WO2003048198A2 - Polypeptides inducing apoptosis, polynucleotides encoding them and therapeutic uses thereof - Google Patents

Abstract

The invention concerns an isolated or purified polypeptide and the polynucleotide encoding it inducing, through activation of initiator caspase 8, the formation of apoptotic bodies capable of stimulating in a vertebrate a humoral immune response and preferably, a B-dependent humoral immune response.

Description

 POLYPEPTIDES INDUCING APOPTOSIS, POLYNUCLEOTIDES ENCODING SAME AND THERAPEUTIC APPLICATIONS THEREOF

BACKGROUND OF THE INVENTION a) Field of the invention

The present invention relates to the induction of apoptosis, the production of apoptotic bodies and their therapeutic applications.

b) Brief description of the prior art

When a cell is infected with a virus, cell physiology is disturbed and these changes can lead to apoptosis. It is known that apoptosis can be induced by the rabies virus. Indeed, it has been established that the strains of rabies virus ERA and CVS infect lymphocytes but that only infection of lymphocytes by the rabies vaccine strain induces apoptosis (46). Paradoxically, the immunogenicity of strains of rabies virus is inversely proportional to the degree of cell death they induce (Thoulouze et a /. (1997) J. Virol., 71: 7372-80). The initiation of the apoptotic process or its blocking, either by the addition of the caspase inhibitor zVAD-FMK, or by overexpression of Bcl-2 has no effect on the viral production of the cells.

This indicates that the apoptotic process viro-induced by the ERA strain is an event which does not modify the viral cycle, certainly because it intervenes too late and that apoptosis does not improve the release of viral particles. Rabies apoptosis cannot be induced by simply incubating cells with non-replicating virus and this makes it possible to exclude that the rabies virus induces apoptosis by interaction with membrane receptors.

Rabies apoptosis involves the activation of caspases 1, 3 and 8, so it is dependent caspase. In zVAD-FMK treatment experiments it was found that a subpopulation of apoptotic cells were resistant to the action of this inhibitor. This suggested that an independent caspase pathway should also be activated. Recently it has just been shown in the laboratory that the strains which induce apoptosis activate the AIF pathway (Prehaud et al., Forum Neurosciences, 2001). Both forms of apoptosis are blocked by Bcl-2.

Recently, a correlation has been reported between the degree of induced apoptosis and the overexpression of glycoprotein G (Dietzschold, Faber & Schnell, XII International Meeting on Advances in Rabies Ressearch and Control in the Americas, 2001). In addition, a study carried out with Jurkat cell clones overexpressing Bcl-2 has shown that this resistance to apoptosis is accompanied by a modification of the distribution of the viral antigens N and G. Thus in cells which do not enter not in apoptosis, the viral antigens are concentrated in perinuclear globular structures. On the contrary, in cells which enter apoptosis, they accumulate in the cytoplasm and this accumulation leads to death by apoptosis. These results recall those observed in the system of lymphoblastoid cells not transfected with Bcl-2, between a non-apoptogenic strain (CVS) and a pro-apoptotic strain (ERA). We thus see that the transition to an apoptogenic phenotype is characterized by the accumulation of viral antigens in a diffuse way throughout the cytoplasm and that the non-apoptogenic phenotype corresponds to an accumulation of proteins in perinuclear globular structures. This suggested that the rabies apoptosis pathway is of the toxic type.

The disintegration of the integrity of the infected cell leads to the generation of apoptotic bodies which contain, in addition to pieces of nuclei and cytoplasm, viral antigens. Numerous experimental evidences now attest that these structures are extremely effective immunopotentiators when they are taken up by certain cells presenting CPA antigens (Dendritic cells) and that there is presentation of viral antigens by MHC class molecules. I, therefore an induction of the cellular immune response (Albert et al. (1998) Nature 392: 86; Bérard and al. (2000) J. Exp. Med. 192: 1535; Arrode et al. (2000) J. Virol. 74 (21): 10018-24; Paczesky et al. (2000) Cancer Res. 61: 2386; Sasaki et al. (2001) Nature Biotechnology 19: 543-547).

The possibility of inducing apoptosis in a specific and controlled manner so that it can be used in order to increase the immunostimulatory power of vaccines, such as an adjuvant, in order to stimulate a humoral immune response explains the need to characterize a protein which induces the production of apoptotic bodies capable of stimulating the response of B cells of the immune system.

The present invention meets this need and other needs as will be apparent to a person skilled in the art on reading the present description of the invention

SUMMARY OF THE INVENTION

The present invention relates to an isolated or purified polypeptide and to the polynucleotide coding for it, inducing, through the activation of initiating caspase 8, the formation of apoptotic bodies capable of stimulating in a vertebrate a humoral immune response and preferably, a humoral immune response type B dependent.

The polypeptide of the invention can be selected, according to various preferred embodiments, from the group consisting of: a) a glycoprotein G, having a sequence comprising at least one proline in position 27, an isoleucine in position 48, a valine in position 219 or an arginine at position 491; b) a fragment of at least 100 amino acids of the glycoprotein G as defined in a); and c) a polypeptide comprising an amino acid sequence having at least 75% homology with an amino acid sequence as defined by SEQ ID NO: 2.

According to another preferred embodiment, the polypeptide of the invention also comprises a threonine in position 89. Another object of the invention is to provide an isolated or purified chimeric polypeptide comprising at least one polypeptide as defined according to the invention associated with an exogenous polypeptide sequence. Preferably, the exogenous polypeptide sequence is the sequence of an B epitope and / or a T epitope.

According to another preferred embodiment, the polypeptide of the invention comprises a nucleotide sequence having at least 75% homology with the nucleotite sequence as defined by SEQ ID NO: 1.

Another object of the invention is to provide an isolated or purified oligonucleotide comprising a sequence chosen from the group consisting of the sequences SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO 7.

The object of the invention is further to provide an expression or cloning vector which contains a polynucleotide of the invention. According to certain preferred embodiments, the vector is selected from the group consisting of recombinant viruses and recombinant plasmids having an origin of eukaryotic replication.

The invention relates, in particular, to the vectors pRevTRE.G.ERA, pRevTRE.N.ERA, pRev-TRE-G-CVS and pRev-TRE-N-CVS deposited at the CNCM, on November 30, 2001 under the numbers d accession 1-2760, 1-2761, 1 -2758 and 1 -2759 respectively.

The invention further relates to a host cell containing a polynucleotide or a vector according to the present invention. In a preferred embodiment, the polynucleotide or vector is stably integrated into the genome of the host cell. In another preferred embodiment, the host cell is of lymphoblastoid origin. The invention also relates to an immunogenic or vaccine composition containing at least one element chosen from the group consisting of a) a polypeptide, b) a polynucleotide, c) a vector, d) a host cell or an apoptotic body comprising a), b ), c) and or d), and a pharmaceutically acceptable vehicle.

Furthermore, the object of the invention is also to provide a method of inducing humoral immunity comprising the administration to a vertebrate of an immunogenic or vaccine composition according to the present invention.

The invention also relates to the use of a polypeptide, a vector, a host cell or an apoptotic body according to the present invention for the preparation of a vaccine or an immunity adjuvant. . Preferably, the vaccine or adjuvanted immunity induces a humoral immune response in a vertebrate. More preferably, the vaccine or adjuvant of immunity is intended for the prevention or treatment of viral encephalitis, rabies, cancer or neuropathology. It will be understood that the vaccine or the adjuvant according to the present invention relates to human or animal use.

One of the major advantages of the present invention is that the nucleic acids and the polypeptides according to the invention make it possible, via the activation of caspase 8, which initiates the induction of the formation of apoptotic bodies capable of stimulating the humoral immune response of the target host.

Many other objectives and advantages of the present invention will appear on reading the nonlimiting description of the invention which follows. BRIEF DESCRIPTION OF THE FIGURES

Figure 1 is a diagram illustrating the mechanism of the RevTet-ON gene expression system.

Figure 2 is a diagram illustrating the formation of infectious but replication-deficient retroviral particles.

Figure 3 is a diagram illustrating the demonstration of the expression of the various proteins in transgenic Jurkat lines. The expression of the glycoproteins is visualized using the biotinylated primary antibody 8.2 (1/100) followed by strep-PE (1/100). For the N proteins, we used a mixture of three Japanese anti-N antibodies (each 1/50) followed by a biotinylated anti-mouse (1/100) and strep-PE (1/100). The M1 marker is used to quantify the fluorescence of cells treated with Dox compared to untreated cells.

FIG. 4 represents the analysis of the size and the granulosity of the cells by flow cytometry. The living cells, forming the R1 population, are characterized by a small granulosity and a large size. The dead cells, forming the R2 population, for their part, have a greater granulosity and a reduction in their size. The quantification of the cells in apoptosis was made by the difference between the percentage of cells in apoptosis present in the cultures treated with Dox and the percentage of apoptosis in the cultures not treated. These results are representative of several analyzes.

Figures 5A, 5B and 5C show an analysis of cells in apoptosis by Hoechst 33342 labeling (0.5 mg / ml) and counting using the Metamorph offline software. (A) Cells in apoptosis, a white arrow shows cells with a characteristic fragmentation of nuclear DNA. (B) Non-apoptotic cells without nucleus fragmentation (white arrow). (C) Cells counted using Metamorph offline software.

FIG. 6 represents the measurement of the activation of caspase I by flow cytometry. The M1 marker is used to measure the fluorescence in cells treated with Dox compared to untreated cells.

FIG. 7 represents the measurement of the activation of caspase 8. The marker is analyzed by flow cytometry. The M1 marker is used to measure the fluorescence in cells treated with Dox compared to untreated cells.

Figure 8 is a graph showing the relative growth of the anti-rabies antibody titer (AU / ml, ELISA) (all proteins combined) produced in the days following the injection of apoptotic bodies (square points) or whole cells ( diamond dots) in syngeneic BALB / c mice of Wehi cells and following the recall.

Figure 9 is a graph showing the production of anti-protein N antibodies produced after injection of apoptotic bodies or whole cells into BALB / c syngeneic mice of Wehi cells.

Figure 10 shows the nucleotide sequence (SEQ ID NO: 1) of the gene coding for the protein G of the rabid strain ERA.

Figure 11 shows the peptide sequence (SEQ ID NO: 2) of the G-ERA protein deduced from the sequence shown in Figure 12.

Figure 12 shows the nucleotide sequence (SEQ ID NO: 1) of the gene coding for the protein G-ERA and the corresponding deduced peptide sequence (SEQ ID NO: 2). Figure 13 shows the specific labeling of NT2-N cells infected with the ERA virus, against non-infected NT 2-N cells.

Figure 14 represents the quality of measurements in% of NT2-N cells in apoptosis according to whether they are uninfected, infected with CVS or infected with ERA, evaluated after staining with Hoechst.

FIG. 15 represents an example of the detection obtained of the G-ERA antigen in green monkey kidney cells (CV1) infected with the monoclonal antibody 8.2 (collection of the Viral Neuroimmonology Unit).

FIG. 16 represents the flow cytometry profiles of lymphoblastoid cells of mice, of infected EL4 line. The R 2 populations are the cells entering apoptosis, while the R1 populations are the living cells.

FIG. 17 represents the determination by flow cytometry of the quantity of cells exhibiting both PE (and FITC) fluorescence. In the population co-infected with the vv PHA and vv G-ERA viruses, the population of cells expressing at the both the G-ERA and HA antigens on the surface of their plasma membranes represent 82% of the population.

Figure 18 shows an example of the type of cells observed when making EL4 cells with the vv PHA and vv G-ERA viruses. Hoechst 33342 staining allows visualization of cell nuclei. The cells have a green fluorescence characterizing the detection of the HA antigen and a more or less significant nuclear fragmentation depending on the cells and visualized in blue. Apoptotic bodies contain two types of viral antigens: HA (red, 18-B-D) and G-ERA (green, 18-C-D).

Figure 19 shows the apoptogenic capacity of different viruses in Jurkat cells. The recombinant chimera R-N2C is a strain where in a SN10 genome, the glycoprotein gene has been replaced by that of the G protein of the CVS-N2C strain. If the strain SN10 is pro-apoptogenic in lymphoblastoid cells, this is not the case for the chimeric strain R-N2C. This is verified by apoptosis measurements carried out by different methods (percentage of activated caspase 8, tunnel or nuclear fragmentation).

FIG. 20 represents the apoptogenic capacity of different viruses in neuronal cells of SK-N-SH line. The recombinant chimera R-N2C is a strain where in a genome of SN10 type, the gene of glycoprotein has been replaced by that of the protein G of the strain CVS-N2C (see FIG. 19). If the strain SN10 is pro-apoptogenic in neuronal cells, this is not the case for the chimeric strain R-N2C. This is verified by apoptosis measurements carried out by different methods (percentage of activated caspase 8, tunnel or nuclear fragmentation).

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to an isolated or purified polypeptide which induces the formation of at least one apoptotic body capable of stimulating a humoral immune response, for example a humoral type B response in a vertebrate and preferably in a mammal.

By "isolated or purified" is meant the molecules which have been altered by humans from their native state, that is to say that if such a molecule exists in nature, it has been changed and / or removed from its initial environment. For example, a polynucleotide or polypeptide naturally present in a living organism is not "isolated". However, the same polynucleotide or polypeptide when separated from its normal environment and / or obtained by cloning, amplification and / or by chemical synthesis is considered according to the present invention as being "isolated". Besides, a polynucleotide or polypeptide which is introduced into an organism by transformation, manipulation genetic or by any other method of genetic engineering is considered "isolated" even if it is still present in said organism.

By "humoral immune response" is meant an in vivo or in vitro reaction in response to contact with an immunogen. A humoral reaction is generally induced by the production of antibodies generated against said immunogen.

By “apoptotic body” is meant a nucleic acid / protein complex formed by cell fragmentation during the apoptotic process of the cell. 1. Polypeptide and polynucleotide

In a first aspect, the present invention relates to an immunostimulatory polypeptide which stimulates the response of humoral immunity by inducing the formation of apoptotic bodies by the activation of the caspase pathway and very particularly that of the initiating caspase-8. In a preferred embodiment of the present invention, the polypeptide is selected from the group consisting of: a) a glycoprotein G which has a sequence comprising at least one proline at position 27, an isoleucine at position 48, a valine at position 219 or an arginine at position 491, preferably, this glycoprotein is derived from a rabies virus; b) a fragment of at least 100 amino acids of the protein as defined in a); c) a polypeptide comprising a nucleotide sequence having at least 75% homology with an amino acid sequence as defined by the sequence SEQ ID NO.2

The polypeptide of the present invention can also comprise a threonine at position 89. Advantageously, the inventors have discovered that a polypeptide having as sequence peptide the sequence SEQ ID NO: 2 was of particular interest. The present invention also relates to a chimeric polypeptide having a sequence comprising at least one polypeptide as defined previously associated with an exogenous sequence, for example that of a B epitope and / or a T epitope

The polypeptides according to the present invention can be prepared and / or obtained by any suitable method known to a person skilled in the art. They can in particular be obtained by chemical synthesis but it is also possible to obtain them by molecular biology techniques, using in particular RT-PCR and various appropriate vectors and cell lines as described below.

The present invention also relates to oligonucleotides having a sequence selected from the group consisting of the sequences SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO 6 and SEQ ID NO: 7. These oligonucleotides being preferably used as a probe or as a primer, for example, in the context of an RT-PCR amplification of a polynucleotide which codes for the polypeptide of the invention. To this end, the present invention also relates to any polynucleotide encoding a polypeptide as defined above. More specifically, the polynucleotide preferably comprises a nucleotide sequence having at least 75% homology with that defined by the sequence SEQ ID NO: 1.

According to the context of the present invention, the term "polynucleotide" refers to any polyribonucleotide or poly-deoxyribonucleotide, which may be a modified DNA or RNA molecule. This definition includes without limitation, single or double stranded DNA or RNA; DNA or RNA consisting of a mixture of single-double stranded region; single-double-triple strand and the DNA / RNA hybrid molecules made up of a mixture of single-double and / or triple-strand region. Polynucleotides can also include triple strand regions of DNA and / or RNA. The strands of such regions can come from the same molecule or from a different molecule. The molecule comprising a triple helix region can be an oligonucleotide. The polynucleotides can further comprise DNA and / or RNA molecules comprising modified bases such as tritylated bases. Thus, such molecules modified for the purpose of stability or for any other reason are included within the scope of the present invention. In addition, DNA and / or RNA molecules comprising particular bases such as inosine are included within the scope of the invention. Any polynucleotide which has been chemically, enzymatically or metabolically modified but which has retained the biochemical properties of the original polypeptide, that is to say which has retained its immunostimulatory power through the activation of caspase 8 initiator and therefore the formation of apoptotic bodies is included within the scope of the present invention.

In the context of the present invention, the term "polypeptide" is defined as being any peptide or protein comprising at least two amino acids linked by a modified or unmodified peptide link. The term polypeptide refers to short chain molecules such as peptides, oligopeptides or oligomers or to long chains such as proteins. A peptide according to the present invention generally comprises from 2 to 20 amino acids, preferably from 2 to 10 amino acids, and even more preferably from 2 to 5 amino acids. A polypeptide according to the present invention can comprise modified amino acids. Thus, the polypeptide of the present invention can also be modified by a natural process such as post-transcriptional modifications or by a chemical process. Some examples of these modifications are: acetylation, acylation, ADP-ribosylation, amidation, covalent bond with flavin, covalent bond with a heme, covalent bond with a nucleotide or a nucleotide derivative, bond with a lipid or a lipid derivative, covalent bond with a phosphotidylinositol, crosslinking, cyclization, disulfide bond formation, demethylation, cysteine molecule formation, pyroglutamate formation, formylation, gamma-carboxylation, hydroxylation, iodination, methylation, oxidation, phosphorylation, racemization, hydroxylation, etc. Thus, any modification of the polypeptide which does not have the effect of eliminating the biochemical characteristics of the original polypeptide, that is to say the capacity to activate the initiating caspase 8 and by the very fact of causing the formation apoptotic bodies which have the effect of stimulating a humoral immune response are covered within the scope of the present invention. 2. Vector and Cell

The subject of the invention is also vectors of expression and / or cloning

(plasmid, cosmid, virus, etc.) comprising the nucleotide sequences coding for the polypeptides of the invention. It also targets the host cells comprising these expression vectors from which the polypeptide of the present invention can be recovered by means known to those skilled in the art.

The invention also relates to any vector (cloning and / or expression) and any cellular host (prokaryotic or eukaryotic) transformed, transfected or infected with such a vector, and comprising the regulatory elements allowing the expression of the sequence of nucleotides coding for a peptide according to the invention. Preferably, the vectors are expression or cloning vectors comprising a polynucleotide of the present invention. Advantageously, such vectors are selected from the group consisting of recombinant viruses and recombinant plasmids having an origin of eukaryotic replication. More preferably, the vectors are those named pRevTRE.G.ERA and pRevTRE.N.ERA (used as a negative control), pRev-G-CVS, pRev-TRE-N-CVS which have been deposited at the C.N.C.M. November 30, 2001 under accession numbers I-2760, 1-2761, 1-2758 and I-2759 respectively.

The invention also relates to host cells, preferably of lymphoblastic origin, which contain either a polynucleotide defined by SEQ ID NO: 1 or a fragment thereof, preferably this is integrated in a stable manner, or either a polypeptide or a vector as defined above.

In the context of the present invention, the term "vector" refers to a polynucleotide construct designed to be transfected into different cell types. As a result, these vectors target the expression vectors designed for the expression of a nucleotide sequence in a host cell; cloning vectors designed for the isolation, propagation and replication of inserted nucleotides; viral vectors designed for the production of recombinant virus or viral-like particle; or shuttle vectors that include attributes from more than one vector. 3. Composition and use

The present invention also relates to the use of these polypeptides and of the polynucleotides encoding them, of these host cells and of the apoptotic bodies induced according to the present invention for the preparation of vaccines or adjuvants of immunity.

By “adjuvant” is meant any substance or molecule introduced into the composition of a vaccine or of an immunogenic composition leading to an increase in the immune response. It is therefore understood that the polypeptide of the present invention can serve as an adjuvant to thereby enhance the humoral immune response against an antigen of interest. Among the antigens against which one may want to stimulate an immune response, include, for example, viral, tumor and cellular antigens whose overexpression or alteration lead to a pathology (for example, neuropathology). It is also clear that other types of adjuvants well known to those skilled in the art can be added to the vaccine or immunogenic composition of the present invention described below.

Thus, the present invention relates to immunogenic or vaccine compositions containing apoptotic bodies as described below and which stimulate a humoral immune response and a pharmaceutically acceptable vehicle. Preferably, said apoptotic bodies are prepared from a host cell as defined above and more particularly by transfection or micro-injection of said polypeptide of the invention into the host cell.

In a preferred embodiment, the immunogenic or vaccine composition contains an element chosen from the group consisting of:

- a polypeptide according to the present invention;

- a polynucleotide according to the present invention; - a vector according to the present invention; and

- a host cell according to the present invention. These compositions can be advantageous for preventing or treating viral encephalitis including rabies, cancers or neuropathologies. For example, it is possible to envisage a method of inducing humoral immunity against an antigen of interest comprising a step comprising the administration to a human or an animal of such an immunogenic or vaccine composition. The means of preparation and administration of the compositions of the present invention will not be described further because already known to those skilled in the art.

The following examples will make it possible to demonstrate other characteristics and advantages of the present invention.

EXAMPLES

The examples which follow serve to illustrate the scope of use of the present invention and not to limit its scope. Modifications and variations can be made without departing from the spirit or scope of the invention. Although other methods or products equivalent to those found below can be used to test or carry out the present invention, the preferred equipment and methods are described.

Example 1: Determination of a pro-apoptotic protein of the rabies virus

It has been shown previously in the laboratory (46) that rabies apoptosis in Jurkat cells requires the virus to replicate and the caspases to be activated. The entry into apoptosis is concomitant with an accumulation and a particular distribution of viral proteins in the cytoplasm. This suggests that engorgement of the cytoplasm by viral antigens induces apoptosis. Certain rabies strains, such as the attenuated strain ERA induce apoptosis, which is not the case with the highly pathogenic strain CVS.

As, it has been shown that Jurkat cells are infected with both viral strains, we hypothesized that the entry into apoptosis was linked to the intrinsic nature of viral proteins: those of ERA inducing apoptosis unlike proteins of the CVS strain. In order to identify and characterize pro-apoptotic proteins, we have developed a system for the inducible expression of rabies virus proteins under the control of the Tet operator in Jurkat cells. The objective of the work was to establish a system which makes it possible to study the molecular bases of apoptosis induced by the rabies virus outside the context of infection with a complete virus.

Material and methods

1. Viruses and cells

Laboratory strains of rabies virus ERA (vr332) and CVS (vr959) come from the American Type Cell Collection (ATCC) (Rockville, MD, USA) and were amplified on BSR cells (derived from hamster kidney fibroblast cells) BHK-21). These cells are cultured in MEM medium supplemented with 8% fetal calf serum (SVF), tryptose-phosphate, 2 mM L-Glutamine, 40 mg / ml gentamycin. The PT67 packaging cells (Retro Pack cell line from Clontech) are cultured on DMEM medium supplemented with 10% of FCS, 4 mM L-glutamine, 100 U / ml of penicillin, 100 μg / ml of streptomycin. NIH3T3 cells are cultured in DMEM medium supplemented with 10% FCS, 100U / ml of penicillin, 100 μg / ml of streptomycin, 1mM sodium-pyruvate. Jurkat rtTA cells (47) are cells cultured in suspension in RPMI 1640 medium supplemented with 10% of FCS, 2 mm of L-Glutamine, 100 U / ml of penicillin, 100 μg / ml of streptomycin and 2.5 μg / ml of puromycin. All these media and solutions come from Invitrogen (France).

2. Mechanism of the RevTet-ON gene expression system

The regulatory plasmid pRevTet-ON codes, under the control of the strong CMV promoter, for the reverse transactivator Tetracycline rtTA (a fusion between the reverse Tet repressor and activator domain of VP16) (Figure 1). The latter is fixed on its TRE response element present on the expression vector pRevTRE. The addition of Doxycycline (1 μg / ml) allows the transcription of the gene of interest. The protocol was carried out following the instructions in the RevTet manual (Clontech # K1627-1). Plasmid sequences and maps are available at http://www.clontech.com.

3. Cloning of viral genes and sequencing

The molecular biology techniques used for RT-PCR, cloning and sequencing are commonly described in the manual of Perbal (48).

4. Obtaining recombinant retroviruses from the PT67 lines

The different expression vectors pRevTRE containing the genes of interest, a selection gene with an antibiotic (HygromycinR), and Ψ + (signal necessary for the formation of retroviral particles) or the regulatory vector pRevTet-ON having the selection gene to an antibiotic (Neomycin R) were transfected into the PT67 packaging cell lines using the GenePorterTM 2 Transfection Reagent kit (GTS.Ozyme). The packaging cell line provides the structural genes necessary for the formation of a viral particle: gag (structural proteins), pol (reverse transcriptase, integrase) and env (envelope glycoprotein) integrated into the genome of the cell without no other viral sequence. Retroviral vectors can integrate stably or be transiently expressed. The viruses released by these cells are infectious but incompetent for replication, thus preventing viral production in the cell lines subsequently infected. Selection with hygromycin (0.4 mg / ml) or G418 (0.4 mg / ml) makes it possible to isolate cells which have integrated the retroviral genome and which produce the recombinant MomuLV-NCVS, MomuLV-NERA retroviruses respectively, MomuLV-GERA, MomuLV-GCVS and MomuLV-TetON.

5. Titration of retroviral stocks In order to verify the viability of the MomuLV recombinant retroviral particles produced by the PT67 packaging lines, we performed a plate titration on NIH3T3 cells. 1 ml of the culture supernatant is added with 1 μl of sterile polybrene (cationic agent facilitating fusion) at a final concentration of 4 μg / ml. Serial dilutions of the virus are then carried out (10 ^"1 to 10 ^{" 6} ) in NIH3T3 + polybrene medium (4 mg / ml). The NIH3T3 target cells (10 ⁵ cells / well) are then infected with the various dilutions. 48 hours after infection, the cells are subjected to a selection with G418 (0.8 mg / ml) for pRevTet-ON or with hygromycin (0.4 mg / ml) for pRevTRE, for one week. The virus titer corresponds to the number of colonies present at the highest dilution multiplied by the dilution factor, it is expressed in CFU / ml (Colony-forming Units / ml).

6. Obtaining stable transgenic Jurkat cell lines We first infected Jurkat cells with the recombinant MomuLV-TetON. Two days after infection, the cells which have integrated the reverse transactivator rtTA are selected with G418 (0.8 mg / ml). After one month of subculturing and amplification on selective medium, we obtained a non-clonal population of cells resistant to G418. This Jurkat TetON line is currently being characterized. Jurkat rtTA cells possessing the reverse transactivator in episomal form were infected with the various recombinant MomuLV retroviruses. Selection by puromycin (2.5 μg / ml) and hygromycin (0.4 mg / ml) made it possible to isolate stable non-clonal lines having both rtTA in episomal form and having integrated the gene for interest.

7. Analysis of the Size and Sizing of the Cells by Flow Cytometry The transgenic Jurkat cells and the control cells having reached a density of 5 × 10 ⁵ cells / ml are separated into two homogeneous batches of which only one is treated with Doxycycline at 1 μg / ml. The cells are recovered after 18 h of induction, then centrifuged for 5 min at 800 rpm (4 ° C). The cells are resuspended in PBS buffer and Cell Fix (Becton Dickinson, BD, USA). The cells are then analyzed using the FACSCalibur cytofluorimeter and the Cell Quest Pro software (BD, USA).

8. Marking with Hoechst 33342 and counting.

The transgenic Jurkat cells and the control cells treated in the same way as above are incubated for 10 min at 37 ° C. with Hoechst 33342 (0.5 mg / ml), a DNA intercalating agent making it possible to visualize cells having a fragmentation characteristic of the nucleus. The cells are centrifuged, washed with PBS, fixed for 30 min with 0.4% paraformaldehyde (PFA). After washing in PBS, the cells are taken up in buffered glycerol and mounted between slide and coverslip. The cells are observed under an inverted fluorescence microscope and a count of the cells exhibiting a characteristic fragmentation of nuclear DNA was carried out using the Metamorph offlline software (Metamorph software imaging System, Universal Imaging Corporation, Downingtown, PA). Twenty fields were randomly counted for each condition and the mean was achieved.

9. Detection of protein expression by flow cytometry.

This protocol is described in Thoulouze et al (46). The transgenic Jurkat cells and the control cells, treated in the same way as above, are centrifuged, washed with PBS buffer, fixed for 30 min with 0.4% paraformaldehyde, and permeabilized with 0.1% saponin (PB buffer ). The cells are resuspended and incubated for 1 hour at 4 ° C. with the specific monoclonal antibodies: mAb 8.2 biotinylated at (1/100) for G, and a mixture of three Mab 1/50 supplied by Dr N. Minamoto, University of Gifu ( Japan) for N proteins. The cells are washed in PB then incubated for 30 min with the appropriate components: streptavidin-PE (1/100) for the detection of the biotinylated antibody and for the N proteins, the anti mouse -biotinylated (1/100). The cells are washed in PB and finally resuspended in fluorescence labeling buffer (SB). Cells are incubated for 30 min with streptavidin-PE (1/100) to reveal the N proteins, then washed in SB buffer. The cells are analyzed using the cytofluorimeter.

10. Marking of total caspases and caspases 1 and 8 We first used a total caspase inhibitor, FITC-VAD-FMK (CaspACETM FITC-VAD-FMK in Situ Marker kit, Promega). This molecule is a fluorescent analogue of z-VAD-FMK (N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone) where the N-terminal blocking group (N-benzyloxycarbonyl = z) has been replaced by fluorescein isothiocyanate (FITC) to create a fluorescent marker. This compound contains an amino acid sequence recognized by cysteine proteases, it enters the cell, and irreversibly binds to activated caspases. Then we used the FAM-YVAD-FMK (CaspaTagTM Caspase 1 (YVAD) Activity Kit, Intergen), fluorescent analogue of N-benzyloxycarbonyl-Tyr-Val-Ala- Asp- fluoromethylketone: z-YVAD-FMK which is a specific inhibitor caspase 1, and FAM-LETD-FMK (CaspaTagTM Caspase 8 (LETD) Activity Kit, Intergen), fluorescent analogue of N-benzyloxycarbonyl-Leu-Glu-Thr-Asp-fluoromethylketone: z-LETD-FMK, which is a specific caspase 8 inhibitor. The transgenic Jurkat cells and the control cells having reached a density of ⁵ × 10 ⁵ cells / ml are separated into two homogeneous batches, one of which is treated with Doxycycline at 1 mg / ml. Eighteen hours after induction, the cells are centrifuged and incubated for 1 hour at 37 ° C. with the various caspase inhibitors. The cells are then washed twice with PBS buffer, then once with SB, before being taken up in SB buffer and Cell Fix (BD, USA). The cells are analyzed with a cytofluorimeter. Results

1. Cloning of the genes coding for the N and G proteins of the rabies strains CVS and ERA. The clones of the genes coding for the nucleoproteins and the glycoproteins of the ERA and CVS strains, used in the previous work (46) to demonstrate the pro-apoptogenic nature of the ERA strain, did not yet exist in the laboratory. The work was therefore to carry out the molecular cloning of these genes. Cloning was carried out in the plasmid pRevTRE which was to be used for the establishment of the inducible TetON system. Analysis of the rabies genome reference sequence in the EMBL database (Accession numbers M13215, M21634 by Tordo et al.) Indicated that the BamHI and Hpal restriction sites were neither present in the gene coding for nucleoprotein, or that of glycoprotein. This allowed us to perform cloning using these two restriction sites of pRevTRE. Oligonucleotides containing either the BamHI site and the start codon of the corresponding gene, or the Hpal site and the stop codon of this same gene were synthesized and are described in the following table.

Table 1: Synthetic oligonucleotides for cloning

G-BamH1-start (valid ERA CVS) SEQ ID NO: 4

5 "GGCCGGATCCATGGTTCCTCAGGCTCTC 3 '

N-BamH1-start (valid ERA / CVS) SEQ ID NO: 5

5 ¹ GGCCGGATCCATGGATGCCGACAAGATT 3 '

G-Hpa1-CVS-stop SEQ ID NO: 6

5 'GGCCGTTAACTCACAGTCTGGTCTGACC 3'

G-Hpa1-ERA-stop SEQ ID NO: 7

5 'GGCCGTTAACTCACAGTCTGGTCTCACC 3'

N-Hpa1-stop (valid ERA / CVS) SEQ ID NO: 8

5 'GGCCGTTAACTTATGAGTCACTCGAATA 3'

The total intracellular RNAs were isolated 48 hours after infection of BSR cells with the two strains of rabies virus ERA and CVS at a multiplicity of infection of 3. RT-PCRs were performed on these extracts after denaturation of the secondary structures of the RNA molecules with methyl mercury. PCR products having a size compatible with the entire coding sequence for the G and N genes (1575nt and 1353nt respectively) were obtained. These PCR products were subjected to a double BamHI / Hpal enzymatic digestion. The digestion products did not show a substantial change in their size by 0.8% agarose gel electrophoresis. This observation therefore proved that the choice of cloning sites had been relevant and that neither the N and G genes of the two strains ERA and CVS had these restriction sites. These PCR products were introduced into the vector pRevTRE digested with BamHI / Hpal and dephosphorylated. After transformation of competent XL1 Gold bacteria and selection by resistance to ampicillin, clones bacteria have been isolated. The presence of recombinant pRevTRE plasmids was verified by establishing restriction profiles and sequencing the ends of the plasmid on either side of the cloning site. At least two clones were selected by construction, and the entire sequence of the insert was carried out.

The sequences of these genes were compared with each other and with the reference sequence of the genome of the strain of rabies virus PV. The results are presented in Tables 2 and 3. This made it possible to establish that sequences corresponding to the N and G genes of the ERA and CVS strains had indeed been cloned. A clone representative of each gene was used in the rest of this work, they are called pRevTRE.N.ERA, pRevTRE.G.ERA, pRevTRE.N.CVS and pRevTRE.G.CVS. From the analysis of the sequence comparisons of the N and G proteins of ERA and CVS, it can be concluded that the N proteins and the glycoproteins diverge slightly from each other by 5 and 6 amino acids respectively. Compared to PV, reference sequence, we see that the ERA strain would be the most "distant" (9 different amino acids for N and 18 for G between these two strains) compared to CVS (4 different amino acids for N and 16 for G).

Table 2: Comparison of N ERA and CVS protein sequences with each other and with the reference strain PV.

Table 3: Comparison of sequences of the ERA and CVS glycoproteins with one another and with the reference strain PV.

2. Isolement de lignées productrices de particules rétrovirales recombinantes.

2. Isolation of lines producing recombinant retroviral particles.

As Jurkat cells are difficult to transfect by conventional methods (electroporation, transfection), we have chosen to penetrate the transgenes using the infection capacity of recombinant retroviral particles of the MomuLV virus. We had the PT67 line which contains the structural genes gag, pol and env of MomuLV. The vector pRevTRE having the regulatory sequences 3'LTR / 5'LTR as well as the packaging signal +, the transfection of the recombinant pRevTRE plasmid in these cells results in the production of retroviral particles having a transgene in their genome. We carried out the transfection of PT67 cells with the plasmids pRevTRE.N.ERA, pRevTRE.G.ERA, pRevTRE.N.CVS and pRevTRE.G.CVS. We also included a series of transfections with the plasmid pRevTet-ON which has as its transgene the tetracycline rtTA activator under the control of the CMV promoter. The first series of plasmids has the gene coding for resistance to hygromycin and the plasmid pRevTet-ON has that coding for resistance to G418. Two days after transfection, a selective medium containing either hygromycin (0.4 mg / ml) or G418 (0.8 mg / ml) is added to the culture medium. After 4 to 7 days, cells that have not integrated the defective retroviral genome die and detach. On the contrary, those which have undergone this integration and which express the drug resistance genes remain adherent and generate clusters of cells. After three phases of amplification and subculturing over a period of approximately 3 weeks, we had non-clonal lines resistant either to hygromycin or to G418 for all the plasmids originally transfected. The capacity of these cells to produce virus in the culture supernatant was checked by carrying out a titration on NIH3T3 cells. The results of these titrations are presented in Table 4. Table 4: Titration of the recombinant MomuLV retroviruses produced in the culture supernatant by the packaging lines on NIH3T3 cells.

1 ml of each virus stock is added sterile polybrene at a final concentration of 4 μg / ml. Serial dilutions are made (10 ^"1 to 10 ^{" 6} ) and each dilution is placed in a well containing 10 ⁵ NIH3T3 cells. 48 h after infection, the cells are subjected to a selection by hygromycin (0.4 mg / ml) or G418 (0.8 mg / ml). The virus titer corresponds to the number of colonies present at the highest dilution multiplied by the dilution factor.

For all the isolated lines, viral production was successfully obtained. This therefore attests to the biological viability of our constructions and their capacity to generate functional recombinant MomuLV retroviral particles. It should be noted, however, that the viral titers remain low (of the order of 102 CFU) which is significantly lower than the 105

CFUs which are described in the literature. The fact that we have isolated non-clonal packaging lines, which consist of a heterogeneous mixture of highly producing cells and cells which are weakly producing virions, could explain this result. Viral titles would therefore represent an average title. This hypothesis was verified for at least one line since by subcloning the cells we were able to isolate subclones producing viruses with a higher titer. 3. Obtaining transgenic Jurkat cell lines

Our goal was to be able to isolate transgenic lines of Jurkat cells possessing at the same time the reverse transactivator tetracycline rtTA and the N and G transgenes of the ERA or CVS strains. Jurkat cells are cells of the human lymphoblastoid line which grow in suspension with a doubling time of approximately 24 hours. These cells are sensitive to low cell densities and to cytotoxic or cytolytic mediators. First, we infected Jurkat cells with MomuLV-TetON (multiplicity of infection of 1) in order to integrate rtTA. Two days after infection, these cells were transferred to a selective medium containing G418 (0.8 mg / ml). After 7 days of culture under these conditions, living cells whose growth is slowed down by the presence of dead cells appear. After one month of subculturing and amplification in a selective medium (two passages per week), and attempt to eliminate the dead cells by centrifugation at low speed (800rpm / min, for 10 min), we obtained a non-clonal population of cells. resistant to G418. This Jurkat TetON line is currently being characterized.

In order to be able to more quickly test the phenotype of our viral transgenes, we decided to also use an already existing Jurkat rtTA line where rtTA is present in episomal form. In these cells, rtTA is maintained by selection with puromycin. This line was graciously provided to us by Dr John Hiscott (47) through Pr Alain Israël (IP, Paris). These cells were infected with the recombinant MomuLV-N.CVS, MomuLV-N.ERA, MomuLV-G-ERA or MomuLV-G.CVS retroviruses. Two days after infection, these cells were cultured in a selective medium containing 0.4 mg / ml of hygromycin. A procedure identical to that performed for Jurkat TetON cells was then applied. After 1, 5 months we had stable non-clonal lines resistant to both puromycin and hygromycin which had the integrated viral transgenes. These lines were named JrtTA-N.CVS, JrtTA-N.ERA, JrtTA-G.ERA, JrtTA-G.CVS. 4. Characterization of the transgenic Jurkat cell lines

The transgenic Jurkat cell lines were characterized for their capacity to induce the expression of the viral proteins N and G of the ERA and CVS strains after treatment with Doxycycline (Dox). Indeed, Dox has, compared to tetracycline, a 100 times higher affinity for the reverse transactivator rtTA. Therefore, it induces a greater activation of the rtTA protein and therefore a stronger induction of the gene to be expressed.

The transgenic lines have a 24-hour doubling rate which is similar to the JrtTA control cell line which only has rtTA. Like parental cells, these lines are sensitive to low cell concentrations and cytolytic factors. The JrtTA-N.CVS, JrtTA-N.ERA, JrtTA-G.ERA, JrtTA-G.CVS cells as well as the JrtTA control cells were cultured. When they reached a density of ⁵ × 10 ⁵ cells / ml, these cultures were individually separated into two homogeneous batches, only one of which was treated with Dox at 1 μg / ml. After an induction for 18 h, it is observed that the cells placed in the presence of Dox have a slight growth retardation compared to the untreated cells and have a slightly higher level of dead cells (10-15%) which translates into a certain toxicity of doxycycline treatment. The cells were centrifuged, fixed with paraformaldehyde and finally permeabilized with 0.1% saponin to verify the expression of proteins by immunochemistry. Labeling with monoclonal antibodies specific for intracellular and membrane viral N and G antigens (in the case of glycoprotein) was carried out. The analysis was then carried out by flow cytometry. The results are shown in Fig 3.

In the case of the JrTA control cells, the analysis of the labeling does not show any difference in the absence or in the presence of Dox, which attests to the specificity of the reactions. On the contrary, for transgenic Jurkat lines, a shift in the fluorescence peak is observed, which corresponds to an increase in the number of cells labeled with antibodies in the presence of Dox. We find that 46% of Jurkat cells express the N-CVS protein, and 64% express the N-ERA protein. For G proteins, the expression rate is 40% for G-ERA and 25% for G-CVS. These results indicate that the four proteins are well expressed in an inducible way in our system but that the expression rates are variable (from 25 to 64%). This could result from two situations: 1) variable effectiveness of integration sites

The system consists of an integration of the gene of interest into the genome of the Jurkat cell. This random integration can locate the transgene in a transcriptionally active region or not. In some cases, the transgene may be difficult to access, and the rtTA may more or less bind to its TRE target. The transgenic lines being for the moment non-clonal, it is therefore very likely that the integration products are located at different loci.

2) variable reactivity of the antibodies We could also explain this weak expression by the fact that the antibodies used recognize more or less well the proteins synthesized. Indeed, these viral proteins are expressed in the absence of their other viral protein partners and, for the moment, we do not know if all the molecules expressed are correctly structured or not. Otherwise, a more or less important part of the synthesized protein may not have the epitope recognized by the monoclonal antibody.

Despite the slight variations, our results demonstrate that the transgenic Jurkat cell lines inducibly produce the four proteins that we decided to study. So we have in hand a system that allows us to study the molecular basis of apoptosis.

5. Demonstration of a pro-apoptogenic phenotype We have just shown that we have a cell system

Jurkat where the expression of N and G proteins of the ERA and CVS strains was inducible. We have therefore focused on demonstrating that this expression could have an implication on cell physiology and more particularly on programmed cell death. The transgenic Jurkat lines as well as the control JrtTA line were cultured. When the cell cultures reached a density of ⁵ × 10 ⁵ cells / ml, they were separated into two homogeneous batches, one of which was treated with Dox at 1 μg / ml. The cell phenotype of apoptosis induction was analyzed 18 hours after induction either by measuring the size and particle size of the cells by flow cytometry, or by counting the cells having a fragmentation of their nuclei (Hoechst staining technique). During apoptosis, cells undergo characteristic morphological changes, some of which can be measured by flow cytometry. The dead cells, forming the R2 population, are characterized by a high granulosity (SSC: Side Light Scatter) and by a reduction in their size (FSC: For ard Light Scatter). On the other hand, the living cells, forming the R1 population, have a smaller grain size and a larger size. The results are presented in Fig 4. In all cultures we observe the presence of an R2 population. After treatment with Dox, Jurkat rtTA cells have a basic level of 10% of dead cells which could correspond to cells in apoptosis. This result shows that the presence of the reverse transactivator within the cells and the induction by Dox is not neutral for the viability of the cell population. On the other hand, a variation in this R2 population which is dependent on the transgene expressed is observed in the transgenic Jurkat lines. In fact, we note the gradient of the following number of dead cells:

G-ERA> G-CVS> N-CVS> N-ERA> JrtTA 35% 20% 15% 11% 10%

It is therefore interesting to note that the expressed viral glycoproteins cause greater cell death than nucleoproteins whatever their origin. However, it is undeniably the cells expressing the glycoprotein of the ERA strain which have the most marked phenotype. The one of the phenotypes characteristic of apoptosis is the condensation of chromatin and the fragmentation of nuclear DNA which lead to a morphological modification of cell nuclei. We have studied these phenomena by labeling the nuclei with Hoechst 33342 (0.5 mg / ml). The results obtained are shown in FIG. 5. Compared to a cell culture which does not have cells dying of apoptosis, apoptotic cell cultures have many cells where the cell nucleus is completely destructured and even fragmented into several globular components. The cells were observed under an inverted fluorescence microscope and then we performed an image analysis using the Metamorph offline software to count the cells in terminal apoptosis, that is to say having a fragmented nucleus. The data in FIG. 5 clearly show that it is the JrtTA-G.ERA line which has the most cells in apoptosis (28.2%). These results confirm the results obtained by analyzing the size and grain size of the cells, the G-ERA protein is the main pro-apoptogenic protein.

We have moreover verified that putting the culture supernatant of Dox-induced JrtTA-G.ERA cells in the presence of naive Jurkat cells does not cause apoptosis of these cells. This result indicates that the apoptosis demonstrated in the previous studies is indeed consecutive to the expression of the protein and is not due to a long-term action by release of soluble mediators.

6. Molecular characterization of viral-induced apoptosis

We have established that among the four viral proteins expressed, the glycoprotein of the ERA strain was the most pro-apoptogenic protein. So we decided to start studying what could be the mechanisms that trigger the apoptotic process. It had previously been shown in the laboratory that viral-induced apoptosis by the rabies virus was dependent on the caspase pathway (caspases 8 and 3). We therefore focused on studying what it was for the expressed G-ERA protein. The activation of the caspases was measured 18 h post-induction by flow cytometry in the cultures of JrtTA-N.CVS, JrtTA-N.ERA, JrtTA-G.ERA, JrtTA-G.CVS cells as well as JrtTA control cells. This was done using the caspase inhibitor labeled with a fluorochrome FITC-VAD-FMK (Promega, France) directly on the unfixed cells and in their culture medium. This molecule, which allows the detection of all the activated caspases, penetrates into the cell and irreversibly binds to them, thus carrying out an "in situ" labeling of these proteases. After washing the cells and fixing, the labeling can be detected with a flow cytometer. We have seen that this methodology allows the detection of activated caspases by flow cytometry. However, we also observed that the quantification of specific activation was difficult due to the presence of significant background noise. In this experiment, it was not possible to conclude for the transgenic Jurkat lines expressing the nucleoproteins. However, this analysis made it possible to show that G-ERA induces a clear activation of caspases while this is weaker for G-CVS.

In order to further characterize the caspases involved, we analyzed the activation state of two of them: caspase 1 (cytokine activator and pro-inflammatory) and caspase 8 (initiating caspase). These experiments were carried out by following the same experimental procedure as above but this time using substrates with high specific activity such as FAM-YVAD-FMK for caspase 1, or FAM-LETD-FMK for caspase 8 (Intergen, France). The results are presented in Figures 6 and 7.

Analysis with a cytofluorimeter reveals clear differences between ERA and CVS, we find that the expression of the G-ERA protein in Jurkat cells causes activation of caspase 8 significantly. Thus we observe 19% activation while the expression of the G-CVS protein only leads to an activation of 5% and that of N-CVS to 3%. We observe another profile concerning the pro-inflammatory caspase 1. The expression of the G-ERA protein only leads to an activation of 7% whereas the G-CVS protein induces a strong activation (23%); the expression of the N-CVS protein, for its part, gives an activation of 12%. N-ERA protein neither caspase 1 nor caspase 8 is activated. Following this methodology, it has proved quite difficult to quantify precisely by double labeling cells which were positive both for the detection of the viral protein and for the activation of caspases 1/8. Indeed, as we had said previously, the measurement of the expression is done rather on the cells of small granulometry and of large size. Whereas activated caspases are mainly detected on cells of larger particle size and smaller size.

However, it is clear from these results that the expression of the G-ERA protein is a main factor triggering apoptosis in Jurkat cells. This process involves the caspase pathway and more precisely the initiating caspase 8. Unlike the CVS strain, the ERA strain glycoprotein does not appear to activate caspase 1.

Discussion and Perspectives This work made it possible to set up a system of inducible expression of the proteins of the rabies virus in a cellular model of human lymphocytes (Jurkat cells). Using this model, we have established that 1) the expression of the G-ERA protein in Jurkat rtTA target lymphocytes induces apoptosis, 2) during this apoptosis, the caspase pathway is activated, 3 ) the G-ERA protein induces the activation of the initiating caspase 8, 4) the G-ERA protein has finally lost the capacity to activate the pro-inflammatory caspase 1 which remains the characteristic of the proteins of the non-apoptotic CVS strain.

These results confirm the results previously established in the laboratory according to which, viral-induced apoptosis by the rabies virus is mainly characterized by an activation of the caspase pathway (8 and 3) and that caspase 1 is rather activated by the CVS strain. . Since the CVS strain is non-apoptotic, these results suggest that caspase 1 is certainly not involved in the phenomenon of apoptosis. On the contrary, it could be involved in the inflammation. The fact that the protein N-ERA does not activate apoptotic caspases demonstrates what had been suggested on the basis of the analysis of kinetics (46), that is to say that the protein G and not N induces l apoptosis. The primary sequences of the CVS and ERA glycoproteins show that these two proteins diverge from each other by only 6 amino acids. However, these 6 differences translate into a very distinct phenotype vis-à-vis apoptosis. One may wonder why and how this glycoprotein can accomplish this task. To explain the different functioning of the two proteins, we could formulate two hypotheses:

1) G-ERA, unlike G-CVS, would be cleaved by the initiating caspase 8 at a potential non-consensual site of cleavage. 2) G-ERA would interact with pro-caspase 8 and cause its activation, while G-CVS would lack this property.

- Hypothesis 1: G-ERA, unlike G-CVS, would be cleaved by caspase 8. It has been established in two human diseases, Huntington's disease and cystic fibrosis, that cellular dysfunction is directly linked to the aggregation of proteins in the cell. It would appear, in fact, that the mutant form of huntingtin (Huntington's disease) or CFTR (cystic fibrosis transmembrane conductance regulator) causes the loss of functionality of the structure which degrades the abnormal proteins, the proteasome (49). It therefore seems that the toxic accumulation of certain proteins can induce the loss of the entire cell and cause the death of the cell. A malfunction of the proteasome could be at the origin of other neurodegenerative diseases such as Parkinson's disease, Alzheimer's disease and Spinocerebral ataxia (50-53). Huntington's disease is characterized by the generation of N-terminal proteolytic fragments of huntingtin, containing repeats of several glutamines, which aggregate in the nucleus and cytoplasm of neurons. Accumulation of the pathological form of huntingtin is thought to result from abnormal cleavage by the caspases (54). Huntingtin is said to be cleaved by caspase 6 at the IVLD sequence which is a potential non-consensual site for cleavage of group III caspases (55, 56). We could hypothesize that in a similar way to what is observed with Huntingtin, G-ERA would be cleaved by caspases which would induce an abnormal aggregation of the protein leading to a deterioration of the function of the proteasome and the death of the cell. On the contrary, G-CVS is not cleaved by caspases, does not accumulate abnormally and therefore does not induce apoptosis. This hypothesis can be advanced since the analysis of the protein sequences of the glyproteins of the CVS and ERA strains highlights the existence of a potential non-consensual site of cleavage by group III caspases. The precise analysis shows that for ERA, the IVLD site is in fact transferred to 48IVED51, while for CVS it is transferred to 48WED51. It should therefore now be established that the IVED site is functional while the WED site is not, by studying whether these glycoproteins are cleaved by group III caspases (the initiating caspases). Subsequently, it would be interesting to identify 2) the nature of the caspase which would be responsible for cleavage, 3) if the cleavage products accumulate in the cytoplasm (and possibly in the nucleus), and 4) if in this the functionality of the proteasome is impaired.

This hypothesis, if true, would explain why a engorgement of the cytoplasm by viral antigens is observed in the case of infection of Jurkat cells by the ERA strain and not in the case of CVS.

- Hypothesis 2: G-ERA, would interact with pro-caspase 8 and cause its activation.

Our experiments have shown that cells which express the G protein of the ERA strain activate caspase 8. During the pathway of activation of initiating caspase 8 by signal transduction via the Fas-L / Fas interaction, there is recruitment of the zymogen (pro-caspase8) at the DISC level by the interaction of the effector domain DED of FADD with the DEDs of pro-caspase 8. This leads to the proteolytic activation of pro-caspase 8 and the release caspase 8 activated in the cytoplasm (57). Rabies apoptosis cannot be induced by simply incubating cells with non-replicating virus. This makes it possible to exclude that the rabies virus induces apoptosis by interaction with membrane receptors like Fas. It is likely that Fas did not is not involved in rabies apoptosis since the addition of anti-Fas antibodies does not increase apoptosis (58). We must therefore think that it is the intracellular synthesis of the G protein of the ERA strain which directly or indirectly causes this activation of caspase 8. It is interesting to note that between amino acids 254 and 417 of the G protein, there is a sequence which has homology with the two DEDs (A and B) of procaspase 8. Nevertheless, this sequence is conserved between the CVS and ERA strains, it is therefore likely that if there is a functional difference, between ERA and CVS glycoproteins, this does not happen at the level of the primary sequences. Predictions of the secondary structures of these 2 proteins made with the peptidestructure software from GCG (Genetic Computer Group, University of Wisconsin, USA) show that the mutations present on either side of this sequence cause different folding of the G protein. We can therefore hypothesize that this could possibly lead to exposure of this sequence of homology with the DEDs for the strain ERA on the surface of the trimers of G and on the contrary to its masking for CVS. It has been demonstrated that in the absence of Fas-L / Fas and FADD, pro-caspase 8 can be activated by an autoproteolytic activity mediated by the interaction of pro-caspase 8 molecules with each other following the interaction of their DED (59, 60). It could be envisaged that during its synthesis and its post-translational maturation, the glycoprotein of the strain ERA could play a role of catalyst in the oligomerization of the molecules of pro-caspase 8. The demonstration of an interaction between the G-ERA protein and the pro-caspase 8 at the level of their DED could be envisaged by 1) carrying out co-immunoprecipitation experiments, using an anti-caspase 8 antibody, for example, the monoclonal antibody Ab-3 ( Oncogene) reacting with pro-caspase 8. We would expect in the case, where our hypothesis is verified, to co-immunoprecipitate the G-ERA protein and not the G-CVS protein. This result would be verified by performing reciprocal immunoprecipitation using an anti-G antibody to see if there is co-immunoprecipitation of pro-caspase 8 with the protein G-ERA and not with G-CVS. This hypothesis would provide an explanation of the activation phenomenon one of the major initiating caspases: caspase 8 which could cascade activate the other caspases.

The second hypothesis is not contradictory to the first since it would explain how the first glycoprotein molecules can be cleaved by caspases. Indeed, the G-ERA glycoprotein which may be responsible for the oligomerization of pro-caspase 8 which by autoproteolytic action becomes active, could then be possibly cleaved by caspase 8 at the potential non-consensual site of cleavage.

EXAMPLE 2 Demonstration of the Superiority of the Presentation of Antigens in the Form of Apoptotic Bodies Compared to That of Infected Infected Cells in Term of Antibody Response

When a cell is infected with a virus, cell physiology is disturbed and these changes can lead to apoptosis. The disintegration of the integrity of the infected cell leads to the generation of apoptotic bodies which contain, in addition to pieces of nuclei and cytoplasm, viral antigens. Numerous experimental evidences now attest that these structures are extremely effective immunopotentiators when they are taken up by certain cells presenting CPA antigens (dendritic cells) and that there is presentation of viral antigens by MHC class molecules. I. It is therefore relevant to consider that the possibility of inducing apoptosis in a specific and controlled manner can be used to increase the immunostimulatory power of genetically engineered vaccines. It would therefore be interesting and innovative to characterize a pro-apoptotic viral protein which would stimulate the response of B cells of the immune system. This would be particularly important for vaccines derived from recombinant vectors or nucleic acid molecules.

The Applicant shows that the apoptotic bodies, generated during the apoptosis of lymphoblastoid cells infected with the rabies virus, contain at least two of the viral proteins: the glycoprotein G and the nucleoprotein N. Experimentally, the Applicant has established in a mouse model that animals vaccinated with apoptotic bodies produce much more antibody than animals injected with whole infected cells. The Applicant therefore demonstrates an innovative advantage of the stimulation of B cells. These results suggest that the apoptic bodies supported by CPA can be an important source of antigens effectively presented to the immune system. They also suggest that the inclusion of a pro-apoptotic protein in a genetically engineered vaccine could be a determining factor in improving this vaccine. The addition of the protein G-ERA which is derived from a vaccine strain commonly used in veterinary medicine is a guarantee of satisfaction of biosecurity constraints.

Results

Immunogenicity of apoptotic bodies

Wehi cells were infected with the ERA strain of the rabies virus at a multiplicity of infection of three, which makes it possible to obtain 80% of infected cells after 48 hours of culture. Two days after infection, viro-induced apoptosis of these cells remains in the minority, although viral antigens are already being synthesized. Forty-eight hours after infection, apoptosis is then induced by a treatment with ceramide which generates the formation of apoptotic bodies. At the same time, uninfected cells are treated in the same way. Controls of the potential effect of apoptotic bodies consist of infected and uninfected cells not treated with ceramide. To cancel the vaccination effect of the live virus, the preparations are inactivated by UV (inactivation of 100% of the infectious power of the virus). The presence of apoptotic bodies was checked by immunohistochemistry and measured by ELISA (Cell death detection Elisa kit) which detects nucleosomes. The quantity of viral antigens injected was measured by ELISA quantification of the protein N (Montano-Hirose et al., 1995, Res. Virol., 146, 217-224). The effective dose by intraperitoneal injection is 200 to 400 ng per mouse. Four groups of 8 syngeneic BALB / c mice of Wehi cells were injected with the same volume of the following four preparations:

• Wehi infected intact

• intact uninfected wehi • infected wehi fragmented into apoptotic bodies

• Uninfected Wehi fragmented into apoptotic bodies

The booster effect was measured after injection on day 21 of the whole rabies vaccine (PM strain, production on VERO cell). The specific antibody titres of the rabies virus present in the sera are measured by immunoenzymatic test. The results are expressed as follows: 1) Title for cells = {Title obtained for intact infected cells - Title obtained for uninfected intact cells}, 2) Title for apoptotic bodies = {Title obtained for infected cells fragmented into apoptotic bodies - Title obtained for uninfected cells fragmented into apoptotic bodies}.

The results of these experiments are represented in FIGS. 8 and 9. As can be seen, the injection of apoptotic bodies leads to an immunostimulation of the response B. In addition, the effect of booster vaccination (heterologous for cells or apoptotic bodies) indicates a very strong immunopotentiation in mice having received a primary injection of apoptotic bodies.

Conclusion

We had shown that the apoptotic bodies, generated during the apoptosis of lymphoblastoid cells infected with the rabies virus (strain ERA), contain at least two of the viral proteins: the glycoprotein G and the nucleoprotein N. Experimentally we have established here, in a mouse model, that animals vaccinated with apoptotic bodies produce more antibodies than animals injected with whole infected cells. We we also provide innovative evidence of the stimulation of B cells by apoptotic bodies.

These results suggest that apoptotic bodies supported by CPA may be an important source of antigens that are effectively presented to the immune system. They also suggest that the inclusion of a pro-apoptotic protein in a genetically engineered vaccine could be a determining factor in improving these vaccines, which are most often weak immunogens (recombinant vectors, DNA / RNA vaccine).

EXAMPLE 3 Induction of Apoptosis of NT2-N Cells by the Strain

ERA.

Embryonic carcinoma (EC) cells are the "stem cells" (CS) that exist in some germ line tumors, teratocarcinomas. Like CS cells, human CE cells proliferate extensively in vitro and in teratomas formed in vivo after injection in immunocompromised animals. However, unlike CS cells, human CE cells are aneuploid and only differentiate in certain specific cell types (Stem Cells: Scientific progress and future research directions, June 2001, NIH report, http: //www.nih. gov / news / stemceII / scireport.htm). Thus N Tera 2D1 cells can differentiate into pure cultures of post-mitotic human neurons (NT2-N) after induction of the differentiation of N Tera 2D1 by all-trans retinoic acid, then treatment with mitosis inhibitors and purification by gentle trypsination (Andrews, PW, 1998, APMIS, 106, 158-167). NT2-N cells have all the specific markers of differentiated human neurons (Guillemain, I., The Journal Of Comparative Neurology, 2000, 422, 380-395). They can establish in synaptic contacts with each other and functional contacts with astrocytes in coculture. These cells are currently the best model because they offer the possibility of cultivating large quantities of post-mitotic human neurons with a homogeneous physiological state. These cells were used for clinical phase II clinical trials of nerve cell replenishment (Thompson, TP, Neurosurgery, 1999, 45, 4, 741-752). The culture and obtaining of NT2-N are commonly mastered in the UP of Viral Neuroimmunology.

NT2-N cultures were cultured on individual labteck type chambers treated with polylysine and laminin, a treatment which allows their adhesion. After 48 h of culture, the cell bodies express and the Neurofilament 200 (Nf) antigen of the axons. On uninfected cells, the presence of neurofilament proteins was detected by a specific anti-Nf antibody (Anti Neurofilament 200, sigma, USA, no. N4142). These cells were either infected with the rabies virus strain ERA (pro-apoptotic) at a multiplicity of infection of 5, or uninfected. 24 h postinfection the cells were fixed with 80% acetone (30 min at 4 ° C) followed by an incubation of 20 min in pure ethanol.

The integrity of the cell nucleus or its fragmentation taken as a marker of apoptosis was demonstrated by staining with Hoechst 33342 at 1 μg / μl.

The presence of viral antigens in the infected cells was verified using an anti-rabies nucleocapsid antibody (Anti-nuclear nucleocapsid, Sanofi Diagnostic Pasteur, no. 72114).

The results presented in FIG. 13 show that the NT2-N cells are specifically labeled with the anti-Nf antibody, thus attesting to their character as neurons. The cells infected with the ERA virus exhibit a punctate fluorescence characteristic of the rabies nucleocapsid. The infected cells show a breakdown of the nuclear structure, visualized by Hoechst 33342 staining, which is characteristic of the apoptotic process.

In a second step, the NT2-N cells were infected either with the ERA strain, or with the CVS strain (not apoptogenic in Jurkat cells), or else uninfected and 24 h post infection the quantity of neurons in apoptosis was evaluated. after coloring with Hoechst. The results are represented in FIG. 16. We see that the strain ERA generates a neuronal apoptosis (18.55%) which is significantly greater than the control of uninfected cells (2.8%>) or the cells infected with CVS. (4%). The ERA (non-pathogenic) strain of the rabies virus, unlike the CVS (pathogenic) strain, is therefore capable of triggering neuronal apoptosis as it does in cells of the lymphoblastic lines.

Example 4: vaccine system

Construction of vaccinia recombinants

a) Construction of a recombinant vaccinia virus expressing G-ERA

The strategy we adopted to isolate a recombinant vaccinia virus expressing G-ERA is based on that which had been successfully established by Dr MP Kieny (Transgène SA, France) for the expression of the rabies virus glycoprotein. (Kieny et al, 1984, Nature, 312, 163). To carry out this work, we used the vector pTG 186 described by Kieny et al (Biotechnology, 1986, 4, 790) which has a multiple cloning site and which allows the insertion of the gene of interest under the control of the 7.5K promoter. vaccine. The co-transfection of this shuttle plasmid in the presence of the infectious DNA of the wild Copenhagen vaccinia allows homologous recombination events to occur at the locus of the tk gene and therefore allows the obtaining and selection in the presence of BrdUrd I of recombinant TK virus ^" .

L o cu s tk

7.5 K G -ERA

C onstru ctio n of the vaccine vector r tk- sim p le reco m binant expressing G E R A

Briefly, an insert corresponding to the open reading phase of the G-ERA glycoprotein gene was isolated from the plasmid pRev-TRE.G.ERA (CNCM, n ° 1- 2760) by enzymatic cleavage by the restriction enzymes BamHI and Hpal . This insert was cloned into the vector pTG 186 previously cut by the restriction enzymes BamHI and Smal and dephosphorylated. After ligating and transforming competent DH5α bacteria, clones resistant to ampicillin were isolated. The plasmids of these bacteria were purified and analyzed by enzyme restriction. Four recombinant plasmids possessing the integration of the G-ERA gene under the control of the 7.5K promoter were characterized. The nucleic acid sequences have been established. A representative clone was selected and used for the experiments of co-transfection with the infectious DNA of vaccinia Copenhagen. Four recombinant viruses were selected. The expression of the G-ERA transgene was measured by flow cytofluorimetry after infection of green monkey kidney cells (CV1) and detection of the antigen by the monoclonal antibody 8.2 (collection of the Viral Neuroimmunology Unit). An example of the type of detection obtained is shown in Figure 15. The four recombinant viruses gave a similar expression of the glycoprotein. c) Construction of a recombinant vaccinia virus expressing G-ERA and HA-influenza

To carry out the construction of a virus expressing both G-ERA and HA-influenza, we decided to use the strategy which had been developed by Dr Barbara Coupar of CSIRO (Australia). Dr. Coupar constructed a plasmid allowing the insertion of a gene of interest under the control of the 7.5K promoter into a multiple cloning site of a shuttle plasmid (pTK-7.5 A) at the Hind III-F locus of the vaccinia (Coupar et al, 1988, Gene, 68, 1). This plasmid also has the tk gene of the HSV virus under the control of the vaccinia PF promoter. After co-transfection of this shuttle plasmid with infectious DNA from a vaccinia virus tk-, homologous recombination events can occur at the Hind III-F locus of the vaccinia virus genome. The selection of tk + viruses is carried out in the presence of methotrexate as described by Boyle and Coupar (J. Gen. Virol., 1986, 67, 1591). The use of the recombinant vaccinia virus vv-PHA (tk-), available to Nicolas Escriou (GMVR Unit) as a substrate for recombination leads to the final production of a double recombinant vector tk + expressing both HA of influenza and the gene of interest which is here G-ERA.

Locus tk Locus Hind III -F

7.5 K HA influenza PF HSV-tk 7.5K G-ERA

Construction of the vaccine vaccine tk + double recombinant HA influenza and G ERA

Briefly, an insert corresponding to the open reading phase of the G-ERA glycoprotein gene was isolated from the plasmid pRev-TRE.G.ERA (CNCM, n ° 1- 2760) by enzymatic cleavage by the restriction enzymes BamHI and Hpal . This insert was cloned into the vector pTK 7.5-A previously cut by the restriction enzymes Hind III (repaired at the Hind III site) and BamHI, and dephosphorylated. After ligation and transformation of competent DH5α bacteria, clones resistant to ampicillin were isolated. The plasmids of these bacteria were purified and analyzed by enzyme restriction. A recombinant plasmid having integration of the G-ERA gene under the control of the 7.5K promoter was characterized. The nucleic acid sequence has been established. The selection of recombinant viruses is underway.

Antibody response measurement

In order to establish the best experimental conditions for performing the antibody response measurement tests, we first tested the injection routes (iv or ip) and the doses of vaccinia virus (vv-PHA or vv -control) in a mouse model. The mice were infected intravenously with 10 ⁷ UFP of virus or 10 ⁷ and 10 ⁵ of virus by intraperitoneal route which is the commonly described route.

Blood samples were taken from the retro-orbital plexus. Specific serum antibodies were measured by ELISA test. We obtain anti-HA titers of 10,000 to 15,000 for an iv injection. 107 UFP of virus, and respectively 500-4000 and 400-4000 for ip injections of 10 ⁷ and 10 ⁵ of virus. The protective power of the induced response was tested by challenge infection with a lethal dose of homologous influenza virus: all the mice injected with the vv-PHA virus survived, unlike the animals injected with the vaccinia control.

G-ERA Glycoprotein Expressed in the Vaccine System Induces Apoptosis

Mouse lymphoblastoid cells, line EL4 (ATCC # TIB-39) were infected with the vv G-ERA virus at a multiplicity of infection of 3. After 24 hours of infection, the cells were fixed using cell fix (Becton Dickinson, USA) to be analyzed by flow cytometry (Figure 16). During apoptosis, EL4 cells undergo characteristic morphological changes, some of which can be measured by flow cytometry (Kishimoto, H. et al, 1995, J. Exp. Med, 181, 649-655). The cells entering apoptosis form the R2 population and are characterized by a high grain size (SSC: Side Light Scatter) and by a reduction in their size (FSC: Forward Light Scatter). On the other hand, living cells form the R1 population and have a smaller grain size and a larger size. As seen on the flow cytometry profiles shown in Figure 16 (A and C), the population R2 represents 67%> in the case of cells infected with the vv G-ERA virus, whereas it does not represents that 20% of all the cells after infection with the vaccinia virus not expressing a transgene (vv TG186 empty). This establishes that the expression of G-ERA using this vector retains well the pro-apoptotic property of G-ERA.

The HA protein of the influenza virus induces less apoptosis than the G ERA protein of the rabies virus

The EL4 cells were then infected with the recombinant vaccinia virus expressing the HA protein of the influenza virus (vv PHA). The results are represented in FIG. 16 B. Under these conditions, the population R2 represents 46% of the totality of the cells, whereas it represented 67% in the case of cells infected with the vv G-ERA virus. This confirms that the pro-apoptotic character is an intrinsic property of G-ERA and that it is not associated with any viral glycoprotein of which HA is the prototype.

The HA and G-ERA proteins are co-expressed on the cell surface in the case of vv PHA and vv G-ERA co-infection.

In order to determine whether the two glycoproteins G-ERA and HA could be co-expressed on the surface of EL4 cells, double infections were carried out either with the viruses vv PHA and vv G-ERA, or vv PHA and vv TG186 or finally with vv G-ERA and vv TG 186 with a multiplicity of infection of 3 for each virus. The presence of antigens on the cell surface has been demonstrated by specific labeling of either the HA antigen using a polyclonal antibody directed against the influenza virus (gift from Nicolas Escriou, GMVR Unit, Institut Pasteur) followed a secondary antibody coupled to biotin, then streptavidin coupled to phycoerythrin (PE), or a monoclonal antibody specific for the G-ERA protein (mAb 8.2) followed by a secondary antibody coupled to FITC. The quantity of cells exhibiting both PE and FITC fluorescence was determined by flow cytometry. The results are presented in FIG. 17. It is observed that in the population co-infected with the vv PHA viruses and vv G-ERA, the population of cells expressing both the G-ERA and HA antigens on the surface of their plasma membrane represents 82% of the population.

The apoptotic bodies generated during co-infection vv PHA and vv G-ERA clearly present the two antigens

The cells co-infected with the vv PHA and vv G-ERA viruses and treated with the antibodies described above were analyzed by immunofluorescence after fixation with 80% acetone and staining with Hoechst 33342. The dye Hoechst 33342 makes it possible to visualize the state of cell nuclei. Cells in apoptosis are characterized by condensation of chromatin and nuclear fragmentation. An example of the type of cells that we observe during this type of co-infection is shown in Figure 18. As we can see, the cells do show a green fluorescence characterizing the detection of G-ERA, a red fluorescence characterizing the detection HA antigen and more or less significant nuclear fragmentation depending on the cells and displayed in blue. On the other hand, we can clearly see in these immunofluorescence photographs that apoptotic bodies do contain the two types of viral antigens: HA (red, 18-B-D) and G-ERA (green, 18-C-D).

Apoptotic bodies co-presenting the G-ERA and HA proteins induce stimulation of the anti-G-ERA and anti-HA response B in mice

EL4 cells were infected with recombinant vaccinia viruses vv PHA, w G-ERA or vv TG-186 at a multiplicity of infection of three, under conditions allowing to obtain more than 95% of infected cells between 30 and 48 hours of culture. The state of infection or co-infection was established by flow cytometry analysis as described above. The measurement of the apoptosis of the population was carried out by flow cytometry (size of R2) and by immunofluorescence experiments after labeling with Hoechst 33342. The following cell preparations were carried out: 1. Cells infected with vv PHA and vv TG-186

2. Cells infected with vv G-ERA and vv TG-186

3. Cells infected with vv G-ERA and vv PHA

4. Cells infected with vv G-ERA plus cells infected with vv PHA 5. idem 3 but with the addition of the apoptosis Z inhibitor VAD-FMK

6. idem 4 but with the addition of the apoptosis Z inhibitor VAD-FMK

7. Cells infected with vv TG-186

8. Uninfected cells

The comparison of conditions 1, 2 and 7 makes it possible to establish that only G-ERA, and not PHA, has an amplifying effect on the response B directed against the antigens of vv TG-186. The comparison of conditions 1, 3 makes it possible to establish that G-ERA is an immunopotentiator against the heterologous viral antigen HA. The comparison of conditions 3 and 4 provides information on the impact of the mode of presentation of G-ERA in relation to the heterologous antigen and answers the question is what the two antigens must be expressed by the same cell or can they be presented by different cells. The comparison of the responses of antibodies directed against HA and against G-ERA under conditions 3 and 5 as well as 4 and 6, demonstrates the role of the induction of apoptosis in the immunogenic power of G-ERA

To cancel the vaccination effect of the live virus, the preparations are inactivated by UV (inactivation of 100% of the infectious power of the virus). Eight groups of 10 syngeneic C57BL mice of EL4 cells were injected intraperitoneally in the same volume. Response B is analyzed for 21 days. Blood samples are taken from the retro-orbital plexus.

The analysis of serum antibodies is carried out by ELISA tests (Montano-Hirose et al, 1995, Res. Virol., 146, 217-224), by fluorescent focal reduction test (RFFIT, Smith et al, 1980, WHO monograph, Aguilar-Setien, A. et al, 2002, J. Wild. Dis., 38 (3), 539-544) and by hemagglutination inhibition test for the HA antigen (Johnston, SL et al, 1992, Diagn. Microbiol. Infect. Dis., 15 (4), 363-365; Astoul, E. et al, 1996, J. Exp. Med., 183 ( 4), 1623-1631).

To test the persistence of the immunopotentiating effect of G-ERA, the mice were then boosted, 90 days after the first injection, a dose of inactivated influenza virus. The anti-HA antibody response was then measured 7 and 15 days after the booster.

Another series of experiments made it possible to test the effectiveness of the protective response against a lethal dose of lethal influenza virus for one mouse in two.

Example 5: apathogenic strain

The apathogenic strain of rabies virus glycoprotein is the only protein inducing caspase-dependent apoptosis in human cells

The previous work made it possible to set up an inducible expression system for rabies virus proteins in a cellular model of human lymphocytes (Jurkat cells). Using this model, we have established that 1) the expression of the G-ERA protein in Jurkat rtTA target lymphocytes induces apoptosis, 2) during this apoptosis, the caspase pathway is activated (caspase 8). We then set out to demonstrate that this phenomenon was not limited only to the ERA strain but also to other apathogenic strains and that, moreover, gycoprotein was the major determinant even in a context where the other viral proteins are expressed. To answer these questions, we benefited from the use of viruses generated by reverse genetics by Dr. B. Dietzschold (Thomas Jefferson University, Philadelphia, USA).

SN10 is the apathogenic prototype strain which allowed the development of the reverse genetic system of the rabies virus (Schnell, MJ et al, 1994, EMBO J., 13: 4195-4203, accession number in GenBank M31046). The nucleotide sequence of the gene coding for protein G is different between ERA and SN10. The CVS-N2C strain is a highly pathogenic, neurotropic and non-apoptogenic strain (Morimoto, K. et al, 1998, PNAS, 95: 3152-3156). The recombinant chimera R-N2C is a strain in which, in an SN10 genome, the glycoprotein gene has been replaced by that of the G protein of the CVS-N2C strain (Morimoto, K. et al, 2000, J. of NeuroVirol., 6: 373-381, see Figure 19). We have analyzed the apoptogenic capacity of these different viruses in Jurkat cells (see Figure 19). We observe that if the strain SN10 is pro-apoptogenic in lymphoblastoid cells, this is not the case for the chimeric strain R-N2C. This is verified by apoptosis measurements carried out by different methods (percentage of activated caspase 8, tunnel or nuclear fragmentation). This absence of apoptosis is not linked to a difficulty in infection of cells by these viruses). Thus the mere absence of G from a non-pathogenic pro-apoptotic strain is necessary and sufficient to abolish the pro-apoptogenic character of the SN10 strain. This thus demonstrates that, in this system, glycoprotein is indeed the only determining factor of the apoptogenic phenotype. The rabies virus being a neurotropic virus, we then set out to show that these results could be extended to human neuronal cells. We therefore carried out the same experiments in a line of human neuroblastomas (SK-N-SH, ATCC N ° HTB-11). The results represented in FIG. 20 are identical to those obtained with Jurkat cells. We therefore demonstrate here that glycoprotein is indeed the major determinant of viro-induced apoptosis both on lymphoblastoid cells and on neuronal cells.

Although the present invention has been described in relation to concrete and privileged embodiments, it will however appear obvious to those skilled in the art or science in question that it is possible to introduce a number of variations and modifications without derogating from the scope of the invention described in this document. BIBLIOGRAPHY

J. Gaudiπ, Y., TuITereau. C, Durrer, P., Bmrrner, J., Flamand, Λ. & uigrak, R. (1999) Mol Metnbr βiυl 1 (5, 21- 1.

2. Lafoπ, M., Wiktor, T. J. & Macfarlan, R. I (1983) J Gen Virol 64, 843-51.

3. Barge, Λ., Gagnon, J., ChaflfoUe, Λ., Tïmmins, P., Langowski, J., Ruigrok, R .. ^" W. & Gau in, Y. (1 96) Virology Z 19, 465 -70.

4. Tordo, N _.; Poch, O., Erminc, Λ „Keil, G. & Rougcon, F. (19S6) Froc Natl Λcad Sci USA 83, 3914-8.

5. Tordo, N., Poch, O., Ermine, Λ., Keilli, G. & Rυugeon, F. (1988) Virology 165, 565-76.

6. Charllon, K. M. (1994) Cuir o Microbiol Immunol 187, 95-1 19,

7. Murphy, F. Λ., Harrison, Λ. K., Wli, W. C. & Bauer, S. P. (1973) ab Invest 29, 1-16.

8. Schneider, L. G (1991) in //; The Nalural Hislory ofjiabic, ed. Bπer, G. M., Boca Raton), pp. 133.

9. Murphy, F. Λ. (1977) Arch Virol 54, 79-97.

10. Gosztoπyi, G. (1994) Curr Top Microbiol Immunol 187, 43-68.

11. aux, H., Flamand, Λ. & Blonde !, Iλ (2000) J Viiol 74, 10212-6.

12. Jacob, Y., Badrane, H., Ceccaldi, P. E. & Tordo, N. (2000) J Tvol 74, 10217-22.

13. Hemachudha, T., Plianuphak, P., Sriwanthatia, B., Manutsathit, S., Pliantlminclimda, K., Siriprasomsup, W., Ukachoke, C, Rasameechan, S. & Kaoroptham, S.] 98d) Am J Med 94, 673-7.

14. Camelo, S., Castellanos, J., Lαfage, M. & Laion, M. (2001) J Virol 75, 3427-34.

15. Galelli, Λ., Baloul, L. & Lafon, M. (2000) JNeurovirol 6, 359-72.

16. Bursch, W., Kleine, L. & Tenniswood, M. (1990) Biochem Cell Biol 68, 107 j -4.

17. Kerr,). P., Wyllie, Λ. 11. & Cuπic, Λ. R. (1972) Br J Cancer 26, 23-57.

18. Kroeπier, G., Petit, P., Zamzami, N., Vayssiere, J. L. & Mignotte, B. (1995) Faseb J 9, 1277-87.

19. Proud, W., Beyacrt. R., Declercη. W. & Vandεnabcele, P. (1999) Oncogene 18, 7719-30.

20. Roulslon, Λ., Marcellus, R. C. & Biaiilon, P. E. (1999) Annu Rev Microbiol 53, 577-628.

21. Hard ick, J. M. (1998) Semin Cell Dev Biol 9, 339-49.

22. Evcrell, IL & McFaddeti, G. (1999) Trend Microbiol 7, 160-5.

23. Green, D. R. (1998) Cell 94, 695-8.

24. Λs ketiazi, A. & Dixit, V. M. (1998) Science 28i, 1305-8.

25. PaM, H. L. &. Baeuerle, P. Λ. (1997) Trends Biochem Sci 22, 63-7.

26. Eaπishaw, W. C, Martïns, L. M. & Kaufrnann, S. H. (1999) Annu Rev Biochem 68, 383-424.

27. Nicholsoπ, D. W. (1999) CellDeath Differ 6, 1028-42.

28. Slee, EA, Adrain, C. & Martin, SJ (1999) CellDeath Differ 6 _t 1067-74.

29. Rcstifo, K P. (2000) Curr Opin Immunol 12, 597-603.

30. Loefϋer, M. & Kroe er, G. (2000) Exp Cell Res 256, 19-26.

31. Migπαtte, B. & Vayssiere, I L. (1998) Eut J Biochem 252, 1-15.

32. Kroemer, G., Zamzami, N. & Susiπ, S. A. (1997) Immunol Today 18, 44-51.

33. Susin, S. Λ., Loreπzo, HK, Zamzami, N., Marzo, L, Snow, BE, Brothers, GM, Mangion, J., Jacotαt, E., Coslanlirii, P., LoefTIer, M., Larochelte, N., Goodlelt, D. R-, Aebersold, R., Siderovski, DP, Peπnïnger, JM & Kroemer, G. (1999) Λ ^r oΛ rβ 397, 441-6.

34. Gross, Λ., Me Dounell, j M. & Koismeyer, S. J. (1999) Genes & Dev 13, 1899-19 I I. 35. Kroemer, G. (1999) Biochem Soc Synψ 66, 1- 15,

36. Sπiylh, M. J. & Trapani, j. Λ. (1995) Immunol Today 1, 202-6.

38. Hardwick, J. M. (1997) Aώ' Ph rmac l 41, 295-336.37. Yuan. S O997) Cvrrcnt opinion in cell 7.47-51.

38. Hardwick, JM (1997) Aώ 'Ph rmac l 41, 295-336.

39. Miller, L. K. (1997) J Cell Physiol 173, 178-82.

40. Tyler. K. L „Squier, M. K., Rucfgers, S. E, Sclmeider, B. E., Obeihaus, S, M., Grdina, T. A., Cohen, J. J. & Dcrmody, T. S. (1995) ./ Virol 69, 6972-9.

41. Uo, M., Yamamolo, T., Watanabe, M, Ihara, T., Kamiya, FI. & Saknmî, M. (1996) FEMS immunol Me t Microbiol 15, 1 15-22.

42. Ohniiπus, IL, lleinkcle, M. & Jassoy, C. (1997) Immunol 159, 5246-52.

43. Notebom, M. H., Todd, D., Verschueren, C, A., de Gau, II. W., Curran, W. L ,, Veldka p, S., Douglas, Λ. J., McNulty, M. S., van der, E. A. & Kαch, G. (1994) ./ Virol 68, 346-5 1.

44 Despres, P., Flamand, M., Ceccaldi, P. Ji. & Deubel, V. (1996) J Virol 70, 4090-6,

45. Marianneau, P., Flamand, M., Coυrageot, M. P., Deubel, V. & Despres, P. (1998) Aim Biol Clin (Paris) 56, 395-405.

46. Thoulouze, M. I., Lafrge, M., Monlaπo-Hirαse, J. A. & Lafoπ, M. (1997) J Virol 71, 7372- 80.

47. Kwon, H., Pelletier, N., DeLuca, C, Genïn, P., Cisternas, S., Lin, R., Wainberg. M. A. & Hiscott, J. (1998) JBiol Chem 273, 7431-40.

48. Perbal, B, (1 88) A pracfical giride to moleciilar cloning.

49 Bcnce, N. F., Sanφat, R. M. & Kopito, R. R. (2001) Science 292, 1552-5.

50. Femandez-Funez, P., Niπo-Rosales, ML, de Gouyon, B., She, W. C, Lucha, J, M., Martinez, P., Turiegano, E. _r Benito, J., Capovilla , M., Skiαner, PJ, McCall, A., Canal. L, Oir, HT, Zoghbi, II. Y. & Bolas, J. (2000) Nature 408, 101-6.

51. Keller, J. N., f Ianπi, K. B. & Matkesbery, W. R. (2000) JNeurochem 75, 436-9.

52. Kitada, T, Asakawa, S., Hattori, N., Matsumiπe, H., Yamamura, Y., Miπoshima, S., Yokoc i, M., Mizuno, Y, &. Shimizu, N. (1998) Nature 392, 605-8.

53. Leroy, E., Boyer, R., Auburger, G., Leube, B., Ulm, G., Mezey, E., Harta, G., Bro nstein, MJ, Jonnalagada, S., Cheinova, T ., Dehejia, A., Lavedaπ, C, Gasser, T., Sleiubach, VJ, Wilkinsoπ, K, D. &. Poiymerσpαulos, M. H. (1998) Nature 395, 451-2.

54. Wellington, C. L., Elleiby, L. M., Iîackam, Λ. S., Margolis, RL, T firo, M. Λ., Singaraja, R., McCutcheoπ, K., Salvesen, GS, Propp, SS, Brorrun, M., Rowland, K, J „Zhaπg, T., Rasper ,, Roy, S., Thorπberry, M, Pinsky, L., Kakizuka, A., Ross, C. A, Nicholson, DW, Bredesen, DE & Ilayden, M. II. (1998) JBiol Chem 273, 158-67.

55. Wellington, CL, Singaraja, R., Ellerby, L., Savill, S., Roy, S., Leavitt, B., Cattaneo, E., Hackam, A., Sharp, A, Thorπberry, N., Nicholson, DW, Bredesen, DE & Ilayden, MR (2000) JBiol Chem 275, 19831-8,

56. Wellington, C. L., Leavilt, B. R. & Hayden, M. R. (2000) JNeυral Transm Suppl, 1-1,

57. Krammer, P. f 1. (2000) Nature 407, 789-95.

58. Jacotot, E., Cardona, A, Rebouilîat, D., Terradillos, O., Marianneau, P., Thoulouze, M. L, Lafoπ, M., Deubel, V. & Edelmari, L. (1999) Λpσpfosis 4, 169-178.

59. Yang, X., Chang, H. Y. & Baltimore, D. (1998) Mol Cell 1, 319-25.

60. Muzio, M., Stock eU. BR, Steunicke, HR, Salvesen, GS & Dixit, VM (1998) JBiol Chem 273, 2926-30. ABBREVIATIONS

DNA Deoxyribonucleic acid

AD? Λpoptosis Indncing Facior

Apaf-1 Apoptotic protease activa fingfactor l

RNA Ribonucleic Acid

BHK-21 Baby Hamster Kidmy ~ 2l (newborn hamster kidney)

BSR Hamster Kidney Fibroblast

CAV Chicken Anemia Virus

CMV Cytomegalo Virus cvs Challenge Virus Standard (CVS strain of rabies virus)

DED Death Kffec / or Domain

DISC Deaih Indncing Signait ng Comphx

Dox Doxycycline

ERA Evelyn-Rol tnicld-Abelseth (Rabies vius strain)

FADD Fas-associσted death domain priein

FITC Fluorescein isothiocyanate (Fluorescein Isothiocyanate)

Flury LEP Flury Lou> Egg Passage (Low passage in the chicken embryo)

Flury IIEP Flury High Egg Pas age (High passage in the chicken embryo)

ICAM- I Inter-Cellular Adh sion molecule (iπter-ceilulaire adhesion molecule)

IFN-y Iπterférσπ-γ

IL Interleukin

MCP-1 Monocyle Chemoiaclic Proiein- 1

MomuLV Moloney Mu fine Leukemia Virus

PBS Phosphate Btiffer Saline (6th Phosphate Buffer Solution)

PE Phycoerytlume

PFA Paraformaldehyde

PM Piltman-Moore

PV Pasteur production Virus (Pasteur pioduclion virus)

PTP Pore Transient permeability

RE Endoplasmic Reticulum

RT Reverse Transcήplase

PCR Polymerase Chain Reaction rtTA Reverse Teïracycline Transactivator

SAD Street Âlabama Dog

SNC Central Nervous System

SVF Fetal Calf Serum

TNF-α T mor Necrosis Factor (Tumor Necrosis Factor)

TRE Te (racycline-Re.ψonsive El ment (Wine Tetracycline Response Element) wine Human Immunodeficiency Virus

FITC-VAD-FMK Fluorescent analog of N-benzyloxycarboπyl-Val-Ala-Λsp-fluorornediylketone

Claims

CLAIMS: 1- Isolated or purified polypeptide characterized in that it induces the formation of apoptotic bodies capable of stimulating in a vertebrate a humoral immune response against an antigen of interest. 2- A polypeptide according to claim 1, characterized in that it activates the caspase pathway. 3- Polypeptide according to claim 2, characterized in that it activates the initiating caspase-8. - Polypeptide according to any one of claims 1 to 3, characterized in that it is selected from the group consisting of: a) a glycoprotein G derived from a rabies virus, said glycoprotein having a sequence comprising at least one praline in position 27, an isoleucine in position 48, a valine in position 219 or an arginine in position 491; and b) a fragment of at least 100 amino acids of the protein as defined in a). - Polypeptide according to claim 4, characterized in that its sequence also comprises a threonine in position 89. - Polypeptide according to claim 5, characterized in that it has the sequence SEQ ID NO: 2. - Polypeptide according to claim 4, comprising an amino acid sequence having at least 75% homology with an amino acid sequence as defined in SEQ ID NO: 2. 8- Isolated or purified chimeric polypeptide, characterized in that it comprises at least one polypeptide as defined in any one of claims 1 to 7 associated with an exogenous polypeptide sequence. 9- chimeric polypeptide according to claim 8, characterized in that said exogenous polypeptide sequence is the sequence of an epitope B and / or an epitope T. 10- Isolated or purified polynucleotide characterized in that it codes for a polypeptide according to any one of claims 1 to 9. 11- Polypeptide according to any one of claims 1 to 9, characterized in that it comprises a nucleotide sequence having at least 75% homology with a nucleotite sequence as defined by SEQ ID NO: 1. 12- Isolated or purified oligonucleotide, characterized in that it comprises a sequence chosen from the group consisting of the sequences SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO 7. 13- An expression or cloning vector, characterized in that it contains a polynucleotide as defined in claim 10 or 11. 14- Vector according to claim 13, characterized in that it is selected from the group consisting of recombinant viruses and recombinant plasmids having an origin of eukaryotic replication. 15- Vector pRevTRE.G.ERA deposited at the C.N.C.M., on November 30, 2001 under the accession number 1-2760. 16-Vector pRevTRE.N.ERA deposited at the CNCM on November 30, 2001 under the accession number 1-2761. - Vector pRev-TRE-G-CVS deposited at the CNCM on November 30, 2001 under the accession number 1-2758. - Vector pRev-TRE-N-CVS deposited at the C.N.C.M. November 30, 2001 under accession number 1-2759. - Host cell, characterized in that it contains a polynucleotide according to claim 10 or 11 or a vector according to any one of claims 13 to 18. - Host cell according to claim 19, characterized in that said polynucleotide or vector is stably integrated into the genome of the host cell. -Host cell according to claim 19 or 20, characterized in that it is of lymphoblastoid origin. -Host cell characterized in that it contains a polypeptide according to any one of claims 1 to 9. - Host cell according to any one of claims 19 to 22, characterized in that it also contains an antigen of interest or else a polynucleotide coding for said antigen of interest. -An immunogenic or vaccine composition, characterized in that it contains at least one element chosen from the group consisting of: a) a polypeptide according to any one of claims 1 to 9 and 11; b) a polynucleotide according to claim 10; d) a vector according to any one of claims 13 to 18; e) a host cell according to any one of claims 19 to 23; f) an apoptotic body comprising a), b), c) and / or d); and a pharmaceutically acceptable vehicle. -An immunogenic or vaccine composition, characterized in that it contains at least one apoptotic body capable of stimulating in a mammal a humoral immune response against an antigen of interest and a pharmaceutically acceptable vehicle. -An immunogenic or vaccine composition according to claim 25, characterized in that said apoptotic bodies are prepared from a host cell according to any one of claims 19 to 23. -An immunogenic or vaccine composition according to claim 26, characterized in that said apoptotic bodies are prepared by transfection or micro-injection of a polypeptide according to any one of claims 1 to 9 and 11. -An immunogenic or vaccine composition according to claim 25 characterized in that said apoptotic bodies are prepared from cells expressing a glycoprotein G derived from a non-pathogenic strain of the rabies virus. -An immunogenic or vaccine composition according to claim 25, characterized in that said apoptotic bodies are prepared from cells co-expressing a glycoprotein G derived from a non-pathogenic strain of the rabies virus and from an antigen of interest. - Use of a polypeptide according to any one of claims 1 to 9 and 11, or a vector according to any one of claims 13 to 18 or a host cell according to any one of claims 19 to 23 or an apoptotic body as defined according to any one of claims 24 and 25 to 28 for the preparation of a vaccine or an immunity adjuvant. -Use according to claim 30, characterized in that said vaccine or said immunity adjuvant induces a humoral immune response in a vertebrate. - Use according to claim 31, characterized in that said vaccine or said immunity adjuvant is intended for the prevention or treatment of viral encephalitis, rabies, cancer or neuropathology. -Use according to any one of claims 30 to 32, characterized in that said vaccine or said immunity adjuvant is intended for human or animal use. -A method of inducing humoral immunity, characterized in that it comprises the administration to a vertebrate of an immunogenic or vaccine composition according to any one of claims 24 and 25 to 29.