Mycobacterial antigen vaccine
10357555 · 2019-07-23
Assignee
Inventors
- Emmanuel Tupin (Stockholm, SE)
- Romain Micol (Lyons, FR)
- Charles Antoine Coupet (Lyons, FR)
- Geneviève Inchaupse (Lyons, FR)
- Marie Gouanvic (Lyons, FR)
- Nathalie Silvestre (Ergersheim, FR)
- Jean-Baptiste Marchand (Obernai, FR)
- Cécile Beny (Cerisiers, FR)
Cpc classification
C07K2319/40
CHEMISTRY; METALLURGY
International classification
A61K39/00
HUMAN NECESSITIES
Abstract
The present invention relates generally to immunogenic combinations comprising at least five antigens of a Mycobacterium species as well as fusion thereof and nucleic acid molecules encoding such combined antigens and fusion. The present invention also relates to nucleic acid molecules, vectors, host cells and compositions comprising or encoding said combinations of mycobacterial antigens and fusion polypeptides as well as to methods for recombinantly producing them. The present invention also relates to methods of using said combinations of mycobacterial antigens, fusion polypeptides, vectors, host cells, compositions particularly for inducing or stimulating an immune response against a Mycobacterium infection or any disease caused by or associated with a Mycobacterium infection. The present invention also concerns antibodies directed to such mycobacterial antigens and fusion polypeptides that can be used in the diagnosis of a Mycobacterium infection and method of detection as well as kits of reagent comprising said combinations of mycobacterial antigens, fusion polypeptides, vectors, host cells, compositions or antibodies.
Claims
1. An immunogenic combination comprising a viral vector or a combination of two or more viral vectors comprising one or more nucleic acid molecules encoding at least 5 antigens, wherein said antigens are independently obtained from a Mycobacterium species and wherein said at least 5 mycobacterial antigens are selected from the group consisting of ESAT-6 (Rv3875), TB10.4 (Rv0288), Ag85B (Rv1886), RpfB, RpfD, Rv0111, Rv0569, Rv1733c, Rv1807, Rv1813, Rv2029c, Rv2626, Rv3407, and Rv3478.
2. The immunogenic combination according to claim 1, wherein the mycobacterial antigens are obtained from a Mycobacterium species of the tuberculosis complex selected from the group consisting of M. tuberculosis (Mtb), M. bovis, M. bovis BCG, M. africanum, M. canetti, M. caprae, and M. microti.
3. The immunogenic combination according to claim 1, wherein said immunogenic combination encodes mycobacterial antigens from at least 2 different infection phases selected from the group consisting of active, resuscitation, and latent phases.
4. The immunogenic combination according to claim 3, wherein said immunogenic combination is multiphasic encoding at least one antigen from the active infection phase, at least one antigen from the resuscitation infection phase, and at least one antigen from the latent infection phase.
5. The immunogenic combination according to claim 3, wherein said immunogenic combination expresses at least ESAT-6 (Rv3875), Ag85B (Rv1886), and TB10.4 (Rv0288).
6. The immunogenic combination according to claim 3, wherein said antigen(s) of the latent phase is/are selected from the group consisting of Rv0111, Rv0569, Rv1733, Rv1807, Rv1813, Rv2029, Rv2626, Rv3407, and Rv3478.
7. The immunogenic combination according to claim 3, wherein said immunogenic combination expresses at least RpfB and RpfD.
8. The immunogenic combination according to claim 1, wherein the at least 5 mycobacterial antigens are selected from polypeptides comprising an amino acid sequence at least 80% homologous or identical to any one of SEQ ID NO: 1-24.
9. The immunogenic combination according to claim 1, wherein said immunogenic combination encodes the mycobacterial antigens in the form of separate polypeptides or in the form of one or more fusion polypeptides or both in the form of separate antigen(s) and fusion(s) polypeptides.
10. A vector, or a combination of vectors, comprising one or more nucleic acid molecule(s) encoding a fusion polypeptide comprising at least two mycobacterial antigens selected from the group consisting of ESAT-6 (Rv3875), TB10.4 (Rv0288), Ag85B (Rv1886), RpfB, RpfD, Rv0111, Rv0569, Rv1733c, Rv1807, Rv1813, Rv2029c, Rv2626, Rv3407, and Rv3478.
11. The vector of claim 1 or claim 10, wherein said vector is a viral vector selected from the group consisting of retrovirus, adenovirus, adenovirus-associated virus (AAV), poxvirus, herpes virus, measles virus, foamy virus, alphavirus, and vesicular stomatis virus.
12. The vector of claim 11, wherein said vector is a E1-defective adenoviral vector or a poxvirus vector selected from the group consisting of fowlpox, canarypox, and vaccinia virus vector.
13. The vector of claim 12, wherein said vaccinia virus vector is selected from Copenhagen, Wyeth, NYVAC, and modified Ankara (MVA) strains.
14. The vector according to claim 10, which is in the form of infectious virus particles.
15. A composition comprising at least one of the immunogenic combinations of claim 1, the vector of claim 10, or any combination thereof.
16. The composition of claim 15, which further comprises a pharmaceutically acceptable vehicle.
Description
DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
MATERIALS AND METHODS
(16) Analytical Methods
(17) Existing data on Mtb antigens were investigated from the available literature and data bases with the goal of identifying a first selection of Mtb genes/antigens that may be useful in an immunotherapeutic vaccine capable of raising anti-TB immunity during all phases of the natural course of infection.
(18) The selected antigens were then submitted to a data-mining scoring system that was developed to transcribe and compare data from different sources. An overall final score was generated reflecting the value of each antigen. This score takes into account the immunogenic potential of the antigen as well as its capacity to protect against an infectious challenge in animal models and in humans (for example protection data in humans will be better scored than inducing immunogenicity in animal models). Once all data for a particular antigen were collected, a grade from 0 to 5 was attributed to each category, 0 being the worse possible grade while 5 being the best. The choice of the grade was also based on the quality of the data (e.g. right controls used in the experiments, rigorous interpretation) but also on the robustness of the data (e.g. number of times experiments were run, number of publications confirming/supporting the findings).
(19) Antigen Biochemical in Silico Analyses
(20) Biochemical and biological data are also key data for optimizing expression and fusion design permitting to anticipate potential expression problems. For example, the biological functions of a protein may lead to a potential toxicity resulting in genetic unstability and/or safety profile upon vector-mediated expression. Moreover, protein unfolding may impact stability and expression levels due to a higher cellular degradation rate.
(21) An extensive bibliographic search was carried out for all Mtb antigens in order to better understand and characterize the structure and the functions of these proteins.
(22) Additionally, biochemical and bioinformatic predictions were also performed for characterization of Mtb antigens. Bioinformatic prediction tools (Nielsen et al., 2007 PLos One 2: e796; Nielsen et al., 2008, PLoS Comput Biol 4: e1000107) were used to look at predicted epitopes for class I and II HLA molecules. Identification of these epitopic regions may be useful for optimization of the selected Mtb antigens or to facilitate the development of immune based assays.
(23) Moreover, extensive in silico structure prediction analyses were performed in order to predict biochemical properties and/or biological functions and thus allow Mtb antigens selection and design (e.g. whether a full length native form is likely to be expressed or whether modifications appear to be required).
(24) More specifically Search for structural homologs in protein data bank (PDB). The program used was BLASTP with the default parameters and the selected database was NPS@ 3D SEQUENCES (from PDB). This search allows finding NMR or crystalline structures of the antigen or of proteins with sequence homologies higher than 25%. 3D structures were visualized using CN3D 4.1. or PDB viewer. Search in UNIPROT-SWISSPROT and TB databases. Protein homologs to the Mtb antigens of interest were searched on UNIPROT-SWISSPROT database using primary structures as a query and BLAST search on NPS@. The program used was BLASTP with the default parameters and the selected database was UNIPROT-SWISSPROT; UNIPROT-SWISSPROT entries give access to general information on protein function, domains, potential signal peptide, posttranslational modifications as well as bibliographic references. General information (e.g. gene functions, genetic link between genes, phenotypes and mutations associated with genes, immunogenicity, as well as bibliographic references) about the selected Mtb antigens were retrieved using the Rv protein name as a query in the TB database. Prediction of signal peptides: these short N-terminus sequences are often predicted as transmembrane domains but are not present in native and mature proteins. The presence of signal peptide was notified in the UNIPROT-SWISSPROT database for certain antigens or using the hidden Markov model of signal v3.0 algorithm. If no hit for homolog search, additional searches were undertaken: Prediction of potential transmembrane domains TM using three different programs (e.g. dense Alignment surface (DAS) method, Algorithm TMHMM and TopPred0.01). It is our observations that the presence of such hydrophobic TM domains with may impair genetic stability of the corresponding antigen when expressed in viral vectors such as MVA. Search for known protein motif associated with protein domains, families and functional sites using PROSITE SCAN Prediction of secondary structures using several prediction methods (namely: SOPM, MLRC, HNN, DSC, PHD, PREDATOR). A secondary structure was considered as highly probable if the 6 methods predicted it. A secondary structure was considered possible if 3 out of the 6 methods predicted it. Hydrophobic cluster analysis (HCA). The HCA method is based on essential features of protein folding: the hydrophilic/hydrophobic dichotomy and the hydrophobic compactness of protein globular domains. HCA plots were used to identify hydrophobic clusters along the protein sequence. These clusters are characteristic of folded proteins with a hydrophobic buried core. An antigen was considered as probably having a folded state if hydrophobic clusters were present at least at some part of the protein. Prediction of natively disordered regions that permit to identify unfolded regions (MetaPrDOS predictions). This analysis complements the HCA plot by predicting areas of the protein that are not folded. All areas above the threshold 0.5 (disordered tendency) were considered as unfolded parts of the protein. To be noted that most of the time, N- and C-terminus parts of proteins are non-folded in their native state. If such stretches at the extremities were present and smaller than 10 residues, they were kept as potential linkers for fusion. Prediction of coiled coil (using the COILS program). Oligomerization state of the antigen could impact the antigen fusion design. A coiled coil domain was predicted if the output probability displayed a value of 1 for a part of the protein with at least 14 residues window analysis.
(25) Further, sequence alignments were carried out to verify that the selected Mtb antigens are conserved among different Mtb strains and isolates. More precisely, multiple sequence alignments were performed using Clustal W2 (@.ebi.ac.uk/Tools/msa/clustalw2/) between the amino acid sequence of each selected antigen (the exemplified Mtb antigens originate from the H37Rv strain) and their equivalent of 11 other Mtb strains (clinical isolates) and M bovis that have been identified in protein databases (BLASTP search). As a result, the TB antigens showed high conservations among the 12 Mycobacteria strains analyzed with a percentage of identity ranging from 100% to 96% depending on the antigen and the Mycobacterium strain. The major exception was seen with Rv3478 for which only 88% identity was found between H37Rv and CDC1551 sequences.
(26) Finally, another key criteria to reach final TB antigen selection was to ensure a balanced representativeness of antigens from the various phases of infection. For example, some latent antigens were selected despite lower final data mining score than most of the active phase antigens.
(27) Construction of Fusion of Mtb Antigens
(28) 12 fusions of Mtb antigens were engineered as illustrated in
(29) On the other hand, signal peptides (also called signal sequence or SS) and membrane-anchoring peptides (also called trans-membrane or TM peptide/domain) were added respectively at the N-terminus and C-terminus of the Mtb fusion proteins to ensure anchorage at the cell surface which is assumed to optimize immunogenic activity in certain cases. However, addition of TM domain was not necessary for fusions ending with Rv0111 or Rv1733, as these proteins already contain membrane-anchoring peptides. For comparative purposes, four fusions were also engineered without any signal sequence (SS) and TM domain so as to study the influence of cell location (membrane presentation in the presence of SS and TM peptides versus cytoplasmic location in the absence of such peptides) on expression level and immunogenic activity. For example, pTG18269 encodes the same Mtb antigens (Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807) as pTG18295 except that the pTG18269-encoded fusion is equipped with a SS at its N-terminus and a TMat its C terminus between Myc and His tags whereas the pTG18295-encoded fusion is devoid of such SS and TM peptides.
(30) Synthetic genes coding for the different Mtb antigens and fusions were synthesized by Geneart (Regensburg, Germany). The sequences were optimized for human codon usage and a Kozak sequence (ACC) was added before the ATG starting codon. Moreover some motives were excluded: TTTTTNT, GGGGG, CCCCC which are deleterious for expression in poxvirus vector and AAAGGG, AAAAGG, GGGAAA, GGGGAA, (and complementary sequences TTCCCC, TTTCCC, CCTTTT, CCCCTT) which can be deleterious for expression in some others vectors.
(31) The fusions were cloned in pGWiz plasmid (Gelantis) digested by NotI and BamH. This plasmid contains a modified CMV promoter, followed by intron A from the CMV immediate early gene, and a high-efficiency artificial transcription terminator.
(32) Construction of pTG18266 (Fusion No 2)
(33) The amino acid sequence of the fusion no 2 is shown in SEQ ID NO: 28. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus ERA strain (described in Genbank no M38452), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 317 correspond to Ag85B*, amino acids 318 to 412 correspond to TB10.4, amino acids 413 to 506 correspond to ESAT6, amino acids 507 to 516 correspond to the c-myc TAG, a Ser linker, amino acids 518 to 583 correspond to the membrane-anchoring peptide derived from the rabies glycoprotein of ERA strain and amino acids 584 to 589 correspond to the His TAG. The fusion no 2-encoding nucleotide sequence shown in SEQ ID NO: 40 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18266.
(34) Construction of pTG18267 (Fusion No 3)
(35) The amino acid sequence of the fusion no 3 is shown in SEQ ID NO: 30. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus PG strain (described in Genbank no ay009097 and SEQ ID NO: 2 in WO2008/138649), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 380 correspond to RPFB-Dhyb*, amino acids 381 to 390 correspond to the c-myc TAG, a Ser linker, amino acids 392 to 457 correspond to the membrane-anchoring peptide derived from the rabies glycoprotein of PG strain (SEQ ID NO: 3 in WO2008/138649) and amino acids 458 to 463 correspond to the His TAG. The fusion no 3-encoding nucleotide sequence shown in SEQ ID NO: 42 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by Nod and BamH1 to give pTG18267.
(36) Construction of pTG18268 (Fusion No 4)
(37) The amino acid sequence of the fusion no 4 is shown in SEQ ID NO: 32. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus PG strain (described in Genbank no ay009097), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 380 correspond to RPFB-Dhyb*, amino acids 381 to 666 correspond to Ag85B*, amino acids 667 to 761 correspond to TB10.4, amino acids 762 to 855 correspond to ESAT6, amino acids 856 to 865 correspond to the c-myc TAG, a Ser linker, amino acids 867 to 932 correspond to the membrane-anchoring peptide derived from the rabies glycoprotein of PG strain and amino acids 933 to 938 correspond to the His TAG. The fusion no 4-encoding nucleotide sequence shown in SEQ ID NO: 44 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18268.
(38) Construction of pTG18269 (Fusion No 5)
(39) The amino acid sequence of the fusion no 5 is shown in SEQ ID NO: 34. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus ERA strain (described in Genbank no M38452), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 118 correspond to Rv0569, amino acids 119 to 227 correspond to Rv1813*, amino acids 228 to 325 correspond to Rv3407, amino acids 326 to 717 correspond to Rv3478, amino acids 718 to 1115 correspond to Rv1807, amino acids 1116 to 1125 correspond to the c-myc TAG, a Ser linker, amino acids 1127 to 1192 correspond to the membrane-anchoring peptide derived from the rabies glycoprotein of PG strain (SEQ ID NO: 3 in WO2008/138649) and amino acids 843 to 848 correspond to the His TAG. The fusion no 5-encoding nucleotide sequence shown in SEQ ID NO: 46 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18269.
(40) Construction of pTG18270 (Fusion No 6)
(41) The amino acid sequence of the fusion no 6 is shown in SEQ ID NO: 36. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus ERA strain (described in Genbank no M38452), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 317 correspond to Ag85B*, amino acids 318 to 459 correspond to Rv2626, amino acids 460 to 808 correspond to RPFB-Dhyb*, amino acids 809 to 956 correspond to Rv1733*, amino acids 957 to 966 correspond to the c-myc TAG, a Ser linker, and amino acids 968 to 973 correspond to the His TAG. The fusion no 6-encoding nucleotide sequence shown in SEQ ID NO: 48 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18270.
(42) Construction of pTG18272 (Fusion No 8)
(43) The amino acid sequence of the fusion no 8 is shown in SEQ ID NO: 37. Amino acids 1 to 23 correspond to the signal peptide present at the N-terminus of the glycoprotein precursor of rabies virus ERA strain (described in Genbank no M38452), amino acids 24 to 31 correspond to the Flag TAG, amino acids 32 to 317 correspond to Ag85B*, amino acids 318 to 459 correspond to Rv2626, amino acids 460 to 607 correspond to Rv1733*, amino acids 608 to 617 correspond to the c-myc TAG, a Ser linker and amino acids 619 to 624 correspond to the His TAG. The fusion no 8-encoding nucleotide sequence shown in SEQ ID NO: 49 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18272.
(44) Construction of pTG18323 (Fusion No 13)
(45) The amino acid sequence of the fusion no 13 is shown in SEQ ID NO: 38. Amino acids 1 to 28 correspond to the signal peptide present at the N-terminus of the F protein of measles virus (Hall strain, described in Genbank no X05597-1), amino acids 29 to 36 correspond to the Flag TAG, amino acids 37 to 349 correspond to Rv2029*, amino acids 350 to 491 correspond to Rv2626, amino acids 492 to 639 correspond to Rv1733*, amino acids 640 to 932 correspond to Rv0111*, amino acids 933 to 942 correspond to the c-myc TAG, a Ser linker and amino acids 944 to 949 correspond to the His TAG. The fusion no 13-encoding nucleotide sequence shown in SEQ ID NO: 50 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18323.
(46) Construction of pTG18324 (Fusion No 14)
(47) The amino acid sequence of the fusion no 14 is shown in SEQ ID NO: 39. Amino acids 1 to 28 correspond to the signal peptide present at the N-terminus of the F protein of measles virus (Hall strain, described in Genbank no X05597-1), amino acids 29 to 36 correspond to the Flag TAG, amino acids 37 to 349 correspond to Rv2029*, amino acids 350 to 444 correspond to TB10.4, amino acids 445 to 538 correspond to ESAT6, amino acids 539 to 831 correspond to Rv0111*, amino acids 832 to 841 correspond to the c-myc TAG, a Ser linker and amino acids 843 to 848 correspond to the His TAG. The fusion no 14-encoding nucleotide sequence shown in SEQ ID NO: 51 was generated by synthetic way and the synthetic gene was cloned in pGWiz restricted by NotI and BamH1 to give pTG18324.
(48) Construction of Fusions 9-12
(49) The targeting sequences were deleted from plasmids pTG18267, pTG18269, pTG18266 and pTG18268 by directed mutagenesis (Quick Change Site-Directed mutagenesis kit, Stratagene) using appropriate pairs of primers, OTG20188 (CGCGGCCGCACCATGGATTACAAGGATGACGACG; SEQ ID NO: 52) and OTG20189 (CGTCGTCATCCTTGTAATCCATGGTGCGGCCGCG; SEQ ID NO: 53) for deleting signal peptide sequence and OTG20190 (CATCTCAGAAGAGGATCTG-CATCATCATCATCATCATTG; SEQ ID NO: 54) and OTG20191 (CAATGATGATGAT-GATGATGCAGATCCTCTTCTGAGATG; SEQ ID NO: 55) for deleting TM sequence. The resulting plasmids were respectively pTG18307 (fusion no 12=cytoplasmic fusion no 3), pTG18295 (fusion no 9=cytoplasmic fusion no 5), pTG18296 (fusion no 10=cytoplasmic fusion no 2) and pTG18297 (fusion no 11=cytoplasmic fusion no 4), corresponding to amino acid sequences SEQ ID NO: 31, 35, 29 and 33 encoded by the nucleotide sequences SEQ ID NO: 43, 47, 41 and 45.
(50) Construction of Individual Mtb Gene Expression Plasmids
(51) The Flag sequence and c-myc-His sequences separated by a NheI restriction site were introduced downstream the CMV promoter in pGWiz plasmid. A synthetic DNA fragment containing the end of CMV promoter, Flag and c-myc-His sequences was synthesized by Geneart and inserted into the plasmid FLAG_TAG_1. This plasmid was digested by PvuII and BgIII and the resulting fragment was inserted in pGWiz restricted by the same enzyme, giving rise to pTG18282. The individual Rv3407, Rv0569, Rv1807, Rv1813*, Rv3478 and Rv2626 genes were then amplified by PCR from pTG18269 except Rv2626 for which the pTG18323 was used as template.
(52) The amplification primer pairs used for isolation of each TB gene are illustrated in Table 1.
(53) TABLE-US-00001 TABLE1 TBgene Primername Primersequence Rv3407 OTG20232 GATGACGACGATAAGGCTAGCA SEQIDNO:56 GAGCCACCGTGGGACTGG OTG20233 GATGAGTTTTTGTTCGCTAGCC SEQIDNO57 TGTTCATCCCGCATCTCGT Rv0569 OTG20234 GATGACGACGATAAGGCTAGCA SEQIDNO:58 AGGCCAAAGTCGGCG OTG20235 GATGAGTTTTTGTTCGCTAGCT SEQIDNO:59 GTTCCTCTGGCGTGC Rv1807 OTG20236 GATGACGACGATAAGGCTAGCG SEQIDNO:60 ATTTTGCCACCCTCCCACC OTG20237 GAGATGAGTTTTTGTTCGCTAG SEQIDNO:61 CGCCAGCTGCAGGAGGTCTGG Rv1813* OTG20238 GATGACGACGATAAGGCTAGCG SEQIDNO:62 CCAACGGCAGCATGAGCG OTG20239 GAGATGAGTTTTTGTTCGCTAG SEQIDNO:63 CGTTGCAGGCCCAGTTCACGA Rv3478 OTG20240 GATGACGACGATAAGGCTAGCG SEQIDNO:64 TGGACTTCGGCGCCCTGC OTG20241 GAGATGAGTTTTTGTTCGCTAG SEQIDNO:65 CGCCAGCGGCTGGAGTTCTGG Rv2626 OTG20242 GATGACGACGATAAGGCTAGCA SEQIDNO:66 CAACCGCCAGAGACATCATG OTG20243 GATGAGTTTTTGTTCGCTAGCA SEQIDNO:67 GAGGCCAGGGCCATGGGG
(54) The resulting amplicons were cloned by In fusion Advantage PCR cloning method (Clontech) in pTG18282 linearized by NheI. This allows the fusion of Tag sequences with Mtb genes. The generated plasmids were named respectively pTG18300 (Rv3407), pTG18301 (Rv0569), pTG18302 (Rv1807), pTG18303 (Rv1813*), pTG18304 (Rv3478) and pTG18305 (Rv2626).
(55) Six plasmids containing expression cassettes for ESAT6, Rv1733*, Ag85B*, TB10-4, Rv0111* and Rv2029* fused to Flag in 5 and c-myc-His sequences in 3 were synthesized by Geneart and inserted in pGWiz. They were named respectively pTG18308 (ESAT6), pTG18309 (Rv1733*), pTG18310 (Ag85B*), pTG18315 (TB10.4), pTG18329 (Rv0111*), pTG18317 (Rv2029*). As Rv1733* and Rv0111* proteins contain a TM domain, the signal peptide presents at the N-terminus of the glycoprotein precursor of rabies virus ERA strain was fused upstream to the Flag sequence to avoid expression issues.
(56) Whether encoding individual or fused Mtb genes, plasmids used for immunization were produced in endotoxin-free conditions.
(57) Construction of Recombinant MVA
(58) Deletion of TAG Sequences
(59) TAG sequences were removed from the Mtb antigen fusions to avoid their presence in the MVA vectors. TAG sequences located inside the Mtb fusion cassettes (i.e. Flag present between the signal peptide and the first amino acid of the Mtb fusion and cmyc TAG present between the last amino acid of the Mtb fusion and membrane-anchoring peptide) were deleted by directed mutagenesis using the QuikChange Site-directed Mutagenesis kit (Stratagene) and appropriate primers pairs as illustrated in the following Table 2. TAG sequences located outside the Mtb fusion cassettes (for cytoplasmic fusion and His TAG) were deleted by PCR using primers allowing the addition of an initiator Met and a terminator codon on both extremity of the fusions.
(60) TABLE-US-00002 TABLE 2 Resulting Fusion Primer pairs for deletion of Flag Primer pairs for deletion of cmyc plasmid 4 OTG20313 (SEQ ID NO: 68) OTG20315 (SEQ ID NO: 70) pTG18339 OTG20314 (SEQ ID NO: 69) OTG20316 (SEQ ID NO: 71) 5 OTG20317 (SEQ ID NO: 72) OTG20319 (SEQ ID NO: 74) pTG18340 OTG20318 (SEQ ID NO: 73) OTG20320 (SEQ ID NO: 75) 6 OTG20321 (SEQ ID NO: 76) NA pTG18341 OTG20322 (SEQ ID NO: 77) 13 OTG20333 (SEQ ID NO: 78) NA pTG18342 OTG20334 (SEQ ID NO: 79) 14 OTG20333 (SEQ ID NO: 78) NA pTG18343 OTG20334 (SEQ ID NO: 79)
(61) Construction of MVATG18355 (Fusion No 13)
(62) The nucleotide sequence encoding fusion no 13 (SF-Rv2029*-Rv2626-Rv1733*-Rv0111* as illustrated by the portion of SEQ ID NO: 38 from 1 to 28 and 37 to 932) was placed under the control of the p7.5K promoter (SEQ ID NO: 80; CCACCCACTTTTTATAGTAAGTTTTTCACCCATAAATAATAAATACAATAATTAA TTTCTCGTAAAAGTAGAAAATATATTCTAATTTATTGCACGGTAAGGAAGTAGA ATCATAAAGAACAGT). This latter was amplified by PCR from VV (Vaccinia virus) Copenhagen strain DNA using a pair of appropriate primers OTG20405 (SEQ ID NO: 81) and OTG20406 (SEQ ID NO: 82) while the fusion no 13 sequence was amplified from plasmid pTG18342 by PCR with OTG20407 (SEQ ID NO:83) and OTG20408 (SEQ ID NO: 84). Then p7.5K and fusion no 13-encoding sequence were reassembled by double PCR using the primers OTG20405 (SEQ ID NO: 81) and OTG20408 (SEQ ID NO: 84). The resulting fragment was inserted into the BgIII and NotI restriction sites of a vaccinia transfer plasmid, pTG17960, resulting in pTG18355.
(63) The MVA transfer plasmid, pTG17960, is designed to permit insertion of the nucleotide sequence to be transferred by homologous recombination in deletion III of the MVA genome. It originates from the plasmid pTG1E (described in Braun et al., 2000, Gene Ther. 7:1447) into which were cloned the flanking sequences (BRG3 and BRD3) surrounding the MVA deletion III (Sutter and Moss, 1992, Proc. Natl. Acad. Sci. USA 89:10847). The transfer plasmid also contains a fusion between the Aequorea victoria enhanced Green Fluorescent Protein (eGFP gene, isolated from pEGP-C1, Clontech) and the Escherichia coli xanthine-guanine phosphoribosyltransferase gene (gpt gene) under the control of the early late vaccinia virus synthetic promoter p11K7.5 (kindly provided by R. Wittek, University of Lausanne). Synthesis of xanthine-guanine phosphoribosyltransferase enables GPT.sup.+ recombinant MVA to grow in a selective medium containing mycophenolic acid, xanthine, and hypoxanthine (Falkner et al, 1988, J. Virol. 62, 1849-54) and eGFP enables the visualisation of recombinant MVA plaques. The selection marker eGFP-GPT is placed between two homologous sequences in the same orientation. After clonal selection, the selection marker can be easily eliminated by several passages without selection allowing the growth of eGFP-GPT recombinant MVA.
(64) Generation of MVATG18355 was performed by homologous recombination in primary chicken embryos fibroblasts (CEF) infected with MVA and transfected by nucleofection with pTG18355 (according to Amaxa Nucleofector technology). Viral selection was performed by plaque purification after growth in the presence of a selective medium containing mycophenolic acid, xanthine and hypoxanthine. As mentioned above, the selection marker was then eliminated by passage in a non-selective medium. Absence of contamination by parental MVA was verified by PCR.
(65) Construction of MVATG18364 (Fusion No 13+Fusion No 4)
(66) The nucleotide sequence encoding fusion no 4 (SR-RPFB-Dhyb*-Ag85B*-TB10.4-ESAT6-TMR as illustrated by the portion of SEQ ID NO: 32 from 1 to 23 followed by 32 to 855 and 866 to 932) was placed under the control of pH5R promoter (SEQ ID NO: 85, TTTATTCTATACTTAAAAAATGAAAATAAATACAAAGGTTCTTGAGGGTTGTGTT AAATTGAAAGCGAGAAATAATCATAAATTATTTCATTATCGCGATATCCGTTAA GTTTG) cloned from genomic DNA of wild type MVA by PCR with primer pair OTG20445 (SEQ ID NO: 86) and OTG20446 (SEQ ID NO: 87). The amplified product was digested by NotI and PacI. The fusion no 4-encoding sequence was amplified from pTG18339 by PCR using OTG20447 (SEQ ID NO: 88) and OTG20380 (SEQ ID NO: 89) primers. The amplified product was digested by PacI and XhoI. Both fragments were cloned together into pTG18355 restricted by NotI and XhoI, resulting in pTG18364.
(67) Generation of MVATG18364 virus was performed in CEF by homologous recombination as described above.
(68) Construction of MVATG18365 (Fusion No 13+Fusion No 11)
(69) The nucleotide sequence encoding fusion no 11 (RPFB-Dhyb*-Ag85B*-TB10.4-ESAT6 as illustrated by the portion of SEQ ID NO: 33 from position 10 to position 833 preceded with the Met initiator in position 1) was placed under the control of pH5R promoter. The promoter was obtained from pTG18364 by PCR with OTG20445 (SEQ ID NO: 86) and OTG20446 (SEQ ID NO: 87) primers and the amplified fragment digested by NotI and PacI. The fusion no 11-encoding sequence was cloned from pTG18297 by PCR using primer pair OTG20448 (SEQ ID NO: 90) and OTG20382 (SEQ ID NO: 91) and the amplified product digested by PacI and XhoI. Both fragments were cloned together into pTG18355 restricted by Nod and XhoI to give pTG18365.
(70) Generation of MVATG18365 virus was performed in CEF by homologous recombination as described above.
(71) Construction of MVATG18376 (Fusion No 13+Fusion No 4+Fusion No 5)
(72) The nucleotide sequence encoding the fusion no 5 (SR-Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807-TMR as illustrated by the portion of SEQ ID NO: 34 from positions 1 to 23 followed by 32 to 1115 and 1126 to 1192) was placed under the control of the B2R promoter (SEQ ID NO: 92, TATATTATTAAGTGTGGTGTTTGGTCGATGTAAAATTT-TTGTCGATAAAAATTAAAAAATAACTTAATTTATTATTGATCTCGTGTGTACAAC CGAAATC). The promoter was amplified from VV Western Reserve strain DNA by PCR using primer pair OTG20469 (SEQ ID NO: 93) and OTG20470 (SEQ ID NO: 94) and the amplified fragment was digested by XhoI and NheI. The fusion no 5-encoding sequence was amplified from pTG18340 using primer pair OTG20472 (SEQ ID NO: 95) and OTG20473 (SEQ ID NO: 96) before being restricted by NheI and BamHI. Both digested fragments were cloned together into pTG18364 linearized by XhoI and BamHI to generate pTG18376.
(73) Generation of MVATG18376 virus was performed in CEF by homologous recombination as described above.
(74) Construction of MVATG18377 (Fusion No 13+Fusion No 11+Fusion No 5)
(75) The B2R promoter was amplified from pTG18376 using primer pair OTG20469 and OTG20470 described above and digested by XhoI and NheI. The nucleotide sequence encoding the fusion no 5 (SR-Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807-TMR) was amplified as described above and cloned under the control of the B2R promoter into pTG18364 linearized by XhoI and BamHI to generate pTG18377.
(76) Generation of MVATG18377 virus was performed in CEF by homologous recombination as described above.
(77) Construction of MVATG18378 (Fusion No 13+Fusion No 4+Fusion No 9)
(78) The nucleotide sequence encoding the fusion no 9 (Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807 as illustrated by the portion of SEQ ID NO: 35 from positions 10 to 1093 preceded with the Met initiator in position 1) was amplified from pTG18295 by PCR using primer pair OTG20483 (SEQ ID NO: 97) and OTG20474 (SEQ ID NO: 98). The amplified product was digested by NheI and BamHI and cloned with the XhoI and NheI-restricted B2R promoter (amplified from pTG18376 as described above) into pTG18364 linearized by XhoI and BamHI, resulting in pTG18378.
(79) Generation of MVATG18378 virus was performed in CEF by homologous recombination as described above.
(80) Construction of MVATG18379 (Fusion No 13+Fusion No 11+Fusion No 9)
(81) The nucleotide sequence encoding the fusion no 9 (Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807) and the B2R promoter were both amplified as described above and cloned together into pTG18365 linearized by XhoI and BamHI, resulting in pTG18378.
(82) Generation of MVATG18379 virus was performed in CEF by homologous recombination as described above.
(83) Construction of MVATG18404 (Fusion No 14+Fusion No 6)
(84) The nucleotide sequence encoding the fusion no 14 (SF-Rv2029*-TB10.4-ESAT6-Rv0111* as illustrated by the portion of SEQ ID NO: 39 from positions 1 to 28 and 37 to 831) was amplified from pTG18343 by PCR using primer pair OTG20407 (SEQ ID NO: 83) and OTG20525 (SEQ ID NO: 99). The p7.5K promoter was obtained from pTG18355 by PCR with OTG20524 (SEQ ID NO: 100) and OTG20406 (SEQ ID NO: 82) primers. The fusion no 14-encoding sequence was then cloned under the control of the p7.5K promoter by double PCR using OTG20524 (SEQ ID NO: 100) and OTG20525 (SEQ ID NO: 99). The resulting fragment was restricted with BamHI and NotI and inserted into the BgIII and NotI restriction sites of the vaccinia transfer plasmid, pTG17960, resulting in pTG18395.
(85) The nucleotide sequence encoding the fusion no 6 (SS-Ag85B*-Rv2626-RPFB-Dhyb*-Rv1733* as illustrated by the portion of SEQ ID NO: 36 from positions 1 to 23 and 32 to 956) was amplified from pTG18341 by PCR using primer pair OTG20527 (SEQ ID NO. 101) and OTG20376 (SEQ ID NO: 102) and the amplification product was digested with PacI and XhoI. The pH5R promoter was amplified from pTG18355 as described above and digested by NotI and PacI. Both digested fragments were cloned together into pTG18395 linearized by NotI and XhoI, resulting in plasmid pTG18404.
(86) Generation of MVATG18404 virus was performed in CEF by homologous recombination as described above.
(87) Construction of MVATG18417 (Fusion No 14+Fusion No 6+Fusion No 5)
(88) The nucleotide sequence encoding the fusion no 5 (SR-Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807-TMR) placed under the control of B2R promoter was obtained by digestion of pTG18376 with XhoI and BamHI. The resulting fragment was inserted in pTG18404 restricted by the same enzymes, giving rise to pTG18417.
(89) Generation of MVATG18417 virus was performed in CEF by homologous recombination as described above.
(90) Construction of MVATG18418 (Fusion No 14+Fusion No 6+Fusion No 9)
(91) The nucleotide sequence encoding the fusion no 9 (Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807) placed under the control of B2R promoter was obtained by digestion of pTG18379 with XhoI and BamHI. The resulting fragment was inserted in pTG18404 restricted by the same enzymes, giving rise to pTG18418.
(92) Generation of MVATG18418 virus was performed in CEF by homologous recombination as described above.
(93) Production and Protein Purification
(94) Four E. coli strains have been tested for the expression of the individual Mtb antigens. All the strains carry the DE3 prophage in their genome that allows the induction of expression of T7 polymerase by lactose or analogue of lactose (i.e. IPTG). The four strains were B121(DE3) (Lucigen) as a classic strain for protein expression, C41(DE3) (Lucigen) for the expression of toxic protein, B121(DE3) Rosetta (Merck Chemical) for expression of protein with a codon usage that is different of the E. coli one, and C43(DE3) (Lucigen) for the expression of protein with trans-membrane peptides (e.g. Rv1733). Moreover, three different temperatures and production time were tested for optimizing antigen production.
(95) Expression Assays for Determining Optimal Conditions
(96) Each E. coli strain was transformed with the plasmid encoding the Mtb antigen to be produced. Five colonies were isolated from a freshly transformed plate, inoculated in 50 ml of LB (Luria Broth) medium in the presence of ampicillin and allowed to grow overnight at 37 C. under shaking A flask of autoinducible medium (AI medium containing glucose/lactose and antibiotic; Studier, 2005, Protein Expr Purif. 41: 207-34) was inoculated with preculture specimen and was then cultured at either 18 C., 30 C. and 37 C. for 24, 8 and 8 hours, respectively. At the end of incubation, the absorbance at 600 nm was measured and the cells were harvested by centrifugation. The cell pellet was resuspended in PBS and the OD 600 nm adjusted around 50 for each culture condition tested before lysing the cells by sonication. The cell lysate was then centrifuged at 10,000 g for 10 minutes at 4 C. and a specimen (typically 10 L) of the supernatant and the pellet were then loaded on a SDS-PAGE to estimate optimal conditions.
(97) Production and Purification of Mtb Antigens
(98) Purification of His tag-containing Mtb antigens was undertaken from 500 mL culture grown in 2 L flasks applying the optimal conditions determined previously. The cells were harvested by centrifugation and pellets corresponding to 250 mL of culture were kept at 20 C. until use. The harvested bacteria were resuspended in PBS or in guanidine depending of the solubility of the antigen, submitted to sonication for cell lysis and purified by IMAC affinity chromatography on Ni sepharose 6 fast Flow resin (GE Healthcare; reference 17-5318) either in native or denaturing conditions according to the provider's recommendations. Proteins were eluted by applying increasing concentrations of Imidazole (50 mM, 100 mM and 250 mM). Fractions containing the pure protein were pooled and dialysed against PBS or Urea depending of the solubility of the antigen.
(99) Protein Characterization
(100) A variety of tests can be performed to estimate the quantity and quality of the purified Mtb antigens present in the eluted fractions.
(101) Endotoxin levels were measured using Portable Test System (PTS) from Charles River Laboratories. Cartridges with a range of detection of 0.005 to 0.5 EU/mL were used according to the manufacturer's recommendations.
(102) Protein concentrations were determined by Bradford assay (Bioroad) according to the manufacturer's recommendations. Bovine serum albumin (BSA) diluted in the sample buffer was used as a standard.
(103) Purity of the eluted fractions and dialysed solution can be evaluated by electrophoresis on SDS-PAGE (4-12% Invitrogen).
(104) Mass of the purified proteins was measured using MALDI (Matrix-Assisted Laser Desorption/ionization) or electrospray methods. Measured and calculated masses were compared in order to determine if the protein is intact or not. Identity of the protein either in solution or in a band of gel was checked by mass measurement of peptides generated after trypsin digestion. Masses of peptides were determined by MALDI and/or liquid chromatography coupled to tandem mass Spectrometry (LC/MS/MS). Measured and calculated masses of peptides were compared in order to verify the identity of the protein.
(105) Production of Antibodies Against Mtb Antigens
(106) Antibodies directed against the various Mtb antigens were produced following immunization of rabbits with a mixture of two different antigen-specific peptides (Eurogentec; Seraing, Belgium). Such peptides of 15 or 16 amino acid residues were selected after running epitope B prediction programs. Antisera against Rv1733*, Rv2029*, Rv0569, Rv1807, Rv0111, RPFB-Dhyb*, Rv1813* and Rv3407 antigens were generated following rabbits immunization with the two specific peptides at day 0 and three boosts at day 7, 10 and 18. Blood samples were taken before first peptide injection and at day 21. Final bleeding of rabbits was done at day 29. For Rv3478, the rabbits were injected at day 0, 22, 49 and 77 with the two specific 16 mer peptides. Blood samples were taken before first peptide injection and at day 31 and 59. Final bleeding of rabbits was done at day 87.
(107) The final sera were evaluated by ELISA using the specific peptides and by Western-blot analysis using the individual Mtb gene expression plasmids.
(108) In Vitro Testing of the Mtb Fusion Proteins
(109) Western Blot on DNA-Mediated Expression Products
(110) 210.sup.6 HEK293 cells were transfected with 5 g of the various plasmids encoding Mtb antigen fusions or individual genes using Lipofectamine 2000 (Invitrogen; #11668-019) in presence of proteasome inhibitor MG132 (10 M) added to growth medium 18 h after transfection. pGWIZ plasmid was used as negative control After 48 hours medium was discarded and cells were lysed with 450 L/dish of Tris-Glycin-SDS 2+ buffer (ref: LC2676; Novex) supplemented with -mercaptoethanol (5% v:v). The lysate was then sonicated and boiled for 5 min at 95 C. Thirty microliters of cell lysates were submitted to electrophoresis onto precasted 10% Criterion gel using the Criterion Precast gel system (Biorad). Following electrophoresis, proteins were transferred onto a PVDF membrane (Macherey Nagel, 741260). Immunodetection was performed with 1/500 diluted monoclonal anti-Flag M2 peroxydase (HRP) antibody (Sigma; #A8592) or with 1/5000 diluted monoclonal anti-His peroxydase antibody (Invitrogen; #R931-25). Immune-complexes were revealed using the ImmunStar WesternC kit (Biorad, ref 170.5070).
(111) Sera (diluted 1/1000) obtained after immunization of rabbit, as described above, were also used for Western Blot detection of Rv1733*, Rv2029*, Rv0569, Rv1807, Rv0111*, Rpf-B-D, Rv1813*, Rv3407 and Rv3478. Commercial antibodies were used for detecting ESAT6, Ag85B*, TB10.4 and Rv2626, respectively, mouse monoclonal antibody HYB076-08 (Santa-Cruz; #sc-57730, diluted 1/500) for ESAT6, rabbit polyclonal anti-serum NR-13800 (BEI, diluted 1/5000) for Ag85B*, mouse monoclonal antibody 26A11 (Lifespan-Biosciences; #LS-C91052 diluted 1/1000) for Rv2626 and polyclonal rabbit antibody ABIN361292 (Antibodies-online, diluted 1/1000) for TB10.4.
(112) Western Blot on MVA-Mediated Expression Products
(113) 10.sup.6 A549 cells were infected at MOI 1 with the various MVA producing Mtb antigen fusions in presence of proteasome inhibitor MG132 (10 M) added to growth medium 30 min after infection. MVATGN33.1 empty vector was used as negative control. After 24 hours, medium was discarded and cells were lysed with 300 L/dish of Tris-Glycin-SDS 2 buffer (ref: LC2676; Novex) supplemented with -mercaptoethanol (5% v:v). The lysate was then sonicated and heated for 5 min at 95 C. Twenty microliters of cell lysates were submitted to electrophoresis onto precasted 4-15% Criterion gel using the Criterion Precast gel system (Biorad). Following electrophoresis, proteins were transferred onto a PVDF membrane (Trans-Blot Turbo Transfer System (#170-4155, Biorad)). Immunodetection was performed with Mtb specific antibodies, as described above in connection with expression products of DNA plasmids. Immune-complexes were revealed using the ImmunStar WesternC kit (Biorad, ref 170.5070).
(114) Immunogenicity Evaluation in a Mouse Model
(115) DNA Immunization Protocols
(116) Mice were immunized three times at 2 or 3-week interval either with the fusion encoding plasmid or with a mix of plasmids encoding the individual Mtb antigens included in the fusion. 100 g of DNA in 100 L of sterile PBS were injected via intramuscular route in the tibialis anterior muscle. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays.
(117) MVA Immunization Protocols
(118) Immunogenicity of MVA TB candidates was evaluated in BALB/c, transgenic HLA-A2, C57BL/6 and C3H/HeN mice. Each MVA vector was administered subcutaneously at the base of the tail once at a dose of 110.sup.7 pfu in 100 L of a Tris-HCl-buffered and sucrose-containing buffer. Cellular immune responses were evaluated 7 days after MVA injection by ELISpot IFN assay.
(119) Peptide Libraries
(120) A peptide library was used to restimulate ex-vivo the splenocytes from immunized mice. More precisely, 679 peptides (15 mers overlapping by 11 amino acids) covering all 14 Mtb antigens contained in the fusions described above were synthesized (ProImmune). Pools of peptides were prepared in DMSO with a final concentration of 1 mol/L. One to 4 pools were needed so as to cover the full length of each Mtb antigen.
(121) Rv1733 was covered by 2 pools of 18 and 17 peptides. Pool 1: 18 peptides covering Rv1733 residues 62 to 144; Pool 2: 17 peptides covering Rv1733 residues 134 to 210.
(122) Rv2029 was covered by 4 pools of 19 peptides. Pool 1: 19 peptides covering Rv2029 residues 1 to 87; Pool 2: 19 peptides covering Rv2029 residues 77 to 163; Pool 3: 19 peptides covering Rv2029 residues 153 to 239; Pool 4: 19 peptides covering Rv2029 residues 229 to 314.
(123) Rv0569 was covered by 1 pool of 20 peptides covering Rv0569 from residues 1 to 88.
(124) Rv1807 was covered by 4 pools of 25 peptides for the first 3 pools and 22 peptides for the fourth pool. Pool 1: 25 peptides covering Rv1807 residues 1 to 111; Pool 2: 25 peptides covering Rv1807 residues 101 to 211; Pool 3: 25 peptides covering Rv1807 residues 201 to 311; Pool 4: 22 peptides covering Rv1807 residues 301 to 399.
(125) Rv0111 was covered by 4 pools of 20 peptides for the first 3 pools and 19 peptides for the fourth pool. Pool 1: 20 peptides covering Rv0111 residues 361 to 451; Pool 2: 20 peptides covering Rv0111 residues 441 to 531; Pool 3: 20 peptides covering Rv0111 residues 521 to 611; Pool 4: 19 peptides covering Rv0111 residues 601 to 685.
(126) RpfB-Dhyb was covered by 4 pools of 22 peptides for the first 3 pools and 19 peptides for the fourth pool. Pool 1: 22 peptides covering RpfB residues 30 to 127; Pool 2: 22 peptides covering RpfB residues 117 to 215; Pool 3: 22 peptides covering RpfB residues 205 to 284 and RpfD residues 53 to 71; Pool 4: 19 peptides covering RpfD residues 61 to 146.
(127) Rv1813 was covered by 1 pool of 25 peptides covering Rv1813 residues 34 to 143.
(128) Rv3407 was covered by 1 pool of 22 peptides covering Rv3407 residues 1 to 99.
(129) Rv3478 was covered by 4 pools of 24 peptides. Pool 1: 24 peptides covering Rv3478 residues 1 to 107; Pool 2: 24 peptides covering Rv3478 residues 97 to 203; Pool 3: 24 peptides covering Rv3478 residues 193 to 299; Pool 4: 24 peptides covering Rv3478 residues 289 to 393.
(130) Rv2626 was covered by 2 pools of 17 and 16 peptides. Pool 1: 17 peptides covering Rv2626 residues 1 to 79; Pool 2: 16 peptides covering Rv2626 residues 69 to 143.
(131) Ag85B was covered by 3 pools of 23 peptides. Pool 1: 23 peptides covering Ag85B residues 39 to 141; Pool 2: 23 peptides covering Ag85B residues 131 to 233; Pool 3: 23 peptides covering Ag85B residues 223 to 325.
(132) ESAT-6 was covered by 1 pool of 21 peptides covering ESAT-6 from residues 1 to 95.
(133) TB10.4 was covered by 1 pool of 21 peptides covering TB10.4 from residues 1 to 95.
(134) IFNELISpot Assays
(135) Splenocytes from immunized mice were collected and red blood cells were lysed (Sigma, R7757). 210.sup.5 cells per well were cultured in triplicate for 40 h in Multiscreen plates (Millipore, MSHA S4510) coated with an anti-mouse IFN monoclonal antibody (BD Biosciences; 10 g/mL, 551216) in MEM culture medium (Gibco, 22571) supplemented with 10% FCS (JRH, 12003-100M), 80 U/mL penicillin/80 g/mL streptomycin (PAN, P06-07-100), 2 mM L-glutamine (Gibco, 25030), 1 non-essential amino acids (Gibco, 11140), 10 mM Hepes (Gibco, 15630), 1 mM sodium pyruvate (Gibco, 31350) and 50 M -mercaptoethanol (Gibco, 31350) and in presence of 10 units/mL of recombinant murine IL2 (Peprotech, 212-12), alone as negative control, or with: The above-described pool of peptides at a final concentration of 1 mol/L 5 g/ml of Concanavalin A (Sigma, C5275) for positive control. Irrelevant peptide
(136) IFN-producing T cells were quantified by ELISpot (cytokine-specific enzyme linked immunospot) assay as previously described (Himoudi et al., 2002, J. Virol. 76: 12735-46). Results are shown as the mean value obtained for triplicate wells. An experimental threshold of positivity for observed responses (or cut-off) was determined by calculating a threshold value which corresponds to the mean value of spots observed with medium alone +2 standard deviations, reported to 10.sup.6 cells. A technical cut-off linked to the CTL ELISpot reader was also defined as being 50 spots/10.sup.6 cells (which is the value above which the CV (coefficient of variation) of the reader was systematically less than 20%). Statistical analyses of ELISpot responses were conducted by using a Kruskal-Wallis test followed, when a significant difference was obtained, by a Mann-Whitney test. P value equal or inferior to 0.05 will be considered as significant.
(137) Evaluation of Therapeutic Efficacy of Mtb Antigens-Containing Vaccines Against Mycobacterium tuberculosis Infection in Mice
(138) Female C57BL/6 mice (6 to 8 weeks old) were aerosol challenged at week 0 using a contained Henderson apparatus in conjunction with an Aero control unit (Hartings et al., 2004, J Pharmacol Toxicol Methods 49: 39-55). Mycobacterium tuberculosis challenge strain H37Rv (NCTC 7416) was cultured in a chemostat (James et al., 2000, J Appl Microbiol 88: 669-77) and fine particles of a mean diameter of 2 m were generated in a Collison nebulizer and delivered directly to the animal snout. The suspension in the Collison nebulizer was adjusted to deliver an estimated inhaled dose of approximately 100 CFU/lung to each group of mice.
(139) Mice were immunized at 10 and 14 weeks post-infection with MVATG18364, MVATG18376 or MVATG18377 given subcutaneously at the base of tail in one site (10.sup.7 pfu/100 L/mouse). A group of mice was injected with MVATGN33.1 as negative control (10.sup.7 pfu/100 L/mouse).
(140) Mice were treated with isoniazid (INH, 25 mg per kg body weight) and rifapentine (RIF, 10 mg per kg body weight) once a week by oral gavage for 10 weeks, from week 6 to week 15 (protocol adapted from Aagaard et al., 2011, Nat Med, 17: 189-194). 5 mice were sacrificed at week 6 before drug treatment.
(141) Mice from each of group were sacrificed, five at the end of the antibiotic treatment (15 weeks post-infection) to determine clearance at the end of the treatment phase and the others at week 21. Organs (e.g. spleen) were aseptically removed, frozen on the day of necropsy and processed for bacterial load analysis. Serial dilutions of samples of organ homogenates were plated onto 7H11 Middlebrook OADC selective agar and incubated for up to 3 weeks for enumeration of viable mycobacteria (CFU). Bacterial load data was expressed as Log 10 total Colony Forming Unit (CFU).
(142) The therapeutic efficacies of the MVA candidates were compared and ranked using the Mann-Whitney test. A p value inferior to 0.05 was considered significant.
(143) Results
Example 1: Selection of Suitable Mtb Antigens for Immunogenic Combination
(144) The Mtb genome expresses approximately 4000 genes but the function and role in Mtb life cycle of the great majority of the gene products have not yet been characterized. As described in Materials and Methods, existing data on Mtb antigens were investigated from the available literature and data bases with the goal of identifying the most appropriate set of genes/antigens from Mtb genome for providing an immunotherapeutic vaccine capable of raising anti-TB immunity during all phases of the natural course of infection.
(145) These bibliographic analyses permit to pre-select a set of 33 Mtb antigens belonging to all three phases of infection, namely seven antigens of the active phase, five resuscitation (Rpf) antigens and 19 latent antigens as well as two PE/PPE antigens. Antigens of the active phase: ESAT-6 (Rv3875), CFP-10 (Rv3874), TB10.4 (Rv0288), Ag85A (Rv3804), Ag85B, (Rv1886) and two ESAT-6 like antigens (Rv3620 and Rv3619); Two PE/PPE antigens (Rv2608 and Rv3478) which appear to be associated with virulence. Antigens of the resuscitation phase: the five existing Rpfs genes (RpfA (Rv0867c), RpfB (Rv1009), RpfC (Rv1884c), RpfD (Rv2389), RpfE (Rv2450c)) were preselected. Rpfs are secreted or membrane bound muralytic enzymes which expression is required for the resuscitation of the dormant cells. Nineteen latent antigens were preselected from the more than 150 existing latent genes described. More precisely, twelve belong to the DosR regulon, a set of 45 genes which expression is increased during latency period and five were selected among genes which expression was modulated during culture condition thought to mimic the latency condition that Mtb encounters in vivo. Three latent antigens were also selected based on preclinical and early clinical phases recently described (Bertholet et al., 2008, J. Immunol. 181: 7948-57; Bertholet et al., 2010, Sci Transl Med 2: 53ra74, Mollenkopf et al., 2004, Infect Immun 72: 6471-9). In summary, the 19 latent antigens preselected were Rv1733c, Rv2029c, Rv1735, Rv1737, Rv2628, Rv0569, Rv2032, Rv2627c, Rv0111, Rv3812, Rv1806, Rv1807, Rv0198, Rv2626, Rv0081, Rv2005c, Rv2660, Rv3407 and Rv1813.
(146) Then, a second selection was undertaken in order to rank the 33 preselected Mtb antigens. The second selection of Mtb antigens was based on a data mining-based selection process (see Materials and Methods) reflecting their immunological and protective potential (highest score retained) as well as biochemical prediction.
(147) The following antigens were chosen: Latent phase antigens: Rv1733, Rv2029, Rv0569, Rv0111, Rv1807 and Rv3407. Rv2626 and Rv1813 were also chosen due to their very good data mining score and biochemical prediction score. Active phase antigens: ESAT-6 (Rv3875), TB10.4 (Rv0288), Ag85B (Rv1886) and Rv3478. It has to be noted that the pre-selected active phase Rv3619 antigen had a good data mining score, but being an ESAT-6 like protein while not showing better score than ESAT-6 itself, it was not retained in the selected list. As another example, active Rv2608 and Rv3478 antigens had the same data mining score but Rv3478 was selected on the basis of its capacity of inducing a stronger percentage of responders in human cohort studies. Resuscitation phase antigens: RpfB and RpfD. Among the 5 resuscitation gene products, three Rpfs stood out (RpfB, D and E) with very similar score after the running data mining scoring process but only RpfB and D were selected for 2 main reasons. Firstly, the reported cross-reactivity in term of cellular and humoral responses between 4 out of 5 Rpfs (Yeremeev et al., 2003, Infect Immun 71: 4789-94), except for RpfB justified in our view the selection of the latter. Secondly, RpfD was chosen instead of RpfE after sequence analysis based on a lower sequence homology in the lysozyme domain (LD) between RpfB and D than between RpfB and E. It is thus assumed that keeping Rpfs B and D would be sufficient to generate immune response toward the 5 Rpfs.
Example 2: Fusion Design
(148) Extensive in silico structure prediction and bibliographical analyses were performed in order to predict biochemical properties and/or biological functions of the selected Mtb antigens as described in Materials and Methods.
(149) The selected 14 antigen candidates were classified into three groups that required different types of analysis. Antigens with available data concerning their expression in various viral vectors namely Ag85B ESAT-6 and TB10-4 in MVA (Kolilab et al. 2010, Clin Vaccine Immunol 17: 793-801); vaccinia virus (Malin et al. 2000, Microbes Infect 2: 1677-85) and adenovirus (Mu et al., 2009, Mol Ther 17: 1093-100; Dietrich et al. 2005, J Immunol 174: 6332-9; and Havenga et al. 2006; J Gen Virol 87: 2135-43). In these cases, analysis of the bibliography was the main source of information to design the sequence to insert in vector constructions. Antigens with no data reported on viral vectorization but identical or homologous to a protein with a known structure. In these cases, structural data were the main source of information to design the Mtb sequence to insert in vector constructions (Rv2626, Rv2029, RpfB, RpfD and Rv0569). Antigens with no data reported on viral vectorization and with no homology with any protein with a known structure. In these cases, in silico biochemical analyses and predictions were used to characterize the antigens, and to design the Mtb sequence to insert in vector constructions (Rv0111, Rv3407, Rv3478, Rv1807 and Rv1813).
(150) Design of Ag85B Antigen
(151) Ag85B displays a 40 residues long peptide signal that was conserved in the Kolilab's MVA vector but not in the Malin's vaccinia virus and the adenovirus constructs. As Ag85B signal peptide was predicted as a TM domain, the inventors recommended not to keep the Ag85B peptide signal in the vector constructions of this invention. The recommended primary structure of Ag85B* to be used in vector constructions described herein corresponds to the amino acid sequence shown in SEQ ID NO: 20.
(152) Design of ESAT-6 Antigen
(153) ESAT-6 forms a heterodimeric complex with CFP-10 and this heterodimeric interaction is expected to induce the folding of both proteins. Alone ESAT-6 adopts a molten globule-like state and a helix-turn-helix when complexed with CFP10. Thus, ESAT-6 bound to its partner could be more stable than ESAT-6 expressed alone. However, the recommended primary structure of ESAT-6 to be used in vector constructions described herein corresponds to the full length protein, (amino acid sequence shown in SEQ ID NO: 14) eventually without its initiator Met (e.g. if internal position in the fusion).
(154) Design of TB10-4 (Rv0288)
(155) TB10-4 belongs to the same family of protein as ESAT-6. NMR structure of TB10-4 showed that it forms a heterodimeric complex with Rv0287 that is expected to stabilize the structure. There is no publication reporting TB10-4 expression by poxviruses whereas expression of the full length TB10-4 was reported in adenovirus vectors in a form fused to the C-terminus part of either Ag85A or Ag85B. On this basis, the recommended primary structure of TB10.4 to be used in vector constructions described herein corresponds to the full length protein (amino acid sequence shown in SEQ ID NO: 2), eventually without its initiator Met.
(156) Design of Rv2626
(157) Crystallization of Rv2626 (Sharpe et al., 2008, J Mol Biol 383: 822-36) showed that it is expressed as a homodimer with an intra and an inter subunit disulfide bonds. No signal peptide was predicted for Rv2626. Since Rv2626 has a very well defined fold, the recommended primary structure of Rv2626 to be used in vector constructions described herein corresponds to the full length protein (amino acid sequence shown in SEQ ID NO: 10), eventually without its initiator Met.
(158) Design of Rv0569
(159) Rv0569 structure is not known but this protein displays a 62% identity (81% similarity) with Rv2302 in a 76 amino acid overlap region (out of 88 residues). The structure of this latter has been solved by NMR (Buchko et al., 2006, Bacteriol 188: 5993-6001) and showed a very well folded structure in solution with antiparallel -sheet core and a C-terminal -helix. No coiled coil prediction is associated with this protein. No known function is associated with Rv0569 protein. Due to the potential very well defined fold, the recommended primary structure of the Mtb Rv0569 to be used in vector constructions described herein corresponds to the full length protein (amino acid sequence shown in SEQ ID NO: 3, eventually without its initiator Met.
(160) Design of Rv2029
(161) Rv2029 structure is not known, but this protein displays a 35% identity with phosphofructokinase-2 (pfk2) of Escherichia coli in a 310 aa overlap region (out of 339). Moreover, PROSCAN search yielded to the identification of a fully conserved carbohydrate kinases signature. Therefore, Rv2029 has probably a phosphofructokinase activity in Mtb. Phosphofructokinase catalyzes the phosphorylation of fructose-6-phosphate during glycolysis. E. coli pfk2 structure is tetrameric when ATP is bound and dimeric when ATP is not present in the medium (allosteric regulation of the enzyme activity). In the E. coli enzyme, deletion of the last C-terminal 4 residues completely inhibits ATP induced tetramerization. Thus, in order to avoid oligomerization heterogeneity of Rv2029 (mix of dimeric and tetrameric forms), the deletion of the C-terminus part is recommended (i.e. deletion of the last 25 residues). Moreover, in order to abolish enzymatic activity of Rv2029, the mutation D265N (position 265 starting from the Met initiator or 264 without Met) is recommended since it abolishes almost totally the enzymatic activity in E. coli pfk-2 (Cabrera et al., 2010, Arch Biochem Biophys 502: 23-30). On this basis, the recommended primary structure of the Rv2029 antigen (Rv2029*) to be used in vector constructions described herein corresponds to the amino acid sequence shown in SEQ ID NO: 21.
(162) Design of RpfB and RpfD
(163) The Resuscitation Promoting Factors (Rpf) are secreted proteins that are produced during the reactivation phase of the bacteria (transition from dormancy to growth). M. tuberculosis has five different Rpf (A to E) that all contain a conserved catalytic domain (lysozyme like domain). Apart from this domain, there is no significant similarity among these five proteins. RpfB structure has been obtained for about half of the molecule (residues 194-362) and a signal peptide was predicted (residues 1-29; Ruggiero et al. 2009, J Mol Biol 385: 153-62). The full length protein (without its signal peptide) behaves as a monomer when expressed in E. coli.
(164) In silico predictions and analyses were performed on RpfB to analyse the part of the protein (30-193) for which no structure was available. Except for the signal peptide, no transmembrane domain was predicted. HCA plots, secondary structure prediction and natively disordered regions predictions are in agreement with a well-defined fold of the 30-193 region. Coiled coils predictions and search for known motifs using PROSCAN did not yield any significant result.
(165) Activity of the catalytic domain has been shown to depend on a conserved residue essential in the resuscitation activity of Micrococcus luteus Rpf in a Mycobacterium smegmatis resuscitation assay (mutation E292K; Mukamolova et al. 2006, Mol Microbiol 59: 84-98). Furthermore, the two residues T315 and Q347 are involved in substrate binding in lysozyme, and conserved in RpfB (Cohen-Gonsaud, et al. 2005, Nat Struct Mol Biol 12, 270-3).
(166) In addition, it has been chosen to design a RPFB-D hybrid that corresponds to the RpfB molecule with its catalytic domain replaced by the most divergent catalytic domain among Rpfs (i.e. RpfD catalytic domain). Therefore, the RPFB-D hybrid to be expressed in viral vectors is a hybrid protein with a neutralized catalytic activity by three mutations (E292K, T315A and Q347A) and without signal peptide. The recommended primary structure for this RPFB-D hybrid protein used in fusions corresponds to the amino acid sequence shown in SEQ ID NO: 31 from residue 10 to residue 283 of RpfB fused to residue 51 to residue 147 of RpfD, eventually with a initiator Met.
(167) Design of Rv1807
(168) Rv1807 structure is not publicly available, but a BLAST search against the PDB database yielded a match with only the first 150 residues of a Mtb PPE protein (Rv2430). PE/PPE is a large family of Mtb proteins (around 100 PE and 60 PPE members) that have in common a PE (Proline, Glutamic acid) or PPE (Proline, Proline, Glutamic acid) motif, at their N-terminus parts. PE proteins are expressed as heterodimers with PPE, and their function is not known yet. BLAST search against UNIPROT-SWISSPROT yielded several matches but all of them were additional Mtb PPE that did not allow to gain additional information.
(169) In E. coli, expression of a soluble PPE (Rv2430) or PE (Rv2431) is apparently possible only when expressed as a heterodimer (Strong et al. 2006, Proc Natl Acad Sci 103: 8060-5). These authors reported that Rv1807 expressed alone in E. coli forms inclusion bodies. PROSCAN search did not yield any significant match with a known motif. No signal peptide or transmembrane domain were reported or predicted for this protein. HCA plots, as well as secondary structure predictions were in agreement with a well-defined fold of the whole protein except the last 60-70 residues region. Moreover, the last 60 residues are predicted to be unfolded using natively disordered regions predictions whereas coiled coils predictions on Rv1807 did not yield any significant result.
(170) As for ESAT6 and TB10-4, the coexpression of Rv1807 with its partner (i.e. Rv1806) would probably favourably impact the protein stability and therefore potentially its immunogenicity. The expression of a misfolded protein (a monomeric one) could impair the recombinant vector stability (protein toxicity). Moreover, the unfolded C-terminus part of Rv1807 could also have an unfavourable impact on either immunogenicity and/or on the recombinant virus stability. The recommended primary structure for Rv1807 used in fusions corresponds to the full length protein (SEQ ID NO: 6). In case of problem encountered with the full length antigen, one may use a C-terminus truncated antigen deleted of the last 60 residues (as shown in SEQ ID NO: 18).
(171) Design of Rv3478
(172) Rv3478 is another PPE protein. Its PPE domain is 57% identical to the PPE domain of Rv1807 (41% identity between the two whole proteins). BLAST search against UNIPROT-SWISSPROT yielded several matches that were all other Mtb PPE. HCA plot demonstrated the presence of hydrophobic patches all along the protein sequence. In other words, HCA plot does not indicate unfolded hydrophilic region in Rv3478. But, as for Rv1807, the last 40 to 50 residues of Rv3478 are predicted to be unfolded (based on both secondary structure and natively disordered predictions). No signal peptide or transmembrane domain were reported or predicted for this protein. Coiled coils predictions on Rv3478 did not yield any significant result. As for Rv1807, the recommended primary structure is the full length protein (SEQ ID NO: 13) or, if problem are encountered, a C-terminus truncated antigen deleted of the last 40 residues (as shown in SEQ ID NO: 24).
(173) Design of Rv0111
(174) Rv0111 is predicted to be a membrane protein with a possible acyltransferase activity. Ten transmembrane domains are predicted by DAS, TMHMM and TopPred spanning from residues 58 to 427. No signal peptide was predicted. Secondary structures are predicted all along the primary structure, with a gap at 449-469 that corresponds to a predicted natively disordered region. Coiled coils predictions on Rv0111 did not yield any significant result.
(175) Proscan analysis yielded four hits with 80% similarity: Aldo/keto reductase enzyme site, acyltransferase lipoyl binding site, sugar transport protein signature and the eukaryote lipocalin proteins. As the three first signatures are in the first 300 residues of the protein, it is thus recommended to remove at least this part of the protein in order to avoid any potential biological activity. This would also allow to get rid of the majority of the transmembrane domains of the protein. Therefore the recommended primary structure of Rv0111 to be used in viral vectors is the C terminus part of the protein (e.g. residues 393-685 of the native antigen as shown in SEQ ID NO: 15) with only one TM for plasmatic membrane anchorage in case of secreted construction. If expression problems are encountered, one may use an even more truncated antigen without any TM domain (residues 429-685 of the native Rv0111 starting at residue 37 of SEQ ID NO: 15).
(176) Design of Rv1813
(177) Rv1813 structure is not publicly available and BLAST search against PDB yielded no match. Rv1813 is a small protein (143 residues), that is predicted to contain a signal peptide (1-32) and no transmembrane domain. It displays no significant homology with other proteins in the Uniprot-Swissprot database. HCA plots, secondary structure prediction and natively disordered regions predictions are all in agreement with a well-defined fold of the whole protein. Coiled coils predictions did not yield any significant result. No function is reported in the TB base and a PROSCAN search yielded no significant match with a known motif. Therefore the recommended primary structure of Rv1813 to be used in viral vectors is the full-length protein without its signal peptide (residues 1 to 34) which amino acid sequence is shown in SEQ ID NO: 19.
(178) Design of Rv3407
(179) Rv3407 structure is not publicly available and a BLAST search against PDB did not yield any match. Rv3407 is a small protein (99 residues) with no significant homology with other protein in Uniprot-Swissprot database. No signal peptide or transmembrane domain was reported or predicted for this protein. HCA plot and secondary structure predictions were in agreement with a well-defined fold of the whole protein. However, natively disordered regions predictions indicated that the last 33 residues may not be folded in a defined structure. This last result that is not in agreement with HCA and secondary structure predictions could be the signature of a MORE (Molecular Recognition Element) that folds upon binding to a partner protein. In the case of Rv3407 a C-terminal alpha helix could be present only when Rv3407 is bound to its partner. Coiled coils predictions did not yield any significant result. No function is reported in TB base for this protein and PROSCAN search did not yield any significant match with a known motif. The recommended primary structure of Rv3407 is the full length protein (SEQ ID NO: 12). If stability issue are encountered, one may use a C-terminus truncated antigen deleted of the last 33 residues (as shown in SEQ ID NO: 23).
(180) Design of Rv1733
(181) Rv1733 is predicted to be a membrane protein according to UNIPROT-SWISSPROT and TB base, with two transmembrane domains (that are also predicted using DAS, TMHMM and TopPred). The first TM domain was predicted as a signal peptide. Apart from these transmembrane domains, few secondary structures are predicted for this protein. HCA plot demonstrates the presence of few hydrophobic patches between the two transmembrane helices. Finally, a natively disordered region (about 20 residues long) was predicted between the two transmembrane helices. All together, theses results indicate a probably loose fold beside the transmembrane domains. PROSCAN search on Rv1733 without its signal peptide did not yield any significant match with a known motif Coiled coil prediction on Rv1733 did not yield any significant result. Therefore the recommended primary structure of Rv1733 to be used in viral vectors is the whole protein minus its signal peptide (62 first residues) as shown in SEQ ID NO: 17. Alternatively, one may also use the full length Rv1733 (SEQ ID NO: 5).
Example 3: Construction of Mtb Gene Fusions
(182) Twelve different fusion proteins were engineered as illustrated in
(183) The following biochemical rationales have been followed to design the fusions Two dimeric proteins (i.e. Rv2626 and Rv2929*) should not be fused in the same fusion to avoid any steric clashes. Two membrane proteins should not be used in the same fusion (i.e. Rv1733* and Rv0111*) to avoid potential toxicity issue. Proteins with TM (Rv1733* and Rv0111*) should be inserted at the end of the fusion to allow plasmatic membrane anchorage for secreted constructions. If possible, the fusion protein should start with a well folded protein (i.e. Ag85B*, Rv2029*, Rv2626, Rv0569, RPFB-D hybrid*) to have a chaperone effect on the rest of the fusion. Make three fusions, two of which with the minimal set of antigens (i.e. Ag85B*, Rv2029*, Rv2626, Rv0111*, Rv1733*, TB10-4, ESAT-6, RPFB-D hybrid*) and the last fusion with the rest (optional) of the antigens (i.e. Rv0569, Rv1813*, Rv3407, Rv1807, Rv3478).
(184) On the other hand, fusions were also designed relative to the phase of TB disease. Fusion no 2 contains active antigens (Ag85B*-TB10.4-ESAT6) while fusion no 4 contains active and resuscitation antigens (RPFB-Dhyb*-Ag85B*-TB10.4-ESAT6). The fusion no 13 is constituted by latent antigens (Rv2029*-Rv2626-Rv1733*-Rv0111*).
(185) As described in Materials and Methods, a series of peptides were added to the Mtb antigen fusion, respectively a N-terminal Flag Tag and C-terminal c-myc and His Tag peptides aimed to facilitate detection of the encoded gene products as well as N-terminal signal and C-terminal membrane-anchoring peptides to enhance immunogenic activity (to be noted that addition of a TM domain was not necessary for fusions ending with Rv0111* or Rv1733*, as these proteins already contain such domains).
(186) For comparative purposes, fusions were also constructed without any SS and TM peptides in order to evaluate cytoplasmic expression of the encoded Mtb antigens. The fusions no 3 (pTG18267), no 5 (pTG18269), no 2 (pTG18266) and no 4 (pTG18268) were deleted from the SS and TM peptides, giving fusions no 12 (pTG18307), no 9 (pTG18295), no 10 (pTG18296) and no 11 (pTG18297). The N-terminus Flag TAG and the C-terminus c-myc and His TAG were kept in these constructions.
(187) Table 3 provides a summary of the various fusions constructed in this study
(188) TABLE-US-00003 Fusion # TB antigens plasmids Fusion by 13 Rv2029*-Rv2626-Rv1733*-Rv0111* pTG18323 phase 2 Ag85B*-TB10.4-ESAT6 pTG18266 4 RPFB-Dhyb-Ag85B*-TB10.4- pTG18268 ESAT6 Max list 5 Rv0569-Rv1813*-Rv3407-Rv3478- pTG18269 Rv1807 Fusion by 6 Ag85B*-Rv2626-RPFB-Dhyb- pTG18270 biochem- Rv1733* istry rules 14 Rv2029-TB10.4-ESAT6-Rv0111* pTG18324 8 Ag85B*-Rv2626-Rv1733* pTG18272 3 RPFB-Dhyb pTG18267 Fusion 9 Rv0569-Rv1813*-Rv3407-Rv3478- pTG18295 without SS Rv1807 and TM 10 Ag85B*-TB10.4-ESAT6 pTG18296 11 RPFB-Dhyb-Ag85B*-TB10.4- pTG18297 ESAT6 12 RPFB-Dhyb pTG18307
(189) For comparative purposes, plasmids encoding the individual Mtb genes used in the above-described fusions were amplified by PCR or gene sequence synthesized by Geneart. More precisely, pTG18269 was used as template to amplify Rv3407, Rv0569, Rv1807, Rv1813* and Rv3478 whereas pTG18323 was used to amplify Rv2626. ESAT6, Rv1733*, Ag85B*, TB10-4, Rv0111* and Rv2029* were produced as synthetic genes.
(190) The individual genes were placed in the same context as the fusions, i.e. inserted in pGWiz downstream the CMV promoter and fused to Flag in 5 and c-myc-His sequences in 3. As Rv1733* and Rv0111* proteins contain a TM domain, the signal peptide presents at the N-terminus of the glycoprotein precursor of rabies virus ERA strain was fused upstream to the Flag sequence to avoid expression issues. The generated plasmids were named respectively pTG18300 (Rv3407), pTG18301 (Rv0569), pTG18302 (Rv1807), pTG18303 (Rv1813*), pTG18304 (Rv3478), pTG18305 (Rv2626), pTG18308 (ESAT6), pTG18309 (Rv1733*), pTG18310 (Ag85B*), pTG18315 (TB10.4), pTG18329 (Rv0111*), pTG18317 (Rv2029*).
(191) The various fusion proteins was assessed in eukaryotic expression system after introduction of the corresponding expression plasmids. Expression was assessed by Western Blot whereas immunogenic activity was evaluated by ELISpot IFN assays after DNA immunization of mice. When possible, expression and immunogenicity of the cytoplasmic (without SS and TM) and membrane-anchored versions were compared as well as immunogenicity provided by the fusions with that obtained with a mix of plasmids expressing the individual Mtb antigens.
Example 4: Analysis of Expression of Mtb Antigens and Fusions
(192) Whether expressed individual or in fusion, expression of Mtb genes was analyzed by Western Blot from cell lysates obtained from transfected HEK293 cells.
(193) 4.1 Western Blot Analysis of Cell Lysate Transfected with Plasmids Encoding Individual Mtb Antigens.
(194) Immunodetection of the individual Mtb antigens was performed with either antibodies directed to the tag peptides included in the expression cassettes (e.g. anti-Flag M2 peroxydase (HRP) antibody, monoclonal anti-c-myc peroxidase antibody and monoclonal anti-His peroxydase antibody) or antibodies specific for Mtb antigens. Specifically, the sera obtained after immunization of rabbits (see Materials and Methods) were used for detection of Rv1733*, Rv2029*, Rv0569, Rv1807, Rv0111*, Rpf-B-D, Rv1813*, Rv3407, and Rv3478 whereas commercial antibodies were used for the detection of ESAT6, Ag85B*, TB10.4 and Rv2626.
(195) The results are summarized in Table 4. More specifically, a band corresponding to the expected size was detected for all individual proteins, whatever the immunodetection system used (anti-Flag, anti-His antibodies, specific rabbit sera and commercial antibodies). Additional products were also detected for some proteins and depending on the immunodetection system used. Moreover, high levels of expression were detected, except for Rv3407 and, to a lesser extend TB10.4 and ESAT6. Examples of expression detection are shown in
(196) TABLE-US-00004 TABLE 4 TB antigen Expected size Additional products with anti Flag Additional products (plasmid) (level expr.) and anti-His antibodies with anti Mtb antibodies Rv3407 14.4 kDa (+) (pTG18300) Rv0569 12.9 kDa (+++) 1 weak band 10 kDa (pTG18301) Rv1807 43.3 kDa (++) (pTG18302) Rv1813* 15.1 kDa (+++) (pTG18303) Rv3478 42.8 kDa (+++) 2 N-terminal clived products 1 band 30 kDa (pTG18304) (recognized by anti-Flag antibody) of about 16 and 26 kDA Rv2626 18.9 kDa (+++) Additional band corresponding to Dimer 43 kDa (pTG18305) Rv2626 dimers ESAT6 13.0 kDa (++) (pTG18308) Rv1733* 21.2 kDa (+++) One N-terminal clived product of 2 bands 10 and 20 (pTG18309) about 20 kDa and 3 C-terminal products kDa comprised between 8 and 10 kDa Ag85B* 33.9 kDa (+++) 5 minor N-terminal clived products 3 weak bands 26, 28 (pTG18310) of about 26, 24, 20, 17 and 12 kDa as and 34 kDa well as a C-terminal clived product (detected with anti-His antibody) of about 34 kDa. TB10.4 13.5 kDa (++) (pTG18315) Rv0111* 37.6 kDa (+++) one N-terminal clived product of 1 band 34 kDa and 2 (pTG18329) about 8 kDa and one C-terminal very weak bands 18 products of about 34 kDa and 20 kDa Rv2029* 35.8 kDa (+++) (pTG18317) RfpB-Dhyb* 39.4 kDa (+++) 2 weak bands 40 kDa (pTG18307)
(197) 4.2 Western Blot Analysis of Cell Lysate Transfected with Plasmids Encoding Mtb Antigen Fusions.
(198) HEK293 cells were transfected with the plasmids expressing the different Mtb gene fusions and expression products were analysed by Western blot in the same conditions as above. Transfections were done in the presence, but also in the absence of proteasome inhibitor MG132. Here again, immunodetection was performed with anti-Flag M2 peroxydase (HRP) antibody, monoclonal anti-c-myc peroxidase antibody and monoclonal anti-His peroxydase antibody as well as anti-Mtb specific antibodies.
(199) The expected sizes of the tagged fusions are indicated below: fusion no 2 (pTG18266): 63.6 kDa fusion no 3 (pTG18267): 49.0 kDa fusion no 4 (pTG18268): 99.7 kDa fusion no 5 (pTG18269): 122.0 kDa fusion no 6 (pTG18270): 103.5 kDa fusion no 8 (pTG18272): 67.3 kDa fusion no 9 (pTG18295): 112.9 kDa fusion no 10 (pTG18296): 53.8 kDa fusion no 11 (pTG18297): 90.0 kDa fusion no 12 (pTG18307): 39.3 kDa fusion no 13 (pTG18323): 101.5 kDa fusion no 14 (pTG18324): 90.6 kDA
(200) All the Mtb antigen fusions were detected with the anti-Flag and anti-His monoclonal antibodies. Mtb fusion products were also detected with the anti-c myc monoclonal antibody except for pTG18266, pTG18267, pTG18268 and pTG18269. The c-myc epitope might be inaccessible in these fusions due to adjacent TM domains since the cytoplasmic counterparts (pTG18296, pTG18307, pTG18297 and pTG18295) are well detected with the anti-myc antibody.
(201)
(202) Whatever the immunodetection system, a band corresponding to the expected size was highlighted for all fusions and, in some cases, additional fusion products were also observed. In particular, dimers were detected for pTG18270, pTG18272 and pTG18323. These three fusions contain Rv2626 which has the ability to form dimers resistant to reducing conditions. Immunodetection with anti-Flag and anti-His antibodies highlighted some additional minor proteolytic products for pTG18323 and pTG18269. Moreover, additional products higher than the expected size were detected for pTG18266, pTG18268, pTG18269, pTG18270, pTG18272, pTG18323 and pTG18324 with anti-Flag and anti-His antobodies. These bands correspond to N-glycosylated products as it was demonstrated by in vitro treatment with N-Glycosidase F (i.e. expression products at the expected size were obtained after N-glycosidase treatment of cellular extracts). All fusions containing a signal peptide lead to N-glycosylated products, except fusion no 4 (pTG18267, RpfB-D*). N-glycosylated products were also detected with antigen-specific antibodies as well as dimers for Rv2626-containing fusions pTG18270, pTG18272 and pTG18323. Proteolytic products were also evidenced for some fusions with specific sera (data not shown) depending on the fusions and the sera. For example, additional bands of 40 kDa for pTG18269 and 36 and 38 kDa for pTG18295 were detected with Rv3407 specific serum but are not seen with Rv0569 specific serum.
(203) Similar and high levels of expression were obtained for all fusions and higher amounts of products were detected in the presence of MG132. The expression levels of membrane-anchored fusions (pTG18269, pTG18268) were comparable to those detected with their cytoplasmic counterparts (pTG18295, pTG18297), except for pTG18266 which was better expressed than the cytoplasmic fusion (pTG18296). Fusion no 5 (pTG18269) was very weakly detected with Rv1807 specific antibody while it is not the case for the cytoplasmic fusion (pTG18295). Rv1807 specific epitopes might be inaccessible in this fusion due to adjacent TM sequence.
Example 5: DNA Immunization Evaluation
(204) Immunogenic activity of the various Mtb antigen fusions was evaluated in various mouse models following DNA immunization.
(205) 5.1. Evaluation of the Immunogenicity Induced by Fusions Based on Mtb Antigens of the Active Phase.
(206) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmid expressing the fusion Ag85B-TB10.4-ESAT6 either in an anchored form at the cell membrane (SS/TM: pTG18266) or cytoplasmic form (pTG18296). For comparative purposes, mice were also immunized with a mix of plasmids encoding the individual Mtb antigens included in the fusion (pTG18310 (Ag85B)+pTG18315 (TB10.4)+pTG18308 (ESAT6)) and with empty pGWiz as negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the various peptide pools described in Materials and Methods.
(207) As illustrated in
(208) Thus, at least for Mtb antigens of the active phase, these results highlight the benefit of designing antigen fusions expressed at the cell surface (with SS and TM peptides) to optimize the immunogenic activity of the resulting Mtb antigen fusions.
(209) 5.2. Evaluation of the Immunogenicity Induced by Fusions Based on Mtb Antigens of the Active and Resuscitation Phases.
(210) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmid expressing the fusion RpfB-Dhyb-Ag85B-TB10.4-ESAT6 either in an anchored form at the cell membrane (SS/TM: pTG18268) or cytoplasmic form (pTG18297). For comparative purposes, mice were also immunized with a mix of plasmids encoding the individual TB antigens included in the fusion (pTG18307 (RpfB-Dhyb)+pTG18310 (Ag85B)+pTG18315 (TB10.4)+pTG18308 (ESAT6)) and with empty pGWiz as negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the various peptide pools described in Materials and Methods.
(211) As illustrated in
(212) 5.3. Evaluation of the Immunogenicity Induced by Fusions Based on Mtb Antigens of the Resuscitation Phase.
(213) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmid expressing the fusion RpfB-Dhyb either in an anchored form at the cell membrane (SS/TM: pTG18267) or cytoplasmic form (pTG18307). Empty pGWiz was used as a negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the four peptide pools described in Materials and Methods.
(214) As illustrated in
(215) 5.4. Evaluation of the Immunogenicity Induced by Fusions Based on Mtb Antigens of the Latent Phase.
(216) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmid expressing the fusion Rv0569-Rv1813-Rv3407-Rv3478-Rv1807 either in an anchored form at the cell membrane (pTG18269) or cytoplasmic form (pTG18295). For comparative purposes, mice were also immunized with a mix of plasmids encoding the individual Mtb antigens included in the fusion (pTG18300 (Rv3407)+pTG18301 (Rv0569)+pTG18302 (Rv1807)+pTG18303 (Rv1813)+pTG18304 (Rv3478)) and with empty pGWiz as a negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the various peptide pools described in Materials and Methods.
(217) As illustrated in
(218) Other strains of mice were also used for investigation of the anti Mtb antigen responses in order to cover different MHC haplotypes: BALB/c mice are H-2.sup.d, C57BL/6 mice are H-2.sup.b, CBA/J and C3H/HeN mice are H-2.sup.k.
(219) Mice were immunized three times at 2-week interval via intramuscular route with pTG18323 expressing the antigens from the latent phase Rv2029-Rv2626-Rv1733-Rv0111 or with empty pGWiz as a negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the various peptide pools described in Materials and Methods section. As illustrated in
(220) Immune responses specific for Rv2029, Rv2626 and Rv1733 antigens were also detected to similar levels as seen in H-2.sup.k CBA/J mice immunized with pTG18323. In contrast, in BALB/c mice, IFN producing cells were specifically detected only after re-stimulation with Rv2626 peptides while in C57BL/6 mice no signal was detected.
(221) Overall, these results highlight the fact that the tested Mtb antigen fusion sequences are able to induce robust cell-based immune responses in different haplotype of mice.
(222) 5.5. Evaluation of the Immunogenicity Induced by Fusions Based on Biochemistry Rules.
(223) BALB/c mice or C57BL/6 mice were immunized three times at 2-week interval via intramuscular route with plasmids coding for the fusion number 6, 8 or 14, designed according to biochemistry properties of Mtb antigens, i.e. pTG18270 (Ag85B-Rv2626-RpfB-Dhyb-Rv1733), pTG18272 (Ag85B-Rv2626-Rv1733) and pTG18324 (Rv2029-TB10.4-ESAT-6-Rv0111). For comparative purposes, mice were also immunized with empty pGWiz as negative control. Cellular immune response was evaluated 2 weeks following the last DNA injection by ELISpot IFN assays after ex vivo re-stimulation with the various peptide pools described in Materials and Methods section.
(224) A strong cellular response specific of Ag85B and RpfB-Dhyb antigens was induced in both BALB/c and C57BL/6 mice immunized with pTG18270, whereas high level of IFN producing cells specific of Rv2626 were detected only in BALB/c mice. Immunization with pTG18272 resulted in activation of IFN producing cells specific of Ag85B in BALB/c mice and specific of Ag85B and Rv2626 antigens in C57BL/6 mice, but to a lower level compared to response induced by pTG18270. In mice immunized with pTG18324, high levels of IFN producing cells specific of TB10.4 and ESAT-6 antigens was detected, whereas IFN producing cells specific of Rv2029 and Rv0111 were also induced but to lower levels. As expected, immunization with the empty plasmid did not induce any specific immune response.
(225) Overall, the tested fusions, designed according a biochemical-based rationale in order to increase stability and production of the fusions, display a good immunogenic response specific of the Mtb antigens from the different phases of infection.
(226) 5.6. Evaluation of the Anti-Rv1733 Humoral Response Induced by Mtb Antigen Fusions.
(227) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmids expressing the fusion Ag85B*-Rv2626-Rv1733* (pTG18270) and the fusion Ag85B*-Rv2626-RPFB-Dhyb*-Rv1733* (pTG18272). For comparative purposes, mice were also immunized with a mix of plasmids encoding the individual Mtb antigens included in the fusion (pTG18310 (Ag85B*)+pTG18305 (Rv2626)+pTG18309 (Rv1733*)) and with empty pGWIZ as negative control. Humoral immune response was evaluated 2 weeks following the last DNA injection. Sera of immunized-mice were pooled and analysed by Western-blot. More specifically, 100 ng/lane of recombinant protein Rv1733 (produced in E. coli, see example no 8) were loaded on an acrylamide gel and immunodetection was performed with 1/200 diluted sera. As a result, specific detection on Rv1733 protein was observed with the sera of mice immunised with pTG18270, pTG18272 and the mix of plasmids encoding the individual Mtb antigens.
(228) 5.7. Evaluation of the Anti-Rv1813 Humoral Response Induced by Mtb Antigen Fusions.
(229) BALB/c mice were immunized three times at 3-week interval via intramuscular route with the plasmid expressing the fusion Rv0569-Rv1813-Rv3407-Rv3478-Rv1807 either in an anchored form at the cell membrane (SS/TM: pTG18269) or cytoplasmic form (pTG18295). For comparative purposes, mice were also immunized with a mix of plasmids encoding the individual Mtb antigens included in the fusion (pTG18300 (Rv3407)+pTG18301 (Rv0569)+pTG18302 (Rv1807)+pTG18303 (Rv1813)+pTG18304 (Rv3478)) and with empty pGWiz as a negative control. Humoral immune response was evaluated 2 weeks following the last DNA injection. Sera of immunized mice were pooled and analysed by Western-blot with 100 ng/lane of recombinant protein Rv1813 (produced in E. coli, see example no 8) loaded on an acrylamide gel. Immunodetection was performed with 1/200 diluted sera. As a result, Rv1813 protein was specifically detected with the sera of mice immunised with pTG18269 (encoding fusion in an anchored form at the cell membrane).
Example 6: Generation of Recombinant MVA Expressing Mtb Antigens
(230) A total of 10 MVA vaccine candidates were engineered for expression of one or up to three Mtb fusions and expression of the various Mtb antigens was analyzed by Western Blot from cell lysates obtained from infected A549 cells.
(231) 6.1 Generation of Recombinant MVA by Phase of the TB Disease
(232) Seven recombinant MVA candidates were engineered so as to contain one, two or three cassettes for expression of Mtb fusions representative of the various phases of TB disease. Fusion no 4 and fusion no 11 both contain active and resuscitation antigens (RPFB-Dhyb*-Ag85B*-TB10.4-ESAT6) either expressed anchored in the cell membrane (fusion no 4 equipped with N-terminal signal and C-terminal membrane anchoring peptides) or in the cytoplasm (fusion no 11 corresponds to the cytoplasmic version of fusion no 4). Fusion no 13 contains latent antigens (Rv2029*-Rv2626-Rv1733*-Rv0111*). Fusion no 5 and fusion no 9 both contain additional latent antigens (Rv056-Rv1813*-Rv3407-Rv3478-Rv1807) expressed at different cell location either anchored in the cell membrane (fusion no 5 contains N-terminal signal and C-terminal membrane anchoring peptides) or in the cytoplasm (fusion no 9).
(233) All together, the seven MVA candidates are the followings: MVATG18355 contains the fusion no 13 under the control of p7.5K promoter. MVATG18364 contains the fusion no 13 under the control of p7.5K promoter and the fusion no 4 under the control of pH5R promoter. MVATG18365 contains the fusion no 13 under the control of p7.5K promoter and the fusion no 11 under the control of pH5R promoter. MVATG18376 contains the fusion no 13 under the control of p7.5K promoter, the fusion no 4 under the control of pH5R promoter and the fusion no 5 under the control of B2R promoter. MVATG18377 contains the fusion no 13 under the control of p7.5K promoter, the fusion no 11 under the control of pH5R promoter and the fusion no 5 under the control of B2R promoter MVATG18378 contains the fusion no 13 under the control of p7.5K promoter, the fusion no 4 under the control of pH5R promoter and the fusion no 9 under the control of B2R promoter. MVATG18379 contains the fusion no 13 under the control of p7.5K promoter, the fusion no 11 under the control of pH5R promoter and the fusion no 9 under the control of B2R promoter.
(234) 6.2 Generation of Recombinant MVA on Biochemical Rational
(235) Three recombinant MVA candidates were engineered so as to contain two or three cassettes for expression of Mtb fusions designed relative to biochemical rationales. Fusion no 6 contains the following antigens Ag85B*-Rv2626-RPFB-Dhyb*-Rv1733* while fusion no 14 contains Rv2029*-TB10.4-ESAT6-Rv0111*. N-terminal signal peptides were added for both fusions while no TM domain were added since these fusions end with Rv0111 or Rv1733 which already contain membrane-anchoring peptides. MVATG18404 contains the fusion no 14 under the control of p7.5K promoter and the fusion no 6 under the control of pH5R promoter. MVATG18417 contains the fusion no 14 under the control of p7.5K promoter, the fusion no 6 under the control of pH5R promoter and the fusion no 5 under the control of B2R promoter. MVATG18418 contains the fusion no 14 under the control of p7.5K promoter, the fusion no 6 under the control of pH5R promoter and the fusion no 9 under the control of B2R promoter.
(236) 6.3 Western Blot Analysis of MVA-Expressed Mtb Antigens and Fusions
(237) A549 cells were infected (MOI 1) with the various MVA candidates described above and expression products were analyzed by Western blot under the conditions described in Materials and Methods. Immunodetection was performed with antibodies specific of the various Mtb antigens described herein. Specifically, the sera obtained after immunization of rabbits (see Materials and Methods) were used for detection of Rv1733*, Rv2029*, Rv0569, Rv1807, Rv0111*, RPFB-Dhyb*, Rv1813*, Rv3407, and Rv3478 whereas commercial antibodies were used for the detection of ESAT6, Ag85B*, TB10.4 and Rv2626.
(238) As a result, a band corresponding to the expected size was highlighted for all fusions whatever the recombinant MVA tested. More specifically, a band of approximately 98.4 kDa (expected size for fusion no 13) was detected following anti-Rv2626 and anti-Rv0111 immunodetection in the cell lysates originating from cells infected with MVATG18355, MVATG18364, MVATG18365, MVATG18376, MVATG18377, MVATG18378 and MVATG18379. A band of approximately 96.7 kDa (expected size for fusion no 4) and a band of approximately 87 kDa (expected size for fusion no 11) were detected following anti-ESAT6 immunodetection in the cell lysates originating from cells infected respectively with fusion no 4-containing MVATG18364, MVATG18376 and MVATG18378 and fusion no 11-containing MVATG18365, MVATG18377 and MVATG18379. Moreover a band of approximately 119.7 kDa (expected size for fusion no 5) and a band of approximately 109.9 kDa (expected size for fusion no 9) were detected following anti-Rv3407 immunodetection in the cell lysates originating from cells infected respectively with fusion no 5-containing MVATG18376 and MVATG18377 and fusion no 9-containing MVATG18378 and MVATG18379. Finally, a band of approximately 100.4 kDa (expected size for fusion no 6) and a band of approximately 87.5 kDa (expected size for fusion no 14) were detected in the cell lysates originating from cells infected with MVATG18404 following anti-Rv2626 and anti-Rv0111 immunodetection, respectively.
(239) Moreover, in some case, additional fusion products were also observed. In particular, dimers were detected for fusion no 13 and fusion no 6 likely resulting of the ability of Rv2626 to form dimers resistant to reducing conditions. Concerning fusion no 13, expression of the entire fusion no 13 (expected size 98.4 kDa) was indeed detected but at low level. Major proteolytic products were observed with anti-Rv2626 (around 70 kDa) and with anti-Rv0111 (around 30 kDa), suggesting a proteolytic cleavage of the fusion no 13.
(240) Similar level of expression was detected for fusions no 4 and no 11 which contain the same antigens (RPFB-Dhyb*-Ag85B*-TB10.4-ESAT6) but either membrane-anchored (fusion no 4) or cytoplasmic (fusion no 11). A higher band than the expected size (115 kDa instead of 96.7 kDa) was observed for fusion no 4, corresponding probably to N-glycosylated products. Minor proteolytic products were also detected for both fusions.
(241) Similar level of expression was also revealed with anti-Rv3407 in cell lysates of MVA expressing fusions no 5 and no 9 (both corresponding to the fusion of Rv0569-Rv1813*-Rv3407-Rv3478-Rv1807 antigens but expressed in membrane anchored form (fusion no 5) or in cytoplasmic (fusion no 9)). A higher band than the expected size (120 kDa instead of 98.4 kDa) was present in fusion no 5-expressing cell lysates, corresponding probably to N-glycosylated products. On the other hand, fusion no 5 was very weakly detected with anti-Rv1807 antibody while it is not the case for the cytoplasmic version. It is assumed that Rv1807 specific epitopes might be inaccessible in the membrane-anchored fusion due to the adjacent TMR sequence.
Example 7: Evaluation of the Immunogenicity of MVA Candidate Vaccines Expressing Mtb Antigens
(242) 7.1 Evaluation of Immunogenicity of MVA Candidate Vaccines Expressing Mtb Antigens in BALB/c Mice
(243) BALB/c mice were immunised with MVATG18365 and MVATG18364 both expressing Rv2029-Rv2626-Rv1733-Rv0111 (corresponding to fusion no 13) as well as RpfB-Dhyb-Ag85B-TB10.4-ESAT6 (corresponding to fusion no 4 or no 11). In the fusion no 13, a SS domain is present at the N-terminus and Rv1733 and Rv0111 are expressed with a TM domain which should direct expression of the fusion to the cell surface. Fusion no 4 expressed by MVATG18364 contains both a SS and a TM domain whereas fusion no 11 (MVATG18365) does not and should theoretically retain a bcytoplasmic expression. Specific-cellular immune responses were evaluated one week after injection by IFN ELISpot assays following restimulation with peptide pools described herein. Mice were also immunized with empty MVA vector (MVATGN33.1) as a negative control.
(244) As illustrated in
(245) In addition, all recombinant MVA candidate vaccines described in Example 6 section were injected in BALB/c mice and cellular immune responses specific of all Mtb antigens were assessed by IFN ELISpot assays as described in Materials and Methods section. A summary of the scope and intensity of responses induced by each MVA candidate in BALB/c mice is described in
(246) 7.2 Evaluation of Immunogenicity of MVA Candidate Vaccines Expressing Mtb Antigens in Transgenic HLA-A2 Mice
(247) As we have observed in the DNA-based studies, the mice haplotype has an influence on immunogenicity of the selected Mtb antigens (see section 5). In order to further analyze immunogenicity of Rv1733 and Rv0569 antigens induced by MVA candidates, transgenic mice expressing human MHC class I molecule, HLA-A2, were injected with recombinant MVAs expressing both antigens. Cellular immune response was evaluated one week after injection by IFN ELISpot assay after restimulation with the peptide pools described herein. Mice were also immunized with empty MVA vector (MVATGN33.1) as a negative control. Specifically, HLA-A2 mice were immunized with MVATG18376 or MVATG18378 vaccines expressing Rv2029-Rv2626-Rv1733-Rv0111 (corresponding to fusion no 13), RpfB-Dhyb-Ag85B-TB10.4-ESAT6 (corresponding to fusion no 4) as well as Rv0569-Rv1813-Rv3407-Rv3478-Rv1807 (corresponding to fusion no 5 or no 9). In the fusion no 5 expressed by MVATG18376, SS and TM domains were expressed at the N-terminus and C-terminus part, respectively.
(248) 7.3 Evaluation of Immunogenicity of MVA Candidate Vaccines Expressing Mtb Antigens in C57Bl/6 Mice
(249) H-2.sup.b haplotype C57BL/6 mice were immunized with MVATG18377 or MVATG18379 vaccines expressing Rv2029-Rv2626-Rv1733-Rv0111 (corresponding to fusion no 13), RpfB-Dhyb-Ag85B-TB10.4-ESAT6 (corresponding to fusion no 11) as well as Rv0569-Rv1813-Rv3407-Rv3478-Rv1807 (corresponding to fusion no 5 or no 9) in order to demonstrate immunogenicity of Rv0569 and Rv1733 antigens. Cellular immune response was evaluated one week after injection by IFN ELISpot assays after restimulation with the peptide pools described herein. Mice were also immunized with empty MVA vector (MVATGN33.1) as a negative control. Cellular IFN responses are summarized in
(250) 7.4 Evaluation of Immunogenicity of MVA Candidate Vaccines Expressing Mtb Antigens in C3H/HeN Mice
(251) As immunogenicity specific of Rv1733 has been demonstrated in H-2.sup.k haplotype C3H/HeN mice vaccinated with plasmids (see section 5.4), MVATG18376, MVATG18378, MVATG18377 and MVATG18379 expressing fusions containing the Rv1733 protein were injected to this mouse strain. Cellular immune response was evaluated one week after injection by IFNELISpot assays after restimulation with the peptide pools described herein. Mice were also immunized with empty MVATGN33.1 vector as a negative control. As illustrated in
(252) In addition to MVATG18377, immune responses induced by MVATG18376, MVATG18378 and MVATG18379 in C3H/HeN mice are illustrated in
(253) Overall, immunization with MVA vectors as well as with DNA plasmids leads to induction of strong and specific cellular responses targeting all Mtb antigens included in the fusions described in the present application. Humoral immune responses specific of two tested antigens were also detected in DNA-immunized mice. As with DNA plasmids, membrane-anchorage of the MVA-expressed Mtb fusions improves to some extent the level of induction of specific immune responses.
Example 8: Production and Purification of the Mtb Antigens
(254) 8.1 Optimal Conditions for Biomass Production of the Selected Mtb Antigens
(255) Four E. coli strains have been tested for the expression of the individual Mtb antigens as well as different culture conditions (e.g. temperature).
(256) These assays highlight that all the 14 selected antigens could be expressed at least in one bacterial strain at one defined temperature but significant differences were observed from one Mtb antigen to another. Indeed, some Mtb antigens could be easily produced in various E. coli strains and whatever the culture conditions (e.g. Rv0111, Rv0569, Rv1807, Rv2029, Rv2626, RpfB-D fusion) while other antigens require very specific host cells and conditions (e.g. Rv1733, Rv1813, TB10-4). On the other hand, high expression levels could be obtained for most of Mtb antigens in the different E. coli strains except Rv3407, Ag85B and Rv1813 expressed at lower but detectable levels. Moreover, certain Mtb antigens are produced as soluble material (e.g. Rv2626, Rv3407 and Ag85B that could be collected directly from cell lysate supernatants) while others are in insoluble material (e.g. RPFB-D, Rv0111, Rv1733, Rv2029, Rv3478, Rv1807, ESAT6 and TB10.4 that are collected from the pellet after cell lysis). Interestingly, Rv0569 is soluble when produced from transformed B121 cells cultured at 18 C. and both in soluble and insoluble material (in supernatant and pellet after lysis) when the B121 cells are cultured at 37 C.
(257) 8.2 Purification of the Mtb Antigens
(258) As described in Materials and Methods, Mtb antigens were purified by IMAC chromatography on nickel columns, eventually followed by gel filtration columns.
(259) Representative purification assays are shown for Rv2626 (purified from soluble material produced in C41 (DE3) cells at 37 C.), RPFB-D fusion (denatured RpfB-D purified from solubilized inclusion bodies produced in B121 (DE3) at 37 C.) and for TB10.4 (purified from soluble and insoluble material produced in C41 (DE3) cells at 37 C. The eluted fractions were assayed on SDS-PAGE as shown in
(260) As illustrated in
(261) When visualized on SDS-PAGE (lanes 1 to 8 represent intermediate purification fractions and lanes 9 to 11 5, 10 and 15 L of purified pool), the RPFB-Dhyb purified pool did not show any visible contaminant (see
(262) TB10-4 was purified in denaturing conditions followed by a final step in native conditions. As illustrated in
(263) In the three cases, endotoxin levels were measured in the purified pools and showed to be at a maximum level of 10 EU/mg protein.
(264) Therefore, the three proteins have been purified with acceptable amount, purity and endotoxin level.
(265) As a summary, the present invention provides an optimized combination of Mtb antigens. 14 Mtb antigens were selected after extensive bibliographic, data mining scoring and biochemical in silico analyses and cloned in plasmid vectors either individually or in the form of fusions. As demonstrated by Western blotting, all fusions were expressed at high levels and detected at the correct expected size following immunodetection with a series of antibodies directed against tags present at the N and C termini or against each Mtb antigen present in the fusion. Immunization assays in BALB/c mice support the immunogenic potential of the selected Mtb antigen combinations and fusions for inducing T cell responses.
(266) Moreover, the selected Mtb antigens (RpfB and RpfD in fusion) were individually produced in bacteria by recombinant means. Conditions for expression in E. coli were optimized by studying criteria such as bacterial strains and culture conditions (e.g. growth temperature). All proteins have been successfully expressed and produced at a litter scale.
Example 9: Evaluation of Therapeutic Efficacy of Mtb Antigen-Containing Vaccines Against Mycobacterium tuberculosis Infection in Mice
(267) Therapeutic efficacies of three Mtb-expressing MVA candidates, MVATG18364, MVATG18376 and MVATG18377 were investigated in a therapeutic setting co-administered with antibiotics in mice that were previously infected with Mtb strain H37Rv. As negative control, MVATGN33.1 was also injected in one group. Bacterial load was evaluated in spleens collected from the treated mice at the end of antibiotic treatment (at week 15 post-infection) and 6 weeks after (at week 21 post-infection). Groups of mice, drug regimen and immunization schedule are described in Table 5.
(268) TABLE-US-00005 TABLE 5 INH/RIF Treatment Pre-treat (week 6 to week 15) Week 0 CFU MVA MVA CFU CFU N Aerosol (day 1, Week Week Week (relapse) Group mice Treatment challenge Week 6) 10 14 15 Week 21 1 5 None 5 5 2 20 INH/RIF 20 5 15 3 20 INH/RIF + 20 20 20 5 15 MVATG18376 4 20 INH/RIF + 20 20 20 5 15 MVATG18377 5 20 INH/RIF + 20 20 20 5 15 MVATG18364 6 20 INH/RIF + 20 20 20 5 15 MVATGN33.1
(269) Six weeks post-Mtb infection and before starting chemotherapy and MVA immunization, the mycobacteria developed in the spleen of all mice groups (2.64 log.sub.10 total cfu). As expected, mycobacterial load decreased during chemotherapy treatment. In group 2 treated only with chemotherapy, the Mtb level decreased progressively to reach 1.18 log.sub.10 total cfu at week 15 and 0.70 log.sub.10 total cfu at week 21. Interestingly, at week 15, the mycobacterial loads were lower in mice co-treated with antibiotics and Mtb antigens-expressing MVA (0.70 log.sub.10 total CFU at week 15 in groups 3-5) than in mice treated with antibiotics only (group 2) and with the control empty MVA in combination with antibiotics (group 6), suggesting that Mtb-expressing MVA contributed to a stronger antibacterial effect. It is noteworthy that 6 weeks after the end of treatment (week 21), mycobacteria did not proliferate and loads were controlled in MVA-vaccinated mice (ranging from 0.70 to 0.85 log.sub.10 total cfu) at a level similar to the one observed in antibiotics therapy alone-treated mice. As control, the empty MVATGN33.1 vector combined with drugs therapy did not induce any anti-mycobacterial effect better than antibiotic regimen only at week 15 and week 21.