Multiple malaria pre-erythrocytic antigens and their use in the elicitation of a protective immune response in a host

Abstract

The invention relates to Plasmodium antigenic polypeptides identified through the use of a specifically devised functional immunization screening assay. In particular, the invention relates to antigenic polypeptides of malaria parasites wherein said antigenic polypeptides that exhibit a protective effect, especially that of eliciting a protective immune response in a host against challenge by Plasmodium sporozoites. The invention relates to a combination of compounds, comprising at least 2 distinct active ingredients wherein each active ingredient consists of an antigenic polypeptide of a Plasmodium parasite, a polynucleotide encoding the antigenic polypeptide, or a vector, in particular a viral vector, especially a lentiviral vector, expressing such antigenic polypeptide of a Plasmodium parasite, wherein one antigenic polypeptide is the circumsporozoite protein (CSP) or a polypeptidic derivative thereof and another antigenic polypeptide is either protein Ag40 (11-09) or protein Ag45 (11-10).

Claims

1. A polynucleotide which is a mammalian codon-optimized synthetic nucleic acid encoding a pre-erythrocyte stage antigen of a Plasmodium parasite, wherein said polynucleotide is selected from the group of: a. SEQ ID No. 10 for CSP of P. berghei, SEQ ID No. 19 for TRAP of P. berghei, SEQ ID No.28 for ICP of P. berghei, SEQ ID No.37 for Falcilysin of P. berghei, SEQ ID No.57 for GPI-anchored protein P113 of P. berghei, SEQ ID No.48 for pore-forming like protein SPECT2 of P. berghei, SEQ ID No.66 for protein Ag40 11-09 of P. berghei, and SEQ ID No.75 for protein Ag45 11-10 of P. berghei, or, b. SEQ ID No.13 for CSP of P. falciparum, SEQ ID No.22 for TRAP of P. falciparum, SEQ ID No.31 for ICP of P. falciparum, SEQ ID No.40 for Falcilysin of P. falciparum, SEQ ID No.51 for pore-forming like protein SPECT2 of P. falciparum, SEQ ID No.60 for GPI-anchored protein P113 of P. falciparum, SEQ ID No.69 for protein 11-09 of P. falciparum, and SEQ ID No.78 for protein 11-10 of P. falciparum or, c. SEQ ID No.16 for CSP of P. vivax, SEQ ID No.25 for TRAP of P. vivax, SEQ ID No.34 for ICP of P. vivax, SEQ ID No.43 for Falcilysin of P. vivax, SEQ ID No.54 for pore-forming like protein SPECT2 of P. vivax, SEQ ID No.63 for GPI-anchored protein P113 of P. vivax, SEQ ID No.72 for protein 11-09 of P. vivax, and SEQ ID No.81 for protein 11-10 of P. vivax.

2. A method comprising administering to a subject a collection of lentiviral vectors comprising the polynucleotides according to claim 1 provided the collection of lentiviral vectors enables the expression of all antigenic polypeptides encoded by the polynucleotides of one subgroup selected among a/, b/, and c/.

3. A combination of compounds, comprising at least 2 distinct active ingredients, wherein each active ingredient comprises a polynucleotide encoding the antigenic polypeptide of a Plasmodium parasite, wherein the polynucleotides are mammalian codon-optimized synthetic nucleic acids selected from the group consisting of: a. SEQ ID No. 10 for CSP of P. berghei, SEQ ID No.19 for TRAP of P. berghei, SEQ ID No.28 for ICP of P. berghei, SEQ ID No.37 and 46 for Falcilysin of P. berghei, SEQ ID No.57 for GPI-anchored protein P113 of P. berghei, SEQ ID No.48 for pore-forming like protein SPECT2 of P. berghei, SEQ ID No.66 for protein Ag40 11-09 of P. berghei, and SEQ ID No.75 for protein Ag45 11-10 of P. berghei, b. SEQ ID No.13 for CSP of P. falciparum, SEQ ID No.22 for TRAP of P. falciparum, SEQ ID No.31 for ICP of P. falciparum, SEQ ID No.40 for Falcilysin of P. falciparum, SEQ ID No.51 for pore-forming like protein SPECT2 of P. falciparum, SEQ ID No.60 for GPI-anchored protein P113 of P. falciparum, SEQ ID No.69 for protein Ag40 11-09 of P. falciparum, and SEQ ID No.78 for protein Ag40 11-10 of P. falciparum, and C. SEQ ID No.16 for CSP of P. vivax, SEQ ID No.25 for TRAP of P. vivax, SEQ ID No.34 for ICP of P. vivax, SEQ ID No.43 for Falcilysin of P. vivax, SEQ ID No.54 for pore-forming like protein SPECT2 of P. vivax, SEQ ID No.63 for GPI-anchored protein P113 of P. vivax, SEQ ID No.72 for protein Ag40 11-09 of P. vivax, and SEQ ID No.81 for protein Ag40 11-10 of P. vivax.

4. The combination of compounds according to claim 3, wherein one of the polynucleotides encodes the circumsporozoite protein (CSP) and one of the polynucleotides encodes protein Ag40 (11-09).

5. The combination of compounds according to claim 3, wherein one of the polynucleotides encodes the circumsporozoite protein (CSP) and one of the polynucleotides encodes protein Ag45 (11-10).

6. The combination of compounds according to claim 4, further comprising a polynucleotide encoding an antigenic polypeptide selected from the group consisting of thrombospondin related anonymous protein (TRAP), inhibitor of cysteine protease (ICP), metallopeptidase (Falcilysin), GPI-anchored protein P113, and pore-forming like protein SPECT2.

7. The combination of compounds according to claim 5, further comprising a polynucleotide encoding an antigenic polypeptide selected from the group consisting of thrombospondin related anonymous protein (TRAP), inhibitor of cysteine protease (ICP), metallopeptidase (Falcilysin), GPI-anchored protein P113, and pore-forming like protein SPECT2.

8. The combination of compounds according to claim 3, wherein the distinct active ingredients of Plasmodium are provided as separate active ingredients.

9. The combination of compounds according to claim 3, wherein the distinct active ingredients of Plasmodium are provided as separate compositions of multiple active ingredients in the combination of compounds.

10. The combination of compounds according to claim 3, wherein the distinct active ingredients of Plasmodium are provided as a single composition.

11. The combination of compounds according to claim 3, wherein the active ingredients are lentiviral vector(s).

12. The combination of compounds according to claim 11, wherein the active ingredients are HIV-1 lentiviral vector(s).

13. The combination of compounds according to claim 12, wherein each lentiviral vector is a replication-incompetent pseudotyped lentiviral vector, wherein said vector contains a genome comprising a human-codon optimized synthetic nucleic acid.

14. The combination of compounds according to claim 13, wherein each lentiviral vector is an integrative pseudotyped lentiviral vector.

15. The combination of compounds according to claim 14, wherein the genome of the lentiviral vector genome is obtained from the pTRIP vector plasmid wherein the Plasmodium synthetic nucleic acid encoding the antigenic polypeptide has been cloned under control of the human beta-2 microglobulin promoter, and under the control of post-transcriptional regulatory element of the woodchuck hepatitis virus (WPRE).

16. The combination of compounds according to claim 15, wherein the lentiviral vector is pseudotyped with the glycoprotein G from a Vesicular Stomatitis Virus (V-SVG) of Indiana or of New-Jersey serotype.

17. A formulation suitable for administration to human host comprising a combination of compounds according to claim 3 together with excipient(s) suitable for administration to a human host.

18. A method for the protective immunisation against malaria parasite infection or against malaria parasite-induced condition or disease, in a human host comprising administering the combination of compounds according to claim 3 to said human host.

19. The method of claim 18, wherein the active ingredients are lentiviral vector(s) and are administered in a priming and boosting steps, wherein the lentiviral vector(s) are pseudotyped with distinct envelope protein(s) which do not cross-seroneutralise, and wherein said priming and boosting steps are separated in time by at least 6 weeks.

20. The method of claim 19, comprising separately provided doses of the lentiviral vector(s), wherein the dose for boosting the cellular immune response is higher than the dose for priming.

21. The method of claim 19, wherein the dose for priming and the dose for boosting the cellular immune response each comprises from 10.sup.5 to 10.sup.9 transducing units (TU) when integrative-competent vector particles are used and the dose for priming and for boosting the cellular immune response comprises from 10.sup.7 to 10.sup.10 TU when integrative-incompetent vector particles are used.

Description

LEGENDS OF THE FIGURES

(1) FIG. 1. Schema of plasmids used in the production of Lentiviral Particles.

(2) FIG. 2. C57BL/6 mice (n=5) were immunized intramuscularly with 5×10.sup.7 TU of VSV.sup.IND pseudotyped lentiviral particles coding for the antigens, CSP, Celtos SPECT, HSP20 and Ag13. As a positive control of protection, mice were immunized with 50 k irradiated sporozoites via intravenous injection. Thirty days after immunization, the animals were challenged with 10,000 bioluminescent sporozoites micro-injected subcutaneously in the mice footpad. The parasite load in the liver was quantified two days later by bioluminescence as shown in the picture for CSP, Celtos and Ag13. The graph shows the quantification of the liver infection represented as the log of average radiance (squares). Dotted line represents the average of background signal (Bk) of a non-infected region. *P<0.05 and ***P<0.001 (ANOVA).

(3) FIG. 3. C57BL/6 mice (n=5 per group) were immunized or not (naïve) with 5×10.sup.7 TU of VSV.sup.IND LPs carrying Ag13 (negative control) and CSP (positive control). The groups receiving concentrated LPs were inoculated intramuscularly in the thigh muscle with 50 uL of vector (Ag13 im c and CSP im c). The groups receiving non-concentrated LPs were inoculated intraperitoneally with 700 uL of vector (Ag13 ip nc and CSP ip nc). Thirty days after immunization, the animals were challenged with 5,000 luciferase-expressing sporozoites, micro-injected subcutaneously in the mice footpad. The parasite load in the liver was quantified two days later by bioluminescence as shown in the FIG. 2. The graph shows the average and sd of the log of average radiance in the liver two days after SPZ inoculation. Dotted line represents the average of background signal (Bk). ns, not significant (ANOVA).

(4) FIG. 4. C57BL/6 mice (n=4-5 per group) were intraperitoneally immunized or not (naïve) with 1×10.sup.7 TU of non concentrated VSV.sup.IND CSP LPs under the control of CMV or B2M promoters (CMV CSP and B2M CSP, respectively). Thirty days after immunization, the animals were challenged with 5,000 luciferase-expressing sporozoites micro-injected subcutaneously in the mice footpad. The parasite load in the liver was quantified two days later by bioluminescence as shown in the FIG. 2. The graph shows the average and sd of the log of average radiance in the liver two days after SPZ inoculation. Dotted line represents the average of background signal (Bk). *P<0.05; ns, not significant (ANOVA).

(5) FIG. 5. 4 and 7 weeks-old C57BL/6 mice (n=4-per group) were acclimated for 3 weeks (old groups) and 3 days (new groups). These age-matched groups were then intraperitoneally immunized with 1×10.sup.7 TU of non concentrated VSV.sup.IND B2M CSP or GFP LPs. Thirty days after immunization, the animals were challenged with 5,000 luciferase-expressing sporozoites micro-injected subcutaneously in the mice footpad. The parasite load in the liver was quantified two days later by bioluminescence as shown in the FIG. 2. The graph shows the average and sd of the log of average radiance in the liver two days after SPZ inoculation. Dotted line represents the average of background signal (Bk). *P<0.05; ns, not significant (ANOVA).

(6) FIG. 6. 4 weeks-old C57BL/6 mice (n=4-per group) were acclimated for 3 weeks (old groups) and intraperitoneally immunized with different doses of non-concentrated VSV.sup.IND B2M CSP (black) or GFP (white) LPs. Thirty days after immunization, the animals were challenged with 5,000 luciferase-expressing sporozoites micro-injected subcutaneously in the mice footpad. The parasite load in the liver was quantified two days later by bioluminescence as shown in the FIG. 2. The graph shows the average and sd of the log of average radiance in the liver two days after SPZ inoculation. Dotted line represents the average of background signal (Bk). *P<0.05; ***P<0.001; ns, not significant (ANOVA).

(7) FIG. 7. Analysis of Sporozoite, Liver Stage and Blood Stage cDNA libraries of Plasmodium berghei (Pb) and falciparum (Pf) deposited in Plasmodb. The percentage of each expression sequence tag (EST) was normalized to the total number of ESTs and represented cumulatively. Each symbol represents one gene, ranked by EST abundance (higher to lower) and represented as % of total ESTs. Of note ˜10% of genes (most abundant) are responsible for ˜50% of total ESTs (dotted lines).

(8) FIG. 8. Expression and surface localization of antigens. GFP-expressing Pb sporozoites were fixed with 2% of PFA and permeabilized with 0.1% of Triton X100 (perm) or not (live). Parasites were incubated with the indicated immune-sera (1/50) for one hour on ice, washed and revealed with goat anti-mouse secondary antibody labelled with AlexaFluor 647. Sporozoites were then analysed by cytometry as shown in the right histograms (surface, staining using live non-permeabilized SPZ; permeabilized, staining using fixed and permeabilized SPZ) or by fluorescence microscopy, as depicted in the pictures. Notice that CSP and antigen 9-6 present a surface pattern staining both by cytometry and microscopy.

(9) FIG. 9. Targeted screening of protective antigens. 4 weeks-old C57BL/6 mice (n=5 per group) were acclimated for 3 weeks and intraperitoneally immunized with a single dose of 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs. Thirty days after immunization, the animals were challenged with 5,000 GFP-expressing sporozoites micro-injected subcutaneously in the mice footpad. The parasite infection was measured by flow cytometry. The graph shows the average of the log of parasitemia (trace, individual mice represented by circles) immunized with the indicated plasmodial antigens. Bold dotted lines represent the 95% tolerance interval of GFP log normal distribution. Mice with parasitemia below the lower limit of the tolerance interval are considered protected. Top dotted line is the average of control and bottom dotted line represents non-infected (NI) mice.

(10) FIG. 10. Comparison of protection induced by one or two immunization doses. 4 weeks-old C57BL/6 mice (n=5 per group) were acclimated for 3 weeks and intraperitoneally immunized with a first dose of 5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs. Thirty days after the first immunization, the animals received a second dose of 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs. Thirty days later, mice were challenged with 5,000 GFP-expressing sporozoites micro-injected subcutaneously in the footpad. The parasite infection was measured by flow cytometry. The graph shows the log of parasitemia at day 5 post-inoculation of individual challenged mice that received two immunization doses (Squares, PB). Circles represent mice that received only one immunization dose of LPs (data from experiment shown in FIG. 9). Traces represents the average of the Log Parasitemia. Bold dotted lines represent the 95% tolerance interval of GFP log normal distribution. Mice below the lower limit of tolerance interval are considered protected. NI, non-infected mice.

(11) FIG. 11. Targeted Screening of Protective Antigens. 4 weeks-old C57BL/6 mice (n=5-10 per group) were acclimated for 3 weeks and intraperitoneally immunized with a first dose of 5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs. Thirty days after the first immunization, the animals received a second dose of 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs. Third days later, mice were challenged with 5,000 GFP-expressing sporozoites micro-injected subcutaneously in the footpad. The parasite blood infection was measured by flow cytometry. (a.) The upper graph shows the log of parasitemia of individual mouse at day 5 post-infection. Traces represent the mean of the log parasitemia. The average of the GFP group (control of protection) is represented by the dotted middle line. The superior and inferior dotted lines delineate the 95% tolerance interval (grey box) of the GFP control group. The CSP group is the positive control of protection. NI (not infected=no parasitemia at day 10 post-infection, located at the limit of detection of our method of parasitemia quantification). Black circles represent antigens where there was a significant decrease in the averaged log parasitemia and therefore are considered protective (ANOVA). (b) The bottom graph represents the percentage of protected mice (% of animals below the 95% tolerance interval). Black bars represent protective antigens (Fisher's Exact test). *P<0.05, **P<0.01, ****P<0.0001.

(12) FIG. 12. Structure of P. berghei protective antigens. Conserved structural and functional domains are represented by boxes according to the code on the right. GPI (glycosylphosphatidylinositol), TSR (thrombospondin type I repeat), MACPF (membrane attack complex/perforin).

(13) FIG. 13. Protective antigens are conserved among plasmodial species. Amino acid sequences of protective orthologous antigens from rodent-infecting P. berghei, macaque-infecting P. cynomolgi, and human-infecting P. falciparum and P. vivax parasites were aligned by MUltiple Sequence Comparison by Log-Expectation (MUSCLE). Vertical black bars represent identical amino acids conserved in the four plasmodial species, short dark gray bars represent repetitive regions and short light gray bars, insertional gaps used for the alignment.

(14) FIG. 14. Protection induced by combination of down-selected protective antigens with a sub-optimal dose of CSP. Mice were immunized twice, four weeks apart, with a sub-optimal dose of CSP (5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LP in the first immunization and 5×10.sup.6 TU of non-concentrated VSV.sup.IND B2M LP in the second immunization, white triangle, CSP) and the usual dose of protective plasmodial antigens (CSP+11-03, +11-05, +11-06, +11-07, +11-09 and +11-10; triangles). As negative control mice were immunized with the usual, two doses of GFP. 4 weeks after the second immunization dose, animals were challenged with 5,000 sporozoites.

(15) FIG. 15. Sterile protection induced by a multigenic combination. Mice were immunized twice, four weeks apart, with 7× the individual dose (1 dose=5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs in the first immunization/1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs in the second immunization) of the control antigen AL11-luciferase (Luc, white triangles), with the individual dose of CSP plus 6×Luc (gray triangles), or with the individual doses of CSP and of 6 conserved PE antigens (11-05, 11-06, 11-07, 11-09, 11-10 and 18-10; black triangles, 7cPEAg). 4 weeks after the second immunization dose, mice were challenged with 5,000 GFP SPZs. Both graphs show the individual log of parasitemia at day 5 post-challenge. (a) The graph shows the pooled results of three independent experiments. Number of sterile protected/challenged mice: 7×Luc (0/21), 1×CSP 6×Luc (1/20) and 1×7cPEAg (18/21). (b) Three and one day before sporozoite challenge, 1×7cPEAg immunized mice were injected with 400 μg of control (Ctr), CD4-depleting (a-CD4+, clone GK1.5) and CD8-depleting (a-CD8+, clone 2.43) monoclonal antibodies. GFP data comes from experiment showed in FIG. 11 (gray circles). Number of sterile protected/challenged mice: 7×Luc (0/7), 1×CSP 6×Luc (1/7) and 1×7cPEAg (ctr,7/7; a-CD8, 0/7 and a-CD4, 7/7). Notice that depletion of CD8+ cells abolished sterile protection. *P<0.05, **P<0.01, ****P<0.0001 (ANOVA).

(16) FIG. 16. Sterile protection induced by a multigenic combination in a single immunization dose. (a) Mice were immunized twice, four weeks apart, with 7× the individual dose (1 dose=5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs in the first immunization/1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs in the second immunization) of the control antigen AL11-luciferase (Luc, black triangles), with the individual dose of CSP plus 6×Luc (CSP, black triangles), or with the individual dose of CSP and of 6 conserved PE antigens (11-05, 11-06, 11-07, 11-09, 11-10 and 18-10; black triangles; 2 im 7cPEAg). Alternatively, mice were administered only with the second individual immunization dose (1×10.sup.7 TU) of CSP and of 6 conserved PE antigens (11-05, 11-06, 11-07, 11-09, 11-10 and 18-10; grey diamonds; 1 im 7cPEAg). 4 weeks after the second immunization dose, mice were challenged with 5,000 GFP SPZs. The graph shows the individual log of parasitemia at day 5 post-challenge. Black bars are the average of log of parasitemia. Number of sterile protected/challenged mice: Luc (0/7), CSP (0/7), 2im 7cPEAg (6/7) and 1 m 7cPEAg (6/7). (b) Mice were immunized once with 9× the individual dose (1 dose=1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs) of the control antigen AL11-luciferase (Luc, black diamonds), or with the individual doses of CSP+ of 7 conserved PE antigens (11-05, 11-06, 11-07, 11-09, 11-10, 18-10, 30-03A and 30-03B; grey diamonds; 8cPEAg). Three and one day before sporozoite challenge, 8cPEAg immunized mice were injected with 400 μg of control (Ctr), CD4-depleting (a-CD4+, clone GK1.5) and CD8-depleting (a-CD8+, clone 2.43) monoclonal antibodies. 4 weeks after the single immunization dose, mice were challenged with 5,000 GFP SPZs. The graph shows the individual log of parasitemia at day 5 post-challenge. Black bars are the average of log of parasitemia. Number of sterile protected/challenged mice: 7×Luc (0/7) and 8cPEAg (ctr,6/7; a-CD8, 0/7 and a-CD4, 4/7). Notice that depletion of CD8+ cells abolished protection. *P<0.05; **P<0.01; ****P<0.0001; ns, P>0.05 (ANOVA).

(17) FIG. 17. Sterile protection induced by a minimal combination of 5 PE antigens. (a) Mice were immunized twice, four weeks apart, with the individual dose multiplied by the number indicated in the circles (1 dose=5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs in the first immunization/1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs in the second immunization). For example, for the control antigen AL11-luciferase (LUC), animals were immunized with 5× the individual dose. All groups received 5 doses, with exception of the positive control of protection that received 7 doses of LPs (7PEAg). 4 weeks after the second immunization dose, mice were challenged with 5,000 GFP SPZs. Bars represents the percentage of sterile protected mice. The numbers of sterile protected/challenged mice are shown at the right of bars. **P<0.01 (Fisher's Exact test). (b) Mice were immunized twice, four weeks apart, with the individual dose (1 dose=5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs in the first immunization/1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs in the second immunization) of the control antigen GFP (GFP, black circles) or with the individual dose of CSP, TRAP, 18-10, 11-09 or 11-10 (grey triangles). Three and one day before sporozoite challenge, immunized mice were injected with 400 μg of control (Ctr), CD4-depleting (a-CD4+, clone GK1.5) and CD8-depleting (a-CD8+, clone 2.43) monoclonal antibodies. 4 weeks after the second immunization dose, mice were challenged with 5,000 GFP SPZs. Graphs show the average±sd of log of parasitemia at day 5 post-challenge. *P<0.05; ns, P>0.05 (ANOVA). (c) The 5 down-selected protective antigens were split according the presence of predicted CD8 T cell epitopes and respecting conserved structural domains as depicted by the schematic representation of the antigens. The graphs above the schematic proteins represent the distribution of epitopes predicted to bind to H2Kb (8 aa) and H2Kd (9 aa) MHC class I molecules using SYFPEITHI (score) and IEDB ANN IC 50 (nM). The graphs on the left of schematic proteins represent the protection induced by these constructs, where bars are the average±sd of log of parasitemia at day 5 post-challenge. Data shown for antigen 11-09 come from FIG. 11. Dotted line represents the inferior limit of the tolerance interval of the control calculated in the FIG. 11. *P<0.05; ns, P>0.05 (ANOVA). (d) Correlation of the best epitope predicted to bind to MHC class I molecules in the segments of CD8+ T cell dependent PE antigens and mean protective activity obtained from 17c. Circles show the IC50 using IEDB ANN software and squares the score values using SYFPEITHI. Dotted line shows the average of Luc control.

(18) FIG. 18. Clustering of CD8 T cell epitopes in conserved amino acid regions and binding of predicted Pf epitopes to HLA A02:01. Amino acid sequences of protective orthologous antigens from rodent-infecting P. berghei, macaque-infecting P. cynomolgi, and human-infecting P. falciparum and P. vivax parasites were aligned by MUltiple Sequence Comparison by Log-Expectation (MUSCLE). Vertical black bars represent identical amino acids conserved in the four plasmodial species. The graph shows the distribution of Pb epitopes predicted to bind to H2Kb (8 aa) and H2Kd (9 aa) MHC class I molecules or of Pf epitopes predicted to bind to the HLA A02:01 (9 mers) using SYFPEITHI (score) and IEDB ANN IC50 (nM). The best predicted HLA binders were tested in the assay of stabilization of MHC class I molecule in the presence of peptide and β2-microglobulin (REVEAL® Score). The score of 100 corresponds to the binding of a positive control peptide. Notice the clustering of epitopes in regions of conserved amino acids.

(19) FIG. 19. plasmid used to produce VSV-pseutdotyped lentiviral particles: pTRIP CMV GFP

(20) The sequence of the plasmid is constituted by the following functional regions wherein the cis-active lentiviral regions are derived from the HIV genome, and the promoter driving the expression of the protein (GFP) is CMV:

(21) The insert in the plasmid that provides the vector genome is composed as follows: LTR-ψ-RRE-cPPT/CTS-CMV-GFP-WPRE-ΔU3LTR, wherein

(22) LTR is Long Terminal Repeat

(23) Psi (ψ) is Packaging signal

(24) RRE is Rev Responsive Element

(25) CMV is Immediate early CytoMegaloVirus promoter

(26) cPPT is central PolyPurine Tract, and wherein the nucleotide segment from cPPT to CTS forms the flap sequence

(27) CTS is Central Termination Sequence

(28) WPRE is Woodchuck hepatitis virus Post Regulatory Element

(29) The nucleotide sequence is provided as SEQ ID No. 1

(30) FIG. 20: alternative plasmid (to the plasmid of FIG. 19) used to produce VSV-pseutdotyped lentiviral particles: pTRIP B2M GFP

(31) The insert in the plasmid that provides the vector genome is composed as follows: LTR-ψ-RRE-cPPT/CTS-B2M-GFP-WPRE-ΔU3LTR.

(32) The nucleotide sequence is provided as SEQ ID No. 2.

(33) FIG. 21. plasmid used to produce VSV-pseutdotyped lentiviral particles: packaging 8.74 plasmid

(34) The plasmid provides the required GAG and POL coding sequences of the HIV-1 lentivirus under the control of the CMV promoter.

(35) The nucleotide sequence is provided as SEQ ID No. 3.

(36) FIG. 22. plasmid used to produce VSV-pseutdotyped lentiviral particles: encapsidation plasmid pCMV-VSV-INDco

(37) The envelope protein is the VSV-G of the Indiana strain and the coding sequence has been mouse-codon optimized.

(38) The nucleotide sequence is provided as SEQ ID No. 4.

(39) FIG. 23. alternative plasmid (to plasmid of FIG. 22) used to produce VSV-pseutdotyped lentiviral particles: encapsidation plasmid pCMV-VSV-NJco

(40) The envelope protein is the VSV-G of the New-Jersey strain and the coding sequence has been mouse-codon optimized.

(41) The nucleotide sequence is provided as SEQ ID No. 5.

(42) The following table provides the list and identification of the sequences contained in the sequence listing.

(43) TABLE-US-00001 SEQ ID No. Sequence designation Origin Type  1 pTRIP CMV GFP DNA  2 pTRIP B2M GFP DNA  3 PACKAGING 8.74 PLASMID DNA  4 pCMV-VSV-INDco DNA  5 pCMV-VSV-Njco DNA  6 eGFP DNA  7 eGFP protein protein  8 AL11-Luciferase protein  9 AL11-Luciferase protein 10 circumsporozoite (CS) protein (CSP) P. berghei DNA mouseCO + Kozak ANKA strain 11 PbCSP (mouseCO + Kozak) P. berghei protein ANKA strain 12 PbCSP P. berghei protein ANKA strain 13 PfCSP humanCO + Kozak P falciparum DNA 3D7 strain 14 PfCSP (humanC0 + Kozak) P falciparum protein 3D7 strain 15 PfCSP P falciparum protein 16 PvCSP humanCO + Kozak P vivax Sal-1 DNA strain 17 PvCSP (humanCO + Kozak) P vivax Sal-1 protein strain 18 PvCSP P vivax Sal-1 protein strain 19 thrombospondin-related anonymous P. berghei DNA protein (PbTRAP) mouseCO + Kozak ANKA strain 20 PbTRAP (mouseCO + Kozak) P. berghei protein ANKA strain 21 PbTRAP P. berghei protein ANKA strain 22 PfTRAP humanCO + Kozak P falciparum DNA 3D7 strain 23 PfTRAP (humanCO + Kozak) P falciparum protein 3D7 strain 24 PfTRAP P falciparum protein 25 PvTRAPhumanCO P vivax Sal-1 DNA strain 26 PvTRAP P vivax Sal-1 protein strain 27 PvTRAP P vivax protein 28 inhibitor of cysteine proteases (ICP) P. berghei DNA mouseCO + Kozak ANKA strain 29 PbICP (mouseCO + Kozak) P. berghei protein ANKA strain 30 PbICP P. berghei protein ANKA strain 31 PfICP humanCO P falciparum DNA 3D7 strain 32 PfICP P falciparum protein 3D7 strain 33 PfICP P falciparum protein 34 PvICP humanCO + Kozac P vivax Sal-1 DNA strain 35 PvICP (humanCO + Kozac) P vivax Sal-1 protein strain 36 PvICP P vivax protein 37 Bergheilysin-A-mouseCO + Kozak P. berghei DNA ANKA strain 38 Bergheilysin-A (1-777, P. berghei protein mouse CO + Kozak) ANKA strain 39 Bergheilysin entire ORF (1-1149) P. berghei protein ANKA strain 40 Falcilysin human CO + Kozak P falciparum DNA 3D7 strain 41 Falcilysin (human CO + Kozak) P falciparum protein 3D7 strain 42 Falcilysin P falciparum protein 3D7 strain 43 PvFalcilysin human CO + Kozak P vivax Sal-1 DNA strain 44 PvFalcilysin (humanCO + Kozak) P vivax Sal-1 protein strain 45 PvFalcilysin P vivax Sal-1 protein strain 46 Bergheilysin-B-mouseCO + Kozak + P. berghei DNA signal peptide (SP) ANKA strain 47 Bergheilysin-B (SP + 778-1149, P. berghei protein mouse CO + Kozak) ANKA strain 48 perforin like protein 1 (SPECT2) P. berghei DNA mouseCO + Kozak ANKA strain 49 PbSPECT2 (mouseCO + Kozak) P. berghei protein ANKA strain 50 PbSPECT2 P. berghei protein ANKA strain 51 PfSPECT2 human CO + Kozak P falciparum DNA 3D7 strain 52 PfSPECT2 (humanCO + Kozak) P falciparum protein 3D7 strain 53 PfSPECT2 P falciparum protein 54 PvSPECT2 human CO + Kozak P vivax Sal-1 DNA strain 55 PvSPECT2 (human CO + Kozak) P vivax Sal-1 protein strain 56 PvSPECT2 P vivax protein 57 GPI_P113 mouseCO + Kozak P. berghei DNA ANKA strain 58 Pb GPI_P113 (mouseCO + Kozak) P. berghei protein ANKA strain 59 Pb GPI_P113 P. berghei protein ANKA strain 60 PfP113 human CO + Kozak P falciparum DNA 3D7 strain 61 PfP113 (human CO + Kozak) P falciparum protein 3D7 strain 62 P113 P falciparum protein 63 PvP113 human CO + Kozak P vivax Sal-1 DNA strain 64 PvP113 (human CO + Kozak) P vivax Sal-1 protein strain 65 P113 P vivax protein 66 PbAg40 mouse CO + Kozak P. berghei DNA ANKA strain 67 PbAg40 (mouse CO + Kozak) P. berghei protein ANKA strain 68 PbAg40 P. berghei protein ANKA strain 69 PfAg40 human CO + Kozak P falciparum DNA 3D7 strain 70 PfAg40 (human CO + Kozak) P falciparum protein 3D7 strain 71 Ag40 P falciparum protein 72 PvAg40 human CO + Kozak P vivax Sal-1 DNA strain 73 PvAg40 (human CO + Kozak) P vivax Sal-1 protein strain 74 PvAg40 P vivax Sal-1 protein strain 75 PbAg45 mouse CO + Kozak P. berghei DNA ANKA strain 76 PbAg45 (mouse CO + Kozak) P. berghei protein ANKA strain 77 PbAg45 P. berghei protein ANKA strain 78 PfAg45 human CO + Kozak P falciparum DNA 3D7 strain 79 PfAg45 (human CO + Kozak) P falciparum protein 3D7 strain 80 PfAg45 P falciparum protein 81 PvAg45 human CO + Kozak P vivax Sal-1 DNA strain 82 PvAg45 (human CO + Kozak) P vivax Sal-1 protein strain 83 PvAg45 P vivax protein 84 Kozak consensus sequence DNA 85 Kozak consensus sequence DNA 86 BamHI site DNA 87 Xhol site DNA 88-94 CD8 T cell epitopes protein

(44) Additional information relating to some of the sequences disclosed in the above table are provided in the table below.

(45) TABLE-US-00002 SEQ ID GenBank strain pubmed 15 BAM84930.1 Plasmodium falciparum isolate 23295064 Pal97-042 origin: Philippines ACO49323 Plasmodium falciparum” isolate A5 19460323 origin: Thailand 18 AAA29535.1 P.vivax (strain Thai; 2290443 isolate NYU Thai) origin: Thailand 24.sup.(1) EWC74605.1 Plasmodium falciparum UGT5.1 strain origin: Uganda 27 AIU97014.1 Plasmodium vivax isolate = “TMS38” origin: Thailand 36.sup.(3) KMZ87332.1 Plasmodium vivax Brazil I strain 56 KMZ82648.1 Plasmodium vivax India VII 65 KMZ78214.1 Plasmodium vivax India VII 83 KMZ90984.1 Plasmodium vivax Mauritania I .sup.(1)Worldwide web: ncbi.nlm.nih.gov/biosample/SAMN01737342 .sup.(2)Worldwide web: ncbi.nlm.nih.gov/biosample/SAMEA2394724 .sup.(3)Worldwide web: ncbi.nlm.nih.gov/biosample/SAMN00710434

EXAMPLES

(46) To approach the complex problem of identifying protective antigens, the inventors devised a functional screening to identify and combine novel PE protective antigens using a rodent malaria model where mice (C57BL/6) are extremely susceptible to Plasmodium berghei (Pb) sporozoite infection. In this model, sterilizing protection induced by live irradiated sporozoites is mediated by antibodies and mainly by CD8 T cell responses against sporozoites and liver stages, respectively. The inventors' screening strategy was designed based on four main features: i) parameterized selection of 55 PE antigens based on abundance, orthology, predicted topology and function, ii) synthesis of codon-optimized antigens to avoid AT-rich plasmodial sequences and maximize the expression in mammalian cells, iii) immunization using HIV-based lentiviral vector that elicits strong humoral and cellular responses.sup.11, 12, and iv) measurement of protection after a stringent challenge of sporozoites inoculated sub-cutaneously in the immunized mice.

(47) 1. Setting Up the Parameters of the Screening.

(48) In a proof-of-concept experiment aimed at validating the viability of the strategy to screen antigens at a medium-throughput, the inventors ordered mouse-codon optimized synthetic genes of Pb CSP (SEQ ID No. 11), a known protective antigen, and of more 4 other sporozoite antigens (Celtos, SPECT, HSP20 and Ag13), which were previously correlated with protection.sup.13. The synthetic plasmodial genes were cloned in to the pTRIP vector plasmid, which drives their expression in mammalian cells via the immediate-early cytomegalovirus promoter (CMV) and the post-transcriptional regulatory element of woodchuck hepatitis virus (WPRE) (FIG. 1, SEQ ID No. 1). These two elements assure a strong expression of the antigen in a wide variety of mouse cells in vivo. HIV-1 derived lentiviral particles were produced by transient co-transfection of HEK 293T cells with three helper plasmids encoding separate packaging functions, the pTRIP vector plasmid containing the synthetic plasmodial gene, the envelope expression plasmid encoding the glycoprotein G from the Vesicular Stomatitis Virus, Indiana (VSV.sup.IND) or New Jersey (VSV.sup.NJ) serotypes, and the p8.74 encapsidation plasmid (FIG. 1). This co-transfection generates integrative but replication-incompetent pseudotyped lentiviral particles capable of transducing dividing and non-dividing cells—including dendritic cells—and inducing potent cellular.sup.6 and humoral.sup.7 memory responses. The particles were collected 48 hours after co-transfection and each batch of vector were titrated in HeLa cells by quantitative PCR. This functional titration assay gives the concentration of particles capable to transfer one copy of the gene per cell and will be expressed in Transducing Units (TU)/mL. Plasmid sequences are shown in the figures and their sequences are provided in the sequence listing.

(49) Groups of five mice were immunized with a single intra-muscular dose of 5e7 TU of ultracentrifugation-concentrated VSV.sup.IND pseudotyped lentiviral particles (LPs). Thirty days after immunization, mice were challenged with 10,000 bioluminescent sporozoites inoculated sub-cutaneously in the footpad. Two days later, the parasite load in the liver was measured by bioluminescence. Surprisingly, CSP-immunization decreased 15×-fold the parasite load in the liver after a challenge using 10,000 bioluminescent sporozoites, versus a 5×-fold decrease in animals immunized intravenously with 50,000 irradiated sporozoites, our golden standard of protection (FIG. 2). This preliminary and promising result validated the high performance of the present method to functionally identify new protective antigens and showed the feasibility to scale-up our test samples.

(50) The inventors next aimed at the transposition of these optimal experimental conditions to those of a larger screening. This transposition included the validation of the use of non-concentrated LPs, the choice of the best promoter driving the expression of the plasmodial antigens, and the dose of immunization.

(51) The first parameter tested was the use of non-concentrated, instead of concentrated LPs, to avoid a costly and time consuming ultracentrifugation concentration step in the protocol of LP production, which requires large volumes of non-concentrated LP suspensions. FIG. 3 shows that there is no significant difference between protection induced by the same dose (5×10.sup.7 TU) of concentrated LPs injected intramuscularly (CS im c, 50 μL) and non-concentrated LPs injected intraperitoneally (CS ip nc, 700 μL). Protection was measured by reduction in the liver infection, as assessed by bioluminescence after a challenge of 5,000 sporozoites injected subcutaneously 30 days following immunization. As negative control of protection the inventors used mice immunized with Pb Ag13, determined previously as a non-protective antigen (FIG. 1).

(52) Next two promoters were tested to identify which one induced the best protection using the codon optimized Pb CSP. The inventors compared the use of the strong and constitutive cytomegalovirus (CMV) promoter versus a human beta-2 microglobulin (B2M) promoter, which direct gene expression in many cell types, particularly in dendritic cells. FIG. 4 shows that CSP-induced protection was slightly better, although not statistically significant, using the B2M promoter at an immunization dose of 1×10.sup.7 TU of non-concentrated LP. Therefore the inventors further adopted this promoter in our constructs.

(53) During this period of optimization the inventors observed some variations in the CSP-induced protection using the same stock of LPs, as can be seen in the FIG. 4. The inventors asked if this variability could be due to the process of mouse acclimation, including the modification of mouse microbiota. To test this hypothesis a group of mice purchased from Elevage Janvier (4 weeks-old) was reared in the animal facility for 3 weeks before immunization (group old). A second group of mice (7 weeks-old) was purchased and put in cages 3 days before the immunization (group new). Both groups were intraperitoneally immunized with 1×10.sup.7 TU of non-concentrated LPs. As shown in the FIG. 5, mouse acclimation of 3 weeks resulted in a significant and more homogeneous protection when compared to 3 days of acclimation. Consequently, the inventors adopted this period of acclimation in all our subsequent experiments.

(54) Next, the best protective immunization dose was tested, ranging from 1×10.sup.8 to 1×10.sup.5 TU of B2M CSP non-concentrated LPs. As shown in FIG. 6, significant protection was observed using 10.sup.7 and 10.sup.8 TU, and the best protective activity was observed using an immunization dose of 1×10.sup.7 TU. In this experiment the inventors also observed a gradual loss of SPZ infectivity over time, as evidenced in the GFP groups, due to the use of a single SPZ stock to challenge all animals. To reduce the multiple shocks of temperature due to the manipulation of the stock tube, kept on ice between injections, the inventors prepared a SPZ stock for each group in the subsequent challenges and this variation disappeared.

(55) In summary, an immunization protocol was set up based on CSP that relied on a single intraperitoneal injection of 10.sup.7 TU of non-concentrated VSV.sup.IND B2M LP in C57BL/6 mice of 7 weeks-old, acclimated for 3 weeks in the animal facility. In the pooled data, this protocol leaded in average to a ˜5-fold decrease in the parasite liver load, as assessed by bioluminescence imaging, using a subcutaneous challenge of 5,000 luciferase-expressing SPZ.

(56) However, this bioluminescent method of detection of parasites presents some disadvantages such as the use and associated costs of anesthesia and luminescent substrate, limited capacity of analysis of a few animals per acquisition, being time-consuming and not sensible enough to predict sterile protection. Therefore, the inventors decided to use fluorescent parasites to check protection by measuring parasitemia at day 4, 5, 6 and 10 post-inoculation by flow cytometry. The inventors analyze at least 100,000 red blood cells, which gives the sensibility to detect a parasitemia of 0.001%. At day 4 to 6, parasites grow exponentially in the blood therefore the log transform of parasitemia can be fitted using a linear regression where the slope represents the time of parasite replication per day. Consequently, the inventors use this parameter to determine if the immunization impacts the parasite growth in the blood. For quantifying protection, the inventors used the log of parasitemia at day 5 post inoculation. This represents an indirect measure of liver infection and it is more robust than the measure at day 4 because more events of infected blood cells are registered. Finally the inventors defined that immunized mice are sterile protected if infected red blood cells are not detected after 10 days post inoculation. After defining the protocol of immunization and the method for the quantification of parasite infection the inventors started to screen the protective activity of down-selected antigens.

(57) 2. Parameterized Selection of Antigens

(58) By merging proteomic and transcriptomic data using PlasmoDB (Worldwide web: plasmodb.org), the inventors identified ˜9000 genes expressed in plasmodial pre-erythrocytic stages—salivary gland sporozoites and liver-stages—of three different plasmodial species, with 3654 syntenic orthologs in Plasmodium falciparum (Pf), the most lethal human-infecting plasmodial species. By analyzing the repertoire of pathogen transcripts, as inferred by the amount of expressed sequence tags (ESTs) obtained in cDNA libraries of different stages and species of malaria parasites, the inventors have observed that ˜50% of the total amount of ESTs are coming from only ˜10% of genes represented in these libraries, corresponding to approximately 100 genes in these libraries (FIG. 7). Therefore, by focusing on the ˜100 most abundant transcribed genes the inventors could target about 50% of the putative (to be translated) antigenic mass of a given parasite stage. Accordingly, the inventors selected ˜50 abundantly transcribed genes coding for conserved proteins with high probability of being expressed/presented on the surface of the parasite/infected cell, giving priority to candidates containing T cell epitopes predicted by IEDB MHC binding algorithm (Worldwide web: tools.iedb.org/mhci/). A Kozak consensus sequence, a translational start site, was added to these down-selected genes, which were then mammalian codon-optimized and synthesized by MWG Eurofins (listed in the figures). These synthetic codon-optimized down-selected plasmodial genes were then cloned into the B2M pTRIP plasmid and produced as non-concentrated VSV.sup.IND LPs.

(59) 3. First Screening of Protective Antigens Using a Single Dose of LPs

(60) Usually, 6-10 plasmodial antigens were tested by experiment, with a negative (GFP) and positive (CSP) control of protection. After three weeks post-immunization, the immune-sera were tested on permeabilized and non-permeabilized sporozoites, allowing the determination of (i) the efficiency of the host humoral response and therefore the immunogenicity of the lentivirus-delivered antigen, and (ii) the localization of the parasite antigen (surface vs intracellular). As shown in the FIG. 8, where the inventors immunized mice with putative GPI-anchored antigens, surface antigens were identified by flow cytometry and immunofluorescence (CSP and 9-6). The sera of GFP and CSP group served, respectively, as positive control for intracellular and surface antigen localization.

(61) Four weeks post-immunization the animals were challenged with 5,000 GFP-expressing sporozoites, microinjected in the footpad of immunized mice. Parasitemia was determined by flow cytometry. To define protection, parasitemia of all GFP groups (day 5 post-infection, n=35) was log transformed, pooled and the 95% tolerance interval was calculated (FIG. 9). All animals below the inferior limit of the tolerance interval, which represents a ˜8-fold decrease in parasitemia compared to the mean log of parasitemia of the GFP group, were considered protected. As positive control, 43% of animals (15/35) were protected by CSP immunization with a mean decrease of ˜5 fold in comparison to the GFP group. 9% of them (3/35) became sterile protected after sporozoite challenge.

(62) In the first set of 43 antigens tested (FIG. 9), we identified 9 PE antigens that protected at least one out of five immunized mice (black circles; 07-03, 09-06, 10-05, 10-10, 12-03, 12-04, 12-05, 12-07 and 13-08). Three of them were also identified as sporozoite surface antigens (09-06, 10-05 and 10-10).

(63) To verify the robustness of our screening, the inventors selected 4 protective antigens (CSP, 09-06, 10-05 and 07-03), 6 non-protective antigens (GFP, 09-07, 07-05, 07-06, 06-06 and 10-06), and instead of only one immunization dose, the inventors administered one dose of 5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs and one month later, a second dose of 1×10.sup.7 TU of VSV.sup.IND B2M LPs. As shown in the FIG. 10, the inventors observed three patterns of infection profile when the inventors compared one (circles, data from FIG. 9) and two immunization doses (squares, PB). For the non-protective antigens GFP, 09-07 and 07-05, the second dose of LP did not change the profile of infection, as expected. For the protective antigens CSP (***P<0.001), 09-06, 10-05 and 07-03, the second dose of LP increased the number of protected mice and/or decreased the average parasitemia, also, as expected. Notably, the non-protective antigens 07-06, 06-06 and 10-06 as assessed by one dose of LP immunization, showed a strong protective activity, including a sterile protected mice (PB 7-6), when administered twice in mice.

(64) These results validated some of our protective antigens detected with a single immunization dose, but also showed that some good protective antigens were not detected in our first screen, leading to the decision of repeating the screening using two immunization doses.

(65) 4. Second Screening of Protective Antigens Using Two Doses of LP

(66) By functionally screening the protective activity of 55 down-selected plasmodial PE antigens using the protocol of two immunization doses, the inventors identified 16 antigens that protected at least one immunized mice per group. Among these 16 antigens, 7 of them (black circles/bars in the FIG. 11) conferred significant protection when compared to animals immunized with the GFP, both when analysing the number of protected mice (Fisher's Exact test) or the mean of the log parasitemia (ANOVA).

(67) All of them presented a similar or an inferior protective activity when compared individually to our standard of protection, the CSP (FIG. 11). Five of them are molecules with assigned function (11-05, 11-06, 11-07, 30-03 and 18-10) and two are proteins with no predicted function (11-09 and 11-10). The structure of these Pb protective antigens is shown in the FIG. 12 and the alignment of these proteins with their respective orthologs from human-infecting parasites, P. falciparum (Pf) and P. vivax (Pv), and macaque-infecting parasite P. cynomolgi, is represented in the FIG. 13. As shown in table I, the percentage of identical amino acids between orthologs varied from 75 to 38% (Pb vs Pf), 78 to 33% (Pb vs Pv) and 79 to 26% (Pf vs Pv). The most conserved genes (>50% identity) are 30-03, 11-09, 11-10 and 11-06 orthologs. Antigens with divergent repetitive sequences are penalized in the alignment by insertional gaps, presenting less percentage of identity.

(68) TABLE-US-00003 TABLE I Percent Identity Matrix created by CLUSTAL 2.1. Amino acid sequence of Pb antigens were pBlasted against Pf and Pv taxids (organism) and the best matched sequence was used to align the orthologous proteins using MUSCLE (Worldwide web: ebi.ac.uk/Tools/msa/muscle/). The table shows the percentage of identical amino acids between species. Raw data is presented in the figures. Amino acid identity (%) Antigen Pb/Pf Pb/Pv Pf/Pv 30-03 74.65 77.90 73 11-09 66.19 66.19 79.05 11-06 64.92 63.40 64.44 11-10 56.94 50.15 62.28 CSP 42.06 33.53 26.36 18-10 39.60 41.23 49.30 11-05 38.75 44.67 42.86 11-07 37.53 33.42 46.30

(69) In a decreasing order of protection the first antigen identified is TRAP.sup.14 (thrombospondin related anonymous protein; 11-05) (SEQ ID No. 20 and 21), which validated our method of screening since immunization with TRAP is known to induce protection both in rodents.sup.15 and humans.sup.16. TRAP is a type I transmembrane protein harbouring two extracellular adhesive domains, a von Willebrand factor type A domain and a thrombospondin type 1 domain, followed by a proline-rich repetitive region. TRAP is stored in micronemal secretory vesicles and following parasite activation, the protein is translocated to the surface of sporozoites where it serves as a linker between a solid substrate and the cytoplasmic motor of sporozoites. Intriguingly, anti-TRAP antibodies do not impair parasite motility and infectivity.sup.17 CD8+ T cells seem to mediate the protection mediated by TRAP immunization.sup.10,15,16,18.

(70) The second protective antigen identified is an inhibitor of cysteine protease (ICP, 18-10).sup.19 (SEQ ID No. 29, 30). ICP seems to be involved in the motility and infectivity capacity of sporozoites via the processing of CSP.sup.20,21,, as well as, in the parasite intra-hepatic development.sup.22. Although the protein does not present structural signatures of membrane localization, there is evidence that the protein is located on the surface of sporozoites.sup.19,20. Opposing results are published regarding the secretion of the protein following parasite activation.sup.21,22. Similarly, there are contradictory results regarding the inhibition of host cell invasion by sporozoites in vitro in the presence of anti-ICP immune sera.sup.20,23.

(71) The third protective antigen identified is a metallopeptidase (Falcilysin/Bergheilysin, 30-03).sup.24 (SEQ ID No. 38 for Bergheilysin A, No. 47 for Bergheilysin B, and No. 39 for the entire Bergheilysin ORF). This protease seems to be involved in the catabolism of hemoblobin in the parasite blood stages.sup.25. A H-2K.sup.b-restricted CD8 T cell epitope was recently described during the parasite blood infection.sup.25 suggesting that CD8 T cells could mediate the protection elicited by the antigen 30-03 during the hepatic infection.

(72) The fourth protective antigen is a GPI-anchored protein (P113, 11-07) (SEQ ID No. 58 and 59) initially described in blood stages.sup.16 and also expressed in PE stages. P113 seems to be important for liver infection, dispensable for blood infection, but its precise function is unclear.sup.17.

(73) The fifth antigen is the pore-forming like protein SPECT2 (11-06).sup.28 (SEQ ID No. 49 and 50). This protein has a membrane attack complex/perforin (MACPF) domain and is involved in the sporozoite cell traversal activity, being important for the progression of sporozoites in the dermis.sup.29 and survival to phagocytosis in the liver.sup.30.

(74) The sixth antigen identified is a hypothetical protein that the inventors called 11-09 or Ag40 (SEQ ID No. 67 and 68). This protein has 4-5 annotated transmembrane domains. Deletion of the gene coding for the antigen 11-09 caused impairment of Pb parasite development in the liver.

(75) The seventh antigen is also a hypothetical protein that the inventors called 11-10 or Ag45 (SEQ ID No. 76 and 77). This protein doesn't have annotated domains, but possesses a central region with negatively charged amino acids. Recently the 11-10 ortholog of Plasmodium yoelii, another rodent-infecting plasmodial species, was also identified as a protective antigen.sup.21. The deletion of the gene coding for the antigen 11-10 blocked the Pb sporozoite invasion of salivary glands and completely abolished the capacity of sporozoites to infect the liver.

(76) To determine if CSP based protection could be additively or synergistically improved by the combination of antigens, the inventors assessed the protection elicited by a sub-optimal dose of CSP (5×10.sup.5 TU of VSV.sup.NJ/5×10.sup.6 TU of VSV.sup.IND B2M LPs) in the absence or presence of a usual dose of protective antigens (5×10.sup.5 TU of VSV.sup.NJ/1×10.sup.7 TU of VSV.sup.IND B2M LPs). This protection induced by CSP+protective antigens was compared to the protection elicited by these antigens alone (data from FIG. 11). As negative control the inventors used animals immunized with the usual dose of GFP LPs. FIG. 14 shows that 4 antigens when combined with a sub-optimal dose of CSP (CSP+11-03, +11-10, +11-07 and +11-05, triangles) did not change the average of protection when compared to the protective activity elicited by these antigens administered alone. For two antigens, the antigen combination (CSP+11-09 and CSP+1-06) showed a tendency of better protection (˜10 fold), but not statistically significant.

(77) 5. Identification of Multi-Antigenic Formulations Capable of Sterilizing Sporozoite Infection Via a CD8+ T Cell Response

(78) Since testing all possible combinations of antigens was technically unfeasible, the inventors decided to evaluate the protection elicited by the combination of these multiple protective antigens. Remarkably, two immunizations of mice with the combination of CSP and 6 of these antigens (11-05, 11-06, 11-07, 11-09, 11-10, 18-10) elicited sterile protection in the vast majority of challenged animals (7PEAg, 86-100%, FIG. 15). This percentage of sterile protection was far superior to the protection conferred by CSP in the same experimental conditions (0-14%). Depletion of CD8+ cells (α-CD8) just before the challenge, but not of CD4+ cells, decreased this protection to the level of that induced by CSP, suggesting that CD8+ T cells mediate the extra-protection elicited by the addition of these 6 PE antigens.

(79) The same protective efficacy was observed using only a single immunization for the 7PEAg or for the 7PEAg+30-03 (FIGS. 16a and 16b, 8PEAg), as well as, the dependence on CD8+ T cells for the sterilizing immunity of 8PEAg (FIG. 16b). Since the antigen 30-03 is a large molecule and did not improve sterile protection when administered with the 7PEAg, the inventors excluded it from further analysis.

(80) 6. Design of a Chimeric Antigen Containing the Protective Domains of Down-Selected PE Antigens

(81) To determine a minimal antigenic composition capable of eliciting this additional protective CD8+ T cell response, the inventors first identified the antigens whose protective activity was dependent on these T cells. Protection induced by two immunizations using TRAP, 18-10 and 11-09 was significantly reduced after depletion of CD8+ cells, as shown in the FIG. 17b. Protection induced by two immunizations using 11-10 was reduced after depletion of CD8+ cells but it was not statistically significant (FIG. 17b). Therefore the inventors grouped CSP with the CD8+ dependent protective antigens, TRAP, 18-10 and 11-09 and added separately 11-10, 11-07 and 11-06 to identify a minimal antigenic combination capable of sterile protect immunized animals like the complete combination of antigens. As shown in FIG. 17a, the combination of 5 antigens, CSP+TRAP, 18-10, Ag40 and Ag45 induced comparable level of sterile protection elicited by the combination of the 7PEAg.

(82) In order to combine the protective domains of each of these 5 antigens in a single chimeric molecule and thus avoid the costs associated with the production and delivery of five different antigens, the inventors mapped the protective regions of each antigen according to the localization of predicted epitopes binding to MHC class I molecules (FIGS. 17c and 18) and structural-functional conserved motifs (FIGS. 12, 13, 17c and 18).

(83) As shown in the FIG. 17c, all tested domains presented either a better (11-10CT) or similar protective activity when compared to the entire antigen. The level of mean protection elicited by the domains of antigens inducing protective CD8+ T cells correlated with the score (P<0.01) or affinity (P=0.01) of CD8+ T cell epitopes respectively predicted by SYFPEITHI and IEDB (FIGS. 17c and 17d). Importantly, the mapping of protective domains allowed the reduction of ˜2000 basepairs in the final chimeric PE antigen construct. Due to its small size, Ag40 was not split in domains and the data presented in the FIG. 17c comes from the experiment showed in the FIG. 11.

(84) Analysis of the distribution of epitopes of Pb antigens predicted to bind to MHC class I molecules of C57BL/6 mice (H2-K.sup.b, H2-D.sup.b) or of the Pf orthologues predicted to bind to HLA A02:01, a high prevalent human HLA allele, revealed that most of predicted good binders are clustering in the regions that are conserved among different plasmodial species (Pb, Pc, Pv and Pf, FIG. 18). This renders possible the utilization of the Pb protective regions mapped in the FIG. 17 to select the correspondent regions in the Pf orthologues. In addition, the inventors validated the binding of the best predicted Pf epitopes to the HLA A02:01 class I molecule using the REVEAL® binding assay developed by Proimmune, which allows the quantification of the binding and stabilization of the complex formed by the tested peptide, HLA A02:01 and β2-microglobulin (FIG. 18).

(85) In summary, using a parameterized selection of antigens, a screening based on lentiviral vaccination and a direct measurement of protection in vivo against a stringent sporozoite challenge, the inventors identified 8 protective antigens, including the vaccine candidates CSP and TRAP, out of 55 tested antigens. All these 8 antigens are conserved across several plasmodial species. Remarkably, immunization using a combination of seven or eight of these antigens elicited sterile protection in the vast majority of challenged mice, either using one or two immunizations. More importantly, this protection was far superior than the one elicited by CSP, so far the best protective PE antigen. Depletion of CD8+ T cells abolished sterilizing immunity, indicating that these cells are essential for this protective phenotype, similarly to the protection conferred by irradiated sporozoites. A minimal combination of 5 of these antigens was also capable of eliciting sterile protection in most of challenged animals. Mapping of the protective domains of these 5 antigens allowed the design of a chimeric antigen containing the fused protective domains of these 5 down-selected antigens. The human-infecting parasite orthologs of these protective antigens, or of their protective domains are potential candidates for being used in the development of a malaria vaccine formulation containing multiple protective antigens or multiple protective domains fused in a single molecule.

(86) Material and Methods

(87) Parasite Strains:

(88) Plasmodium berghei ANKA strain constitutively expressing the GFP under the control of the hsp70 promoter (Ishino et al, 2006) was used in the challenges using parasitemia, quantified by flow cytometry, as protective readout. Plasmodium berghei ANKA strain constitutively expressing a GFP-luciferase fusion under the control of the eef-1alfa promoter (Franke-Fayard et al, 2008) was used in the challenges using liver infection, assessed by bioluminescence, as protective readout. Of note, parasitemia quantified using hsp70-gfp parasites was at least 10 times more sensible than using eef-1a gfp:luc parasites due to more intense expression level of GFP. Ishino T, Orito Y, Chinzei Y, Yuda M (2006) A calcium-dependent protein kinase regulates Plasmodium ookinete access to the midgut epithelial cell. Mol Microbiol 59:1175-1184. Franke-Fayard B, Djokovic D, Dooren M W, Ramesar J, Waters A P, et al. (2008) Simple and sensitive antimalarial drug screening in vitro and in vivo using transgenic luciferase expressing Plasmodium berghei parasites. Int J Parasitol 38:1651-1662.

(89) Mouse Strains:

(90) C57BL/6 Rj and Swiss mice were purchased from Elevage Janvier (France). All experiments were approved by the Animal Care and Use Committee of Institut Pasteur (CETEA 2013-0093) and were performed in accordance with European guidelines and regulations (MESR-01324).

(91) Production of Lentiviral Particles Stock:

(92) Down-selected plasmodial antigens were synthesized by Eurofins MWG as mouse codon-optimized genes with the addition of a Kozak consensus sequence (GCCACCATGGCT(C) (SEQ ID No. 85 and 86), representing the first 12 nucleotides in the coding sequences of the antigenic polypeptides), encompassing the first translated ATG. This modification adds an extra alanine after the first methionine. A BamHI (GGATCC-SEQ ID No. 87) and Xho I (CTCGAG-SEQ ID No. 88) restriction sites were also inserted in the 5′ and 3′ extremities of the construct, respectively. These synthetic codon-optimized genes were then cloned into the BamHI and Xho I restriction sites of the pTRIP plasmid harboring either the CMV or B2M promoter (FIGS. 16 and 17). Lentiviral particles were produced by transient calcium co-transfection of HEK 293T cells with three helper plasmids encoding separate packaging functions, the pTRIP vector plasmid containing the synthetic plasmodial gene, the envelope expression plasmid encoding the glycoprotein G from VSV (Vesicular Stomatitis Virus, Indiana (FIG. 19) or New Jersey (FIG. 20) serotypes) and the p8.74 encapsidation plasmid (FIG. 18), containing the structural, accessory and regulatory genes of HIV. This co-transfection will generate integrative but replication-incompetent pseudotyped lentiviral particles. At 24 hours post-transfection, the cell culture medium was replaced by serum-free DMEM. Supernatants were collected at 48 hours post-transfection, clarified by low-speed centrifugation, and stored at −80° C. The lentiviral vector stocks were titrated by real-time PCR on cell lysates from transduced HEK 293T cells and titer were expressed as transduction unit (TU) per ml.

(93) Immunization Protocol:

(94) For the screening using one single dose of LPs, 4 weeks-old C57BL/6 mice (n=5 per group per experiment) were acclimated for 3 weeks and intraperitoneally immunized with a single dose of 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs. For the protocol using two immunization doses. 4 weeks-old C57BL/6 mice (n=5 per group per experiment) were acclimated for 3 weeks and intraperitoneally immunized with a first dose of 5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs. Thirty days after the first immunization, the animals received a second dose of 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs. For testing combinations of a sub-optimal dose of CSP+ an optimal dose of down-selected antigens, mice were immunized twice, four weeks apart, with a sub-optimal dose of CSP (5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LP in the first immunization and 5×10.sup.6 TU of non-concentrated VSV.sup.IND B2M LP in the second immunization) and the usual dose of protective plasmodial antigens (5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LP in the first immunization and 1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LP in the second immunization). For testing the combination of multiple antigens, mice were immunized twice, four weeks apart, with 7× the individual dose (1 dose=5×10.sup.5 TU of non-concentrated VSV.sup.NJ B2M LPs in the first immunization/1×10.sup.7 TU of non-concentrated VSV.sup.IND B2M LPs in the second immunization) of the control antigen Al11-luciferase (Luc), with the individual dose of CSP plus 6 doses of Luc or with the individual doses of CSP and of the 6 conserved PE antigens (11-05, 11-06, 11-07, 11-09, 11-10 and 18-10). For this experiment the inventors used ultrafiltration and lenti-X (Clontech) concentrated stocks. The average volume of injection was 500 uL of LPs diluted in DMEM.

(95) In all cases, thirty days after last immunization, mice were challenged with 5,000 GFP-expressing sporozoites micro-injected subcutaneously in the mice footpad.

(96) Sporozoite Challenge:

(97) Anopheles stephensi (Sda500 strain) mosquitoes were reared using standard procedures. 3-5 days after emergence, mosquitoes were fed on infected Swiss mice with a parasitemia superior to 2%, and kept as described in Amino et al, 2007. Between 20 and 23 days post-feeding, the salivary glands of infected mosquitoes were dissected in PBS, collected in 20 uL of sterile PBS on ice and disrupted using an eppendorf pestle. The suspension of parasites was filtered through a nylon mesh of 40 um, counted using Kova glasstic slide (Hycor) and adjusted to a concentration of 5,000 or 10,000 sporozoites/uL with sterile PBS. This suspension was divided in individual tubes, one for each group of immunized mice (n=4-7 per group), and kept on ice until the challenge. One microliter of parasite suspension was injected in the right footpad of mice using a Nanofil syringe (World Precision Instruments) with a 35 GA bevelled needle (NF35BV).

(98) Amino R, Thiberge S, Blazquez S, Baldacci P, Renaud O, et al. (2007) Imaging malaria sporozoites in the dermis of the mammalian host. Nat Protoc 2:1705-1712.

(99) Measurement of Parasite Infection:

(100) Hepatic parasite loads were quantified at ˜44 h by bioluminescence in fur shaved mice infected with GFP LUC parasites. Infected mice were first anesthetized with isoflurane and injected subcutaneously with D-luciferin (150 mg/kg, Caliper LifeSciences). After a 5 minutes incubation allowing the distribution of the substrate in the body of the anesthetized animals, mice were transferred to the stage of an intensified charge-coupled device photon-counting video camera box where anesthesia was maintained with 2.5% isoflurane delivered via nose cones. After 5 minutes of signal acquisition controlled by the Living Image software (Xenogen Corporation), animals were returned to their cage. Automated detection of bioluminescence signals by the system resulted in the generation of bioluminescence signal maps superimposed to the gray-scale photograph of the experimental mice. These images were then quantified using the Living Image software. Briefly, regions of interest (ROI) encompassing the liver were manually defined, applied to all animals and the average radiance within these ROIs was automatically calculated. Background signal was measured in the lower region of the abdomen, and the average values of background signal obtained.

(101) Alternatively, blood infection was assessed by flow cytometry using hsp70-GFP parasites. At day 4, 5, 6 and >10 post-challenge, a millimetric excision was performed in the tail of mice allowing the collection of a drop of blood that was readily diluted in 500 uL of PBS. This diluted blood was analyzed using a flow cytometer. 500,000 events were analyzed at day 4 post-challenge and 100,000 events in the subsequent days. Non-infected mice after day 10 were considered as sterile protected.

(102) Statistical Analysis:

(103) Parasitemia data from GFP immunized control were log transformed and pooled for the calculation of 95% tolerance of interval with 95% of certitude. For the immunization protocol of one dose this limit comprised the interval of the mean value±2.49 SD (mean=−0.3906, SD=0.3392, n=35). Similarly, for the immunization protocol of two doses this limit comprised the interval of the mean value±2.51 (mean=−0.3002, SD=0.3305, n=33). All mice with a log parasitemia inferior to the lower limit (mean−2.5 SD) were considered as significantly different from the control mice (P<0.05), and therefore considered as protected. In the protocol using two immunization doses, the difference in the numbers of protected mice, following the definition above, between the test group and the GFP control group was assessed using the Fisher's exact test. The average of the log parasitemia of the groups with significant differences in the Fisher's Test were compared to the GFP group using one-way ANOVA (Holm-Sidak's multiple comparison test).

REFERENCES

(104) 1. Worldwide web: who.int/mediacentre/factsheets/fs094/en/ 2. RTS,S Clinical Trials Partnership. Lancet. 2015; 4; 386(9988):31-45. 3. Moorthy V S, Ballou W R. Malar J. 2009; 8: 312. 4. Worldwide web: who.int/immunization/topics/malaria/vaccine_roadmap/en/ 5. Seder R A, et al. Science. 2013; 341:1359-65. 6. Amino R, Ménard R. Nature. 2012; 484(7395):S22-3 7. Kester K E, et al. 2014. pii: S0264-410X(14)00822-6. 8. Mishra S, et al. Vaccine. 2011; 29(43):7335-42. 9. Murphy S C, Kas A, Stone B C, Bevan M J. Proc Natl Acad Sci USA. 2013; 110(15):6055-60. 10. Hafalla J C, et al. PLoS Pathog. 2013; 9(5):e1003303. 11. Iglesias, M. C., et al. J Gene Med. 2006; 8, 265-274. 12. Firat, H., et al. J Gene Med. 2002; 4, 38-45. 13. Doolan D L, et al. Proc Natl Acad Sci USA. 2003; 100:9952-7. 14. Robson K J, et al. Nature. 1988; 335(6185):79-82. 15. Khusmith S, et al. Science. 1991; 252(5006):715-8. 16. Ewer K J, et al. Nat Commun. 2013; 4:2836. 17. Gantt S, et al. Infect Immun. 2000; 68(6):3667-73. 18. Khusmith S, Sedegah M, Hoffman S L. Infect Immun. 1994; 62(7):2979-83. 19. LaCrue A N, et al. Mol Biochem Parasitol. 2006; 148(2):199-209. 20. Rennenberg A et al. PLoS Pathog. 2010; 6(3):e1000825. 21. Boysen K E, Matuschewski K. MBio. 2013; 4(6):e00874-13. 22. Lehmann C, et al. PLoS Pathog. 2014; 10(8):e1004336. 23. Pei Y, et al. Cell Microbiol. 2013 September; 15(9):1508-26. 24. Eggleson K K, Duffin K L, Goldberg D E. J Biol Chem.; 274(45):32411-7. 25. Poh C M, Howland S W, Grotenbreg G M, Rénia L. Infect Immun. 2014; 82(11):4854-64. 26. Gilson P R, et al. Mol Cell Proteomics. 2006; 5(7):1286-99. 27. Offeddu V, Rauch M, Silvie O, Matuschewski K. Mol Biochem Parasitol. 2014; 193(2):101-9. 28. Ishino T, Chinzei Y, Yuda M. Cell Microbiol. 2005; 7(2):199-208. 29. Amino R, et al. Cell Host Microbe. 2008 Feb. 14; 3(2):88-96. 30. Tavares J, et al. J Exp Med. 2013; 210(5):905-15. 31. Speake C, et al. PLoS One. 2016; 11(7):e0159449.