AU2136488A

AU2136488A - Vaccination against rabies-related viruses

Info

Publication number: AU2136488A
Application number: AU21364/88A
Authority: AU
Inventors: Bernhard Dietzschold; Hilary Koprowski
Original assignee: Wistar Institute of Anatomy and Biology
Current assignee: Wistar Institute of Anatomy and Biology
Priority date: 1987-07-30
Filing date: 1988-07-29
Publication date: 1989-03-01
Anticipated expiration: 2008-07-29
Also published as: NO891355L; AU608354B2; NO891355D0

Description

VACCINATION AGAINST RABIES-RELATED VIRUSES

This invention was made under grants from the National Institutes of Health. The United States Government has certain rights in the invention.

TECHNICAL FIELD OF THE INVENTION

This invention relates to rabies virus and rabies-related viruses. More particularly it relates to methods of immunizing host animals to protect against infections with rabies and rabies-related viruses.

BACKGROUND OF THE INVENTION

Rabies virus continues to be endemic in most areas of the world. It causes an acute central nervous system disease which is normally fatal to humans and domestic and wild animals.

The virus structure is bullet-shaped, consisting of a nucleocapsid core surrounded by a membrane envelope. The nucleocapsid is comprised of a single, non-segmented strand of RNA together with RNA transcriptase (L), phosphoprotein (NS), and nucleoprotein (N). The N protein, which is the major portion of the nucleocapsid, is noncovalently bound to the RNA to form the helicar ribonucleo- protein (RNP) complex. Two viral proteins are associated with the viral envelope, the major surface antigen (G) which is glycosylated and the matrix protein (M) which is thought to be located on the inner leaflet of the lipid bilayer, associating with both the C-terminal domain of the membrane-inserted G protein and the RNP structure.

Virus-neutralizing antibodies raised in animals against rabies virus, only recognize the G protein. The level of antibody production has been thought to correlate with the degree of protection afforded against live virus infection. See Crick, Post Graduate Medical Journal, Vol. 49, p. 551 (1973) and Sikes, et al., Journal of American Veterinary Medical Association, Vol. 150, p. 1491 (1971). However, there are indications that antibody alone is not sufficient to protect from viral infection. For instance, passive immunization with anti-rabies antibodies without a vaccine as post-exposure therapy, does not decrease the probability of infection. Nicholson, et al., Journal of Infect. Diseases, Vol. 140, p. 176 (1979). In addition, there exist effective vaccines which do not induce high level antibody production. One such vaccine is produced by Norden Labs.

It is possible that resistance to infection by rabies virus may require both virus-neutralizing antibodies and effector T-eell responses. Both the G protein and the nucleoprotein (N) have been shown to stimulate proliferation of rabies antigen-specific T cell lines. Most such T cell lines respond strongly to N protein and less strongly to G protein. A minority of rabies reactive T cell lines respond to G protein, but not at all to N protein. Celis et al., Journal of Immunology, Vol. 136, pp. 692-697, {1986). It has not been shown previously that N protein provides any protection against rabies infection in immunocompetent animals or humans. Current vaccines against rabies consist of an inactivated rabies virus. These vaccines do not provide any cross-protection against the rabies-related virus Mokola. Preparation of current vaccines involves growth of the virus on permissive animal cells or embryos. This method of production requires the handling of the virus by workers which entails an undesirable risk of infection. Further, some inactivated virus vaccines can cause adverse side effects, such as demyelinating allergic encephalitis and systemic reactions, in a proportion of the vaccine recipients.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a method for inducing protective immunity to rabies and rabies-related viruses.

It is another object of the present invention to provide a rabies-related virus vaccine which does not require growth of viruses.

It is yet another object of the present invention to provide a rabies-related virus vaccine which can be chemically synthesized.

It is still another object of the present invention to provide a rabies-related virus vaccine which can be produced in a microorganism.

It is an object of the present invention to provide a rabies-related virus vaccine which stimulates proliferation of both T and B cells.

It is another object of the present invention to provide a method of inducing protective immunity to rabies-related virus which does not cause demyelinating allergic encephalitis. In accordance with these objects, the present invention provides an improved method for inducing protective immunity to rabies-related viruses comprising administering a priming injection of an antigen and one or more booster injections, the improvement comprising employing a polypeptide having amino acid sequence homology to rabies-related virus nucleoprotein as said priming injection, said polypeptide being substantially free of rabies-related virus glycoprotein, said polypeptide having the ability to induce proliferation of rabies-related virus-specific human T cells.

In another embodiment of the present invention, a biologically pure sample of a polypeptide is provided capable of inducing a proliferative response in rabies-related virus-specific human T cells, said polypeptide having sequence homology with the nucleoprotein of rabies-related virus and having the ability to induce protective immunity to rabies-related viruses.

In yet another embodiment of the present invention a method is provided for inducing protective immunity to a rabies-related virus comprising administering an injection of a preparation comprising a polypeptide having sequence homology with the nucleoprotein of a rabies-related virus, said polypeptide being capable of inducing a proliferative response in rabies-related virus-specific human T cells, said polypeptide being substantially free of rabies-related virus glycoprotein.

The vaccines of this invention can be produced without growing the rabies virus and they do not cause the adverse side effects associated with inactivated virus vaccines. In addition, the vaccine provides protection against heterologous rabies-related virus strains. DETAILED DESCRIPTION OF THE INVENTION

The inventors have found that the nucleoprotein of rabies virus, a protein which is internal to the membranous envelope of the virus, can serve as an effective inducer of both humoral and cellular immune responses. Further, a particular region of the nucleoprotein (N protein), amino acid residues 313-337, has been found to contain an epitope identified by certain anti-rabies antibodies. In addition, this isolated region of the protein (as well as two other regions, corresponding to amino acid residues 369-383 and 394-408) is able to stimulate proliferation of rabies antigen-specific human T cells in the presence of irradiated autologous mononuclear cells. Other suitable peptides may be found which are able to stimulate proliferation of rabies antigen-specific human T cells and which can provide protective immunity to rabies infection. Stimulation of proliferation of T cells can be tested according to the method of Celis et al, J. Immunol. 56:426-433 (1985), and in Example 6 below. Protective immunity can be tested according to standard methods known in the art and described in Example 7 below.

As normally practiced in the art, rabies prophylaxis is performed by a series of injections, the first one being termed a priming injection and subsequent ones termed boosters. Methods for administering such injections, including appropriate formulations, dosages, anatomical localizations and timing are generally known in the medical and veterinary sciences. Generally, the dosage desired is such that will induce a protective response without causing side effects or immune tolerance. Standard inactivated rabies virus vaccine, for example, may be used in the practice of the present invention as a booster inoculation. The inactivation of virus is generally performed by treatment with a chemical, such as beta-propiolactone. Such a vaccine is available, for example, from Institut Merieux, Lyon, France.

For the purposes of the present invention the term "rabies- related virus" is meant to encompass both rabies as well as rabies-related viruses such as the Mokola and European bat virus (e.g. the Duvenhage 6 strain). This definition is used because of the finding that the vaccines of the present invention provide cross-protection to heterologous virus strains. This is a great benefit provided by the present invention over prior vaccines.

Subunits of the rabies-related virus, such as N protein or RNP complex- may be prepared and purified according to known methods. See, for example, Virology 124:330 (1983) and J. Virol. 7:241 (1971). Even purer preparations of N protein can be obtained by preparative gel electrophoresis, as described in J. Virol, 44:596 (1982). Other purification steps may be used, as are known in the art.

The useful polypeptides of the present invention can be most easily prepared by chemical synthesis, according to the solid phase method described by Merrifield, Adv. Enzymol., 32:221-296 (1969). Alternatively, proteolysis of N protein can be accomplished using enzymes or chemicals, e.g., trypsin, 3-bromo-3-methyl-2 [(2-nitrophenyl) thiol] -3 H-indole skatole, cyanogen bromide, or protease V8. Fragments after cleavage can be separated by gel electrophoresis, for example. Although many fragments can be found which are reactive with monoclonal antibodies raised against rabies N protein, only some fragments are also able to stimulate rabies-specific T cell proliferation. The inventors have found one polypeptide to be particularly useful among a set of proteolytic cleavage products of N protein. This fragment, termed N-V12b, was originally produced by protease V-8 treatment of N protein, and later chemically synthesized. Fragments N-V10c and D30 were also found to stimulate T cell proliferation, although to a lesser degree. Fragment N-V12b corresponds to the amino acids 313-337 of N protein; fragment N-V10c corresponds to amino acids 369-383; and fragment D30 corresponds to amino acids 394-405. The sequences are shown below in Table 4. The sequence of the N protein has been previously determined by Tordo et al., Nucleic Acids Research, Vol. 14, pp. 2671-2683 (1986).

When the synthetic peptides are coupled to a protein, e.g. keyhole limpet hemocyanin, either via thiol groups (Biochemistry 18:690-697 (1979)) or via amino groups with glutardialdehyde, they are found to stimulate anti-N protein antibodies in rabbits. However, the antibody titer was lower in anti-N-V12b serum than in anti-N-V10c serum.

Although the sequences of the polypeptides are defined herein with particularity (Table 4) it will be apparent to one skilled in the art that some of the amino acids can be conservatively substituted without losing the beneficial characteristics. It is also apparent to one skilled in the art that changes in length can be made without altering the antigenicity. The minimum number of amino acids necessary to comprise the epitopes has not been determined.

The polypeptides of the present invention can also be prepared through genetic engineering. That is to say that parts or all of the N gene can be cloned into suitable expression vectors, as are known in the art. Organisms transformed with the cloned gene or portion thereof can be grown to produce the polypeptide products. Recovery of the polypeptide products is within the ordinary skill of the art. Combinations or concatemers of antigenic polypeptide fragments can also be used. These can be made by synthesis, cloning, or post-synthetic covalent bonding. Liposomes, which can be employed in the practice of the present invention, are known in the art. They are used as a macromolecular carrier for the polypeptides to ensure antigenicity. Other proteins or synthetic nanoparticles may also be used as carriers. In the case of RNP administration no carrier is required. Liposomes may be formed, e.g., according to the method of Thibodeau, "Genetic variation among influenza viruses," Acad. Press, N.Y. (1981) p. 587. Generally, a mixture of lipids in particular ratios, such as phosphatidyl choline, cholesterol, and lysophosphatidyl choline, are mixed under conditions to form closed lipid vesicles. Many such conditions and methods are known in the art.

In the practice of the present invention the polypeptides are eoupled to a saturated fatty acid having from about 15 to about 21 carbon atoms, before mixing with the lipid composition to form liposomes. Suitable fatty acids include, palmitic, stearic and oleic acids. Conjugates may be formed by the method of Hopp, Molecular Immunology, 21:13-16 (1984), wherein peptide fragment spacers of gly-gly-lys-(NH₂)₂ are added to the N-terminus of the polypeptide, and the alpha and epsilon amino groups of lysine are used to couple with the fatty acids.

The biologically pure samples of the polypeptides of the present invention are substantially free of glycoprotein of rabies virus. They are sufficiently pure that after injection into mice, rabies glycoprotein-specific antibody or T cells are not induced. In addition, rabies glycoprotein specific T cells would not be stimulated in vitro to proliferate in the presence of the preparation.

The following examples are illustrative only and are not intended to limit the scope of the invention. The invention is defined by the claims appended below. EXAMPLE 1

This example describes the selection of a multiple variant virus of rabies virus.

Seven anti-glycoprotein monoclonal antibodies capable of neutralizing parent CVS-11 rabies virus were used to sequentially select antigenic variants. Serial dilutions of virus were mixed with monoclonal antibody diluted 1:100, overlayed with nutrient agar, and after 4 to 5 days of incubation at 35°C in a 5% carbon dioxide atmosphere, plaques were selected. Viruses able to replicate in the presence of the monoclonal antibodies used for their selection were recovered. The seventh generation variant CVS-V7 was not neutralized by any of 40 different rabies virus-specific neutralizing monoclonal antibody. The nucleoprotein antigen of the CVS-V7 virus remained immunologically indistinguishable from that of the CVS-11 parent virus.

EXAMPLE 2

This example shows the protection afforded by vaccination with the CVS-V7 virus, as well as the virus neutralizing antibodies it induced against parent strain CVS-11.

Vaccines prepared from the CVS-V7 variant viruses, were used to immunize mice. Groups of nine 4-week old female ICR mice were immunized with 0.2 ml of 5 serial dilutions (2000-3 ng) of the CVS-V7 inactivated virus vaccine on days 0 and 7. The inactivated virus vaccine was prepared with beta-propiolactone and adjusted to a protein concentration of 100 ug/ml. The viruses had been grown on BHK 21 cell monolayers and purified as described in Journal of Virology, Vol. 21, pp. 626-635 (1977). On day 14, vaccinated and unvaccinated control mice were bled and infected intracerebrallv with 0.03 ml (50 MIC LD₅₀) of CVS-11. (CVS-11 parent virus was used in the challenge because the CVS-V7 variant virus was not pathogenic for adult mice.) The animals were observed for 3 weeks and mortalities were recorded daily. The effective dose was calculated for each vaccine as described in Atansiu, P., "Quantitative assay and potency test of antirabies serum," In: Laboratory Techniques In Rabies, 2nd Ed., Geneva: World Health Organization, 1966, pp. 167-72. The neutralizing activity of the mouse immune sera for CVS-11 was determined as described in Lumio et al., Lance, Vol. i, p. 378 (1986).

The geometric mean tighter (GMT) was calculated for each vaccination group. Multiple group comparisons for differences in GMT were tested by one way analysis of variance and two-group GMT were compared statistically by a one-tailed test.

As can be seen in Table l, virus neutralizing activity as well as protection from death were provided by the CVS-V7 vaccine. As the CVS-V7 virus, from which the vaccine was prepared, was not neutralized by any of 40 rabies glycoprotein-specific monoclonal antibodies and yet was able to confer protection against challenge with the parental CVS-11 strain, it is concluded that the glycoprotein is not the sole factor in determining the relative efficacy of rabies prophylaxis.

EXAMPLE 3

This example demonstrates the protection afforded and the virus neutralizing antibody induced by means of rabies virus subunit vaccines which are formulated with liposomes.

The glycoprotein and ribonucleoprotein (RNP) were purified from the CVS-V7 variant virus as described in Dietzchold et al., CVS-V7-G- CVS-V7-N- some θ-rβ» Mortality Rat* no*)

7/7

e and C VS-V7 su

Isolation and Purification of a Polymeric Form of the Glycoprotein of Rabies Virus, J. Gen. Virol. 1978, 40 131-135 and Schneider et al., Rabies Group-Specific Ribonucleoprotein Antigen and a Test System for Grouping and Typing of Rhabdoviruses 1973, J. Viral. 11, 748-755. The proteins were inserted into liposomes as described in Thibodeau, Genetic Variation Among Influenze Viruses, Acad. Press, N.Y. 1981, p. 587. The amount of rabies protein liposomes used per injection is expressed as weight of the total complex. Mice were injected intracerebrally on days 0 and 7 and challenged at day 14 as described in Example 2. The resulting mortality rates and amount of virus neutralizing antibody produced are shown in Table 1.

The RNP incorporated into liposomes induced no detectable VNA against CVS-11 nor any detectable protection from death. The CVS-V7-derived G protein incorporated into liposomes indueed low titers of VNA and poor protection from death. However, when the RNP was combined with G protein, the treatment was significantly better than either protein alone, and produced results roughly comparable to those obtained with the whole virus vaccine. This too, shows the importance of N protein in immunity to rabies.

EXAMPLE 4

This example demonstrates that rabies virus RNP can augment the function of B cells producing neutralizing antibody.

Groups of mice were primed with either 5 ug of N protein (from the ERA strain of rabies) plus complete Freund's adjuvant (CFA) or CFA alone. Ten days after priming both groups of animals received serial dilutions of inactivated rabies virus vaccine or isolated rabies glycoprotein. Mice that were primed with RNP plus CFA and boosted with rabies virus vaccine developed significantly higher VNA titers than those mice primed with CFA alone and then boosted with a rabies virus vaccine. The results are shown in Table 2. In addition, the effective dose of rabies virus vaccine was determined and found to be about 10-fold lower in the mice which had been primed with RNP.

The mice which received booster vaccinations consisting of just rabies virus glycoprotein developed only low levels of VNA and were poorly protected against a lethal challenge infection with rabies virus, which can be seen in the right half of Table 2. This suggests that there must be common antigens between the priming and boosting immunizations in order to induce an effective immune response.

The mortality rates shown in Table 2 were determined by challenging all mice with an intracerebral innoculation with 50 MIC LD₅₀ of rabies virus and observing the survival up to 3 weeks post-infection.

EXAMPLE 5

This examples demonstrates that protective immunity can be induced using rabies virus RNP to an intramuscular (I.M.) challenge infection with homologous or heterologous rabies virus strains.

Purified RNP of the ERA strain of rabies or of the rabies-related strain Mokola was used to immunize Balb/c mice against an (I.M.) challenge with the rabies strain CVS-24. The RNP was isolated and purified according to the method of Schneider, et al., J. Virol., Vol. 11, pp. 748-755 (1973). Groups of 10 to 20 mice were immunized intraperitoneally (I.P.) with ERA-or Mokola-RNP plus CFA, or with CFA alone, or subcutaneously (S.C.) with ERA-RNP alone. Four weeks after immunization, the mice were challenged (I.M.) with eight mouse I.M. LD₅₀ of CVS-24 rabies virus

(see Table 3) or two mouse I.M. LD₅₀ of rabies-related virus Duvenhage 6 (see Table 4).

As can be seen in Table 3, 80% of the mice that received ERA-RNP plus CFA intraperitoneally or ERA-RNP alone subcatane- ously survived the challenge. In addition, 90% of the mice that were immunized intraperitoneally with Mokola-RNP plus CFA survived the challenge infection. In contrast only 10% of the mice that received only complete Freund Adjuvant (CFA) succumbed to rabies.

It is known that neutralizing antibody produced against the Mokola virus does not neutralize rabies virus; therefore, this experiment clearly demonstrates that the protection conferred by RNP is not due to neutralizing antibody which may have been induced by very small, undetectable amounts of glycoprotein. In addition, the results shown in Table 4 demonstrate that RNP from both the ERA strain and from the Mokola virus induce protective immunity against the rabies-related European bat virus strain Duvenhage 6. Taken together these results demonstrate that RNP purified from rabies virus and rabies-related viruses can induce protective immunity against heterologous viruses.

EXAMPLE 6

The synthetic peptides shown below in Table 5 were synthesized according to the method of Merrified, cited above. In a typical coupling reaction, the ^tboc group of the amino terminus was removed with 50% trifluoroacetic acid (TFA). After neutralization with N, N-diisopropylethylamine (DIEA), a 4- to 6-molar excess of preformed ^tboc-amino acid-pentafluorophenylester (Kisfaludy et al., Liebigs Ann. Chem. pp. 1421-1429 (1973)) and a 1-molar equivalent of DIEA in dichloromethane (1:1) were added. After bubbling with N₂ gas for 1.5-2 h, the resin was analyzed for the presence of free amino groups as described by Kaiser et al. (Anal. Biochem., 34:595-598 (1970)). Couplings were repeated until less than 1% free NH₂ groups were found.

Peptides were cleaved and deblocked with HF/thioanisol (10:1) at 0°C for 30 min, and the peptide was extracted with 0.1 M NH₄HCO₃ and lyophilized. The crude peptide was then dissolved in 0.1 M NH₄HCO₃ and purified on a BioGel™ P4 column calibrated with 0.1 M NH₄HCO₃. The eluted peptide was then applied to a Vydac RP C18 reverse-phase column and eluted with methanol- water (8:2). The elution of the peptide was monitored with a UV detector at 214 nm. To verify the amino acid sequence, aliquots of the peptide were subjected to amino acid analysis and amino acid sequencing.

The peptides were tested for the ability to stimulate T-cell proliferation in the presence of antigen-presenting populations of cells of the same HLA-DR type. At concentrations of 0.4 ug/ml to 25 ug/ml of protein, fragments N-V12b and N-V10c, and N protein isolated from ERA strain virus, stimulated proliferation of the T-cells. Hepatitis B antigens, used as controls, caused no stimulation. At low concentrations, N-V12b was more stimulatory than N protein, whereas at higher concentrations the reverse was true.

The rabies specific T-cell lines were isolated and tested as described in Celis et al., J. Immunol. 56:426-433 (1985).

EXAMPLE 7

This example demonstrates the protective capabilities of the synthetic peptides N-V10 and N-V12b against the rabies strain CVS-24.

Groups of 5 to 10 mice were immunized intraperitoneally with N-V12b-liposomes plus CFA, N-V10c-liposomes plus CFA, or liposomes plus CFA. The mice were then challenged with various amounts of CVS-24 virus.

The liposomes were formed by the method of Thibodeau, Genetic Variations Among Influenza Viruses, Acad. Press, N.Y., 587 (1981). The peptides were incorporated into liposomes as described above in Example 3. To facilitate the incorporation of the peptides into liposomes, palmitic acid was linked to the amino terminal end of the peptide as described by Hopp, Mol. Immunol., Vol. 21, pp. 13-16, 1984. In experiment No. 1, two times the mouse LM. LD₅₀ was used as challenge. As shown in Table 6, 88% of the mice that received the N-V12b-liposome vaccine survived while none of the mice immunized by N-V10c-liposome survived three weeks after the challenge.

In experiment No. 2, four times the mouse I.M. LD₅₀ was used as a challenge. Sixty-two percent of the mice immunized with peptide N-V12b-liposome vaccine survived while only 12% of the mice which received the control vaccine of liposomes plus CFA survived the challenge.

In experiment No. 3, 8 times the mouse I.M. LD₅₀ was used as a challenge. Sixty percent of the mice immunized with the N-V12b-liposome vaccine survived the challenge, while only 10 and 20 percent respectively of the mice immunized with N-V10 liposomes plus CFA and liposome control vaccine plus CFA survived.

Thus, peptide N-V12b provides good protection from rabies virus challenge when incorporated into liposomes and administered with CFA. Peptide N-V10c however does not provide good

Protective activities of N-V12b peptide liposomes in mice to i.m. challenge with CVS-24

Experiment #1, I.m. challenge with 2 MIM LD ₅₀

Experiment #2, I.m. challenge with 4 MIM LD

50

Experiment #3, I.m. challenge with 8 MIM LD

50

proteetion against rabies virus challenge, even though it is able to stimulate rabies specific T cell proliferation.

EXAMPLE 8

Peptides consisting of 15 amino acids were synthesized which together correspond to the entire amino acid sequence of the ERA-N protein. These were synthesized as described above in Example 6. Each peptide was screened for T cell proliferative activity in vitro and for protective activity in vivo as described above for peptides N-V12b and N-V10c.

The results are displayed in Table 7. Peptide number 30D demonstrated both significant T cell stimulatory activity as well as partial but significant protection against rabies virus challenge. Peptide D30 protected 60% of the immunized mouse population from a challenge of 8 times the mouse I.M. LD₅₀ of CVS-24 virus.

Claims

1. In a method for inducing protective immunity to a rabies-related virus comprising administering a priming injection of an antigen and one or more booster injections, the improvement comprising:

employing a polypeptide having amino acid sequence homology to a rabies-related virus nucleoprotein as said priming injection, said polypeptide substantially free of a rabies-related virus glycoprotein, said polypeptide having the ability to induce proliferation of rabies-related virus-specific human T cells.

2. The method of claim 1 wherein said polypeptide is the rabies-related virus nucleoprotein.

3. The method of claim 1 wherein said polypeptide is derived by cleavage of the rabies-related virus nucleoprotein.

4. The method of claim 3 wherein said cleavage is enzymatic.

5. The method of claim 3 wherein said cleavage is chemical.

6. The method of claim 1 wherein said polypeptide is chemically synthesized.

7. The method of claim 1 wherein the sequence of said polypeptide corresponds to amino acids 313 to 337 of rabies virus nucleoprotein.

8. The method of claim 1 wherein the booster injections comprise inactivated rabies-related virus.

9. The method of claim 1 wherein the booster injections comprise a mixture of rabies-related virus glycoprotein and rabies-related virus nucleoprotein.

10. The method of claim 1 wherein the polypeptide is incorporated in a liposome.

11. A biologically pure sample comprising a polypeptide capable of inducing a proliferative response in rabies-related virus-specific human T cells and of inducing protective immunity to rabies-related virus, said polypeptide having sequence homology with the nucleoprotein of rabies-related virus.

12. A biologically pure sample comprising a polypeptide of claim 11 having a sequence corresponding to amino acids 279 to 337 of rabies virus nucleoprotein.

13. A biologically pure sample of a polypeptide of claim 11 having a sequence corresponding to amino acids 313 to 337 of rabies virus nucleoprotein.

14. A method of inducing protective immunity to rabies-related virus comprising:

administering an injection of a preparation comprising a polypeptide having sequence homology with the nucleoprotein of a rabies-related virus, said polypeptide being capable of inducing a proliferative response in rabies-related virus-specific human T cells, said polypeptide being substantially free of rabies-related virus glycoprotein.

15. The method of claim 14 wherein the preparation comprises ribonucleoproteϊn complex from a rabies-related virus.

16. The method of claim 14 wherein the preparation comprises liposomes containing said polypeptide.

17. The method of claim 16 wherein the polypeptide comprises amino acids 313 to 337 of rabies virus nucleoprotein.

18. The method of claim 16 wherein the polypeptide is covalently bonded to a saturated fatty acid having from about 15 to about 21 carbon atoms.