 different methods for conjugating Shigella O-SP to tetanus toxoid are exemplified. Because the methods ofthe present invention better preserve the native structure ofthe antigen, they are preferred over methods which oxidize the polysaccharide with periodate [18]. The polysaccharide can be covalently bound to a carrier with or without a linking molecule. To conjugate without a linker, for example, a carboxyl- group-containing polysaccharide and an amino-group-containing carrier are mixed in the presence of a carboxyl activating agent, such as a carbodiimide, in a choice of solvent appropriate for both the polysaccharide and the carrier, as is known in the art [25]. The polysaccharide is often conjugated to a carrier using a linking molecule. A linker or crosslinking agent, as used in the present invention, is preferably a small linear molecule having a molecular weight of about 500 or less, and is non-pyrogenic and non-toxic in the final product form, for example as disclosed in references [22 - 25].

To conjugate with a linker or crosslinking agent, either or both ofthe polysaccharide and the carrier may be covalently bound to a linker first. The linkers or crosslinking agents are homobifunctional or heterobifunctional molecules, e.g., adipic dihydrazide, ethylenediamine, cystamine, N-succinimidyl 3-(2-pyridyldithio)propionate (SPDP), N-succinimidyl-N-(2-iodoacetyl)- β-alaninate-propionate (SIAP), succinimidyl 4-(N-maleimido-methyl)cyclohexane- 1-carboxylate (SMCC), 3,3'-dithiodipropionic acid, and the like. Also among the class of heterobifunctional linkers area omega-hydroxy and omega-amino alkanoic acids.

More specifically, attachment ofthe E. coli 0157 O-specific polysaccharide to a protein carrier can be accomplished by methods known to the art. In a preferred embodiment, the attachment is accomplished by first cyanating the O- specific polysaccharide with a cyanylation reagent, such as cyanogen bromide, N- cyano-N,N,N-triethylammonium tetrafluoroborate, l-cyano-4-(N,N- dimethylamino)pyridine tetrafluoroborate, or the like. Several such cyanylation reagents are known to those skilled in the art [27]. The resulting cyanated E. coli 0157 O-specific polysaccharide may then be reacted with a linker, such as a dicarboxylic acid dihydrazide, preferably adipic acid dihydrazide, so as to form a hydrazide-functionalized polysaccharide. This hydrazide-functionalized polysaccharide is then coupled to the carrier protein by treatment with a peptide coupling agent, preferably a water-soluble carbodiimide such as l-ethyl-3-(3- dimethylaminopropyl)carbodiimide, l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide methiodide, or the like.

More preferably, the cyanated E. coli 0157 O-specific polysaccharide is directly reacted with the carrier protein, without introduction of a linker. It has been found, surprisingly, that, in the exemplified conjugates, elimination ofthe customary linker provides a more effective immunogen in the case ofthe E. coli 0157 O-specific polysaccharide.

Regardless ofthe precise method used to prepare the conjugate, after the coupling reactions have been carried out the unbound materials are removed by routine physicochemical methods, such as for example gel filtration or ion exchange column chromatography, depending on the materials to be separated. The final conjugate consists ofthe polysaccharide and the carrier bound directly or through a linker. Dosage for Vaccination

The present inoculum contains an effective, immunogenic amount of a polysaccharide-carrier conjugate of this invention. The effective amount of polysaccharide-carrier conjugate per unit dose sufficient to induce an immune response to E. coli 0157 depends, among other things, on the species of mammal inoculated, the body weight ofthe mammal, and the chosen inoculation regimen, as is well known in the art. Inocula typically contain polysaccharide-carrier conjugates with concentrations of polysaccharide from about 1 micrograms to about 10 milligrams per inoculation (dose), preferably about 3 micrograms to about 100 micrograms per dose, and most preferably about 5 micrograms to 50 micrograms per dose.

The term "unit dose" as it pertains to the inocula refers to physically discrete units suitable as unitary dosages for mammals, each unit containing a predetermined quantity of active material (polysaccharide) calculated to produce the desired immunogenic effect in association with the required diluent.

Inocula are typically prepared as solutions in physiologically tolerable (acceptable) diluents such as water, saline, phosphate-buffered saline, or the like, to form an aqueous pharmaceutical composition. Adjuvants, such as aluminum hydroxide, may also be included in the compositions.

The route of inoculation may be intramuscular, subcutaneous or the like, which results in eliciting antibodies protective against E. coli 0157. In order to increase the antibody level, a second or booster dose may be administered approximately 4 to 6 weeks after the initial injection. Subsequent doses may be administered as indicated herein, or as desired by the practitioner. Antibodies

An antibody ofthe present invention in one embodiment is characterized as comprising antibody molecules that immunoreact with E. coli 0157 O-SP or LPS.

An antibody ofthe present invention is typically produced by immunizing a mammal with an immunogen or vaccine containing an E. coli 0157 polysaccharide-protein carrier conjugate to induce, in the mammal, antibody molecules having immunospecificity for the immunizing polysaccharide. Antibody molecules having immunospecificity for the protein carrier, such as the B subunit of Shiga toxins 1 or 2, will also be produced. The antibody molecules may be collected from the mammal and, optionally, isolated and purified by methods known in the art.

Human or humanized monoclonal antibodies are preferred, including those made by phage display technology, by hybridomas, or by mice with human immune systems. The antibody molecules ofthe present invention may be polyclonal or monoclonal. Monoclonal antibodies may be produced by methods known in the art. Portions of immunoglobulin molecules, such as Fabs, may also be produced by methods known in the art.

The antibody ofthe present invention may be contained in blood plasma, serum, hybridoma supernatants and the like. Antibody-containing serum of this invention will be capable of killing, in the presence of complement, 50% of E. coli Ol 57at a serum dilution of 1300: 1 or more, preferably will do so at a dilution of 32,000:1 or more, and most preferably will be capable of killing 50% of E. coli 0157 at a dilution of 64,000:1 or more.

Alternatively, the antibodies ofthe present invention are isolated to the extent desired by well known techniques such as, for example, ion chromatography or affinity chromatography. The antibodies may be purified so as to obtain specific classes or subclasses of antibody such as IgM, IgG, IgA, IgGi, IgG₂, IgG₃, IgG₄ and the like. Antibodies ofthe IgG class are preferred for purposes of passive protection. The antibodies ofthe present invention have a number of diagnostic and therapeutic uses. The antibodies can be used as an in vitro diagnostic agents to test for the presence of E. coli 0157 in biological samples or in meat and meat products, in standard immunoassay protocols. Such assays include, but are not limited to, agglutination assays, radioimmunoassays, enzyme-linked immunosorbent assays, fluorescence assays, Western blots and the like. In one such assay, for example, the biological sample is contacted with first antibodies ofthe present invention, and a labeled second antibody is used to detect the presence of E. coli 0157 to which the first antibodies have bound.

Such assays may be, for example, of direct format (where the labeled first antibody is reactive with the antigen), an indirect format (where a labeled second antibody is reactive with the first antibody), a competitive format (such as the addition of a labeled antigen), or a sandwich format (where both labeled and unlabelled antibody are utilized), as well as other formats described in the art.

The antibodies ofthe present invention are also useful in prevention and treatment of infections and diseases caused by E. coli 0157.

In providing the antibodies ofthe present invention to a recipient mammal, preferably a human, the dosage of administered antibodies will vary depending upon such factors as the mammal's age, weight, height, sex, general medical condition, previous medical history and the like.

In general, it is desirable to provide the recipient with a dosage of antibodies which is in the range of from about 1 mg/kg to about 10 mg/kg body weight ofthe mammal, although a lower or higher dose may be administered. The antibodies ofthe present invention are intended to be provided to the recipient subject in an amount sufficient to prevent, or lessen or attenuate the severity, extent or duration ofthe infection by E. coli 0157. Antibodies which immunoreact with Shiga toxin 1 or 2 are intended to be provided to the recipient subject in an amount sufficient to prevent, or lessen or attenuate the severity, extent or duration ofthe infection by Shigatoxin producing organisms, such as E. coli strains 0157, Ol l l, 026, and 017.

The administration ofthe agents ofthe invention may be for either "prophylactic" or "therapeutic" purpose. When provided prophylactically, the agents are provided in advance of any symptom. The prophylactic administration ofthe agent serves to prevent or ameliorate any subsequent infection. When provided therapeutically, the agent is provided at (or shortly after) the onset of a symptom of infection. The agent ofthe present invention may, thus, be provided prior to the anticipated exposure to E. coli 0157 (or other Shiga toxin producing bacteria), so as to attenuate the anticipated severity, duration or extent of an infection and disease symptoms, after exposure or suspected exposure to these bacteria, or after the actual initiation of an infection.

For all therapeutic, prophylactic and diagnostic uses, the polysaccharide-carrier conjugates of this invention, as well as antibodies and other necessary reagents and appropriate devices and accessories may be provided in kit form so as to be readily available and easily used.

The following examples are exemplary ofthe present processes and incorporate suitable process parameters for use herein. These parameters may be varied, however, and the following should not be deemed limiting.

EXAMPLES

Example 1 Conjugation of E. coli 0157 O-SP with Various Polypeptides

0157 LPS were detoxified by hydrolysis with acetic acid (designated O-SP) or with hydrazine (designated DeA-LPS) and then covalently bound to Clostridium welchii exotoxin C (Pig Bel toxoid [CW]), Pseudomonas aeruginosa recombinant exoprotein A (rEPA), or bovine serum albumin (BSA) [8]. These E.coli O157:H7 polysaccharide-protein conjugates were given the following designations:

O-SP-BSA₂

DeA-LPS-BSA O-SP-CW

DeA-LPS-CW

O-SP-rEPA

DeA-LPS-rEPAi

DeA-LPS-rEPA₂ Mice were immunized with these conjugate compositions containing 2.5ug of polysaccharide, with booster injections, and the determination of antibody levels and bactericidal antibody titers in mice were determined. Geometric mean antibody level (ELISA units) and immunoglobulin class composition of LPS antibodies elicited by E. coli 0157-rEPA conjugates in mice are shown in Table 1.

TABLE 1.

Immunoglobulin class composition of LPS antibodies elicited by E. coli 0157-rEPA conjugates in mice

Immunogen Geometic mean antibody level (ELISA units) (25^th-75^th centiles) After 1^st injection After 2^nd injection After 3^rd injection

IgG

O-SP-rEPA 0.08 (0.05-0.10) 2.50* (1.06-4.79) 6.26** (3.37-9.6)

DeA-LPS-rEPA, 0.07 (0.04-0.13) 1.37* (0.50-2.63) 4.49*** (1.49-16.4)

DeA-LPS-rEPA₂ 0.07 (0.06-0.07) 0.66* (0.07-3.73) 5.10** (2.23-10.0)

IgM

O-SP-rEPA 0.53 (0.36-0.72) 0.51 (0.31-1.12) 0.38 (0.22-0.59)

DeA-LPS-rEPA, 0.11 (0.04-0.34) 0.32 (0.08-0.89) 0.94 (0.28-2.94)

DeA-LPS-rEPA₂ 0.09 (0.06-0.11) 0.32 (0.06-1.53) 0.28 (0.21-0.45) a. IgG and IgM components of the hyperimmune 0157 sera (see Materials and Methods) were used as standards and assigned a value of 100 ELISA U each. Injection of O-SP, DeA-LPS, or saline did not elicit detectable antibodies.

*, O.Ol when compared with the value for O-SP-rEPA after the first injection; **, >0.02 when compared with the value for the same immunogen after the second injection; ***, P<0.07 when compared with the value for the same immunogen after the second injection. Bactericidal activity of serum LPS antibodies elicited in mice by immunization with heat-killed E. coli O157:H7 or O-specific polysaccharide-protein conjugates are shown in Table 2 below:

TABLE 2.

Bactericidal activity of serum LPS antibodies elicited in mice by immunization with heat-killed E.coli O157:H7 or O-specific polysaccharide-protein conjugates

Vaccine Antibody titer (ELISA units) Reciprocal bacterial

Total IgG IgM titer^b

Expt l

O-SP-CW 79.25 100

DeA-LPS-CW 15.1 >100

DeA-LPS-CW 19.4 80

E.co/ O157:H7 100.0 35

Expt 2

DeA-LPS-rEPA 18.8 0.07 320

DeA-LPS-rEPA 56.8 0.33 640

DeA-LPS-rEPA 32.8 0.45 640

O-SP-rEPA 18.6 0.44 640

O-SP-rEPA 15.8 0.59 640 ^a E. coli 0157:H7 is pooled hyperimmune sera from mice injected with heat-killed E. coli 0157. All other sera were from individual mice taken after the third conjugate injection. Serum dilutions were mixed with an equal volume of ~10³ E. coli 0157:H7 organisms per ml and complement. ^b The reciprocal bactericidal titer is expressed as the highest serum dilution yielding 50% killing. Absorption with LPS or DeA-LPS removed all of the bactericidal activity from sera from conjugate-injected mice and 90% from the hyperimmune sera prepared by injection of heat-killed E. coli 0157.

Example 2 Coniugation of E. coli 0157 O-SP with rEPA: Preparation of Vaccine Compositions

As discussed above, O-SP of E. coli 0157, prepared by acetic acid hydrolysis, and DeA-LPS 0157, prepared by hydrazinolysis, have been previously described. Conjugates of these polysaccharides to rEPA (O-SP 0157-rEPA, DeA-LPS 0157-_!, and DeA-LPS O157-rEPA₂) were prepared by the published procedure [8]. These conjugates were approved for investigation by the NTH (OH94-CH-N040), the FDA (BB-IND-5528) and the Institutional Review Board, Carolinas Medical Center, Charlotte, NC (08-94-08B). Pyrogen, sterility and safety testing ofthe final containers were performed by the Center for Biologies Evaluation and Research, FDA. All three conjugates elicited serum IgG anti-LPS with bactericidal activity when injected by a clinically relevant scheme and dosage in mice[8]. Clinical protocol

Volunteers of either gender and any ethnic group between ages 18 and 44 years were recruited from the staff of Carolinas Health Care System and the city of Charlotte, NC. Exclusion criteria were: pregnancy or planned pregnancy in the next six months, positive stool culture for E. coli 0157 or a history of hemorrhagic colitis, chronic disease including HIV 1, hepatitis or inflammatory bowel disease, acute illness including diarrhea, taking controlled substances, hospitalization within the year, taking regular medications, participation in another research protocol during the next two months, abnormal liver function test or having received cholera vaccine [32, 28]. After giving Informed Consent, a medical history and physical examination were performed and blood was obtained for assay of HIV 1, hepatitis b surface antigen, pregnancy test, liver function tests (LFT), antibodies to E. coli 0157 LPS and P. aeruginosa exotoxin A (ETA) and a culture ofthe stool for E. coli 0157. Eighty-seven volunteers were determined healthy and randomized into 3 groups of 29 to receive a injection of 0.5 ml of one ofthe experimental vaccines containing 25 μg of O-SP. Injections were delivered intramuscularly into the deltoid muscle. The volunteers were observed for 30 minutes after vaccination. Temperature and local or systemic reactions were recorded at 6, 24, 48 and 72 hours following vaccination.

All volunteers returned at 1, 4 and 26 weeks following vaccination for a health history and reaction, and blood was drawn. LFTs were performed, total protein/albumin), total bilirubin/direct and indirect, alkaline phosphatase (AP), SGOT (AST), SGPT (ALT), and GGT at each visit. Volunteers who had abnormal LFT levels at one week had repeated LFT tests at subsequent visits. Serum was collected for LPS and ETA antibody assays. Stool cultures for E. coli 0157 were obtained prior to and 4 and 26 weeks following vaccination. E. coli 0157 LPS and P. aeruginosa exotoxin A (ETA) antibodies ofthe volunteers were determined by ELISA [8]. Statistical methods

Antibody levels are expressed as geometric means (GM). Levels below the sensitivity of ELISA were assigned the value of one-half of that level. Comparison of GM was performed with either the two-sided t-test, paired t-test or the Wilcoxon test where appropriate. Results - clinical responses

One volunteer reported 3-6 cm diameter of erythema at the injection site within 24 hours following vaccination; one reported 1-3 cm and one reported >6 cm. Four volunteers reported erythema and induration after 72 hours observation: one (1-3 cm), two (3-6 cm) and one (>6 cm); all erythemas resolved by the 17th day.

Six volunteers (6.9%) had asymptomatic elevations (up to 35% above the normal range) of one or more serum LFT following vaccination. Four of these 6 volunteers had mild elevation of LDH and/or AP that returned to normal at 4-5 weeks. One volunteer had a serum bilirubin of 2.2 mg/dl (normal 1.5 mg/dl) with indirect bilirubin of 1.9 mg/dl at four weeks, and normal values at 14 weeks. Another volunteer had ALT (SGPT) and GGT evaluations of 33% and 26% respectively at four weeks, and elevations 13% and 47% respectively at 24 weeks following vaccination.

Ninety percent of volunteers reported oral temperatures less than 37.2° C at different observation times post- vaccination. The remainder ofthe volunteers reported oral temperatures 37.2-38° C with symptoms of acute respiratory infections.

There was no significant correlation between the reported post- vaccination observations and the lots of vaccine administered and no volunteer was hospitalized during the study.

One recipient of DeA-LPS 0157-rEPAι had a stool culture positive for E. coli 0157 at the 1 week post- vaccination visit. This volunteer had no adverse reactions following vaccination and no complaints throughout the study, and subsequent stool cultures were negative for E. coli 0157. Results - antibody levels (Tables 3 a and 3b

IgG: Pre-vaccination GM IgG anti-LPS levels in the three groups were low and similar. One week after vaccination, 71/87 (82%) responded with a ≥4-fold rise. Four weeks after vaccination, there were further rises in GM levels in all three groups (pθ.0001): all vaccinees responded with a >4-fold rise over the 1 week level. The GM levels in recipients of O-SP-rEPA were slightly higher than in those injected with either ofthe two DeA-LPS-rEPA conjugates (61.9 vs. 46.3 NS, 61.9 vs. 36.3, p<0.05). At 26 weeks, the GM levels ofthe 3 groups were similar (32.8, 31.2, 33.1, NS). Although the decline from the four week level was significant for each group (p<0.05), the levels at 26 weeks were higher than those at one week following vaccination in all three groups (32.8, 31.2, 33.1 vs. 7.93, 5.73, 4.12, pO.Ol); and 97% of volunteers had ≥lO-fold higher levels at 26 weeks than their pre-injection levels. Within the 25-75 percentile range, geometric mean titers were increased 68-fold to 132-fold after 4 weeks, and the overall result for the three conjugates at 4 weeks was a 93-fold increase in geometric mean titer. At 26 weeks, the results were increases of 61 -fold to 70-fold, and 64-fold increase overall for all conjugates. The volunteer who had a stool culture positive for E. coli 0157 at 1 week had IgG anti-LPS levels at pre-immunization, 1-, 4-, and 26-week post-immunization of 0.81, 1.15, 7.73 and 7.01 respectively, that are lower than the GM of all 3 groups.

IgM: Each conjugate elicited a significant rise in IgM anti-LPS at the 4 and 26 weeks intervals (p<0.0001). O-SP-rEPA elicited the highest level at each post vaccination interval but the difference was significant only at 4 weeks (32.8 vs. 18.1,19.1, p<0.05). At the 4 week interval, there was a >4-fold rise in 61/87 (70%) and in 34/86 (39.5%) at 26 weeks compared to pre-vaccination levels. There was a significant decrease in serum IgM anti-LPS at 26 weeks in all ofthe three groups (p<0.02) but there were no significant differences in the GM levels among the three groups. The volunteer who had a stool culture positive for E. coli 0157 at 1 week had a pre-immunization anti-LPS IgM level which was relatively high (11.9). The IgM levels declined 1, 4 and 26 weeks post-immunization (7.04, 10.6 and 5.94 units, respectively). These levels are lower than the GM ofthe three groups. IgA: Pre-vaccination levels of IgA anti-LPS were low. Similar to IgG and IgM anti-LPS, about 90% ofthe volunteers responded with >4-fold rise in IgA anti- LPS at one week, and 99% at four weeks (p<0.001). IgA anti-LPS GM levels declined to about 70% ofthe levels at the 4 week interval.

Table 3 a.

Geometric mean titers of serum IgG, IgM, and IgA lipopolysaccharide (LPS) antibodies elicited in volunteers by injection of E. coli 0157 O-SP-rEPA conjugates.

ELISA units (25^th - 75^th percentiles)

Conjugate Preimmune 1 Week 4 Weeks 26 Weeks

IgG

O-SP-rEPA 0.47 (0.3-0.7) 7.93 (2.8-24) 61.9 (40-91) 32.8 (23-50)

DeA-LPS-rEPA, 0.51 (0.3-0.9) 5.73 (1.8-22) 46.3 (22-84) 31.2 (12-61)

DeA-LPS-rEPA₂ 0.54 (0.3-0.9) 4.12 (2.2-6.0) 36.6 (20-76) 33.1 (15-57)

IgM

O-SP-rEPA 8.10 (4.0-14) 32.8 (23.51) 64.7 (47-109) 28.6 (17-44)

DeA-LPS-rEPA, 7.19 (3.1-12) 19.1 (9.2-29) 43.5 (13-56) 22.5 (11-34)

DeA-LPS-rEPA₂ 7.41 (4.6-13) 18.1 (10-27) 42.7 (26-73) 25.3 (17-35)

IgA

O-SP-rEPA 0.06 (0.0-0.1) 0.98 (0.5-2.4) 1.73 (1.0-2.5) 1.17 (0.9-2.1)

DeA-LPS-rEPA, 0.06 (0.0-0.1) 0.58 (0.3-0.8) 1.26 (0.6-3.7) 1.01 (0.5-1.9)

DeA-LPS-rEPA₂ 0.07 (0.0-0.1) 0.90 (0.4-1.8) 2.13 (1.2-4.9) 1.40 (1.0-2.5)

NOTE: High-titered postvaccination sera were used as standards. IgG, IgM, and IgA were assigned value of 100 ELISA units. Each group had 29 volunteers.

Table 3b.

Fold increases in geometric mean titers of serum IgG, IgM, and IgA lipopolysaccharide (LPS) antibodies elicited in volunteers.

Ab class Conjugate -fold increase in 25^th - 75^th percentiles l Week 4 Weeks 26 Weeks

IgG O-SP-rEPA 17 132 70

DeA-LPS-rEPA, 11 91 61

DeA-LPS-rEPA₂ 7.6 68 61

Geometric mean 11 93 64

IgM O-SP-rEPA 4.0 8.0 3.5

DeA-LPS-rEPA, 2.7 6.0 3.1

DeA-LPS-rEPA₂ 2.4 5.8 3.4

Geometric Mean 3.0 6.5 3.3

IgA O-SP-rEPA 16 29 20

DeA-LPS-rEPA, 9.7 21 17

DeA-LPS-rEPA₂ 13 30 20

Geometric Mean 13 26 19

NOTE: High-titered postvaccination sera were used as standards. IgG, IgM, and IgA were assigned value of 100 ELISA units. Each group had 29 volunteers.

Results - serum bactericidal activity (Table 4

Serum from high-responding volunteers (above the 75th percentile) was diluted serially and the diluted samples were analyzed for their ability to kill E. coli O157:H7. Pre-vaccination sera had no detectable bactericidal activity against E. coli O157:H7. The three conjugates elicited serum bactericidal activity that roughly correlated with the serum IgG and IgM anti-LPS antibody levels.

The results in Table 4 are those for serum from high-responding volunteers. Typically, the bactericidal titer (reciprocal dilution) for 50% killing ranged from about 2400 to about 32000.

Table 4.

Bactericidal activity (reciprocal titer) of serum anti-lipopolysaccharide (LPS) antibodies elicited in volunteers by injection of E. coli 0157 O-SP-rEPA conjugates.

Expressed as reciprocal of highest serum dilution yielding 50% killing.

Results - serum anti-P. aeruginosa exotoxin A (Table 5)

Most volunteers had low or non-detectable ETA antibodies in their pre- vaccination sera. All three conjugates elicited significant increases in GM IgG anti-ETA at the 1-week (p<0.002) and 4-week (p<0.001) intervals. At 26 weeks, the GM levels declined to those observed one week after vaccination. There were no statistically significant differences in the GM IgG anti-ETA at each bleeding interval among the three groups.

Table 5

Serum antibodies to Pseudomonas aeruginosa exotoxin A (ETA) elicited by Escherichia coli 0157 O-specific polysaccharide-rEPA conjugates in volunteers

GM antibody level (ELISA Units*)

Conjugate n Preimmune 1 week 4 weeks 26 weeks

O-SP-rEPA 29 0.29 0.93 1.90 0.88

DeA-LPS-rEPAi 29 0.39 0.91 1.48 0.87

DeA-LPS-rEPA₂ 29 0.29 0.65 0.93 0.67

*A high titered volunteer serum was used as a standard and assigned a value of 100 ELISA Units.

Example 3 Coniugation of E. coli 0157 O-SP with STXB1 and Preparation of Vaccine Compositions

E. coli 0157 O-SP was prepared by treatment of LPS with acetic acid as previously described [8, 9]. The B-subunit of Shigella toxin I (StxBl) was synthesized by Vibrio cholerae strain 0395-N1 (pSBC32 containing the StxBl gene) and purified by affinity chromatography [20, 21]. SDS 7% PAGE of StxBl showed one major band at 9 kDa and a faint band with slightly higher molecular weight.

For conjugation, 0157 O-SP was bound to StxBl directly by treatment with l-cyano-4-dimethylaminopyridinium tetrafluoroborate (CDAP) or by carbodiimide mediated condensation with adipic acid hydrazide linker [29, 30]. For direct conjugation, CDAP (100 mg/ml in acetonitrile) was added to O-SP in saline (5 mg/ml) at 0.3/1 (wt/wt) at room temperature, pH 5.8 to 6. 60 μL of 0.2 M triethylamine (TEA) added to bring the pH to 7.0 for 2 minutes. An equal weight of StxBl was added to the CDAP treated O-SP and the pH maintained at 8.0 to 8.5 for 2 hours. The reaction mixture was passed through a 1.5x90 cm Sepharose 6B column in 0.2M NaCl, the void volume fractions collected, and designated as OSP-StxBl.

Conjugate using a linker, adipic acid dihydrazide (ADH) was prepared as described previously [8, 30]. Briefly after addition of TEA in the above procedure, an equal volume of 0.8 M ADH in 0.5 M NaHCO₃ was added and the pH maintained at 8.0 to 8.5 for 2 hours. The reaction mixture was dialyzed against saline overnight at 4 °C and passed through a 2.5x31 cm P10 column in water. The void volume fractions were pooled, freeze-dried, and designated as OSP-AH. OSP-AH (10 mg), dissolved in 2 ml of saline, was added to an equal weight of StxBl and the pH brought to 5.1. The reaction mixture was put on ice and l-ethyl-3-(3- dimethylaminopropyl)carbodiimide (EDC) was added to 0.05M and the pH maintained at 5.1 to 5.5 for 2 hours. The reaction mixture was passed through a 1.5x90 cm Sepharose 6B column in 0.2 M NaCl, the void volume fractions collected and designated as OSP-AH-StxBl. Double immunodiffusion and ELISA were performed as described [8].

Female general purpose mice (n=10/group) were injected subcutaneously with saline or one ofthe conjugates containing 2.5 μg saccharide on days 0, 14, and 28. The mice were exsanguinated 7 days after each injection. Pooled sera from hyperimmunized mice were used as reference and assigned 100 ELISA units for IgG and IgM respectively. Neutralization of Stxl and Stx2 toward HeLa cells was measured using HeLa (CCL-2) cell monolayers in 96-well flat-bottom microtiter plates [21]. Each well was seeded with l-6xl0⁴ cells in 0.1 ml. Monolayers were established by overnight incubation in 5% CO₂. Toxin neutralization was determined by incubating dilutions of mouse serum with Stx-I or Stx-II at a final concentration of 100 pg/ml. The serum and toxin mixture was incubated at room temperature for 30 minutes and 0.1 ml was added to each well. Following incubation overnight, the surviving cells were determined spectro- photometrically using the crystal violet staining method of Gentry and Dalrymple[31]. Toxin neutralization was determined from a dose response curve of either Shiga toxin on each 96-well plate. Bactericidal activity was assayed as described [8, 10]. Results with 0157 O-SP — STXB1 conjugates

Derivatization of O-SP with adipic acid dihydrazide was 3.1% (wt/wt), similar to previous E. coli 0157 preparations [8]. The saccharide/protein ratios (wt/wt) were about 0.5 for both conjugates. The yields, based on saccharide in the conjugates, were 2.3% for OSP-StxBl and 3.4% for OSP-AH-StxBl. A single line of precipitation in double immunodiffusion was formed by rabbit anti-Stxl and mouse hyperimmune anti-0157 reacted against either conjugate.

After three injections, both conjugates elicited statistically significant rises of IgG and IgM anti-LPS (Table 6). The geometric mean (GM) anti-LPS level elicited by OSP-StxBl was 0.63 for IgG and 0.14 for IgM and for O-SP-AH-StxBl were 1.7 for IgG and 0.25 for IgM: the differences between two conjugates were not statistically significant.

Table 6.

Geometric mean IgG and IgM serum LPS antibody levels and neutralization titers against Shiga toxin 1 elicited in mice by conjugates of Escherichia coli 0157 O-SP with StxBl.

* Geometric mean of sera from 10 mice. Expressed in ELISA units using pooled hyperimmune mouse sera as reference (100 units for IgG and IgM respectively). f Geometric mean (n=10) neutralization titer determined with Stxl and HeLa cells, No neutralization at 1/100 dilution.

Sera from mice injected with saline or human sera from volunteers injected with E. coli 0157 O-SP-rEPA conjugates showed no neutralization to Stxl or to Stx2. Sera from mice injected 3 times with either ofthe 0157 O-SP — StxBl conjugates showed complete neutralization of Stxl at 1/100 dilution. At 1/1,000 dilution, the GM of neutralization titer was 90% for OSP-AH-StxBl and 98% for OSP-StxBl. At 1/10,000 dilution, the sera from mice injected with OSP-StxBl had a significantly higher neutralization titer (70%) than the sera elicited by O-SP-AH StxBl (34%o). None ofthe sera from mice injected with either conjugate showed neutralization against Stx2. Both conjugates elicited levels of bactericidal antibodies against E. coli 0157 that were roughly proportional to the content of IgG anti-LPS; this activity was removed by absorption with E. coli 0157 LPS.

DISCUSSION

The O-SP of E. coli 0157 LPS is a linear copolymer composed ofthe tetrasaccharide repeat unit: (^~ 3)-α-£⁾-Gak7NAc-(l→2)-α-D-Per )NAc-(l→3)-α- Z-Fucp-(l- 4)-β-Z -Glcp-(l->). It is non-immunogenic, probably due to its comparatively low molecular weight. As with other polysaccharides, its immunogenicity is increased by binding it to proteins to form a conjugate. Ofthe three conjugates ofthe present invention shown in Table 1, none elicited fever or significant local reactions in human volunteers, and all volunteers responded with a >4-fold rise in serum IgG anti-E. coli 0157 LPS that was sustained 26 weeks after injection. (Re- injection ofthe E. coli 0157 O-SP conjugates was not attempted because ofthe failure of other polysaccharide conjugates to induce a booster response in adults.)

These volunteers, like most adults, had low levels of "natural" serum anti E. coli 0157 LPS probably induced by cross-reacting antigens [32, 33, 34, 35]. This is typical for other bacterial pathogens as well. Higher levels of anti-0157 LPS antibodies are found in patients with HUS, and in individuals involved in raising cattle in certain areas, probably as a result of previous contact with these organisms. Although the unusual monosaccharide perosamine is found in the O-SP of both E. coli 0157 and V. cholerae Ol, we have not been able to detect a cross-reaction between human antibodies to these two antigens. The conjugate prepared from the O-SP obtained by acetic acid hydrolysis (O-SP-rEPA) elicited significantly higher levels of anti-0157 LPS at four weeks than did conjugates prepared with hydrazine-treated LPS. The LPS and ETA antibody levels, however, at 26 weeks post-injection were similar in all three groups (Table 1). As reported for patients with shigellosis and for adults vaccinated with Shigella conjugates, serum IgG anti-LPS rose to the highest level and was the most sustained ofthe three serum immunoglobulins [13, 15, 34, 36, 37]. Similar results were obtained in mice for the IgG anti-LPS responses elicited by E. coli Ol 11 conjugates [38].

The protective action of existing vaccines may be due to the induction of a critical level of specific IgG antibodies that, in many cases, inactivate the inoculum of the pathogen on epithelial surfaces including the intestine [39, 40]. It is not commonly appreciated that serum IgG is a major immunoglobulin component of secretory fluids including that ofthe small intestine. As has been observed in mice [8], all three conjugates induced IgG anti-LPS with bactericidal activity in the volunteers (Table 2). Serum IgG anti-polysaccharide is the major, if not the sole host component, that confers immunity induced by these conjugates. Accordingly, it should be possible to standardize the potency of E. coli 0157 conjugates by chemical assay and by measurement of serum IgG anti-polysaccharide as has been done for Haemophilus influenzae type b conjugate vaccines.

The 1995 outbreak of E. coli 0157 infection in Japan lasted several months, partly due to the failure to identify the bacterial sources [41]. Most ofthe volunteers (81%) responded with nearly a 10-fold increase in IgG anti-LPS 1 week after vaccination, indicating that the vaccine of this invention could serve to control E. coli 0157 infection during an outbreak. Another use for the E. coli 0157 conjugates of this invention would be to prepare high-titered IgG anti-LPS globulin for prophylaxis and treatment of case contacts during an outbreak. It has been suggested that antibiotic treatment of patients increases the incidence of HUS, possibly by causing lysis ofthe E. coli 0157 with release of additional Shiga toxins. Clinical and experimental data point to LPS as the pathogenic agent for HUS and the other extraintestinal lesions following infection with enteric Gram-negative pathogens [42, 43]. There is also a suggestion of a direct role of Shiga toxins on renal tissue involvement in HUS [44]. The present invention provides a solution to this problem in the form of a conjugate of E. coli 0157 O-SP with the B subunit of Shiga toxin 1. In mice, this conjugate induces both serum IgG anti-LPS and neutralizing antibodies to Shiga toxin 1.

The data show that the various E. coli 0157 LPS -protein conjugates disclosed herein will generate high antibody levels in humans (i.e., approximately 5- 10 times more IgG in humans than in mice) and high neutralization antibody titers in humans (i.e., 10³ to 10⁴ in humans as opposed to 10² in mice). The data also show that the various E. coli 0157 LPS-protein conjugates disclosed herein will generate a greater than 4-fold rise in IgG antibody levels in about 80% of human subjects one week after a single injection and in all human subjects 4 weeks after a single injection.

REFERENCES AND NOTES 1. For reviews, see:

(a) J. B. Robbins, R. Schneerson, S. Szu, V. Pozsgay, In: Vaccinia, vaccinations and vaccinology: Jenner, Pasteur and their successors (Ed.: S. Plotkin, B. Fantini), Elsevier, Paris, p. 135-143 (1996).

(b) R. K. Sood, A. Fattom, V. Pavliak, R. B. Naso, Drug Discovery Today , 1, 381-387 (1996).

(c) H. J. Jennings, R. K. Sood, In Neoglycoconjugates. Preparation and Applications (Eds. Y. C. Lee, R. T. Lee), Academic Press, New York, pp. 325-371 (1994).

2. K. Landsteiner, The specificity of serological reactions, Harvard University Press, Cambridge, (1970).

3. R. Schneerson, O. Barrera, A. Sutton, J.B. Robbins, J. Exp. Med. 1980, 152, 361- 376.

4. J.B. Robbins, R. Schneerson, P. Anderson, D.H. Smith, J. Am Med. Assoc. 1996, 276, 1181-1185.

5. For example:

(a) Cohen, D., et al, Lancet, 349, 155-0159 (1997).

(b) Cohen, D., et al, Infect. Immun., 64, 4074-4077(1997).

6. Riley. L.W., et al, N. Engl. J. Med., 308, 681-685 (1983).

7. Takeda, Y., World Health Statistics Quarterly, 50, 74-80 (1997)

8. Konadu et al., Infection & Immunity, 62, 5048-5054 (1994)

9. Robbins, J.B., et al, Clin. Infect. Dis., 15, 346-361 (1992)

10. Konadu et al., Journal of Infectious Diseases, 111 383-387 (1998)

11. Butler, T., Islam, M.R., Azad, M.A.K., Jones, P.K., J. Pediatr., 110, 894-897 (1987)

12. Proulx, F., et al, J. Pediatr., Ill, 299-303 (1992).

13. Cohen, D, C. Block, M.S. Green, G. Lowell, and I. Ofek, J. Clin. Microbiol, 27, 162-167 (1989).

14. Cohen, D., M.S. Green, C. Block, T. Roauch, and I. Ofek, J. Infect. Dis., 157, 1068-1071 (1988). 15. Robbins, J.B., and R. Schneerson, J. Infect. Dis., 161, 821-832 (1990).

16. Taylor, D.N., et al, Infect. Immun., 61, 3678-3687 (1993).

17. Cohen,D., S. Ashkenazi, et al, Lancet, 349, 155-159 (1997).

18. Cryz, S. J., et al, J. Infect. Dis., 163, 1040-1045 91991).

19. Weinstein, D.L., Jackson, M.P., Perera, L.P., Holmes, R.K., O'Brien, A.D., Infect. Immun., 57, 3743-3750 (1989)

20. Acheson, et al, Infect. Immun. , 61, 1098-1104 (1993).

21. Pozsgay, V., Trinh, L., Shiloach, J., Robbins, J.B., Donohue-Rolfe A, Calderwood SB., Bioconjugate. Chem., 7, 45-55 (1996).

22. Fattom, A., C. Lue, S.C. Szu, J. Mestecky, G. Schiffman, D. A. Bryla, W.F. Vann, D. Watson, L.M. Kimzey, J.B. Robbins, and R. Schneerson, Infect. Immun. , 55, 2309-2312 (1990).

23. Devi, S.J., J.B. Robbins and R. Schneerson., Proc. Natl. Acad. Sci. USA 88:7175- 7179, 1991 (1992).

24. Szu, S.C, X. Li, A.L. Stone, and J.B. Robbins, Infect. Immun. 59 4555- 4561(1991).

25. Szu, S.C, A.L. Stone, J.D. Robbins, R. Schneerson, and J.B. Robbins, J. Exp. Med. 166 1510-1524 (1987).

26. C. Chu, et al, Infect. Immun., 59, 4450-4458 (1991).

27. Kohn, J., Wilchek, M., FEBS Letters, 154, 209 (1983).

28. Aleksic, S., Karch, H., Bockemϋhl, J.,. Int. J. Med. Microbiol, 276, 221-230 (1992).

29. Lees, A., Nelson, B., Mond, J.J., Vaccine., 14, 190-198 (1995).

30. Konadu, E., Shiloach, J., Bryla, D.A., Robbins, J.B., Szu, S.C, Infect. Immun., 64, 2709-2715 (1996).

31. Gentry M., Dalrymple J.M., J. Clin. Micro. , 12, 361-366 (1980).

32. Chart, H, Rowe, B., Lancet, 341, 1282 (1993). 33. Robbins J.B., Schneerson R., J Infect Dis., 161, 821-832 (1990).

34. Greatorex J.S., Thorni G.M., J Clin Microbiol, 32, 1172-1178 (1994).

35. Cohen, D., et al, Infect Immun., 64, 4074-4077 (1996).

36. Ekwall E, et al, Serodiag. Immunother. Infect. Dis., 2, 171-182 (1988).

37. Cohen D., et al, Infect Immun., 64, 4074-4077 (1996).

38. Gupta, R.K., Egan W, Bryla DA, Robbins JB, Szu SC, Infect. Immun., 63, 2805- 2810 (1995).

39. Farmer, J.J., et al, JClin Microbiol, 21, 46-76 (1985).

40. Chart, H., Scotland, S.M., Rowe, B., J Clin Microbiol, 27, 285-290 (1989).

41. Watanabe, H., Wada, A., Inagak, Y., Tamura, K., Lancet, 348, 831-832 (1996).

42. Koster, .F, et al, N. Engl J. Med, 298, 927-933 (1978).

43. Jalkanen, K.S., et al, Lancet, i, 685-688 (1990).

44. Pickering, L.K., Obrig, T.G., Stapleton, F.B., Pediatr. Infect. Di.s J, 13, 459-476 (1994).

All ofthe references referred to above are hereby incorporated by reference in their entirety.

Claims

CLAIMSWe claim:

1. A conjugate molecule comprising the E. coli 0157 O-specific polysaccharide, covalently bound to a carrier selected from the group consisting of: the B subunit of Shiga toxin 1, the B subunit of Shiga toxin 2, a non-toxic mutant Shiga toxin 1 holotoxin, and a non-toxic mutant Shiga toxin 2 holotoxin.

2. The conjugate molecule of claim 1 wherein the E. coli 0157 O-specific polysaccharide is covalently bound to the B subunit of Shiga toxin 1 by means of a dicarboxylic acid dihydrazide linker.

3. The conjugate molecule of claim 2 wherein the dicarboxylic acid dihydrazide is adipic acid dihydrazide.

4. The conjugate molecule of claim 1 wherein the E. coli 0157 O-specific polysaccharide is covalently bound to the B subunit of Shiga toxin 1 by a process which comprises the steps of

(a) cyanation ofthe E. coli 0157 O-specific polysaccharide with a cyanylation reagent; and

(b) reaction ofthe B subunit of Shiga toxin 1 with the resulting cyanated E. coli 0157 O-specific polysaccharide.

5. The conjugate molecule of claim 4 wherein the cyanylation reagent is l-cyano-4- (N,N-dimethylamino)pyridinium tetrafluoroborate.

6. A pharmaceutical composition comprising a conjugate molecule of any one of claims 1-5, further comprising a pharmaceutically acceptable carrier.

7. The pharmaceutical composition of claim 6, further comprising an adjuvant.

8. The pharmaceutical composition of claim 6, wherein the composition is capable, upon injection into a mouse of an amount of said composition containing 2.5 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said mouse antibodies which neutralize the toxicity of Stxl toward HeLa cells.

. The pharmaceutical composition of claim 6, wherein the composition is capable, upon injection into a mouse of an amount of said composition containing 2.5 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said mouse antibodies which neutralize the toxicity of Stxl toward HeLa cells.

10. A vaccine composition comprising a conjugate molecule, said conjugate molecule comprising the E. coli 0157 O-specific polysaccharide covalently bound to a carrier protein, in a pharmaceutically acceptable carrier.

11. The vaccine composition of claim 10, wherein the carrier protein is selected from the group consisting of native or mutant forms of: tetanus toxoid, diptheria toxoid, pertussis toxoid, : the B subunit of Shiga toxin 1 , the B subunit of Shiga toxin 2, a non-toxic mutant Shiga toxin 1 holotoxin, a non-toxic mutant Shiga toxin 2 holotoxin, Clostridium perfringens toxoid, Clostridium welchii exotoxin C, Pseudomonas aeruginosa recombinant exoprotein A, hepatitis B surface antigen, hepatitis B core antigen, and bovine serum albumin.

12. The vaccine composition of claim 11, wherein the carrier protein is selected from the group consisting of Clostridium welchii exotoxin C, Pseudomonas aeruginosa recombinant exoprotein A, B subunit of Shiga toxin 1, and bovine serum albumin.

13. The vaccine composition of any one of claims 10 - 12, wherein the composition is capable, upon injection into a human of an amount of said composition containing 25 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said human bactericidal activity against E. coli 0157 such that the serum kills 50% or more of E. coli 0157 at a serum dilution of 1300:1 or more.

14. The vaccine composition of any one of claims 10 - 12, wherein the composition is capable, upon injection into a human of an amount of said composition containing 25 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said human bactericidal activity against E. coli 0157 such that the serum kills 50% or more of E. coli 0157 at a serum dilution of 32,000:1 or more.

15. The vaccine composition of any one of claims 10 - 12, wherein the composition is capable, upon injection into a human of an amount of said composition containing 25 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said human bactericidal activity against E. coli 0157 such that the serum kills 50% or more of E. coli 0157 at a serum dilution of 64,000:1 or more.

16. The vaccine composition of any one of claims 10 - 12, wherein the composition is capable, upon injection into a human of an amount of said composition containing 25 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said human at least a 50-fold rise in IgG which immunoreacts with E. coli 0157 LPS, when said IgG is measured 4 weeks post injection.

17. The vaccine composition of any one of claims 10 - 12, wherein the composition is capable, upon injection into a human of an amount of said composition containing 25 ╬╝g of E. coli 0157 O-specific polysaccharide, of inducing in the serum of said human at least a 60-fold rise in IgG which immunoreacts with E. coli 0157 LPS, when said IgG is measured 26 weeks post injection.

18. The vaccine composition of any one of claims 10 - 12, further comprising an adjuvant.

19. A method of inducing in a mammal serum antibodies that are bacteriostatic or bactericidal to E. coli 0157, comprising administering to said mammal, in a physiologically acceptable carrier, a conjugate molecule of any one of claims 1-5.

20. The method of claim 18 wherein said conjugate molecule is administered at a dose of about 5 micrograms to about 50 micrograms of E. coli 0157 O-specific polysaccharide.

21. The method of claim 18 wherein the antibodies protect the mammal against infection by E. coli 0157.

22. A composition comprising antibodies which are immunoreactive with E. coli 0157 O-specific polysaccharide.

23. The composition of claim 22, further comprising antibodies which are immunoreactive with the B subunit of Shiga toxin 1.

24. The composition of claim 22, wherein the composition is chosen from the group consisting of mammalian plasma, mammalian serum, and mammalian gamma globulin fraction.

25. The composition of claim 23, wherein the composition is chosen from the group consisting of mammalian plasma, mammalian serum, and mammalian immunoglobulin fraction.

26. An antibody which is immunoreactive with E. coli 0157 O-specific polysaccharide.

27. A method of passively immunizing a mammal against E. coli 0157, comprising administering to said mammal an immunologically sufficient amount of a composition according to any one of claims 22 - 25.

28. The method of claim 27 wherein the antibody is administered at a dose in the range of from about 1 mg/kg to about 10 mg/kg body weight ofthe mammal.

29. The method of claim 28 wherein the mammal is a human.

30. A method for vaccinating a mammal against E. coli 0157 infection, comprising administering to the human an immunizing amount of a composition according to claim 6.

31. The method of claim 30 wherein the mammal is a human.

32. A method for vaccinating a mammal against E. coli 0157 infection, comprising administering to the human an immunizing amount of a vaccine composition according to any one of claims 10 - 12.

33. The method of claim32 wherein the mammal is a human.

34. A conjugate molecule comprising an O-specific polysaccharide, covalently bound to the B subunit of Shiga toxin 1 or Shiga toxin 2, or to a non-toxic mutant Shiga holotoxin, wherein the O-specific polysaccharide is an O-specific polysaccharide of a bacterium chosen from the group consisting of: E. coli 0157, E. coli Ol 11, E. coli 017, E. coli 026, and Shigella dyenteriae.

35. The conjugate molecule of claim 34 wherein the O-specific polysaccharide is covalently bound to the B subunit of Shiga toxin 1 by means of a dicarboxylic acid dihydrazide linker.

36. The conjugate molecule of claim 35 wherein the dicarboxylic acid dihydrazide is adipic acid dihydrazide.

37. The conjugate molecule of claim 36 wherein the O-specific polysaccharide is covalently bound to the B subunit of Shiga toxin 1 by a process which comprises the steps of

(a) cyanation ofthe O-specific polysaccharide with a cyanylation reagent; and

(b) reaction ofthe B subunit of Shiga toxin 1 with the resulting cyanated O- specific polysaccharide.

38. The conjugate molecule of claim 37 wherein the cyanylation reagent is l-cyano-4- (N,N-dimethylamino)pyridinium tetrafluoroborate.

39. A pharmaceutical composition comprising a conjugate molecule of any one of claims 34-37 further comprising a pharmaceutically acceptable carrier.

40. A composition coprising antibodies which are immunoreactive with Shiga toxin 1 or Shiga toxin 2.

41. A method of administering a composition of claim 40 to a mammal in an immunologically sufficient amount.