WO1994016084A1

WO1994016084A1 - Hybrid particle-forming protein comprising a particle-inducing segment and a plurality of antigens

Info

Publication number: WO1994016084A1
Application number: PCT/GB1994/000068
Authority: WO
Inventors: Sally Elizabeth Adams; Guy Timothy Layton
Original assignee: British Bio-Technology Limited
Priority date: 1993-01-16
Filing date: 1994-01-13
Publication date: 1994-07-21
Also published as: AU5838794A

Abstract

A hybrid particle-forming protein comprising a first amino acid sequence substantially homologous with a yeast retrotransposon Ty p1 protein or a GAG protein of a RNA retrovirus fused to a plurality of different antigenic amino acid sequences.

Description

HYBRID PARTICLE-FORMING PROTEIN COMPRISING A PARTICLE-INDUCING SEGMENT AND A PLURALITY OF ANTIGENS .

The present invention relates to biologically useful particles. In particular it relates to modified particles derived from the yeast retrotransposon Ty and from retroviruses. Particles formed from such proteins are immunogenic and can be used as diagnostic agents, in immunotherapeutic or prophylactic vaccines or in targetted drug delivery systems.

An ideal immunogen is a polymer of multiple antigenic determinants assembled into a high molecular weight, particulate complex. A substantial disadvantage of most antigens produced by recombinant DNA techniques for vaccines is that they are usually produced as simple monomeric proteins. This is not the ideal configuration of an immunogen since monomeric B cell epitopes are less efficient than polymeric epitopes for the activation of B cells. Further, multiple determinant presentation can induce cross-reactive antibodies and cover MHC restriction of T-cell epitopes. For these reasons it would be advantageous to develop polyvalent, particulate carrier systems for vaccine delivery.

WO-A-8803562 and WO-A-8803563 describe the use of certain fusion proteins derived from retrotransposons or RNA retroviruses for pharmaceutical, diagnostic or purification applications. Such particles are designated virus-like particles (VLPs) when derived from the yeast retrotransposon Ty and virus-derived particles (VDPs) when derived from the gag protein of retroviruses. Specifically, the above published PCT applications note that polyvalent particles are useful for immunisation purposes because their polyvalent nature provides that more antibodies will be raised against the particulate antigens used. The particles are formed of fusion proteins having a particle-forming sequence and, in some embodiments at least, an antigenic sequence. In the examples, only one copy of the antigenic sequence is fused to the particle-forming sequence.

While the above approach is promising, a potential difficulty is that a number of aetiological agents have antigenic epitopes which are rapidly evolving and there is therefore widespread variation between strains. For a vaccine to be effective, it should provide protection against all the known strains of an aetiological agent; the production of cross- neutralising antibodies is a goal sought in the art. However, the domains which induce neutralising antibodies may be located within the hypervariable regions, for example, the third variable domain of the envelope protein gp160 of HIV (Boudet et al. , Int. Immunol. 4 283-294

(1992); Freed et al. , J. Virol. 65 190-194 (1991); Masuda et al. , J. Immunol. 145 3240-3246 (1990)). Variation in key neutralising epitopes is also seen in many other viral proteins, for example influenza haemagglutinin. There is also a high degree of serotype variation in the L1, L2, E6 and E7 proteins of HPV. This high level of diversity between strains means that vaccination with an antigen from any single strain may not lead to production of cross-neutralising antibodies.

Methods of optimising antibody responses have also been sought. Multiple epitope expression is advantageous for optimal B cell activation and inclusion of increased numbers of such epitopes within the particle might be expected to result in higher antibody titres.

Particles carrying a plurality of antigens containing T-cell epitopes might also have enhanced ability to stimulate a cell-mediated immune response, such as a T-helper cell reponse or a cytotoxic T-lymphocyte (CTL) reponse. The inclusion of a plurality of T-cell epitopes would also circumvent potential problems with MHC-restriction. WO-A-8803562 discloses two specific ways in which particles comprising a plurality of different antigens may be provided. First, it is suggested that a cocktail of different particulate antigens may be used. Secondly, it is suggested as an alternative that a homogeneous population of particulate antigens having more than one epitope be prepared by allowing a mixture of different hybrid proteins to aggregate into particles. Both of these approaches may be difficult to control to the satisfaction of the regulatory authorities; additionally they require considerable resources to build the necessary plurality of genetic constructs.

Particles made of a homogeneous population of hybrid proteins bearing a plurality of identical epitopes have been found to be no better than single epitopes. Engineering a plurality of homologous epitopes onto a hybrid protein is not, therefore, a promising approach. However, it has now surprisingly been found that when a plurality of different epitopes is engineered onto a single particle-forming sequence, the immunogenicity of the resulting particle is much improved over the case where the epitopes are the same as each other. The present invention relates to the discovery that incorporation of a plurality of different antigenic sequences fused to a particle forming protein results in enhanced immunogenicity, which may be either humoral or cell-mediated. If more than one variable sequence from different viral strains is present in a fusion protein of the invention, the production of cross-neutralisating antibodies may be favoured. This may enable a single fusion protein to be useful in the protection against more than one viral isolate; for example, a plurality of analogous sequences from different viral isolates can be present on the same molecule. Alternatively, a plurality of different epitopes may be useful in immunising a population with multiple MHC types with a single particle type, and therefore overcoming MHC restriction. Finally, it may be possible to immunise a patient with a single vaccine carrying epitopes from two or more pathogens, and thus reduce the number of vaccinations required and increase patient compliance.

According to a first aspect of the invention, there is provided a hybrid particle-forming protein comprising a first amino acid sequence substantially homologous with a yeast retrotransposon Ty p1 protein or a GAG protein of a RNA retrovirus fused to a plurality of different antigenic amino acid sequences.

A number of different arrangements of antigenic sequence are possible. The hybrid protein may comprise multiple variable sequences from the same epitope of two or more strain variants of an aetiological agent. Alternatively or in addition, the variable sequence may be derived from two or more different aetiological agents. While the hybrid protein may comprise more than one copy of any single antigenic sequence, hybrid molecules where all the antigenic sequences are different are preferred. A further possibility is the use of two or more different antigenic epitopes so that the hybrid particle induces both humoral and cell- mediated responses.

The expression "substantially homologous", when describing the relationship of an amino acid sequence to a natural protein, means that the amino acid sequence can be identical to the natural protein or can be an effective but truncated or otherwise modified form of the natural protein or can share at least 50%, 60%, 70%, 80%, 90%, 95% or 99%, in increasing order of preference, of the residues of the natural protein or its modified form. "Effective" means that the particle forming ability of the natural protein is retained (or at least not substantially lost).

Alternatively or in addition, a nucleic acid sequence encoding the amino acid sequence may hybridise, for example under stringent conditions, to a nucleic acid sequence encoding the natural protein or its truncated form, or would do so but for the degeneracy of the genetic code. Stringent hybridisation conditions are known in the art and are exemplified by approximately 0.9 molar salt concentration at approximately 35° to 65°C.

The retrovirus from which the particle-forming GAG protein may be HIV-1, HIV-2, HTLV-I, HTLV,II, HTLV-III, SIV, BIV, LAV, ELAV,

CIAV, murine Leukaemia virus, Moloney murine leukaemia virus and Feline Leukaemia virus. More preferred are those GAG proteins of lentiviral origin such as HIV-1, HIV-2, SIV, BIV and FIV. It is most preferred that the GAG protein be derived from HIV-1. Stability of particles formed from the fusion proteins may be enhanced by removing some or all of the basic C-terminus of the GAG precursor protein (p55 in HIV-1). At least the first 437 amino acids of the GAG protein will preferably be present in the preferred truncated form. Natural GAG proteins, at least of HIV-1, are post-translationally modified. Such modification usually includes myristylation in naturally HIV-1 GAG. Naturally post-translationally modified GAG proteins and non-post-translationally modified GAG proteins are included within the scope of the invention. The myristylation or other post-translational modification site or sites may cause non-optimal expression in certain non-animal host cells (such as yeast), and for this reason may be absent in certain embodiments of the invention. Its absence is far from mandatory, however, as certain cell lines (such as insect cell lines) readily express myristylated GAG proteins.

The antigenic sequence may correspond to a sequence derived from or associated with an aetiological agent or a tumour. The aetiological agent may be a microorganism such as a virus, bacterium, fungus or parasite. The virus may be: a retrovirus, such as HIV-1, HIV-2, HTLV-I, HTLV-II, HTLV-III, SIV, BIV, LAV, ELAV, CIAV, murine leukaemia virus, Moloney murine leukaemia virus, and feline leukaemia virus; an orthomyxovirus, such as influenza A or B; a paramyxovirus, such as parainfluenza virus, mumps, measles, RSV and Sendai virus; a papovavirus, such as HPV; an arenavirus, such as LCMV of humans or mice; a hepadnavirus, such as Hepatitis B virus; a herpes virus, such as

HSV, VZV, CMV, or EBV. The tumour-associated or derived antigen may for example be a proteinaceous human tumour antigen, such as a melanoma-associated antigen, or an epithelial-tumour associated antigen such as from breast or colon carcinoma.

The antigenic sequence may be also derived from a bacterium, such as of the genus Neisseria, Campilobacter, Bordetella, Listeria, Mycobacteria or Leishmania, or a parasite, such as from the genus Trypanosoma, Schizosoma, Plasmodium, especially P. falciparum, or from a fungus, such as from the genus Candida, Aspergillus, Cryptococcus,

Histoplasma or Blastomyces.

Preferred antigenic sequences correspond to epitopes from a retrovirus, a paramyxovirus, an arenavirus or a hepadna virus, or from a human tumour cell. Examples include epitopes from: 1) HIV (particularly HIV-1) gp120,

2) HIV (particularly HIV-1) p24,

3) Influenza virus nucleoprotein and haemagglutinin, 4) LCMV nucleoprotein,

5) HPV L1, L2 E4, E6 and E7 proteins,

6) p97 melanoma associated antigen,

7) GA 733-2 epithelial tumour-associated antigen,

8) MUC-1 epithelial tumour-associated antigen,

9) Mycobacterium p6,

10) Malaria CSP or RESA antigens,

11 ) VZ V gpl, gpll and gpIII

Particularly preferred antigenic sequences are derived from variable epitopes of at least two different strains of a virus, for example from the third variable domain of a lentivirus. This region of lentiviruses, known as the V3 loop or GPGR loop is found between amino acids 300 and 330 of the envelope glycoprotein gp120 of HIV-1 and in analogous positions of other lentiviruses. The V3 loop is defined by two flanking cysteine residues linked by a disulphide bond and, for HTV-1 at least, is the major neutralising epitope of the virus (Putney et al 1986 Science

234, 1392; Rusche et al 1988 Proc. Natl. Acad. Sci. 85, 3198; Palker et al

1988 Proc. Natl. Acad. Sci. 85 1932; and Goudsmit et al 1988 AIDS 2 157).

The antigenic portion of choice may constitute the whole of the V3 loop when derived from different strains. However, multiple copies of a conserved sequence of the V3 loop may be useful in conferring immunity against more than one isolate of a virus (such as HIV-1).

A number of isolates of HIV-1, in which the sequence of the V3 loop varies from isolate to isolate, are known. The most common isolates are HXBII, RF and MN; MAL, ELI and BH10 are also important, but the MN isolate may be the most clinically relevant. Laboratory isolate IIIB is a mixture of strains BH10 and HXBII. The invention is not limited to sequences derived from the V3 loop of any particular isolate, some of which are shown below.

BH10 SNCTRPNNNTRKSIRIQRGPGRAFVΗGKIGNMRQAHCNISG

HXBII SNCTRPNNNTRKRIRIQRGPGRAFVΠGKIGNMRQAHCNISG

MN SNCTRPNYNKRKRIHIGPGRAFYTTKNΠGTIRQAHCNISG MAL SNCTRPGNNTRRGIHFGPGQALYTTGIVDIRRAYCTING

RF SNCTRPNNNTRKSITKGPGRVΓYATGQΠGDIRAHCNLSGS

ELI STCARPYQNTRQRTPIGLGQSLYTTRSRSΠGQAHCNISG.

Neither is the invention limited to natural V3 loop sequences. Examples of variant V3 loop sequences which can be used in the invention include:

MAL(var) SNCTRPGNNTRRGΓHFGPGQALYTTGIVDEIRRAYCNISG RF(var)SNCTRPNNNTRKSITKQRGPGRVLYATGQIIGDIRKAHCNSIG ELI(var) STCARPYQNTRQRTPIGLGQSLYTTRGRTKπGQAHCNISG.

A comparison of the sequences of the V3 loop from many different HIV-1 isolates shows great variation between isolates. Antibodies raised against the V3 loop are therefore usually type-specific. However, approximately 60% of isolates to date have the consensus sequence GPGRAF, and more than 80% have a GPGR sequence at the tip of the loop. Recent studies have shown that immunisation with peptides containing the GPGRAF consensus sequence or cross-immunisation with recombinant gp120 from different isolates can induce antibodies which cross react between isolates. The GPGRAF consensus sequence may itself be used in the invention. Other embodiments of the invention involve the use of short sequences of V3 which are not necessarily conserved between various isolates. Whatever V3-derived or V3-related sequence is used, the resulting fusion proteins, or at least particles assembled from them, will be similar antigenically to natural V3 loop sequences in the sense that they cross-react with one or more common antibodies.

As fusion proteins in accordance with the invention spontaneously assemble into particles, it is possible by means of the invention to prepare multivalent particles: a plurality of different fusion proteins could be prepared which comprise different multiple antigens.

According to a second aspect of the invention, there is provided a particle comprising a plurality of hybrid proteins as described above. Particles in accordance with the invention may contain a heterologous, or, preferably, homologous population of proteins. Each protein may have any of the configurations described above.

According to a third aspect of the invention, there is provided nucleic acid (particularly DNA) coding for a fusion protein as described above. It will generally be the case that the nucleic acid will be capable of being expressed without splicing or anti-termination events. There will generally be no frame shifting, but frame shifting is not necessarily always excluded.

Further according to the present invention is provided a vector including nucleic acid as described above.

Expression vectors in accordance with the invention will usually contain a promoter. The nature of the promoter will depend upon the intended host expression cell. For yeast, PGK is a preferred promoter, but any other suitable promoter may be used if necessary or desirable. Examples include GAPD, GAL 1-10, PH05, ADH1, CYC1, Ty delta sequence, PYK and hybrid promoters made from components from more than one promoter (such as those listed). For insect cells, a preferred promoter is the polyhedrin promoter from Autographica calif ornica nuclear polyhedrosis virus (AcNPV). Those skilled in the art will be able to determine other appropriate promoters adapted for expression in these or other cells. Vectors not including promoters may be useful as cloning vectors, rather than expression vectors.

The invention also includes host cells, for examples bacterial cells, such as E. coli, which may be used for genetic manipulation, yeast cells such as Saccharomyces cerevisiae or Pichia pastoris or animal cells. Although mammalian cells such as COS or CHO cells can be used, expression in insect cell lines is preferred in many cases. A suitable cell line is Spodoptera frugiperda, such as SF9.

Because of the augmented immunogenic nature of the particles in accordance with the invention, it is likely that it will be easier to produce antibodies than with conventional antigens and that those antibodies will have specific characteristics. The invention thus further provides antibodies raised against particulate antigens of the invention.

The antibodies may be polyclonal (obtained for example by injecting antigens into a rabbit) or monoclonal antibodies, produced by hybridoma cells. It is also likely that in vitro immunisation can be achieved more readily than with other forms of antigen; this may facilitate the production of human monoclonal antibodies. Hybridoma cells may be prepared by fusing spleen cells from an immunised animal with a tumour cell. Appropriately secreting hybridoma cells may thereafter be selected.

Particulate antigens in accordance with the invention may be useful in the preparation of vaccines, for example immunotherapeutic vaccines, which form a further aspect of the invention. The vaccine may comprise a particulate antigen and a physiologically acceptable non-toxic diluent such as sterile physiological saline or sterile PBS. Sterility will generally be essential for parenterally administrable vaccines. One or more appropriate adjuvants may also be present. Examples of suitable adjuvants include muramyl dipeptide compounds such as prototype muramyl dipeptide, aluminium hydroxide and saponin. For induction of a CTL response, it may be preferred to immunise in the absence of adjuvant. Fusion protein and particulate antigens of this invention are useful as diagnostic reagents. Particulate antigens for diagnostic purposes are particularly advantageous because they can be physically separated by centrifugation or filtration and can be directly dispersed on solid supports such as glass or plastic slides, dip sticks, macro or micro beads, test tubes, wells of microtitre plates and the like. The particulate antigens of this invention may also be dispersed in fibrous or bibulous materials such as absorbent disk (US-A-4,632,901), strips or chromatography columns as the solid support. The particles and fusion proteins readily adhere to solid supports. The particles may after purification be disrupted into fusion proteins and the fusion proteins may be dispersed on surfaces as indicated above. These reagents are useful for a variety of diagnostic tests. For example, a test sample suspected of having antibody to the particulate antigen and fluorescent, enzyme or radio-labelled antibody is competitively reacted with the particulate antigen or fusion protein on a solid support and the amount of labelled antibody which binds to the particulate antigen on the solid support. Particulate antigens of this invention are also useful for agglutination reactions with antibodies. Those skilled in the diagnostic arts will recognise a wide variety of application of particulate antigens and fusion proteins of this invention for diagnostic purposes.

Preferred features for each aspect of the invention are as for the first aspect mutatis mutandis.

The following examples illustrate the invention, but are not intended to limit the scope in any way. The examples refer to the accompanying figures, in which; Table 1 shows MN, RF and IIIB antibody responses in rabbits immunised with MN and combination MN/RF/IIIB VLPs

Table 2 shows MN and IIIB antibody responses in macaques immunised with MN and combination MN/RF/IIIB VLPs

Table 3 shows the construction of V3: Sendai NP-VLPs.

Table 4 shows the construction of VSV: Sendai NP-VLPs. Figure 1 shows the production of MHC-restricted CTLs, specific to the H-

2d restricted V3 sequence in BALB/c mice or to Sendai core NP (H-2b) in C57BL/6 mice.

Figure 2 shows the production of CTLs specific to the H-2b-restricted VSV and Sendai NP epitopes in C57BL/6 mice. Comparative Example 1; Multiple V3 Loop Constructions

VLPs were prepared by insertion of 1 MN V3 loop (OGS254 and OGS530), 2 MN V3 loops (OGS255), 3 MN V3 loops (OGS256), 3 MN V3 loops (OGS257) and 1 MN V3 loop with a glycine spacer between the p1 and MN sequences (OGS335) into pOGS40, which is disclosed in copending patent application PCT/ GB92/01545. The MN sequence SNCTRPNYNKRKRIHIGPGRAFYTTKNIIGTIRQAHCNISG (SEQ ID 1) was inserted into the 3' end of the TY gene via a Bg1 II/ Bam H1 cassette. Rats were immunised with OGS254, OGS255, OGS335 OGS256 and

OGS257 in alum. None of these innoculations induced any significant anti-V3 antibody responses.

V3-specific CTL Responses to Multiple MN V3 Loop-VLP Constructions Three MN-based V3-VLP constructions were tested for their ability to induce V3-specific CTL following immunisations i.m. with 20μg of VLPs in saline. The VLP constructions contained the following sequences: OGS530 - 1 x 39mer (entire V3 loop)

OGS255 - 2 x 39mer

OGS256 - 3 x 39mer

The MN CTL epitope has been shown to be within RIHIGPGRAFYTTKN which is contained in all of the MN V3-VLP constructions. Splenocytes (2 spleens/group) from the three groups of mice were expanded for 6 days with MN 39mer peptide and tested against P815 cells either pulsed with the same peptide or not pulsed. V3 peptide-specific lysis of target cells was only seen with effector cells derived from mice immunised with OGS530 (39mer MN V3) or OGS335 (39mer MN V3 + glycine spacers) VLPs, despite the presence of the CTL epitope in all of the other constructs. This indicates that the presentation of multiple copies of a short sequence containing a CTL epitope has a deleterious effect on CTL induction. Example 2 V3/T Cell Epitope Constructions

Oligonucleotides encoding the T-1, T-2 and Th4.1 epitopes of HIV-1 gp160 were made. The T-1 oligonucleotides (encoding amino acids 428- 443) were inserted as Bglll-BamH1 cassettes into pOGS40 (disclosed in copending patent application WO-A-93/20840). The plasmid was designated pOGS342. OGS342 VLPs (p1:T-1) react with both anti-Ty and human HIV positive sera by Western blot. Oligonucleotides encoding the T-2 epitope (amino acids 102-120 of gp160) were inserted at the 3' end of T-1 in pOGS342 to generate pOGS345 (p1:T-1/T-2). Oligonucleotides encoding the Th4.1 epitope (amino acids 824-841) were then inserted at the 3' end of T-2 in pOGS345 to generate pOGS346 (p1:T-1/T-2/Th4.1).

The final step in the construction was to insert the (GPGRAF)₃ sequence at the 3' end of Th4.1 to produce pOGS347.

2.1 Antibody Responses Induced by OGS347 VLPs

Rats were immunised with OGS347 (GPGRAF x 3/T1 /T2/T_h4.1). The sera obtained were assayed against MN V3 peptide, IIIB V3 peptide, rgp120 and Ty. OGS347-immunised animals all responded to rgp120, with a higher mean mid-point titre (MPT) in the high-dose (250μg) group than the low-dose (50μg) group. Antibody responses against MN V3 peptide and IIIB V3 peptide were detected in three of ten animals.

Sera from the high-dose group were assayed for the presence of antibodies against the T1 peptide. In addition, an inhibition assay was carried out using T1 and T2 peptides to inhibit the binding of anti-gp120 antibodies in sera from the high dose group. No response was observed against the T1 peptide by ELISA. The T1 peptide also failed to inhibit rgp120 binding. In contrast, the T2 peptide inhibited the binding of OGS347/1 sera to rgp 120, even at the highest dilution of peptide used. This indicates that a large proportion of the anti-gp120 activity in the anti-OGS347 sera is directed against the T2 epitope, rather than the

GPGRAF sequence at the tip of the V3 loop. This was substantiated by testing the terminal bleed sera at a 1/10 dilution in an MN virus neutralization assay. All of the sera tested were found to be negative.

2.2 CTL Responses Induced by OGS347 VLPs

Three strains of mice, C3H-Hej (H-2^k), C57BL/6(H-2^b) and BALB/c (H- 2^d), were immunised with 50μg doses of OGS347 VLPs, by the i.m. route, without adjuvant. 27 days following immunisation, splenocytes were prepared from mice of each strain. The splenocytes were divided into three aliquots and expanded with either peptide T1 , T2 or T4.1. After 7 days, the effector cells were tested against appropriate cell lines (LBRM [H-2^k], EL4 [H-2^b] and P815 [H-2^d]) labelled with peptides T1, T2 or T4.1. Specific CTLs against T1, T2 or T4.1 were not detected.

2.3 T-helper Cell Responses Induced by OGS347 VLPs

Three strains of mice, C3H-Hej (H-2^k), C57BL/6 (H-2^b) and BALB/c (H- 2^d), were immunised with 50μg of OGS347 VLPs as an alum precipitate. After 7 days, lymph nodes draining the site of injection (din) were removed and single cell suspensions prepared. These cells were stimulated in vitro with the peptides T1, T2 or T4.1, or with Ty particles. Strong proliferative responses were seen against the T2 peptide in all three strains. Much lower responses were observed against T1 and T4.1, with only the BALB/c mice (H-2^d) responding with a stimulation index (SI) of >2 to the T1 peptide. Measurement of IL2 in the supernatant, using a bioassay based on the response of the CTL line to IL2, gave a similar pattern of results.

Example 3 ; Multiple V3 Constructions from different isolates

OGS 561 is a derivative of pOGS 40, which is disclosed in copending patent application WO-A-93/20840. At the 3' end of the TyA gene are three consecutive V3 loops, in order HXBII, MN, RF. These comprise the amino acid sequences

SNCTRPNNNTRKRIRIQRGPGRAFVTIGKIGMMRQAHCNISG

(SEQ ID 2)

SNCTRPNYNKRKRIHIGPGRAFYTTKNIIGTIRQAHCNISG (SEQ ID1 ) SNCTRPNNNTRKSITKGPGRVIYATGQIIGDIRKAHCNLSGS

(SEQ ID 3)

which are linked by Bam H1 sites which encode two redundant amino acids glycine and serine.

OGS 561 induced cross-reactive anti-MN and IIIB antibodies in rabbits, mice and macaques more frequently than OGS 530 (MN alone).

Groups of rabbits were immunised intramuscularly with OGS 530 or OGS 561 in either alum adjuvant or FCA. The results are summarised in Table 1. In most experiments the animals received four or five doses of 250 or 500ug of VLPs per dose.

In one experiment (561/1), rabbits immunised with OGS 561 in alum produced anti-IIIB, RF and MN antibodies with titres of between 1/100 and 1/350 in 60-80% of the animals. Sera from 1/5 animals neutralised the IIIB isolate and 2/5 sera neutralised the MN isolate. There was no RF neutralisation. TABLE 1: MN, RF & IIIB ANTIBODY RESPONSES IN RABBITS IMMUNISED WITH MN & COMBINATION MN/IIIB/RF VLPS

F. Freund's Adjuvant

i.m. Intramuscular

Rb. Rabbit

* Assay antigen

1 Recombinant gp120 based ELISA

2 peptide based ELISA

3 Mean of responders

N.T. Not tested

In a subsequent experiment (561/2), the results were similar, except that 100% of animals responded to the IIIB component and 70% to the MN component. Titres ranged from 1/100 to 1/350 for MN and from 1 /100 to 1/500 for IIIB.

Four macaques were immunised intramuscularly with either OGS 530 or OGS 561 in alum adjuvant. Two doses of 1.0mg were given at week 0 and week 6. The titres at week 8 are shown in Table 2. Low anti-V3 titres are detectable at this relatively early time-point.

Table 2

Example 4 Overcoming MHC restriction using V3 : Sendai nucleoprotein constructions

Sendai NP:HIV-1V3:Ty VLPs were prepared. The CTL epitope from HTV is H-2d restricted whereas the epitope from Sendai NP is H-2b restricted. Two different mouse haplotypes (BALB/c H-2d and C57BL/6 H-2b) were immunised. pOGS 265 is a derivative of pOGS 40, which is disclosed in copending patent application WO-A-93/20840. At the 3' end of the TyA gene is a sequence encoding a core CTL epitope from Sendai nucleoprotein (amino acids 325-332) inserted into the BamH1 restriction site. Oligomers encoding the IIIB V3 region (amino acids 301-322) were cloned into the new Bam H1 restriction site to create pOGS280.

In addition, oligomers encoding the core CTL epitope of Sendai NP were cloned into the Bam H1 site of a construct carrying a sequence encoding the IIIB V3 loop, pOGS 260. pOGS 260 is a derivative of pOGS 40, which is disclosed in copending patent application WO-A-93/20840. At the 3' end of the TyA gene is a sequence encoding amino acids 301- 322 of the IIIB V3 region. This dual epitope construct was designated pOGS 279. A third transformant, pOGS 281, contained two copies of the Sendai NP insert. These constructs are shown in Table 3.

Table 3

The epitopes are linked by Bam H1 sites which encode two redundant amino acids glycine and serine, shown in lower case.

Mice were immunised intramuscularly with OGS 279, OGS 280 or OGS 281 in the absence of adjuvant. On day 28 after immunisation, spleens from each group of mice (BALB/c and C57BL/6) were removed and single cell suspensions prepared. Effector cells were restimulated with 25μg of V3 (H-2d) or Sendai NP (H-2b). After 7 days, the effector cells induced in this way were incubated with control or target cells (p815 for H-2d, EL4 for H-2b) labelled with 50μg of V3 or nucleoprotein peptide, as appropriate. This resulted in the lysis of those target cells where the MHC class I antigen bears peptides corresponding to the CTL epitope.

Figure 1 shows that BALB/c mice immunised with OGS 279, OGS 280 or OGS 281 produce CTLs specific to the H-2d restricted V3 sequence but none to Sendai core NP, regardless of the orientation of the epitope with respect to the nucleoprotein. The C57BL6 mice are not capable of presenting the sequence from V3 to CTLs but instead produce a CTL response to the core epitope sequence from Sendai NP.

Example 5 Sendai nucleoprotein : vesicular stomatitis virus nucleoprotein constructions

Sendai NP.VSV NP:Ty VLPs were prepared. The CTL epitopes from

Sendai NP and VSV NP are both H-2b restricted.

Oligomers encoding the Sendai NP CTL epitope region (amino acids 325-

332) and VSV NP (amino acids 52-59) were cloned into the Bam HI of pOGS 40 at the 3' end of the TyA gene. This construct was designated pOGS282.

This construct is shown in Table 4.

Table 4

The epitopes are linked by two amino acids from the Sendai NP sequence, shown in lower case. C57BL/6 mice were immunised intramuscularly with 50μg OGS282 in the absence of adjuvant. On day 17 after immunisation, spleens were removed from the mice and single cell suspensions prepared. Splenocytes were restimulated with 25μg of VSV or Sendai NP, or with both peptides. After 7 days, the effector cells induced in this way were incubated with control or target EL4 cells labelled with 50μg of VSV or

Sendai nucleoprotein peptide, or with both, as appropriate. This resulted in the lysis of the target cells in all cases, showing that Sendai- specific and VSV-specific CTLs are produced.

Sequence IDs

1 SNCTRPNYNKRKRIHIGPGRAFYTTKNIIGTIRQAHCNISG

2 SNCTRPNNNTRKRIRIQRGPGRAFVTIGKIGMMRQAHCNISG

3 SNCTRPNNNTRKSITKGPGRVIYATGQIIGDIRKAHCNLSGS 4 NNTRKRIRIQRGPGRAFVTIGKgsAPGNYPAL

5 APGNYPALgsNNTRKRIRIQRGPGRAFVTIGK

6 NNTRKRIRIQRGPGRAFVTIGKgsAPGNYPALgsAPGNYPAL

7 APGNYPALwsRGYVYQGL

Claims

1. A hybrid particle-forming protein comprising a first amino acid sequence substantially homologous with a yeast retrotransposon Ty p1 protein or a GAG protein of a RNA retrovirus fused to a plurality of different antigenic amino acid sequences corresponding to sequences derived from or associated with an aetiological agent or a tumour.

2. A particle-forming protein as claimed in claim 1 where the antigenic amino-acid sequences comprise variants of the same epitope of different strain variants of an aetiological agent.

3. A particle-forming protein as claimed in claim 1 where the antigenic amino-acid sequences comprise two or more different antigenic epitopes.

4. A particle-forming protein as claimed in claim 3 where the antigenic amino-acid sequences comprise a B-cell and a T-cell epitope.

5. A particle-forming protein as claimed in claim 3 where the antigenic amino-acid sequences have different MHC restriction.

6 A particle-forming protein as claimed in claim 3 where the antigenic amino-acid sequences are derived from different aetiological agents.

7 A particle-forming protein as claimed in any one of claims 1-6 where the particle-forming protein is the yeast retrotransposon Ty p1 protein.

8 A particle-forming protein as claimed in any one of claims 1-6 where the particle-forming protein is the GAG protein of a retrovirus.

9. A particle-forming protein as claimed in claim 8 where the retrovirus from which the particle-forming GAG protein is derived is

HIV-1, HIV-2, HTLV-I, HTLV,II, HTLV-III, SIV, BIV, LAV, ELAV, CIAV, murine Leukaemia virus, Moloney murine leukaemia virus and Feline Leukaemia virus.

10. A particle-forming protein as claimed in claim 9 where the retrovirus from which the particle-forming GAG protein is derived is of lentiviral origin.

11. A particle- forming protein as claimed in claim 10 where the retrovirus from which the particle-forming GAG protein is derived is

HIV-1, HIV-2, SIV, BIV or FIV.

12. A particle-forming protein as claimed in claim 11 where the stability of particles formed from the fusion proteins is enhanced by removing some or all of the basic C-terminus of the GAG precursor protein (p55 in HIV-1).

13. A particle-forming protein as claimed in claim 12 where the GAG protein is myristylated.

14. A particle-forming protein as claimed in any one of claims 1 to 13 where the aetiological agent is a microorganism such as a virus, bacterium, fungus or parasite.

15. A particle-forming protein as claimed in claim 14 where the virus is a retrovirus, such as HIV-1, HIV-2, HTLV-I, HTLV-H, HTLV-III, SIV, BIV, ELAV, CIAV, murine leukaemia virus, Moloney murine leukaemia virus, and feline leukaemia virus; an orthomyxovirus, such as influenza A or B; a paramyxovirus, such as parainfluenza virus, mumps, measles, RSV and Sendai virus; a papovavirus, such as HPV; an arenavirus, such as LCMV of humans or mice; a hepadnavirus, such as Hepatitis B virus; a herpes virus, such as HSV, VZV, CMV, or EBV.

16. A particle-forming protein as claimed in claim 14 where the tumour-associated or derived antigen is a proteinaceous human tumour antigen, such as a melanoma-associated antigen, or an epithelial-tumour associated antigen such as from breast or colon carcinoma.

17. A particle-forming protein as claimed in claim 14 where the antigenic sequence is derived from a bacterium, such as of genus Neisseria, Bordetella, Listeria, Mycobacteria or Leishmania, from a parasite, such as from the genus Trypanosoma, Schizosoma or Plasmodium, or from a fungus, such as from the genus Candida, Aspergillus, Cryptococcus, Histoplasma or Blastomyces.

18. A particle-forming protein as claimed in claim 14 or claim 15 where at least one of the antigenic sequences is an epitope from: 1) HIV (particularly HIV-1) gp120,

2) HIV (particularly HIV-1) p24,

3) Influenza virus nucleoprotein and haemagglutinin,

4) LCMV nucleoprotein,

5) HPV L1, L2, E4, E6 and E7 proteins,

6) p97 melanoma associated antigen, 7) GA 733-2 epithelial tumour-associated antigen,

8) MUC-1 epithelial tumour-associated antigen,

9) Mycobacterium p6,

10) Malaria CSP or RES A antigens,

11) VZV gpI, gpII or gpIII

19. A particle-forming protein as claimed in claim 18 where the epitope is the V3 loop or GPGR loop of the envelope glycoprotein gp120 of a lentivirus.

20. A particle comprising a population of homologous hybrid proteins as claimed in any of claims 1 to 19.

21. A particle comprising a population of heterologous hybrid proteins as claimed in any of claims 1 to 19 .

22. Nucleic acid coding for a fusion protein as claimed in any of claims 1 to 19.

23. A vector including nucleic acid as as claimed in claim 22.

24. A host cell carrying a vector as claimed in claim 23.

25. A host cell as claimed in claim 24 where the host cell is E. coli

26. A host cell as claimed in claim 24 where the host cell is a yeast cell such as Saccharomyces cerevisiae or Pichia pastoris

27. Host cells as claimed in claim 24 where the host cell is a mammalian cell such as a COS or CHO cell or an insect cell such as Spodoptera frugiperda.

28. Antibodies raised or directed against particulate antigens as claimed in any of claims 1 to 21.

29. The use of hybrid proteins and /or particles as claimed in any one of claims 1 to 21 in the preparation of an immunotherapeutic or prophylactic vaccine.

30 The use of particulate antigens as claimed in claims 1 to 21 as a diagnostic agent.

31. A pharmaceutical or veterinary composition comprising a protein as claimed in any one of claims 1 to 21 together with a pharmaceutically and /or veterinarily acceptable carrier.