Molecular Complex for Targeting Antigens Towards Cells Comprising Antigens and Uses Thereof for Vaccination

Abstract

The invention relates to a molecular complex for targeting the antigen towards cells comprising antigens, including at least one antigen associated with at least two ligands of surface molecules of cells comprising antigens, said complex including at least one first ligand of a sulphated sugar of the glycosaminoglycan family and a second ligand of a specific surface molecule of cells comprising antigens, and said first ligand being covalently bonded with said antigen and/or said second ligand.

Claims

1-20. (cancelled)

21. A method of enhancing an antigen-specific immune response in a subject, comprising administering to the subject an immunogenic composition comprising at least: an isolated fusion protein comprising at least an antigen and a ligand of heparan sulfates (first ligand) selected from the group consisting of a fragment of Human Immunodeficiency Virus (HIV) Tat protein comprising Tat 49-57 basic region (SEQ ID NO: 1) and a peptide comprising diphtheria toxin R domain (SEQ ID NO: 6), and a ligand of a surface molecule specific of antigen-presenting cells (second ligand) bound to the fusion protein, in an amount effective to induce an antigen-specific immune response in the subject, wherein said antigen, first and second ligands are derived from three different molecules and said composition does not comprise the full length Tat protein or diphtheria toxin, and the administration of said composition comprising the fusion protein bound to the second ligand enhances the antigen-specific immune response in the subject compared to the administration of a composition comprising the fusion protein not bound to the second ligand or the antigen bound to the second ligand without the first ligand.

22. The method of claim 21, wherein said second ligand is selected from the group consisting of: antibodies directed against MHC class II molecules, C-type lectin receptors or immunoglobulin constant region receptors and fragments thereof comprising at least the paratope, antibodies not specific or specific for said antigen and fragments thereof comprising at least the Fc region, and Staphylococcus aureus protein A, the BB fragment thereof (SEQ ID NO: 4) and the ZZ derivative thereof (SEQ ID NO: 3).

23. The method of claim 21, wherein the first ligand is the peptide Tat 49-57 (SEQ ID NO: 1) or Tat 37-57 (SEQ ID NO:7).

24. The method of claim 21, wherein said antigen is an antigen specific for a pathogenic agent or a tumor.

25. The method of claim 24, wherein said antigen is a viral, bacterial, fungal or parasite antigen.

26. The method of claim 21, wherein the immune response comprises an antibody response, helper T-cell and/or cytotoxic T-cell response specific for the antigen.

27. The method of claim 21, wherein the composition comprises a fusion protein comprising the antigen, the first ligand and the second ligand.

28. The method of claim 27, wherein the composition comprises a fusion protein comprising the antigen, the first ligand, and the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3).

29. The method of claim 28, wherein the composition comprises a fusion protein comprising the antigen, the peptide Tat 49-57 (SEQ ID NO: 1) or Tat 37-57 (SEQ ID NO: 7), and the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3).

30. The method of claim 28, wherein the composition comprises a fusion protein comprising the antigen, the peptide consisting of diphtheria toxin R domain (SEQ ID NO: 6), and the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3).

31. The method of claim 21, wherein the composition comprises: i. a fusion protein comprising the antigen, the first ligand, and an element for binding to the second ligand chosen from the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3), and ii. the second ligand chosen from immunoglobulin G or fragment thereof comprising at least the Fc region; anti-MHC class II, anti-FcgammaR, anti-DEC-205 antibody or fragment thereof comprising at least the paratope.

32. The method of claim 21, wherein said subject is a human subject.

33. The method of claim 21, which comprises an amount of fusion protein bound to the second ligand sufficient to protect said subject against an infectious disease.

34. The method of claim 21, which comprises an amount of fusion protein bound to the second ligand sufficient to treat cancer.

35. An immunogenic or vaccine composition comprising: an isolated fusion protein comprising at least an antigen specific for a pathogenic agent and a ligand of heparan sulfates (first ligand) selected from the group consisting of a fragment of HIV Tat protein comprising Tat 49-57 basic region (SEQ ID NO: 1) and a peptide comprising diphtheria toxin R domain (SEQ ID NO: 6), and a ligand of a surface molecule specific of antigen-presenting cells (second ligand) bound to the fusion protein, in an amount effective to induce an antigen-specific immune response in a subject, and a pharmaceutically acceptable vehicle, wherein said antigen, first and second ligands are derived from three different molecules and said composition does not comprise the full length Tat protein or diphtheria toxin.

36. The immunogenic or vaccine composition of claim 35, wherein said antigen is a viral, bacterial, fungal or parasite antigen.

37. The immunogenic or vaccine composition of claim 35, wherein said antigen induces at least an antibody response, helper T-cell and/or cytotoxic T-cell response specific for the antigen.

38. The immunogenic or vaccine composition of claim 35, which comprises an isolated fusion protein comprising the antigen specific for a pathogenic agent, the peptide Tat 49-57 (SEQ ID NO: 1) or Tat 37-57 (SEQ ID NO: 7), and the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3).

39. The immunogenic or vaccine composition of claim 35, which comprises an isolated fusion protein comprising the antigen specific for a pathogenic agent, the peptide consisting of diphtheria toxin R domain (SEQ ID NO: 6), and the BB fragment of Staphylococcus aureus protein A (SEQ ID NO: 4) or the ZZ derivative thereof (SEQ ID NO: 3).

Description

[0086] In addition to the above arrangements, the invention also comprises other arrangements which will emerge from the description that follows, which refer to examples of use of the complex which is the subject of the present invention and also to the appended drawings, in which:

[0087] FIG. 1 shows the cell binding of Tat101 and of four peptides derived from Tat. A. Sequence of Tat101, Tat101C(22-37)S and of three overlapping peptides. The cysteine-rich (22-37), central (core: 38-48) and basic (49-57) regions are underlined. B, C, D: Cell binding of Tat101 and of four peptides derived from Tat. Series of dilutions of the five peptides were incubated in the presence of, respectively, splenocytes (B), B lymphoma A20 cells (C) and T hybridoma T1B2 cells (D). After incubation for 30 minutes at 4 C., an anti-Tat monoclonal antibody covalently coupled to fluorescein was added and the binding of the peptides to the cells was evaluated by flow cytometry (FACS). Similar results were obtained in two separate experiments;

[0088] FIG. 2 shows the binding to various antigen-presenting cells of alpha toxin in free form or in a form coupled to the Tat peptide 37-57. A. Splenocytes were incubated with series of dilutions of alpha toxin (alpha) or alpha37-57, in the presence or absence of an excess of heparin (3 M). After incubation for 30 minutes at 4 C., a rabbit anti-alpha toxin polyclonal antibody was added. A F(ab)2 fragment of a goat anti-rabbit IgG polyclonal antibody coupled to fluorescein was added for 30 minutes and the binding of the toxin to the cells was analyzed by FACS. B. Binding to dendritic cells (DCs), analyzed according to the same protocol. C. Binding to A20 cells, analyzed according to the same protocol. Similar results were obtained in at least two separate experiments;

[0089] FIG. 3 shows that the capacity of alpha toxin to stimulate T cells is increased when it is coupled beforehand to Tat fragment 37-57 and that the stimulating effect is modified in the presence of an excess of heparin or of 37-57. Splenocytes (510.sup.5/well) were incubated with serial dilutions of alpha toxin, of alpha toxin plus peptide 37-57, or of alpha37-57, in the presence or absence of an excess (3 M) of sulfated polysaccharides (heparin, chondroitin A, chondroitin B (A and D). After three hours, T1B2 cells (A) or T1C9 cells (D) were added and the mixtures were incubated for 24 hours. The level of T stimulation which is reflected by IL-2 secretion was evaluated by means of a test for proliferation of an IL-2-dependent CTL line. A similar protocol was used to examine the antigen presentation in the presence or absence of a fixed amount (0.2 M) of free peptide 37-57 (B and E). A further analysis of the inhibition obtained with peptide 37-57 is shown in C and F. In these experiments, the splenocytes (10.sup.5/well) were preincubated at 4 C. for 30 minutes in the presence or absence of peptide 37-57. Alpha37-57 was then added to the wells for one hour at 4 C. The plates were then washed, and the T1B2 (C) or T1C9 (F) cells were added for 24 hours at 37 C. The level of T stimulation which is reflected by IL-2 secretion was evaluated by means of a test for proliferation of an IL-2-dependent CTL line. The results are expressed in cpm (number of counts per minute). Similar results were observed in at least two separate experiments;

[0090] FIG. 4 shows that the stimulating effect requires antigen processing. To examine the need for processing, previously fixed splenocytes were incubated for three hours, either with a series of dilutions of alpha toxin, alpha37-57 or peptide 32-49, before the addition of T1B2 (A), or with a series of dilutions of alpha toxin, alpha37-57, peptide Ac24-36 or peptide Ac24-36-Tat37-57 before the addition of T1C9 (B). To compare the stimulating capacities of Ac24-36 and Ac24-36-Tat37-57 in the presence of live APCs, the splenocytes were incubated for 3 hours, with these two peptides, respectively (C). T1C9 was added and the T-cell stimulation was evaluated as described in FIG. 3. Similar results were observed in at least two separate experiments;

[0091] FIG. 5 shows that the Tat fragment 37-57 increases the presentation of alpha toxin to T1B2 by B lymphocytes and splenic DCs. The B lymphocytes (A) or the DCs (B) were incubated with serial dilutions of alpha toxin, alpha toxin plus peptide 37-57, or alpha37-57. After three hours at 37 C., the T1B2 cells were added and the T-cell stimulation was evaluated as described in FIG. 3; similar results were obtained in at least two separate experiments;

[0092] FIG. 6 shows that the immune complexes containing alpha37-57 bind to A20 cells lacking FcRII or expressing FcRII and that the alpha37-57+anti-alpha complex binds the Fc receptor type II in a manner similar to alpha+anti-alpha for low antigen concentrations. Serial dilutions of alpha37-57 were incubated overnight in the presence of fixed amounts of a rabbit anti-alpha polyclonal antibody (10 nM or 25 nM). The A20 cells lacking FcRII (A) or expressing FcRII (B) were then added to the mixtures and incubated for 30 minutes at 4 C. The cells were then washed and a F(ab)2 fragment of a goat anti-rabbit IgG polyclonal antibody coupled to fluorescein was added in order to evaluate the binding of the immune complexes by FACS analysis. Similar results were obtained in four separate experiments;

[0093] FIG. 7 shows that the alpha37-57+anti-alpha complex exhibits an increased T-stimulating capacity. Series of dilutions of alpha toxin or alpha-37-57 were preincubated overnight at 4 C. in the presence or absence of a fixed amount of anti-alpha rabbit polyclonal antibodies or rabbit IgGs (12.5 nM). The A20 cells lacking FcRII (A, B) or expressing FcRII (C, D) were then added. After incubation for 3 hours at 37 C., T1B2 (A, C) or T1C9 (B, D), were added and incubated for 24 hours at 37 C. IL-2 secretion was evaluated using a line of which the growth is dependent on the presence of IL-2. Similar results were obtained in at least two separate experiments;

[0094] FIG. 8 shows that the ZZDTR-BD fusion protein is capable of binding heparin and heparan sulfate. Series of dilutions of ZZDTR-BD were incubated in microtitration plates previously adsorbed with heparin coupled to albumin in order to evaluate the capacity of the fusion protein to bind the sulfated sugar (A). The binding to the plates was detected using a rabbit polyclonal antibody and an anti-rabbit antibody goat antibody coupled to peroxydase and a substrate for this enzyme (ABTS). In order to evaluate the specificity of binding to the plates and the region of ZZDTR-BD involved in the interaction, a fixed amount of ZZDTR-BD was incubated in the plates in the presence of series of dilutions of ZZ, Hep6000 and HS, respectively (B). The binding to the plates was detected in the same way as in A;

[0095] FIG. 9 shows that the T-stimulating capacity of ZZDTR-BD depends on its ability to bind HSs and immunoglobulins. ZZDTR-BD was preincubated in the presence or absence of an excess of Hep6000 (3 M), of rabbit IgG (0.8 M) or of a mixture of Hep6000 (3 M) and rabbit IgG (0.6 M). The mixtures were added to A20 cells. After 1 hour at 37 C., the T4B6 hybridoma was added. 24 hours later, the supernatants were removed and the presence of IL-2 was evaluated using an IL-2-dependent CTL line;

[0096] FIG. 10 shows the determination of the splenocyte population bound by ZZDTR-BD. A fixed amount of this complex was incubated in the presence or absence of splenocytes and of three phycoerythrin-labeled antibodies specific for CD4+ T lymphocytes (anti-CD4-PE), for CD8+ T lymphocytes (anti-CD8-PE) and for B lymphocytes (anti-CD19-PE). After 30 minutes at 4 C., the cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. 30 minutes later, the cells were washed and analyzed by flow cytometry;

[0097] FIG. 11 shows that the T-stimulating capacity of ZZDTR-BD is increased when it is complexed beforehand with antibodies specific for determinants expressed at the surface of APCs. ZZDTR-BD was preincubated in the presence or absence of equimolar amounts of the 14-4-4S antibody, the 10-1.D.2 antibody and the M2-3 antibody, respectively. The mixtures were added to A20 cells or to splenocytes. After 1 hour at 37 C., the T4B6 hybridoma was added. 24 hours later, the supernatants were removed and the presence of IL-2 was evaluated using an IL-2-dependent CTL line;

[0098] FIG. 12 shows the binding of the alpha/anti-alpha and alpha37-57/anti-alpha complexes to BALB/c mouse splenocytes, analyzed by flow cytometry using an F(ab)2 fragment of a goat anti-rabbit IgG polyclonal antibody coupled to fluorescein. Cal FITC: splenocytes (control). Alpha/anti-alpha: splenocytes+alpha/anti-alpha immune complexes. Alpha37-57/anti-alpha: splenocytes+alpha37-57/anti-alpha immune complexes;

[0099] FIG. 13 shows the determination of the splenocyte population bound by the anti-alpha toxin/alpha toxin37-57 complex. A fixed amount of this complex was incubated in the presence or absence of splenocytes and of three phycoerythrin-labeled antibodies specific for CD4+ T lymphocytes (anti-CD4-PE), for CD8+ T lymphocytes (anti-CD8-PE) and for B lymphocytes (anti-CD19-PE). After 30 minutes at 4 C., the cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. 30 minutes later, the cells were washed and analyzed by flow cytometry;

[0100] FIG. 14 shows the determination of the splenocyte population bound by the 14-4-4S/ZZDTR-BD complex. A fixed amount of ZZDTR-BD or of the 14-4-4S/ZZDTR-BD complex was incubated in the presence or absence of splenocytes and of three phycoerythrin-labeled antibodies specific for CD4+ T lymphocytes (anti-CD4-PE), for CD8+ T lymphocytes (anti-CD8-PE) and for B lymphocytes (anti-CD19-PE). After 30 minutes at 4 C., the cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. 30 minutes later, the cells were washed and analyzed by flow cytometry;

[0101] FIG. 15 shows that the anti-MHC/ZZDTR-BD complex induces, in the absence of adjuvant, an anti-diphtheria toxin antibody response which is greater than that induced by the IgG2a/ZZDTR-BD complex. The two complexes were respectively injected in the absence of adjuvant into two groups of four BALB/c mice. The animals were taken forty-five days after the immunization and the sera were pooled. The presence of anti-diphtheria toxin antibodies was evaluated by immunoenzymatic assay;

[0102] FIG. 16 shows that a compound having the ability to target HSs and a molecule expressed specifically at the surface of APCs is capable of inducing a greater cytotoxic immune response than a compound which targets only HSs or only the molecule specifically expressed at the surface of APCs. The Tat47-57-SIINFEKL-alpha and SIINFEKL-alpha proteins (final dilution 1 M), complexed or not complexed with the anti-alpha toxin rabbit polyclonal antibody (anti-alpha, 50 nM) were added to JAWS II dendritic cells used as ACPs. After 5 h at 37 C., the cells were fixed and OT1 mouse splenocytes, which contain CD8+ T lymphocytes that recognize the T epitope, of sequence SIINFEKL, in association with class I molecules of type I-A.sup.b, were added. The level of CD8+ T stimulation of the OT1 splenocytes was evaluated by means of a cell proliferation test. The results are expressed in cpm (number of counts per minute);

[0103] FIG. 17 shows that a compound which has the ability to target HSs and a molecule expressed specifically at the surface of APCs increases the expression of the CD80 and CD86 co-stimulatory molecules more strongly than compounds which target only HSs or only the molecule specifically expressed at the surface of APCs. The wild-type alpha toxin and the Tat47-57-SIINFEKL-alpha and SIINFEKL-alpha proteins (final concentration 1 M) complexed or not complexed with the antibody (25 nM final concentration) were added to JAWS II dendritic cells used as APCs. After 24 hours at 37 C., the cells were washed, incubated in the presence of fluorescein-labeled anti-CD80 (A) and anti-CD86 (B) antibodies, and the binding of the antibodies to the cells was analyzed by flow cytometry.

EXAMPLE 1

Selection of a Tat Fragment Capable of Binding Cell-Surface Heparan Sulfates and Coupling to Alpha Toxin

[0104] 1. Materials and Methods

[0105] 1.1 Synthesis of Tat Peptides

[0106] The Tat protein (SEQ ID NO: 8; FIG. 1A) corresponds to that of the NDK isolate of HIV-1 (Groenink et al., J. Virol., 1991, 65, 1968-175) which represents the consensus sequence obtained from 66 sequences of primary isolates of HIV-1 reported in the SWISSPROT and TrEMBL databases between 1999 and 2000. The chemical synthesis of the Tat peptides was carried out by means of the Fmoc/tert-butyl strategy using an Applied Biosystems 433A automatic peptide synthesizer. The chemical process uses 0.1 mmol of Fmoc-Asp(OtBu)-PAL-PEG-PS resin, a 10-fold excess of each amino acid, dicyclohexylcarbodiimide/1-hydroxy-7-azabenzotriazole and diisopropylethylamine/N-methylpyrrolidone. The cleavage and the deprotection were carried out using a trifluoroacetic acid/triisopropylsilane/water mixture (9.5/0.25/0.25, v/v/v). The raw material was precipitated twice with tert-butyl methyl ether cooled to 4 C. and then dissolved in a 15% aqueous acetic acid solution. The peptides were then purified by reverse-phase high performance liquid chromatography (HPLC), on a Vydac C18 column (Hesperia). The peptides and the proteins synthesized were characterized by mass spectrometry and by amino acid analysis. They are stored at 20 C., in lyophilized form.

[0107] 1.2 Coupling of the Tat.sub.37-57 Peptide to Alpha Toxin

[0108] The alpha toxin of Naja nigricollis (Swiss-Prot P01468; SEQ ID NO: 9) was purified as described in Fryklund et al., Biochemistry, 1975, 14, 2865-2871. Alpha toxin monothiolate, comprising an N-terminal thiol group, was obtained using N-succinimidyl 3-(2-pyridyldithiol)propionate (SPDP) as bifunctional reagent, according to the protocol previously described (Lonetti et al., J. Exp. Med., 1999, 189, 12177-). The additional disulfide bridge was then reduced using an acetate buffer, pH 4.5, containing 0.1 M NaCl and 25 mM dithiothreithiol. The mixture was then stirred at ambient temperature for 20 min and the solution was filtered on a PD10 column equilibrated with 0.1 M phosphate buffer, pH 6.1, containing 0.1 M NaCl. The toxin monothiolate eluted from the dead volume was then incubated for 1 hour at ambient temperature in the presence of a three-fold excess of the Tat.sub.37-57 peptide, in PBS buffer. The mixture was then filtered on a PD10 column equilibrated with 0.1M phosphate buffer, pH 7, the purity of the conjugate was evaluated by reverse-phase high performance liquid chromatography (HPLC).

[0109] 1.3 Cell Binding

[0110] Cells Used

[0111] The murine B lymphoma line called A20, obtained as described in K. J. Kim et al., J. Immunol., 1979, 122, 549-, is available under No. ATCC TIB-208.

[0112] The T hybridoma, specific for a toxin and for erabutoxin a, called T1B2, was obtained as described in B. Maillre et al., J. Immunol, 1993, 150, 5270-.

[0113] The mouse splenocytes were isolated from mouse spleens. For this, the animals were sacrificed and then the spleens were removed sterilely. The spleens were dilacerated and the red blood cells were lysed at 4 C. using a Gey's buffer. The splenocytes were recovered after centrifugation for 10 minutes at 4 C.

[0114] The dendritic cells were isolated from mouse splenocytes using MACS microbeads, according to the protocol of the manufacturer (Miltenyi Biotec). Briefly, the spleens were incubated in the presence of collagenase D (2 mg/ml) for 30 minutes at 37 C. in the presence of an excess of nonspecific mouse IgGs. The cells were then incubated at 4 C. Fifteen minutes later, anti-CD11c microbeads were added and incubated for 15 minutes at 6-8 C. The splenocytes were washed, centrifuged and passed through magnetic columns. The enrichment was evaluated by FACS analysis, using four anti-CD antibodies (anti-CD4-FITC, anti-CD19-PE, anti-CD11c-FITC and anti-IA/IE-PE, Becton-Dickinson). The purity was greater than 95%.

[0115] Cell Binding of Tat101 and of Tat Peptides with Different C-Terminal Extensions

[0116] Series of dilutions of each Tat peptide were incubated in the presence of various types of cells (210.sup.5 splenocytes per well, 10.sup.5 A20 or T1B2 cells per well) for 30 min at 4 C. in PBS buffer/0.5% BSA. The cells were washed three times and incubated in the presence of a murine monoclonal antibody specific for the N-terminal region of Tat, covalently coupled to fluorescein, as described in Lecoq et al., Vaccine 2008, 26, 2615-2626 (1 g/well). After incubation for 30 min at 4 C., the cells were washed and analyzed by FACS.

[0117] Cell Binding of Alpha Toxin in Free Form or Covalently Coupled to Tat.sub.37-57 (Alpha37-57)

[0118] Series of dilutions of alpha toxin and alpha37-57 were incubated in the presence of various cell types (210.sup.5 splenocytes per well, 10.sup.5 A20 cells or 10.sup.5 dendritic cells) for 30 min at 4 C. in PBS buffer/0.5% BSA. The cells were washed and incubated in the presence of a rabbit anti-alpha toxin polyclonal antibody (1 g/well). After incubation for 30 min at 4 C., the cells were washed and an F(ab)2 fragment of a goat anti-rabbit IgG polyclonal antibody coupled to fluorescein was added. After incubation for 30 min at 4 C., the cells were washed and analyzed by FACS.

[0119] 2. Results

[0120] Tat is a heparin-binding protein (Albini et al., Oncogene, 1996, 12, 289-; Rusnati et al., J. Biol. Chem., 1997, 272, 11313-) which is internalized in cells via a mechanism requiring heparan sulfate proteoglycans (HSPGs) at the surface of the cells (Tyagi et al., J. Biol. Chem., 2001, 276, 3254). A wild-type Tat protein of 101 residues (Tat101) and four Tat-derived peptides were produced by chemical synthesis in order to select a monomeric Tat fragment capable of binding the cells (FIG. 1A).

[0121] In these derivatives, the seven cysteine residues of Tat located in the cysteine-rich region of the molecule (amino acids 22 to 37) were replaced with seven serines, given that Tat has a natural tendency to form a large variety of disulfide-bridge-mediated oligomers (Kittiworakarn et al., J. Biol. Chem., 2006, 281, 3105-). The first derivative is a complete molecule called Tat101C(22-37)S (Kittiworakarn et al., J. Biol. Chem., 2006, 281, 3105-). The other three polypeptides are C-terminal-truncated Tat fragments (1-57C(22-37)S, 1-48C(22-37)S, 1-37C(22-37)S). After having evaluated the purity of the five polypeptides, their ability to bind various cell types was compared. In these experiments, the cell binding was revealed using a monoclonal antibody directed against the N-terminal region of Tat and flow cytometry analysis (FACS).

[0122] FIG. 1B shows that Tat 101 binds splenocytes, whereas Tat101C(22-37)S interacts weakly, indicating that the cysteine-rich region plays an important role in the interaction of Tat with cells. Unlike Tat101C(22-37)S, the 1-57C(22-37)S peptide binds splenocytes like wild-type Tat101, indicating that the loss of binding due to the absence of the cysteine residues has been counterbalanced by the absence of region 58-101. It requires approximately 50 times more 1-48C(22-37)S peptides to obtain binding similar to that of the 1-57C(22-37)S peptide, indicating that the absence of residues 49 to 57, corresponding to the basic region of Tat, greatly decreases the binding capacity. Finally, no binding was observed with the 1-37C(22-37)S peptide, showing that the interaction is abolished in the absence of residues 38 to 48 corresponding to the core domain of Tat.

[0123] Similar results were obtained using the murine B lymphoma cells of the A20 line (FIG. 1C) or cells of the T hybridoma called T1B2 (FIG. 1D), indicating that Tat and fragments thereof are capable of binding a variety of cells, in accordance with the ubiquitous distribution of HSPGs. Owing to these results which show the involvement of the cysteine-rich (residues 22-37), core (residues 38 to 48) and basic residue-rich (residues 49-57) regions in cell binding, and the prior observations that Tat has a spontaneous propensity to form a large variety of disulfide-bridge-mediated oligomers (Kittiworakarn et al., J. Biol. Chem., 2006, 281, 3105-), only the basic residue-rich and core regions were selected for preparing a homogeneous monomeric fragment capable of binding cells. Cystein 37 of Tat was incorporated into this fragment in order to couple the alpha toxin of Naja nigricollis (Thai et al., J. Biol, Chem., 2004, 279, 50257-). The new peptide synthesized, called 37-57, was coupled to alpha toxin via a disulfide bridge between cysteine 37 and a thiol group previously incorporated onto leucine 1 of the alpha toxin (Kessler et al., Bioconjug. Chem., 1994, 5, 199-), since this residue is remote from the T epitopes of the toxin that are recognized by the H-2d haplotype (Lonetti et al., J. Immunol., 1990, 145, 4214-; Maillre et al., J. Immunol., 1993, 150, 5270). The derivative was called alpha37-57.

[0124] In order to study whether the peptide 37-57 confers on alpha toxin the ability to bind cells, various dilutions of alpha toxin or of alpha37-57 polypeptide were incubated in the presence, respectively, of splenocytes, of A20 cells or of dendritic cells. After incubation for 30 min at 4 C., a rabbit anti-alpha toxin polyclonal antibody and an F(ab)2 fragment of a goat anti-rabbit IgG polyclonal antibody coupled to fluorescein were added, and then the analysis by FACS was carried out.

[0125] FIG. 2 shows that the alpha37-57 polypeptide increases the fluorescence intensity of various cell types, whereas free alpha toxin (alpha) is not very effective, strongly suggesting that peptide 37-57 confers, on alpha toxin, the ability to bind cells. In order to reinforce the role of Tat portion 37-57 in cell binding, the cell binding of alpha37-57 was also examined in the presence of heparin, a soluble sulfated polysaccharide, representative of heparan sulfates and capable of binding Tat (Albini et al., Oncogene, 1996, 12, 289). The fluorescence intensity decreases when the splenocytes are preincubated in the presence of alpha37-57 and of an excess of heparin (FIG. 2A), demonstrating that the binding of alpha37-57 is mediated by the Tat portion.

EXAMPLE 2

Alpha Toxin is Presented More Effectively to Two Alpha-Toxin-Specific T-Cell Hybridomas when it is Covalently Coupled To Peptide 37-57

[0126] 1. Materials and Methods

[0127] 1.1 Isolation of B Lymphocytes and of Dendritic Cells from Splenocytes

[0128] The dendritic cells were isolated as described in example 1. The B lymphocytes were isolated using anti-CD19 microbeads according to a similar protocol.

[0129] 1.2 T-Cell Stimulation Tests

[0130] All the experiments were carried out using DCCM1 (Biological Industries) as synthetic culture medium. Serial dilutions of the various antigens were incubated in the microculture wells (Nunc) for 3 hours at 37 C. in the presence, either of A20 (510.sup.4/well), or of splenocytes (510.sup.5/well) or of DCs (310.sup.4/well). An alpha-toxin-specific T-cell hybridoma (T1C9: Maillre et al., Mol. Immunol., 1995, 32, 1073-; T1B2: Maillre et al., J. Immunol., 1993, 150, 5270-; 510.sup.4/well) was then added to the wells and the cells were then cultured for 24 h at 37 C. The level of T stimulation which is reflected by IL-2 secretion was evaluated by sampling the culture supernatants and measuring the presence of IL-2 using a cytotoxic T-cell (CTL) line of which the growth is dependent on this interleukin, as described in Gillis et al., Nature, 1977, 268, 154-156. The proliferation of the IL-2-dependent CTL line was evaluated by measuring the incorporation of tritiated methyl-thymidine (5 Ci/mmol). The data are expressed in cpm.

[0131] 2. Results

[0132] The presentation of alpha37-57 and of free alpha toxin to two specific T-cell hybridomas, called T1B2 and T1C9, which recognize respectively a thiol-dependent epitope (Maillre et al., J. Immunol., 1993, 150, 5270-) and a thiol-independent epitope (Maillre et al., Mol. Immunol., 1995, 32, 1073-) was evaluated using splenocytes as antigen-presenting cells (APCs). As shown in FIG. 3A, the amount of free alpha toxin required to stimulate T1B2 is similar in the presence and in the absence of equimolar amounts of free peptide 37-57, thereby indicating that this free Tat fragment has no influence on antigen presentation. On the other hand, 5-fold smaller amounts of alpha37-57 are sufficient to stimulate T-cell hybridomas as effectively as free alpha toxin, indicating that the increase in capacity for stimulating T cells is due to the covalent coupling to the Tat peptide. Two observations indicate that the stimulating effect is due to the targeting of the hybrid protein onto the HSPGs expressed at the surface of the APCs. The first observation was made using three different sulfated soluble polysaccharides incubated in excess (3 M) with serial dilutions of alpha and of alpha37-57. Neither chondroitin A nor chondroitin B, which are representatives of chondroitin sulfates and which bind Tat weakly (Rustani et al., J. Biol. Chem., 1997, 272, 11313-), are capable of influencing the stimulating effect (FIG. 3A). On the other hand, heparin, which is representative of heparan sulfates and is capable of inhibiting the binding of the alpha37-57 hybrid to splenocytes (FIG. 2A), is capable of decreasing the T-stimulating capacity of alpha37-57 at the level observed with free alpha toxin. The second observation was made using excess amounts of free peptide 37-57. In a first series of experiments carried out with a protocol similar to that used for the sulfated polysaccharides, free peptide 37-57 has no effect on the presentation of free alpha toxin. Indicating that it is not toxic at the concentration used (0.2 M) and that it weakly, or even not at all, inhibits the presentation of alpha37-57 (FIG. 3B). However, this weak inhibition capacity is linked to the protocol used, given that an important impairment of the T-stimulating capacity was observed when the cell-binding step was separated from that of processing and presentation (FIG. 3C). Thus, the stimulation of the T-cell hybridoma is decreased when the splenocytes are incubated at 4 C. with peptide 37-57 and alpha37-57 and they are subsequently washed to remove the unbound protein before the addition of the T1B2 cells. These observations were also made when the antigen presentation was studied by means of the T1C9 hybridoma (FIGS. 3D, 3E, 3F). Thus, the hybrid protein is 13 times more powerful than alpha toxin in terms of stimulating T1C9, and this phenomenon is affected in the presence of an excess of heparin (FIG. 3D) or of free peptide 37-57 (FIG. 3F). Consequently, the stimulating effect is not associated with a particular epitope. Since alpha toxin is a very stable protein (Thai et al., J. Biol. Chem., 2004, 279, 50527-) which absolutely requires processing in order to stimulate T cells (Maillre et al., J. Immunol., 1993, 150, 5270-), it was subsequently examined whether the stimulating effect was dependent on antigen processing. To do this, fixed splenocytes were used as APCs. In their presence, the two T-cell hybridomas are not stimulated by alpha toxin nor by alpha37-57 (FIGS. 4A and 4B). This absence of stimulation is not due to an impairment of the presentation capacity of the fixed splenocytes, given that they remain capable of presenting peptide 32-49 and peptide Ac24-36 to the T1B2 (FIG. 4A) and T1C9 (FIG. 4B) hybrodomas, respectively. These results therefore indicate that the increase in T-stimulating capacity is dependent on processing of the toxin by the APCs. Next, it was evaluated whether Tat fragment 37-57 can also increase the presentation of an antigen which does not require processing. When comparing the stimulating capacity of peptide Ac24-36 in its free form or its form coupled beforehand to Tat37-57, it was observed that Ac24-36 and Ac24-36-Tat37-57 do not significantly differ in terms of their ability to stimulate T1C9, in the presence of fixed (FIG. 4B) or live (FIG. 4C) splenocytes, indicating that the stimulating effect is not linked to an increase in presentation at the surface of the cells. All of these results indicate that the Tat37-57 fraction targets HSPGs and thus increases MHC class II-restricted presentation of a protein antigen which requires processing. A similar increase in T-stimulating capacity was observed when alpha toxin is coupled to the basic region of Tat.

[0133] Next, the lymphocytes and the DCs were purified from splenocytes in order to examine whether the stimulating effect is observed in the presence of these two different types of APCs. With the B lymphocytes, the stimulation of T1B2 requires an amount of alpha37-57 which is approximately 10 times less than that of free alpha toxin, or of free alpha toxin plus peptide 37-57 (FIG. 5A). With the DCs, the effect is more pronounced, since the T-cell stimulation requires 180 times less hybrid than free alpha toxin in the presence or absence of peptide 37-57 (FIG. 5B). Consequently, the stimulating effect is observed in the presence of these two types of APCs and is dependent on covalent coupling of the Tat fragment to the protein antigen.

EXAMPLE 3

APCs Bearing an Fc Receptor Type II (FcRII) More Efficiently Present an Immune Complex Containing an Antigen Covalently Coupled to Peptide 37-57

[0134] 1. Materials and Methods

[0135] Cells Used

[0136] See example 1.

[0137] The A20 B lymphoma lines lacking or expressing FcRII are described in Amigorena et al., Science, 1992, 256, 1808-.

[0138] Binding of Immune Complexes to Cells

[0139] Series of dilutions of alpha toxin or of alpha37-57 were incubated overnight at 4 C., in the presence of a rabbit anti-alpha toxin polyclonal antibody (10 nM or 25 nM). The A20 cells and A20 cells expressing FcRII were then added (10.sup.5 per well) and incubated for 30 min at 4 C. in PBS buffer supplemented with 0.5% bovine serum albumin (PBS/0.5% BSA). The mixtures were then washed and a goat anti-rabbit IgG antibody coupled to fluorescein was added. After incubation for 30 min at 4 C., the cells were washed and analyzed by FACS.

[0140] T-Cell Stimulation Tests

[0141] See example 2.

[0142] Serial dilutions of alpha or of alpha37-57 in the presence or absence of anti-alpha (12.5 nM) or of rabbit antibody not specific for the toxin (called rabbit IgG) were preincubated overnight at 4 C. A20 cells lacking or expressing FcRII (510.sup.4 cells per well) were added. After incubation for 3 hours at 37 C., the T1B2 and T1C9 hybridomas were respectively added in a proportion of 510.sup.4 cells per well. After culture for 24 hours at 37 C., the level of T stimulation which is reflected by IL-2 secretion was evaluated by sampling the culture supernatants and measuring the presence of IL-2 using a CTL line of which the growth is dependent on this interleukin. The proliferation of the IL-2-dependent CTL line was evaluated by measuring the incorporation of tritiated methyl thymidine (5 Ci/mmol).

[0143] 2. Results

[0144] Since HSPGs can act as coreceptors which modulate the meeting between extracellular proteins and their receptors by forming heparan sulfate (HS)/protein complexes (Park et al., J. Biol. Chem., 2000, 275, 29923-; Carey, D. J. Biochem. J., 1997(Pt1):1), alpha37-57 was used to study whether HSPGs can regulate receptor-mediated antigen presentation. The present study concerns presentation mediated by the Fc receptor type II (FcRII) given that this receptor binds immune complexes (ICs) and it influences antigen presentation (Sallusto et al., J. Exp. Med., 1994, 179, 1109; Amigorena et al., Science, 1992, 256, 1808-; Amigorena et al., J. Exp. Med., 1998, 187, 505-; Regnault et al., J. Exp. Med., 1999, 189, 371-). Consequently, an immune complex was used. This complex contains a rabbit anti-alpha toxin polyclonal antibody (anti-alpha) and the alpha toxin from Naja nigricollis which has been coupled beforehand to Tat fragment 37-57 capable of binding HSs (alpha37-57). This immune complex, called alpha37-57+anti-alpha, therefore has the ability to bind, firstly, HSs expressed at the surface of most cells, and, secondly, receptors which recognize the antibody Fc region and which are selectively expressed at the surface of APCs.

[0145] This complex was evaluated using, as APCs, the A20 B lymphoma line lacking or expressing the Fc receptor type II (FcRII; Amigorena et al., Science, 1992, 256, 1808-). Alpha37-57+anti-alpha was compared with a complex lacking the 37-57 region, called alpha+anti-alpha, in order to evaluate what is provided by the region capable of binding HSs. Two characteristics were studied: firstly, the capacity for binding two types of

[0146] A20 cells, secondly the capacity for stimulating two T hybridomas, called T1C9 and T1B2, which are specific for alpha toxin.

[0147] 2.1 Capacity for Binding A20 Cells Lacking or Expressing FcRII: for Low Antigen Concentrations, the Alpha37-57+Anti-Alpha and Alpha37-57+Anti-Alpha Complexes Bind the Fc Receptor Type II in a Similar Manner

[0148] The binding of the two complexes to A20 cells was evaluated (FIG. 6). As can be seen in FIG. 6A, the A20 cells which do not express FcRII are bound only by the alpha37-57+anti-alpha complex, thereby indicating that the interaction of the complex is mediated by the HS-binding 37-57 region. The scenario is different in the presence of the A20 cells expressing FcRII (FIG. 6B). Indeed, the alpha+anti-alpha complex binds to the cells in a manner similar to alpha37-57+anti-alpha for Ag concentrations less than or equal to 1 nM, thereby indicating that, for low antigen concentrations, the two complexes both bind the cells. These observations therefore indicate that the two complexes interact similarly with FcRII expressed at the surface of the A20 cells, thereby indicating that the presence of the HS ligand does not disrupt the interaction of the antibodies with FcRII. When the amount of antigen becomes greater than 1 nM, the interaction of alpha+anti-alpha decreases, thereby indicating that the antibodies bind less efficiently the FcRIIs expressed at the surface of the cells. On the other hand, the binding of alpha37-57+anti-alpha increases, thereby indicating that the interaction with the cells becomes mainly mediated by the 37-57 region.

[0149] 2.2 T-Simulating Capacity: the T-Stimulating Capacity of Alpha Toxin is Increased when the Protein is Coupled to 37-57 and Complexed with the Anti-Alpha Antibody

[0150] The T-stimulating capacity of the alpha+anti-alpha and alpha37-57+anti-alpha complexes was evaluated (FIG. 7). As can be seen in this figure, the A20 cells negative (FIG. 7A) and positive (FIG. 7C) for FcRII present alpha37-57 to the T1B2 hybridoma better than they present alpha to the T1B2 hybridoma, thereby indicating that the presence of the HS ligand increases presentation of the protein and its T-stimulating capacity. In the presence of A20 cells lacking FcRII, the presentation of the alpha37-57+anti-alpha immune complex remains similar to that of free alpha37-57 and the presentation of the alpha+anti-alpha complex remains similar to that of alpha (FIGS. 7A and 7B), thereby indicating that the T-stimulating capacity of the antigen (Ag) is not modified when it is associated with the antibody and presented by these APCs. The result is similar for the alpha+anti-alpha complex in the presence of A20 cells expressing FcRII (FIGS. 7C and 7D), thereby indicating that the T-stimulating capacity of the wild-type alpha toxin is not modified when it is associated with the antibody which interacts with the Fc receptor expressed at the surface of the APCs. The result is, on the other hand, substantially different for the alpha37-57+anti-alpha complex. Indeed, the T-stimulating capacity of alpha37-57 is greatly increased when it interacts beforehand with anti-alpha (FIGS. 7C and 7D). Thus, to stimulate T1B2 with alpha37-57+anti-alpha, amounts of Ag that are respectively 130 times lower than free alpha37-57 (alpha+rabbit IgG), and more than 1000 times lower than free alpha (alpha+rabbit IgG) or complexed with anti-alpha (alpha+anti-alpha), are sufficient. The same phenomenon is observed for the T1C9 hybridoma, thereby indicating that the increase in T-stimulating capacity is provided for all the T epitopes of alpha toxin. These results therefore indicate that the targeting of FcRII makes it possible to increase the T-stimulating capacity of alpha toxin complexed with the antibody only in the case where it is coupled beforehand to the 37-57 fragment. Furthermore, the increase in the T-stimulating capacity disappears when the complex is incubated in the presence of an excess of heparin (see alpha37-57+anti-alpha+heparin, FIG. 7D), thereby indicating that the 37-57 fragment contributes to the effect.

[0151] All these observations therefore indicate that the T-stimulating capacity of alpha toxin is increased when it is associated with an HS ligand and with an Fc receptor II ligand, and that this phenomenon is brought about by the targeting of FcRII in combination with the capacity for binding HSs.

EXAMPLE 4

The Diphtheria Toxin Domain Capable of Binding to the Cell Receptor is Capable of Binding Heparan Sulfates

[0152] Fusion proteins incorporating, firstly, an antigenic protein and, secondly, a ZZ double domain derived from Staphylococcus aureus protein A have been previously constructed (Lonetti et al., J. Immunol, 1998, 160, 3820-). ZZ can bind to immunoglobulins in a manner similar to protein A and it has been shown that fusion proteins of ZZAg type have an increased T-stimulating capacity. It has been shown that this increase is mediated by the ZZ region which binds APCs bearing immunoglobulins and thus increases the amount of Ag incorporated into the cells. This fusion protein system therefore enables surface-immunoglobulin-mediated Ag targeting. On the basis of this system, another fusion protein, called ZZDTR in the publication by Lobeck et al. (Infection and Immunity, 1998, 66, 418-423), was constructed. This fusion protein, now called ZZDTR-BD, contains the diphtheria toxin domain (DTR-BD) capable of binding the cell receptor which is the target of this toxin. This receptor is not expressed at the surface of APCs. It was evaluated whether the ZZDTR-BD fusion protein was capable of binding heparan sulfates.

[0153] The binding of ZZDTR-BD to heparan sulfates was evaluated in two steps.

[0154] In a first step, the binding of ZZDTR-BD to microtitration plates bearing heparin was evaluated. For this, the microtitration plates were pre-adsorbed with heparin-albumin (1 g/100 l/well) and then saturated with bovine serum albumin (200 l/well at 0.3%). The plates were then washed and a series of dilutions of ZZDTR-BD was deposited in the wells and incubated overnight at 4 C. The plates were washed and a rabbit polyclonal antibody was added. After 30 minutes at ambient temperature, the plates were washed and an anti-rabbit antibody goat antibody was added. 30 minutes later, the plates were washed, a substrate was added, and the coloration was measured at 414 nm after incubation for 30 minutes.

[0155] In a second step, it was evaluated whether the binding of ZZDTR-BD to the plates adsorbed with heparin-albumin can be modified when the fusion protein is incubated in the presence of solutions containing either a heparin fragment having a molecular weight of 6000 Da (Hep6000), or heparan sulfate or ZZ. For this, a fixed concentration of ZZDTR-BD was deposited in the presence of series of dilutions of Hep6000, of heparan sulfate (HS) or of ZZ. After incubation overnight at 4 C., the plates were washed and a rabbit polyclonal antibody was added. After 30 minutes at ambient temperature, the plates were washed and an anti-rabbit antibody goat antibody was added. 30 minutes later, the plates were washed, a substrate was added and the coloration was measured at 414 nm after incubation for 30 minutes.

[0156] As can be seen in FIG. 8A, ZZDTR-BD binds the plates adsorbed with heparin-albumin, thereby indicating that the fusion protein is capable of interacting with this sulfated sugar which is similar to heparan sulfates. The interaction of ZZDTR-BD with the plates is inhibited by the solutions of Hep6000 and of HS, thereby indicating that the fusion protein interacts with HSs and heparin using a common site. However, this site is not located in the ZZ region, since the free ZZ double domain does not modify the binding of ZZDTR-BD to the plates. It can therefore be concluded therefrom that the HS-binding capacity is mediated by the DTR-BD domain of ZZDTR-BD.

EXAMPLE 5

The ZZDTR-BD Fusion Protein has a T-Stimulating Capacity which is Influenced by its Ability to Bind HSs and Immunoglobulins Expressed at the Surface of APCs

[0157] Since the fusion protein is capable of binding HSs and immunoglobulins expressed at the surface of APCs, it was evaluated whether its T-stimulating capacity is dependent on these two characteristics. For this, the presentation of ZZDTR-BD to a diphtheria-toxin-specific T hybridoma was studied in vitro. The hybridoma, called T4B6, recognizes the 92-106 T epitope located in the DTR-BD domain of diphtheria toxin.

[0158] To evaluate the T-stimulating capacity, series of dilutions of ZZDTR-BD in the presence or absence of an excess of Hep6000 (3 m final concentration), of HS (3 M final concentration) or of rabbit IgG (0.8 M final concentration) were incubated for 3 h at 4 C. in cell culture plates. The A20 line, which is a B lymphoma that has surface immunoglobulins (510.sup.4 cells/well, FIG. 9) and has presentation capacities, was then added. After incubation for 1 hour at 37 C., the T4B6 hybridoma is added and the mixture is incubated for 24 hours at 37 C. The stimulation of the hybridoma is then evaluated by measuring the presence of IL-2 in the supernatants. For this, the supernatants are sampled and incubated in the presence of a CTL line of which the growth is dependent on the presence of IL-2. The cell growth is evaluated by measuring the incorporation of tritiated thymidine into the IL-2-dependent CTL line.

[0159] As can be seen in FIG. 9, in the presence of the A20 B lymphoma, ZZDTR-BD is capable of stimulating the diphtheria-toxin-specific T4B6 hybridoma. The stimulation is weaker when ZZDTR-BD is incubated beforehand in the presence of an excess of Hep6000, thereby demonstrating that the T-stimulating capacity of the protein is influenced by its ability to bind heparin. Since it has been previously shown that the ZZAg fusion protein system enables targeting of Ags to immunoglobulins expressed at the surface of APCs (Lonetti et al., J. Immunol., 1998, 160, 3820-), these results demonstrate that ZZDTR-BD has a T-stimulating capacity which is increased by its capacity for binding immunoglobulines and compounds of the heparan sulfate family.

[0160] In order to determine the population of splenocytes bound by ZZDTR-BD, a fixed amount of ZZDTR-BD (10 nM) was incubated in the presence of splenocytes and of three antibodies respectively specific for CD4+ T lymophocytes (anti-CD4), for CD8+ T lymphocytes (anti-CD8) and for B lymphocytes (anti-CD19). These three antibodies are labeled with phycoerythrin. After 30 minutes at 4 C., the cells were washed and incubated in the presence of a rabbit polyclonal antibody which can interact with the ZZ region of the fusion protein. 30 minutes later, an anti-rabbit antibody polyclonal antibody coupled to fluorescein was added. After incubation for 30 minutes at 4 C., the cells were washed and analyzed by flow cytometry. As can be seen in FIG. 10, the splenocytes contain approximately 14% CD8+ T lymphocytes, and 22% of these cells are bound by ZZDTR-BD. The splenocytes contain approximately 32% of CD4+ T lymphocytes, and 15.8% of these cells are bound by ZZDTR-BD. The splenocytes contain approximately 52% B lymphocytes, and 51.5% of these cells are bound by ZZDTR-BD. Since the B cells express surface antibodies and the CD4+ T and CD8+ T cells are devoid of antibodies, it may be concluded that ZZDTR-BD targets the surface antibodies and thus preferentially binds to B lymphocytes, which are APCs.

EXAMPLE 6

Increase in the T-Stimulating Capacity of the ZZDTR-BD Fusion Protein Associated Beforehand with Antibodies Specific for APC Surface Determinants

[0161] Protein A and the Z domain which is derived therefrom can bind to the Fc region of various classes of antibodies. It has previously been shown that this characteristic can be exploited to form complexes between Abs and fusion proteins of ZZAg type (Lonetti et al., J. Immunol., 1998, 160, 3820-). It has also been shown that an Ab/ZZAg complex containing an Ab specific for an APC surface determinant is presented better to T cells than free ZZAg or ZZAg complexed with a nonspecific antibody, and that it is capable of inducing, in animals, an immune response in the absence of adjuvant. The same principle of formation of complexes between ZZDTR and antibodies was therefore used to evaluate whether the fusion protein which has the capacity for binding HSs can thus experience an increase in its T-stimulating capacity when it interacts with antibodies capable of targeting APC surface determinants.

[0162] In order to carry out this study, three mouse monoclonal antibodies, of IgG2a subclass, which all have the capacity for binding ZZ, were selected. The first, called M2-3, described in Trmeau et al., FEBS Lett., 1986, 208, 236-240, is used as a control since it does not bind APCs. The other two antibodies are specific for APC surface determinants. The first, called 14-4-4S, recognizes the MHC class II molecule I-E.sup.d (Ozato et al., J. Immunol., 1980, 122, 549). The second, called 10-1.D.2, recognizes the Lyb-2.1 antigen which is expressed at the surface of B lymphocytes (number ATCC TIB-165).

[0163] In order to evaluate the T-stimulating capacity of ZZ-DTR in the presence of antibodies, ZZDTR is diluted in the presence or absence of 14-4-4S, 10-1.D.2 and M2-3, respectively. The antibodies are incubated at final concentration of 50 nM. Following these dilutions, the compounds are incubated for 3 h at 4 C. and then APCs are added. The APCs are, firstly, A20 cells (510.sup.4/50 l/well) and secondly, BALB/c mouse splenocytes (510.sup.5/50 l/well). After 2 h at 37 C., the T4B6 hybridoma (510.sup.4/50 l/well) is added and the mixture is incubated for 24 h at 37 C. The stimulation of the T4B6 cell is then evaluated by measuring its interleukin-2 secretion. This measurement is carried out on the culture supernatants which are incubated on a CTL line of which the growth is dependent on the presence of IL-2.

[0164] As can be seen in FIG. 11, in the presence of A20 cells, ZZDTR-BD stimulates T4B6. Amounts of ZZDTR-BD which are approximately 13.4 times and 3.8 times lower are required to reach the same level of stimulation when the fusion protein is respectively complexed with anti-MHC (anti-MHC/ZZDTR-BD) and with anti-Lyb-2.1 (anti-Lyb-2.1/ZZDTR-BD). The increase in T-stimulating capacity is not due to the fact that anti-MHC and anti-Lyb-2.1 are immunoglobulins of subclass 2a, since the effect is not found when ZZDTR-BD is complexed with an IgG2a not specific for an APC surface determinant (IgG2a/ZZDTR-BD). These results therefore demonstrate that the T-stimulating capacity of ZZDTR-BD is increased by targeting the MHC class II or Lyb-2.1 molecules which are expressed selectively at the surface of the A20 cells used as APCs.

[0165] In the presence of splenocytes, ZZDTR-BD stimulates T4B6. Amounts of ZZDTR-BD that are approximately 5.2 times lower are required to reach the same level of stimulation when the fusion protein is complexed with anti-MHC (anti-MHC/ZZDTR-BD). On the other hand, when ZZDTR-BD is complexed with the anti-Lyb-2.1 antibody (anti-Lyb-2.1/ZZDTR-BD), the T-stimulating capacity is slightly decreased. The increase in the T-stimulating capacity mediated by the anti-MHC and also the decrease in T-stimulating capacity mediated by the anti-Lyb-2.1 are not due to the fact that these immunoglobulins are of subclass 2a, since the effect is not found when ZZDTR-BD is complexed with an IgG2a not specific for an APC surface determinant (IgG2a/ZZDTR-BD). These results therefore demonstrate that the T-stimulating capacity of ZZDTR-BD is increased by targeting the MHC class II molecules that are expressed selectively at the surface of splenocytes.

EXAMPLE 7

Two Different Ags which are Respectively Associated with an HS Ligand with an Antibody Capable of Binding an APC Surface Protein Bind, Preferentially and in an Increased Manner, APCs Expressing this Surface Protein

[0166] Two splenocyte-binding experiments were carried out in order to evaluate whether the Ags which are associated with an HS ligand and with an antibody specific for an APC surface protein are capable of preferentially targeting APCs.

[0167] 7.1 Splenocyte-Binding Study Using the Protein System Described in Example 3

[0168] 7.1.1 Study of Binding in the Presence or Absence of the Heparan Sulfate Ligand

[0169] The wild-type alpha toxin and alpha37-57 toxin were preincubated in the presence or absence of the anti-alpha toxin rabbit polyclonal antibody. The mixtures were then added to splenocytes (510.sup.5/100 l/well) and incubated for 30 minutes at 4 C. The cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. After 30 minutes at 4 C., the cells were washed and the binding of the anti-alpha toxin/alpha toxin and anti-alpha toxin/alpha37-57 toxin complexes was evaluated by flow cytometry.

[0170] As can be seen in FIG. 12, the cells incubated in the presence of the anti-rabbit antibody polyclonal antibody coupled to fluorescein exhibit a low level of fluorescence which results in a geometric mean of 3.6. The splenocytes preincubated with the alpha/anti-alpha complex exhibit a medium level of fluorescence which is increased by a factor of 2.75 (geometric mean of 9.9). Finally, the cells preincubated with the alpha37-57/anti-alpha complex exhibit a mixed profile. A first cell population exhibits weak labeling (geometric mean of 4.44) which is not very different than that of the cells incubated in the absence of complex. On the other hand, a second population fluoresces much more strongly (geometric mean of 232.1, intensity multiplied by 64.5-fold). These observations therefore indicate that the alpha37-57/anti-alpha complex has the particularity of binding in increased amount to a subpopulation of splenocytes.

[0171] 7.1.2 Determination of the Splenocyte Subpopulations Bound by the Anti-Alpha Toxin/alpha37-57 Toxin Complex

[0172] In order to determine the splenocyte population bound by the anti-alpha toxin/alpha37-57 toxin complex, a fixed amount of this complex (10 nM) was incubated in the presence or absence of splenocytes and of three antibodies respectively specific for CD4+ T lymphocytes (anti-CD4), for CD8+ T lymphocytes (anti-CD8) and for B lymphocytes (anti-CD19). These three antibodies are labeled with phycoerythrin. After 30 minutes at 4 C., the cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. 30 minutes later, the cells were washed and analyzed by flow cytometry.

[0173] As can be seen in FIG. 13, the splenocytes contain approximately 14% CD8+ T lymphocytes, and 6.4% of these cells are bound by the alpha37-57+anti-alpha complex. The splenocytes contain approximately 26% CD4+ T lymphocytes, and 5.5% of these cells are bound by the complex. The splenocytes contain approximately 49% of B lymphocytes and virtually all these cells are bound by the complex. Since B cells bear Fc receptors, whereas CD4+ T and CD8+ T cells are devoid of Fc receptors, it can therefore be concluded therefrom that the alpha37-57/anti-alpha complex binds preferentially and in an increased manner to B lymphocytes, which are Fc receptor-bearing APCs.

[0174] 7.2 Study of Splenocyte-Binding Using the 14-4-4S/ZZDTR-BD Protein System Described in Example 5

[0175] In order to determine the splenocyte population bound by the 14-4-45/ZZDTR-BD complex, a fixed amount of this complex (10 nM) was incubated in the presence or absence of splenocytes and four antibodies respectively specific for CD4+ T lymphocytes, for CD8+ T lymphocytes, for B lymphocytes and for MHC class II I-Ad and I-Ed molecules. These antibodies are labeled with phycoerythrin. After 30 minutes at 4 C., the cells were washed and incubated in the presence of an anti-rabbit antibody polyclonal antibody coupled to fluorescein. 30 minutes later, the cells were washed and analyzed by flow cytometry.

[0176] As can be seen in FIG. 14, ZZDTR-BD interacts weakly with the CD8+ T lymphocytes (approximately 12.4% of the CD8Ts are labeled) and this is also the case for the 14-4-4/ZZDTR-BD complex (approximately 13.3% of the CD8Ts are labeled). ZZDTR-BD interacts weakly with the CD4+ T lymphocytes (approximately 11.7% of the CD8Ts are labeled) and this is also the case for the 14-4-4/ZZDTR-BD complex (approximately 11% of the CD8Ts are labeled). ZZDTR-BD interacts more strongly with the B lymphocytes (approximately 50% of the B lymphocytes are labeled) and this is also the case for the 14-4-4/ZZDTR-BD complex (approximately 68.5% of the CD8Ts are labeled). The targeting antibody therefore makes it possible to increase by 37% the number of B cells bound by ZZDTR-BD. As it happens, B cells are APCs bearing MHC class II molecules. The 14-4-4S/ZZDTR-BD molecular complex therefore preferentially binds to APCs expressing MHC class II molecules.

EXAMPLE 8

A Compound Having the Capacity to Target HSs and a Molecule Expressed Specifically at the Surface of APCs (14-4-4S/ZZDTR-BD) is Capable of Inducing an Immune Response that is Greater than that Induced by a Compound which has the Capacity to Target a Broader Variety of Cells (M2-3/ZZDTR-BD)

[0177] In order to evaluate whether the targeting of HSs and of a receptor expressed specifically at the surface of APCs is capable of inducing an increased immune response in vivo, the immunogenicity of ZZDTR-BD when it is complexed with the 14-4-4S antibody, which targets APCs expressing class II molecules, and when it is complexed with a control antibody of the same isotype (IgG2a), was compared.

[0178] Before injection, the ZZDTR-BD fusion protein and the two Abs were diluted in HBSS medium. ZZDTR-BD was then incubated for one hour at ambient temperature in the presence of equimolar amounts of the anti-MHC or of the control IgG2a. Two groups of four mice were then injected in the absence of adjuvant (0.01 nmol of complex/mouse/100 l) with the anti-MHC/ZZDTR-BD complex or with the IgG2a/ZZDTR-BD complex, respectively. Forty-five days after the injection, the blood of the animals was taken and the sera were pooled. The presence of anti-diphtheria toxin antibodies was then evaluated by immunoenzymatic assay using microtitration plates which were adsorbed beforehand with a non-toxic mutant of diphtheria toxin, called CRM197, described by Uchida et al., Science, 1972, 175, 901-903 (0.1 pg of CRM197/well/100 l PBS). In order to carry out this evaluation, the two pooled sera were respectively diluted and incubated overnight at 4 C. in the microtitration plates. The plates were subsequently washed and then incubated in the presence of a goat anti-mouse IgG antibody coupled to peroxidase. After 30 minutes, the plates were washed, a substrate (ABTS) was added, and the coloration was measured at 414 nm after incubation for a further 30 minutes. On the basis of these measurements, the antibody titers are defined as the serum dilution resulting in an OD of 0.6.

[0179] As can be seen in FIG. 15, a titer of 1/3400 is measured for the serum derived from the immunization with the IgG2a/ZZDTR-BD complex, thereby indicating that this complex is capable of inducing an anti-diphtheria toxin antibody response in the absence of adjuvant. This titer is, however, lower than that measured for the serum derived from the immunization with the anti-MHC/ZZDTR-BD complex (1/12 000), thereby indicating that the targeting of the MHC class II molecules makes it possible to increase the humoral immune response induced against the complex.

EXAMPLE 9

A Compound Having the Capacity to Target HSs, a Molecule Expressed Specifically at the Surface of APCs and a CD8+ T Epitope is Capable of Inducing an Immune Response which is Greater than a Compound which Targets Only HSs

[0180] In order to evaluate whether the double targeting also makes it possible to increase the capacity for stimulating cytotoxic CD8+ T cells, several fusion proteins were constructed. The first fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope), a CD4+ T epitope of sequence SYKKVWRDHRGTI (SEQ ID NO:13), and the Tat22-57.sub.C(22-37)S fragment which contains the Tat region capable of binding to heparan sulfates. The second fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope), and a CD4+ T epitope of sequence SYKKVWRDHRGTI (SEQ ID NO: 13). The third fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), ovalbumin, and the Tat22-57.sub.C(22-37)S fragment which contains the Tat region capable of binding to heparan sulfates. The fourth fusion protein contains the immunoglobulin-binding ZZ double domain and ovalbumin. Ovalbumin is used as a control. These fusion proteins were used free or complexed either with the AF6-120.1 monoclonal antibody specific for the MHC class II I-Ab molecule (#553549; Becton-Dickinson Biosciences), or with a monoclonal antibody which is nonspecific but of the same subclass (IgG2a) as AF6-120.1, or with a rabbit anti-mouse IgG polyclonal antibody, or with a nonspecific rabbit antibody. The formation of the complexes was carried out under conditions identical to those described in the sixth example.

[0181] In order to evaluate the stimulating capacity of the various compounds, the B3Z hybridoma which recognizes the T epitope of sequence SIINFEKL in association with class I molecules of type I-Ab was used. Alternatively, splenocytes derived from OT1 mice, which contain CD8+ T cells which also recognize the SIINFEKL epitope, were used. For evaluating the stimulation of the B3Z hybridoma, the fusion proteins (antigens) were incubated in the presence or absence of each of the four antibodies described previously. The mixtures were added to the APCs. The APCs are, firstly, a dendritic cell line (5104/50 l/well), and secondly, C57B1/6 mouse splenocytes (5105/50 l/well).

[0182] To evaluate the stimulation of the B3Z hybridoma, the antigens (Ags) were incubated for 2 h at 37 C. in the presence of APCs, the B3Z cells (510.sup.4/50 l/well) were added and the mixtures were incubated for 24 h at 37 C. The stimulation of the B3Z cell is then evaluated either by measuring interleukin-2 secretion, or by measuring the expression of the Lac Z gene which encodes -galactosidase. The IL-2 secretion measurement is carried out using the culture supernatants which are subsequently incubated on a CTL line of which the growth is dependent on the presence of IL-2. The measurement of the -galactosidase activity is carried out using chlorophenolred--D-galactopyranoside as substrate.

[0183] To evaluate the stimulation of the OT1 cells, the Ags were incubated for 5 h at 37 C. in the presence of JAWS II dendritic cells. The cells were then fixed with glutaraldehyde and then OT1 mouse splenocytes (510.sup.5/50 l/well) were added and the mixtures were incubated for 3 days at 37 C. Tritiated thymidine was then added (1 Ci/well). After incubation for 18 hours at 37 C., the cells were removed and the radioactivity incorporated into the cells was measured in order to evaluate the OT1 cell proliferation.

EXAMPLE 10

A Compound Having the Capacity to Target HSs, a Molecule Expressed Specifically at the Surface of APCs and an Ag is Capable of Inducing a Greater Immune Response In Vivo than a Compound which Contains the Same Ag but Targets Only the Molecule Expressed Specifically at the Surface of APCs

[0184] In order to evaluate whether the double targeting also makes it possible to increase the immune response in vivo, several fusion proteins were constructed. The first fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope), a CD4+ T epitope of sequence SYKKVWRDHRGTI (SEQ ID NO: 13), and the Tat22-57.sub.C(22-37)S fragment which contains the Tat region capable of binding to heparan sulfates. The second fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope), and a CD4+ T epitope of sequence SYKKVWRDHRGTI (SEQ ID NO: 13). The third fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3), ovalbumin, and the Tat22-57.sub.C(22-37)S fragment which contains the Tat region capable of binding to heparan sulfates. The fourth fusion protein contains the immunoglobulin-binding ZZ double domain (SEQ ID NO: 3) and ovalbumin. Ovalbumin is used as a control. These fusion proteins were used free or complexed either with the AF6-120.1 monoclonal antibody specific for the MHC class II I-Ab molecule (#553549; Becton-Dickinson Biosciences), or with a monoclonal antibody which is nonspecific but of the same subclass (IgG2a) as AF6-120.1, or with a rabbit anti-mouse IgG polyclonal antibody, or with a nonspecific rabbit antibody. The formation of the complexes was carried out under conditions identical to those described in the sixth example.

[0185] In order to evaluate the immune response in animals, the various mixtures were injected, in the presence or absence of adjuvant, into mice (6 animals per group). Fourteen days after the second immunization, a blood sample was taken from the animals in order to evaluate the humoral response. The animals were then sacrificed and their spleen was removed in order to evaluate the cellular response. The humoral response was evaluated by immunoenzymatic assay of the anti-ovalbumin antibodies. In order to evaluate the cellular response, the presence of specific T lymphocytes was determined by counting cells secreting interferon gamma or IL-4 using an ELISPOT assay according to the protocol described in Turbant et al., Vaccine, 2009, 27, 5349-56.

EXAMPLE 11

A Compound Having the Capacity to Target HCs and a Molecule Expressed Specifically at the Surface of APCs is Capable of Inducing a Greater Cytotoxic Immune Response than a Compound which Targets Only HSs or Only the Molecule Expressed Specifically at the Surface of APCs

[0186] In order to evaluate whether the double targeting makes it possible to increase the capacity for stimulating cytotoxic CD8+ T cells, two chimeric proteins were chemically synthesized. The first protein, called Tat47-57-SIINFEKL-alpha, contains the sequence of the alpha toxin from Naja nigricollis (SEQ ID NO: 9), a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope), and the Tat47-57 fragment (SEQ ID NO: 14) which contains the Tat region capable of binding to heparan sulfates. The second protein, called SIINFEKL-alpha, contains the sequence of the alpha toxin from Naja nigricollis (SEQ ID NO: 9), and a CD8+ T epitope of sequence SIINFEKL (SEQ ID NO: 10) which is derived from ovalbumin (the flanking sequences LEQLE (SEQ ID NO: 11) and TEWTS (SEQ ID NO: 12) are respectively inserted at the N-terminal and C-terminal of this epitope). These two proteins were used free or complexed with the rabbit anti-alpha toxic polyclonal antibody, called anti-alpha, described in example 3. The formation of the complexes was carried out by incubation overnight at 4 C.

[0187] In order to evaluate the stimulating capacity of the two compounds, OT1 mouse splenocytes were used. These splenocytes contain CD8+ T lymphocytes which recognize the T epitope of sequence SIINFEKL, in association with class I molecules of type I-A.sup.b. The two proteins (final dilution 1 M) complexed or not complexed with the antibody (50 nM) were respectively incubated in cell culture plates in the presence of a dendritic cell line used as APCs. This line, called JAWS II, was incubated in a proportion of 310.sup.4/50 l/well. After 5 h at 37 C., the plates were washed and the cells were fixed with glutaraldehyde. After fixing, OT1 mouse splenocytes were added (910.sup.4/50 l/well). After incubation for 4 days at 37 C., a solution of tritiated thymidine was added (1 Ci/25 l/well) and the plates were incubated for 18 hours at 37 C. The cells were then removed and the radioactivity was measured in order to evaluate cell proliferation. As can be seen in FIG. 16, the free SIINFEKL-alpha compound is capable of inducing CD8+ T-cell proliferation. In the presence of a fixed concentration of anti-alpha (50 nM final concentration), SIINFEKL-alpha stimulates OT1 cells more strongly, thereby indicating that the Fc-receptor-mediated targeting also makes it possible to increase cross-presentation, as had been observed previously for other antigens (Regnault, J. Exp. Med., 1999, 371-380). An increase in the stimulation of OT1 cells is also observed for the Tat47-57-SIINFEKL-alpha compound, thereby indicating that the targeting mediated by binding to HSs makes it possible to increase cross-presentation. The greatest increase in OT1 cell stimulation is, however, observed when Tat47-57-SIINFEKL-alpha is complexed with the anti-alpha antibody, thereby indicating that the targeting mediated jointly by HSs and Fc receptors provides a synergistic effect on cross-presentation of the antigenic protein.

EXAMPLE 12

A Compound Having the Capacity to Target HSs and a Molecule Expressed Specifically at the Surface of APCs Increases More Strongly the Expression of the Costimulatory Molecules of CD80 and CD86 than Compounds which Target Only HSs or Only the Molecule Expressed Specifically at the Surface of APCs

[0188] Antigen-presenting capacity represents an essential aspect in setting up an adaptive immune response. However, the setting up of this response also requires the activation of antigen-presenting cells (APCs). For this reason, the capacity of the molecular targeting complexes to activate APCs was evaluated. In order to evaluate this aspect, the ability of the molecular targeting complex described in example 11 to increase the expression of the costimulatory molecules CD80 and CD86 by JAWS II APCs, described previously, was analyzed. The chimeric proteins and antibodies used to carry out this evaluation are those described in example 11. The two proteins were used free or complexed with the rabbit anti-alpha toxin polyclonal antibody. The formation of the complexes was carried out by incubation overnight at 4 C. The wild-type alpha toxin was also used, as a control.

[0189] The two proteins (1 M final concentration), complexed or not complexed with the antibody (25 nM final concentration), were respectively diluted and incubated in cell culture plates in the presence of the JAWS II line (510.sup.5/100 l/well) for 24 hours at 37 C. The cells were then washed and incubated for 30 minutes at 4 C. in the presence of anti-CD80 antibody labeled with fluorescein and anti-CD86 antibody labeled with fluorescein, respectively. Finally, the antibody binding to the cells was analyzed by flow cytometry. As can be seen in FIG. 17A, when the cells were incubated in the presence of SIINFEKL-alpha, the CD80 molecule is expressed at the surface of the APCs. The expression of the CD80 molecule is increased when the JAWS II were incubated with Tat47-57-SIINFEKL-alpha or with the SIINFEKL-alpha/anti-alpha complex. However, the expression of CD80 is more strongly increased for the Tat47-57-SIINFEKL-alpha/anti-alpha complex, thereby indicating that the molecular targeting complex provides a synergistic effect. The same behavior is observed for the CD86 molecule (cf. FIG. 17B). Thus, when the cells were incubated in the presence of SIINFEKL-alpha, the CD86 molecule is expressed at the surface of the APCs. The expression of the CD86 molecule is increased when the JAWS II were incubated with Tat47-57-SIINFEKL-alpha or with the SIINFEKL-alpha/anti-alpha complex. However, the expression of CD86 is more strongly increased for the Tat47-57-SIINFEKL-alpha/anti-alpha complex, thereby indicating that the molecular targeting complex provides a synergistic effect.

[0190] This example, which demonstrates the synergistic effect on the expression of the costimulatory molecules CD80 and CD86, therefore indicates that the molecular targeting complex brings about activation of Ag-presenting cells.