WO2007122392A1

WO2007122392A1 - Composition comprising a cytotoxic t-cell - inducing adjuvant

Info

Publication number: WO2007122392A1
Application number: PCT/GB2007/001384
Authority: WO
Inventors: James William Wells; Alistair Noble
Original assignee: King's College London
Priority date: 2006-04-20
Filing date: 2007-04-17
Publication date: 2007-11-01
Also published as: WO2007122392A8

Abstract

The present invention relates to a composition for eliciting a strong CD8 cytotoxic T cell response to an antigen. The present invention also provides the use of the composition to elicit a strong CD8 cytotoxic T-cell response and for treating immunological diseases. The composition may also be used to raise a strong CD4 Th1 type response and a Th1-associated antibody response.

Description

Composition

The present invention relates to a composition for eliciting a strong CD8 cytotoxic T cell response to an antigen. The present invention also provides the use of the composition to elicit a strong CD8 cytotoxic T-cell response and for treating immunological diseases. The composition may also be used to raise a strong CD4 ThI type response and a ThI -associated antibody response.

Vaccines are one of the most successful types of medicine in history and have been used to prevent or even eliminate diseases for over 200 years. The most effective vaccines consist of either (i) a live microorganism that is closely related to the one that causes the disease, and therefore provides immunity to both, or (ii) an inactivated form of the pathogenic organism itself, which is not virulent enough to cause disease but elicits an immune response which is protective. Other vaccines have consisted of killed pathogens, which are safe but tend to induce weak immunity that is not long-lasting. The problem with live vaccines is that they could potentially mutate or revert to a dangerous infectious organism, and are difficult to produce and store in a viable state for long periods. Producing inactivated or "attenuated" organisms is problematic and unpredictable. More recent vaccines have therefore used proteins or peptides derived from the infectious organism (subunit vaccines) which are safe, well-defined and can be reliably manufactured in large quantities. A major problem with these types of vaccine is the weak immunity they tend to induce. This is because although the body recognises the proteins/subunits of the vaccine as foreign, it does not see them as dangerous or highly infectious, and so does not mount its most potent type of immune response to fight them.

Immune responses can be defined into two broad categories. A response to a particular organism, or antigen, is usually dominated by one of these types. Type 1 immunity, or cell-mediated immunity, is characterised by the presence of highly inflammatory and cytotoxic (killer) T cells, which specifically recognise and kill infected cells in the body. This type is generally most effective at fighting and eliminating viruses, intracellular parasites and bacteria. Type 2, or humoral immunity, is characterised by high levels of antibodies in the circulation, which neutralise and inactivate microbes or toxins produced by them, or can mark them out for destruction by cells. This type of immunity is particularly important for large extracellular parasites. Although useful in viral infections, antibodies cannot eliminate viruses once they have infected cells. Type 1 immunity is generally induced by the most effective, live vaccines, while the inactivated and subunit vaccines tend to induce type 2 immune responses which are only weakly protective. Thus a major goal in vaccine development is to find a method for making subunit vaccines which stimulate strong type 1 immunity. In particular, it is essential that such vaccines induce large numbers of cytotoxic T cells which destroy the infection and provide long-term immunological memory.

A key cell involved in induction of type 1 immunity is the dendritic cell (related to macrophages) which is the most efficient antigen-presenting cell (APC). Dendritic cells pick up antigens very effectively, process them into peptides and induce specific T cell responses from naϊve populations. In addition, they become activated by non-antigen- specific factors to express costimulatory molecules and the cytokine IL- 12, which enhance T cell activation and development of type 1 immunity. Dendritic cells express a wide variety of Toll- like receptors (TLR' s), which bind to pathogen-associated molecules present on foreign organisms but not mammalian cells. When these receptors are triggered the dendritic cells induce more type 1 immunity. Inflammatory signals such as cytokines also help activate dendritic cells. Together these amplifying pathways are known as "danger signals".

Synthetic peptides or proteins are poorly immunogenic and on their own they do not stimulate strong immune responses. It is therefore necessary to combine the peptide or protein with materials which enhance the immune response to the antigen. These materials are called adjuvants. Typical adjuvants consist of oils, detergents and killed bacteria or molecular components of bacterial cell walls. They function by: (i) retaining the antigen in a complex which is not easily removed, prolonging antigen exposure; (ii) activating immune cells, thus alerting the immune system to the presence of a "dangerous" substance; and (iii) enhancing uptake of antigen into cells, where they can be more effectively presented to the T cells which have the cytotoxic function. There are currently very few adjuvants available which induce a strong type 1 immune response and large numbers of cytotoxic T cells directed to the antigen. The adjuvant most commonly used in humans is aluminium hydroxide, which stimulates type 2, antibody responses. Other adjuvants, such as complete Freund's adjuvant, stimulate very strong antibody responses and also some type 1 immunity. However these adjuvants do not induce very large numbers of cytotoxic T cells and cannot be used in humans as they can cause damaging tissue inflammation. Therefore, the development of new vaccines is being hindered by the lack of a highly effective, type 1 , cytotoxic T cell-inducing adjuvant which is safe for human as well as animal use.

Development of vaccines to new or poorly studied organisms is also hindered by the need to manufacture protein subunits derived from them which elicit protective immune responses. Genes derived from organisms that encode one or more antigens can be incorporated into vectors and injected, so that the antigens are produced within cells of the host. This approach facilitates intracellular expression of antigen, enhancing cytotoxic T cell development, but lacks the activation signals that trigger very strong responses. Another approach is to identify epitopes that are recognised most strongly by cytotoxic T cells and manufacture peptides displaying the epitopes in large quantities. These tend to be very weakly immunogenic even when combined with adjuvant. One reason for this is that CD8 cytotoxic T cells recognise peptides complexed to major histocompatibility complex (MHC) class I molecules, but only respond strongly if non- cytotoxic, CD4+ "helper" T cells are activated at the same time. Since helper T cells recognise different epitopes (peptide complexed to MHC Class II molecules), immunisation with a peptide displaying a single epitope is ineffective.

CD8 T cell responses to virulent infection are known to reach a peak and then decline rapidly to approximately 10% or less of peak antigen-specific CD8 numbers. The remaining cells become a stable, long-lived memory T cell population which is critical for maintaining long-lasting immunity in the absence of further infection. Such immunity can last for decades since the memory CD8 cells replicate themselves to maintain their numbers. If the response is weak to start with, or costimulation is poor, a long-lived stable population is not formed and long-lasting immunity is not induced. Such a response is difficult to induce using subunit or killed vaccines.

Cancer is caused by normal cells of the body transforming into tumour cells, which divide uncontrollably. The immune system protects the body from cancer by recognising tumour antigens (peptides that are highly expressed on tumour cells) or other abnormalities on tumour cells, and killing them. One reason that this system sometimes fails is that tumour cells are difficult to distinguish from healthy cells, and they may actively suppress immune responses against them. Research in animal models has shown that vaccines containing tumour-associated molecules in a form that activates cell-mediated immunity can provide protection from cancer. The problem is that the strength and/or type of immunity that can be induced is often insufficient to eliminate all the tumour cells. This is because the tumour antigens are inherently poorly immunogenic, and/or that they are not administered in a form that stimulates the strongest immunity. Potent type- 1 -inducing adjuvants combined with appropriate tumour antigens or peptides will be of use as tumour vaccines.

Allergic diseases such as asthma, hay fever and atopic eczema are caused by uncontrolled immune responses to innocuous environmental antigens (dust mites, pet hair, pollen, i.e., allergens). Harmful immune responses are characterised by CD4+, Th2 type T cells and by IgE antibody production. Current immunotherapy for treating allergies uses whole antigen, which can reduce allergic symptoms but is hazardous as it can trigger IgE-mediated anaphylaxis. MHC Class Il-restricted peptide vaccines are being developed, which do not trigger IgE but could activate Th2 cells, which cause allergic inflammation. However at present the main treatment for diseases such as asthma is administration of immunosuppressive drugs. These suppress symptoms but do not cure the disease, are not effective in all cases, and have harmful side-effects.

It has been shown that CD 8 T cells also respond to allergens but switch immunity towards a type 1 response, which antagonises the allergic response. The inventors' data suggest that immunotherapeutic vaccines targeting the Class JJCDS T cell axis is beneficial in the treatment of allergic disease. Vaccines using MHC Class I-restricted peptides derived from allergens, and which stimulate strong CD8 T cell responses, require a potent type- 1 -inducing adjuvant in order to be effective, since they would normally be used in established disease.

There is therefore a need for a composition that can be used with a protein or a peptide displaying an antigenic epitope to induce a strong type 1 immune response.

The use of a monoclonal anti-CD40 antibody in combination with a toll-like receptor (TLR) agonist has been used to increase the immunogenicity of a protein (Ahonen et al., J. Exp. Med., 199, 775-784, 2004 and International Patent Application WO 2004/060319). However, the percentage of all CD8 T-cells that are specific for the protein is, at best, around 16%. Accordingly, this method is still of limited value.

The use of interferon-γ (IFN-γ) in combination with the Ribi adjuvant system (comprising monophosphoryl lipid A (MPL) and trehalose dicorynomycolate (TDM)) has been used to generate a ThI immune response against recombinant antigens (Hovav et al., Infection and Immunity, 73, 250-257, 2005).

The use of TLR agonists in combination with CD40 ligands and IFN-γ has been shown to trigger a T helper type 1 -polarizing program in dendritic cells in vitro (Napolitani et al, Nature Immunology, 6, 769-776, 2005).

The use of TLR agonists or CD40 agonists to prime dendritic cells is disclosed but there is no disclosure of a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope (Nakamura et ah, Microbiol. Immunol., 48, 211-219, 2004).

The use of a long peptide in combination with a TLR agonist is disclosed. The combination is said to produce CTL responses (Melief et al., Dev. Biol., 116, 205, 2004).

The potential synergy between various TLR agonists for increasing various immune responses is discussed in Kornbluth et al., (J. Leukocyte Biology, 80, 1-19, 2006). In particular, reference is made to Hokey et al., (Cancer Res., 65, 10059-10067, 2005) which discloses the use of CpG, poly LC and IFN-γ to induce high levels of IL-12p70 secretion in the absence of antigen. The combination of components was also shown to induce Th-I bias when co-cultured with CD4 T-cells. However, the authors found no evidence that the combination of components lead to CD8 stimulation in vivo.

Existing technologies do not result in the production of highly potent CD8-T-cell responses to peptide vaccination. At best antigen-specific CD8 T-cells have been expanded to the extent that they reach 16% of the total CD8 T-cell population. There is a need for a composition enabling the generation of an immune response to an antigen wherein the percentage of antigen-specific CD8 T-cells has been expanded to the extent that they reach substantially more than 16% of the total CD8 T-cell population. It is also desirable that the composition allows the substitution of MHC Class I-restricted peptide antigens with other Class I- or Class II- or Class I and Class II-restricted peptides, whole proteins and potentially killed organisms or crude extracts.

The present invention provides a composition for eliciting a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope and/or a strong CD4 ThI type response to an MHC Class II restricted epitope comprising: two or more toll-like receptor (TLR) agonists; a substance that causes CD40 stimulation; IFN-γ, and a detergent.

It has been found that the use of the composition of the present invention in combination with a MHC Class I restricted epitope leads to a highly potent CD8 T-cell response. In particular, the composition of the present invention can expand antigen- specific CD8 T-cells so that they form over 50%, more preferably over 57%, and even up to 80% of the total CD8 T-cell population.

When the composition of the present invention is combined with a protein containing a MHC Class II epitope, a very strong ThI CD4 T cell response and a rapid ThI- associated antibody response is obtained. The TLR agonist can be any agonist capable of specifically activating a TLR. Preferably the TLR agonist is an agonist of any known TLR (e.g., TLR 1 to 13). Suitable agonists are well known to those skilled in the art. Agonists for TLRs include the inflammatory mediators tri-acyl lipopeptides (TLRl), lipoteichoic acid (TLR2), dsRNA (TLR3), LPS (TLR4), flagellin (TLR5), diacyl lipopetides (TLR6), imidazoquinolines (TLR7, TLR8) and CpGs (TLR9). In particular, kits of TLR agonists are commercially available (see InvivoGen). TLR agonists can be specific for individual TLRs (e.g. TLR3, TLR4, TLR9, etc.) or maybe specific for 2 or more TLRs, e.g., Pam3CysK4 is an agonist for TLRl/ 2 (i.e., TLR 1 and 2, which combine together to form a single signalling complex). Preferably, the composition of the present invention comprises TLR agonists for at least two of the following preferred TLRs: TLR1/2, TLR3, TLR4, TLR7, TLR8 and TLR9. It is further preferred that the composition comprises TLR agonists for activating TLR4 and TLR9; TLR3 and TLR4; TLR3 and TLR9; and TLR1/2 and TLR3.

One preferred TLR agonist is a CpG DNA sequence. CpG DNA sequences are immunostimulatory oligonucleotides that can be produced on a DNA synthesiser. A number of CpG DNA sequences are known to be TLR agonists. TLR agonist CpG 1826, which may be phosphorothioate-modified, is most often used as an immunomodulator in mice and has been shown to be an agonist for TLR9. CpG-28, CpG-7909 and CpG-ODN (1018 ISS) are particularly preferred TLR9 agonists that are suitable for use in humans. A further preferred TLR agonist is monophosphoryl lipid A (MPL), a non-toxic lipid from mutants of Salmonella Minnesota. MPL is a TLR4 ligand and forms a part of the Ribi adjuvant. A further preferred TLR agonist is poly (inosinic-cytidylic) acid (poly LC), a synthetic mimic of viral double-stranded RNA (not species specific), and is an agonist for TLR3. A further preferred TLR agonist is Pam3CysK4 which is an agonist for TLRl and 2.

The substance that causes CD40 stimulation can be any substance that directly or indirectly stimulates CD40. Preferably the substance is a CD40 agonist that directly stimulates CD40. Suitable CD40 agonists are well known to those skilled in the art and include antibody molecules, small molecules, CD40 ligands and transfected cell lines that express a CD40 ligand (Dessureault et al, J. Surg. Res., 125, 173-81, 2005). Preferably the CD40 agonist is an anti-CD40 antibody molecule. The term "antibody molecule" refers to polyclonal or monoclonal antibodies of any isotype, or antigen binding fragments thereof, such as Fv, Fab, F(ab')₂ fragments and single chain Fv fragments provided the antibody molecule is a functional CD40 agonist. The antibody molecule may be a recombinant antibody molecule, such as a chimeric antibody molecule, a CDR grafted antibody molecule or a fragment thereof. Such antibodies and methods for their production are well known in the art. The antibody molecule can be produced in any suitable manner, e.g. using hybridomas or phage technology. One skilled in the art would know how to produce an antibody having specific binding affinity for CD40, see Antibodies: A Laboratory Manual, eds. Harlow et al. Cold Spring Harbour Laboratory 1988. The antibody molecule can be produced from any suitable organism, for example, from sheep, mice, rats, rabbits, goats, donkeys, camels, lamas or sharks or from a library of specificities generated through molecular biology techniques.

Preferably, the CD40 agonist is a monoclonal antibody. Anti-CD40 monoclonal antibodies are commercially available from numerous sources (e.g., Invitrogen). Particularly preferred antibody molecules include SGN-40 (Seattle Genetics), ch5D12 (PanGenetics BV), CHIR- 1212 (Chiron & Xoma), BMS-224819 (Bristol Myers Squibb) and Chi-220 (tereliximab).

Alternatively, the substance causes CD40 stimulation indirectly. Indirect stimulation may be achieved by supplying signals that induce CD40 signalling by other cell types. Preferably the substance that causes CD40 stimulation indirectly is a protein or a peptide comprising a MHC Class II restricted epitope or a glycolipid. It has been found that CD4 T cells can be induced to cause CD40 signalling when an MHC Class II restricted epitope is provided on a helper peptide, such as OVA_323-33P or a whole immunogenic protein, e.g., Keyhole Limpet Hemocyanin (KLH). Peptides from recall antigens, such as PPD or tetanus toxoid, as well as KLH, can be used. The protein or peptide comprising a MHC Class II restricted epitope may only comprise MHC Class II restricted epitopes or may also comprise MHC Class I restricted epitopes. It has also been found that NKT cells recognise glycolipids when they are presented to them in the context of a molecule called CDl, which is present in dendritic cells. The NKT cells then release IFN and activate dendritic cells via CD40. Accordingly, a glycolipid can be used to cause CD40 signalling (see Hermans et ah, J. Immunol., 171, 5140-5147, 2003). Preferably the glycolipid is ce-GalCer (alpha-galactosyl ceramide).

The detergent can be any suitable detergent. For example, the detergent may be Tween and/or trehalose dimycolate (TDM). TDM also acts as an immunostimulator.

Without being bound to a specific theory, it is hypothesised that the detergent allows the one or more TLR agonists of the composition of the present invention to pass through the cell membrane and come into contact with intracellular TLRs.

In one preferred embodiment, the composition of the present invention additionally comprises an oil. The oil can be any suitable oil known to those skilled in the art. hi particular, the oil is preferably a metabolisable oil, such as a lipid, e.g., squalene. It is particularly preferred that the oil and detergent are provided as the Ribi adjuvant system. The Ribi adjuvant system is well known to those skilled in the art and comprises monophosphoryl lipid A (MPL), trehalose dimycolate (TDM), squalene and Tween 80. An additional advantage with using the Ribi adjuvant is that it promotes a

ThI immune response. Furthermore, the Ribi adjuvant comprises MPL which is a

TLR4 agonist.

In an alternative preferred embodiment of the present invention, it is preferred that the composition of the present invention does not comprise an oil. It has been found by the inventors that the presence of an oil is not essential. The absence of oil in the composition will be advantageous since it reduces the possibility of inflammation at the site of injection, renders the composition suitable for mucosal delivery and simplifies preparation of a vaccine because emulsification is not required.

In a particularly preferred embodiment of the present invention, the composition additionally comprises one or more additional cytokines that specifically induce a ThI immune response or suppress a Th2 immune response. Preferably, the one or more cytokines are selected from the group consisting of TNF-α, IL- 12, IL-18, IFN-α and

IFN-(S.

The composition of the present invention is preferably used in combination with a protein or peptide bearing a MHC Class I restricted epitope so that a strong CD8 cytotoxic T-cell response is raised. The protein or peptide may comprise MHC Class I and II epitopes or may only comprise MHC Class I epitopes. Preferably the protein or peptide only bears MHC Class I restricted epitopes. This is particularly advantageous in allergy where it maybe useful to avoid stimulation of allergen-specific CD4 T cells. Furthermore, avoiding CD4 responses to multiple MHC Class II epitopes may conserve the CD8 response as there will be less competition. The protein or peptide may be of any size but is preferably between 5 and 50 amino acids in size, more preferably between 6 and 25 amino acids in size, and most preferably between 7 and 11 amino acids in length. MHC Class I epitopes are generally 7-11 amino acids long and longer peptides need to be processed to shorter ones by dendritic cells. Furthermore, short peptides are easier and cheaper to synthesise.

The term "strong CD8 cytotoxic T-cell response" means that on delivering the composition of the present invention and the epitope bearing protein or peptide to an animal capable of generating a CD 8 cytotoxic T-cell response that epitope specific CD8 T-cells are produced that constitute at least 20%, preferably at least 30%, more preferably at least 40%, even more preferably at least 50% and most preferably at least 55%, of the total CD8 T-cell population. Such a large response will result in the effective treatment or prevention of a disease caused by an agent bearing the epitope. Alternatively it may be desired to induce responses to multiple epitopes in order to prevent evasion of the immune response by the organism/tumour. The level of the response will depend on the immunogenicity of the epitope bearing peptide/protein as well as the strength of the animal's immune system and the number of different epitopes present. The percentage of epitope specific CD8 T-cells can be measured using any suitable technique. For example, MHC/peptide multimer staining, recall CD8 IFN-γ responses to peptide, CTL assays in vivo or in vitro, and protection from infection/tumour formation. Preferably the percentage of epitope specific CD8 T-cells is measured using MHC/peptide multimer staining as described in the examples below. The strong CD8 cytotoxic T-cell response produced will also have the advantage of producing long-lasting immunity.

Alternatively, the composition of the present invention may be used in combination with a protein or peptide bearing a MHC Class II restricted epitope so that a strong CD4 ThI response is raised and, optionally, a ThI -associated antibody response is raised. The protein or peptide may comprise MHC Class I and II epitopes or may only comprise MHC Class II epitopes. Preferably the protein or peptide only bears MHC Class II restricted epitopes. This is particularly advantageous as avoiding CD8 responses to MHC Class I epitopes may conserve the CD4 response as there will be less competition. The protein or peptide may be of any size but is preferably between 5 and 50 amino acids in size, more preferably between 9 and 30 amino acids in size.

The term "strong CD4 ThI response" means that on delivering the composition of the present invention and the epitope bearing protein or peptide to an animal capable of generating a CD4 ThI response that epitope specific CD4 T-cells are produced that secrete large amounts of pro-inflammatory cytokines such as IFN-γ and TNF-α. The term "a ThI -associated antibody response" is used herein to mean that on delivering the composition of the present invention and the epitope bearing protein or peptide to an animal capable of generating a Thl-associated antibody response that large amounts of epitope specific antibody molecules are produced. Such a large CD4 and antibody response will result in the effective treatment or prevention of a disease caused by an agent bearing the epitope. Alternatively it may be desired to induce responses to multiple epitopes in order to prevent evasion of the immune response by the organism/tumour. The level of the response will depend on the immunogenicity of the epitope bearing peptide/protein as well as the strength of the animal's immune system and the number of different epitopes present. The amount of epitope specific CD4 T- cells can be measured using any suitable technique. For example, intracellular cytokine staining or ELISpot. The amount of epitope specific antibody molecules produced can be measured using any suitable technique.

The term "epitope" as used herein means an immunological determinant of an antigen. The epitope is the structure on the antigen against which an immunological response is directed. The epitope is present on an antigenic protein or peptide. The antigenic protein or peptide may be in an isolated form, in the form of a complex, in the form of an organism, preferably a killed organism or in the form of a crude extract. Preferably the antigenic protein or peptide is an isolated form, i.e., in a substantially pure form.

The terms "MHC Class I restricted epitope" and "MHC Class II restricted epitope" are well known terms used in the art to distinguish between epitopes that bind to and are presented by either MHC Class I or MHC Class II molecules.

The protein or peptide may be derived from an infectious agent, a tumour antigen, an allergen or may be a self antigen.

The infectious agent may be prokaryotic, eukaryotic, prion, or viral. Suitable infectious viruses include HIV, Hepatitis A, Hepatitis B, influenza viruses (orthomyoxvirus), Herpes viruses, papovaviruses, Rhaboviruses, vesicular stomatitis virus (VSV)₅ etc. Suitable infectious bacteria include Legionella spp., Bacillus spp., Neisseria spp., Haemophilus spp., Helicobacter spp., Corynebacterium spp., Pneumococcus spp., Salmonella spp., Mycobacterium spp., Chlamydia spp. and Shigella spp. etc. Suitable parasites include Cryptosporidium spp., Toxoplasma spp., Leishmania spp., Theilera spp. etc. and any parasites with an intracellular stage in their lifecycle e.g. Plasmodium spp.

The tumour antigen may be any tumour antigen. Numerous tumour antigens are known to those skilled in the art. Examples of tumour antigens include p53, B7, CEA, ESOl, Her2, Muc-1, OFA-iLRP (oncofetal antigen immature laminin receptor protein), etc.

The allergen may be any allergen known to those skilled in the art. Examples of allergens include grass pollen (e.g., PhI p 5b), tree pollen (e.g., Bet vl), house dust mite (e.g., Der pi), animal dander (e.g., cat FeI dl), moulds, latex, food allergens (e.g., peanut Ara h2, chicken egg ovalbumin and ovomucoid) and bee/wasp venom (e.g., phospolipase A2). The self antigen, namely an antigen from the animal that will receive the composition of the present invention, can be used to produce symptoms of an autoimmune disease, so that the animal can be used as a model in research. For example, the self antigen could be an insulin peptide, and the animal used as a model of diabetes.

Each component of the composition of the present invention can be delivered to an animal capable of raising an immune response simultaneously, sequentially or separately. Preferably, each component is given simultaneously. The composition can be given repeatedly to boost and maintain immunity over long periods, typically at intervals of 10 days or more.

The composition of the present invention can be delivered to any suitable animal, such as a human, livestock or pets. Preferably the animal is a mammal or a bird. In particular, the animal may be selected from the group comprising: human, dog, cat, cow, horse, pig, sheep and birds. It is specifically preferred that the animal is a human.

The composition of the present invention is preferably given to an animal in combination with one or more proteins or peptides bearing a MHC Class I epitope. For example, different epitopes of an infectious agent, different tumour antigens or different allergens may be delivered in combination with the composition. The composition of the present invention may therefore be used to raise a CD8 cytotoxic T- cell response to a number of different epitopes. This may be advantageous when it is not clear what specific agent is causing the disease or if two or more different agents are causing the disease. This may also be advantageous where the agent causing the disease is known to evade the immune system by altering epitopes.

Alternatively, the composition of the present invention may be given to an animal in combination with one or more proteins or peptides bearing a MHC Class II epitope. For example, different epitopes of an infectious agent, different tumour antigens or different allergens may be delivered in combination with the composition. The composition of the present invention may therefore be used to raise a CD4 ThI response to a number of different epitopes. This may be advantageous when it is not clear what specific agent is causing the disease or if two or more different agents are causing the disease. This may also be advantageous where the agent causing the disease is known to evade the immune system by altering epitopes.

The present invention also provides a first vaccine composition comprising the composition of the present invention and one or more proteins or peptides bearing a MHC Class I restricted epitope. This may be used to raise a strong CD8 cytotoxic T cell response.

The present invention also provides a second vaccine composition comprising the composition of the present invention and one or more proteins or peptides bearing a MHC Class II restricted epitope. This may be used to raise a CD4 ThI type response and/or an antibody response. It may also be used as an indirect CD40 agonist.

The protein or peptide bearing the epitope is as defined above.

The composition or vaccine compositions of the present invention may also comprise one or more pharmaceutically acceptable excipients. Suitable excipients are well known to those skilled in the art.

The specific amounts of each component of the composition of the present invention can be determined using standard methodologies and by extrapolating from the specific values used in the example section below. The specific amounts used will depend on a number of factors, including the size and metabolism of the animal to be treated.

The present invention also provides the use of the composition according to the present invention in combination with a protein or peptide bearing a MHC Class I restricted epitope, in the manufacture of a medicament for treating an immunological disorder.

The immunological disorder can be any disorder that is preventable or treatable by raising an immune response comprising a strong CD8 cytotoxic T-cell component. The term "immunological disorder" includes cancers, diseases caused by an infectious agent, e.g., bacterial, parasitic and viral infections, and diseases caused by an allergen. The protein or peptide bearing a MHC Class I restricted epitope is as defined above.

In one embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class I epitope is derived from an infectious agent and the medicament is for treating an infectious disease caused by the infectious agent.

In an alternative embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class I epitope is a tumour antigen and the medicament is for treating a tumour bearing the tumour antigen.

hi a further alternative embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class I epitope is an allergen and the medicament is for treating an allergic disease associated with the allergen.

The present invention also provides a method of treating an individual with an immunological disorder comprising delivering an effective amount of the composition according to the present invention in combination with a protein or peptide bearing a MHC Class II restricted epitope to the individual.

The immunological disorder is as defined above.

hi one embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is derived from an infectious agent and the immunological disorder is an infectious disease caused by the infectious agent.

In an alternative embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is a tumour antigen and the immunological disorder is a tumour bearing the tumour antigen.

In a further alternative embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is an allergen and the immunological disorder is an allergic disease associated with the allergen. The present invention also provides the use of the composition according to the present invention in combination with a protein or peptide bearing a MHC Class II restricted epitope, in the manufacture of a medicament for treating an immunological disorder.

The immunological disorder can be any disorder that is preventable or treatable by raising an immune response comprising a strong CD4 ThI type response. The term "immunological disorder" includes cancers, diseases caused by an infectious agent, e.g., bacterial, parasitic and viral infections, and diseases caused by an allergen.

The protein or peptide bearing a MHC Class II restricted epitope is as defined above.

In one embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class II epitope is derived from an infectious agent and the medicament is for treating an infectious disease caused by the infectious agent.

In an alternative embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class II epitope is a tumour antigen and the medicament is for treating a tumour bearing the tumour antigen.

In a further alternative embodiment of the use according to the present invention, the protein or peptide bearing the MHC Class II epitope is an allergen and the medicament is for treating an allergic disease associated with the allergen.

The immunological disorder is as defined above.

In one embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is derived from an infectious agent and the immunological disorder is an infectious disease caused by the infectious agent. In an alternative embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is a tumour antigen and the immunological disorder is a tumour bearing the tumour antigen.

In a further alternative embodiment of the method according to the present invention, the protein or peptide bearing the MHC Class II epitope is an allergen and the immunological disorder is an allergic disease associated with the allergen.

According to one specific embodiment of the present invention, there is provided a composition for eliciting a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope and/or a strong CD4 ThI type response to an MHC Class II restricted epitope comprising: two or more toll-like receptor (TLR) agonists; a substance that causes CD40 stimulation

IFN-γ, and an oil and detergent carrier.

The composition may also be used to elicit a strong CD4 ThI type response to an MHC Class II restricted epitope and a ThI -associated antibody response to a protein containing a Class II restricted epitope.

The Ribi adjuvant is preferably used to provide one of the TLR agonists (e.g., MPL) and the oil and detergent carrier. Such a composition has been found to be particularly effective at eliciting a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope and a strong CD4 ThI type response to an MHC Class II restricted epitope.

The present invention is now described by way of example only with reference to the following figures.

Figure 1 shows a graph where mice injected with the peptide/adjuvant composition show strong peptide-specific CD8 T-cell expansion in the blood. Naϊve C57BL/6 mice were immunised with OVA/Alum (i/p) on day O, and then immunised with SIINFEKL- peptide in conjunction with the adjuvant composition subcutaneously on day 9. 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are antigen-specific.

Figure 2A is a graph demonstrating that antigen-specific CD8 T-cells induced with the peptide/adjuvant composition show potent cytotoxic activity. Naive C57BL/6 mice were immunised with OVA/Alum on day 0, and then immunised with SIINFEKL- peptide in conjunction with the adjuvant composition on day 9. 7 days later the mice were injected with CSFE-labelled SIINFEEX peptide-pulsed splenocytes, along with control, non-pulsed cells labelled with a higher concentration of CFSE. The following day the presence/absence of target cells in the spleen was determined by flow cytometry. Total numbers of pulsed target cells detected in each sample are shown.

Figure 2B is a graph demonstrating that antigen-specific CD8 T-cells induced with the peptide/adjuvant composition show potent cytotoxic activity. Naϊve C57BL/6 mice were immunised with OVA/Alum on day 0, and then immunised i.d. with SIINFEKL- peptide in conjunction with the adjuvant composition on day 9. On day 21 the mice were injected with equal numbers of CSFE-labelled SIINFEKL peptide-pulsed "target" splenocytes (left-hand peak), and control, non-pulsed cells (right hand peak) labelled with a higher concentration of CFSE. The following day the presence/absence of target cells in the spleen was determined by flow cytometry. Numbers in the profiles represent the percentage of CTL-killing of SIINFEKL-pulsed target cells.

Figure 3A is a graph showing that long-lived CD8 immunity is induced with the peptide/adjuvant composition. C57BL/6 mice were immunised as described for Figure 1 and numbers of antigen-specific CD8 cells were monitored in peripheral blood for up to 90 days, using MHC Pentamer staining. Readily detectable responses were still present (approx 3%) at the end of the experiment.

Figure 3B shows the same graph as Figure 3A except that a recall response to peptide alone is shown following injection on day 104. Mice injected with the peptide/adjuvant composition generate long-lived memory CD8 T-cells. Mice were immunised as described for Figure 1 and numbers of antigen-specific CD8 cells were monitored in peripheral blood for up to 135 days using MHC pentamer staining. Readily detectable responses were still present (approx 3%) at 104 days. Re-challenge with peptide alone resulted in a strong peptide-specific recall response.

Figure 4 A is a graph showing that mice injected with a viral peptide in combination with the composition show strong peptide-specific CD8 T-cell immune responses. Naive C57BL/6 mice were vaccinated twice on days 0 and 10 with VSV peptide in conjunction with the adjuvant composition. On day 18 the level of intracellular IFN-γ released by CD8 T-cells in the spleen in response to co-culture with VSV peptide was determined by flow cytometry. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are producing IFN- γ.

Figure 4B is a graph demonstrating that antigen-specific CD8 T-cells induced with a viral peptide in the adjuvant composition show potent cytotoxic activity. Naive C57BL/6 mice were immunised i.d. on days 0 and 10 with VSV_52-59-peptide in conjunction with the adjuvant composition. On day 20 the mice were injected with equal numbers of CSFE-labelled VSV_52-59 peptide-pulsed "target" splenocytes (left- hand peak), and control, non-pulsed cells (right hand peak) labelled with a higher concentration of CFSE. The following day the presence/absence of target cells in the spleen was determined by flow cytometry. Numbers in the profiles represent the percentage of CTL-killing of VSV_52-59-pulsed target cells.

Figure 5 is a graph showing that mice injected with a tumour-associated peptide in combination with the composition show strong peptide-specific CD8 T-cell immune responses. Naϊve C57BL/6 mice were vaccinated twice on days 0 and 10 with TRP-2 peptide (associated with the B 16, C57BL/6-derived melanoma) in conjunction with the composition. On day 18 the level of intracellular IFN-γ released by CD8 T-cells in the spleen in response to co-culture with TRP-2 peptide was determined by flow cytometry. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are producing IFN-γ. Figure 6 is a graph showing that the anti-CD40 antibody component of the vaccine can be replaced with an MHC-Class II restricted peptide or unrelated immunogenic protein. Mice injected with various peptide/adjuvant composition show strong peptide-specific CD8 T-cell expansion in the blood. Naive C57BL/6 mice were immunised with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-7 and the additional component indicated (days 0 and 9). 10 days later the level of SITNFEKL- specific CD8 T-cells in the blood was measured by Pentamer staining. Numbers in the top right-hand quadrant represent the percentage of total CD 8 T-cells which are antigen-specific.

Figure 7 is a graph showing that IFN-7 significantly enhances vaccine efficacy. Naive C57BL/6 mice were immunised with OVA/ Alum (i/p) on day 0, and then immunised with the composition either with or without IFN-γ subcutaneously on day 9. 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining.

Figure 8 shows the combination of IFN-7, anti-CD40 antibody and secondary TLR agonist CpG 1826, results in a more extensive expansion of peptide-specific CD8 T- cells than individual components alone. Mice injected with various peptide/vaccine- compositions show strong peptide-specific CD8 T-cell expansion in the blood. Naϊve C57BL/6 mice were immunised with OV A/Alum on day 0, and on day 9 with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, and the additional component(s) indicated. 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are antigen-specific.

Figure 9 shows the anti-CD40 antibody component of the vaccine can be replaced with a completely unrelated MHC-Class II restricted peptide derived from Mycobacterium tuberculosis. Mice injected with various peptide/adjuvant compositions show strong peptide-specific CD8 T-cell expansion in the blood. Naϊve C57BL/6 mice were immunised with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-7 and the additional component indicated (days 0 and 9). 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are antigen-specific.

Figure 1OA shows that peptide-speciflc CD4 T-cells are heavily polarised towards a ThI phenotype following vaccination with the vaccine composition. Naive C57BL/6 mice were immunised with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-γ and the additional MHC-class II peptide indicated (days 0 and 9). 10 days later the inguinal lymph nodes were removed and cultured with 2μg of the relevant MHC-class II peptide for 7 days. The level of intracellular IFN-γ and IL-4 released by CD4 T-cells was then determined by flow cytometry. Numbers in each quadrant represent the percentage of total CD4 T-cells which lie within the quadrant.

Figure 1OB shows that protein-specific CD4 T-cells are heavily polarised towards a ThI phenotype following immunisation of naϊve C57BL/6 mice with a vaccine composition comprising lmg OVA protein in conjunction with MPL/TDM adjuvant, CpG 1826, and IFN-γ (days 0 and 9). 14 days later the inguinal lymph nodes were removed and cultured with 2μg of the relevant MHC-class II peptide for 7 days. The level of intracellular interferon-γ and IL-4 released by OVA_323-339-specific CD4 T-cells was then determined by flow cytometry. Numbers in each quadrant represent the percentage of total CD4 T-cells which lie within the quadrant.

Figure 11 shows that stimulation through a range of different TLRs can provide the necessary amplification signal required to boost vaccine efficacy when used in combination with a TLR 4 agonist (MPL present in the adjuvant). Mice injected with various TLR agonist combinations show strong peptide-specific CD8 T-cell expansion in the blood. Naϊve C57BL/6 mice were immunised with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, IFN-γ, anti-CD40 antibody and the additional component indicated (days 0 and 10). 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are antigen- specific. Figure 12 A shows that compared with 2 well known adjuvants, the adjuvant composition, when used with whole OVA protein leads to more extensive expansion of OVA-specific CD8 T-cells. Groups of 5 naϊve C57BL/6 mice were immunised with lmg OVA protein in conjunction with the adjuvant indicated (days 0 and 10). 11 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining.

Figure 12B shows that compared with 2 well known adjuvants, the adjuvant composition, when used with whole OVA protein leads to a stronger ThI pattern of antibody production. Groups of 5 naϊve C57BL/6 mice were immunised with 400μg OVA protein in conjunction with the adjuvant indicated (days 0 and 10). Throughout the course of the experiment serum samples were harvested every 7 days and analysed for the presence of OVA-specific IgGl or IgG2a by ELISA. Numbers in the top right- hand quadrants represent the percentage of total CD8 T-cells which are antigen- specific.

Figure 13 indicates a critical role for MPL and Tween 80 contained within the MPL/TDM (Ribi) adjuvant in the promotion of antigen-specific CD8 T-cell expansion. Naϊve C57BL/6 mice were immunised with SIINFEKL-peptide, IFN-γ, anti-CD40 and CpG 1826 in conjunction with combinations of MPL, TDM, squalene and Tween 80 (days 0 and 10). 10 days later the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Data from 1 experiment, n=4.

Figure 14 shows peptide-specific CD8 T-cell immune responses to poorly immunogenic peptides can be increased by increasing the concentration of peptide in the vaccine. Naϊve C57BL/6 mice were vaccinated twice on days 0 and 9 with the indicated concentration of TRP-2 peptide (associated with the B 16, C57BL/6-derived melanoma) in conjunction with the vaccine composition. A) On day 21 the mice were injected with equal numbers of CSFE-labelled TRP-2 peptide-pulsed "target" splenocytes (left-hand peak), and control, non-pulsed cells (right hand peak) labelled with a higher concentration of CFSE. The following day the presence/absence of target cells in the spleen was determined by flow cytometry. Numbers in the profiles represent the percentage of CTL-killing of target cells. B) The level of intracellular IFN-γ released by CD8 T-cells in the spleen in response to co-culture with TRP-2 peptide was determined by flow cytometry. Numbers in the top right-hand quadrant represent the percentage of total CD8 T-cells which are producing IFN-γ

Figure 15 shows stimulation through a range of different combination of TLR' s can provide the necessary amplification signal required to boost vaccine efficacy. Mice injected with various TLR agonist combinations show strong peptide-specific CD8 T- cell expansion in the blood. Naϊve C57BL/6 mice were immunised with SIINFEKL- peptide in conjunction with IFN-γ, anti~CD40 antibody, Tween 80 and the additional component indicated (Oil-free adjuvant, days 0 and 10). 10 days later the level of SπNFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining

Figure 16 shows that the adjuvant composition, when mixed with whole OVA protein, results in extensive expansion of OVA-specific CD8 T-cells (A). (B) Typically, CD8 T-cell expansion in response to vaccination with whole OVA is around 10-30%. (C) Higher concentrations of OVA lead to clearer evidence of ThI antibody polarisation Naϊve C57BL/6 mice were immunised with lmg OVA protein in conjunction with the additional adjuvant component indicated (days 0 and 10). On days 10 and 20 the level of SIINFEKL-specific CD8 T-cells in the blood was measured by Pentamer staining. Throughout the course of the experiment serum samples were harvested every 7 days and analysed for the presence of OVA-specific IgGl or IgG2a by ELISA. Numbers in the top right-hand quadrant in (A) represent the percentage of total CD8 T-cells which are antigen-specific.

Figure 17 shows that vaccination with the adjuvant composition prevents a primary IgE response. Naϊve C57BL/6 mice were injected with the indicated vaccine composition on day 0. On day 11 the mice were challenged with 400μg/ml OV A/ Alum i.p. and subsequently bled every 7-10 days from the tip of the tail. OVA-specific IgE was measured in the serum by ELISA. Error bars represent SEM.

Figure 18 shows that vaccination with the adjuvant composition suppresses eosinophilia (A) and IL- 13 production (B) in the BAL fluid in a Th2-mediated model of allergic airways inflammation. Naϊve C57BL/6 mice were sensitised with two i.p. injections of 400μ,g/ml OVA/Alum on days 0 and 12. On day 20 the mice were injected with SIINFEKL peptide in conjunction with the adjuvant composition. On days 28 to 33 the mice received a daily intranasal challenge with 50μ.g OVA in PBS. BAL fluid was harvested on day 34 and the infiltration of cells was measured by flow cytometry (n=6). IL- 13 was measured by ELISA (n=12, combined data from two independent experiments). Error bars represent SEM.

Figure 19 shows that vaccination with the adjuvant/peptide composition delays tumour formation and reduces the rate of tumour growth at early time-points in a murine melanoma model. Naϊve C57BL/6 mice (n=8/group) were challenged s.c. with 1x10⁵ viable tumour cells on day 0. Mice were injected i.d. in both flanks on days -16 and -7 (Prophylactic group), or days 3 and 11 (Treatment group), with lOOμg TRP -2 peptide in conjunction with the adjuvant composition of the present invention. Tumour diameters were recorded every 3-4 days and results expressed as Mean Tumour Diameter/group.

Figure 20 shows immunisation with a peptide consisting of a single MHC Class II- restricted epitope in the novel adjuvant formulation induces ThI responses. BALB/c mice were immunised i.d. with lOOμg OVA^323"329 peptide, 25μg anti-CD40, 25μg CpG, lOOng IFN-γ and MPL+TDM emulsion (right panels) or PBS only (left panels). 11 days later draining lymph node cells were cultured with OVA^323"339 (2μg/ml) for 7 days to expand CD4 cells. Cells were then stained for intracellular cytokine production after re-stimulation with anti-CD3/CD28 (lμg/ml each, BD Bioscience; upper panels) or OVA^323"339 and dendritic cells (lower panels). IFN-γ production after peptide re- stimulation indicates that most of the cells induced were antigen-specific.

Figure 21 shows immunisation with the adjuvant formulation induces the release of bioactive IL- 12 in vivo. C57BL/6 mice were immunised with lmg whole OVA protein i.d. with 25 μg CpG, lOOng IFN-γ and MPL+TDM emulsion. Mice were bled after 7 and 14 days and levels of IL- 12 p70 in serum were measured using a commercial ELISA kit. EXAMPLES

Materials and Methods

The components shown to be highly effective are:

(i) MPL + TDM adjuvant (Sigma). 2xl00μl/mouse, injected subcutaneously. This is a stable oil-in-water emulsion containing 2% squalene, 0.2% Tween 80, monophosphoryl lipid A (MPL, a non-toxic lipid from mutants of Salmonella

Minnesota, a TLR4 ligand), and synthetic trehalose dicorynomycolate (TDM, an analogue of trehalose dimycolate from the cord factor of Tubercle bacillus, ligand not known). Peptide/antigen and additional components of the composition (below; in PBS) are mixed 50:50 with this adjuvant (vortexed to an emulsion) before injection.

(ii) CpG DNA (25μg) OR poly LC (50μg). CpG DNA sequences are immunostimulatory oligonucleotides produced on a DNA synthesiser. The inventors have used phosphorothioate-modified CpG 1826, which is most often used as an immunomodulator in mice. It has been shown to be an agonist for

TLR9. Other CpGs are used in other species. Poly LC is poly (inosinic- cytidylic) acid, a synthetic mimic of viral double-stranded RNA (not species specific), and is a ligand of TLR3. MPL, pam3Cys K4, R837 have also been shown to be effective. (iii) Anti-CD40 antibody (25μg). This an agonistic monoclonal antibody which binds to CD40, a costimulatory molecule expressed on B cells and dendritic cells. Ligation of CD40 on dendritic cells, usually by CD40-ligand expressed on activated T cells, is thought to provide signals that amplify APC function, IL- 12 synthesis and "help" for presentation and costimulation of CD8 T cell responses. The inventors have used a rat anti-mouse IgG2a monoclonal antibody, clone 3/23 (Serotec, azide free), or a rat anti-mouse IgG2a monoclonal antibody, clone IClO (R&D Systems).

(iv) IFN-γ. The inventors have used lOOng recombinant mouse IFN-γ per animal.

This is a cytokine that stimulates expression of MHC molecules and activates dendritic cells, favouring IL-12 synthesis. Availability of this cytokine very early in the immune response is thought to be important in skewing immune responses towards type 1. (v) MHC Class ϋ-restricted peptides (alternative to (iii)), related or unrelated to MHC Class I epitope. These may stimulate memory CD4 populations generated by previous immunisations or infections, or may be new antigens.

Details of the components used are set out in Table 1.

The inventors have discovered that a combination of two or more TLR agonists (such as CpG or POLY LC), anti-CD40 monoclonal antibody (to mimic the CD40-Ligand interaction with CD40 on the antigen presenting cell normally provided by CD4+ "helper" T cells), IFN-γ and a detergent (preferably in the form of an adjuvant that promotes ThI immune responses (MPL and TDM Emulsion)), when injected into mice alongside an MHC Class I-restricted peptide, acts on the immune system at the antigen- presenting cell level subsequently resulting in the induction of very potent CD 8 T-cell expansion and cytotoxic activity directed against cells expressing the peptide (see Figure 1 and Figure 2A). Moreover, the expansion of antigen-specific CD8 T-cells in this manner is achieved without the stimulation of CD4 T-cells and results in the production of long-term peptide-specific CD8 memory T-cell populations. Using MHC pentamers and analysis by flow cytometry the inventors have determined that it is possible to induce the initial expansion of peptide-specific CD8 T-cells such that they represent up to 57% (see Figure 1) or even up to 80% (data not shown) of all CD8 T- cells present in the blood after vaccination with the peptide/adjuvant composition.

Combining SIINFEKL peptide and the adjuvant composition generates potent peptide- specific CTL activity. C57BL/6 mice were sensitized to OVA protein though the i.p. injection of 400μg OV A/Alum (day 0). Nine days later (day 9) the mice were vaccinated i.d. with lOOμg/mouse SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-7 and anti-CD40 antibody (the "naϊve" control was sensitized to OVA on day 0 but was not immunised on day 9). On day 21 the mice were injected with SIINFEKL peptide-pulsed (CFSE-low) target cells and unpulsed (CFSE-high) control cells. 24 hours later the presence of CFSE⁺ cell populations in the spleen was determined by flow cytometry (Figure 2B). Target cell lysis in vaccinated mice was >99%. The individual components of the composition have previously been shown to enhance immune responses. However the magnitude of these responses is much less than those demonstrated using the combination of components described herein. The present work demonstrates a strong synergistic effect of the combined components. The strongest CD8 response reported in published data using peptides is around 16% of total CD8 cells, using anti-CD40 and single TLR-ligands. The inventors have induced responses of over 50%. CD8 T cell responses of around 50% magnitude have only been reported in acute live viral infections such LCMV infection of mice. Furthermore, the inventors have shown that using the composition of the present invention, antigen-specific CD8 T cells decline to a plateau level of around 3% (Figures 3 A and 3B) and that long-lived immunity is induced, as indicated by a recall response to antigen in the absence of adjuvant (Figure 3B).

Omission of any of the components of the composition of the present invention results in dramatically weakened responses. The inventors' work suggests that in order to induce maximal MHC class I-restricted cytotoxic T cells, it is not necessary to express the antigen intracellularly as is often assumed. Instead, maximal stimulation of dendritic cells through multiple receptor-mediated pathways (including TLR9/TLR3,

TLR4, CD40 and IFN-γ receptors) are critical to adjuvanticity. Since TLRs are mainly expressed intracellularly, it is likely that the detergent allows TLR ligands to cross cell membranes and access TLRs more effectively, amplifying signalling. Key intracellular signalling molecules that may synergise are Myd88 (TLR9), TRIF (TLR3), TRAF

(CD40), and STAT-I (IFN- γR). Together, these factors clearly result in upregulation of costimulatory molecules, cytokine synthesis (e.g. IL- 12, IFN-α/β) and enhanced antigen cross-presentation required for full CD8 T cell immunity.

Vaccines for infectious agents

The inventors have demonstrated that the ovalbumin-derived peptide that was initially used as a model antigen to develop the composition, can be replaced with peptides of other specificities. The composition can be used to generate powerful CD8 T cell immune responses against any MHC Class I-restricted epitope. The inventors have used a viral peptide derived from vesicular stomatitis virus (VSV), to generate a strong functional CD8 T-cell response in C57BL/6 mice (see Figure 4A). The composition can be used to formulate effective vaccines against virtually any infectious organism in animals or humans. These could contain peptides bearing known immunogenic epitopes, whole proteins derived from the organism, or killed organisms.

Combining a viral peptide derived from vesicular stomatitis virus (VSV_52-59) with the adjuvant composition generates potent peptide-specific CTL activity. C57BL/6 mice were vaccinated i.d. with lOOμg/mouse VSV_52-59-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-γ and anti-CD40 Antibody on days 0 and 10 (the "naive" control did not receive any immunisations). On day 20 the mice were injected with VSV_52-59 peptide-pulsed (CFSE-low) target cells and unpulsed (CFSE-high) control cells. 24 hours later the presence of CFSE⁺ cell populations in the spleen was determined by flow cytometry (Figure 4B). Target cell lysis in vaccinated mice was >99%.

Vaccines that stimulate anti-tumour immunity

The inventors have demonstrated that strong CD8 T cell responses (see Figures 5 and 14) and cytotoxic function (60% lysis or even up to 97% lysis) can be induced by using the composition in combination with a tumour-associated peptide antigen. The inventors chose a peptide which, while derived from a "self antigen, is known to be present at high levels on the Bl 6 tumour cell line (TRP-2, melanoma). Despite its weak immunogenicity the inventors raised responses to this peptide which are much higher than have previously been generated to an equivalent antigen.

Vaccine compositions which contain helper epitope peptides or whole proteins in place ofanti-CD40 antibody

The inventors have determined that the inclusion of the anti-CD40 monoclonal antibody in the vaccine leads to the high level production of total IgE in serum, which might be undesirable where the vaccine is intended for use in allergy. It may also have other side-effects, and could induce antibody responses which neutralise effects in the host. However, the inventors have shown that the anti-CD40 monoclonal antibody can be substituted with a suitable MHC Class Il-restricted peptide to induce CD4 helper T- cell help which provides the signal mediated by anti-CD40. This approach can be used without loss of vaccine potency and without the high level production of total IgE in the serum. The inventors have also shown that the use of whole protein (OVA) will also induce strong responses in the absence of anti-CD40, as long as it contains MHC Class II as well as Class I epitopes. In cases where it is undesirable to use the whole protein or Class II-epitopes (e.g. allergy), an unrelated helper antigen could be used to replace the anti-CD40.

As indicated above, the inclusion of anti-CD40 antibody into the composition is not strictly necessary for efficacious function; however as the ligation of CD40 on dendritic cells is essential for the stimulation of CD8 cytotoxic T-cells, it is necessary to provide an alternative source of CD40 ligation on dendritic cells if anti-CD40 antibody is not included in the composition. The inventors have shown (Figure 6) that CD4 T-cells can be induced to provide this critical signal to dendritic cells when an MHC Class II restricted helper peptide such as OVA_323-339, or a whole unrelated (immunogenic) protein such as Keyhole Limpet Hemocyanin (KLH) that contains MHC Class II restricted helper peptide sequences, is added to the composition in the place of anti- CD40 antibody. The inventors envisage that in humans, peptides from recall antigens such as PPD or tetanus toxoid, as well as KLH, could be used as "helper" antigens.

However, it does not require prior knowledge or inclusion of MHC Class II-restricted helper-peptides if anti-CD40 is used. Short peptide antigens could therefore be synthesised rapidly based on limited knowledge of the genome/proteome of an infectious organism. This would allow for flexible and rapid vaccine development with the potential to induce strong CD8 T-cell immunity to any antigen of choice.

Vaccine compositions which do not contain IFN-y

The inclusion of IFN-γ into the vaccine composition significantly (PO.005) enhances vaccine efficacy (see Figure 7).

CD8 T-cells interact with all nucleated cells within the body, monitoring peptides expressed on MHC Class-I molecules to detect the presence of infected or transformed cells, which are then killed by the CD8 T-cell. As the composition of the present invention stimulates the expansion of antigen-specific CD8 T-cells, it can be used in any circumstance in which a target antigen has been identified or cells have become infected, such as by a virus, or transformed, such as in cancer. It can therefore be used to prevent virtually any infection in livestock, pets or humans, or to prevent or treat cancer and allergic disease. The adjuvant can even be used in combination with self- antigens in order to develop new animal models of autoimmune disease for research purposes.

Vaccine Composition

The combination of IFN-γ, anti-CD40 antibody and a second TLR agonist such as CpG leads to more efficacious vaccine function than when these three components are used individually or in pairs. The inventors have shown that a greater expansion of peptide- specific CD8 T-cells is detectable in the blood of mice which receive IFN-γ, anti-CD40 antibody and CpG (Figure 8) when compared with mice that received IFN-γ, IFN-γ + anti-CD40Ab, or IFN-γ + CpG alone.

Furthermore the inventors have shown that an unrelated MHC Class II restricted helper peptide derived from Mycobacterium tuberculosis ("peptide 25", Ag85B₂₄o_-254) can be used as a "helper antigen" to provide an alternative source of CD40 ligation on dendritic cells (via CD4 T-cells) if anti-CD40 antibody is not included in the vaccine (Figure 9). The use of unrelated "helper" antigens in the vaccine is a particularly important consideration when designing vaccines for use in allergy as they allow the generation of an allergen-specific CD8 response in the absence of corresponding allergen-specific CD4 responses, which are often associated with disease progression.

The inclusion of MHC Class II restricted epitopes in the vaccine leads to the activation and ThI -polarisation of CD4 T-cells. CD4 T-cells from the inguinal (draining) lymph nodes of vaccinated mice were assayed for the production of IFN-γ (indicative of a ThI immune response) or IL-4 (indicative of a Th2 immune response) in response to co- culture with the relevant class II peptide (Figure 10A). CD4 T-cells produced large amounts of IFN-γ in the absence of IL-4, indicating that they are strongly polarised towards a ThI phenotype.

Injection with whole OVA protein and the adjuvant composition polarises CD4 T-cells to a ThI -phenotype. The inclusion of a whole protein containing MHC Class II - restricted epitopes in the vaccine leads to the activation and ThI -polarisation of CD4 T- cells. CD4 T-cells from the inguinal (draining) lymph nodes of mice vaccinated with whole OVA protein and the adjuvant composition were assayed for the production of IFN-γ (indicative of a ThI immune response) or EL-4 (indicative of a Th2 immune response) in response to co-culture with an OVA-derived class II peptide (OVA_323-339, Figure 10B). It is likely that other CD4 clones will have been raised to other MHC class II-restricted epitopes within the OVA protein and that these clones would also secrete IFN-γ upon re-stimulation with the relevant peptides. OVA_323-339-specific CD4 T-cells were found to produce large amounts of IFN-γin the absence of IL-4, indicating that they are strongly polarised towards a ThI (cell-mediated) immuno-phenotype.

Multiple combinations of TLR agonists act synergistically on vaccine efficacy

The synergy displayed by triggering both TLR 4 (MPL present in the adjuvant) and

TLR 9 (CpG added to the vaccine), which results in substantial amplification of antigen-specific CD8 T-cell expansion, is not restricted exclusively to pairing TLR 4 with TLR 9. The inclusion of POLY LC, which ligates TLR 3, or Pam3CysK4, a TLR 1/2 agonist, alongside MPL/TDM also results in a substantial boosting of vaccine efficacy (see Figure 11). Accordingly, a diverse array of agonist combinations targeting a range of TLR' s display the same boosting of vaccine efficacy.

Comparison of the vaccine combination with other well-known adjuvants Two widely used adjuvants are alum and Complete/Incomplete Freund's adjuvant (CF A/IF A). Alum is generally regarded as the "Gold Standard" in human adjuvants and is a Th2-inducing adjuvant while CFA/IFA is said to promote cell-mediated immunity (ThI), both are often used to induce antibody production. CFA/IFA cannot be used in humans as it induces severe inflammatory reactions. In a direct comparison using whole OVA protein, the presently claimed adjuvant, but not alum or CFA/IFA, was shown to result in strong expansion of OVA-specific CD8 T-cells (Figure 12A). Comparison of OVA-specific serum antibody levels revealed that the vaccine composition stimulated the production of high and sustained levels of IgG2a (ThI- associated antibody subclass) coupled with very low levels of IgGl (Th2-associated antibody subclass), indicating that the adjuvant polarises the immune response in favour of cell-mediated (ThI) immunity to a greater degree than alum or CFA/IFA (Figure 12B).

Development of an aqueous, oil-free adjuvant combination with no reduction in efficacy

One preferred constituent of the composition is the commercially available MPL/TDM emulsion which contains MPL, TDM, Squalene (oil) and Tween 80 (detergent). To determine whether all of these components are necessary the inventors used mixtures of these individual components purchased as separate reagents (Figure 13). Preliminary data show that the absence of Squalene and TDM did not reduce vaccine efficacy, suggesting that these two components are not critical for vaccine function. Removal of either MPL or Tween 80 however, greatly reduced vaccine efficacy indicating that these two components alone may be sufficient. The results are important as they show that an aqueous, no oil adjuvant consisting of MPL, CpG, Tween 80 and EFN-γ alone is highly effective. Such an adjuvant could eliminate the problems associated with oil- containing adjuvants in humans, potentially allowing the use of crude antigens such as killed microbes.

Increasing vaccine efficacy to poorly immunogenic peptides Peptide-specific CD8 T-cell responses to weakly immunogenic peptides such as the Melanoma-derived peptide TRP-2₁₈o-i88 can be improved by increasing the dose of peptide in the vaccine. Increasing the concentration of TRP-2 peptide in the vaccine from lOOμg to 400μ.g results in a dose-dependent increase in cytotoxic function from -55-60% (as previously described) up to ~ 97% lysis (Figure 14A). A corresponding increase in IFN-7 release was also shown to be present (Figure 14B). These data show that a combination of the composition with very high doses of tumour-associated peptide antigen can provide effective anti-tumour immunity. It is anticipated that systems that deliver targeted antigen via dendritic cell receptors would also improve the response and/or reduce the antigen dose required.

Multiple combinations of TLR agonists can be used in the adjuvant

MPL, the TLR 4 agonist present in the Ribi adjuvant, is not the only TLR agonist that can be paired with another TLR agonist to result in efficacious adjuvant function (Figure 15). The combination of POLY I:C (TLR 3 agonist) and CpG (TLR 9 agonist), or POLY LC with Pam3CysK4 (TLR 1/2 agonist), results in a similar degree of vaccine efficacy as pairing TLRs 4 and 9. R837 was used as the agonist for TLR7. Some pairs of TLR agonists do appear to be more efficacious than others. TLR 1/2 when paired with TLR 9 for example, leads to significantly weaker antigen-specific CD8 T-cell expansion than coupling TLR 3 with TLR 9.

Comparison of the vaccine compositions with other well-known adjuvants

The strongest expansion of SIINFEKL-specific CD8 T-cells in response to vaccination with whole OVA protein and the adjuvant composition is around 45% (Figure 16A). Typically, however, responses are in the region of 10-30%, and may reduce over time in response to further boosting as CD4 T-cell responses take precedence over CD8 T- cell responses (Figure 16B). Comparison of OVA-specific serum antibody levels following immunisation with a higher concentration (lmg) of OVA protein reveals even stronger polarisation of the immune response in favour of cell-mediated (ThI) immunity as induced by the adjuvant composition (Figure 16C).

Suppression of nascent IgE production by the adjuvant composition

Naϊve C57BL/6 mice were immunised with whole OVA protein in conjunction with MPL/TDM adjuvant, CpG 1826, and IFN-γ (Vaccine - whole OVA), or with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFNPy and an unrelated MHC Class II - restricted helper peptide derived from Mycobacterium tuberculosis ("peptide 25", Ag85B_240-254). 11 days later the mice were challenged with OVA in conjunction with a powerful Th2-promoting adjuvant, Alum, and the subsequent level of OVA-specific IgE in the serum was measured by ELISA (Figure 17). In Naϊve mice, OV A/ Alum induced a peak production of OVA-IgE 10 days after challenge that had declined to basal levels 31 days later. In mice vaccinated with whole OVA and the adjuvant composition, or SIINFEKL peptide and the adjuvant composition (containing peptide 25), the production of IgE was inhibited. The data indicate that the adjuvant composition is capable of priming the immune system towards ThI, such that it can prevent the effects of a subsequent challenge with a normally Th2-inducing adjuvant. Suppression ofTh.2 responses in allergic airways disease by the adjuvant composition To investigate the impact of the vaccine composition would have on Th2-mediated airway inflammation the inventors introduced the vaccine into a murine model of asthma that results in sustained OVA-induced bronchial hyperresponsiveness and eosinophilia. C57BL/6 mice were immunized twice with OV A/alum on days 0 and 12, followed by a daily challenge with 50μg of OVA on day 28 through to day 33. Some mice were immunised i.d. with SIINFEKL-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-γ and anti-CD40 antibody on day 20. The effect of the vaccine on inflammatory cell infiltration was determined using flow cytometric counting of eosinophils (CCR3⁺, Class II¹⁰, CDl lc7CD37B220^"), neutrophils (marker negative granulocytes) and T cells (CD3⁺) in bronchoalveolar lavage (BAL) obtained after culling mice on day 34. Vaccination with SIINPEKL peptide in conjunction with the adjuvant composition resulted in marked suppression of airway eosinophilia (PO.05, Figure 18A). Levels of T-cells in the BAL were significantly increased (P<0.01) following vaccination, however the level of neutrophil infiltration remained unchanged, suggesting that CD 8 cells antagonize the Th2 response without superimposing Thl/Tcl-associated inflammatory pathology. The presence of the Th2- associated cytokine, IL-13, in the BAL fluid was also measured (Figure 18B). Mice that received the vaccine composition also significantly suppressed IL-13 (PO.01), as compared with non-vaccinated mice. Thus allergen-specific CD8 T-cells, induced via immunisation with the vaccine composition, alter the balance of Thl/Th2 effectors in the lung in favour of a ThI profile.

Delay in tumour formation by adjuvant composition.

To investigate the impact of the vaccine composition would have on the generation of anti-tumour immune responses we introduced the vaccine into an aggressive murine melanoma (B 16) model. Naϊve C57BL/6 mice were immunised subcutaneously with B16 cells on day 0. One group of mice received prophylactic vaccinations at -16 and -7 days prior to the adoptive transfer of B16 cells ("Prophylactic") and another group received therapeutic vaccinations 3 and 11 days post adoptive transfer of Bl 6 cells ("Treatment"). Vaccinations were given i.d. and consisted of low-dose (lOOμg/mouse) TRP-2-peptide in conjunction with MPL/TDM adjuvant, CpG 1826, IFN-7 and anti- CD40 antibody. Tumour diameters were recorded every 3-4 days (Figure 19). In both the prophylactic and treatment groups the mice showed a significant delay in tumour establishment as compared with the unvaccinated control group. It is expected that using high-dose (400μg/mouse) TRP-2-peptide in conjunction with the adjuvant composition will lead to even stronger protective tumour immunity as high dose TRP- 2-peptide leads to more powerful CTL induction.

Use of a MHC class II-restricted peptide induces a ThI response in the absence of a CD8 T cell response Figure 20 shows that the use of a peptide consisting of a single MHC class II-restricted epitope (i.e., OVA^323"339) in the absence of an MHC class I-restricted epitope, in combination with the adjuvant formulation of the present invention, induced a ThI response. In particular, on re-stimulating the cells with the peptide or antibodies to CD3/CD28 the level of IFN-7 production indicated that most of the cells were antigen specific.

Immunisation with the Adjuvant formulation induces the release of IL-12 Figure 21 shows that the adjuvant formulation of the present invention induces the release of IL- 12 in vivo. It is believed that the ability of the composition to provide the necessary signals for IL- 12 production may explain its potent induction of both CD8 T cell and ThI -polarised responses.

All documents cited above are incorporated herein by reference.

Table 1

Claims

1. A composition for eliciting a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope and/or a strong CD4 ThI type response to an MHC Class II restricted epitope comprising: two or more toll-like receptor (TLR) agonists; a substance that causes CD40 stimulation; IFN-y, and a detergent.

2. The composition of claim 1, which additionally comprises one or more additional cytokines that induce a ThI immune response or suppress a Th2 immune response.

3. The composition of any one of the preceding claims, wherein the two or more TLR agonists are for any known TLR.

4. The composition of any one of the preceding claims, wherein the two or more TLR agonists is a CpG DNA sequence.

5. The composition of claim 4, wherein the two or more TLR agonists are selected from the group consisting of CpG-28, CpG-7909 and CpG-ODN (1018 ISS).

6. The composition of any one of claims 1 to 2, wherein one of the two or more

TLR agonists is poly inosinic-cytidylic acid.

7. The composition of any one of the preceding claims, wherein the substance that causes CD40 stimulation is an anti-CD40 antibody molecule.

8. The composition of any one of the preceding claims, wherein the substance that causes CD40 stimulation is an anti-CD40 monoclonal antibody.

9. The composition of any one of claim 1 to 6, wherein the substance that causes CD40 stimulation is a protein or a peptide comprising a MHC Class II epitope.

10. The composition of claim 9, wherein the protein or peptide comprises a MHC Class I and a MHC Class II epitope.

11. The composition according to any one of the preceding claims, wherein the detergent is Tween or trehalose dimycolate (TDM).

12. The composition according to any one of the preceding claims, wherein the composition additionally comprises an oil.

13. The composition according to claim 12, wherein the oil is a lipid.

14. The composition according to claim 12, wherein the oil and detergent are provided as a Ribi adjuvant system.

15. The composition according to claim 12, wherein the oil is squalene and the detergent is trehalose dicorynomycolate (TDM) and/or Tween 80.

16. The composition according to claim 2, wherein the additional cytokine is selected from the group consisting of TNF-α, IL-12, IL-18, IFN-alpha, IFN-beta.

17. A vaccine composition comprising the composition according to any one of the preceding claims and a protein or peptide bearing a MHC Class I restricted epitope.

18. A vaccine composition comprising the composition according to any one of claims 1 to 16 and a protein or peptide bearing a MHC Class II restricted epitope.

19. The vaccine composition according to claim 17 or claim 18, wherein the epitope is derived from an infectious agent.

20. The vaccine composition according to claim 19, wherein the infectious agent is a virus, a bacterium or a parasite.

21. The vaccine composition according to claim 17 or claim 18, wherein the epitope is a tumour antigen.

22. The vaccine composition according to claim 17 or claim 18, wherein the epitope is an allergen.

23. The composition according to any one of claims 1 to 16 which additionally comprises one or more pharmaceutically acceptable excipients.

24. The vaccine composition according to any one of claims 17 to 22 which additionally comprises one or more pharmaceutically acceptable excipients.

25. Use of the composition according to any one of claims 1 to 16 in combination with a protein or peptide bearing a MHC Class I restricted epitope the manufacture of a medicament for treating an immunological disorder.

26. Use of the composition according to any one of claims 1 to 16 in combination with a protein or peptide bearing a MHC Class II restricted epitope the manufacture of a medicament for treating an immunological disorder.

27. The use according to claim 25 or claim 26, wherein the epitope is derived from an infectious agent and the medicament is for treating an infectious disease caused by the infectious agent.

28. The use according to claim 25 or claim 26, wherein the epitope is a tumour antigen and the medicament is for treating a tumour bearing the tumour antigen.

29. The use according to claim 25 or claim 26, wherein the epitope is an allergen and the medicament is for treating an allergic disease associated with the allergen.

30. A method of treating an individual with an immunological disorder comprising delivering an effective amount of the composition according to any one of claims 1 to 16 in combination with a protein or a peptide bearing a MHC Class I restricted epitope to the individual.

31. A method of treating an individual with an immunological disorder comprising delivering an effective amount of the composition according to any one of claims 1 to 16 in combination with a protein or a peptide bearing a MHC Class II restricted epitope to the individual.

32. The method according to claim 30 or claim 31, wherein the epitope is derived from an infectious agent and the immunological disorder is an infectious disease caused by the infectious agent.

33. The method according to claim 30 or claim 31, wherein the epitope is a tumour antigen and the immunological disorder is a tumour bearing the tumour antigen.

34. The method according to claim 30 or claim 31, wherein the epitope is an allergen and the immunological disorder is an allergic disease associated with the allergen.

35. A composition for eliciting a strong CD8 cytotoxic T cell response to an MHC Class I restricted epitope and/or a strong CD4 ThI type response to an MHC Class II restricted epitope comprising: a Ribi adjuvant; one or more additional toll-like receptor (TLR) agonists; a substance that causes CD40 stimulation; and IFN-γ.