Live vaccine strains of Francisella

Abstract

A strain of Francisella species wherein a gene which encodes for part of the glutamate metabolic pathway has been inactivated, and which is able to produce a protective immune response in an animal, for use as live prophylactic or therapeutic vaccine against infection by said Francisella species. Particularly effective strains include those where the capB gene is deleted. Other embodiments of the invention describe strains which comprise a further genetic mutation wherein a gene which encodes for another component of the cell is also inactivated. Pharmaceutical compositions comprising said strains, together with methods which utilize such strains are also described and claimed.

Description

This invention relates to live strains of Francisella species, their use as prophylactic or therapeutic vaccines, to compositions comprising these strains, and their use in the prevention or treatment of bacterial infection.

Francisella tularensis is an extremely pathogenic Gram-negative bacterium and is the etiological agent of the zoonotic disease Tularemia. There are four recognised sub-species of F. tularensis, including subspecies tularensis, holarctica, and novicida, which exhibit a high degree of genetic conservation. The most virulent subspecies is Francisella tularensis subspecies tularensis, which has an infectious dose in humans of as little as 10 cells via the airbourne route.

At present there is no available vaccine against F. tularensis infection although it has been demonstrated previously that an undefined attenuated strain of Francisella tularensis, which has been designated Francisella tularensis live vaccine strain (LVS), is capable of providing protection against the most virulent subspecies, tularensis. However, the LVS vaccine is not registered and has only been used to vaccinate at-risk, individuals under special license. This license has now been withdrawn. The LVS strain is likely to remain unlicensed because the genetic changes that are responsible for the attenuating phenotype are not understood at the molecular level. Therefore, there exists a possibility that the vaccine strain could revert back to the fully virulent form. Further, it has been shown that whilst LVS provides effective protection in the mouse model of infection, protection is not complete. The protection afforded by LVS against an aerosol challenge of the most virulent tularensis subspecies is sub-optimal. Clearly, a vaccine which is genetically stable and which provides complete protection is highly desirable.

The fact that naturally occurring, attenuated strains of F. tularensis can induce protective immunity does suggest that an attenuated strain with properly defined genetic mutations in the organism's virulence factors is a feasible approach in vaccine development.

Unfortunately, however, relatively little is known about the virulence mechanisms of F. tularensis and, as such, virulence factors have proven to be very difficult to predict; the recent completion of the genome sequence of the virulent strain F. tularensis subspecies tularensis SchuS4 has so far failed to reveal the presence of classical virulence factors such as toxins or type-III secretion systems, which are predominant in so many other pathogenic bacteria.

Some efforts to identify new vaccine strains have focused on naturally occurring strains (other than LVS) or on spontaneously attenuated strains of F. tularensis, such as the FSC043 mutant of SchuS4 reported by Twine et al (Infection and Immunity Vol 73, 2005, pp 8345-8352). Examination of these attenuated mutants has confirmed that mutations can lead to attenuated strains which afford some level of protection against tularemia. However, the molecular basis of this attenuation and protection is unknown and, in any case, the protection afforded is not better than that provided by LVS. The problem remains, therefore to find defined mutations which give complete protection against the most virulent forms of Francisella.

A live vaccine strain of Francisella derived from the subspecies novicida, which contains a single genetic mutation, is described in co-pending International Application number PCT/GB2004/001264, the contents of which are hereby incorporated by reference. This application shows that a genetic lesion in the purine enzyme pathway provides attenuated strains which are also protective in the mouse model of infection. Whilst this clearly represents a significant advance in the development of a vaccine for tularemia, it is widely recognised that an additional genetic mutation would be required to enable such a strain to obtain licensed status. The problem remains, therefore, to determine further mutations which result in attenuated strains but which also provide complete protection against all strains of Francisella tularensis.

New live vaccines, containing well defined mutations and which are fully protective against tularemia are therefore required.

The applicants have found that by modifying strains of Francisella in a particular way, attenuated strains which are protective can be produced. These live strains can be used as the basis for new vaccines against tularemia.

The present invention therefore provides a strain of Francisella species wherein a gene which encodes for part of the glutamate metabolic pathway has been inactivated, and which is able to produce a protective immune response in an animal, for use as a live prophylactic or therapeutic vaccine against infection by said Francisella species.

As used herein, the term “metabolic pathway” means the sum total of the chemical processes occurring in a cell, in which the processes occur in steps, through which compounds are gradually built up or broken down. Each step of the metabolic pathway is catalysed by an enzyme, whose structure is encoded by a gene.

As used herein the term “glutamate metabolic pathway” means the chemical processes whereby glutamate is synthesised or broken down, including those steps wherein an enzyme catalyses the formation of glutamate or catalyses the use of glutamate to form another material.

Thus, genes that encode for part of the metabolic pathway in Francisella species include those genes which encode for the formation of glutamate and those which encode for enzymes which are utilised in the glutamate metabolic pathway. Inactivation of such a gene is likely to interrupt the normal metabolic pathway such that, for example, glutamate is not produced by the cell or that the amount of glutamate produced is significantly altered when compared with the virulent Francisella species, such as Francisella tularensis subspecies tularensis SchuS4, in which the same gene having been inactivated.

The presence, or absence, of glutamate (or aminoglutaric acid) may be readily determined using 13C, 15N or 1H Nuclear Magnetic Resonance Spectroscopy (NMR), as described by the papers by Robertson, D. E. et al (Applied and Environmental Microbiology 1990, vol. 56 pp 1504-1508 and Biochimica et Biophysica Acta 1989, vol. 992, pp 320-326) and the quantities of glutamate produced by strains of the present invention may be readily determined using techniques such as Liquid Chromatography-Mass Spectrometry (LC-MS), which are routine in the art.

As used herein “glutamate” refers to the amino acid glutamate or glutamic acid, which may exist in the form of β-glutamate (beta-glutamate) or γ-glutamate (gamma-glutamate).

The inventors have found that inactivating a gene which encodes for part of the glutamate metabolic pathway provides a strain which is both attenuated and protective against exposure to virulent Francisella species. Without wishing to be bound by theory, it is thought that part of the glutamate metabolic pathway may include, or have an effect on the formation of extracellular capsule. Thus, genes which encode for the synthesis of capsule, or which are analogous to these genes when compared with established capsule encoding genes from other organisms, may fall within the definition of genes that encode for part of the metabolic pathway, as described above.

As used herein, the term “capsule” means an extracellular component, commonly a layer, of polysaccharide and/or protein which protects a bacterial cell and which, in association with pathogenic bacteria, serves as a barrier against phagocytosis by the white blood cells of an animal host, in which the pathogenic bacteria are present.

As used herein, the term “capsule-encoding gene” means a gene which encodes for a protein, or other molecule, which is involved in the synthesis of the capsule component of a bacterial cell or a gene which is homologous to a gene which has been assigned as having some involvement in bacterial capsule biosynthesis, including involvement in the synthesis of capsule or any component thereof and involvement in the assembly and/or transport of said components to form a capsular structure.

As used herein, the terms “homologous” and “homology” means, at the protein level, the similarity of two amino acid sequences are such that the two sequences share greater than 30% identity. Identity in this instance can be judged for example using the BLAST program (vs. 2.2.12) found at http://www.ncbi.nim.nih.gov/BLAST/ or the algorithm of Lipman-Pearson with, for example, Ktuple:2, gap penalty:4, Gap Length Penalty:12, standard PAM scoring matrix or other suitable parameters as readily determined by a person skilled in the art (Lipman, D. J. and Pearson, W. R., Rapid and Sensitive Protein Similarity Searches, Science, 1985, vol. 227, 1435-1441).

Genes which encode for the glutamate metabolic pathway (including those which encode for a capsule component) of the Francisella bacterium may be determined by analysis of the relevant genome sequence and/or by comparison with other bacteria which have well-defined capsule encoding genes.

Gene inactivation can be carried out using any of the conventional methods known in the art. Typically, the strain is transformed with a vector which has the effect of down-regulating or otherwise inactivating the gene. This can be done by mutating control elements such as promoters and the like which control gene expression, by mutating the coding region of gene so that any product expressed is inactive, or by deleting the gene entirely. Alternatively, the gene can be inactivated at the RNA or protein level, by transforming the cell so that it expresses a sense or anti-sense construct which binds to DNA or RNA encoding the gene to prevent transcription thereof.

Preferably however, the gene is inactivated by complete or partial deletion mutation or by insertional mutation.

Specifically, the applicants have found that it is preferable to inactivate one or more of the genes which has been annotated in the Francisella tularensis genome as being a capsule-encoding gene, for example capB or capC (hereinafter “cap” genes). In particular the applicants have found that it is preferable to inactivate a cap gene in Francisella species such as F. tularensis subspecies tularensis or subspecies holarctica. In a preferred embodiment, the capB gene (FTT0805) and/or the capC gene (FTT0806) of F. tularensis subspecies tularensis is inactivated to provide one strain according to the present invention. In other embodiments, the corresponding genes of other Francisella species are deleted to provide other strains according to the preset invention. The corresponding cap genes from other Francisella species will usually have at least 80%, preferably at least 85% homology and more preferably at least 90% homology to, and similar function to, the cap genes of F. tularensis subspecies tularensis. In a more preferred embodiment, the capB gene (FTT0805) of F. tularensis subspecies tularensis strain SchuS4 is inactivated to provide a preferred strain.

In particular, the applicants have found that a strain of Francisella tularensis which has a cap gene inactivated is attenuated and is protective against challenge from virulent Francisella species. As illustrated hereinafter, a strain of F. tularensis subspecies tularensis which has an inactivated cap gene is protective in mice. In particular the applicants have shown that deletion of the capB gene from F. tularensis subspecies tularensis SchuS4 strain provides an attenuated strain which is protective against aerosol challenge with the virulent SchuS4 strain.

The strain of the invention suitably has a further defined mutation or lesion so as to reduce the risk of the bacterium reverting to a virulent form. In this case, the mutation is in a gene which is selected so that the strain is suitably attenuated, but can still retain the ability to stimulate a sufficient immune response to provide long term protection. Suitable additional mutations can be identified using conventional methods, and examination and analysis of the current live vaccine strain (LVS) or other attenuated strains may assist in the identification. Examples of these mutations include, but are not limited to, mutations to the FTT0918 and FTT0919 genes. Alternatively the further defined mutation may advantageously involve a gene which encodes for another component of the cell, which is not part of the glutamate metabolic pathway. Such further mutations include, but are not limited to, mutations which inactivate pilin genes, for example mutations to pilA and/or pilE and/or pilC genes or other mutations such as those described in co-pending British patent application number GB0511722.1 (the contents of which are hereby incorporated by reference), mutations which inactivate genes which encode enzymes in the purine pathway, for example mutations to purA and/or purF genes and other purine pathway genes such as those described in co-pending International Application number PCT/GB2004/001264 (the contents of which are hereby incorporated by reference).

Particular preferred examples of further defined mutations are inactivation of the purF and/or inactivation of the pilA gene and/or inactivation of the FTT0918 gene.

It will be understood by the skilled person that such further defined mutations can be achieved by using any conventional method as hereinbefore described but that in preferred embodiments of the inventions the gene which encodes for other components of the cell is inactivated by complete or partial deletion mutation or by insertional mutation.

Since the strains of the present invention have been found to be protective against infection by Francisella species in the mouse model of infection, the strains also provide useful vaccines against the diseases caused by Francisella infections and, in particular, tularemia. It is therefore preferred that the strains are formulated into pharmaceutical compositions, in which they are combined with a pharmaceutically acceptable carrier. Such pharmaceutical compositions form a second aspect of the invention.

Suitable carriers may be solid or liquid carriers as is understood in the art. They may suitably be formulated for administration to mucosal surfaces (for example for oral use, of for administration by inhalation or insufflation) or for parenteral administration.

In particular they are formulated as sterile aqueous or oily solutions for intravenous, subcutaneous, intramuscular or intramuscular dosing.

Alternatively they are formulated for administration to mucosal surfaces and in particular for intranasal application. Such formulations may include microencapsulation of the strain in the composition, or microencapsulation of the entire composition. Such microencapsulation techniques are commonly known in the art.

Compositions are suitably prepared in unit dosage forms, as conventional in the art. They are administered at dosages which are determined using clinical practice, and depend upon factors such as the nature of the patient, the severity of the condition, and the precise vaccine strain being employed. Typically dosage units will comprise 10⁵-10⁸ cfu. Dosages may be boosted as appropriate or necessary.

Compositions may also contain further immunogenic reagents which are effective against F. tularensis infection or other diseases. They may further contain other agents such as adjuvants and the like, which enhance the host's immune response to the vaccine.

In a further aspect the present invention relates to the use of a strain of Francisella species wherein a gene which encodes for part of the glutamate metabolic pathway has been inactivated, and which is able to produce a protective immune response in an animal, in the preparation of a live prophylactic or therapeutic vaccine against infection by Francisella species. In particular, such strains find use in the preparation, or manufacture, of a vaccine for the treatment of Tularaemia.

In yet a further aspect, the invention provides a method of preventing or treating infection caused by Francisella species, which method comprises administering to an animal, including a human being, an effective amount of a strain or of a pharmaceutical composition, each as hereinbefore described.

In particular, the method is useful in the treatment of infection caused by Francisella tularensis subspecies tularensis.

Novel strains which are suitable for vaccine use form a further aspect of the invention. In particular, the invention provides a strain of Francisella species wherein a gene which encodes for part of the glutamate metabolic pathway has been inactivated. In particular the strains which are suitable for use as vaccines are as hereinbefore described. In a preferred embodiment the inactivated gene is capB or capC. It is further preferred that the strain is a strain of Francisella tularensis subspecies tularensis, for example the SchuS4 strain.

The invention will now be particularly described by way of non-limiting Example, with reference to the accompanying diagrammatic drawings in which:

FIG. 1 shows the strategy for the construction of a suicide plasmid for the deletion of the capB gene from the F. tularensis subspecies tularensis strain SchuS4.

FIG. 2 shows a southern blot of genomic DNA from wild type and capB mutant strains of F. tularensis subspecies tularensis strain SchuS4.

FIG. 3 shows the effects of osmotic stress on wild type F. tularensis subsp. tularensis strain SchuS4 and the ΔcapB mutant strain of the invention, when grown in a range of salt conditions.

FIG. 4 shows survival data of BALB/c mice infected subcutaneously with wild type and caps mutant strains of F. tularensis subspecies tularensis strain SchuS4.

FIG. 5 shows survival data of BALB/c mice, administered with a capB mutant strain of F. tularensis subspecies tularensis strain SchuS4, and subsequently challenged with 70 mean lethal doses (MLD) of the virulent strain, F. tularensis subspecies tularensis strain SchuS4.

FIG. 6 shows survival data of Female BALB/c mice (6-8 weeks old), injected subcutaneously with 100 μl of PBS containing 10⁴ CFU F. tularensis subsp. tularensis strain SchuS4 ΔcapB, F. tularensis live vaccine strain (LVS) or PBS alone and eight weeks later challenged with 10⁴ CFU administered by the subcutaneous route.

FIG. 7 shows a graph depicting the colonization and clearance of bacteria from the spleens of mice administered with either wild type F. tularensis subsp. tularensis strain SchuS4 or the ΔcapB mutant strain of the present invention.

FIG. 8 shows a graph which quantifies the IL-2 and IFN-γ recall response of spleen cells harvested from mice immunized subcutaneously with 100 μl of PBS containing 10⁴ CFU F. tularensis subsp. tularensis strain SchuS4 ΔcapB, F. tularensis live vaccine strain (LVS) or PBS alone and then subsequently stimulated with wild type F. tularensis subsp. tularensis strain SchuS4.

EXAMPLE 1

Construction of a Plasmid Containing a Mutated Allele of F. tularensis capB

The capB gene of F. tularensis subspecies tularensis strain SchuS4 encodes a protein of 405 amino acids (aa) that has 36% identity to the 397 aa CAPB protein from Bacillus anthracis strain ‘Ames Ancestor’. Regions of DNA flanking the capB gene of F. tularensis were PCR amplified from F. tularensis subspecies tularensis strain SchuS4 using the primer pairs P1/P2 and P3/P4 as shown in Table 1. The chloramphenicol resistance cassette (Cam-r) was PCR amplified from the plasmid pKK202 with the primer pair CamF/CamR.

TABLE 1
Primers used for the construction of pSMP42 -
Sequences in bold indicate the sequence
complementarv to Francisella DNA and the
underlined sites in the 5′ extensions are
restriction sites.
P1   5′-CTGACGCGT-AGGCAGTGTGGTTATGGGTAG-3′
P2   5′-GACGGTAACC-CAAATACGACGACAATTAAC-3′
P3   5′-CTGGGTAACC-TCCAGCAAACTCTTATATTC-3′
P4   5′-TAGACGCGT-ACCCAATCAACCCAGTACAAG-3′
CamF 5′-GCTGGTTACC-TAAGAGGTTCCAACTTTCAC-3′
CamR 5′-CTAGGTTACC-TTTAAGGGCACCAATAACTG-3′

The left and right flanks and the chloramphenicol resistance cassette were assembled as shown in FIG. 1, and cloned into the suicide plasmid pSMP22 to give the construct pSMP42. This gave a plasmid-borne mutant allele that could be used for recombinational exchange with the wild type chromosomal allele.

EXAMPLE 2

Generation of a F. tularensis Strain Deleted for capB

The suicide plasmid pSMP42 was electroporated into the E. coli mobilising strain S-17 λpir. The plasmid was then introduced from the mobilising strain to F. tularensis subspecies tularensis strain SchuS4 by conjugal transfer. Transconjugants were selected on chloramphenicol and merodiploids arising from chromosomal integration of the suicide plasmid were resolved by plating on Thayer Martin agar containing sucrose at 5%. Allelic replacement mutants were confirmed by Southern blot analysis as shown in FIG. 2. Genomic DNA of wild type and capB deletion mutants (ΔcapB) of F. tularensis subspecies tularensis strain SchuS4 was digested with MluI and NcoI, separated by agarose gel electrophoresis and transferred to a nylon membrane. The MluI insert of pSMP42 was labelled with DIG-11-dUTP during PCR amplification with the primer pair P1/P4 and used as a probe to hybridise to the membrane. DNA fragments to which the probe hybridised were detected in a chemiluminescent assay (CSPD substrate, 30 min exposure, X-ray film). The ΔcapB mutant contained two hybridising fragments of 12.5 Kb and 3.2 Kb as shown in FIG. 2.

EXAMPLE 3

Comparison of Phenotype—Effect of Osmotic Stress

F. tularensis subsp. tularensis strain SchuS4 ΔcapB and F. tularensis subsp. tularensis strain SchuS4 (wild type) were suspended in a range of salt (nacl) concentrations (up to 5M) and then plated onto BCGA media using 20 μl droplets. Bacterial growth was assessed after 72 hours and the results are portrayed in FIG. 3. It was apparent that, as compared with the wild type, SchuS4 ΔcapB grew less efficiently at the highest salt (5 M) concentration. It appeared that ΔcapB was also more sensitive at the low salt concentrations. This indicates that the ΔcapB mutant has a different phenotype to wild type F. tularensis and exhibits different growth characteristics in osmotic environments.

EXAMPLE 4

Determination of Virulence of a capB Mutant in the Mouse Model of Tularemia

Several investigators have determined that the MLD of F. tularensis subspecies tularensis is ˜10 colony forming units (CFU) in the BALB\c mouse, irrespective of route of administration. In order to determine whether capB is required for virulence, groups of six female BALB\c mice, aged 6-8 weeks, were infected subcutaneously with wild-type SchuS4 and ΔcapB SchuS4 strains of F. tularensis subspecies tularensis. As previously reported mice infected with 10² (115) CFU of SchuS4 succumbed to infection by day 5 post-infection. In contrast, 100% of the mice infected with 1.6×10⁵ and 80% (4 out of 5) of mice infected with 1.6×10⁶ and 1.6×10⁴ CFU of the ΔcapB isogenic strain survived as shown in FIG. 4. This level of survival following deletion of a single gene demonstrates that capB is an important gene in the virulence of F. tularensis subspecies tularensis.

EXAMPLE 5

Protection Afforded by a capB Mutant of Francisella tularensis

The chromosomal copy of the gene capB was deleted by allelic exchange as described in example 1 and 2. Genetic analysis was performed to ensure that the gene was removed (as per example 2) and a clonal population of the resultant strain was prepared. Upon subcutaneous administration to female Balb/c mice (6-8 weeks), 1.6×10⁵ bacteria of the ΔcapB mutant strain failed to cause death in all five mice. This is in contrast to the parental wild-type strain of which only 115 bacteria were required to kill all 5 mice of a control group in 5 days. 46 days after immunisation with the capB negative strain, survivors of the infection detailed in Example 4 were challenged with 70 mean lethal doses (MLD) of a virulent strain of F. tularensis (SchuS4 strain). Non-immunised (naïve controls) mice succumbed to infection and died within 5 days whereas mice immunised with the capB mutant were protected and did not die as shown in FIG. 5.

EXAMPLE 6

Survival Against Virulent Challenge

Female BALB/c mice (6-8 weeks old) were injected subcutaneously with 100 μl of PBS containing 10³ CFU F. tularensis subsp. tularensis strain SchuS4 ΔcapB (prepared as described above), F. tularensis live vaccine strain (LVS) or PBS alone. Eight weeks later mice were challenged with 10⁴ CFU wild type F. tularensis subsp. tularensis strain SchuS4 administered by the subcutaneous route. Survival data, shown in FIG. 6, clearly indicates that superior protection is afforded by the capB deletion mutant than the live vaccine strain, and that 100% survival was observed.

EXAMPLE 7

Colonisation and Clearance In Vivo of F. tularensis Strains

Female BALB/c mice (6-8 weeks old) were injected subcutaneously with 100 μl of PBS containing 10⁴ CFU wild type F. tularensis subsp. tularensis strain SchuS4 or F. tularensis subsp. tularensis strain SchuS4 ΔcapB. Groups of 4 mice were killed 3, 7, 14 and 31 days following inoculation of the bacteria. Spleens were removed and the number of bacteria per spleen determined by serial dilution in PBS followed by microbiological culture on BCGA agar plates for 96 hours at 37 OC. None of the mice injected with wild type SchuS4 survived longer than five days post inoculation, which precluded determination of splenic bacterial burdens in these mice on days 7, 14 and 31 post inoculation. However, at day 3 post inoculation there was a highly significant (P<0.001) difference in the numbers of bacteria in the spleens of mice injected with wild type and the ΔcapB mutant; mice injected with F. tularensis subsp. tularensis strain SchuS4 ΔcapB had substantially lower numbers of bacteria in their spleens. Animals injected with F. tularensis subsp. tularensis strain SchuS4 ΔcapB steadily cleared the organism and no viable bacteria were detected in the spleens of these animals at 31 days post inoculation; indicating that they had effectively cleared the mutant. These results shown in FIG. 7 indicate that the capB mutant strain is less likely to cause latent infection after administration.

EXAMPLE 8

IL-2 and IFN-γ Recall Response of Mice Immunized with the capB Mutant Strain of Example

Female BALB/c mice (6-8 weeks old) were injected subcutaneously with 100 μl of PBS containing 10⁴ CFU F. tularensis subsp. tularensis strain SchuS4 ΔcapB, F. tularensis live vaccine strain (LVS) or PBS alone. 40 days later groups of immunised 4 mice were killed and their spleens removed. Single cell suspensions of spleen cells were prepared in culture media (RPMI-1640) (Sigma, UK) supplemented with 10% heat inactivated foetal bovine serum (FBS) (Sigma, UK); 1% penicillin/streptomycin/glutamine (Sigma, UK) and 50 μM 2-mercaptoethanol (Sigma, UK). Cells were stimulated overnight in triplicate with either heat killed F. tularensis subsp. tularensis strain SchuS4 (5 μg ml⁻¹ protein) in supplemented RPMI 1640 or supplemented RPMI 1640 alone. Il-2 and IFN-γ secretion from the cells was determined using cytokine bead array technology (BO Biosciences, Oxford UK). As compared with mice injected with PBS, spleen cells from animals immunized with SchuS4 ΔcapB or LVS secreted significant quantities of IL-2 and IFN-γ when re-stimulated in vitro with inactivated F. tularensis strain SchuS4. However, when spleen cells from animals immunized with SchuS4 ΔcapB or LVS were cocultured with media alone, no significant IL-2 or IFN-γ secretion was detected. These results are shown in FIG. 8.

Claims

1. A method of stimulating an immune response against or treating infection caused by Francisella tularensis, comprising administering to an animal an effective amount of a strain of Francisella tularensis wherein capB or capC gene has been inactivated by deletion, and wherein administration of the effective amount of the strain is able to produce an immune response in an animal.

2. A method of stimulating an immune response against or treating infection caused by Francisella tularensis, comprising administering to an animal an effective amount of a pharmaceutical composition comprising a live strain of a Francisella tularensis wherein capB or capC gene has been inactivated by deletion, and wherein administration of the effective amount of the strain is able to produce an immune response in an animal, in combination with a pharmaceutically acceptable carrier.

3. The method of claim 1 wherein the strain of Francisella tularensis is a subspecies selected from the group consisting of tularensis, holarctica and tularensis SchuS4.

4. The method of claim 1 wherein the strain comprises a further deletion of a gene that encodes another component of the cell.

5. The method of claim 4 wherein the gene which encodes for another component of the cell is selected from the group consisting of FTT0918, FTT0919 and FTT1564.

6. The method of claim 4 wherein the gene that encodes for another component of the cell is a gene that encodes for a pilin subunit or is a gene that encodes for an enzyme in the purine pathway.

7. The method of claim 4 wherein the gene that encodes for another component of the cell is a gene selected from the group consisting of pilA, pilE, pilC, purA and purF.

8. The method of claim 2 wherein the strain of Francisella tularensis is a subspecies selected from the group consisting of tularensis, holarctica and tularensis SchuS4.