WO2003106685A1 - Cell wall mutants for delivery of biologically active compounds - Google Patents

Abstract

The present invention relates to a recombinant microorganism comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and comprising an expression vector leading to the production of a polypeptide having a prophylactic or therapeutic activity.The invention further relates to a method for the delivery of polypeptides at mucosal surfaces, comprising administering to a mucosal surface of an individual (human or animal) a composition comprising a recombinant microorganism comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein the polypeptide to be delivered is produced by said recombinant microorganism.

Description

CELL WALL MUTANTS FOR DELIVERY OF BIOLOGICALLY ACTIVE COMPOUNDS

Field of the invention

The present invention relates to the field of recombinant microorganisms for delivery of compounds at mueosal surfaces.

Background to the invention

The delivery at mueosal surfaces is the potentially most attractive route for administration of biologically active compounds as it involves non-invasive techniques and enjoys a high patient compliance.

The development of effective strategies for the delivery of vaccine antigens to the mueosal tissues has received considerable attention over the past decade (reviewed in Michalek et al., 1994; Wells & Pozzi 1997; O'Hagan 1994). The main advantage of this route of administration is that it has the potential to elicit local immune responses leading to the production of antigen-specific secretory immunoglobulin A (S-lgA) as well as systemic immune responses.

Moreover, delivery systems that direct cytokines, enzymes and other biologically active molecules to the mueosal tissues are also needed to elicit potent local effects and avoid any deleterious effects of systemic administration.

WO 9714806 relates to the use of non-invasive gram-positive bacteria such as Lactococcus lactis for the delivery of biologically active polypeptides in the body.

WO 9709437 describes methods for obtaining surface expression of a desired protein or polypeptide in gram-positive bacteria such as L lactis. It also relates to the possible immobilization of the gram-positive host microorganisms on a solid phase by displaying on their cell surface molecules such as streptavidin.

The immune system can be divided into two functionally independent compartments: (i) the systemic, which comprises the bone marrow, spleen and lymph nodes, and (ii) the mueosal, which comprises lymphoid tissue associated with mueosal surfaces and external secretory glands. Mueosal surfaces are associated with the gastrointestinal, genitourinary and respiratory tracts. Each compartment is associated with both humoral (antibodies) and cell-mediated (for example, cytotoxic T-cells) responses, however there are differences in the nature of the immune responses induced in each compartment. Antibodies associated with the systemic compartment are predominantly of the IgG isotype, which function to neutralize pathogens in the circulatory system. In contrast, the production of secretory immunoglobulins A (S-lgA) characterizes mueosal immune responses and is recognized to be a key factor for preventing entry of pathogens at the mueosal surfaces which are the major sites at which microbial infections are initiated. A consequence of this compartmentalization is that systemic routes of immunization are usually of limited value for the prevention of mucosa-contracted diseases. Indeed successful systemic immunization (i.e., delivery of antigen to the systemic compartment) will induce systemic immunity but does not usually yield potent mueosal immune responses. In contrast, antigens delivered at mueosal surfaces can trigger both mueosal (at the immunization site and also at distant sites) and systemic responses.

Most vaccines developed to date are delivered parentally, for example by intramuscular or intradermal injection, and as such induce primarily systemic immunity. However, the majority of infectious disease is acquired via mueosal surfaces, as they constitute the major entry site of most pathogenic microorganisms.

The few vaccine strategies available which are able to induce protective responses at mueosal surfaces have been focused mainly on the use of synthetic (non-living) delivery systems or of live vaccine vehicles comprising attenuated pathogenic bacteria such as Salmonella, Bortedella, and Mycobacterium bovis BCG. This causes a real problem when it comes to the vaccination of less immunocompetent individuals such as elderly and children, as such vaccines can be dangerous as they can revert to pathogenic organisms. In addition, such vaccines can be of limited efficacy because of pre-existing antibodies or strong reactions against the bacterial vector itself. Vaccines based on killed micro- organisms avoid the safety problems associated with live vaccines; however such vaccines often fail to elicit an appropriate effective immune response in targeted individuals.

There is a new focus on the development of non-pathogenic bacteria, such as lactic acid bacteria (LAB), which could be useful as vaccines. These bacteria have the advantage of being well known non-pathogenic species which are commonly used as starters for the preparation of fermented food and feed products. In addition, certain strains or species of LAB belong to the natural endogenous microflora of individuals where they play a critical role in maintaining a balanced ecosystem. WO 0023471 describes the construction of recombinant Lactococcus lactis strains producing IL-10 and their use in anti-inflammatory treatment.

EP 1084709 relates to oral vaccines comprising lactic acid bacteria as specific component for eliciting immunogenicity against a heterologous antigen. Preferably, Lactobacillus plantarum strains are used. Said strains comprise an expression vector capable of expressing the heterologous antigen intracellularly and/or such that the heterologous antigen is exposed on the cell surface of the bacterium.

In addition to vaccines, many gene and drug therapies require efficient and specific delivery vehicles to ensure the greatest possible benefit. Lack of an adequate delivery vehicle is a major roadblock to the application of gene therapy and significantly limits the therapeutic potential of many drugs which exert limited effects when administered through systemic routes or/and have a very short live time when given by mueosal routes.

It is a main object to provide new systems for delivering compounds at mueosal surfaces. It is a further object to provide recombinant microorganisms for use in the preparation of medicaments.

Summary of the invention

According to a first embodiment the invention relates to a recombinant gram-positive bacterium for use as a medicament wherein said microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein said recombinant microorganism is producing intracellularly a polypeptide for prophylactic or therapeutic applications.

The invention further relates to a method for the delivery of polypeptides at mueosal surfaces using said recombinant microorganism, comprising administering to a mueosal surface of an individual a composition comprising said recombinant microorganism having a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein the polypeptide to be delivered is produced by said recombinant microorganism. Furthermore the invention relates to the use of said recombinant for the manufacture of a medicament for preventing or treating for example infectious, inflammatory, auto-immune or allergic diseases, metabolic deficiencies or cancers.

Moreover the present invention relates to use of a mutant microorganism as a vehicle for delivery of compounds or polypeptides at mueosal surfaces characterized in that said mutant microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component.

Detailed description

According to a first embodiment, the present invention relates to a gram positive recombinant microorganism capable of producing a polypeptide having a prophylactic or therapeutic activity, and bearing a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component.

The expression "recombinant microorganism" used herein, relates to a microorganism comprising a vector or an expression vector that has been modified by the introduction of a heterologous nucleic acid or a transgenic strain carrying the heterologous nucleic acid integrated in its chromosome. Thus, for example recombinant microorganisms express genes that are not found in identical form or mode of expression within the native (non- recombinant) microorganism.

It should be understood that the heterologous nucleic acids can be obtained from any natural source and/or can be prepared synthetically using well known DNA synthesis techniques.

The term "expression vector" as used herein relates to a vector wherein the heterologous nucleic acid sequence comprises at least one heterologous gene operably linked to one or more control sequences allowing the expression in prokaryotic host cells. Said nucleic acid sequence comprises coding sequences or open reading frames (ORF) as well as regulatory sequences or elements such as a promoter, an operator or a terminator. According to a preferred embodiment the microorganism according to the invention is a bacterial strain, preferably a non-pathogenic strain, preferably a non-invasive strain, preferably a food-grade strain, more preferably a gram-positive bacterial strain, most preferably a lactic acid bacterial strain. Examples of suitable bacterial strain include but are not limited to Lactobacillus, Lactococcus, Bifidobacterium and non-pathogenic staphylococci species. In a preferred embodiment, said microorganism is a Lactobacillus or a Lactococcus species. Other examples of suitable food grade strains can be found in Bergey's Manual of Systematic Bacteriology, Volume 2 (1986), Gram-positive Bacteria other than Actinomycetes, Williams & Wilkins, Baltimore, hereby incorporated by reference.

The use of recombinant lactic acid bacterial strain, preferably Lactobacillus or Lactococcus strains according to the invention offers considerable advantages. They can be commensal strains isolated from host-specific body cavities (humans or animals), environmental isolates or starter strains used for the production of fermented food and feed. It is becoming well established that given strains are able to persist transiently in body cavities. Moreover, some of these strains have been reported to exhibit probiotic proprieties (beneficial to the host's health) including immuno-adjuvant and immuno- modulating capacities, which vary according to the species or strains used.

The present invention is thus applicable to any of the Lactobacillus, Lactococcus or Streptococcus species or subspecies or strains selected from the list comprising Lactobacillus acetotolerans, Lactobacillus acidipiscis, Lactobacillus acidophilus, Lactobacillus agilis, Lactobacillus algidus, Lactobacillus alimentarius, Lactobacillus amylolyticus, Lactobacillus amylophilus, Lactobacillus amylovorus, Lactobacillus animalis, Lactobacillus arizonensis, Lactobacillus aviaries, Lactobacillus aviarius subsp. araffinosus, Lactobacillus aviarius subsp. aviarius, Lactobacillus batatas, Lactobacillus bavaricus, Lactobacillus bifermentans, Lactobacillus bifidus, Lactobacillus brevis, Lactobacillus buchneri, Lactobacillus bulgaricus, Lactobacillus carnis, Lactobacillus casei, Lactobacillus casei subsp. alactosus, Lactobacillus casei subsp. casei, Lactobacillus casei subsp. pseudoplantarum, Lactobacillus casei subsp. rhamnosus, Lactobacillus casei subsp. tolerans, Lactobacillus catenaformis, Lactobacillus cellobiosus, Lactobacillus coleohominis, Lactobacillus collinoides, Lactobacillus confusus, Lactobacillus coprophilus, Lactobacillus coprophilus subsp. confusus, Lactobacillus corynoides, Lactobacillus corynoides subsp. corynoides, Lactobacillus corynoides subsp. minor, Lactobacillus coryniformis, Lactobacillus coryniformis subsp. coryniformis, Lactobacillus coryniformis subsp. torquens, Lactobacillus crispatus, Lactobacillus curvatus, Lactobacillus curvatus subsp. curvatus, Lactobacillus curvatus subsp. melibiosus, Lactobacillus cypricasei, Lactobacillus delbrueckii, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus delbrueckii subsp. delbrueckii, Lactobacillus delbrueckii subsp. lactis, Lactobacillus desidiosus, Lactobacillus diolivorans, Lactobacillus divergens, Lactobacillus equi, Lactobacillus farciminis, Lactobacillus fermentum, Lactobacillus ferintoshensis, Lactobacillus fornicalis, Lactobacillus frigidus, Lactobacillus fructivorans, Lactobacillus fructosus, Lactobacillus frumenti, Lactobacillus gallinarum, Lactobacillus gasseri, Lactobacillus graminis, Lactobacillus halotolerans, Lactobacillus hamsteri, Lactobacillus helveticus, Lactobacillus heterohiochii, Lactobacillus hilgardii, Lactobacillus homohiochii, Lactobacillus hordniae, Lactobacillus iners, Lactobacillus intestinalis, Lactobacillus inulinus, Lactobacillus japonicus, Lactobacillus jensenii, Lactobacillus johnsonii, Lactobacillus jugurti, Lactobacillus kandleri, Lactobacillus kefiranofaciens, Lactobacillus kefirgranum, Lactobacillus kefiri, Lactobacillus kimchii, Lactobacillus kun/ eei, Lactobacillus lactis, Lactobacillus leichmannii, Lactobacillus lindneri, Lactobacillus malefermentans, Lactobacillus mail, Lactobacillus maltaromicus, Lactobacillus manihotivorans, Lactobacillus minor, Lactobacillus minutus, Lactobacillus mucosae, Lactobacillus murinus, Lactobacillus nagelii, Lactobacillus oris, Lactobacillus panis, Lactobacillus parabuchneri, Lactobacillus paracasei, Lactobacillus paracasei subsp. paracasei, Lactobacillus paracasei subsp. tolerans, Lactobacillus parakefiri, Lactobacillus paralimentarius, Lactobacillus paraplantarum, Lactobacillus pentosus, Lactobacillus perolens, Lactobacillus piscicola, Lactobacillus plantarum, Lactobacillus pontis, Lactobacillus psittaci, Lactobacillus reuteri, Lactobacillus rhamnosus, Lactobacillus rimae, Lactobacillus rogosae, Lactobacillus ruminis, Lactobacillus sakei, Lactobacillus sakei subsp. carnosus, Lactobacillus sakei subsp. sakei, Lactobacillus salivarius, Lactobacillus salivarius subsp. salicinius, Lactobacillus salivarius subsp. salivarius, Lactobacillus sanfranciscensis, Lactobacillus sharpeae, Lactobacillus suebicus, Lactobacillus trichodes, Lactobacillus uli, Lactobacillus vaccinostercus, Lactobacillus vaginalis, Lactobacillus vermiforme, Lactobacillus viridescens, Lactobacillus vitulinus, Lactobacillus xylosus, Lactobacillus yamanashiensis, Lactobacillus yamanashiensis subsp. mail, Lactobacillus yamanashiensis subsp. yamanashiensis, Lactobacillus zeae, Lactococcus garvieae, Lactococcus lactis subsp. lactis, Lactococcus lactis subsp. cremoris, Lactococcus lactis subsp. hordniae, Lactococcus piscium, Lactococcus plantarum, Lactococcus raffinolactis, and Streptococcus thermophilus. According to a more preferred embodiment the lactic acid bacterial strain is a Lactobacillus plantarum or Lactococcus lactis strain. L. plantarum NCIMB8826 has a very good resistance to the gastric barrier and a very good survival in the ileum in humans, and is able to persist for about a week in the mouse intestine after oral administration.

According to the invention, the recombinant microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component.

The term "a cell wall component" in the present invention relates to "at least one cell wall component" whereas it may happen that a mutation in a given gene has a pleitropic effect on the biosynthesis, modification or degradation of several distinct components of the cell wall.

The term "mutation" refers to a sequence variation within a gene such as substitution in which one or more nucleotides are substituted with (an)other nucleotide(s), deletion in which one or more of existing nucleotides are deleted, insertion in which one or more of additional nucleotides are inserted. Also, "mutation" refers to insertion into any site of the chromosome of a nucleic acid sequence originating from a different locus or from a different genome.

The term "a mutation" in the present invention relates to "at least one mutation" whereas it might be necessary to introduce several mutations in a given gene to have an effect on the biosynthesis, modification or degradation of a cell wall component. Furthermore, it might be necessary to have mutations in several genes to affect a single cell wall component.

The expression "a mutation modulating the expression" relates to a mutation which preferably has an effect on the production of the mature or functional cell wall component. Said effect can be the over-production, the non-production or prevention of production or delayed or decreased production of a cell wall component or of an enzyme or structural element involved in the biosynthesis, modification or degradation of a cell wall component. Said effect can also be the production of a non-functional or premature or incomplete polypeptide due to a missense or frame-shift mutation or the generation of a premature stop codon. It should be understood that all said mutations may result in a modification of the bacterial cell wall which may lead to a variety of phenotypes, including reduction in viability, especially under stress conditions, weakening of the cell wall, leakage of cytoplasmic content, auto/co- aggregation, a change in its porosity/permeability, a change in its global physical-chemical surface properties, such as change in hydrophobycity or electric charge.

The expression "expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component" relates to recombinant microorganisms which have been transformed with at least one foreign gene, and wherein said gene is capable of modulating the production of a cell wall component or of an enzyme or structural element involved in the biosynthesis, modification or degradation of a cell wall component. For example, said foreign gene can be responsible of the production, the over-production, the non-production or prevention of production or delayed or decreased production of an endogenous or non-endogenous polypeptide or protein interfering with the biosynthesis, modification or degradation of the microorganism cell wall. As used herein, "foreign gene" means a polynucleotide encoding a protein or fragment thereof or antisense RNA or catalytic RNA, which is foreign to the said recombinant microorganism.

The term "gene" relates to a nucleic acid sequence comprising coding sequences or open reading frames as well as regulatory sequences or elements such as a promoter, an operator, or a terminator. As such, mutations in regulatory sequences, for instance resulting in over-production, non-production or a decreased production of said polypeptide may also have an effect on the biosynthesis, modification or degradation of cell wall components.

As used herein the expression "coding sequence" or "ORF" is defined as a nucleotide sequence that can be transcribed into mRNA and translated into a polypeptide when placed under the control of appropriate control sequences or regulatory sequences. Said coding sequence or ORF is bounded by a 5' translation start codon and a 3' translation stop codon. A coding sequence or ORF can include, but is not limited to RNA, mRNA, cDNA, recombinant nucleotide sequences, synthetically manufactured nucleotide sequences or genomic DNA.

Reference herein to a "promoter" is to be taken in its broadest context and includes the transcriptional start signal and additional regulatory or control elements (i.e. operators).

More specifically, the term "promoter" includes the prokaryotic transcriptional start signal, in which case it may include a -35 box sequence and/or a -10 box sequence. Promoters may contain one or several copies of regulatory elements, to control expression of the associated transcription unit in response to intra or extra cellular signals or molecules.

The term "terminator" refers to a DNA sequence at the end of a transcriptional unit which signals termination of transcription.

The recombinant microorganism according to the present invention comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component, wherein the expression "cell wall component" generally refers to all structures and molecules found in a bacterial cell wall, such as peptidoglycans, teichoic acids, lipoteichoic acids, teichuronic acids, polysaccharides, and cell wall associated proteins.

In order for the recombinant microorganism to elicit a mueosal immune response, the said microorganism has to come in close contact with the cells from the body cavity mucosa which are responsible for triggering the mueosal immune response, and to deliver to them the said compounds or polypeptides. The compounds or the polypeptides produced in the cytoplasmic compartment of recombinant microorganism are confined in said microorganism by the bacterial cell wall. In order to allow an efficient delivery of said compounds or polypeptides produced in the cytoplasm of said microorganism, a cell wall mutant of the bacterium is used. Said recombinant microorganism comprising said cell wall mutation or expressing said foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component is therefore bearing an altered cell wall which may be subject to a premature lysis and which may result in a more efficient and more controlled release of the cellular compounds or polypeptides when administered to an organism in need thereof. The mutation may induce changes in the global cell wall properties such that the resulting mutant is able to interact more efficiently with the host immune system or the epithelial surface of the targeted body cavity, and/or to better persist in the targeted body cavity.

In a preferred embodiment the recombinant microorganism according to the invention comprises a mutation in a gene involved in the biosynthesis of peptidoglycans, teichoic acids, or lipoteichoic acids, more preferably the mutated gene is involved in the biosynthesis of D-alanine, D-glutamate or D-aspartate, or in the alanylation or glucosylation of teichoic or lipoteichoic acids. Examples of genes involved in the biosynthesis, modification or degradation of a cell wall component comprise the following gene products: alanine racemase, glutamate racemase, aspartate racemase, D-lactate dehydrogenase, lactate racemase, aspartate kinase, aspartate semialdehyde dehydrogenase, dihydrodipicolinate synthase, dihydrodipicolinate reductase, tetrahydrodipicolinate succinylase, N- succinyldiaminopimelate-aminotransferase, N-succinyl-L-diaminopimelic acid desuccinylase, diaminopimelate epimerase, diaminopimelate decarboxylase, N- acetylglucosamine-1 -phosphate uridyltransferase, UDP-N-acetylglucosamine enolpyruvoyltransferase, UDP-N-acetylmuramate dehydrogenase, UDP-N- acetylmuramate-alanine ligase, UDP-N-acetylmuramoylalanine D-glutamate ligase, UDP- N-acetylmuramoylalanyl-D-glutamate 2,6-diaminopimelate ligase, UDP-N- acetylmuramoylalanyl-D-glutamate lysine ligase, D-alanine-D-alanine ligase, D-alanine-D- lactate ligase, D-alanine-D-serine ligase, D-alanine-D-alanine adding enzyme, D-alanine- D-lactate adding enzyme, D-alanine-D-serine adding enzyme, undecaprenyl pyrophosphate synthetase, undecaprenylpyrophosphate phosphorylase, undecaprenyl- phosphate phospho-N-acetylmuramoylpentapeptide transferase, undecaprenyl-phospho- N-acetylmuramoylpentapeptide N-acetylglucosaminyl transferase, flippase, lipid II translocase, penicillin binding proteins, peptidoglycan transglycosylase, peptidoglycan transpeptidase, D,D-carboxypeptidase, L,D-carboxypeptidase, peptidoglycan endopeptidase, N-acetyl muramidase, N-acetyl glucosaminidase, N-acetyl-muramoyl L- alanine amidase, D-alanine-D-alanine dipeptidase, phosphoglucomutase, UTP-glucose-1- phosphate uridyltransferase, UDP-glucose 4-epimerase, UDP-glucose-hexose-1- phosphate-uridyltransferase, UTP-hexose-1-phosphate-uridyltransferase, galactokinase, aldose-1 epimerase, undecaprenylphosphate phospho-N-acetylglucosaminyltransferase, undecaprenyl-phospho-N-acetylglucosamine N-acetyl-D-mannosaminyltransferase, glycerol-3-phosphate cytidylyltransferase, undecaprenyl-phospho-N-acetylglucosaminyl- N-acetyl-D-mannosamine glycerophosphotransferase, poly(glycerolphosphate) glycerophosphotransferase, poly(glycerolphosphate) glucosyltransferase, poly(glycerolphosphate) galactosyltransferase, poly(glycerolphosphate) N-acetyl- glucosaminyltransferase, poly(glycerololphosphate) glycosyltransferase, UDP-N-acetyl-D- glucosamine 2-epimerase, poly(glycerolphosphate) translocase, ribitol-5-phosphate cytidylyltransferase, ribitol-5-phosphate dehydrogenase, (poly)ribitol-phosphate phosphoribitoltransferase, poly(ribitolphosphate) glucosyltransferase, poly(ribitolphosphate) glycosyltransferase, poly(ribitolphosphate) galactosyltransferase, poly(ribitolphosphate) N-acetyl-glucosaminyltransferase, poly(glycerolphosphate) D-alanyl transferase, D-alanine: D-alanyl carrier protein ligase "DltA", D-alanyl carrier protein "DltC", poly(ribitolphosphate) translocase, UDP-glucose dehydrogenase, UDP-N- acetylglucosamine-4-epimerase, undecaprenyl-phosphate phospho-N- acetylgalactosaminyltransferase, undecaprenyl-phospho-N-acetylgalactosamine glucuronyltransferase, teichuronic acid polymerase.

As such, according to a more preferred embodiment, the recombinant microorganism used in the present invention has a mutation in the gene encoding alanine racemase (air), or glutamate racemase, or aspartate racemase, or in the dlt cluster of genes responsible for D-alanylation of teichoic acids and lipoteichoic acids, or in the tagE gene responsible for glucosylation of teichoic acids. The present inventors clearly demonstrated that the use of a recombinant Air- mutant producing an antigen led to a very significant increase of specific immune responses after mueosal administration to mice when compared to the recombinant Alr+ strain. These results were corroborated by the successful use of a second type of cell wall mutant, i.e. the Lactococcus lactis DltD- strain.

According to an embodiment of the present invention, said recombinant microorganism produces a least one polypeptide for prophylactic or therapeutic application. Produced polypeptides can be derived from eukaryotic sources or prokaryotic sources, or from viruses.

The term "polypeptide", "peptide" or "protein" when used herein refers to amino acids in a polymeric form of any length. Said term also includes any naturally occurring post- translational modifications. Furthermore said term refers to biologically active polypeptides, wherein "biologically active" refers to ability to perform a biological function. Examples of such biologically active compounds are antigenic compounds, immunomodulators, nutritional compounds, medicaments, as well as analogues or derivatives thereof.

The various classes of biologically active polypeptides which can be delivered by the method of the invention include, but are not limited to any therapeutic or prophylactic compound selected from the group comprising antigens, epitopes, allergens, immune modulators including adjuvants, enzymes, receptor ligands or variants thereof. Other examples include anti-inflammatory compounds, analgesics, antiarthritics, antispasmodics, antidepressants, antipsychotics, tranquilizers, antianxiety compounds, narcotic antagonists, antiparkinsonism compounds, cholinergic agonists, chemotherapeutic compounds, immunosuppressive compounds, antiviral compounds, antibiotics, antimicrobial compounds, appetite suppressants, antiemetics, anticholinergics, antihistaminics, antimigraine compounds, coronary, cerebral or peripheral vasodilators, hormones, contraceptives, antithrombotic compounds, diuretics, haemostatics, hemolytics, antihypertensive compounds, cardiovascular compounds, allergoids, vitamins, antibodies, enzyme inhibitors, anti-receptors or receptor blockers, neuropeptides and the like. Suitable examples of polypeptides include but are not limited to the C subunit of tetanus toxin (TTFC) and the Urease B subunit (Urease B or UreB) from Helicobacter pylori.

Furthermore, the produced polypeptide may be fused to another peptide, polypeptide or protein to form a chimeric protein.

The polypeptide produced exerts a beneficial effect or activity on the health of the individual, in a prophylactic or therapeutic manner, wherein "prophylactic" means protective or preventive, and "therapeutic" means curative.

The invention further relates to the use of any recombinant microorganism herein described as a medicament.

According to an alternative embodiment the invention relates to a recombinant microorganism for use as a medicament wherein said recombinant microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein said recombinant microorganism is producing a polypeptide for prophylactic or therapeutic application.

The invention further relates to a method for the delivery of polypeptides at mueosal surfaces using said recombinant microorganism, comprising delivering to a mueosal surface of an individual an effective amount of a composition comprising said recombinant microorganism having a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein the polypeptide to be delivered is synthesized by said recombinant microorganism. The term "individual" as used herein refers to a human or an animal host. The individual will preferably be a human, but veterinary applications are also in the scope of the present invention targeting for example domestic livestock, laboratory or pet animals.

Mueosal surfaces include mueosal membranes such as buccal, gingival, nasal, tracheal, bronchial, gastrointestinal, rectal, urethral, ureteral, vaginal, cervical, uterine, etc. Administration of said composition to a mueosal surface comprises any route of administration to the body of a human or animal for which it is not required to puncture the skin (e.g. as with intravenous, intramuscular, subcutaneously or intraperitoneal administration). Usually, said composition is administered to the body via one of the body cavities, such that it comes into contact with the mucosa. Examples of routes to administer said composition include nasal, oral or intragastric, genital, rectal, aural (i. e. via the ear), ocular, sublingual, buccal, bronchial, inhalatory, and urethral routes.

The invention also relates to the use of a recombinant microorganism as defined previously for the manufacture of a medicament for prophylactic or therapeutic application. It may for example be used for vaccine application as well as for treating illness such as infectious diseases, allergy, chronic inflammation, auto-immune diseases, metabolic deficiencies and cancers.

According to yet another embodiment, the invention relates to the use of a mutant microorganism as a vehicle for delivery of compounds or polypeptides at mueosal surfaces characterized in that said mutant microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component. The compound or polypeptide to be delivered is chosen from a protein, polypeptide, or peptide.

In a preferred embodiment, the mutant microorganism is a gram-positive commensal bacterium, preferably a lactic acid bacterium, most preferably a Lactobacillus or Lactococcus strain chosen from Lactobacillus plantarum or Lactococcus lactis.

Preferably the mutant microorganism according to the invention, comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component wherein said cell wall component is preferably peptidoglycan, teichoic acid, or lipoteichoic acid. More preferably said mutation is in a gene involved in the biosynthesis of D-alanine, D-glutamate, or D-aspartate, or in the alanylation or glucosylation of teichoic or lipoteichoic acids. Yet more preferably said gene encodes alanine racemase, or aspartate racemase, or glutamate racemase, or (poly)glycerolphosphate/ (poly)ribitolphosphate D-alanyltransferase, or

(poly)glycerolphosphate/ (poly)ribitolphosphate glucosyltransferase.

The invention, also relates to the use of a mutant microorganism as previously defined as a vaccine, wherein said mutant microorganism is expressing a nucleic acid encoding an antigen which is able to elicit an immune response when administered to a human or animal host.

As used herein the term "antigen" is meant to include peptides, polypeptides and proteins, encoded by a nucleic acid expressed by a recombinant microorganism of the present invention, against which an immune response can be elicited, such as an antibody response, and which can act as a trigger for immunizing the individual against the target pathogen.

The antigen produced is preferably a peptide, a polypeptide or a protein derived from a viral or a bacterial pathogen or from fungi or parasites or from other microorganisms capable of infecting human or animal species. Alternatively, an antigen can also be a peptide, a polypeptide or a protein, for instance, derived from a hyperproliferative disease such as cancer, thus also including tumor antigens and auto-immune antigens. They may be post-transcriptionally derivatized antigens like glycosylated, lipidated, glycolipidated or hydroxylated antigens. Besides, such native antigens and antigenic components (including antigenic parts, fragments or epitopes thereof), antigenic mutants or analogs thereof - obtained synthetically or via recombinant DNA techniques - may be produced. In addition, a combination of two or more such antigens may be produced, i.e. by a single type or strain of the recombinant microorganism according to the invention, or by several different types or strains said microorganism. Examples of antigens produced may be chosen from the group comprising antigens derived from toxins such as toxins from diphtheria or from Clostridium (such as Clostridium septicum, Clostridium tetani, Clostridium perfringens), Pneumococcus and other Streptococcus species, Pasteurella maltocida and Corynebacterium pseudotuberculosis. In one aspect, because the uses and the methods according to one aspect of the present invention are related to the delivery of antigen at mueosal surfaces and therefore are directed to the induction of mueosal immunity, selected antigens produced by said recombinant microorganism can also be derived from pathogens which invade the mammal via the mucosa. For example, the pathogen may be selected from viruses: Human Papilloma Viruses (HPV), HIV, HSV2/HSV1 , influenza virus (types A, B, and C), Polio virus, RSV virus, Rhinoviruses, Rotaviruses, Hepatitis A virus, Norwalk Virus Group, Enteroviruses, Astroviruses, Measles virus, Para Influenza virus, Mumps virus, Varicella-Zoster virus, Cytomegalovirus, Epstein- Barr virus, Adenoviruses, Rubella virus, Human T-cell Lymphoma type I virus (HTLV-I), Hepatitis B virus (HBV), Hepatitis C virus (HCV), Hepatitis D virus, Pox virus, Marbug and Ebola; bacteria: Bacillus anthracis, Bordetella pertussis, Brucella, Campylobacter; Escherichia coli, Giacardia lamblia, Klebsiellae species, Haemophilus influenzae, Helicobacter pylori, Listeria monocytogenes, Legionella pneumophila, Franciscella tulorensis, Meningococcus, Moraxella catarrhalis, Mycobacterium tuberculosis, Mycobacterium leprae, Neisseria gonorrhea, Neisseria meningitidis, Proteus species, Streptococcus pneumoniae, Streptococcus pyogenes, Plasmodium sp. (PI. falciparum, PI. vivax, etc.), Pseudomonas aeruginosa, Pneumococcus, Salmonellae species, Shigellae species, Staphylococcus aureus, Treponema pallidum, Vibrio cholerae, Yersinina pestis, Donovanosis, and Actinomycosis; fungal pathogens including Candida and Aspergillus; parasitic pathogens including Taenia, Flukes, Roundworms, Antamoeba, Giardia, Leishmania, Cryptosporidium, Schistosoma, Pneumocystis carinii, Trichomonas and Trichinella.

The recombinant microorganism provided by the invention is readily applicable as a vaccine against any pathogen against which immunization via the mueosal route is effective. Therefore, the invention encompasses the production of antigens derived from a wide range of human or animal pathogens to which mueosal immunity is desired. The term "antigen" is further intended to encompass peptide or protein analogs of known or natural antigens such as those described above, which analogs may be more soluble or more stable than wild type antigen, and which may also contain mutations or modifications rendering the antigen more immunologically active. Thus, the invention is not limited by the identity of the particular antigen produced by said microorganism.

According to the present invention, the immune response elicited may be prophylactic or therapeutic, wherein prophylactic immune response refers to an immune response which targets an antigen to which the individual has not yet been exposed such as a pathogen antigen in an uninfected individual, or a disease cell associated protein in an individual who does not have the disease such as a tumor associated protein in a patient who does not have a tumor. The use of a recombinant microorganism for prophylactic applications should be understood herein that, for instance protective immune responses (including antibodies) could be generated. The expression "therapeutic immune response" refers to an immune response which targets an antigen to which the individual has been exposed such as a pathogen antigen in an infected individual, or a disease cell associated protein in an individual who has the disease or a tumor associated protein in a patient who has a tumor. The use of a recombinant microorganism for therapeutic application, for instance, may result in the reduction or curing of a disease. Moreover, the recombinant microorganism of the present invention can be used to modulate an immune response not only to stimulate (i.e., elicit, produce, and/or enhance) a protective immune response but also to suppress (i.e., reduce, inhibit, block) an overactive, or harmful immune response (induction of tolerance against an allergen, reduction of inflammation, restoration of the immune homeostasis).

A response (e.g. an antibody or cellular immune response) is considered significant when it leads to a detectable change or response in a human or animal, and in particular to a detectable immunological change or response, such as the production of antibodies, cytokines, chemokines, cytotoxic or helper T cell responses, etc. A significant response as used herein may be, but is not necessarily, also a protective response. A response (e.g. an immunological response against a pathogen or an antigen) is considered protective when it is capable of protecting the individual which shows said response against said pathogen and/or against a pathogen associated with said antigen.

The vaccines according to the invention can be formulated such that a single dose is sufficient. However embodiments where multiple applications over a period of time, e.g. with a view to the persistence of the bacterial host in the gastro-intestinal tract, are also envisaged as falling within the scope of the invention. The provision of booster vaccinations is also envisaged as a potential embodiment with the vaccine formulations according to the invention.

The invention further relates to a method for the preparation of a vaccine comprising admixing a recombinant microorganism as defined previously with a pharmaceutically acceptable carrier. Said vaccine or said medicament may also contain one or more adjuvants, including immune adjuvants such as, but not limited to cytokines, co- stimulatory molecules, as long as these are compatible with the recombinant microorganism and do not interfere with its desired immunogenic properties. According to an embodiment, the adjuvants may be a lactic acid bacterium, such as the bacterial host strain itself, or one of the other lactic strains mentioned herein. In another embodiment, said vaccine may be prepared or formulated/processed as in use in the food fermentation technology. It can be formulated such as to obtain a fermented product.

The invention further relates to a method for the preparation of a medicament comprising admixing a recombinant microorganism as defined previously with a pharmaceutically acceptable carrier or formulated/processed as in use in the food fermentation technology.

Pharmaceutically acceptable carriers may include sterile aqueous or non-aqueous solutions, suspensions, and emulsions suitable for ingestion, inhalation, or administration as a suppository to the rectum or vagina. Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and certain organic esters such as ethyl oleate. Aqueous carriers include water, alcoholic/aqueous solutions, emulsions or suspensions, including saline and buffered media.

Preservatives and other additives may also be present such as, for example, antimicrobials, antioxidants, chelating agents, and inert gases and the like. One skilled in the art will select among these available compounds depending upon the particular mueosal inductor site targeted, i.e., whether for ingestion or inhalation. Further, said vaccine or medicament comprising recombinant microorganisms may be lyophilized, freeze dried, dry sprayed or submitted to any formulation or process in use in the starter technology or for probiotic preparations, according to the route of administration, for example inhalation, absorption, genital administration etc....

The vaccine composition or the medicament may suitably be provided in the form of a spray, an aerosol, a mixture, tablets (entero-or not-enterocoated), capsule (hard or soft, entero-or not-enterocoated), a suspension, a dispersion, granules, a powder, a solution, an emulsion, chewable tablets, tablets for dissolution, drops, a gel, a paste, a syrup, a cream, a lozenge (powder, granulate, tablets), an instillation fluid, a gas, a vapor, an ointment, a stick, implants (ear, eye, skin, nose, rectal, or vaginal), vagitories, suppositories, or uteritories suitable for administration via the oral, nasal, vaginal, sublingual, ocular, rectal, urinal, intramammal, pulmonary, otolar, or buccal route. The vaccine composition or the medicament may also be in the form of a fermented product.

The invention further relates to the use of a recombinant microorganism comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component, as a vehicle for delivery of therapeutic or prophylactic compounds or polypeptides at mueosal surfaces, wherein said therapeutic or prophylactic compound or polypeptide is produced by said microorganism and is chosen from a protein, polypeptide, or peptide. Preferably, said therapeutic or prophylactic compound or polypeptide is able to elicit or modulate an immune response when administered to a human or animal host.

The invention further relates to the use of a mutant microorganism comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component, as a vehicle for delivery of DNA molecules such as for example a vector carrying immunostimulatory DNA sequences ISS-DNA, also known as CpG-DNA.

The present invention further encompasses the mutant microorganism according to the invention for use as a medicament, wherein said mutant microorganism comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component. In an embodiment, said cell wall component is preferably peptidoglycan, teichoic acid, or lipoteichoic acid. More preferably said mutation is in a gene involved in the biosynthesis of D-alanine, D-glutamate, or D-aspartate, or in the alanylation or glucosylation of teichoic or lipoteichoic acids. Yet more preferably said gene encodes alanine racemase, or aspartate racemase, or glutamate racemase, or (poly)glycerolphosphate/ (poly)ribitolphosphate D-alanyltransferase, or

(poly)glycerolphosphate/ (poly)ribitolphosphate glucosyltransferase.

Examples of suitable bacterial strain are the same as that described above. According to a preferred embodiment, said microorganism is a gram-positive bacterium, such as a lactic acid bacterial strain. In a preferred embodiment said lactic acid bacterial strain is a Lactobacillus plantarum or Lactococcus lactis strain.

The invention, also relates to the use of a mutant microorganism as previously defined as a medicament. The use of a cell wall mutant microorganism for therapeutic application, for instance, may result in the prevention, reduction or curing of a disease. Moreover, said mutant microorganism can be used to modulate an immune response not only to stimulate (i.e., elicit, produce, and/or enhance) a protective immune response but also to suppress (i.e., reduce, inhibit, block) an overactive, or harmful immune response (induction of tolerance against an allergen, reduction of inflammation, restoration of the immune homeostasis and the like). As such, according to a more preferred embodiment, the mutant microorganism used in the present invention has a mutation in the gene encoding alanine racemase (air), or glutamate racemase (glr), or aspartate racemase, or in the dlt cluster of genes responsible for D-alanylation of teichoic acids and lipoteichoic acids, or in the tagE gene responsible for glucosylation of teichoic acids. The present inventors clearly demonstrated that a Dlt- mutant of L. plantarum NCIMB8826 was able to strongly elicit regulatory cytokines such as IL-10, without having pro-inflammatory properties (i.e. reduced induction of pro- inflammatory cytokines), compared to the wild type strain. Furthermore, the use of said Dlt- mutant led to a more effective prevention of chemically induced colitis when administered to mice, as compared to the wild type strain.

The invention also relates to the use of a mutant microorganism according to the invention for the manufacture of a medicament for prophylactic or therapeutic application. It may for example be used for preventing or treating illness such as allergy, chronic inflammation, chronic infection (HCV), auto-immune diseases, metabolic deficiencies and cancers. Such mutants might also be used as mueosal adjuvants.

In a preferred embodiment, said mutant microorganism can be used for the manufacture of a medicament for the prevention or treatment of various intestinal disorders. In an embodiment of the present invention, said mutant microorganism can be used for the manufacture of a medicament for the prevention or treatment of inflammatory bowel disease.

The invention further relates to a method for the preparation of a medicament comprising admixing a cell wall mutant microorganism as defined previously with a pharmaceutically acceptable carrier or formulated/processed as in use in the food fermentation technology. Suitable examples of pharmaceutically acceptable carriers are the same as that described above. Preservatives and other additives may also be present as described above. Suitable forms for said medicament are the same as that described above.

Examples and figures are provided below as illustrating aspects of the invention. They should be regarded as illustrating rather than limiting the invention, which is defined by the appended claims. Brief description of the drawings

Figure 1 corresponds to a Western Blot illustrating the intracellular production level of the C subunit of tetanus toxin (TTFC) in different host strains. The blot was obtained by immunoblotting cell extracts (equivalent to 10⁸ CFU) obtained from control strain NCIMB8826/ pTG2247 (1) or MG1363/ pTX (8) or TTFC-producing strains NCIMB8826/ pMEC4 (2), NCIMB8826 Air-/ pMEC4 (3), NCIMB8826/ pMEC46 (4), NCIMB8826 Air-/ pMEC46 (5), NCIMB8826/ pMEC127 (6), NCIMB8826 Air-/ pMEC127 (7), MG1363/ pMEC46 (9), MG1363 Air-/ pMEC46 (10) or from 50 ng purified TTFC (11).

Figure 2 is a bar graph depicting the effect of immunization with different strains on serum anti-TTFC IgG titers. Anti-TTFC serum IgG responses were measured following intragastric immunization with wild type NCIMB8826/ pMEC4, NCIMB8826/ pMEC46, NCIMB8826/ pMEC127 or mutant NCIMB8826 Air-/ pMEC4, NCIMB8826 Air-/ pMEC46 or NCIMB8826 Air-/ pMEC127 recombinant Lactobacillus plantarum strains, in comparison to control non-expressor strain [Lactobacillus plantarum (pTG2247)] or buffer alone. Individual sera were collected 10 days after the first administration (I), the first (^), the second (D) or the third (ED) boost.

Figure 3 is a bar graph depicting the effect of immunization with different strains on specific anti-TTFC IgA titers. Anti-TTFC IgA levels were measured in intestinal lavages from groups of eight mice immunized intragastrically with recombinant mutant Lactobacillus plantarum NCIMB8826 Air-/ pMEC127 in comparison to control non- expressor strain NCIMB8826/ pTG2247 or buffer alone. Intestinal lavages were performed 10 days after the last administration (third boost).

Figure 4 is a bar graph depicting the effect of immunization with different strains on serum anti-TTFC IgG titers. Anti-TTFC IgG response were measured, following intragastric immunization with wild type NCIMB8826/ pMEC127 or mutant NCIMB8826 Air-/ pMEC127 recombinant Lactobacillus plantarum and wild type MG1363/ pMEC46 or mutant MG1363 Air-/ pMEC46 recombinant Lactococcus lactis strains in comparison to control non- expressor strains [NCIMB8826/ pTG2247 or MG1363/ pTX] or buffer alone. Individual sera were collected 10 days after the first administration (M), the first (ϋ), or the second (D) boost.

Figure 5 is a bar graph depicting the effect of immunization with different strains on serum anti-TTFC IgG titers. Anti-TTFC IgG response were measured following intravaginal immunization with wild type NCIMB8826/ pMEC127 or mutant NCIMB8826 Air-/ pMEC127 recombinant Lactobacillus plantarum and wild type MG1363/ pMEC46 or mutant MG1363 Air-/ pMEC46 recombinant Lactococcus lactis strains in comparison to control non- expressor strains [NCIMB8826/ pTG2247 or MG1363/ pTX] or buffer alone. Individual sera were collected 10 days after the first administration (H), the first (ϋ) or the second (EH) boost.

Figure 6 is a bar graph depicting the effect of immunization with different strains on serum anti-TTFC IgG titers. Anti-TTFc IgG response were measured following intrarectal immunization with wild type NC1MB8826/ pMEC127 or mutant NCIMB8826 Air-/ pMEC127 recombinant Lactobacillus plantarum and wild type MG1363/ pMEC46 or mutant MG1363 Air-/ pMEC46 recombinant L. lactis strains in comparison to control non-expressor strains [NCIMB8826/ pTG2247 or MG1363/ pTX] or buffer alone. Individual sera were collected 10 days after the first administration (H), the first <ffl), or the second (D) boost.

Figure 7 is a bar graph depicting the effect of immunization with different strains on specific anti-TTFC fecal IgA titers. Anti-TTFC IgA levels were measured in feces from groups of mice immunized intrarectally with wild type NCIMB8826/ pMEC127 or mutant NCIMB8826 Air-/ pMEC127 recombinant Lactobacillus plantarum strain or wild type MG1363/ pMEC46 or mutant MG1363 Air-/ pMEC46 recombinant L lactis strains in comparison to control non-expressor strain [NCIMB8826/ pTG2247 or MG1363/ pTX] or buffer alone. Feces were collected from pooled groups of 3 mice, 10 days after the first (_^) or second (H) boost.

Figure 8A corresponds to a Western Blot illustrating the intracellular production level of UreB in different Lactobacillus plantarum strains. The blot was obtained by immunoblotting 5 μg (1) or 10 μg (2) of recombinant UreB (rUreB) and cell extracts (equivalent to 10⁶ CFU) obtained from UreB-producing strains NCIMB8826/ pMEC142 and NCIMB8826 Air-/ pMEC142 prepared at day 1 , and stored at 4°C until day 2 or 3.

Figure 8B and 8C are bar graphs depicting the effect of immunization with different strains on specific anti-UreB serum IgA (B) and IgG (C). Anti-UreB antibody responses were measured following intragastric immunization with wild type NCIMB8826/ pMEC142, or mutant NCIMB8826 Air-/ pMEC142 recombinant Lactobacillus plantarum strains, in comparison to control non-expressor strain Lactobacillus plantarum I pTG2247 or medium alone. As a positive control (protection against a Helicobacter felis challenge), mice were immunized with 50 μg recombinant UreB + 10 μg cholera toxin (CT). Individual sera were collected prior immunization (D), 3 days after the last immunization (E3) or at sacrifice 2 weeks after H. felis infection (H). Results are expressed as OD units using 1 :200 (IgA) or 1:1 '000 (IgG) dilutions of mouse sera, respectively.

Figure 9 represents genomic DNA amplification by PCR resolved on 2 % agarose gels containing ethidium bromide. Representative examples for mice vaccinated with Lactobacillus plantarum NCIMB8826 Air-/ pMEC142 (A) and UreB/CT (B) are shown. Single band (marked by a star) detected by UN. fluorescence was considered as PCR- positive. End-points of analysis were set at 32 and 40 cycles with constant amount of template genomic DΝA.

Figure 10 corresponds to a Western Blot illustrating the intracellular production level of TTFC in different host strains. The blot was obtained by immunoblotting cell extracts (equivalent to 10⁸ CFU) obtained from the control strain Lactococcus lactis MG1363 / pTX (1) or TTFC-producing Lactococcus lactis strains ΝZ3900/ pMEC46 (2), MG1363 DltD-/ pMEC46 (3) or 200 ng purified TTFC (4).

Figure 11 is a bar graph depicting the effect of immunization with different strains on serum anti-TTFC IgG responses. Anti-TTFC IgG titers were measured, following intragastric immunization with the wild type (NZ3900/ pMEC46) or mutant (MG1363 DltD-/ pMEC46) recombinant Lactococcus lactis strains in comparison to control non-expressor strain (MG1363/ pTX) or buffer alone. Individual sera were collected 10 days after the first administration (I), the first (^), or the second (11) boost.

Figure 12 is a bar graph depicting the effect of immunization with recombinant Lactococcus lactis strains on specific anti-TTFC intestinal IgA titers. Anti-TTFc IgA levels were measured in intestinal lavages from groups of mice immunized intragastrically with wild type NZ3900/ pMEC46 or mutant MG1363 DltD-/ pMEC46 recombinant Lc. lactis strains in comparison to control non-expressor strain (MG1363/ pTX) or buffer alone. Intestinal lavages were collected individually 10 days after the last boost.

Figure 13 is a bar graph depicting the cytokine response of human PBMC (2 x 10^δ cells / ml) to Lactobacillus plantarum NCIMB8826 wild type and Dlt- mutant (NCIMB8826 DltB-) strains. PBMC from healthy human donors were stimulated with or without (medium only)

10⁷ CFU / ml of thawed bacteria. IL-10, IL-12 (p70), TNFα and IFNγ were measured by ELISA in the PBMC supernatants collected after 24h stimulation. The results represent the mean + SEM for the response of 11 donors to each bacterial strain.

Figure 14 is a bar graph depicting the cytokine response of human monocytes (0.5 x 10⁶ cells / ml) to Lactobacillus plantarum NCIMB8826 wild type and DltB-mutant strains. Monocytes were stimulated with or without (medium only) 10⁷ CFU / ml of thawed bacteria. IL-10, IL-12 (p70), TNFα and IFNγ were measured by ELISA in monocytes supernatants collected after 24h stimulation. The results represent the mean + SEM for the response of 4 donors to each bacteria strain.

Figure 15 is a bar graph depicting the effect of preventive administration of Lactobacillus plantarum NCIMB8826 wild type and DltB- mutant strains on acute TNBS-induced colitis in Balb/c mice. (A) Weight variation between day 5 (TNBS administration) and day 7 (sacrifice). (B) Wallace and (C) Ameho inflammation scores. Results are means ± SEM of two separated experiments (9 mice/ group). Significant p values < 0.05 ** or < 0.1*, as compared to the TNBS control group that received no bacteria.

Examples

Example 1 Construction of recombinant strains

Lactobacillus plantarum (Lb. plantarum) NCIMB8826 and Lactococcus lactis (Lc. lactis) MG1363 were used. The Lb. plantarum NCIMB8826 Air- (MD007) (with defective gene coding for alanine racemase) and the Lc. lactis MG1363 Air- (PH3960) were obtained as described in Bron P. et al, Appl. Environ. Microbiol. 2002 Nov; 68 (11): 5663-70, both strains were deposited on June 16, 2003 at the Belgian Coordinated Collections of Microorganisms (BCCM). The replicative plasmid (pMEC4) carrying the TTFC-encoding gene (TTFC, C subunit of tetanus toxin) under the control of a constitutive promoter, the LDH promoter, was first constructed. pMEC127 was obtained by deleting the sequence coding for 25 amino acids of the N-terminus of the LDH gene (Reveneau et al., Vaccine, 2002, 20(13): 1769-1777). The replicative plasmid (pMEC46) carrying the TTFC-encoding gene under the control of nisin inducible promoter (pNis A, Pavan S. et al., Appl. Environ. Microbiol., 2000; 66(10): 4427-32) was also used in the experiment. These plasmids were introduced by electroporation in wild type and Air- mutant strains of Lb. plantarum NCIMB8826 (plasmids pMEC4, pMEC46, pMEC127) and Lc. lactis MG1363 (plasmid pMEC46). The production of heterologous antigen (TTFC) was verified by western blot (Figure 1 ). It was observed that recombinant wild type and recombinant mutant strains produce the same quantity of antigen (for Lb. plantarum NCIMB8826/ pMEC4, / pMEC46, / pMEC127 and Lc. lactis MG1363/ pMEC46). The levels of TTFC produced by Lb. plantarum NCIMB8826/ pMEC4, NCIMB8826/ pMEC46 and NCIMB8826/ pMEC127 were respectively low, moderate or high. The level of TTFC production observed for the Lc. lactis MG1363/ pMEC46 strain was similar to that obtained with the NCIMB8826/ pMEC127 strain. No antigen production was detected in the control strains NCIMB8826/ pTG2247 and MG 1363/ pTX.

Example 2: In vivo delivery of TTFC by lactic acid bacteria

Wild type strain Lb. plantarum harboring the control plasmid (NCIMB8826/ pTG2247) and TTFC-producing wild type strains (NCIMB8826/ pMEC4, NCIMB8826/ pMEC46 and NCIMB8826/ pMEC 27) were grown in MRS broth (Difco, Detroit, Michigan), while the recombinant Air- mutant strains (NCIMB8826 Air-/ pMEC4, NCIMB8826 Air-/ pMEC46 and NCIMB8826 Air-/ pMEC127) required the addition of D-alanine (200 μg/ml) for their growth. Lc. lactis wild type strains (MG1363/ pTX and MG1363/ pMEC46) were grown in M17 (Difco) supplemented with 0.5% glucose; the Air- strain (MG1363 Air-/ pMEC46) required the addition of D-alanine (400 μg/ml) in the growth medium. For strains harboring the pMEC46 plasmid, expression of the TTFC encoding gene was induced by adding nisin at 20 ng/ml for 4h or nisin at 5 ng/ml for 3h in the case of Lb. plantarum or Lc. lactis strains, respectively. The expression of the altered phenotype linked to the Air- mutation requires the growth of the strains without alanine. Starvation times have been fixed to 3 hours for Lb. plantarum and 2 hours for Lc. lactis. The preparation of bacterial inocula before intra-gastric administration was done by growth and harvest of the bacteria at exponential phase (optical density at 600 nm [OD₆₀₀] of « 1-2), with the Air- mutants being grown in media supplemented with alanine. The latter were subsequently starved by incubation in media without alanine. After two washes in corresponding media, the bacteria were resuspended in a gavage buffer (for intra-gastric administration) allowing buffering of the gastric pH (PBS containing 0.2 M NaHCO₃, 0.5% casein hydrolysate and 0.5% glucose) or in PBS (for intra-vaginal and intra-rectal administration) and the concentrations were adjusted to 10¹⁰ CFU (colony forming unit)/ml, 10¹¹ CFU/ml or 2 10¹⁰ CFU/ml for intra-gastric, intra-vaginal or intra-rectal administration respectively. Groups of 8 mice C57/B16 were immunized either by intra-gastric gavage with 100 μl of bacterial suspension (10⁹ CFU) or 100 μl buffer, or by intra-vaginal or intra-rectal administration with, respectively, 10 μl or 50 μl of bacterial suspension (10⁹ CFU) or with PBS. An aliquot of each suspension was stored at -20°C and analyzed by Western blot to check the antigen dose actually administered to mice. The immunization protocol consisted of 3 consecutive doses of 10⁹ CFU every 24 h, followed by 2 boosts at 3-week-intervals. Serum samples were collected by retro-orbital bleeding of the mice 10 days after each administration. The intestinal washes were performed 10 days after the last administration. The intestine of individual mice was cut in its length and washed with 1 ml of PBS containing protease inhibitors (Complete Protease Inhibitor Cocktail, Boehringer). Each homogenate was further cleared by 2 consecutive centrifugations first at 3000 rpm and further at 12000 rpm. For the fecal IgA measurements, the feces were collected from 3 mice per group, pooled, weighed and homogenized in PBS containing protease inhibitors at a final suspension (Weight/Volume) of 100 mg feces per ml. Each homogenate was further cleared by 2 consecutive centrifugations first at 3000 rpm and further at 12000 rpm (labtop centrifuge eppendorf). Example 3: Evaluation of TTFC specific IgG and IgA response by ELISA .

TTFC antigen was coated onto a plastic support (Microtiter plates 96 wells Immulon I Dynatech) by incubation overnight at 4°C, of 100 μl per well of a 2 μg /ml solution of TTFC in 0.1 M NaHCO₃/Na₂CO₃ buffer pH 9.5. After blocking of the wells with a solution of PBS containing 3% bovine serum albumin (BSA, Sigma), followed by a wash with 0.1% PBS/Tween, the serum samples or intestinal wash samples were added in a 2 fold serial dilution in 1% PBS/BSA buffer from 100 μl of a 1/50 dilution (sera) or a 1/4 dilution (intestinal wash, fecal suspensions). After 2 hours incubation at room temperature, the plates were washed three times in 0.1% PBS/Tween. The bound antibody was detected by addition of 100 μl of mice biotintylated anti-lgG (1/10000) or anti-lgA (1/2000) (Southern Biotechnology) in 1% PBS/BSA-0.1% Tween buffer, and incubation for 1 hour at room temperature. After 4 washes in 1% PBS/Tween buffer, 100 μl of streptavidin peroxidase conjugate (Amersham) were added at a 1/2000 dilution in 1% PBS/Tween buffer. After 30 min of incubation at room temperature, followed by 6 washes in 1% PBS/Tween buffer, 100 μl of 1 mg/ml of o-phenylenediamine substrate (Sigma) in 0.2 M Na2HPO4, 0.1 M Citrate buffer (pH 5.5) was added. After 30 min of incubation at 37 °C, the color-reaction was followed by measuring the absorbance at 490 nm. End point titers of specific antibodies were calculated as corresponding to the final dilution that gave an absorbance three times higher than background, that is the absorbance measured for a well containing only the reagents and not the samples. The concentration of specific IgA in the intestinal washes was calculated by taking into account the total IgA concentration in each sample. Total IgA levels were determined by ELISA, as described above, with microplates coated with non-labeled anti-lgA from mice (Sigma). The concentration of IgA was calculated using a standard curve of mouse lgA_k (Sigma) with consecutive serial dilutions from 100 to 0.78 ng/ml. The results were expressed as specific activity, calculated by dividing the endpoint titer by the level of total IgA concentration expressed in μg/ml. The statistical significance was evaluated by Mann-Whitney U test.

Example 4: Analysis of the in vivo immune response induced by intragastric administration of recombinant TTFC-producing wild-type and mutant Lb. plantarum strains

In the first construction pMEC4, the heterologous gene was placed under the control of a constitutive promoter (L-LDH promoter). Deletion of the sequence coding for 25 amino acids of the N-terminus of the LDH gene resulted in the second-generation plasmid pMEC127 which, after transformation in Lb. plantarum NCIMB8826, resulted in a recombinant strain (NCIMB8826/ pMEC127) capable of producing much higher TTFC amounts than those produced by the NCIMB8826/pMEC4 strain (Reveneau et; al., Vaccine, 2001, 20(13): 1769-1777). In the pMEC46 construct, the heterologous gene was under the control of an inducible promoter (pnisA). Transformation of pMEC46 in Lb. plantarum NCIMB8826 resulted in a recombinant strain NCIMB8826/ pMEC46 which after induction by nisin produced more TTFC than strain NCIMB8826/ pMEC4 but less than strain NCIMB8826/ pMEC127. Nasal administration of these strains led to very strong systemic immune responses (Grangette C. et al. Infect. Immun. 2001, 69(3): 1547-53). Furthermore, only the strain NCIMB8826/ pMEC127 was capable to induce excellent systemic and local responses after intra-gastric administration of 10⁹ CFU (Figure 2), leading to anti-TTFC serum IgG end point titers of 2.10⁴. On the other hand no significant response was induced by either the first generation strain NCIMB8826/ pMEC4 or the NCIMB8826/ pMEC46 strain. The use of Air- mutant led to a very clear increase of specific immune responses after intra-gastric administration to mice for the two strains synthesizing medium or high amounts of antigen (NCIMB8826 Air-/ pMEC46 or NCIMB8826 Air-/ pMEC127, respectively) when compared to the wild type homologue. Combining the two improvements, gene expression system and bacterial host, it was demonstrated that when the TTFC coding gene is under the control of the LDH constitutive promoter in the second generation vector (pMEC127), the recombinant Alr- mutant strain elicited the serum IgG end point titers twenty times higher (p < 0.05) than those obtained with the recombinant wild type strain. After the third boost, the serum IgG titers were equal to 4 x 10⁵ and 2 x 10⁴ for strains NCIMB8826 Air-/ pMEC127 and NCIMB8826/ pMEC127, respectively (Figure 2). The NCIMB8826 Air-/ pMEC127 strain was moreover able to induce a significant response (titer 2 x 10³) already after priming. Moreover, when the TTFC-coding gene was under the control of the nisin-inducible promoter (pMEC46), only the recombinant Air- mutant strain (NCIMB8826 Air-/ pMEC46) was able to induce a significant immune response. Furthermore, intra-gastric administration of the recombinant mutant strain (NCIMB8826 Air-/ pMEC127) induced significant local IgA response (p < 0.005) in intestinal lavages of immunized mouse (Figure 3). In contrast, no local response was observed in the intestinal lavages of mice who received the buffer or the control strain (NCIMB8826/ pTG2247).

To initiate protective immune responses after oral administration it is therefore necessary that high levels of antigen (TTFC) are produced by the recombinant strains, at least when the antigen is produced intracellularly, and this was achieved by optimizing the plasmid constructs. Moreover, the use of mutant strain of Lb. plantarum NCIMB8826 deficient in the alanine racemase permitted the effective delivery of a heterologous antigen after intragastric administration. The responses observed were significantly higher than those described in the literature, especially after the use of Lc. lactis recombinant strains (final titer ten times higher) or alternative wild type Lb. plantarum strains.

Example 5: In vivo comparative analysis of the immune response elicited by intra-gastric administration of wild type and mutant recombinant Lb. plantarum and Lc. lactis strains

In order to understand the impact of the mutation on the induction of the immune response and more specially on the antigen delivery after local administration, the uses of wild type and Air- mutant strains of Lb. plantarum NCIMB8826 (persistent prototype, i.e. a strain which is able to transiently persist in the mouse intestine) and Lc. lactis MG1363 (non-persistent prototype, i.e. a strain which is no longer detectable in faecal samples after 36 hours) were compared. Three experiments have been performed, the first one by intra-gastric administration, the second one by intra-vaginal administration and the third one by intra-rectal administration.

In these experiments, the serum IgG immune responses obtained after local administration of recombinant wild type and recombinant Air- mutant strains of Lb. plantarum NCIMB8826 wherein the heterologous gene coding for TTFC was under the control of a constitutive promoter (pMEC127), were compared with the serum immune responses obtained with recombinant wild type and recombinant Air- mutant strains of Lc. lactis MG1363 wherein said TTFC gene was under the control of an inducible promoter (pMEC46). All the constructs allowed the production of a comparable level of antigen as estimated by Western blotting. The immune responses induced by these recombinant strains producing the antigen intracellularly were compared to those induced by the control strains, respectively Lb. plantarum NCIMB8826/ pTG2247 and Lc. lactis MG1363/ pTX, which are strains that do not produce the antigen, and to those obtained in mice receiving only PBS buffer. After intra-gastric administration (Figure 4) the recombinant wild type Lb. plantarum strain (NCIMB8826/ pMEC127) was able to induce a serum IgG response significantly higher than the one induced by the recombinant wild-type strain Lc. lactis (MG1636/ pMEC46). As stated above (Example 4), the use of Lb. plantarum NCIMB8826 Air- mutant strain allowed a significant increase of the immune responses when compared to the corresponding wild type strain (Lb. plantarum NCIMB8826/ pMEC127). In the same way, the use of the Lc. lactis Air- mutant (MG1363 Air-/ pMEC46) induced a significantly higher immune response (20 to 30 fold enhancement) than that induced by the wild type strain. The same levels of antibody response were obtained with the Lc. lactis MG1363 Air-/ pMEC46 and Lb. plantarum NCIMB8826/ pMEC127 strains, as no significant statistical difference was observed. The two recombinant Lb. plantarum strains, i.e. the wild type NCIMB8826/ pMEC127 and the mutant NCIMB8826 Air-/ pMEC127, induced a protective immune response, whereas only the MG1363 Air-/ pMEC46 Lc. lactis strain induced neutralizing antibodies (Table I).

Table I: Neutralizing activity in serum antibodies elicited by recombinant LAB strains

Neutralizing tetanus antibody levels in pooled sera Protection

Immunization protocol. _{/ιι ι j ι a} (IU/ml)^a (>0,01)

Buffer «0,0025

NCIMB8826/ pTG2247 «0,0025

NCIMB8826/ pMEC127 0,02-0,04

NCIMB8826 Air-/

0,01-0,02 pMEC127

MG1363/ pTX «0,025

MG1363/ pMEC46 «0,025

MG 1363 Air-/ pMEC46 0,04-0,08 ^aSera from intragastrically immunized mice were collected and pooled 10 days after the 2^nd boost and used in the tetanus toxin (TT) neutralization assay (McComb J.A., N. Engl. J. Med., 1964, 270: 175-178). Protection against tetanus toxin requires a minimum neutralizing antibody titer of 0.01 lU/ml (International Units).

These results show that the use of a cell wall mutant promotes an increase in the immune response and is not restricted to Lactobacillus strains but can also be applied to other lactic acid bacteria, such as here demonstrated with a second genus of lactic acid bacteria. Moreover, the use of cell wall mutants seems essential when the expression system used in the wild type strain does not allow the induction of a protective response, such as immunization with Lc. lactis MG1363/ pMEC46 strain.

These results have been confirmed by two other experiments of local immunizations using the vaginal and rectal routes of administration. In these experiments the immune responses induced by the two recombinant species Lb. plantarum (persistent prototype) and Lc. lactis (non-persistent prototype), were analyzed. A controlled comparison was conducted between the wild type (NCIMB8826/ pMEC127, MG1363/ pMEC46) and mutant strains (NCIMB8826 Air-/ pMEC127, MG1363 Air-/ pMEC46). It was verified by intra-vaginal administration (Figure 5) that the recombinant wild type Lb. plantarum strain (NCIMB8826/ pMEC127) was able to induce a serum IgG response significantly higher than the one induced by the recombinant wild type Lc. lactis strain (MG1363/ pMEC46). The use of the Air- mutant also allowed for each of the species, Lb. plantarum and Lc. lactis, to increase significantly the antibody responses, when compared with the corresponding wild type strains.

After intra-rectal administration a significant increase of the serum IgG response was also observed when using the Lb. plantarum Air- mutant (Figure 6). The analysis of the local IgA response clearly showed a higher response when using Lb. plantarum Air- mutant (Figure 7). However in the case of the Lc. lactis Air- mutant, no increase in the local immune response was observed, which might reflect an intrinsic difference between the two bacterial vaccine vectors used.

Example 6: Construction of recombinant Lb. plantarum strains producing Urease B from Helicobacter pylori

The replicative plasmid pMEC142 carrying the ureB gene (encoding Urease B) from Helicobacter pylori, encoding gene under the control of a constitutive promoter, the LDH promoter, was constructed and introduced by electroporation in wild type and Air- mutant strains of Lb. plantarum NCIMB8826. The intracellular production of the heterologous antigen (UreB) was verified by Western blot (Figure 8A). It was observed that the recombinant wild type (NCIMB8826/ pMEC142) and mutant (NCIMB8826 Air-/ pMEC142) strains produced similar quantities of antigen. No antigen production was detected in the control strain NCIMB8826/ pTG2247 (data not shown).

Example 7: In vivo delivery of UreB by recombinant Lb. plantarum strains (Wild type and Air- mutant strains)

The wild type strain Lb. plantarum harboring the control plasmid (NCIMB8826/ pTG2247, non-expressor strain) and the UreB producing wild type strain (NCIMB8826/ pMEC142) were grown in MRS broth (Difco) supplemented with 5 μg/ml erythromycin or 10 μg/ml chloramphenicol, respectively. The recombinant Air- mutant strain (NCIMB8826 Air-/ pMEC142) required the addition of D-alanine (200 μg/ml) for its growth. However, the expression of the altered phenotype linked to the air mutation requires growing the mutant strain without D-alanine. Wild type NCIMB8826/ pMEC142 was harvested at exponential growth phase (OD₆₀₀ of 1-2), while the mutant NCIMB8826 Air-/ pMEC142 was harvested after starvation. After 2 washes in PBS, the latter was subsequently starved by incubation for 3 hours in medium without D-alanine. After two washes in the corresponding medium, the bacteria were resuspended in a gavage buffer allowing buffering of the gastric pH (PBS containing 0.2 M NaHCO₃, 0.5% casein and 0.5% glucose) and the concentrations were adjusted to 10¹⁰ CFU (colony forming unit)/ml. Groups of 10 mice Balb/c were immunized by intra-gastric gavage with 200 μl of bacterial suspension containing 10⁹ CFU or 200 μl of medium. An aliquot of each suspension was stored at -20°C and analyzed by Western blot to check the antigen dose actually administered to mice (Fig. 8A). The immunization protocol consisted of a priming of 3 consecutive doses of 10⁹ CFU every 24 h, followed by 2 boosts (triple doses at day 0, 1 and 2) at 3-week-intervals. As a positive control (protection against an Helicobacter felis challenge), mice were immunized intragastrically 3 times at weekly intervals with 50 μg recombinant UreB (rUreB) and 10 μg cholera toxin (CT) in a total volume of 200 μl PBS. Serum samples were collected by tail bleeding of the mice 3 days after the last administration. Twelve days after the last immunization, mice were infected with freshly cultured H. felis (5 x 10⁷ CFU) by orogastric intubation with polyethylene tubing introduced at a fixed distance of 4.5 cm from the incisors, under light anesthesia with isofluran (Baxter, Switzerland). Mice were sacrificed 2 weeks later and stomachs were collected to examine the degree of protection. The presence of Helicobacter in gastric tissue was assessed by urease test (UT) (Jatrox-test, Procter & Gamble, Weiterstadt, Germany). One half of the stomach was immersed in 500 μl of reconstituted mix according to the supplier's recommendation and incubated at 37°C for 3 h. Specimen were centrifuged and the supernatant was used for spectrophotometric quantification at an optical density of 550 nm, as described previously (Corthesy-Theulaz et al., 1995). The cut-off value of urease test used to discriminate between infection and cure corresponded to the mean ± 2 Standard Deviation of the absorbance values obtained with gastric tissues of naive mice. The other half of the stomach was fixed in neutral buffered 10% formalin prior to processing for genomic DNA extraction. About 20 mg stomach was cut into small pieces, placed into a 1.5-ml plastic tube with 180 μl of ATL buffer (Qiagen, Basel, Switzerland), and homogenized. Following digestion with proteinase K (1.5 mg/ml final concentration) for 30 min at 55°C, 200 μl of AL buffer (Qiagen, Basel, Switzerland) was added. The mixture was incubated for 10 min at 70°C, 200 μl of 100% ethanol was added and the lysate was cleared by centrifugation. The following steps were as described in Sambrook et al. (1989). H. felis genomic DNA served as a positive control to help designing a pair of primers highly specific for urease B subunit (Ferrero and Labigne, 1993). Forward primer 5'- GGAAGCACACCTGCAGCTATTCACC-3' and backward primer 5'- GCGAATCCTCGAATCGGCAAACTGC-3' yielded a single band of 348 bp when combined with DNA template under the following conditions: T_denaturation'- 95°C; T_hybridation: 60°C; T_ex_tensi_on'- 72°C; 30 cycles. Identical conditions were used for amplification of stomach samples, with end-points set after 32 and 40 cycles using 1 μg of genomic DNA as a template. PCR products were analyzed onto 2% agarose gels cast and run in 1 x Tris-borate-EDTA buffer (Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.).

Example 8: Evaluation of UreB specific serum IgG and IgA responses by ELISA

Sera recovered prior to immunization, after the second boost, and at sacrifice, were analyzed by ELISA. Maxisorp immunoplates (Life Technologies, Basel, Switzerland) were coated overnight at 4°C with 0.5 μg recombinant urease (ou rUreB) or 1 μg H. felis lysate in 50 μl of 50 mM sodium bicarbonate (pH 9.6). Nonspecific binding sites were blocked with 200 μl of TBS-0.05% Tween 20 (TBS-T) containing 5% non-fat dry milk (NFDM) for 1 h at 37°C. Plates were washed 3 times with TBS-T, and mouse serum dilutions in TBS- T:0.5% NFDM were incubated for 2 h at 37°C. Urease-specific Ab were detected using either rabbit anti-lgG hooked to HRP (Sigma), or goat anti-lgA (α chain-specific) coupled with HRP (Sigma), respectively. Reactions were developed with 1 ,2-phenylenediamine as a substrate, then stopped with 2 M H₂SO₄. Absorbance was read at 490 nm, with 620 nm as the reference wavelength.

Example 9: Analysis of the in vivo immune response induced by intragastric administration of recombinant UreB-producing wild type and Air- mutant Lb. plantarum strains

The immunogenicity of the UreB producing Air- mutant strain (NCIMB8826 Air- / pMEC142) was compared to the immune responses elicited by the recombinant wild type strain (NCIMB8826/ pMEC142). Negative controls included mice receiving the non- expressor strain (NCIMB8826/ pTG2247) or medium only. Mice of the positive control group received rUreB adjuvanted with CT. All the expressor strains were able to induce a significant serum IgA (Figure 8B) and IgG (Figure 8C) response as compared to those induced in the control groups (non-expressor strain or medium) after intra-gastric administration. Nevertheless, the serum IgA response elicited by the recombinant mutant strain (NCIMB8826 Air-/ pMEC142) was higher than that induced by the recombinant wild type strain (NCIMB8826/ pMEC142) or the adjuvanted pure antigen (rUreB + CT). Significant IgG responses were elicited in all mice immunized with either one of the recombinant strains (NCIMB8826/ pMEC142 or NCIMB882 Air-/ pMEC142) or rUreB + CT. Infection with H. felis did not induce the production of more antibodies, at least during the 2 weeks preceding mouse sacrifice (Figs. 8 B and C).

These results demonstrate that the Air- mutant of Lb. plantarum NCIMB8826 constitutes a more efficient delivery system than its wild type counterpart for another model antigen, UreB.

Example 10: Analysis of the protection against H. felis infection after intragastric administration of recombinant UreB-producing wild type and mutant Lb. plantarum strains

The five groups of immunized mice described in Example 9 were subsequently infected with H. felis. When the bacterial load in mouse half stomachs was assayed using the urease test, a markedly slower color development was observed for samples of the group of mice immunized with the recombinant Air- mutant strain (NCIMB8826 Air-/ pMEC142) as compared to the negative control groups or mice immunized with the NCIMB8826/ pMEC142 strain. Consistent with this observation, end-point values measured at 3h were significantly lower for the group of mice fed with the Lb. plantarum NCIMB8826 Air-/ pMEC142 (p < 0.05) or the positive control (UreB/CT; p < 0.05) as compared to the other groups of mice (data not shown). The analysis was further extended by monitoring the H. felis load through PCR performed on genomic DNA isolated from mouse stomachs of the various experimental groups. End-points were set at 32 and 40 cycles, and representative examples are shown in Figure 9. In the case of animals immunized with the NCIMB8826 Air-/ pMEC142 strain, PCR product analysis obtained after 32 cycles showed amplification in 3 lanes only (Figure 9A), while 2 positive signals were yielded in the UreB/CT- vaccinated group (Figure 9B). Consistent with the urease test, 40 cycles of amplification generated positive signals in all lanes for the NCIMB8826 Air-/ pMEC142 group, whereas lanes negative in the UreB/CT group remained negative. The PCR analysis described above represents a sensitive method to assess reduction in the H. felis load and thus the level of protection against this pathogen. Data gathered from all experimental groups are summarized in Table 2. Specifically, the recombinant wild type strain (NCIMB8826/ pMEC142) allowed a reduction in the H. felis load in 4 out of 9 mice, while mice immunized with the recombinant mutant strain (NCIMB8826 Air-/ pMEC142) exhibited a drop in the H. felis load in 6 out of 9 mice. This is close to the degree of protection elicited by UreB/CT (7/9) used as a positive vaccination control. Under the same analytical conditions, 7 out of 10 mice given MRS medium were infected, while 5/7 mice given the control non-expressor Lb. plantarum strain (NCIMB8826/ pTG2247) were positive.

This demonstrates that the Air- mutant of Lb. plantarum NCIMB8826 behaves as a substantially improved delivery system as compared to its wild type counterpart, even in the case of UreB, an antigen known to be much less immunogenic than TTFC.

Table 2. Protection against Helicobacter felis infection in groups of mice immunized with recombinant wild type or Air- mutant strains of Lb. plantarum NCIMB8826 producing UreB

^(a): Presence of the H. felis genomic DNA in mouse stomach samples as measured by PCR (32 cycles). Groups smaller than 10 mice indicate that some of them died during the experiment.

Example 11: DltD- mutant of Lc. lactis as antigen delivery vehicle: Construction of recombinant Lc. lactis strains producing TTFC

Lc. lactis MG1363 DltD- mutant strain (BG007) (with defective dltD gene, dltD being involved in D-alanylation of LTAs) deposited on June 16, 2003 at the Belgian Coordinated Collections of Microorganisms (BCCM), was obtained with a ISS1 insertion in the dltD gene (Duwat et al. J Bacteriol. 1997 Jul; 179 (14): 4473-9.). NisRK genes were further introduced in the chromosomal pepN gene by homologous recombination as described in Ruyter et al. J. Bacteriol. 1996; 178: 3434-39.

The replicative plasmid (pMEC46) carrying the TTFC-encoding gene under the control of nisin inducible promoter (pNisA, Pavan S et al, Appl. Env. Microbiol., 2000; 66 (10): 4417- 32) was introduced by electroporation in the NZ3900 wild type and in the MG1363 DltD- mutant strains of Lc. lactis. The production of heterologous antigen (TTFC) was verified by Western blot (Figure 10). The blot was obtained by immunoblotting cell extracts (equivalent to 10⁸ CFU) obtained from control strain MG1363 / pTX (1) or TTFC-producing strains NZ3900/ pMEC46 (2), MG1363 DltD-/ pMEC46 (3) or 200 ng purified TTFC (4). The TTFC level observed for the mutant (MG1363 DltD-/ pMEC46) (3) seemed to be higher than that obtained for the wild type strain (NZ3900/ pMEC46) (2). No antigen production was detected in the control strain (MG1363/ pTX) (1).

Example 12: In vivo delivery of TTFC by recombinant L. lactis strains

All Lc. lactis strains were grown in M17 (Difco) supplemented with 0.5% glucose. Erythromycin 10 μg/ml was added for the growth of the wild type Lc. lactis strain harboring the control plasmid (MG1363/ pTX) and chloramphenicol at 20 μg/ ml was added for the growth of the TTFC-producing wild type (Lc. lactis NZ3900/ pMEC46) and mutant (Lc. lactis MG1363 DltD-/ pMEC46) strains. For strains harboring the pMEC46 plasmid, expression of the TTFC encoding gene was induced by adding nisin at 5 ng / ml when the culture reached an OD₆₀₀ of 0.5, followed by 3h incubation at 30°C. The preparation of bacterial inocula before intra-gastric administration was done by growth and harvest of the bacteria after nisin induction. After two washes in PBS, the bacteria were resuspended in the gavage buffer (as previously described) at the concentration of 10¹⁰ CFU / ml. Groups of 8 mice C57/B16 were immunized by intra-gastric gavage with 100 μl of bacterial suspension (10⁹ CFU) or 100 μl buffer. An aliquot of each suspension was stored at - 20°C and analyzed by Western blot to check the antigen dose administered to mice. The immunization protocol consisted of a priming of 3 consecutive doses of 10⁹ CFU every 24h, followed by 2 boosts (triple doses) at 3-week intervals. Serum samples were collected by retro-orbital bleeding of the mice 10 days after each administration. The intestinal washes were performed 10 days after the last administration as previously described.

Example 13: Comparative analysis of the immune response induced by intra-gastric administration of the wild type and DltD- recombinant Lc. lactis strains producing TTFC.

In order to analyze the effect of another cell-wall mutation than air on the induction of the immune response, the immunogenicity of TTFC-producing wild type (NZ3900/ pMEC46) and Dlt- mutant (MG1363 DltD-/ pMEC46) strains of L lactis NZ3900 was compared. In this experiment, the immune responses obtained after intra-gastric administration of the recombinant Dlt- strain wherein the gene coding for TTFC was under the control of an inducible promoter (MG1363 DltD-/ pMEC46), were compared with the immune responses obtained with recombinant wild type strain of Lc. lactis NZ3900 (NZ3900/ pMEC46).

The immune responses induced by the recombinant strains producing the antigen intracellularly were compared to those induced by the control strain MG1363/ pTX and to those obtained in mice receiving only gavage buffer. After intra-gastric administration (Figure 11 ) all the expressor strains were able to induce a significant immune response as compared to those induced in the control groups (non-expressor strain or buffer). Nevertheless the serum IgG response elicited by the recombinant Dlt- mutant strain (MG1363 DltD-/ pMEC46) was significantly higher (titer of 4.4 x 10⁵ after the second boost) than the one induced by the recombinant wild type strain (NZ3900/ pMEC46; titer of 9.8 x 10³ after the second boost). Remarkably, a very high and significant IgG response (titer of 1.4 x 10⁵) was elicited already after the first administration (priming) in the group of mice that received the recombinant mutant strain (MG1363 DltD-/ pMEC46) whereas no significant response could be detected for the recombinant wild type strain (NZ3900/ pMEC46). Moreover, intra-gastric administration of the recombinant mutant strain (MG1363 DltD-/ pMEC46) induced significant local IgA response (p < 0.005) in intestinal lavages (as described in Example 3) while no IgA response could be detected in mice that received the wild type recombinant strain (NZ3900/ pMEC46) (Figure 12).

These results demonstrate that the use of different cell wall mutants as antigen delivery vehicles promotes a substantial increase in the immune response and that this observation is not restricted to the Air- phenotype, as demonstrated with the Dlt- cell wall mutant of Lc. lactis.

Example 14: Lb. plantarum NCIMB8826 DltB- mutant and wild type strains

The Lb. plantarum NCIMB8826 DltB- mutant (EP007) deposited on June 16, 2003 at the Belgian Coordinated Collections of Microorganisms (BCCM), was obtained by using a suicide knockout vector containing an internal fragment of the dltB gene by single-step homologous recombination. Disruption of the locus was confirmed by PCR and Northern blotting and by an in vitro radioactive D-Alanine incorporation test (data not shown). Wild type and mutant strains were grown at 37°C in MRS broth (Difco). Erythromycin (5μg/ml) was added for the mutant NCIMB8826 DltB- strain. Example 15: In vitro evaluation of the immunomodulation capacity of the wild type and the DltB- mutant strains of Lb. plantarum NCIMB8826

L. plantarum NCIMB8826 wild type and DltB- mutant (NCIMB8826 DltB-) strains were grown in MRS medium for 48h and the cells were harvested by centrifugation, washed twice in sterile PBS pH 7.2, resuspended at 10⁹ CFU / ml in PBS containing 20% glycerol and stored at -80°C until used for stimulation assays.

Peripheral Blood Mononuclear Cell (PBMC), were isolated from peripheral blood of healthy donors as previously described (Muller-Alouf, et al. 2001. Infect Immun 69, no.

6:4141.). Briefly, after Ficoll gradient centrifugation (Pharmacia, Uppsala, Sweden), the mononuclear cells were collected at the interphase, washed twice with RPMI 1640 medium (Live technologies, Paisley, Scotland) and adjusted to 2 x 10^ cells per ml in RPMI 1640 supplemented with gentamycin (8 mg / ml), 2 mM L-glutamine, and 10% foetal calf serum (Gibco).

Monocytes (CD14⁺ cells) were purified by using CD14 antibodies coupled to magnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany), according to the manufacturer's recommendations. Briefly, PBMC (300 x 10⁶ cells) were incubated on ice for 30 min with 300 μl of CD14 antibodies coupled to magnetic microbeads. After washings they were applied onto a column placed in magnetic field of a MidiMACS separation unit (Miltenyi Biotec). After elution of the column, the cells obtained by positive selection were adjusted at 0.5 X 10⁶ cells per ml in complete RPMI medium.

Induction of cytokine release was performed by seeding PBMC (2 X 10⁶ cells/ ml) or monocytes (0.5 X 10⁶ cells/ml) in 24-well tissue culture plates. Ten μl of thawed bacterial suspension at 10⁹ bacteria/ml were then added. PBS containing 20% glycerol was used as negative (non stimulated) control. After 24h stimulation at 37°C in 5% CO2, culture supernatants were collected, clarified by centrifugation and stored at -20°C until cytokine analysis. Cytokines were measured by ELISA using BD Pharmingen (San Diego, Ca, USA) antibody pairs for TNFα, IL10 and IFNγ and the Diaclone Eli-pair (Besancon, France) for IL12, according to the respective manufacturer's recommendations.

The Lb. plantarum NCIMB8826 wild type and DltB- mutant strains differentially induced IL10, IL12, TNFα, and INFγ secretion from stimulated PBMC and monocytes. The effect of the DltB- mutant strain on cytokine release after PBMC stimulation was compared to that of the wild type strain (Fig 13). As compared to the wild type strain, the Dlt- mutant strain induced largely increased levels of IL10 secretion by stimulated PBMC, altogether with a significantly decreased release of the Th1 /pro-inflammatory cytokines (IFNγ, IL12 and TNFα). The effect of the wild type and Dlt- mutant Lb. plantarum strains was evaluated on monocyte stimulation, since among PBMC, monocytes are the most likely candidates for the production of IL12 (after bacterial stimulation). Similar results as those obtained with PBMC were obtained for the monocytes (Fig 14).

Human PBMC and monocytes were purified and stimulated with the wild type or the mutant Dlt- Lb. plantarum strains at a concentration of 1 x 10⁸ CFU/ml. IL12 and IL10 concentrations were measured by ELISA in the PBMC culture supernatant after 24h. The results are represented as the ratio of IL10/ IL12 taking the mean of each set of concentrations into account. The ratio of IL10 to IL12 production (Table 3) for the wild type strain was equal to 1.14 for PBMCs and 3.45 for monocytes. Notably, the ratio of IL10 to IL 12 production was dramatically higher for the DltB- mutant (160.8 for PBMC, 122 for monocytes) than for the wild type strain.

Table 3: IL10/IL12 ratio in the cytokine response of human PBMC and monocytes to thewild type and Dlt- mutant strains of Lb. plantarum.

IL10/ IL12 PBMC Monocytes

NCIMB8826 1.14 3.45

NCIMB8826 Dlt- 160.8 122

This result clearly illustrates the profound difference in the immunomodulation capacity of the DltB- mutant as compared to the Lb. plantarum parental strain, leading for the mutant to a substantial improvement of its anti-inflammatory properties associated with a dramatic decrease of the pro-inflammatory effects.

Example 16: Analysis of the beneficial effect of the DltB- mutant of Lb. plantarum NCIMB8826 in a murine colitis model.

Adult Balb/c mice from Iffa Credo (St Germain sur I'Arbresle, France) were used in the murine colitis model. For induction of colitis, anaesthetized mice received an intrarectal administration of 40 μl of a solution of trinitrobenzene sulphonic acid (TNBS) (100 mg/kg,

Fluka, Saint Quentin Fallavier, France) dissolved in 0.9% NaCI/ethanol (60/40 v/v). Control mice received 40% ethanol. Animals were killed by cervical dislocation 2 days after TNBS administration.

L. plantarum NCIMB8826 wild type and DltB- mutant strains were grown for 48h, washed and collected by centrifugation before resuspension at 10⁸ CFU / ml in a 0.2 M NaHCO₃ buffer containing 2% glucose. In order to study the effect of the wild type and DltB- mutant strains on colitis, 10⁷ CFU of bacteria were administered once daily by oral gavage, starting 5 days before and until 1 day after TNBS administration. Mice were weighed before colitis induction and immediately after sacrifice by cervical dislocation. Comparisons between the different animal groups were analyzed by the nonparametric Wilcoxon analysis of variance (ANOVA) test.

Macroscopic lesions of the colon were evaluated according to the Wallace criteria (Wallace et al. 1989, Gastroenterology 96: 29-36), reflecting both the intensity of inflammation and the extend of the lesions. A colon sample located in the most injured area was fixed in 4% paraformaldehyde acid and embedded in paraffin. Four micrometers-sections were stained with May Grϋnwald-Giemsa and blindly scored according to the Ameho criteria (Ameho, OK., et al. 1997. Gut 41:487-493). This grading on a scale from 0 to 6 takes into account the degree of inflammatory infiltrate, the presence of erosion, ulceration or necrosis, and the depth and surface extension of the lesion.

Notably, the percentage of mice displaying diarrhea after TNBS administration was much lower in the group of mice gavaged with the DltB- mutant (11,1%) than in the group of mice receiving the Lb. plantarum wild type strain or no bacteria (55,5% in both cases) (data not shown). Moreover, the weight loss was significantly reduced in mice receiving the Dlt- mutant as compared with mice receiving no bacteria or the wild type strain (Fig 15A).

Control mice receiving ethanol 50% intrarectally displayed neither macroscopic nor histological lesions. A mild colitis was observed 2 days after administration of TNBS in mice receiving no bacteria, leading to a Wallace score of 3.5 ± 0.5 (Fig. 15B) which corresponded to several areas of ulceration accompanied by intestinal wall thickening. Histological analysis performed on these mice revealed large areas of ulceration with important inflammatory infiltrates together with necrosis, leading to an Ameho score of 4.4 ± 0.6 (Fig. 15C). A milder but not significantly different level of inflammation (macroscopic and histological analyses) was observed in mice gavaged with the wild type Lb. plantarum bacteria (Figs. 15B and 15C). In contrast, mice gavaged with the DltB- mutant of Lb. plantarum, displayed less severe lesions. At the macroscopic level, only one mouse displayed necrotic lesions, whereas the others displayed only hyperaemia and thickening of the intestinal wall (Wallace score of 1.6 ± 0.6) (Fig 15B). These results were confirmed by histological observations (Ameho score of 2.4 ± 0.5, corresponding to only mild inflammatory infiltrates and limited oedema (Fig 15C)). The beneficial effect of the DltB- Lb. plantarum strain in the TNBS-induced colitis was similarly obtained with a lower bacterial dose, i.e. 10⁸ CFU per mouse per gavage.

In summary, treatment of mice with the Lb. plantarum NCIMB8826 DltB- mutant before induction of the colitis reduced the severe loss of body weight associated with TNBS administration. This correlated to the significantly lower inflammation scores of mice which received the DltB- mutant strain before TNBS administration as compared to mice gavaged with the wild type strain or receiving no bacteria. These experiments demonstrate that the NCIMB8826 DltB- mutant strain can be successfully used as an effective preventive treatment against intestinal inflammation.

Therefore cell wall mutant strains can be used as effective agents for the treatment of various inflammatory diseases such as for example inflammatory bowel disease and the like.

Claims

1. Recombinant gram-positive bacterium comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and comprising an expression vector leading to the intracellular production of a polypeptide having a prophylactic or therapeutic activity.

2. Recombinant gram-positive bacterium as defined in claim , wherein said bacterium is a food grade bacterium, preferably a lactic acid bacterium, more preferably a Lactobacillus or Lactococcus strain.

3. Recombinant gram-positive bacterium as defined in claim 2, wherein said bacterium is a Lactobacillus plantarum or a Lactococcus lactis strain.

4. Recombinant gram-positive bacterium as defined in any of claims 1 to 3, wherein said mutation is in a gene involved in the biosynthesis, modification or degradation of peptidoglycan, teichoic acids, lipoteichoic acids, teichuronic acids, or cell wall associated proteins.

5. Recombinant gram-positive bacterium as defined in claim 4 wherein said gene encodes alanine racemase, or aspartate racemase, or glutamate racemase, or (poly)glycerolphosphate/(poly)ribitolphosphate D-alanyltransferase, or (poly)glycerolphosphate/(poly)ribitolphosphate glucosyltransferase.

6. Recombinant gram-positive bacterium as defined in claim 5 wherein said gene encodes alanine racemase.

7. Recombinant gram-positive bacterium as defined in any of claims 1 to 6, wherein said polypeptide produced by said bacterium is selected from the group comprising any therapeutic or prophylactic compound selected from the group comprising antigens, epitopes, allergens, immune modulators including adjuvants, enzymes, receptor ligands or variants thereof.

8. Use of a recombinant gram-positive bacterium according to any of claims 1 to 7 as a medicament.

9. Method for the delivery of polypeptides at mueosal surfaces, comprising administering to a mueosal surface of an individual (human or animal) a composition comprising a recombinant gram-positive bacterium comprising a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expressing a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component and wherein the polypeptide to be delivered is produced by said recombinant bacterium.

10. Use of a recombinant gram-positive bacterium as defined in any of claims 1 to 7 for the manufacture of a medicament for prophylactic or therapeutic application to treat diseases selected from the group comprising infectious diseases, chronic inflammation, allergy, auto-immune diseases, metabolic defects or cancers.

11. Use of a mutant gram-positive bacterium as a vehicle for delivery of a polypeptide at mueosal surfaces characterized in that said mutant bacterium comprises a mutation modulating the expression of a gene involved in the biosynthesis, modification or degradation of a cell wall component or expresses a foreign gene interfering with the biosynthesis, modification or degradation of a cell wall component.

12. Use of a mutant gram-positive bacterium as defined in claim 11 as a vaccine.

13. Use of a mutant gram-positive bacterium as defined in claim 12, wherein said mutant bacterium is further expressing a nucleic acid sequence encoding an antigen which is able to elicit an immune response when administered to a human or animal host.

14. Use of a mutant gram-positive bacterium as defined in any of claims 11 to 13, wherein said bacterium is a food grade bacterium, preferably a lactic acid fermenting bacterium, most preferably a Lactobacillus or Lactococcus strain chosen from the Lactobacillus plantarum or Lactococcus lactis species.

15. Use of a mutant gram-positive bacterium according to any of claims 11 to 14, wherein said cell wall component is peptidoglycan or teichoic acid or lipoteichoic acid or a cell wall associated protein.

16. Use of a mutant gram-positive bacterium according to any of claims 11 to 15, further characterized in that said mutation is in a gene involved in the biosynthesis of D- alanine, or D-glutamate, or D-aspartate, or in the D-alanylation or glucosylation of teichoic and lipoteichoic acids.

17. Use of a mutant gram-positive bacterium according to claim 16, wherein said gene encodes alanine racemase, or glutamate racemase, or aspartate racemase, or (poly)glycerolphosphate/(poly)ribitolphosphate D-alanyltransferase, or

(poly)glycerolphosphate/(poly)ribitolphosphate glucosyltransferase.

18. Method for the preparation of a vaccine composition comprising mixing a recombinant gram-positive bacterium as defined in any of claims 1 to 7 with a pharmaceutically acceptable carrier.

19. Method for the preparation of a vaccine composition comprising processing a recombinant gram-positive bacterium as defined in any of claims 1 to 7 as in use in the food fermentation technology.

20. Use of a recombinant gram-positive bacterium as defined in any of claims 1 to 7, as a vehicle for delivery of therapeutic or prophylactic compounds or polypeptides at mueosal surfaces.