GB2570806A - Immunogenic agent and associated compositions, uses and methods - Google Patents

Abstract

The invention relates to an immunogenic agent comprising a Francisella tularensis protein, F. tularensis lipopolysaccharide (LPS), and glucan particles. The F. tularensis protein may be any variant or fragment thereof, and is preferably chosen from FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 or IgIC. The LPS may be a component thereof, preferably O-antigen. Preferably the glucan particles encapsulate the protein and/or LPS. The protein and LPS may be provided as a fusion. The immunogenic agent preferably elicits an antibody and cellular immune response. The immunogenic agent may be formulated as a pharmaceutical composition and may be used in the treatment or prevention of tularemia infection. Methods of producing the immongenic composition or pharmaceutical composition are also disclosed. The F. tularensis proteins FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 or IgIC may be used in the treatment or prevention of tularemia infection.

Description

IMMUNOGENIC AGENT AND ASSOCIATED COMPOSITIONS, USES AND METHODS

Technical Field of the Invention

The invention relates to an immunogenic agent and associated pharmaceutical compositions for use in the prevention or treatment of infection by Francisella tularensis in an animal. In particular, the invention relates to an immunogenic agent and associated compositions comprising: a F. tularensis protein or variant or fragment thereof; F. tularensis lipopolysaccharide (LPS) or component thereof; and glucan particles (GPs). The invention also relates to the use of the immunogenic agent, associated pharmaceutical compositions or specific F. tularensis proteins or variant or fragment thereof as a medicament, in particular with reference to tularemia, the use of same in the prevention or treatment of infection by F. tularensis, and associated methods of production.

Background to the Invention

The bacterium F. tularensis is considered an intracellular pathogen and the causative agent of the disease tularemia. Although typical zoonotic hosts of F. tularensis include rodents and lagomorphs, incidents of tularemia in humans are documented. Of the four known subspecies of F. tularensis (tularensis, holarctica, mediasiatica and novicida), F. tularensis subspecies tularensis is considered the most clinically relevant subspecies, capable of infecting humans with a low infectious dose, in particular via the respiratory route (<50 colony forming units).

Human incidents of tularemia have a high level of mortality (up to 30%) if left untreated, so there is a strong desire to develop effective medical countermeasures against F. tularensis. Although a number of approaches have been investigated, currently there is no licensed vaccine against tularemia. A live vaccine strain (LVS) of F. tularensis subspecies holarctica was developed in the 1950’s and induced protection in human volunteer trials against an aerosol challenge of a virulent

Francisella strain (McCrumb FR. Aerosol infection of man with Pasteurella tularensis. Bacteriological Reviews. 1961;25:262-7). However, LVS induced poorer protection against higher challenge levels. Furthermore, concerns over the use of LVS in humans include a potential reversion to virulence, mixed colony morphology and variable immunology. As a result, LVS is yet to achieve regulatory approval for use in humans. Killed whole cells of F. tularensis have been investigated for tularemia and were shown to be reactogenic, but of dubious efficacy (Foshay L. Tularemia. Ann Rev Microbiology. 1950;4:313-30; Foshay L et al. Vaccine prophylaxis against tularemia in man. Am J Public Health. 1942;32:1131-45).

The identification of Francisella antigens for incorporation in a subunit vaccine has been the focus of other research. One protective antigen identified thus far is Francisella LPS (see for example Dreisbach V. C. et al. Infection and Immunity. 2000;68:1988-1996). However, administration of LPS to animals offers protection against low-virulence, but not highly virulent, F. tularensis strains.

Given the remaining concerns over tularemia, there is a need to develop a welldefined and efficacious medical countermeasure against tularemia, in particular a vaccine capable of inducing protective immunity in an animal against high-virulence subspecies of F. tularensis.

Summary of the Invention

According to a first aspect, the invention provides an immunogenic agent for use in the prevention or treatment of infection by F. tularensis in an animal, the immunogenic agent comprising: a F. tularensis protein or any variant or fragment thereof; F. tularensis lipopolysaccharide or component thereof; and glucan particles, so as to produce an immune response in an animal administered with the immunogenic agent.

It is to be understood that referring to an immunogenic agent comprising a F. tularensis protein or any variant or fragment thereof encompasses at least one, or more than one, Francisella tularensis protein or any variant or fragment thereof.

The term ‘immunogenic’ typically refers to the ability to stimulate an immune response in an animal such as a human, in particular a humoral and/or cell-mediated immune response. The individual components of the immunogenic agent (F. tularensis protein or any variant or fragment thereof; F. tularensis LPS or component thereof; GPs) may each have immunogenic properties.

Although LPS and certain F. tularensis proteins may have immunogenic properties, the components individually are not necessarily beneficial in stimulating protection against higher virulence Francisella strains. The inventors have found that by combining the specific components of the immunogenic agent according to the invention, protection is achieved against tularemia. Furthermore, alternative and/or additional beneficial properties are conferred either to individual components of the immunogenic agent, or as a result of administering the components in combination as the immunogenic agent, for example: extending the half-life of the immunogenic agent; aiding in the delivery and/or presentation of the components to the animal’s immune system; promoting cytokine secretion e.g. pro-inflammatory cytokine secretion as a result of administration in the animal; and/or reducing any adverse side effects of delivering the immunogenic agent to the animal.

The term ‘F tularensis protein variant’ may refer to a sequence of amino acids which differ from the wild-type or base sequence from which they are derived, but wherein the variant retains the desired properties of the wild-type or base sequence. For example, a variant derived from an immunogenic protein would retain a desired level of immunogenicity relative to the original immunogenic protein, and thus be considered an immunogenic variant. As understood by the skilled person, amino acid substitutions can be ‘conservative’ i.e. replacing an amino acid with another amino acid with similar properties; or ‘non-conservative’ i.e. replacing with another amino acid with different properties. As further understood by the skilled person, suitable variants are preferably at least 70% identical, for example at least 75% identical, at least 80% identical, at least 85% identical, at least 90% identical and preferably at least 95% identical compared to the wild-type or base sequence.

The term ‘F. tularensis protein fragment’ may refer to any amino acid portion of the wild-type or base sequence, typically wherein the fragment retains the desired properties of the wild-type or base sequence. For example, an immunogenic fragment may be a sequence of 5-10 amino acids of the wild-type or base sequence of an immunogenic protein which encodes a particular immunogenic epitope, and thus be considered to be an immunogenic fragment.

The term ‘F. tularensis LPS component’ may refer to any element of the LPS molecule, including, but not limited to, O-antigen, O-polysaccharide, core oligosaccharide or Lipid A.

As demonstrated by the inventors, providing a combined immunogenic agent comprising: a F. tularensis protein or any variant or fragment thereof; F. tularensis LPS or component thereof; and GPs provides the advantage of an immunogenic agent capable of being delivered to an animal such that the animal’s immune system can stimulate a humoral and/or cellular immune response, specifically against the F. tularensis bacterium. Furthermore, the stimulated humoral and cellular immune response achieves protection against F. tularensis in an animal. This may reflect that antibody alone is not sufficient to protect against tularemia, and a T-cell memory response must ideally be induced to ensure protection. Furthermore, the injection of the immunogenic agent in animals results in a depot effect at the site of immunisation, in addition to the triggering of proinflammatory cytokine secretion.

US 5032401 describes the use of β-glucan particles (β-GPs) as a vaccine delivery vehicle to target delivery of encapsulated vaccine antigens to the immune system in order to stimulate immunity. β-GPs are purified Saccharomyces cerevisiae cell walls and treated to primarily remove mannans, proteins and nucleic acids. β-GPs are phagocytosed by dendritic cells via the dectin-1 receptor, so that associated antigens can be processed and presented to stimulate both cellular and humoral immune responses. However, US 5032401 does not teach the delivery alongside (preferably by) β-GPs of a combination of a protein or any variant or fragment thereof, and a LPS or component thereof. In particular, US 5032401 does not teach the delivery alongside (preferably by) β-GPs of a F. tularensis protein or any variant or fragment thereof and a F. tularensis LPS or component thereof.

WO2014/114926 A1 teaches a bacterial protein glycan coupling technology directed towards F. tularensis LPS conjugated to a protein carrier as a vaccine candidate. This document discloses a different approach to the present invention and does not disclose co-delivery of vaccine antigen(s) with glucan particles. Instead, the document discloses the use of an emulsion-based adjuvant with the glycoconjugate.

Noah et al. (GroEL and Lipopolysaccharide from Francisella tularensis Live vaccine Strain Synergistically Activate Human Macrophages. Infection and Immunity. 2010; 78:1797-1806) discloses that the ability of one particular F. tularensis protein (GroEL) to induce an inflammatory response in vitro is synergistic with LPS. However, Noah et al. does not investigate or propose the use of F. tularensis GroEL and LPS as a synergistic vaccine antigen and no in vivo benefit or protection by GroEL and LPS in an animal model is disclosed against F. tularensis infection. Noah et al. also does not contemplate the use of glucan particles.

In the invention, the GPs preferably comprise β-1,3-glucans, more preferably β-1,3glucans purified from Saccharomyces cerevisiae cells walls.

Preferably, the F. tularensis protein or variant or fragment thereof and/or the F. tularensis LPS or component thereof is/are encapsulated by the GPs. The F. tularensis protein or variant or fragment thereof and the F. tularensis LPS or component thereof may be co-encapsulated by the GPs. Alternatively, the F. tularensis protein or variant or fragment thereof and the F. tularensis LPS or component thereof are separately encapsulated by the GPs. Such arrangements provide for loading of the protein- and/or LPS-based components of the immunogenic agent within the hollow, porous GP structure in a protected manner when delivering the immunogenic agent to an animal.

Moreover, GPs containing encapsulated LPS have been advantageously shown to be more efficient at stimulating TNF-α secretion by murine dendritic cells, as compared to GPs containing surface-linked LPS.

The F. tularensis protein or variant or fragment thereof may be any suitable protein antigen, for example a protein antigen selected from Table 1 below. Proteins may be selected by any suitable technique, for example by bioinformatics analysis.

Preferably, the F. tularensis protein or variant or fragment thereof comprises at least one putative B- and/or T-cell epitope, more preferably a B- and/or T-cell epitope predicted to bind to at least one human Major Histocompatibility Complex molecule. This feature provides a positive indicator from a potential immunogen perspective, and may be selected using bioinformatics analysis.

Alternative or further attributes which may be used alone or in any combination to select potential F. tularensis protein(s) for inclusion in the immunogenic agent are: whether the F. tularensis protein is exposed or surface-located on the bacterium and hence, potentially more likely to be recognised as non-self by a host immune system; specific characteristics of the B- and/or T- cell motifs such as high affinity binding or promiscuous binding to multiple human Major Histocompatibility Complex (MHC) class I and/or class II molecules; and/or the F. tularensis proteins for which expression by the bacterium is up-regulated in an animal during infection, relative to e.g. non-//? vivo environments, or found to be important in the virulence of the bacterium in vivo.

Based on a variety of selection parameters and animal studies described in more detail below, the F. tularensis protein is preferably chosen from FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 or IgIC. The preferred F. tularensis protein or variant or fragment thereof is advantageously encapsulated by GPs and/or used in conjunction with F. tularensis LPS or component thereof encapsulated by GPs.

By way of explanation, the preferred proteins were identified using immunological analysis of a broader range of proteins in mice. The mouse immunogenicity data were used as a primary tool to identify the protein antigens: FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043 and IgIC. The selection of these antigens was primarily influenced by a combination of the development of antigen specific IFNy ELISPOT responses and/or the detection of an antibody response, particularly where an lgG2a bias was observed. The proteins were also analysed using a more comprehensive computational approach. It was shown that the preferred proteins are highly conserved across many isolates of F. tularensis. This approach also identified that 6 of the 7 antigens are good immunogens with strong B-cell and T-cell epitopes. When murine MHC allele binding was compared to human alleles, the predictive value of the murine response was better for FTT0289, FTT0814, and IgIC than for the other proteins. However, IgIC and FTT0071 may elicit a more evasive or suppressive response for some human alleles due to the presence of very common T-cell exposed motifs and cannot be discounted.

More preferably, the F. tularensis protein is chosen from FTT0814, FTT0071, FTT0438, FTT0890 or FTT1043. Mice studies have shown that FTT0071, FTT0814 and FTT0890 induce the strongest antigen-recall IFNy ELISPOT responses in splenocyte cultures. Moreover, FTT071 and FTT0814 have been found to prime strongly for IFNy memory response. Cytokine responses have shown FTT071, FTT0890 and FTT0814 as potent inducers of IFNy responses, and also induce strong IL-6 and TNF-alpha responses. Immunological studies in rats have shown that LPS, FTT0814 and FTT1043 induce strong serum IgG responses. While immunisation with FTT0438 did not result in a detectable IFNy response when splenocytes were re-stimulated with FTT0438 protein, a response was detected in rats when restimulated with LVS lysate antigen preparation. Accordingly, FTT0438 is also a more preferred protein.

Even more preferably, based on overall mouse immunogenicity data, the F. tularensis protein is chosen from FTT0438, FTT0814 or FTT1043. Most preferably, the F. tularensis protein is chosen from FTT0814 or FTT1043.

The inventors used vaccines containing LPS and FTT0814, FTT1043 or FTT0438 in an in-depth immunology study and evaluated protection. Consistent with the immunology study in mice, evaluation of pre-challenge responses in rats vaccinated with the FTT0814-based GP vaccine demonstrated the strongest and most consistent antigen specific IgG response and the strongest T-cell mediated IFNy responses. The inventors subsequently demonstrated that the GP encapsulated F. tularensis antigen combinations protected against an otherwise lethal aerosol challenge of F. tularensis SCHU S4.

Indeed, a GP vaccine delivering Francisella LPS and the FTT0814 protein was able to induce protection in rats against an aerosol challenge. The protection level was higher than that provided by a GP-LPS only vaccine, and resulted in animals that showed no clinical signs of infection. Analysis of the transient clinical signs of the animal groups therefore shows that FTT0814 augments the protection induced by LPS alone.

In the study, the inventors also demonstrated that FTT1043 is capable of stimulating IgG and IFN-γ in rats.

The F. tularensis protein component of the immunogenic agent may consist of or may consist essentially of FTT0814 or a variant or fragment thereof. Alternatively, the F. tularensis protein component of the immunogenic agent may consist of or consist essentially of FTT1043 or a variant or fragment thereof.

Preferably, the F. tularensis protein or variant or fragment thereof is acquired via recombinant expression. Recombinantly expressing the F. tularensis protein or variant or fragment thereof offers the advantage of large-scale protein production. Furthermore, such expression means confers favourable properties on the F.

tularensis nucleic acid sequence to be recombinantly expressed, for example codon optimisation to any desired expression system (e.g. Escherichia coli, yeast).

More preferably, the F. tularensis protein or variant or fragment thereof is acquired by recombinant expression of at least one F. tularensis nucleic acid sequence. The nucleic acid sequence may be a F. tularensis subspecies tularensis or F. tularensis subspecies holarctica nucleic acid sequence. For example, the nucleic acid sequence may be at least one F. tularensis subspecies holarctica LVS nucleic acid sequence. Alternatively, the nucleic acid sequence may be at least one F. tularensis subspecies tularensis SchuS4 nucleic acid sequence.

Preferably, the protein, variant or fragment thereof is at least one protein, variant or fragment thereof which is recombinantly expressed as a fusion protein. This aids expression and purification. For example at least one protein, variant or fragment thereof may be recombinantly expressed as a His₆-tagged fusion protein, glutathione-S-transferase fusion protein, maltose binding protein fusion, calmodulin binding peptide fusion, or streptavidin/biotin-based binding tags. His₆-tagged fusion proteins or glutathione-S-transferase fusion proteins are particularly preferred. The protein, variant or fragment thereof may be recombinantly expressed by Escherichia coli.

Preferably, the F. tularensis LPS or component thereof is acquired via extraction from F. tularensis. More preferably, the F. tularensis LPS or component thereof is extracted from F. tularensis subspecies tularensis or F. tularensis subspecies holarctica. For example, the F. tularensis LPS or component thereof may be extracted from F. tularensis subspecies holarctica LVS. Alternatively, the F. tularensis LPS or component thereof may be extracted from F. tularensis subspecies tularensis SchuS4.

The F. tularensis LPS or component thereof may be extracted via phenol extraction as understood by the skilled person.

Alternatively, the F. tularensis LPS or component thereof may be acquired via recombinant expression.

Preferably, the F. tularensis LPS or component thereof is O-antigen. The F. tularensis LPS or component thereof may be O-antigen acquired by recombinant expression of at least one Francisella tularensis nucleic acid sequence comprising O-antigen coding region. Preferably, the nucleic acid sequence comprising 0antigen coding region is at least one F. tularensis subspecies tularensis or at least one F. tularensis subspecies holarctica nucleic acid sequence. For example, the nucleic acid sequence may be at least one F. tularensis subspecies holarctica LVS nucleic acid sequence. Alternatively, the nucleic acid sequence may be at least one F. tularensis subspecies tularensis SchuS4 nucleic acid sequence. Further preferably, the F. tularensis LPS or component thereof is LPS or component thereof recombinantly expressed by Escherichia coli.

The F. tularensis protein or variant or fragment thereof and the F. tularensis LPS or component thereof may advantageously be provided as a fusion element, for example as a fusion element of recombinantly-expressed Francisella protein displaying the Francisella O-antigen.

The immunogenic agent of the invention is suitable for use in the prevention or treatment of infection by F. tularensis in an animal, the infection typically being by F. tularensis subspecies tularensis or F. tularensis subspecies holarctica. For example, the infection may be by F. tularensis subspecies tularensis SchuS4. Preferably, the animal is a human.

The immunogenic agent may be administered by routes as understood by the person skilled in the art, for example the parenteral route (e.g. intramuscular, intravenous, subcutaneous and so on), the oral route, or the intranasal route.

According to a second aspect, the invention provides a pharmaceutical composition comprising an immunogenic agent as described above in relation to the first aspect, in combination with a pharmaceutically acceptable carrier.

Preferably, the pharmaceutical composition further comprises an adjuvant. Any suitable adjuvant may be used, typically an inorganic or organic adjuvant. Suitable inorganic adjuvants include, but are not limited to, alum, aluminium hydroxide or aluminium phosphate, preferably alum. An adjuvant that promotes a cell-mediated (T-cell) response rather than an antibody response is preferred. Other examples of adjuvants include the TLR7 agonist Imiquimod, QuilA Saponin, QS21 Saponin, QuilA Immune Stimulatory Complexes (ISCOMs), Iscoamatrix ISCOMs or poly IC (TLR3 agonists). Any combination of the aforementioned adjuvants may also be used.

The pharmaceutical composition may be administered by routes as understood by the person skilled in the art, e.g. the parenteral route (e.g. intramuscular, intravenous or subcutaneous), oral route or intranasal route.

According to a third aspect, the invention provides use of an immunogenic agent according to the first aspect or a pharmaceutical composition according to the second aspect as a medicament.

In a fourth aspect, the invention provides use of an immunogenic agent according to the first aspect or a pharmaceutical composition according to the second aspect as a medicament in the prevention or treatment of infection by F. tularensis. Infection by F. tularensis (i.e. the different subspecies of F. tularensis) leads to the disease tularemia. Use according to the fourth aspect of the invention can prevent or substantially ameliorate the effects of tularemia.

Preferably, use according to the fourth aspect is as a medicament for the prevention or treatment of infection by F. tularensis subspecies tularensis SchuS4.

In a fifth aspect, the invention provides use of a F. tularensis protein or any variant or fragment thereof as a medicament, wherein the protein is selected from the group consisting of FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 and IgIC, or any combination thereof. Preferably, the protein is selected from the group consisting of FTT0814, FTT0071, FTT0438, FTT0890 and FTT1043, or any combination thereof. Even more preferably, the F. tularensis protein is chosen from FTT0438, FTT0814 or FTT1043. Most preferably, the protein is FTT0814 or FTT1043, or a combination thereof. The medicament may be for the prevention or treatment of infection by F. tularensis, for example by F. tularensis subspecies tularensis SchuS4.

In a sixth aspect, the invention provides the use of a F. tularensis protein or any variant or fragment thereof as a medicament in the prevention or treatment of infection by F. tularensis, wherein the protein is selected from the group consisting of FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 and IgIC, or any combination thereof.

Preferably, the protein is selected from the group consisting of FTT0814, FTT0071, FTT0438, FTT0890 and FTT1043, or any combination thereof. Even more preferably, the F. tularensis protein is chosen from FTT0438, FTT0814 or FTT1043. Most preferably, the protein is FTT0814 or FTT1043, or a combination thereof. Infection may be by F. tularensis subspecies tularensis SchuS4

According to a seventh aspect of the invention, there is provided a method of producing an immunogenic agent for use in the prevention or treatment of infection by Francisella tularensis in an animal, the method comprising the steps of: providing a F. tularensis protein or any variant or fragment thereof; providing F. tularensis LPS or component thereof; and providing GPs; and combining said components to provide the immunogenic agent.

Preferably, the GPs are provided in the form of a dried powder. Further preferably, the LPS and/or protein are provided in a solution, which may be used to rehydrate the GPs. Preferably, the solution is provided as a water or urea solution. A suitable urea concentration is 6M.

The method may involve the additional step of providing an adjuvant so as to form a pharmaceutical composition.

Preferably, the F. tularensis protein or any variant and fragment thereof and F. tularensis LPS or component thereof is provided in the same solution. Preferably, the solution is provided in a diluent of water or urea, for example 6M urea.

According to an eighth aspect of the invention, there is provided a method of producing a pharmaceutical composition for use in the prevention or treatment of infection by Francisella tularensis in an animal, the method comprising the steps of: providing a F. tularensis protein or any variant or fragment thereof; providing F. tularensis lipopolysaccharide (LPS) or component thereof; providing GPs; combining said components to provide an immunogenic agent, and providing an adjuvant to the immunogenic agent.

Any feature in one aspect of the invention may be applied to any other aspects of the invention, in any appropriate combination. In particular, immunogenic agent aspects and pharmaceutical composition aspects may be applied to use aspects and method aspects and vice versa. The invention extends to an immunogenic agent, pharmaceutical composition, use or method substantially as herein described, with reference to the Examples.

In all aspects, the invention may comprise, consist essentially of, or consist of any feature or combination of features.

The present invention will now be described, with reference to the following nonlimiting examples and Figures in which:

Brief Description of the Figures

Figure 1 shows a graph demonstrating TN Fa secretion by dendritic cells following stimulation by GPs containing core-loaded or surface-linked F. tularensis LPS;

Figure 2 shows a graph of an immunogenicity screen of GP subunit vaccines in mice;

Figure 3 shows a macrophage/splenocyte killing assay to assess the in vitro antimicrobial potential of vaccine induced immunity;

Figure 4 shows a graph of an immunogenicity screen of GP cocktail vaccines in mice;

Figure 5 shows a graph of IgG antibody responses in mice 70 days post immunisation;

Figure 6 shows a graph of interferon-γ (IFNy) ELISPOT responses in mice (n=5) vaccinated with the GP cocktail containing FTT0071, FTT0438, FTT0814, IgIC and LPS;

Figure 7 shows a graph of Cytokine Bead Assay (CBA) cytokine response in mice vaccinated with GP subunit vaccines;

Figure 8 shows a graph of an immunogenicity evaluation of GP vaccines in F344 rats;

Figure 9 shows a graph of ova serum IgG responses in GP vaccinated F344 rats;

Figure 10 shows a graph of LVS-lysate-stimulated IFNy responses in GP vaccinated F344 rats;

Figure 11 shows a graph of pre-challenge immune responses in vaccinated F344 rats;

Figure 12 shows a graph demonstrating protection levels in F344 rats from a lethal aerosol challenge of F. tularensis using GP-based vaccines;

Figure 13 shows a graph of weight change in rats after aerosol challenged with F. tularensis', and

Figure 14 shows a graph demonstrating protection levels in F344 rats from increasing doses of aerosol-delivered F. tularensis using a GP-based vaccine^

Detailed Description

The immunogenic agent or pharmaceutical composition of the invention, or for use in the invention, comprises a F. tularensis protein or any variant or fragment thereof; F. tularensis LPS or component thereof; and GPs. Various studies have been carried out to demonstrate that it is possible to produce an immune response in an animal administered with the immunogenic agent/pharmaceutical composition of the invention, and to select preferred protein antigens.

Any suitable F. tularensis protein or variant or fragment thereof may be used, preferably a F. tularensis protein or variant or fragment thereof comprising at least one putative B- and/or T-cell epitope, more preferably a B- and/or T-cell epitope predicted to bind to at least one human Major Histocompatibility Complex molecule.

For the purposes of demonstrating the invention, the protein antigens listed in Table 1 were chosen. In general, possible proteins may be selected by any suitable technique, for example by bioinformatics analysis.

1. Production of immunogenic agents

Protein purification

The F. tularensis protein antigens evaluated by the inventors as potential components of an immunogenic agent are summarised in Table 1.

Table 1: F. tularensis proteins tested

Protein FTT No.	Putative protein function
FTT0071	Citrate synthase, GltA
FTT0143	L7/L12 50S ribosomal protein
FTT0209	Homology with pneumococcal surface antigen A, PsaA
FTT0239	MurC/UDP-N-acetyl-muramate:alanine ligase
FTT0289	Putative lipoprotein of unknown function
FTT0374	CTP synthase
FTT0438	UDP-N-acetylmuramate:L-alanyl-gamma-D-glutamyl-mesodiaminopimelate ligase/murein peptide ligase
FTT0464	L-asparaginase II ansB/periplasmic L-asparaginase II precursor
FTT0468	Peptidyl-prolyl cis-trans isomerase
FTT0482	Unknown
FTT0540	Unknown
FTT0547	Unknown
FTT0721	Catalse peroxidase, KatG
FTT0724	Penicillin binding protein (D-alanyl-D-alanine carboxypeptidase)
FTT0814	Unknown
FTT0890	Type IV pilin protein, PilA
FTT0901	Outer membrane 17 kDa lipoprotein LpnA orTul4-A
FTT0904	Outer membrane protein Tul4-B
FTT0918	Outer membrane protein YapH-N
FTT1043	Macrophage infectivity, Mip
FTT1161	Adenylate kinase
FTT1357 / FTT1712	23 Kda protein, IgIC (two loci present in F. tularensis subspecies tularensis SchuS4)

FTT1416	Putative lipoprotein of unknown function
FTT1419	Unknown
FTT1425	NADH oxidase
FTT1696	Heat shock protein 60, Hsp60
FTT1754	Phosphate acetyltransferase
FTT1768	Endochitinase/chitinase family protein

IgIC clone FtCD00062598 was sourced from the DNASU plasmid repository (https://dnasu.org/DNASU/), but the majority of expression plasmids were produced by the UK Defence Science and Technology Laboratory. The majority of the proteins were expressed recombinantly as His₆-tagged proteins, with the exception of FTT1078, FTT0814, FTT0890 and the IgIC protein, which were expressed as glutathione-S-transferase (GST) fusions in E. coli BL21, with GST tags subsequently cleaved. E. coli BL21 DE3 pLysS (Invitrogen) harbouring recombinant pGEX-4-T3 or pCRT7/NT-TOPO plasmids were cultured in LB-broth containing 50 mg/ml ampicillin, 30 mg/ml chloramphenicol and 1 %w/v glucose. Cultures were grown with shaking (180 rev min'¹) at 37°C to an A₆oonm of 1.0 prior to induction with 1.0 mM IPTG. Cultures were incubated for a further 4 hours, followed by harvesting by centrifugation at 18 600 g for 15 mins.

For initial purification of His-tagged proteins, cell pellets were resuspended in Bugbuster® (MerckMillipore) (5 mL per g of cell pellet), incubated for 20 min at room temperature followed by centrifugation at 20,000 g for 1 hour. Soluble proteins were added to Ni Sepharose 6 Fast Flow (GE Healthcare) equilibrated with Bugbuster® and incubated rolling for 2 hours. The resin was then packed into a column and washed with 20 mM sodium phosphate, 500 mM NaCI, 10 % glycerol pH7.7 containing 50 mM imidazole. Bound protein was eluted with 200 mM imidazole in the same buffer. For insoluble proteins following treatment with Bugbuster® pellets were washed with 1 % Triton-Xi 00 in PBS, followed by 1 M NaCI prior to resuspension in 8 M urea in phosphate buffered saline (PBS). Following incubation overnight the suspension was centrifuged at 20,000 g for 1 hour and the supernatants purified as above but with all buffers containing 8 M urea. Subsequently, His-tagged proteins were purified as described above, but with an additional wash with 0.1 % triton-X114 in PBS whilst bound to the Ni Sepharose 6 Fast Flow (GE Healthcare) to remove endotoxin.

For purification of GST tagged proteins, cell pellets were resuspended in PBS (5 mL per g of cell pellet), and lysed by sonication followed by centrifugation at 27,000g for 30 min. The supernatant was loaded onto a GSTrap HP column (GE Healthcare) equilibrated with PBS and washed to baseline with PBS. Thrombin protease (GE Healthcare) 80 units per mL bed volume in PBS was loaded onto the column and incubated overnight. Cleaved protein was washed from the column with PBS.

Affinity purified soluble proteins were dialysed into 20mM Tris pH 7.5 and insoluble ones into the same buffer containing 8M urea prior to loading onto a 1 mL CaptoQ column (GE Healthcare) equilibrated with the respective buffers. Columns were washed with the same buffer plus 0.01 mM NaCI prior to elution with 0.01 to 1 M NaCI gradient in the same buffer. Purified insoluble proteins were refolded following concentration to 3 to 5 mg/mL and diluted in 10 volumes of refolding buffer (PBS, 400 mM NaCI, 20 % glycerol, 0.5 mM EDTA, 0.1 % Tween) and then dialysed into PBS plus 0.5 M urea.

As required, endotoxin levels were depleted by passing through Detoxi-gel™ endotoxin removal columns (Thermo Scientific). The protein concentration was assessed by Bicinchoninic Acid (BCA) assay (Thermo Scientific), and purity assessed by SDS-PAGE and densitometry following Coomassie staining.

Polysaccharide purification

For initial GP formulation and optimisation experiments, LPS was extracted from the LVS strain. Briefly, LPS was extracted from freeze-dried bacteria with 45% phenol at 67°C. The resulting pellet was re-extracted and the water phases from the two extractions were dialysed against water for 3 days. The resulting solution was ultracentrifuged and treated with RNase and proteinase K. O-antigen was isolated from LPS by hot acid hydrolysis in 5% acetic acid. SCHU S4 derived LPS was generously provided as a gift by W. Conlan, Health Canada. The SCHU S4 derived LPS was used in all GP animal vaccination experiments.

Glucan particles

GPs were prepared from S. cerevisiae (Fleischmann’s baker’s yeast) using a series of alkaline and acidic extraction steps. Following centrifugation and washing in water, S. cerevisiae was subjected to two rounds of hot alkali extraction by heating for 1 hour at 90°C in 1 M NaOH. The particles were suspended in water at pH 4.5, heated at 75°C for 1 hour, and then successively washed with water (three times), isopropanol (four times), and acetone (two times). GPs were dried to yield a freeflowing light tan powder. To count the GPs for GP vaccine formulations, a 10 pg/ml suspension of particles in 0.9% saline was lightly sonicated, counted using a hemocytometer, and then kept in aliquots at -20°C until use. One microgram of GPs contains approximately 5x10⁵ particles.

GP vaccine formulations

GPs containing encapsulated core-loaded Francisella antigen complexed with ovalbumin (OVA) and Torula yeast RNA were prepared as follows. 10 mg of dry GPs were swollen with 50 pl of 10 mg/ml Francisella antigen dissolved in water or 6 M urea at ambient temperature to minimally hydrate the GPs, allowing the soluble antigen to diffuse into the hollow GP cavity. The samples were then frozen at -80°C and lyophilized. After lyophilization, the same procedure was repeated to load the selected albumin into the particles. To maximize Francisella antigen and albumin encapsulation into the GP shells, the dry GP Francisella antigen-albumin formulations were swollen, mixed with 25 pl of sterile water and lyophilized. To trap the antigen and albumin inside the GPs, the dry GP antigen-albumin formulations were heated to 50°C and swollen with 50 pl of 25 mg/ml yRNA (derived from torula yeast, type VI) in 0.15 M NaCI for 30 min, and then 10000 pl of 10 mg/ml yRNA was added for 1 hour at 50°C to complete the complexation reaction, trapping both the Ft antigen and albumin inside the GPs. The suspension was centrifuged and washed three times in 0.9% saline, and particles were resuspended in 70% ethanol, incubated 30 min at room temperature to sterilize, aseptically washed three additional times in sterile 0.9% saline, resuspended, counted, diluted to 1x10⁹ particles/ml in sterile 0.9% saline, and stored at -80°C. To calculate the amount of protein encapsulated into the GPs, rhodamine-labelled bovine serum albumin (BSA) or OVA was prepared by reaction with rhodamine B isothiocyanate (RITC) and used as a fluorescent tracer to estimate albumin incorporation into the GP particles as compared to the unbound in the saline washes. Typically, albumin encapsulation efficiency was >95%.

To produce GPs with LPS on the particle surface, an avidin bridge was used. Briefly, 5x10⁹ biotinylated GPs were incubated with 500 pl of 1 mg/ml avidin at 4°C for 1 hour. The avidin-GPs were extensively washed in 0.9% saline to ensure that the unbound avidin was removed. F. tularensis LPS was biotinylated with biotin hydrazide using an excess of biotin hydrazide to LPS ranging from one- to five-fold. Next, the avidin-GPs were incubated with 100 pl of 5 mg/ml biotinylated Francisella LPS for one hour at 4°C. GPs were extensively washed in 0.9% saline to ensure that the unbound biotinylated Francisella LPS was removed.

To confirm encapsulation efficiency and demonstrate antigen identity inside of the loaded GPs, 10% SDS-PAGE analysis was undertaken. For each GP vaccine, input antigen, GPs containing core loaded antigens and OVA and supernatants from the first wash after completing the loading reaction were evaluated to provide a test for successful antigen loading and antigen identity. Sterility of the GP vaccines was confirmed by culture of an aliquot.

Tumor Necrosis Factor a (TNFa) bioassay

Bone marrow derived dendritic cells (BMDCs) were generated as follows. Bone marrow cells obtained from the tibiae and femurs of 8- to 12-wk-old mice were cultured in R10 medium supplemented with 10% GM-CSF conditioned medium from the mouse GM-CSF-secreting J558L cell line. Cells were fed with fresh GM-CSFsupplemented R10 on days 3 and 6. On day 8 adherent cells were collected and plated in 96-well plates, at a density of 10⁵ cells/ml (100 μΙ) in 96 well plates using GM-CSF-supplemented R10 media. Samples (10 μΙ) were added to BMDCs and incubated overnight at 37°C, 5% CO₂ to stimulate TNFa secretion. PBS and empty GPs (5 particles/cell) served as negative control, E. coli LPS (100 ng/ml) served as positive control. The supernatant was assayed for TNFa using an ELISA kit for TNFa (Ebioscience) following manufacturer’s instructions. Murine BMDCs were stimulated with titrated amounts of biotinylated F. tularensis LPS (100-0.001 g/ml), 1x10⁶ GPs containing core-loaded F. tularensis LPS (GP-Ft LPScore) or 1x10⁶ surface-linked F. tularensis LPS (GP-Ft LPS-surface) to produce TNFa. TNFa levels were measured by ELISA. The GP-Ft LPS-core and GP-Ft LPS-surface formulations were tested in duplicate.

Initial expression analysis showed some of the proteins to be relatively insoluble. As a result, standard GP loading conditions were modified to include 6 M urea, which was subsequently removed by washing away the soluble urea from the GP encapsulated antigen-serum albumin-yeast RNA trapped complexes with saline. Efficient GP particle loading with proteins was confirmed using three different methods. Firstly, incorporation of a tracer rhodamine-labelled OVA was monitored to estimate GP antigen encapsulation efficiently: routinely greater than 95% incorporation of the fluorescent albumin protein was observed. Secondly, differential interference contrast (DIC) and fluorescent microscopy was used to qualitatively demonstrate antigen-encapsulation inside the hollow GP cavity. Finally, SDS-PAGE was used to confirm F. tularensis antigen identity, loading efficiency and integrity after GP loading.

The loading of GPs with Francisella LPS was evaluated by two methods, either core loading inside the GPs or on the GP surface via a streptavidin-biotin linkage. The efficiency of the respective loading procedures was assessed quantitatively by indirectly measuring the incorporation of a fluorescent LPS tracer in the unbound wash fractions. The incorporation of LPS inside or on the surface of GPs was qualitatively assessed by confocal microscopy. The GPs core labelled with both Dylight 633-OVA and DTAF-LPS showed punctate red and green fluorescence inside the GPs indicating efficient core loading of the respective fluorescently labelled antigens. In contrast, surface loading of DTAF-LPS resulted in broad green fluorescence around the GP particles with only the core loaded Dylight 633-OVA showing punctate fluorescence.

The immunomodulatory potential of GP core-loaded and surface-linked F. tularensis LPS formulations was evaluated by testing their ability to stimulate bone marrow derived dendritic cells (BMDCs) to secrete TNFa. As shown in Figure 1, GP coreloaded F. tularensis LPS formulations were more immunostimulatory in this particular assay than surface-linked F. tularensis LPS GP formulations, or free biotinylated F. tularensis LPS. Therefore, core-loaded Francisella LPS was selected for further evaluation as a component of the GP delivered vaccine.

2. Evaluation of route of delivery of immunogenic agents based on GP particles

MS Imaging Study

To evaluate the effect of administration route on subsequent dispersal of the GP vaccines, Balb/c mice in groups of 3 were immunised by either the subcutaneous or intramuscular route with either Cy7 labelled GPs (Cy7-GP-OVA alone or with Ft LPS (Cy7-GP-OVA+LPS or GPs containing Cy7 labelled OVA (GP-Cy7-OVA) alone or with Ft LPS (GP-Cy7-OVA+LPS). The GPs were administered on day 1 and mice were imaged daily for 5 days. For imaging, mice were anaesthetised using 0.78 mg ketamine-medetomidine (Ketaset, Fort Dodge Animal Health Ltd, UK) and 0.015mg Domitor (Elanco, UK) given in a total volume of 150 pl intraperitoneal. Mice were imaged using an MS Spectrum (Caliper, Perkin Elmer, USA) and images captured and analysed using the Living Image 4.5 software. Fluorescent signal was detected at excitation 710 nm and emission 760 nm. After imaging, mice were recovered using 0.05 mg in 100 pl of intraperitoneal-delivered Atipamezole (Antisedan, Janssen Animal Health, UK). Whilst under anaesthetic mice were kept in a warming box and closely observed until fully recovered. After imaging on day 5 mice were culled by cervical dislocation and lung, liver, spleen and selected lymph nodes (popliteals and inguinals) were also imaged separately.

After intramuscular administration, signal intensity appeared to increase from day 1 to day 3 before declining, which may reflect an initial quenching effect by the GPs concentrated in the depot at the site of injection. After subcutaneous administration, a gradual decline in signal intensity was observed, probably as the particles dissipated from the injection site. There was no signal detected at other bodily locations by MS during the study or in individual organs or lymph nodes removed post-mortem on day 5. Signal intensity at the immunisation site after day 5 was comparable for both routes of administration.

3. Strains and culture conditions

F. tularensis LVS was derived from an original NDBR 101, Lot 4 vaccine ampoule produced during the 1960s, which had been stored at -20°C. Bacteria were reconstituted according to manufacturer’s instructions, cultured overnight at 37°C on supplemented blood cysteine glucose agar (BCGA) and harvested into sterile phosphate buffered saline (PBS, pH 7.2) and stored at -80°C as single use aliquots. F. tularensis SCHU S4 was cultured at 37°C on BCGA, or modified cysteine partial hydrolysate (MCPH) broth. For preparation of cultures for aerosol infection studies, F. tularensis SCHU S4 was first grown for 24 h on BCGA and the harvested bacteria then used to inoculate MCPH broth which was incubated for a further 48 h at 37°C, shaking at 180 rpm. To allow bacterial enumeration, cultures were serially diluted in PBS and plated on BCGA. BCGA plates were supplemented with lincomycin, colistin sulphate, amphotericin B and trimethoprim (LCAT) selective supplement (Thermo Scientific) to aid bacterial enumeration from animal tissues. Escherichia coli were cultured on Luria Bertani (LB) plates or broth. Unless stated all chemicals were purchased from Sigma-Aldrich.

4. Animal studies

All animal procedures were performed in accordance with the United Kingdom Animals (Scientific Procedures) Act 1986. The animal studies were also performed under the authority of an application approved by the Animal Care and Use Review Office (ACURO). Animals were randomly allocated into cages upon arrival and housed under a 12-h light/dark cycle with free access to food and water. Female 6 to 8-week old Balb/c and C57BI/6 mice were supplied by Charles River Laboratories, UK. For intramuscular dosing, mice were immunized with 50 pl in each hind limb (100 pl per dose). Subcutaneous and intraperitoneal dosing typically used 100 pl volumes. Mice received three immunisations 2 weeks apart. Six weeks following the final vaccination, mice were either culled by cervical dislocation for immunological screening or challenged with F. tularensis SCHU S4 via an intraperitoneal delivery route (100 pl). Mice were checked at least twice daily following challenge, and clinical signs were recorded for each mouse. Animals that were moribund and deemed incapable of recovery were culled according to pre-determined humane end-point criteria.

All statistical analysis of animal study data was performed in Graphpad Prism software. All data sets were tested for normality and appropriate statistical tests applied. For the analysis of bacterial burdens in the tissues from rats, data was log transformed and a value of one was added to all values to allow comparative analysis of data where no bacteria were detected.

5. Selection of F. tularensis proteins for efficacy testing

In efficacy studies, it is not possible to test all potential antigens because such experiments are complex and expensive. Accordingly, only certain antigens were selected for efficacy evaluation. Groups of Balb/c mice (n=5) were vaccinated on 3 occasions, 2 weeks apart, with the panel of GP packaged antigens shown in Table 1. The ability of the protein and the LPS-core GP vaccines to induce humoral and cellular memory immunity was then assessed.

Isolation and culture of lymphocytes

Splenocytes were isolated by maceration of rat or mouse spleens through a 40pm cell sieve. For mouse splenocyte isolation, red blood cells (RBCs) were lysed by incubation of cells for 5 minutes in Red Blood Lysis buffer (Sigma-Adrich) following by 2 washes with PBS. Rat splenocytes were isolated directly from macerated spleens without the use of a RBC lysis step. For isolation of rat PBMCs, blood was collected from a tail vein (200-1 ΟΟΟμΙ) into heparin containing collection tubes (Sarstedt), diluted into 2ml of PBS and overlayed onto 2ml of Ficoll-Paque 1.084 gradient density centrifugation media (GE Healthcare, UK). PBMCs from individual rats were isolated by centrifugation at 800g for 40 minutes followed by 2 washes with PBS. Following lymphocyte enumeration, cells were diluted in RPMI1640 medium (Life Sciences, UK) supplemented with 10% Foetal Bovine Serum (Sigma), nonessential amino acids (Life Technologies), 2-mercaptoethanol (Life Technologies), 100U/ml penicillin and 100mg/ml streptomycin sulphate (Life Technologies). Murine splenocytes (1-5 x 10⁵ cells) were added to duplicate wells of either flat bottomed 96well microtiter plates, or to pre-coated murine IFNy ELISPOT plates (BD Biosciences, UK). Rat splenocytes (2.5x10⁵ cells) were added to duplicate wells of flat bottomed 96-well microtiter plates. Rat and mouse PBMC (1x10⁵ cells) were added to duplicate wells of round bottomed 96-well microtiter plates. Splenocytes or PBMC cultures were then incubated at 37°C/5% CO₂ in the presence of media alone, individual protein antigens (5 pg/ml, Dstl), LVS lysate (10 pg/ml, Dstl) or ConA (5 pg/ml, Sigma-Aldrich). For the mouse and rat PBMC cultures only, cell were costimulated with 5pg/ml of anti-mouse CD28 (clone 37.51, BD Biosciences) or anti-rat CD28 (clone JJ319, eBioscience) antibodies respectively. To allow measurement of murine IFNy ELISPOT responses, culture were incubated for 16-20 hours. To allow measurement of cytokines by IFNy ELISA or CBA assay, cultures were incubated for 72 hours.

IgG antibody assays

Antigen specific IgG responses were measured in serum from mice, or plasma from rats. In addition to measurement of total IgG, mouse serum samples were also assayed for antibody isotypes lgG1 and lgG2a or lgG2c depending on mouse strain. High protein binding 96-well microtitre plates were coated overnight with 5 pg/ml of antigen. In addition, selected wells were coated with goat anti-mouse antibody binding fragment (M4155, Sigma-Aldrich) or goat anti-rat IgG (R5130, Sigma-Aldrich) for standard curve calculation. Serial dilutions of mouse serum, rat plasma, or respective mouse or rat IgG standard, were added to the respective wells of coated plates and incubated overnight. Bound antibody was detected using a sequential combination of horse radish peroxidase conjugated goat anti-mouse IgG (10355), or isotypes (STAR123P and STAR133P, BioRad) or goat anti-rat IgG (A9037, Sigma) followed by 3,3',5,5'-Tetramethylbenzidine (Sigma-Aldrich) development substrate. Responses were detected by measurement of OD450nm on a Multiscan Ascent plate reader and antibody concentrations calculated using the Ascent software.

Figure 2 panel A shows lgG1 and lgG2a antibody isotypes recognising antigens in a LVS lysate measured by ELISA in pooled serum collected 2 weeks after the second vaccination. The mean lgG1 and lgG2a responses (ng/ml) are presented as a stacked bar graph showing the mean response for the pooled serum from 5 mice tested in duplicate for each vaccine group.

Figure 2 panel B shows antigen stimulated expression of IFNy measured by ELISPOT. Spleens were harvested from mice 6 weeks after the final vaccination and pooled splenocytes were cultured in triplicate with the antigen corresponding to the immunising GP. Expression of IFNy was measured by ELISPOT and presented as the mean response (+SEM) per 1x106 splenocytes with the medium-alone response subtracted.

Many of the GP vaccines tested induced a strong antibody response to a crude antigen preparation derived from an LVS-lysate (Figure 2, panel A) demonstrating that the GP vaccines were immunogenic and that humoral immune responses could be induced against the payload antigen. The inventors measured the magnitude of lgG1 and lgG2a isotypes as a high ratio of lgG2a antibody to lgG1 as indicative of Th1-biased immune responses.

The antigens varied in their ability to induce the two antibody isotypes (Figure 2). The LPS packaged GPs induced the strongest antibody response which was exclusively lgG1 biased. To provide a direct measure of antigen-specific T-cell mediated immunity, splenocytes from immunised mice were harvested 6 weeks after the third immunisation and stimulated with the corresponding endotoxin-depleted antigen. The production of lnterferon-γ (IFNy) was measured by Enzyme Linked Immunospot Assay (ELISPOT, Figure 2). Approximately half of the antigens demonstrated an ability to induce a cellular immune response, as evidenced by antigen-specific induction of IFNy. Whilst some proteins induced both cellular and humoral responses, for example FTT0071, there were also proteins that stimulated only IFNy responses or IgG responses. Since LPS is a T-cell independent antigen, it was not surprising to note that immunisation with GP-LPS particles did not induce an IFNy response.

J774A.1 macrophage/splenocyte co-culture killing assay

To obtain more information on the protein antigens, the in vitro antimicrobial potential of vaccine induced immunity using a functional killing assay was assessed. Splenocytes from immunise mice were stimulated overnight with heat-killed F. tularensis SCHU S4 cells and then co-cultured with F. tularensis SCHU S4-infected J774A.1 macrophage cells as follows.

Murine splenocytes were isolated as described and cultured in L15 medium (Life Sciences, UK) with either medium alone, 10 ng/ml phorbol myristate acetate (PMA, Sigma) + 1ug/ml ionomycin (Sigma) or heat killed F. tularensis SCHU S4 for 20 hours at 37°C. The murine macrophage cell line J774A.1 (Public Health England, ECACC, 91051511) was propagated in antibiotic-free RPMI1640 medium supplemented with 10% FBS and seeded in 48-well plates at 5x10⁵ cells/well. F. tularensis SCHU S4 was added to J774A.1 cells at a multiplicity of infection of 10. After 90 minutes, the media was removed and L15 containing 50 pg/ml gentamicin (Life Technologies, UK) was added for 45 min. The cells were washed prior to addition of the antigen-stimulated splenocytes from vaccine-immunised mice. The infected J774A.1 cells and antigen-stimulated splenocytes were then cultured in triplicate for 48 h at 37°C. Cells were then lysed in water and intracellular bacteria enumerated by plating serial dilutions in triplicate on BCGA plates.

As shown in Figure 3, bacterial growth in macrophages exposed to stimulated splenocytes was compared to growth in macrophages exposed to medium-treated splenocytes and with splenocytes from naive control mice (Figure 3). This functional killing assay demonstrated a high degree of response variability for immunised animals across all groups including the controls (Figure 3). Whilst splenocytes from mice vaccinated with FTT0814 induced the strongest magnitude of bacterial control with up to a 22 fold reduction in SCHU S4 cultured from infected macrophages, due to assay variability this did not reach significance when compared with splenocytes from naive control mice (non-parametric ANOVA). Consequently, the mouse immunogenicity data was used as the primary tool to identify preferred protein antigens: FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043 and IgIC. The selection of the antigens was primarily influenced by a combination of the development of antigen specific IFNy ELISPOT responses and/or the detection of an antibody response, particularly where an lgG2a bias was observed. Since there are currently no robust correlates of protection known for tularaemia vaccines, antigens were selected that represented a variety of immune response profiles, albeit with a bias toward cell mediated immunity. IgIC was selected for further evaluation because IgIC has been previously reported to induce partial protection in animals.

6. Further F. tularensis protein studies

To assist the further choice of antigens for efficacy evaluation, the seven preferred proteins were evaluated for their ability to induce protective immunity in mice. A C57BL/6 murine model was chosen for further immunological and efficacy evaluation.

Groups of Balb/c mice (n=5) were vaccinated on 3 occasions, 2 weeks apart, with the panel of GP packaged proteins including FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043 and IgIC.

C57BL/6 mice (n=5) were immunised with each of the GP vaccine combinations shown on the x-axis of Figure 4 panel A. The GP “cocktail” vaccine was comprised of a combination of FTT0071, FTT0438, FTT0814 and IgIC. Panel A: lgG1 and lgG2c antibody isotypes recognising antigens in a LVS lysate were measured by ELISA in individual serum samples collected 2 weeks after the second vaccination. The mean lgG1 and lgG2c responses (ng/ml) are presented as a stacked bar graph showing the mean response for each vaccine group (+ SEM). Figure 4 panel B shows antigen stimulated expression of IFNy was measured by ELISPOT. Spleens were harvested from mice 6 weeks after the final vaccination and splenocytes from individual mice cultured with the antigen corresponding to the immunising GP. Expression of IFNy was measured by ELISPOT and presented for both medium (white bars/circles) and antigen (black bars/circles) stimulated splenocytes. The circles are the responses for individual mice and the bars are the mean response for the group. Where the antigen-specific response is significantly elevated relative to the medium control, this is indicated (*** p<0.001, t-test with Holm-Sidak multiple comparison correction).

Mice were vaccinated with a combination of individual GP-encapsulated protein antigen together with GP-LPS. Co-vaccination with GP-LPS was included with all protein subunit GPs to maximise the likelihood of inducing protective immunity. This was on the basis that LPS is the only known subunit that has consistently demonstrated protective potential against low virulence strains of F. tularensis. Controls groups included a non-vaccinated group, a GP-OVA vector control group and a group vaccinated with GP-LPS. In addition, to assess the potential benefits of an antigen combination vaccine, a “cocktail” comprising GPs individually formulated with FTT0071, FTT0438, FTT0814, IgIC and LPS was included in the study. Measurement of lgG1 and lgG2c serum antibodies demonstrated that, with the exception of PBS and GP-OVA controls, all vaccines induced potent antibody responses that recognized antigens in an LVS-lysate (Figure 4, panel A). The lgG2c subclass is associated with Th1 responses in the C57BL/6J mouse.

Figure 5 shows the IgG antibody responses in C57BL/6 mice (n=5) immunised with each of the GP vaccine combinations shown on the x-axis. The GP “cocktail” vaccine was comprised of a combination of FTT0071, FTT0438, FTT0814 and IgIC. lgG1 and lgG2c antibody isotypes recognising antigens in a LVS lysate were measured by ELISA in serum collected 70 days after the first of 3 vaccinations, Serum samples from individual mice were tested in duplicate and lgG1 and lgG2c responses (ng/ml) presented as a stacked bar graph showing the mean response for each vaccine group (+ SEM). A predominantly lgG1 isotype was observed which persisted for at least 70 days. However, this antibody response most likely reflects the inclusion of

LPS in the vaccine formulation, LPS being a sero-dominant lgG1 inducing antigen, as demonstrated in the previous mouse immunogenicity screening experiments.

A cohort of vaccinated mice (n=5) were culled 6 weeks after the third vaccination to assess T-cell immunity. FTT0071, FTT0814 and FTT0890 GP-vaccines induced the strongest antigen-recall IFNy ELISPOT responses in splenocytes cultures (Figure 4). The dominance of these antigens as T-cell antigens was consistent with responses observed previously in Balb/c mice (Figure 2).

Recall responses to all antigens were also assessed using splenocytes isolated from mice that had been vaccinated with the GP cocktail of FTT0071, FTT0438, FTT0814, IgIC and LPS. In this GP-Cocktail group, only FTT0071 and FTT0814 primed for strong IFNy memory responses (Figure 6). Mice were culled 6 weeks following the final booster vaccination and IFNy responses measured in splenocyte cultures stimulated with each of the antigens shown on the x-axis. Responses for individual animals are shown as circles and the mean response for the group is shown as the corresponding grey bars.

Cytokine assays

Mouse IFNy ELISPOT responses were measured in 16-20 hour antigen-stimulated splenocyte cultures using a commercial detection kit (BD Biosciences). The assay was performed in accordance with kit instructions and spot enumeration was performed using an AID automated reader. Detection of IFNy in 72 hour antigenstimulate culture supernatants was detected using commercial (Mabtech) mouse and rat IFNy Enzyme Linked Immunosorbant Assays (ELISAs). The respective ELISA assays were performed in accordance with kit instructions and responses determined by measurement of optical density at 450nm (OD450nm) using a Multiskan Ascent plate reader (ThermoFisher Scientific). IFNy concentrations were calculated from a standard curve generated using the IFNy standards supplied with the respective kits. Antigen stimulated mouse splenocyte cultures were analysed for IL6, IL10, MCP1, IFNy, TNF and IL-12 using a Cytometric Bead Array Assay (CBA). The CBA mouse inflammation assay kit (BD Biosciences) was performed in accordance with kit instructions and samples analysed on a FACS Canto Flow cytometer (BD Biosciences).

Figure 7 shows the cytokine response of C57BL/6 mice (n=5) immunised with combinations of GP formulated LPS and each of the following antigens; FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043 and IgIC. Mice were culled 6 weeks following the final vaccination and cytokine responses measured in splenocyte cultures stimulated with each of the respective protein antigens. Cytokines were detected using a CBA multiplex assay and are shown for IL-6, IL-10, MCP-1, IFNy and TNFa in each of the indicated panels. Responses for individual animals are shown as circles and the mean response for the group is shown as the corresponding grey bars.

A broad range of cytokine responses was measured in 72-hour antigen stimulated splenocytes cultures using a cytokine multiplex bead assay (CBA) (Figure 7). This assay measured lnterleukin-6 (IL-6), IL-10, Monocyte Chemotactic Protein-1 (MCP1), IFNy, TNFa and IL-12. The CBA data confirmed FTT0071, FTT0814 and FTT0890 as the most potent inducers of IFNy responses. These antigens also induced strong IL-6 and TNFa responses, both of these being potent proinflammatory cytokines. IL-12 responses were also measured, but responses were below the assay detection sensitivity for all vaccine/antigen combinations. FTT0814 stimulated the strongest and most consistent IL-10 response.

7. Computational analysis of proteins

A computational platform was used to evaluate FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 and IgIC. This platform integrates predictions of multiple components of the immune system including predictions of protein topology and epitope exposure, predicted linear B cell epitopes, affinity of MHC I and MHC II binding, probability of cleavage by cathepsin, and the frequency of occurrence of the amino acid motifs which engage T-cell receptors (T cell exposed motifs or TCEM) when peptides are bound in either MHC I or MHC II molecules.

An analysis of the predicted B- and T-cell epitopes and relative dominance of these was generated for BALB/c and C57BL6 mice and for 35 MHC I and 28 MHC II human alleles. In addition, frequency patterns of TCEM were analysed to identify those which occur with high frequency, relative to self-protein and immunoglobulin reference databases, and thus potentially attract a large pre-existing cognate T cell population which may indicate immunosuppression. An estimation of the conservation of each protein was made by comparing the presence of one or more members of the same protein family.

Detailed analysis of the seven preferred proteins, and comparison to the larger panel of proteins in Table 1, indicated that a consideration of immune responses in mice had indeed resulted in preferential selection of antigens predicted to be immunogenic, relative to several of the originally-considered protein antigens from Table 1 which were shown by analysis to have a high content of self-like T-cell exposed motifs predicted as likely to elicit an immunosuppressive response. These proteins were FTT0918, FTT1754, FTT0721, FTT1768 and FTT0540. These proteins are each in the top 10% within the F. tularensis proteome in their content of common TCEM per 100 amino acids, where common is defined as a motif occurring at greater than 1 in 1024 immunoglobulin variable regions.

Relative to these proteins, the content of self-like TCEM in the seven preferred proteins was lower. IgIC carries within peptide QGSLPVCCAASTDKG (amino acid index position 187) an amino acid motif S—V~C~AS which is a T cell exposed motif found in 1 in 16 immunoglobulin variable regions and thus a very common occurrence which would encounter a large cognate T cell population. This peptide is predicted to be bound at high affinity by most DQA and DQB alleles, and less strongly by murine allele H2-IAd. FTT0071 has high frequency TCEM at two positions (amino acid index position 23 LPVYSPSLG and 233 QNASTSTVR). These peptides comprise motifs found in at least 1 in 32 immunoglobulin variable regions. However, their impact would be limited to the MHC alleles to which they bind at high affinity. This is predicted, in both cases, to include several human HLA but not murine alleles H2Kb or H2Kd.

With the exception of FTT0890, the selected proteins each have strong predicted B and T cell epitopes. In FTT0890, predicted MHC binding is confined to the transmembrane domain region of the protein.

The seven preferred proteins differed in the degree to which predicted T cell epitopes for the two mouse strains examined would predict a broad human response. However, three of the antigens, FTT0289, FTT0814, and IgIC do have predicted murine responses that closely resemble the overall human pattern as shown for MHC II. The seven preferred proteins are highly conserved across 100 isolates of F. tularensis examined.

8. Immunological evaluation of immunological agents in Fischer 334 rats

Female 6- to 8-week old Fischer 344 (F344) rats were supplied by Envigo, UK. F344 rats were immunised via the subcutaneous route with GP vaccines in a 100 pl volume using the same vaccination schedule as used for mice. In addition, control groups of rats were immunised with a single 100 pl subcutaneous injection of 1x10⁷ CFU LVS. A tail bleed was performed 2-weeks after the final GP-vaccination, or 4 weeks after the LVS vaccination, for immunology. Six weeks following the final vaccination, rats either underwent euthanasia by intraperitoneal administration of sodium pentobarbitone for immunological assessment or they were challenged with F. tularensis SCHU S4 via an aerosol delivery route. The aerosol was delivered by inhalation in a nose-only exposure unit utilising a 3-jet Collison atomiser attached to a contained Henderson Piccolo arrangement to condition the aerosol to 50% (±5%) relative humidity, and controlled by the (AeroMP) Aerosol Management Platform aerosol system (Biaera Technologies L.L.C.). Groups of 5 rats were exposed to the aerosolised bacteria for 10 minutes, with impingement of the aerosol cloud sampled at the midway point of challenge into PBS via an All-Glass Impinger. (AGI-30; Ace Glass, Vineland, NJ). Following challenge, animals were checked at least twice daily, and clinical signs recorded. Individual rat weights were recorded daily. Rats were monitored for 14 days post infection and culled when they reached pre-determined humane end-point criteria (>15% weight loss and/or moribund and deemed incapable of recovery). All rats that survived to the end of the study were euthanized to allow bacterial enumeration in lung and spleens.

Fischer 344 (F344) rats are susceptible to F. tularensis infection, but unlike in mice, LVS can induce protective immunity even against high virulence strains and could thus be considered a more appropriate rodent model for efficacy evaluation of F. tularensis vaccines. Therefore, the inventors transitioned the evaluation of the GPs into a F344 rat model. Initially, a pilot immunogenicity study was performed to determine the hierarchy of immunological responsiveness of the seven downselected F. tularensis antigens in rats.

Groups of rats (n=3) were immunised on 3 occasions, 2 weeks apart, with each of the GP packaged antigens FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043, IgIC and LPS. Unvaccinated and GP-OVA groups were included as controls. Immune responses were measured in blood taken 2 weeks after the third vaccination (shown in Figure 8).

F344 rats (n=3) were immunised with each of the GP vaccine combinations shown on the x-axis of Figure 8 panel A and immune responses measured 2 weeks after the third vaccination. Total IgG antibody recognizing the antigen corresponding to that in the immunizing GP vaccine was measured in serum by ELISA. The response (OD₄50nm) for the individual rats in each group is shown for the range of indicated serum dilutions. Each bar represents the mean response (+SEM).

Figure 8 panel B shows antigen stimulated expression of IFNy measured in PBMCs. PBMCs from individual rats were cultured with the antigen corresponding to the immunizing GP and expression of IFNy measured by ELISA. PBMC antigen recall responses are reported in the non-vaccinated PBS immunised group (white bars/circles) and in the corresponding GP-antigen immunised groups (black bars/circles). The circles are the responses for individual rats and the bars are the mean response for the group.

The results demonstrate that LPS, FTT0814 and FTT1043 induced strong serum IgG responses.

All GP vaccinated rats developed an IgG response to the OVA marker protein indicating that where no response to GP immunised Francisella antigens was observed, this was not as a consequence of deficiencies with the immunisation protocol (Figure 9). Total IgG antibody recognizing Ova was measured in serum by ELISA. The response (OD450nm) for the individual rats in each group is shown for the range of indicated serum dilutions. Each bar represents the mean response (+SEM).

Induction of cell mediated immunity was assessed by measurement of IFNy from antigen stimulated PBMCs. IFNy antigen recall responses were most consistently elevated for the GPs FTT0814 and FTT1043. Whilst immunisation with FTT0438 did not result in a detectable IFNy response when splenocytes were re-stimulated with FTT0438 protein, a response was detected in all 3 rats when re-stimulated with LVS lysate antigen preparation (Figure 10). PBMCs were isolated 2 weeks after the third vaccination and stimulated with a LVS-lysate crude antigen preparation. IFNy in 72 hour culture supernatants was measured by ELISA. The open circles are the responses for individual rats and the bars are the mean response for the group. Where a significant difference between responses in the unvaccinated (PBS) and GP vaccinated groups was observed, this is indicated (* p<0.05, non-parametric Kruskal-Wallis analysis with Dunn’s multiple comparison post analysis test).

Based on the above, FTT0071, FTT0289, FTT0438, FTT0814, FTT0890, FTT1043 and IgIC were shown to be preferred protein antigens. FTT0814, FTT0071, FTTY0438, FTT0890 or FTT1043 were shown to be more preferred.

FTT0814, FTT1043 and FTT0438 were shown to be a particularly preferred choice of proteins and hence, selected for further efficacy evaluation in the F344 model described below.

9. Efficacy evaluation of immunological agents in F344 rats

FTT0814, FTT0438 and FTT1043 were evaluated in a F344 rat aerosol challenge model. The inventors adopted a strategy of immunising rats with the individual GPencapsulated protein antigen together with GP-LPS.

One group of rats was vaccinated with a cocktail of the individually formulated GP encapsulated antigens FTT0814, FTT0438 and FTT1043 and LPS. Controls groups included a sham-vaccinated group (PBS), a GP-OVA vector control group and a group vaccinated with GP-LPS alone. In addition, one group of rats was vaccinated with 1x10⁷ CFU of LVS. Vaccine induced immunity was measured in tail bleed samples collected 2 weeks after the third GP vaccination and prior to challenge. The results are shown in Figure 11.

Immune responses were measured in blood samples collected 2 weeks after the third vaccination with the following GP vaccines; LPS, FTT0814+LPS, FTT0438+LPS, FTT1043+LPS and a cocktail of FTT0814, FTT0438, FTT1043 + LPS. Figure 11 panel A shows antigen specific serum IgG responses in each of the groups vaccinated with individual protein or LPS GPs. The triangles are the response for individual rats and the bar is the mean response for the group. Figure 11 panel B shows serum IgG responses to each of the component GPs antigens in rats immunised with the GP cocktail (data reporting as per panel A). Figure 11 panel C shows antigen stimulated expression of IFNy measured in PBMCs from rats vaccinated with individual protein or LPS GPs. PBMCs from individual rats were cultured with the antigen corresponding to the immunizing protein GP and expression of IFNy measured by ELISA. Antigen recall responses are reported in the non-vaccinated PBS immunised group (white bars/circles) and in the corresponding GP-antigen immunised groups (black bars/circles). The circles are the responses for individual rats and the bars are the mean response for the group. Figure 11 panel D shows antigen stimulated PBMC IFNy responses measured for each of the component GPs antigens in rats immunised with the GP cocktail (data reporting as per panel D). Where a significant difference between responses in the unvaccinated (PBS) and GP vaccinated groups was observed, this is indicated (*p<0.05, multiple ttest corrected for multiple comparisons using the Holm-Sidak method).

Consistent with the previous immunology study, of the 3 protein antigens, only FTT0814 and FTT1043 induced detectable IgG responses. In the rats immunised with the GP cocktail of antigens, FTT0814 elicited the strongest and most consistent IgG response. Little or no antigen stimulated IFNy responses were detected in PBMCs isolated from rats immunised with the individual protein/LPS vaccine combinations. However, in the GP cocktail immunised mice, elevated IFN responses to FTT0814 were detected (Figure 11).

Figure 12 shows the results of F344 rats vaccinated with GP vaccines combinations, LVS or PBS and then challenged with an aerosol of F. tularensis. With reference to Figure 12 panel A, rats were monitored for 14 days and culled if they reach predefined humane endpoint criteria as presented on the Kaplan-Meier survival curve. With reference to Figure 12 panel B, clinical signs were assessed twice daily and an accumulative clinical score was calculated, presented as the mean score for each group during the 14 day post infection period. With reference to Figure 12 panel C, the accumulative total severity score for the entire 14-day post infection period is shown for rats in vaccine groups that survived the lethal challenge (the box represents the interquartile range, the whiskers the range and the horizontal line the median). With reference to Figure 12 panel D, bacterial enumeration of F. tularensis is shown in lungs and spleen of rats (mean CFU/tissue with SEM). For PBS and Ova groups, tissues were processed on the day that animals met their humane endpoint, for all other groups tissues were processed at day 14 post infection.

All rats were challenged with a calculated retained dose of 1.6 x 10³ CFU F. tularensis SCHU S4 delivered by the aerosol route. As shown in Figure 12 panel A, all PBS sham-immunised and GP-OVA vector control rats succumbed to a lethal infection between days 4-11 post infection. An apparent delay in the time to death between the PBS and GP-OVA groups did not quite reach significance (p=0.0551, Log rank test). All GP vaccine groups that received GP-antigen and/or GP-LPS survived the aerosol challenge of F. tularensis and this effect was significant when compared to the survival of the PBS and GP-OVA control groups (p<0.01, Log rank test).

The scoring of clinical signs in the rats identified that transient clinical indications of infection were observed in many of the rats that survived the challenge. These clinical signs were particularly apparent between days 4-6 post-infection (Figure 12 panel B). To explore this further, a comparative analysis of the clinical scores over the 14 day period was performed (Figure 12 panel C). However, whilst transient clinical signs were observed less frequently in the rats in groups vaccinated with either LVS or the GP FTT0814 I LPS combination, the differences between groups were not significant (p>0.05, Kruskal-Wallis nonparametric test with Dunn’s post analysis multiple comparisons).

All surviving rats were culled at day 14 post-challenge and total bacterial loads enumerated in lungs and spleens (Figure 12 panel D). F. tularensis was cultured from the lung and spleen from all GP-vaccinated rats, although at significantly reduced levels compared with either PBS or GP-OVA groups (p<0.001, 2 wayANOVA with Sidak’s post analysis multiple comparison test). The LVS vaccinated group demonstrated enhanced bacterial clearance with an absence of detectable F. tularensis in the spleens, and significantly reduced bacterial burden in the lung compared with all of the GP-vaccine groups (p<0.001, 2 way-ANOVA with Sidak’s post analysis multiple comparison test).

The rats vaccinated with LVS were also the only group where individual rats continued to steadily gain weight after aerosol challenge (Figure 13). The percentage weight change for individual rats in each of the 8 treatment groups is presented in each of the 8 panels. In the PBS and GP-Ova groups, rats were culled when they reached pre-defined humane endpoints of clinical severity or when the weight loss reached 15%.

10. Reproducibility studies

This study was undertaken to determine the reproducibility of the previous efficacy study. In accordance with the standard vaccination schedule, groups of F344 rats were vaccinated on 3 occasions, two weeks apart. The vaccine treatment groups are shown in Table 2 below. The GP-Ovalbumin (Ova) vector control group only contained 4 rats as one animal had to be culled during the study for animal welfare reasons that were unrelated to the study procedures. All rats were challenged with F. tularensis subspecies tularensis SCHU S4 by the aerosol route six weeks after the third vaccination. The clinical condition of the rats was monitored for up to 21 days. Rats were culled prior to the study end-date if they reached a humane endpoint defined either by the severity of the clinical symptoms and/or where they had reached 15% weight loss.

Table 2: Summary of vaccine treatment groups

Vaccine combination	Comment	N
GP-Ova	Vector control	4
GP-FTT0814	Protein control only	5
GP-Ova + GP-LPS	-	5
GP-FTT0814 + GP-LPS	-	5
GP-Ova + GP-O-antigen	-	5
GP-FTT0814 + GP-O-antigen	-	5
LVS	Gold standard positive control	5

Consistent with the previous efficacy study, LVS provided complete protection with respects to survival. In addition, the LVS vaccine group was the only treatment group in which the rats continued to put on weight immediately following infection indicating effective control of clinical symptoms. The aerosol challenge dose received by the LVS vaccinated rats was 2073 CFU, which was comparable to the challenge dose of 1600CFU which was delivered in the previous efficacy study. 2 out of 4 of the rats in the vector control group (GP-Ova) reached the humane endpoint. This data is suggestive of a protective effective being provided by the GP vaccine platform.

At a challenge dose of 5467 CFU, all rats vaccinated with FTT0814 succumbed to the lethal infection by day 12 post-infection. In comparison, rats vaccinated with the GP combination of FTT0814+LPS (according to the invention) demonstrated a significantly enhanced survival advantage (p<0.05, Matel-Cox Log rank test) at a comparable challenge dose of 4567 CFU. This result demonstrates that the GP vaccine combination of LPS and FTT0814 can reproducibly provide significant protection in the F344 challenge model.

A continued development study (Figure 14) demonstrated the efficacy of the FTT0814+LPS GP vaccine at an escalating aerosol dose of F. tularensis SCHU S4. While protection against lethality begins to be lost at a challenge dose greater than 3000 CFU, all groups were protected compared with controls (not shown on the graph). The data demonstrates the repeatability of the FTT0814+LPS GP vaccine in inducing protection against aerosol-delivered F. tularensis SCHU S4.

It will be understood that the present invention has been described above purely by way of example, and modification of detail can be made within the scope of the invention. Each feature disclosed in the description, and (where appropriate) the claims and may be provided independently or in any appropriate combination.

Moreover, the invention has been described with specific reference to an immunological agent and pharmaceutical composition, and associated uses. Additional applications of the invention will occur to the skilled person.

Claims (22)

1. An immunogenic agent for use in the prevention or treatment of infection by Francisella tularensis in an animal, the immunogenic agent comprising:

a F. tularensis protein or any variant or fragment thereof;

F. tularensis lipopolysaccharide (LPS) or component thereof; and glucan particles (GPs), so as to produce an immune response in an animal administered with the immunogenic agent.

2. An immunogenic agent according to claim 1, wherein the F. tularensis protein or variant or fragment thereof and/or the F. tularensis LPS or component thereof is encapsulated by the GPs.

3. An immunogenic agent according to claim 1 or claim 2, wherein the F. tularensis protein or variant or fragment thereof comprises at least one putative B- and/or T-cell epitope.

4. An immunogenic agent according to any one of claims 1 to 3, wherein the F. tularensis protein is chosen from FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 or IgIC.

5. An immunogenic agent according to any preceding claim, wherein the F. tularensis protein is chosen from FTT0814, FTT0071, FTT0438, FTT0890 or FTT1043.

6. An immunogenic agent according to any preceding claim, wherein the F. tularensis protein is chosen from FTT0814 or FTT1043.

7. An immunogenic agent according to any preceding claim, wherein the at least one F. tularensis protein or variant or fragment thereof is a protein, variant or fragment thereof acquired via recombinant expression.

8. An immunogenic agent according to any preceding claim, wherein the F. tularensis LPS or component thereof is LPS or component thereof acquired via extraction from F. tularensis.

9. An immunogenic agent according to any one of claims 1 to 7, wherein the F. tularensis LPS or component thereof is LPS or component thereof acquired via recombinant expression.

10. An immunogenic agent according to any preceding claim, wherein the F. tularensis LPS or component thereof is O-antigen.

11. An immunogenic agent according to any preceding claim, wherein the F. tularensis protein or variant or fragment thereof and the F. tularensis LPS or component thereof are provided as a fusion element.

12. An immunogenic agent according to any preceding claim, wherein the animal is a human.

13. A pharmaceutical composition comprising an immunogenic agent according to any preceding claim, in combination with a pharmaceutically acceptable carrier.

14. A pharmaceutical composition according to claim 13, further comprising an adjuvant.

15. Use of an immunogenic agent according to any one of claims 1 to 12, or a pharmaceutical composition according to claim 13 or claim 14, as a medicament.

16. Use of an immunological agent according to any one of claims 1 to 12, or a pharmaceutical composition according to claim 13 or claim 14, as a medicament in the prevention or treatment of infection by F. tularensis.

17. Use of a F. tularensis protein or any variant or fragment thereof as a medicament, wherein the protein is selected from the group consisting of FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 and IgIC, or any combination thereof.

18. Use of a F. tularensis protein or any variant or fragment thereof as a medicament in the prevention or treatment of infection by F. tularensis, wherein the protein is selected from the group consisting of FTT0814, FTT0071, FTT0289, FTT0438, FTT0890, FTT1043 and IgIC, or any combination thereof.

19. Use according to claim 17 or claim 18, wherein the protein is selected from the group consisting of FTT0814, FTT0071, FTT0438, FTT0890 and FTT1043, or any combination thereof.

20. Use according to claim 17 or claim 18, wherein the protein is FTT0814 or FTT1043, or a combination thereof.

21. A method of producing an immunogenic agent for use in the prevention or treatment of infection by Francisella tularensis in an animal, the method comprising the steps of:

providing a F. tularensis protein or any variant or fragment thereof;

providing F. tularensis lipopolysaccharide (LPS) or component thereof;

providing GPs; and combining said components to provide the immunogenic agent.

22. A method of producing a pharmaceutical composition for use in the prevention or treatment of infection by Francisella tularensis in an animal, the method comprising the steps of:

providing a F. tularensis protein or any variant or fragment thereof;

providing F. tularensis lipopolysaccharide (LPS) or component thereof;

providing GPs;

combining said components to provide an immunogenic agent, and providing an adjuvant to the immunogenic agent.