WO1999029342A1 - Vaccines containing attenuated bacteria - Google Patents

Abstract

The invention relates to a vaccine comprising a bacterium attenuated by a non-reverting mutation in a gene encoding a protein which promotes folding of extracytoplasmic proteins. Such mutations were intially identified as being useful in vaccines from a bank of randomly inserted, transposon mutants in which attenuation was determined as a reduction in virulence of the organism in the mouse model of infection. Site directed mutation of the gene results in a strain which shows at least 4 logs of attenuation when delivered both orally and intravenously. Animals vaccinated with such a strain are protected against subsequent challenge with the parent wild type strain. Finally, heterologous antigens such as the non-toxic and protective, binding domain from tetanus toxin, fragment C, can be delivered via the mucosal immune system using such strains of bacteria. This results in the induction of a fully protective immune response to subsequent challenge with native tetanus toxin.

Description

VACCINES CONTAINING ATTENUATED BACTERIA

The invention relates to vaccines containing attenuated bacteria.

Background to the invention

The principle behind vaccination is to induce an immune response in the host thus providing protection against subsequent challenge with a pathogen. This may be achieved by inoculation with a live attenuated strain of the pathogen (i.e. a strain having reduced virulence such that it does not cause the disease caused by the virulent pathogen).

Classically, live attenuated vaccine strains of bacteria and viruses have been selected using one of two different methodologies. Mutants have been created either by treatment of the organism using mutagenic chemical compounds or by repeated passage of the organism in vitro. However, use of either method gives rise to attenuated strains in which the mode of attenuation is unclear. These strains are particularly difficult to characterize in terms of possible reversion to the wild type strain as attenuation may reflect single (easily reversible) or multiple mutation events.

Using modern genetic techniques, it is now possible to construct genetically defined attenuated bacterial strains in which stable attenuating deletions can be created. A number of site directed mutants of Salmonella have been created using this type of technology (2, 5, 6, 12, 22, 35, 36, 37). Amongst the most comprehensively studied attenuating lesions are those in which mutations in the biosynthetic pathways have been created, rendering the bacteria auxotrophic (e.g. aro genes). Mutations in these genes were described as early as 1950 (1) as responsible for rendering Salmonella less virulent for mice. Several different auxotrophic mutations such as galE, aro A or pur A have also been described previously (6, 12). Salmonella aroA mutants have now been well characterised and have been shown to be excellent live vaccines against salmonellosis in several animal species. In addition, in order to reduce the chances of a reversion to virulence by a recombination event mutations have now been introduced into two independent genes such as aroAlpurA and aroAlaroC. Identical mutations in host adapted strains of Salmonella such as S. typhi (man) and S.dublin (cattle) has also resulted in the creation of a number of single dose vaccines which have proved successful in clinical (11, 17) and field trials (15).

In animal studies, attenuated S. typhimurium has been used as a vehicle for the delivery of heterologous antigens to the immune system (3, 8, 32). This raises the potential of the development of multivalent vaccines for use in man (9).

Summary of the invention

The original aim of the work that led to the invention was the identification of novel genes that are involved in the virulence pathways of pathogenic bacteria, the identification and deletion of which may render the bacteria avirulent and suitable for use as vaccines. To identify attenuating lesions, random mutations were introduced into the chromosome of S. typhimurium using the transposon TrφhoA (18). This transposon is unique in that it is engineered to identify proteins that are expressed in or at the bacterial outer membrane; such proteins may be those involved in interaction with and uptake by host tissues. By using the natural oral route of infection to screen these mutants, those with important, in vivo induced, attenuating lesions in genes were identified.

One such gene identified through this work is surA. The surA gene product is known to promote the folding of extracytoplasmic proteins. Accordingly, the invention provides a vaccine comprising a pharmaceutically acceptable carrier or diluent and a bacterium attenuated by a non-reverting mutation in a gene encoding a protein which promotes the folding of extracytoplasmic proteins. The vaccine has the ability to confer protection against a homologous wild type oral challenge with the virulent bacterium. In addition, the bacterium used in the vaccine can act as a carrier for heterologous antigens such as fragment C of tetanus toxin.

Detailed description of the invention

Proteins that promote the folding of extracytoplasmic proteins

Periplasmic and outer membrane proteins are secreted across the cytoplasmic (inner) membrane in a mostly unfolded state, and they then fold after secretion. The folding often has enzymatic assistance to catalyse the formation of bonds necessary for the protein to reach its folded state. For example, the folding often requires the participation of enzymes that catalyse the formation of disulphide bonds or enzymes that catalyse the isomerisation of prolyl bonds (peptidyl-prolyl cis-trans isomerases or PPiases).

One known PPiase is SurA. The inventors have now shown that mutation of the surA gene causes attenuation of virulent bacteria and that the attenuated bacteria are useful as vaccines.

SurA was first described as being essential for the survival of E.coli in the stationary phase (33). It is a periplasmic protein. More recently, SurA has been described as belonging to a third, new family of PPiases (30), the parvulin family. Further studies have shown SurA to be involved in the correct folding of outer membrane proteins such as OmpA, OmpF, and LamB (16, 24, 29).

PPiases are divided into three families, the cyclophilins, FK506-binding proteins (FKBPs) and parvulins. Members of all three families have been found in E.coli. Apart from SurA, the parvulin family includes several proteins such as NifM, PrsA and PrtM. Bacteria useful in the invention

The bacteria that are used to make the vaccines of the invention are generally those that infect via the oral route. The bacteria may be those that invade and grow within eukaryotic cells and/or colonise mucosal surfaces. The bacteria are generally Gram-negative.

The bacteria may be from the genera Salmonella, Escherichia, Vibrio, Haemophilus, Neisseria, Yersinia, Bordetella or Brucella. Examples of such bacteria are Salmonella typhimurium - the cause of salmonellosis in several animal species; Salmonella typhi - the cause of human typhoid; Salmonella enteritidis - a cause of food poisoning in humans; Salmonella choleraesuis - a cause of salmonellosis in pigs; Salmonella dublin - a cause of both a systemic and diarrhoel disease in cattle, especially of new-born calves; Escherichia coli - a cause of diarrhoea and food poisoning in humans; Haemophilus influenzae - a cause of meningitis; Neisseria gonorrhoeae - a cause of gonnorrhoeae; Yersinia enterocolitica - the cause of a spectrum of diseases in humans ranging from gastroenteritis to fatal septicemic disease; Bordetella pertussis - the cause of whooping cough; or Brucella abortus - a cause of abortion and infertility in cattle and a condition known as undulant fever in humans.

Salmonella bacteria are particularly useful in the invention. As well as being vaccines in their own right against infection by Salmonella, attenuated Salmonella can be used as carriers of heterologous antigens from other organisms to the immune system via the oral route. Salmonella are potent immunogens and are able to stimulate systemic and local cellular and antibody responses. Systems for driving expression of heterologous antigens in Salmonella in vivo are known; for example the nirB and htrA promoters are known to be effective drivers of antigen expression in vivo.

The invention is also particularly applicable to E.coli, especially exterotoxigenic E.coli ("ETEC"). ETEC is a class of E.coli that cause diarrhoea. They colonise the proximal small intestine. A standard ETEC strain is ATCC HI 0407.

Infections of ETEC are the single most frequent cause of travellers diarrhoea, causing 3-9 million cases per year amongst visitors to developing countries. In endemic areas, ETEC infections are an important cause of dehydrating diarrhoea in infants and young children, resulting in 800,000 deaths a year in the under fives wold- wide. In developing countries, the incidence of ETEC infections leading to clinical disease decreases with age, indicating that immunity to ETEC infection can be acquired. In contrast, naive adults from industrialized countries who visit endemic areas are highly susceptible to ETEC infections. However, with prolonged or repeated visits to endemic areas susceptibility to ETEC infections diminishes, suggesting that a live attenuated approach to ETEC vaccination may prove successful.

Seq. Id. No. 1 shows the sequence of the surA open reading frame in Salmonella typhimurium, and Seq. Id. No. 2 shows the sequence of the surA open reading frame in E.coli.

Second mutations

The bacteria used in vaccines of the invention preferably contain a mutation in one or more genes in addition to the mutation in the gene encoding a protein which promotes folding of extracytoplasmic proteins. This is so that the risk of the bacterium reverting to the virulent state is minimised, which is clearly important for the use of the bacterium as a human or animal vaccine. Although bacteria containing only a mutation in a protein which promotes folding of extracytoplasmic proteins are attenuated and the risk of reversion is small, it will generally be desirable to introduce at least one further mutation so as to reduce the risk of attenuation yet further. A number of genes that are candidates for second and further mutations are known (see e.g. ref 39). These include the aro genes (35), the pur genes, the htrA gene (37), the ompR gene (36), the galE gene, the cya gene, the crp gene or the phoP gene. The aro gene may be aro A, aroC, aroD or aroE. The pur gene may he pur A, purB,purE orpurH. The use of aro mutants, especially double aro mutants, is preferred because such mutants have been shown to be particularly effective as vaccines. Suitable combinations of aro mutations are aroAaroC, aroAaroD and aroAaroE.

The nature of the mutation

The mutations introduced into the bacterial vaccine generally knock-out the function of the gene completely. This may be achieved either by abolishing synthesis of any polypeptide at all from the gene or by making a mutation that results in synthesis on non-functional polypeptide. In order to abolish synthesis of any polypeptide, either the entire gene or its 5 '-end may be deleted. A deletion or insertion within the coding sequence of a gene may be used to create a gene that synthesises only non-functional polypeptide (e.g. polypeptide that contains only the N-terminal sequence of the wild-type protein). In the case of mutations in genes encoding proteins which promote the folding of extracytoplamic proteins, the mutation generally abolishes the ability of the protein to promote such protein folding.

The mutations are non-reverting mutations. These are mutations that show essentially no reversion back to the wild-type when the bacterium is used as a vaccine. Such mutations include insertions and deletions. Insertions and deletions are preferably large, typically at least 10 nucleotides in length, for example from 10 to 600 nucleotides.

The bacterium used in the vaccine preferably contains only defined mutations, i.e. mutations which are characterised. It is clearly undesirable to use a bacterium which has uncharacterised mutations in its genome as a vaccine because there would be a risk that the uncharacterised mutations may confer properties on the bacterium that cause undesirable side-effects.

The attenuating mutations may be constructed by methods well known to those skilled in the art (see ref 31 ). One means for introducing non-reverting mutations into extracytoplamic proteins is to use transposon TrφhoA. This can be introduced into bacteria to generate enzymatically active protein fusions of alkaline phosphatase to extracytoplasmic proteins. The TnphoA transposon carries a gene encoding kanamycin resistance. Transductants are selected that are kanamycin resistant by growing colonies on an appropriate selection medium.

Alternative methods include cloning the DNA sequence of the wild-type gene into a vector, e.g. a plasmid or cosmid, and inserting a selectable marker into the cloned DNA sequence or deleting a part of the DNA sequence, resulting in its inactivation. A deletion may be introduced by, for example, cutting the DNA sequence using restriction enzymes that cut at two points in the coding sequence and ligating together the two ends in the remaining sequence. A plasmid carrying the inactivated DNA sequence can be transformed into the bacterium by known techniques. It is then possible by suitable selection to identify a mutant wherein the inactivated DNA sequence has recombined into the chromosome of the bacterium and the wild-type DNA sequence has been rendered non-functional in a process known as homologous recombination.

Expression of heterologous antigens

The attenuated bacterium used in the vaccine of the invention may be genetically engineered to express an antigen from another organism (a "heterologous antigen"), so that the attenuated bacterium acts as a carrier of the antigen from the other organism. In this way it is possible to create a vaccine which provides protection against the other organism. A multivalent vaccine may be produced which not only provides immunity against the virulent parent of the attenuated bacterium but also provides immunity against the other organism. Furthermore, the attenuated bacterium may be engineered to express more than one heterologous antigen, in which case the heterologous antigens may be from the same or different organisms.

The heterologous antigen may be a complete protein or a part of a protein containing an epitope. The antigen may be from another bacterium, a virus, a yeast or a fungus. More especially, the antigenic sequence may be from tetanus, hepatitis A, B or C virus, human rhinovirus such as type 2 or type 14, herpes simplex virus, polio virus type 2 or 3, foot-and-mouth disease virus, influenza virus, coxsackie virus or Chlamydia trachomatis. Useful antigens include E.coli heat labile toxin B subunit (LT-B), E.coli K88 antigens, P.69 protein from B. pertussis and tetanus toxin fragment C.

The DNA encoding the heterologous antigen is expressed from a promoter that is active in vivo. Two good promoters are the nirB promoter (38, 40) and the htrA promoter (40).

A DNA construct comprising the promoter operably linked to DNA encoding the heterologous antigen may be made and transformed into the attenuated bacterium using conventional techniques. Transformants containing the DNA construct may be selected, for example be screening for a selectable marker on the construct. Bacteria containing the construct may be grown in vitro before being formulated for administration to the host for vaccination purposes.

Formulation of the vaccine

The vaccine may be formulated using known techniques for formulating attenuated bacterial vaccines. The vaccine is advantageously presented for oral administration, for example in a lyophilised encapsulated form. Such capsules may be provided with an enteric coating comprising, for example, Eudragate "S" (Trade Mark), Eudragate "L" (Trade Mark), cellulose acetate, cellulose phthalate or hydroxypropylmethyl cellulose. These capsules may be used as such, or alternatively, the lyophilised material may be reconstituted prior to administration, e.g. as a suspension. Reconstitution is advantageously effected in a buffer at a suitable pH to ensure the viability of the bacteria. In order to protect the attenuated bacteria and the vaccine from gastric acidity, a sodium bicarbonate preparation is advantageously administered before each administration of the vaccine. Alternatively, the vaccine may be prepared for parenteral administration, intranasal administration or intramuscular administration.

The vaccine may be used in the vaccination of a host, particularly a human host but also an animal host. An infection caused by a microorganism, especially a pathogen, may therefore be prevented by administering an effective dose of a vaccine prepared according to the invention. The dosage employed will be dependent on various factors including the size and weight of the host and the type of vaccine formulated. However, a dosage comprising the oral administration of from 10⁷ to 10" bacteria per dose may be convenient for a 70kg adult human host.

Examples

The following Examples serve to illustrate the invention.

Brief description of the drawings

Figure 1 : Southern blot confirming the defined deletion created within surA in the strain BRD 1115. Lanes 1 and 2 have been restricted using the enzyme Pstl, lanes 3- 10 have been restricted with Sail. The filters have been probed using a 500 bp PCR product that contains a 500 bp fragment from the middle of the surA gene. Lanes 2 and 4 show hybridisation of this probe to a band 500 bp smaller than the corresponding wild type lanes 1 and 3. The transposon mutant BRD441 shows hybridisation to 2 bands since the enzyme Sail cuts the transposon into two. HB101 shows no hybridisation whilst the other wild type Salmonella strains show the same hybridisation as C5 when restricted with Sail.

Figure 2: This figure shows the colonisation and persistence of BRDl 115, BRD441 and the wild type C5 in the mesenteric lymph nodes (top left graph), Peyer's patches (bottom right), spleens (bottom left) and livers (top right) in BALB/c mice following oral inoculation. The x-axis is time in days and the y-axis is log₁₀ CFU/ml (CFU stands for colony forming units).

Figure 3: Three strains were constructed to evaluate the ability of mutant Salmonella strains to deliver the heterologous antigen Fragment C in the mouse. BRDl 115 is the parental strain. Two plasmids encoding the Fragment C gene of tetanus toxin under the control of either the htrA or nirB promoter were introduced into the strain BRDl 115 to give the strains BRDl 127 and 1126 respectively. Expression of fragment C was determined in vitro by Western blotting. These strains were then used in an in vivo experiment using BALB/c mice. Groups of 10 mice were immunised orally with log₁₀8 organisms each of the 3 strains. Serum samples were taken weekly and analysed for total antibodies against tetanus toxin fragment C. The titres of anti-fragment C were determined as the highest sample dilution giving an absorbance value of 0.3 above normal mouse serum. The highest sample dilution tested was 1/6250. All mice immunised with BRD 1126 showed antibody titres higher than 6250.

Figure 4: Schematic showing a plasmid map of pLG339/swr/4.

Figure 5: Graph showing the survival of Balb/c mice following oral challenge with log₁₀8 bacteria of the three strains C5, BRDl 115 and K2. Example 1

This Example shows the identification of mutations in surA as attenuating mutations, the construction of a defined surA mutation and the evaluation of a surA mutant as a vaccine (both against homologous challege and as a carrier for heterologous antigens).

Materials and methods

1.1 Bacteria, bacteriophage, plasmids and growth conditions

The bacteria used in this study are listed in Table 1. Bacteria were routinely cultured on L-agar or in L-broth containing lOOμg/ml ampicillin or 50μg/ml kanamycin where appropriate. The bacteriophage P22HT105/linf is a high frequency transducing bacteriophage obtained from Dr Tim Foster (Trinity College, Dublin). The plasmid pGEM-T (Promega Corporation, USA) is designed for direct cloning of PCR fragments and pBluescript^όII SK+ (Stratagene Ltd, Cambridge, U.K.) is a general cloning vector. The other plasmids are described in the text.

1.2 Purification of DNA and DNA manipulation techniques All DNA manipulation including Southern blotting were carried out as described by Sambrook et al (31). Restriction enzymes and T4 DNA ligase were purchased from Boehringer Mannheim (Lewes, UK) and used according to the manufacturers instructions. Chromosomal DNA preparation was prepared according to the method of Hull (13).

1.3 DNA sequencing

Double stranded plasmid sequencing was carried out using the Sequenase kit (Trade Mark, United States Biochemical Corporation) according to the manufacturers' instructions. Labelling of the DNA was achieved using ³⁵S-dATP (Amersham, UK) and fragments separated on an 8% acrylamide/bis-acrylamide gel containing 7M urea, for 2hours at 35 mA. 1.4 DNA amplification by polymerase chain reaction

Polymerase chain reactions (PCR) were carried out with Taq DNA polymerase using the GeneAmp kit (Trade Mark, Perkin Elmer Cetus, USA) according to the manufacturers' instructions. Oligonucleotides were purchased from 5 the Molecular Medicine Unit, Kings College, London and the sequences are shown in Table 1. Mixtures of DNA and specific primers were subjected to multiple rounds of denaturation, annealing and extension in the presence of the enzyme Taq polymerase. 100 ng plasmid DNA and lmg chromosomal DNA were added to a mixture containing 5μl 10 x buffer (lOOmM Tris-HCl, pH 8.3: 500mM KC1; 15mM Mg Cl₂;

10 0.01% gelatine(v/v)); 8μl of deoxy-nucleotide mixture (1.25mM each of deoxy- nucleotide triphosphate; dATP, dCTP, dGTP and dTTP); lμl of a lOμM sense primer; lμl of a lOμM anti-sense primer and 2.5 units Taq polymerase. This mixture is overlaid with 50μl light mineral oil (Sigma) to prevent evaporation and the tubes incubated in an Omnigene Thermal Cycler (Trade Mark, Hybaid). Amplification of

15 the DNA was performed using the following programme: 1 cycle of 95°C for 5 minutes, 50°C for 1.5 minutes, 74°C for 2 minutes; 19 cycles of 95°C for 1.5 minutes, 50°C for 2 minutes, 74^βC for 3 minutes; 10 cycles of 95°C 2 minutes, 50°C for 2 minutes, 74°C for 7 minutes.

20 1.5 Transformation of bacteria 1.5.1. Heat shock

Bacteria are rendered competent to DNA uptake by the calcium chloride method. An overnight bacterial culture was used to seed a fresh 25 ml LB broth culture (a 1 : 100 dilution) which was grown aerobically with shaking until the cells

25 reached mid-log growth phase (OD 650nm = 0.4 to 0.6). The cells were harvested by centrifugation at 3000 x g for 10 minutes at 4°C. The supernatant was discarded and the pellet resuspended in 25 ml ice-cold 75mM CaCl₂. The process was repeated and the cells incubated on ice for 30 minutes. The cells were pelleted by centrifugation at 3000 x g for 10 minutes at 4°C. The cell pellet was resuspended in 1.2 ml ice-cold

30 75mM CaCl₂ and stored on ice until needed. The cells were then competent to DNA uptake. A maximum of 20μl of the ligation mix was added to 200μl of the competent cells and the mixture stored on ice for 30 minutes. The cells were then subjected to heat shock by incubation in a 42 °C waterbath for 2 minutes. The cells were then transferred back to ice for a further 2 minutes. 1 ml of LB broth was added to the mixture and the cells incubated at 37°C for at least 60 minutes to allow 5 expression of the antibiotic marker on the plasmid. lOOμl aliquots of cells were plated onto LB agar plates containing the appropriate antibiotics and incubated overnight at 37°C.

1.5.2. Electroporation

10 Plasmid DNA was introduced into bacterial strains using electroporation.

Mid-log phase growth cultures were generated as for the heat-shock method and the cells pelleted by centrifugation at 3000 x g for 10 minutes at 4°C. The cell pellet was washed twice with an equal volume of ice-cold 10% glycerol and pelleted as before. The cell pellet was resuspended in 300-500μl ice-cold 10% glycerol. Approximately

15 100 ng plasmid (or 1 μg suicide vector) in a volume not greater than 6 μl sterile water was added to 60 μl competent cells in a pre-chilled electroporation cuvette on ice. The plasmid was electroporated into the bacteria using a Bio-Rad Gene Pulser (Trade Mark) with the following conditions 1.75kV, 600Ω, 25μF. 1ml LB broth was then added to the contents of the electroporation cuvette and the mixture incubated at 37°C

20 for 90 minutes to allow the cells to recover. 100 μl aliquots of the electroporation mix were plated out onto selection media and incubated at 37°C overnight.

1.6 P22 Transduction

Transduction experiments were carried out using the bacteriophage P22 25 HT105/1 inf. Phage lysates were prepared using LB5010 as the donor strain. A 5ml overnight culture of LB5010 was grown in L broth containing 0.2% glucose and galactose to increase the expression of phage receptors on the cell surface. Ten fold serial dilutions of the P22 stock were made in TMGS up to 10"8 (stock is approximately 10l0pfti ml). lOμl of each dilution was added to lOOμl of the 30 overnight stock of cells and incubated at 37°C for 30-45 minutes to allow adsorbtion of the phage to the cells. 3mls of top agar was added to each incubation and spread onto L agar plates containing lOOμg/ml ampicillin. The plates were incubated at 37°C for approximately 4-5 hours until plaques were visible. The dilution that gave almost confluent plaques after this length of time was the one chosen for harvesting. The plaques were harvested by scraping the top agar into 2ml of phage buffer with a glass microscope slide. A few drops of chloroform were added and the phage stock stored at 4°C until needed. The recipient strain C5 was grown during the day in L broth at 37°C until late log/stationary phase, lμl, 5μl, lOμl, 20μl, and 50μl aliquots of the new phage stock were added to lOOμl aliquots of the recipient strain and incubated at 37°C for 1 hour. The cells were then spread onto L agar ampicillin plates containing 5mM EGTA (to prevent phage replication) and incubated at 37°C overnight. Colonies were replated onto L agar ampicillin plates containing 5mM EGTA three times to ensure that they were free from phage. The colonies no longer had a jagged appearance thus indicating an absence of phage.

1.7 In vitro analysis of bacterial strain 1.7.1. Agglutination with antisera

Agglutination using anti-sera raised against the O antigen of Salmonella can be used as a rapid test not only for the integrity of the bacterial LPS but also as a diagnostic of the strain, e.g. anti-sera against the 04 and 05 antigens for S. typhimurium. These were obtained from Murex Diagnostics Ltd (Dartford U.K.). A sweep of colonies was harvested from the growth on a plate incubated overnight, and resuspended in lOOμl PBS. This sample was mixed with a drop of antisera on a glass slide and the agglutination compared with a positive and negative sample.

1.7.2 HEp-2 invasion assay

The HEp-2 cell line is an adherent epidermoid carcinoma derived from human larynx (ATCC CCL23). It can be cultured as a monolayer in Dulbecco's modified Eagle's medium with 10% FCS, glutamine and penicillin streptomycin at 37°C in the presence of 5% CO2- Confluent cells were detached from the tissue culture flasks by the use of trypsin EDTA. The cells were first washed in PBS to remove any serum that might affect the action of the trypsin. Trypsin/EDTA was then added to the monolayer and the cells incubated at 37°C for 5 minutes. The cells were removed from the plastic by gentle tapping on the edge of the flask. The trypsin was neutralised with 1.5 volumes of DMEM. Cells are collected by centrifugation at 1000 x g for 5 minutes. The supernatant was removed and the cell pellet resuspended in DMEM. The cell pellet was counted and the concentration adjusted to give 2xl0⁵cells per ml.

1 ml of the cell suspension was added to one well of a 24 well tissue culture plate (Costar 3524), three wells for each bacterial strain being investigated. The cells were incubated overnight to form a confluent monolayer in the well. The cells were then washed 5 times with PBS to ensure removal of the antibiotics and 1 ml DMEM added (without any antibiotics). 1x10' bacteria were added to each well and incubated at 37°C for 3 hours. The cells were washed 3 times with PBS to remove any extracellular bacteria. 1ml of DMEM containing lOOμg/m gentamycin was added and the cells incubated for a further 1 hour. The cells were washed 5 times with PBS. The cells were lysed by the addition of 1ml of 0.1% Triton-X-100 at 37°C for 15 minutes. The cells were further lysed by agitation with a blue pipette tip and the lysate transferred to a 1.5ml centrifuge tube. The viable bacteria that had invaded the cells were counted using the Miles-Misra drop test method (19).

1.8. In vivo analysis of bacterial strains

1.8.1. Preparation of live bacteria for immunisation of mice.

A vial of the appropriate strain was thawed from liquid nitrogen and used to inoculate a 250 ml culture of LB broth containing antibiotic where appropriate. The culture was grown overnight at 37°C without shaking. The bacteria were harvested by centrifugation at 3000 x g for 10 minutes and washed once in sterile PBS. The bacteria were harvested again by centrifugation and resuspended in 5 ml sterile PBS. The concentration of bacteria was estimated by optical density at 650 nm using a standard growth curve for that strain. Based on this estimate the cell concentration was adjusted with PBS to that required for immunisation. A viable count was prepared of each inoculum to give an accurate number of colony forming units per ml (cfu/ml) administered to each animal.

1.8.2. Oral immunisation of mice with live bacteria. The mice were lightly anaesthetised with a mixture of halothane and oxygen and the bacteria administered by gavage in 0.2 ml volumes using a gavage needle attached to a 1ml syringe.

1.8.3. Intravenous (i.v.) immunisation of mice with live bacteria. Mice were placed in a warm chamber and 0.2 ml volumes injected into a tail vein of each mouse using a 27 gauge needle.

1.8.4. Enumeration of viable bacteria in mouse organs.

Groups of four or five mice were sacrificed up to 7 weeks post oral immunisation with three bacterial strains. Spleens, livers, mesenteric lymph nodes and Peyer's patches were removed and homogenised in 10ml sterile PBS using a stomacher (Colworth, U.K.). Dilutions of these homogenates were plated out in LB agar with kanamycin if required and incubated overnight at 37°C. The number of viable bacteria present in each homogenate was then calculated from the dilution.

1.9. Determination of antibody titres against fragment C.

Serum antibody responses against fragment C were measured by enzyme- linked immunosorbant assay (ELISA) as previously described (28) using 96 well EIA/RIA plates (Costar 3590). Absorbance values were read at A₄₉₀ and plotted against dilutions (data not shown). A normal mouse serum control was added to each ELISA plate and used to define the background level response.

1.10 Tetanus toxin challenge

Mice were challenged with 0.05 μg (50 x 50% lethal doses) of purified tetanus toxin as previously described (7), and fatalities recorded for 4 days. Results

2.1 Cloning and mapping of TnphoA insertion sites

A number of S. typhimurium TrφhoA insertion mutants were previously identified as being attenuated when administered orally to BALB/c mice. In addition some of these mutants also exhibited a reduced ability to invade the cultured epithelial cell line HEp-2. To identify the genes that had been disrupted by the TrφhoA insertion, genomic DNA was digested using Sau3A and cosmid banks prepared from each strain. These banks were screened using TiφhoA probes and cosmids exhibiting homology with the 3' and 5' probes were examined. Fragments from these cosmids were cloned into the vector pBluescript^όII SK+. The nucleotide sequence surrounding these insertion sites was determined and the genes identified. Two insertions were found to be within the htrA gene (14), one in the osmZ gene (10) and one in the surA gene.

The surA gene open reading frame of Salmonella typhimurium shown in Seq Id No. 1 is 1281 bases long, encoding a protein of some 427 amino acids with a molecular weight of 47.2Kd. This protein is virtually identical to that found in E.coli (34), and is described as being essential for survival in long term culture (33). The surA gene contains a leader peptidase cleavage site indicating that this is a transported protein. It has now been described as belonging to a peptidyl prolyl isomerase family, with a function to aid the correct folding of outer membrane proteins (16, 24, 29).

2.2 Introduction of a defined deletion into the surA gene.

Restriction analysis and DNA sequencing of the surA gene revealed the presence of single Hpal and Smal restriction enzyme sites within the coding region of the gene which could be used to generate a deletion of 400 bases. The plasmid pGEM-T/212/213 was constructed containing a 3Kb region encompassing the entire surA gene and flanking region. Digestion of the plasmid with the enzymes Hpal and Smal, gel purification of the large 5.5Kb fragment and re-ligation resulted in a plasmid containing a 419bp deletion within the surA gene. This plasmid was designated pGEM-T/ΔsurA.

2.3 Introduction of the surA deletion into the chromosome of S. typhimurium C5. The plasmid pGEM-T was digested with the two restriction enzymes Sphl and

Sail. The 2.6kb fragment containing the deleted surA gene was gel purified and ligated into the suicide replicon pGP704 that had previously been digested with the same enzymes. The suicide replicon pGP704 has been used previously to introduce deletions into the chromosome of S. typhi (4) and S. typhimurium (26) which lack the ir gene, the product of which is essential for the replication of pGP704. The ligation mix was used to transform the strain SY327, an E.coli strain that contains the pir gene, and a plasmid of the expected size identified by restriction analysis. This plasmid was designated pGP704/ΔsurA. Since suicide replicons cannot replicate in S. typhimurium the drug resistance marker is only expressed if there has been a single homologous recombination event, incorporating the plasmid into the bacterial chromosome.

The plasmid pGP704/ΔsurA was used to transform the semi-rough S.typhimurium strain LB5010 by the calcium chloride method. Three transformants were selected on agar containing ampicillin. These single crossovers were moved from this intermediate strain into the wild type C5 using P22 transduction (20). P22 lysates were prepared from the three transductants and introduced into C5. One ampicillin resistant colony was obtained from this process. This transformant was sub-cultured twice into L-broth containing no selection and grown for 48 hours. Serial dilutions of this culture were made and the 10^"6 dilution was spread onto L- agar plates containing no selection. 500 colonies were streaked by hand on to duplicate plates, one containing agar, the other agar with ampicillin. One colony was found to be ampicillin sensitive indicating the loss of the drug resistance marker of the plasmid following a second homologous recombination event.

This potential surA mutant was confirmed as a S.typhimurium strain by agglutination with 04 and 05 antiserum. The deletion was confirmed by PCR using the primers MGR92 and MGR93, giving a 1 kb product. The deletion was also confirmed cloning the PCR product into the vector pGEM-T to give the plasmid pGEM-T/92/93, and sequencing across the deletion using the primers MGR130 and 5 135. Figure 1 shows the results of probing Pstl and Sail digested genomic DNA from C5 and the surA mutant strain with a PCR product obtained from the wild type C5. The band seen in the surA mutant track is approximately 400 bases smaller than that seen in the wild type. This deleted strain was designated BRDl 115.

10 2.4 Characterisation of the strain BRD 1115

2.4.1 In vitro analysis of the invasion of cultured epithelial cells

The strain BRDl 115 was tested for its ability to invade the cultured epithelial cell line HEp-2. The levels of invasion were found to be reduced by 80% in comparison to the wild type strain C5. The transposon mutant BRD441 showed a

15 90%) reduction in invasion compared to C5.

2.4.2. Evaluation of the in vivo properties of BRDl 115 in BALB/c mice.

2.4.2.1. Determination of oral and i.v. LD50's

The oral and i.v. LD ₅₀'s of BRD 1115, C5 and BRD441 were calculated 20 using the mouse susceptible strain BALB/c. 5 mice per group were inoculated either orally or i.v. with doses ranging from log₁₀4 to log₁₀ 10 orally and log₁₀l to log₁₀5 i.v. Deaths were recorded over 28 days and the LD₅₀'s calculated by the method of Reed and Meunch (27). BRDl 115 was determined to show nearly 5 logs of attenuation orally and 3.5 logs i.v compared to C5. BRD441 showed 4.5 logs attenuation orally 25 and 1 log i.v.. The results are presented in Table 2.

2.4.2.2. Persistence of strains in the organs of BALB/c mice following oral inoculation

Groups of 4 BALB/c mice were orally inoculated with log₁₀8 organisms of the 30 three strains. Mice were killed at days 0,1,4,7,10,16,21 and 28 and the organs examined for bacterial load. The wild type strain C5 colonised the spleen, liver, mesenteric lymph nodes and Peyer's patches in high numbers (>log_I04 cfu ml), eventually resulting in the death of the animals. BRD 1115 and BRD 441 on the other hand persisted in the liver and spleens for more than 40 days in low numbers (<log₁₀2 cfu/ml). These results are presented in Figure 2.

2.5. Evaluation of BRDl 115 as a potential vaccine strain

2.5.1. BRDl 115 protects against homologous challenge

Groups of BALB/c mice were orally immunised with log₁₀8 organisms of BRDl 115 and challenged with the wild type strain C5 at 4 weeks and 10 weeks post inoculation. The mice were challenged with log_I04 to log₁₀10 organisms C5 and a new oral LD₅₀ calculated. The levels of protection are presented in Table 3, showing log₁₀4 protection after 4 weeks and log₁₀5 after 10 weeks.

2.5.2. BRDl 115 as a potential carrier strain for heterologous antigens Two plasmids encoding the C fragment of tetanus toxin were introduced into two isolates of BRDl 115 by electroporation. The plasmids are pTET/w>15 (38) in which fragment C is under the control of the nirB promoter, and pTEThtrA in which fragment C is under the control of the htrA promoter. The plasmids were found to be maintained at levels greater than 90% in BRDl 115 even when the selection pressure of ampicillin was removed from the growth medium. In vitro expression of fragment C was determined by Western blotting. The strains were cultured under both inducing (42°C for BRD 1126 and anaerobiosis for BRD 1127) and non-inducing conditions (37°C for BRD 1126 and aerobiosis for BRD 1127). A higher level of expression was seen for both strains under inducing conditions with BRD 1127 showing higher levels of fragment C expression than BRD 1126.

Groups of 10 BALB/c mice were orally immunised with log₁₀8 organisms and bled weekly. The titres of anti-fragment C antibodies present in the serum of each animal was determined by ELISA. The titres were determined as the reciprocal of the highest sample dilution giving an absorbance of 0.3 above normal mouse serum. The results are presented in Figure 3. Four weeks post immunisation the mice were challenged with 50LD₅₀'s of tetanus toxin subcutaneously and the deaths noted over 4 days. The results are presented in Table 4, showing that 100% protection was given after immunisation with BRDl 127 (fragment C under the control of the htrA promoter) and 60% protection after immunisation with BRDl 126 (under nirB promoter). No naive mice survived the challenge.

Example 2

This Example confirms that the mutation in surA is responsible for the attenuation. This was determined by complementation of the deleted gene with an intact version of the gene expressed on a plasmid. The complemented strain was as virulent as the wild-type organism given orally to mice.

Materials and Methods

3.1 Construction of plasmid containing the intact surA gene pLG339 (41) is a low copy number plasmid based on pSC105. A 3kb fragment of the plasmid pGEM-T/212/213 (section 2.2) containing the intact surA gene and flanking region was cloned into the SphllSaR sites of the plasmid pLG339 to create the plasmid

A schematic of this plasmid is shown in Figure 4.

3.2 Introduction of the plasmid pLG339 swr4 into defined mutant strain BRDl 115 The plasmid was electroporated into electrocompetent BRD 1115 as previously described in section 1.5.2. Transformants containing the plasmid were selected by plating the electroporation mix onto agar plates containing 15μg/ml kanamycin. Plasmid DNA was recovered from a single colony of this transformation and checked for identity by restriction analysis. This strain was called K2. 3.3 Plasmid stability within the strain K2.

The ability ofthe intact surA gene on the plasmid to complement the action of the deleted surA gene in the chromosome relies on the plasmid being retained within the bacterial strain. The plasmid contains the gene encoding resistance to the antibiotic kanamycin. Culturing the strain in the presence ofthe antibiotic should ensure that the plasmid is retained. However it is important that the plasmid be retained in the absence ofthe antibiotic selection as antibiotic selection is not possible in vivo.

A single colony ofthe strain K2 was inoculated into duplicate 10 ml cultures of L broth with and without kanamycin. The cultures were grown with shaking at 37°C for a total of 72 hours. Samples were taken at 30 and 48 hours post inoculation and serial dilutions plated onto L agar plates with and without kanamycin. The cultures were diluted 1/100 into Fresh L broth with and without kanamycin and cultured for a further 24 hours. Dilutions ofthe culture were again plated out onto L agar plates with and without kanamycin. Numbers of colony forming units (cfti) were recorded and are reported in Table 5.

3.4 Oral immunisation of mice with the strain K2. The strain K2 was grown as described in 1.8 and used to challenge orally groups of 5 Balb/c mice (as previously described) with a dose range from 10⁴ to 10¹⁰ /dose. Deaths were recorded over 28 days and the LD₅₀s calculated according to the method of Reed and Meunch (described in 2.3.2).

Results

4.1 Strain

The plasmid pLG339/s^**«r_4 was recovered from the strain K2 and digested with the two enzymes Sphl and Sail. Separation ofthe resultant bands by agarose gel electrophoresis revealed the correct sized bands of 6.2 and 3 kb. 4.2 Plasmid Stability

The presence ofthe plasmid pLG339/swr,4 was investigated in the strain K2. The results show that in the absence of antibiotics the plasmid is retained by the bacteria. In these studies, at least 82% ofthe bacteria retain the plasmid when grown without antibiotics. This suggests that this plasmid should be maintained when the bacteria are used to infect mice.

4.3 Complementation data

Groups of 5 Balb/c mice were orally challenged with various doses ofthe putative complemented strain K2. The oral LD₅₀ ofthe complemented strain K2 was calculated to be log₁₀4.35 compared to that of log_I04.17 for the parental strain C5.

Deaths ofthe mice within the group of mice challenged with log₁₀8 bacteria ofthe three strains C5, BRDl 115 and K2 are represented in Figure 5. Although the surA gene expressed from the plasmid appears to commplement the defined mutation in vivo, the apparent delay in the time to death (when compared to the wild type parent strain) suggests the level of surA expression may be reduced in the strain K2.

Tables

Table 1: Bacterial strains, plasmids and oligonucleotide primers used in this study

Bacterial strains Properties Source or ref

E.coli SY327 λpir lysogen Miller V.L.(23)

S.typhimurium LB5010 semi-rough the inventor laboratory

C5 wild type C.Hormaeche, Cambridge, U.K.

BRD441 TnphoA mutant, kan^R Miller I (21)

BRD 1115 this study

BRD 1126 amp" Oxer M.D. (25) BRD 1127 amp^R in press

Plasmids pBluescript^όII SK+ amp^κ Stratagene Ltd pGEM-T amp^R Promega Corp. pGP704 amp^R Miller V.L. (23) pGEM-T/212/213 amp^R this study pGEM-T/ΔsurA amp^κ this study pGP704/ΔsurA amp^R this study pGEM-T/92/93 amp^R this study pTETmrl5 amp^R Oxer M.D. (25) pTET/ztrA amp^R in press

Oligo Primers

MGR 92 TCGGCACGCAAGAAATGT Kings College, London MGR 93 AGACGACCAGTTCAATCG

MGR 130 CGATGGGCTGAACTATTC MGR 135 TATGCAGCTTCGTTAGCG

Table 2: The oral and i.v. LD ₅₀'s ofthe three strains C5, BRD 441 and BRD 1115 were determined in BALB/c mice. Groups of 5 mice were immunised with doses ranging from log₁₀4 to log₁₀10 cfu ofthe strains BRD 441 and BRD 1115, and doses log₁₀l to log₁₀5 ofthe strain C5. The results are presented in the following table.

Strain oral LD₅₀ i.v.LD₅₀ (log,₀ cfu) (log₁₀ cfu)

C5 4.16 <1.87 BRD 441 8.62 2.46 BRD1115 8.98 5.22 Table 3: The ability of the defined surA mutant strain to confer protection against homologous challenge with the wild type strain C5 was determined. Groups of 5 BALB/c mice were orally immunised with log₁₀8 organisms ofthe strain BRDl 115 then challenged with log₁₀4 to log₁₀10 of the mouse virulent strain C5 either 4 or 10 weeks post inoculation. The new LD_{5 0} was then calculated and the results presented in the table below.

Immunising oral LD«n of C5 protection strain (no of LD₅₀'s)

4 weeks post 10 weeks post immunisation immunisation

BRD1115 8.58 -3800 none 4.74

BRD 1115 9.51 -4800 none 4.68

Table 4: Three groups of 10 mice were immunised with the strains BRDl 115, BRDl 126 and BRDl 127 and then challenged 4 weeks post immunisation with 50 LD₅₀ doses of tetanus toxin subcutaneously. Deaths were noted over 4 days. The numbers of mice surviving the challenge are presented in the table below.

Strain Survivors after challenge BRD 1115 0/10 BRD 1126 (nirB) 6/10 BRD 1127 (htrA) 10/10

Table 5: The numbers of bacteria (cfu) present in the cultures ofthe complemented strain K2 following culture in L broth with and without the antibiotic kanamycin were calculated. The cultures were then plated onto L agar with and without kanamycin to show presence ofthe plasmid pLG339 swr/4. The results are presented as a total number and also the kanamycin resistant colonies as a percentage ofthe total bacteria present.

Kanamycin Kanamycin numbers of bacteria (cfu/ml) in broth in agar 30 hours (%) 48 hours (%) 72 hours (%)

4.75xl0⁷ (95%) 6xl0⁷ (71%) 8.25xl0⁷ (82.5%)

5 xlO⁷ 8.5xl0⁷ lOxlO⁷

5.25xl0⁷ (124%) 9.5xl0⁷ (111%) 6xl0⁷ (89%) 4.25xl0⁷ 8.5xl0⁷ 6.75xl0⁷

References

1 Bacon, G.A., Burrows, T. W. and Yates, M. (1950) Br. J. Exp. Pathol.., 31,

714-24.

2. Chatfield, S.N., Charles, I.G., Makoff, A.J. et. al. (1992a) Biotech, 10, 888- 892.

3. Chatfield, S.N. Strahan,K., Pickard, D., Charles, I.G., Hormaeche, C.E. and Dougan, G. (1992b) Microbiol. Pathog. , 12, 145-151.

4. Chatfield, S.N. Fairweather, N., Charles, I., Pickard, D., Levine, M. Hone, D., Posanda, M., Strugnell, R.A. and Dougan G. (1992) Vaccine, 10, 53-60.

5. Curtiss III, R. and Kelly, S.M. (1987) Infect. Immun. 55, 3035-3043.

6. Dougan, G. Chatfield, S., Pickard, D., Bester, J., O'Callaghan, D. and Maskell, D. (1988) J. Inf. Dis, 158,1329-1335.

7. Fairweather N.F., Lyness V. A., and Maskell D.J., (1987) Infect. Immun. 55, 2541-2545

8. Fairweather, N.F., Chatfield, S.N. Makoff, A.J. et. al. (1990) Infect. Immun., 58, 1323-1329.

9. Gomaz-Duarte, O.G., Galen, J. Chatfield, et. al. (1995) Vaccine, 13:1596- 1602.

10. Harrison J.A., Pickard D., Higgins C.F., Khan A., Chatfield S., Ali T., Dorman C.J. Hormaeche C, and Dougan G., (1994) Mol. Micro., 13, 133-140

11. Hohmann, E ., Oletta, C.A., Killeen, K.P. and Miller, S.I. (1996) Vaccine 14, 19-24.

12. Hone, D., Morona, R., Attridge, S. and Hackett, J. (1987) J. Infect. Dis., 156, 167-1

13. Hull R.A. Gill R.E. Hsu P., Minshew B.H., and Falkow S., (1981) Infect. Immun. 33, 933-938

14. Johnson K., Charles I., Dougan G., Pickard D., O'Gaora P., Costa G., Ali T., Miller I., and Hormaeche C. (1991) Mol. Micro., 5, 401-407

15. Jones, P.W., Dougan, G. Haywood, C, .MacKensie, N., Collins, P. and Chatfield, S.N. (1991) Vaccine 9, 29-36. 16. Lazar S.W., and Kolter R., (1996) J.Bact. 178, 1770-1773

17. Levine, M. M., Galen, J., Barry, E., et al (1995) J. Biotech., 44, 193-196.

18. Manoil, C. and Beckwith, J. (1985) Proc. Natl. Acad. Sci., USA 82, 8129- 8133.

19. Miles, A.A., Misra, S.S. and Irwin, J. (1938) J. Hygiene, 38, 732-749.

20. Miller I., Chatfield S., Dougan G., Desilva L., Joysey H.S., and Hormaeche C, (1989a) Mol. Gen. Genet., 215, 312-316

21. Miller, I., Maskell, D., Hormaeche, C, Pickard, D. and Dougan, G. (1989b) Infect. Immun. 57, 2758-2763.

22. Miller, S.I., Kukral, A.M. and Mekalanos, J.J. (1989). Proc. Natl. Acad. Sci., USA 86, 5054-5058.

23. Miller V.L., and Mekalanos J.J. (1988) J.Bact. 170, 2575

24. Missiakis D., Betton J.M., and Raina S., (1996) Mol. Micro., 21, 871-884

25. Oxer, M.D., Bentley, CM., Doyle, J.G. Peakman, T.C., Charles, I.G. and Makoff, A.J. (1991) Nucl. Acids Res. 19, 2889-2892.

26. Pickard, D., Li., J.L., Roberts, M., Maskell, D., Hone, D., Levine, M., Dougan, G. and Chatfield, S. (1994), 62, 3984-3993.

27. Reed L.J., and Meunch H., (1938) Am. J. Hygiene 27, 493-497

28. Roberts M., Bacon A., Rappuoli R., Pizza M., Cropley I., Douce G., Dougan G., 27 Marinaro M., McGhee J., and Chatfield S., (1995) Infect. Immun. 63, 2100-2108

29. Rouviere P.E., and Gross C.A., (1996) Genes Dev., 10, 3170-3182

30. Rudd K.E., Sofia H.J., Koonin EN., Plunkett III G., Lazar S., and Rouviere P.E. (1995) TIBS 20, 12-14.

31. Sambrook J., Fritsch E.F., and Maniatis T., (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, ΝY, USA

32. Strugnell, R.A. Dougan, G., Chatfied, S.Ν. et. al. (1992) Infect. Immun., 60, 3994-4002.

33. Tormo A., Almiron M., and Kolter R., (1990) J.Bact. 172, 4339-4347 34. Yura T., Mori H., Nagai H., Nagata T., Ishihama A., Fujita N., Isono K., Mizobuchi K., and Nakata A.(1992) Nucl. Acids Res., 20, 3305-3308

35. EP-B-0322237 (Dougan et al)

36. EP-B-0400958 (Dougan et al)

37. EP-B-0524205 (Dougan et al)

38. WO 92/15689 (Charles et al)

39. Chatfield, S.N., Strugnell. R.A. and Dougan, G (1989) Naccine, 7, 495-498

40. Everest, P., Allen, J., Papakonstantinopoulou, A., Mastroeni, P., Roberts, M. and Dougan, G. (1995) FEMS Microbiol. Letts., 126, 97-101

41. Stoker Ν.G., Fairweather N.F., and Spratt B.G. (1982) Gene 18(3) 335-341

Claims

Claims

1. A vaccine comprising a pharmaceutically acceptable carrier or diluent and a bacterium attenuated by a non-reverting mutation in a gene encoding a protein which promotes folding of extracytoplasmic proteins.

2. A vaccine according to claim 1 wherein the protein encoded by the mutant gene is a periplasmic protein.

3. A vaccine according to claim 1 or 2 wherein the protein encoded by the mutant gene promotes the folding of secreted proteins.

4. A vaccine according to claim 1, 2 or 3 wherein the protein encoded by the mutant gene is a peptidyl-prolyl cis-trans isomerase (PPiase).

5. A vaccine according to claim 4 wherein the PPiase is a member ofthe parvulin family of PPiases.

6. A vaccine according to any one ofthe preceding claims wherein the protein encoded by the mutant gene is SurA.

7. A vaccine according to any one ofthe preceding claims wherein the bacterium is further attenuated by a non-reverting mutation in a second gene.

8. A vaccine according to claim 7 wherein the second gene is an aro gene, a. pur gene, the htrA gene, the ompR gene, the g╬▒lE gene, the cy╬▒ gene, the crp gene or the phoP gene.

9. A vaccine according to claim 8 wherein the ╬▒ro gene is ╬▒roA, ╬▒roC, ╬▒roD or ╬▒roE.

10. A vaccine according to any one ofthe preceding claims wherein the mutation in the gene encoding a protein which promotes folding of extracytoplasmic proteins and/or the mutation in the second gene is a defined mutation.

5 11. A vaccine according to any one ofthe preceding claims wherein the bacterium has no uncharacterised mutations in the genome thereof.

12. A vaccine according to any one ofthe preceding claims wherein the bacterium is a bacterium that infects via the oral route.

10

13. A vaccine according to any one of the preceding claims wherein the bacterium is from the genera Salmonella, Escherichia, Vibrio, Haemophilus, Neisseria, Yersinia, Bordetella or Brucella.

15 14. A vaccine according to claim 13 wherein the bacterium is Salmonella typhimurium, Salmonella typhi, Salmonella enteritidis, Salmonella choleraesuis, Salmonella dublin, Escherichia coli, Haemophilus influenzae, Neisseria gonorrhoeae, Yersinia enterocolitica, Bordetella pertussis or Brucella abortus. 0

15. A vaccine according to any one ofthe preceding claims wherein the bacterium is genetically engineered to express an antigen from another organism.

16. A vaccine according to claim 15 wherein the antigen is fragment C of tetanus toxin. 25

17. A vaccine according to claim 15 or 16 wherein expression ofthe antigen is driven by the nirB promoter or the htrA promoter.

18. A bacterium as defined in any one ofthe preceding claims for use in a method 30 of vaccinating a human or animal.

19. Use of a bacterium as defined in any one of the preceding claims for the manufacture of a medicament for vaccinating a human or animal.

20. A method of raising an immune response in a host, which method comprises administering to the host a bacterium attenuated by a non-reverting mutation in a gene encoding a protein which promotes folding of extracytoplasmic proteins.