Methods and reagents for vaccination which generate a cd8 t cell immune response   

This application is a continuation of U.S. application Ser. No. 10/833,745, filed Apr. 28, 2004, which is a continuation of U.S. application Ser. No. 10/686,943 (currently pending), filed Oct. 16, 2003, which is a continuation of U.S. application Ser. No. 09/454,204, filed Dec. 9, 1999 (which issued as U.S. Pat. No. 6,663,871 B1), which is a continuation of International Application No. PCT/GB98/01681, which designated the United States and was filed Jun. 9, 1998, published in English, and which claims priority under 35 U.S.C. §119 or 365 to Great Britain Application No. GB9711957.2 filed Jun. 9, 1997.

The entire teachings of the above applications are incorporated herein by reference.

BACKGROUND OF THE INVENTION

A general problem in vaccinology has been an inability to generate high levels of CD8 T cells by immunization. This has impeded the development of vaccines against several diseases including malaria.

Plasmodium falciparum malaria causes hundreds of millions of malaria infections each year and is responsible for 1-2 million deaths annually. The development of an effective vaccine against malaria is thus a major priority for global public health. A considerable body of immunological research over the last twenty years had led to the identification both of candidate vaccine antigens from the parasite and immunological mechanisms on the host that are likely to protect against infection and disease. However, despite this progress there is still no means of vaccinating against malaria infection which has been shown to be effective in field trials.

A major problem has been the identification of a means of inducing a sufficiently strong immune response in vaccinated individuals to protect against infection and disease. So, although many malaria antigens are known that might be useful in vaccinating against malaria the problem has been how to deliver such antigens or fragments of them known as epitopes, which are recognized by cells of the immune system, in a way that induces a sufficiently strong immune response of a particular type.

It has been known for many years that it is possible to protect individuals by immunizing them with very large doses of irradiated malaria sporozoite given by bites from infected mosquitoes. Although this is a wholly impractical method of mass vaccination it has provided a model for analyzing the immune responses that might be mediating protective immunity against sporozoite infection (Nardin and Nussenzweig 1993).

A considerable amount of research over the last decade or more has indicated that a major protective immune response against the early pre-erythrocytic stage of P. falciparum malaria is mediated by T lymphocytes of the CD8+ ve type (CD8+ T cells). Such cells have been shown to mediate protection directly in mouse models of malaria infection (Nardin and Nussenzweig 1993). Such T cells have also been identified in individuals naturally exposed to malaria and in volunteers immunized with irradiated sporozoite (Hill et al. 1991; Aidoo et al. 1995; Wizel et al. 1995). There is much indirect evidence that such CD8+ T cells are protective against malaria infection and disease in humans (Lalvani et al. 1994).

CD8+ T cells may function in more than one way. The best known function is the killing or lysis of target cells bearing peptide antigen in the context of an MHC class I molecule. Hence these cells are often termed cytotoxic T lymphocytes (CTL). However, another function, perhaps of greater protective relevance in malaria infections is the ability of CD8+ T cells to secrete interferon gamma (IFN-ÿ). Thus assays of lytic activity and of IFN-ÿ release are both of value in measuring a CD8+ T cell immune response. In malaria these CD8+ve cells can protect by killing the parasite at the early intrahepatic stage of malaria infection before any symptoms of disease are produced (Seguin et al. 1994).

The agent of fatal human malaria, P. falciparum infects a restricted number of host species: humans, chimpanzees and some species of New World monkey. The best non-human model of malaria is the chimpanzee because this species is closely related to humans and liver-stage infection is observed consistently unlike in the monkey hosts (Thomas et al. 1994). Because of the expense and limited availability of chimpanzees most laboratory studies of malaria are performed in mice, using the rodent malaria species P. berghei or P. yoelii. These latter two models are well studied and it has been shown in both that CD8+ve lymphocytes play a key role in protective immunity against sporozoite challenge.

Previous studies have assessed a large variety of means of inducing CD8+ T cell responses against malaria. Several of these have shown some level of CD8+ T cell response and partial protection against malaria infection in the rodent models (e.g. Li et al. 1993; Sedegah et al. 1994; Lanar et al. 1996). However, an effective means of immunizing with subunit vaccines by the induction of sufficiently high levels of CD8+ T lymphocytes to protect effectively against malaria sporozoite infection has not previously been demonstrated.

In recent years improved immune responses generated to potential vaccines have been sought by varying the vectors used to deliver the antigen. There is evidence that in some instances antibody responses are improved by using two different vectors administered sequentially as prime and boost. A variety of combinations of prime and boost have been tested in different potential vaccine regimes.

Leong et al. (Vaccines 1995, 327-331) describe immunizing mice firstly to DNA expressing the influenza haemagglutinin (HA) antigen and then with a recombinant fowlpox vector expressing HA. An enhanced antibody response was obtained following boosting.

Richmond et al. (Virology 1997, 230: 265-274) describe attempts to raise neutralizing antibodies against HIV-1 env using DNA priming and recombinant vaccinia virus boosting. Only low levels of antibody responses were observed with this prime boost regime and the results were considered disappointing.

Fuller et al. (Vaccine 1997, 15:924-926 and Immunol Cell Biol 1997, 75:389-396) describe an enhancement of antibody responses to DNA immunization of macaques by using a booster immunization with replicating recombinant vaccinia viruses. However, this did not translate into enhanced protective efficacy as a greater reduction in viral burden and attenuation of CD4 T cell loss was seen in the DNA primed and boosted animals.

Hodge et al. (Vaccine 1997, 15: 759-768) describe the induction of lymphoproliferative T cell responses in a mouse model for cancer using human carcinoembryonic antigen (CEA) expressed in a recombinant fowl pox virus (ALVAC). The authors primed an immune response with CEA-recombinant replication competent vaccinia viruses of the Wyeth or WR strain and boosted the response with CEA-recombinant ALVAC. This led to an increase in T cell proliferation but did not result in enhanced protective efficacy if compared to three wild type recombinant immunizations (100% protection), three recombinant ALVAC-CEA immunizations (70% protection) or WR prime followed by two ALVAC-CEA immunizations (63% protection).

Thus some studies of heterologous prime-boost combination have found some enhancement of antibody and lymphoproliferative responses but no significant effect on protective efficacy in an animal model. CD8 T cells were not measured in these studies. The limited enhancement of antibody response probably simply reflects the fact that antibodies to the priming immunogen will often reduce the immunogenicity of a second immunization with the same immunogen, while boosting with a different carrier will in part overcome this problem. This mechanism would not be expected to be significantly affected by the order of immunization.

Evidence that a heterologous prime boost immunization regime might affect CD8 T cell responses was provided by Li et al. (1993). They described partial protective efficacy induced in mice against malaria sporozoite challenge by administering two live viral vectors, a recombinant replicating influenza virus followed by a recombinant replicating vaccinia virus encoding a malaria epitope. Reversing the order of immunization led to loss of all protective efficacy and the authors suggested that this might be related to infection of liver cells by vaccinia, resulting in localization of CTLs in the liver to protect against the hepatocytic stages of malaria parasites.

Rodrigues et al. (J. Immunol. 1994, 4636-4648) describe immunizing mice with repeated doses of a recombinant influenza virus expressing an immunodominant B cell epitope of the malarial circumsporozoite (CS) protein followed by a recombinant vaccinia virus booster. The use of a wild type vaccinia strain and an attenuated but replication-competent vaccinia strain in the booster yielded very similar levels of partial protection. However the attenuated but replication competent strain was slightly less immunogenic for priming CD8 T cells than the wild type vaccinia strain.

Murata et al. (Cell. Immunol. 1996, 173: 96-107) reported enhanced CD8 T cell responses after priming with replicating recombinant influenza viruses and boosting with a replicating strain of vaccinia virus and suggested that the partial protection observed in the two earlier studies was attributable to this enhanced CD8 T cell induction.

Thus these three studies together provide evidence that a booster immunization with a replicating recombinant vaccinia virus may enhance to some degree CD8 T cell induction following priming with a replicating recombinant influenza virus. However, there are two limitations to these findings in terms of their potential usefulness. Firstly, the immunogenicity induced was only sufficient to achieve partial protection against malaria and even this was dependent on a highly immunogenic priming immunization with an unusual replicating recombinant influenza virus. Secondly, because of the potential and documented side-effects of using these replicating viruses as immunogens these recombinant vectors are not suitable for general human use as vaccines.

Modified vaccinia virus Ankara (MVA) is a strain of vaccinia virus which does not replicate in most cell types, including normal human tissues. MVA was derived by serial passage >500 times in chick embryo fibroblasts (CEF) of material derived from a pox lesion on a horse in Ankara, Turkey (Mayr et al. 1975). It was shown to be replication-impaired yet able to induce protective immunity against veterinary poxvirus infections (Mayr 1976). MVA was used as a human vaccine in the final stages of the smallpox eradication campaign, being administered by intracutaneous, subcutaneous and intramuscular routes to >120,000 subjects in southern Germany. No significant side effects were recorded, despite the deliberate targeting of vaccination to high risk groups such as those with eczema (Mayr et al. 1978; Stickl et al. 1974; Mahnel et al. 1994;). The safety of MVA reflects the avirulence of the virus in animal models, including irradiated mice and following intracranial administration to neonatal mice. The non-replication of MVA has been correlated with the production of proliferative white plaques on chick chorioallantoic membrane, abortive infection of non-avian cells, and the presence of six genomic deletions totaling approximately 30 kb (Meyer et al. 1991). The avirulence of MVA has been ascribed partially to deletions affecting host range genes K1L and C7L, although limited viral replication still occurs on human TK-143 cells and African Green Monkey CV-1 cells (Altenburger et al. 1989). Restoration of the K1L gene only partially restores MVA host range (Sutter et al. 1994). The host range restriction appears to occur during viral particle maturation, with only immature virions being observed in human HeLa cells on electron microscopy (Sutter et al. 1992). The late block in viral replication does not prevent efficient expression of recombinant genes in MVA. Recombinant MVA expressing influenza nucleoprotein, influenza haemagglutinin, and SIV proteins have proved to be immunogenic and provide varying degrees of protection in animal models, although this has never been ascribed to CD8+ T lymphocytes alone (Sutter et al. 1994, Hirsch et al. 1995; Hirsch et al. 1996). Recombinant MVA is considered a promising human vaccine candidate because of these properties of safety and immunogenicity (Moss et al. 1995). Recombinant MVA containing DNA which codes for foreign antigens is described in U.S. Pat. No. 5,185,146 (Altenburger).

Poxviruses have evolved strategies for evasion of the host immune response that include the production of secreted proteins that function as soluble receptors for tumor necrosis factor, IL-1ÿ, interferon (IFN)-ÿ/ÿ and IFN-ÿ, which normally have sequence similarity to the extracellular domain of cellular cytokine receptors (Symons et al. 1995; Alcami et al. 1995; Alcami et al. 1992). The most recently described receptor of this nature is a chemokine receptor (Graham et al. 1997). These viral receptors generally inhibit or subvert an appropriate host immune response, and their presence is associated with increased pathogenicity. The Il-1ÿ receptor is an exception: its presence diminishes the host febrile response and enhances host survival in the face of infection (Alcami et al. 1996). We have discovered that MVA lacks functional cytokine receptors for interferon ÿ, interferon ÿÿ, Tumor Necrosis Factor and CC chemokines, but it does possess the potentially beneficial IL-1ÿ receptor. MVA is the only known strain of vaccinia to possess this cytokine receptor profile, which theoretically renders it safer and more immunogenic than other poxviruses. Another replication-impaired and safe strain of vaccinia known as NYVAC is fully described in Tartaglia et al. (Virology 1992, 188: 217-232).

It has long been recognized that live viruses have some attractive features as recombinant vaccine vectors including a high capacity for foreign antigens and fairly good immunogenicity for cellular immune responses (Ellis 1988 new technologies for making vaccines. In: Vaccines. Editors: Plotkin S A and Mortimer E A. W B Saunders, Philadelphia, page 568; Woodrow G C 1977. In: New Generation Vaccines 2nd Edition. Editors: Levine M M, Woodrow G C, Kaper J B, Cobon G, page 33). This has led to attempts to attenuate the virulence of such live vectors in various ways including reducing their replication capacity (Tartaglia J et al. 1992 Virology 188: 217-232). However such a reduction in replication reduces the amount of antigen produced by the virus and thereby would be expected to reduce vaccine immunogenicity. Indeed attenuation of replicating vaccinia strains has previously been shown to lead to some substantial reductions in antibody responses (Lee M S et al, 1992 J Virology 66: 2617-2630). Similarly the non-replicating fowlpox vector was found to be less immunogenic for antibody production and less protective than a replicating wild-type vaccinia strain in a rabies study (Taylor J et al. 1991 Vaccine 9: 190-193).

SUMMARY

It has now been discovered that non-replicating and replication-impaired strains of poxvirus provide vectors which give an extremely good boosting effect to a primed CTL response. Remarkably, this effect is significantly stronger than a boosting effect by wild type poxviruses. The effect is observed with malarial and other antigens such as viral and tumor antigens, and is protective as shown in mice and non-human primate challenge experiments. Complete rather than partial protection from sporozoite challenge has been observed with the novel immunization regime.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows the construct used to express Ty-VLP with the malaria epitope cassette CABDHFE. CTL epitopes are from P. falciparum STARP (sporozoite threonine- and asparagine-rich protein) (st), LSA-1 (liver stage antigen 1) (1s), CSP (circumsporozoite protein) (cp), TRAP (thrombospondin-related adhesive protein) (tr), LSA-3 (liver stage antigen 3) (la) and Exp-1 (exported protein 1) (ex). Helper epitopes are from the P. falciparum CS protein, the M. tuberculosis 38 Kd antigen and Tetanus Toxoid. NANP is the antibody epitope from CS and AM is the adhesion motif from P. falciparum TRAP (Muller et al 1993). The length of the complete string is 229 amino acids.

FIG. 2 shows a schematic outline of the H, M and HM proteins. The bar patterns on the schematic representations of the polyepitope proteins indicate the origin of the sequences. The positions of individual epitopes and their MHC restrictions are depicted above and below the proteins. Pb is the only epitope derived from the protein of P. berghei. All other epitopes in the M protein originate from proteins of P. falciparum: cs—circumsporozoite protein, st—STARP, Is—LSA-1 and tr—TRAP. BCG—38 kDa protein of M. tuberculosis; TT—tetanus toxin.

FIG. 3 shows malaria CD8 T cell ELISPOT data following different immunisation regimes. Results are shown as the number of peptide-specific T cells per million splenocytes.

FIGS. 4A-4D show that malaria CD8 T cell ELISPOT (FIGS. 4A and 4C) and CTL levels (FIGS. 4B and 4D) are substantially boosted by a recombinant MVA immunisation following priming with a plasmid DNA encoding the same antigen. The ELISPOT counts are presented on a logarithmic scale.

FIG. 5 shows the CTL responses induced in BALB/c mice to malaria and HIV epitopes by various immunisation regimes employing plasmid DNA and recombinant MVA. Levels of specific lysis at various effector to target ratios are shown.

FIG. 6 shows the results of ELISPOT assays performed to measure the levels of specific CD8+ T cells to the malaria epitope pb9 following different immunisation regimes. Groups of BALB/c mice (n=3) were immunised as indicated (g.g.=gene gun). The time between all immunisations was 14 days. ELISPOT assays were done two weeks after the last immunisation.

FIGS. 7A-7E show the CTL responses against influenza NP in different mouse strains. Mice of different strains were immunised twice two weeks apart with a DNA vaccine V1J-NP encoding for the influenza nucleoprotein (open circles) or primed with the same DNA vaccine and two weeks later boosted with recombinant MVA expressing influenza virus nucleoprotein (closed circles). The CTL activity was determined in a standard 51Cr-release assay with MHC class I-matched target cells.

FIGS. 8A-8H show CTL responses against different antigens induced in different inbred mouse strains. Mice were immunised with two DNA vaccine immunisations two weeks apart (open circles) or primed with a DNA vaccine and two weeks later boosted with a recombinant MVA expressing the same antigen (closed circles). The strains and antigens were: FIG. 8A, C57BL/6, P. falciparum TRAP; FIG. 8B, DBA/2, E. coli b-galactosidase; FIG. 8C, BALB/c, HM epitope string CTL activity against malaria peptide (pb9); FIG. 8D, DBA/2, HM epitope string CTL activity against pb9; FIG. 8E, BALB/c;,HM epitope string CTL activity against HIV peptide; FIG. 8F, DBA/2, HM epitope string CTL activity against HIV peptide; FIG. 8G, BALB/c, tumour epitope string CTL activity against P1A-derived peptide; and in FIG. 8H, DBA/2, tumour epitope string CTL activity against P1A-derived peptide. Each curve shows the data for an individual mouse.

FIGS. 9A-9E show sporozoite-primed CTL responses are substantially boosted by MVA. Mice were immunised with: FIG. 9A, two low doses (50+50) of irradiated sporozoites; FIG. 9B, two high doses (300+500) of sporozoites; FIG. 9D, low-dose sporozoite priming followed by boosting with MVA.PbCSP; FIG. 9E, high dose sporozoite priming followed by boosting with MVA.PbCSP. CTL responses following immunisation with MVA.PbCSP are shown in FIG. 9C.

FIGS. 10A and 10B show CTL responses primed by plasmid DNA or recombinant Adenovirus and boosted with MVA. Groups of BALB/c mice (n=3) were primed with plasmid DNA(FIG. 10A) or recombinant Adenovirus expressing ÿ-galactosidase (FIG. 10B). Plasmid DNA was administered intramuscularly, MVA intravenously and Adenovirus intradermally. Splenocytes were restimulated with peptide TPHPARIGL [SEQ ID NO: 69] two weeks after the last immunisation. CTL activity was tested with peptide-pulsed P815 cells.

FIGS. 11A-11C show CTL responses in BALB/c mice primed with plasmid DNA followed by boosting with different recombinant vaccinia viruses. Animals were primed with pTH.PbCSP 50 ÿg/mouse i.m. and two weeks later boosted with different strains of recombinant vaccina viruses (106 pfu per mouse i.v.) expressing PbCSP. The different recombinant vaccinia virus strains were: FIG. 11A, MVA; FIG. 11B, NYVAC; and WR in Figure C. The frequencies of peptide-specific CD8+ T cells were determined using the ELISPOT assay.

FIG. 12 shows frequencies of peptide-specific CD8+ T cells following different routes of MVA boosting. Results are shown as the number of spot-forming cells (SFC) per one million splenocytes. Each bar represents the mean number of SFCs from three mice assayed individually.

FIG. 13 shows the survival rate of the two groups of mice. Sixty days after challenge eight out of ten mice were alive in the group immunised with the tumour epitopes string.

FIG. 14 shows results of an influenza virus challenge experiment. BALB/c mice were immunised as indicated. GG=gene gun immunisations, im=intramuscular injection, iv=intravenous injection. Survival of the animals was monitored daily after challenge.

FIGS. 15A-15C show detection of SIV-specific MHC class I-restricted CD8+ T cells using tetramers. Each bar represents the percentage of CD8+ T cells specific for the Mamu-A*01/gag epitope at the indicated time point. One percent of CD8 T cells corresponds to about 5000/106 peripheral blood lymphocytes.

FIGS. 16A-16C show CTL induction in macaques following DNA/MVA immunisation. PBMC from three different macaques (CYD, DI and DORIS) were isolated at week 18, 19 and 23 and were restimulated with peptide CTPYDINQM [SEQ ID NO: 54] in vitro. After two restimulations with peptide CTPYDINQM [SEQ ID NO: 54] the cultures were tested for their lytic activity on peptide-pulsed autologous target cells.

DETAILED DESCRIPTION

It is an aim of this invention to identify an effective means of immunizing against malaria. It is a further aim of this invention to identify means of immunizing against other diseases in which CD8+ T cell responses play a protective role. Such diseases include but are not limited to infection and disease caused by the viruses HIV, herpes simplex, herpes zoster, hepatitis C, hepatitis B, influenza, Epstein-Barr virus, measles, dengue and HTLV-1; by the bacteria Mycobacterium tuberculosis and Listeria sp.; and by the protozoan parasites Toxoplasma and Trypanosoma; and certain forms of cancer e.g. melanoma, cancer of the breast and cancer of the colon.

We describe here a novel method of immunizing that generated very high levels of CD8+ T cells and was found to be capable of inducing unprecedented complete protection against P. berghei sporozoite challenge. The same approach was tested in higher primates and found to be highly immunogenic in this species also, and was found to induce partial protection against P. falciparum challenge. Induction of protective immune responses has also been demonstrated in two additional mouse models of viral infection and cancer.

We show further than the novel immunization regime that is described here is also effective in generating strong CD8+ T cell responses against HIV epitopes. Considerable evidence indicates that the generation of such CD8+ T cell responses can be expected to be of value in prophylactic or therapeutic immunization against this viral infection and disease (Gallimore et al 1995; Ada 1996). We demonstrate that strong CD8+T cell responses may be generated against epitopes from both HIV and malaria using an epitope string with sequences from both of these micro-organisms. The success in generating enhanced immunogenicity against both HIV and malaria epitopes, and also against influenza and tumor epitopes, indicates that this novel immunization regime can be effective generally against many infectious pathogens and also in non-infectious diseases where the generation of a strong CD8+ T cell response may be of value.

A surprising feature of the current invention is the finding of the very high efficacy of non-replicating agents in both priming and particularly in boosting a CD8+ T cell response. In general the immunogenicity of CD8+ T cell induction by live replicating viral vectors has previously been found to be higher than for non-replicating agents or replication-impaired vectors. This is as would be expected from the greater amount of antigen produced by agents that can replicate in the host. Here however we find that the greatest immunogenicity and protective efficacy is surprisingly observed with non-replicating vectors. The latter have an added advantage for vaccination in that they are in general safer for use in humans than replicating vectors.

The present invention provides in one aspect a kit for generating a protective CD8+ T cell immune response against at least one target antigen, which kit comprises: (i) a priming composition comprising a source of one or more CD8+ T cell epitopes of the target antigen, together with a pharmaceutically acceptable carrier; and (ii) a boosting composition comprising a source of one or more CD8+ T cell epitopes of the target antigen, including at least one CD8+ T cell epitope which is the same as a CD8+ T cell epitope of the priming composition, wherein the source of CD8+ T cell epitopes is a non-replicating or replication-impaired recombinant poxvirus vector, together with a pharmaceutically acceptable carrier; with the proviso that if the source of epitopes in (i) is a viral vector, the viral vector in (ii) is derived from a different virus.

In another aspect the invention provides a method for generating a protective CD8+ T cell immune response against at least one target antigen, which method comprises administering at least one dose of component (i), followed by at least one dose of component (ii) of the kit according to the invention.

Preferably, the source of CD8+ T cell epitopes in (i) in the method according to the invention is a non-viral vector or a non-replicating or replication-impaired viral vector, although replicating viral vectors may be used.

Preferably, the source of CD8+ T cell epitopes in (i) is not a poxvirus vector, so that there is minimal cross-reactivity between the primer and the booster.

In one preferred embodiment of the invention, the source of CD8+ T cell epitopes in the priming composition is a nucleic acid, which may be DNA or RNA, in particular a recombinant DNA plasmid. The DNA or RNA may be packaged, for example in a lysosome, or it may be in free form.

In another preferred embodiment of the invention, the source of CD8+ T cell epitopes in the priming composition is a peptide, polypeptide, protein, polyprotein or particle comprising two or more CD8+ T cell epitopes, present in a recombinant string of CD8+ T cell epitopes or in a target antigen. Polyproteins include two or more proteins which may be the same, or preferably different, linked together. Particularly preferred in this embodiment is a recombinant proteinaceous particle such as a Ty virus-like particle (VLP) (Burns et al. Molec. Biotechnol. 1994, 1: 137-145).

Preferably, the source of CD8+ T cell epitopes in the boosting composition is a vaccinia virus vector such as MVA or NYVAC. Most preferred is the vaccinia strain modified virus ankara (MVA) or a strain derived therefrom. Alternatives to vaccinia vectors include avipox vectors such as fowlpox or canarypox vectors. Particularly suitable as an avipox vector is a strain of canarypox known as ALVAC (commercially available as Kanapox), and strains derived therefrom.

Poxvirus genomes can carry a large amount of heterologous genetic information. Other requirements for viral vectors for use in vaccines include good immunogenicity and safety. MVA is a replication-impaired vaccinia strain with a good safety record. In most cell types and normal human tissues, MVA does not replicate; limited replication of MVA is observed in a few transformed cell types such as BHK21 cells. It has now been shown, by the results described herein, that recombinant MVA and other non-replicating or replication-impaired strains are surprisingly and significantly better than conventional recombinant vaccinia vectors at generating a protective CD8+ T cell response, when administered in a boosting composition following priming with a DNA plasmid, a recombinant Ty-VLP or a recombinant adenovirus.

It will be evident that vaccinia virus strains derived from MVA, or independently developed strains having the features of MVA which make MVA particularly suitable for use in a vaccine, will also be suitable for use in the invention.

MVA containing an inserted string of epitopes (MVA-HM, which is described in the Examples) has been deposited at the European Collection of Animal Cell Cultures, CAMR, Salisbury, Wiltshire SP4 0JG, UK under accession no. V97060511 on 5 Jun. 1997.

The term “non-replicating” or “replication-impaired” as used herein means not capable of replication to any significant extent in the majority of normal mammalian cells or normal human cells. Viruses which are non-replicating or replication-impaired may have become so naturally (i.e. they may be isolated as such from nature) or artificially e.g. by breeding in vitro or by genetic manipulation, for example deletion of a gene which is critical for replication. There will generally be one or a few cell types in which the viruses can be grown, such as CEF cells for MVA.

Replication of a virus is generally measured in two ways: 1) DNA synthesis and 2) viral titre. More precisely, the term “non-replicating or replication-impaired” as used herein and as it applies to poxviruses means viruses which satisfy either or both of the following criteria: 1) exhibit a 1 log (10 fold) reduction in DNA synthesis compared to the Copenhagen strain of vaccinia virus in MRC-5 cells (a human cell line); 2) exhibit a 2 log reduction in viral titre in HELA cells (a human cell line) compared to the Copenhagen strain of vaccinia virus.

Examples of poxviruses which fall within this definition are MVA, NYVAC and avipox viruses, while a virus which falls outside the definition is the attenuated vaccinia strain M7.

Alternative preferred viral vectors for use in the priming composition according to the invention include a variety of different viruses, genetically disabled so as to be non-replicating or replication-impaired. Such viruses include for example non-replicating adenoviruses such as E1 deletion mutants. Genetic disabling of viruses to produce non-replicating or replication-impaired vectors has been widely described in the literature (e.g. McLean et al. 1994).

Other suitable viral vectors for use in the priming composition are vectors based on herpes virus and Venezuelan equine encephalitis virus (VEE) (Davies et al. 1996). Suitable bacterial vectors for priming include recombinant BCG and recombinant Salmonella and Salmonella transformed with plasmid DNA (Darji A et al. 1997 Cell 91: 765-775).

Alternative suitable non-viral vectors for use in the priming composition include lipid-tailed peptides known as lipopeptides, peptides fused to carrier proteins such as KLH either as fusion proteins or by chemical linkage, whole antigens with adjuvant, and other similar systems. Adjuvants such as QS21 or SBAS2 (Stoute J A et al. 1997 N Engl J Medicine 226: 86-91) may be used with proteins, peptides or nucleic acids to enhance the induction of T cell responses. These systems are sometimes referred to as “immunogens” rather than “vectors”, but they are vectors herein in the sense that they carry the relevant CD8+ T cell epitopes.

There is no reason why the priming and boosting compositions should not be identical in that they may both contain the priming source of CD8+ T cell epitopes as defined in (i) above and the boosting source of CD8+ T cell epitopes as defined in (ii) above. A single formulation which can be used as a primer and as a booster will simplify administration. The important thing is that the primer contains at least the priming source of epitopes as defined in (i) above and the booster contains at least the boosting source of epitopes as defined in (ii) above.

The CD8+ T cell epitopes either present in, or encoded by the priming and boosting compositions, may be provided in a variety of different forms, such as a recombinant string of one or two or more epitopes, or in the context of the native target antigen, or a combination of both of these. CD8+ T cell epitopes have been identified and can be found in the literature, for many different diseases. It is possible to design epitope strings to generate a CD8+ T cell response against any chosen antigen that contains such epitopes. Advantageously, the epitopes in a string of multiple epitopes are linked together without intervening sequences so that unnecessary nucleic acid and/or amino acid material is avoided. In addition to the CD8+ T cell epitopes, it may be preferable to include one or more epitopes recognized by T helper cells, to augment the immune response generated by the epitope string. Particularly suitable T helper cell epitopes are ones which are active in individuals of different HLA types, for example T helper epitopes from tetanus (against which most individuals will already be primed). A useful combination of three T helper epitopes is employed in the examples described herein. It may also be useful to include B cell epitopes for stimulating B cell responses and antibody production.

The priming and boosting compositions described may advantageously comprise an adjuvant. In particular, a priming composition comprising a DNA plasmid vector may also comprise granulocyte macrophage-colony stimulating factor (GM-CSF), or a plasmid encoding it, to act as an adjuvant; beneficial effects are seen using GM-CSF in polypeptide form.

The compositions described herein may be employed as therapeutic or prophylactic vaccines. Whether prophylactic or therapeutic immunization is the more appropriate will usually depend upon the nature of the disease. For example, it is anticipated that cancer will be immunized against therapeutically rather than before it has been diagnosed, while anti-malaria vaccines will preferably, though not necessarily be used as a prophylactic.

The compositions according to the invention may be administered via a variety of different routes. Certain routes may be favoured for certain compositions, as resulting in the generation of a more effective response, or as being less likely to induce side effects, or as being easier for administration. The present invention has been shown to be effective with gene gun delivery, either on gold beads or as a powder.

In further aspects, the invention provides: a method for generating a protective CD8+ T cell immune response against a pathogen or tumor, which method comprises administering at least one dose of a recombinant DNA plasmid encoding at least one CD8+ T cell epitope or antigen of the pathogen or cancer, followed by at least one dose of a non-replicating or replication-impaired recombinant pox virus encoding the same epitope or antigen; a method for generating a protective CD8+ T cell immune response against a pathogen or tumor, which method comprises administering at least one dose of a recombinant protein or particle comprising at least one epitope or antigen of the pathogen or cancer, followed by at least one dose of a recombinant MVA vector encoding the same epitope or antigen; the use of a recombinant non-replicating or replication-impaired pox virus vector in the manufacture of a medicament for boosting a CD8+ T cell immune response; the use of an MVA vector in the manufacture of a medicament for boosting a CD8+ T cell immune response; a medicament for boosting a primed CD8+ T cell response against at least one target antigen or epitope, comprising a source of one or more CD8+ T cell epitopes of the target antigen, wherein the source of CD8+ T cell epitopes is a non-replicating or a replication-impaired recombinant poxvirus vector, together with a pharmaceutically acceptable carrier; and the priming and/or boosting compositions described herein, in particulate form suitable for delivery by a gene gun; and methods of immunization comprising delivering the compositions by means of a gene gun.

Also provided by the invention are: the epitope strings described herein, including epitope strings comprising the amino acid sequences listed in table 1 and table 2; recombinant DNA plasmids encoding the epitope strings; recombinant Ty-VLPs comprising the epitope strings; a recombinant DNA plasmid or non-replicating or replication impaired recombinant pox virus encoding the P. falciparum antigen TRAP; and a recombinant polypeptide comprising a whole or substantially whole protein antigen such as TRAP and a string of two or more epitopes in sequence such as CTL epitopes from malaria.

Priming Composition: DNA plasmid 1 mg/ml in PBS Boosting Composition: Recombinant MVA, 108 ffu in PBS Protocol: Administer two doses of 1 mg of priming composition, i.m., at 0 and 3 weeks followed by two doses of booster intradermally at 6 and 9 weeks.

Priming Composition: Ty-VLP 500ÿg in PBS Boosting Composition: MVA, 108 ffu in PBS Protocol: Administer two doses of priming composition, i.m., at 0 and 3 weeks, then 2 doses of booster at 6 and 9 weeks. For tumor treatment, MVA is given i.v. as one of most effective routes.

Priming Composition: Protein 500ÿg+adjuvant (QS-21) Boosting Composition: Recombinant MVA, 108 ffu in PBS Protocol: Administer two doses of priming composition at 0 and 3 weeks and 2 doses of booster i.d. at 6 and 9 weeks.

Priming Composition: Adenovirus vector, 109 pfu in PBS Boosting Composition: Recombinant MVA, 108 ffu in PBS Protocol: Administer one or two doses of priming composition intradermally at 0 and 3 weeks and two doses of booster i.d. at 6 and 9 weeks. The above doses and protocols may be varied to optimise protection. Doses may be given between for example, 1 to 8 weeks apart rather than 2 weeks apart.

The invention will now be further described in the examples which follow.

The malaria epitope string was made up of a series of cassettes each encoding three epitopes as shown in Table 1, with restriction enzyme sites at each end of the cassette. Each cassette was constructed from four synthetic oligonucleotides which were annealed together, ligated into a cloning vector and then sequenced to check that no errors had been introduced. Individual cassettes were then joined together as required. The BamHI site at the 3′ end of cassette C was fused to the BglII site at the 5′ end of cassette A, destroying both restriction enzyme sites and encoding a two amino acid spacer (GS) between the two cassettes. Cassettes B, D and H were then joined to the string in the same manner. A longer string containing CABDHFE was also constructed in the same way.

TABLE 1 CTL Epitopes of the Malaria (M) String Amino acid HLA Cassette Epitope Sequence DNA sequence Type restriction A Ls8 KPNDKSLY AAGCCGAACGACAAGTCCTTGTAT CTL B35 SEQ ID NO.: 2 SEQ ID NO.: 1 Cp26 KPKDELDY AAACCTAAGGACGAATTGGACTAC CTL B35 SEQ ID NO.: 4 SEQ ID NO.: 3 Ls6 KPIVQYDNF AAGCCAATCGTTCAATACGACAACTTC CTL B53 SEQ ID NO.: 6 SEQ ID NO.: 5

Download full PDF for full patent description/claims.




You can also Monitor Keywords and Search for tracking patents relating to this Methods and reagents for vaccination which generate a cd8 t cell immune response patent application.
###


Other recent patent applications listed under the agent :