PROTEIN COMPRISING AT LEAST ONE REGULATORY T CELL ACTIVATING EPITOPE

Abstract

The present invention relates to the field of immunology, in particular, to the field of modulation of immune responses, in particular, suppression of immune responses and/or induction of tolerance. It provides a tregitope (regulatory T cell activating epitope) carrying polypeptide based on sequences derived from the Fc part of human IgG, wherein said TCP comprises at least one tregitope heterologous to human IgG that is located within at least one of three specific sequence frames. The invention provides such polypeptides for multiple purposes, e.g., in monomeric or dimeric form, wherein both are optionally be linked to an agent, e.g., to which an immune response is to be modulated or suppressed, or co-administered to such an agent, or for use as a stand-alone therapeutic. Nucleic acids encoding the TCP of the invention, pharmaceutic compositions and uses of said TCP are also provided.

Claims

1. A tregitope carrying polypeptide (TCP) comprising an amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, wherein said TCP comprises at least one tregitope heterologous to SEQ ID NO: 1 that is located within at least one of sequence frames A, B, or C, wherein (a) sequence frame A corresponds to positions 168 to 203 of SEQ ID NO: 1, and (b) sequence frame B corresponds to positions 272 to 307 of SEQ ID NO: 1, and (c) sequence frame C corresponds to positions 212 to 249 of SEQ ID NO: 1, wherein sequence frames A, B, and C are not taken into account for determining the sequence identity.

2. The TCP according to claim 1, wherein the heterologous tregitope does not occur identically in the same position in an amino acid sequence having at least 85% sequence identity to SEQ ID NO: 1.

3. The TCP according to claim 1, comprising at least two heterologous tregitopes, preferably, at least three or four, wherein, optionally, a first heterologous tregitope is located in one of frames A, B, or C, and wherein at least a second tregitope is located in a different frame of frames A, B, C, or C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1.

4. The TCP according to claim 1, wherein (a) if sequence frame A contains no heterologous tregitope, said frame A has at least 85% sequence identity with positions 168 to 203 of SEQ ID NO: 1, and (b) if sequence frame B contains no heterologous tregitope, said frame B has at least 85% sequence identity with positions 272 to 307 of SEQ ID NO: 1, and (c) if sequence frame C contains no heterologous tregitope, said frame C has at least 85% sequence identity with positions 212 to 249 of SEQ ID NO: 1.

5. The TCP according to claim 1, wherein the at least one heterologous tregitope substitutes a sequence in any of frames A B or C having the same length as said tregitope or having the length of the tregitope plus or minus one or two amino acids.

6. The TCP according to claim 1, wherein at least one heterologous tregitope is selected from the group consisting of: SEQ ID NO: 10 (Treg289), SEQ ID NO: 7 (Treg084), SEQ ID NO: 2 (Treg009A), SEQ ID NO: 9 (Treg088x), SEQ ID NO: 8 (Treg134), SEQ ID NO: 3 (Treg029B), SEQ ID NO: 4 (Treg088), SEQ ID NO: 5 (Treg167), SEQ ID NO: 6 (Treg289n-native), SEQ ID NO: 11 (trimmed Treg009A), SEQ ID NO: 12 (trimmed Treg029B-v1), SEQ ID NO: 13 (trimmed Treg029B-v2), SEQ ID NO: 14 (trimmed Treg088), SEQ ID NO: 15 (trimmed Treg088x-v1), SEQ ID NO: 16 (trimmed Treg088x-v2), SEQ ID NO: 17 (trimmed Treg167), SEQ ID NO: 18 (trimmed Treg289n), SEQ ID NO: 19 (trimmed Treg289), SEQ ID NO: 20 (trimmed Treg084), and SEQ ID NO: 21 (trimmed Treg134); wherein, preferably, all tregitopes are selected from said group.

7. The TCP according to claim 1, wherein said TCP is selected from the group consisting of (I) a TCP comprising (a) a tregitope according to SEQ ID NO: 2 (Treg009A) located in frame A, and (b) a tregitope according to SEQ ID NO: 2 (Treg009A) located frame B, and (c) a tregitope according to SEQ ID NO: 7 (Treg084) located in frame C, and (d) a tregitope according to SEQ ID NO: 9 (Treg088x)C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, optionally linked to said sequence via a linker of 3-18 amino acids; (II) a TCP comprising (a) a tregitope according to SEQ ID NO: 9 (Treg088x) located in frame B, and (b) a tregitope according to SEQ ID NO: 2 (Treg009A) located in frame C; (III) a TCP comprising (a) a tregitope according to SEQ ID NO: 10 (Treg289) located in frame B, and (b) a tregitope according to SEQ ID NO: 9 (Treg088x)C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, optionally linked to said sequence via a linker of 3-18 amino acids; (IV) a TCP comprising (a) a tregitope according to SEQ ID NO: 10 (Treg289) located in frame A, and (b) a tregitope according to SEQ ID NO: 7 (Treg084) located in frame C, and (c) a tregitope according to SEQ ID NO: 8 (Treg134)C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, optionally linked to said sequence via a linker of 3-18 amino acids; (V) a TCP comprising (a) a tregitope according to SEQ ID NO: 10 (Treg289) located in frame A, (b) a tregitope according to SEQ ID NO: 8 (Treg134) located in frame B, and (c) a tregitope according to SEQ ID NO: 7 (Treg084) located in frame C; (VI) a TCP comprising (a) a tregitope according to SEQ ID NO: 10 (Treg289) located in frame A, and (b) a tregitope according to SEQ ID NO: 7 (Treg084) located in frame C, and (c) a tregitope according to SEQ ID NO: 9 (Treg088x)C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, optionally linked to said sequence via a linker of 3-18 amino acids; and (VII) a TCP comprising (a) a tregitope according to SEQ ID NO: 7 (Treg084) located in frame C, and (b) a tregitope according to SEQ ID NO: 8 (Treg134)C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, optionally linked to said sequence via a linker of 3-18 amino acids.

8. The TCP according to claim 1, wherein the TCP optionally comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 23 to 44 and 46 to 58 and 111, optionally SEQ ID NO: 54.

9. The TCP according to claim 1, wherein the TCP consists of from 195 to 350 amino acids, wherein the TCP optionally essentially consists of the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1, wherein said TCP comprises at least one tregitope heterologous to SEQ ID NO: 1 that is located within at least one of sequence frames A, B, or C, wherein (a) sequence frame A corresponds to positions 168 to 203 of SEQ ID NO: 1, and (b) sequence frame B corresponds to positions 272 to 307 of SEQ ID NO: 1, and (c) sequence frame C corresponds to positions 212 to 249 of SEQ ID NO: 1, wherein sequence frames A, B, and C are not taken into account for determining the sequence identity, optionally, with one further tregitope C-terminal to the amino acid sequence having at least 85% sequence identity with amino acids 135 to 330 of SEQ ID NO: 1 that may be linked to said sequence via a linker of 3-18 amino acids.

10. The TCP according to claim 1, wherein the TCP wherein the TCP further comprises a VH domain and CH1 domain of an antibody, preferably, an antigen-binding part of an antibody.

11. The TCP according to claim 1, wherein said TCP forms a multimer comprising at least two, three, four, five, six, or more TCP monomers, preferably, a dimer comprising at least two TCP monomers according to any one of the preceding claims, wherein, optionally, said TCP monomers are covalently linked via at least one disulfide bridge.

12. The TCP according to claim 1, wherein the TCP is covalently or non-covalently linked to an agent selected from the group comprising (a) an allergen, (b) an intolerance inducing agent, (c) a target protein of an autoimmune response, e.g., of an autoantibody, (d) a target epitope of an autoimmune response, or (e) a therapeutic agent, wherein, optionally, the TCP and the agent form a fusion protein.

13. A nucleic acid encoding the TCP according to claim 1, wherein the nucleic acid optionally is an expression vector suitable for expressing the TCP in an eukaryotic host cell.

14. A host cell comprising the nucleic acid according to claim 13.

15. A method of manufacturing a tregitope carrying polypeptide (TCP), comprising the steps of (a) cultivating the host cell according to claim 14 under conditions suitable for expression of the TCP (b) harvesting the cell or medium comprising the TCP expressed in step (a), (c) isolating said TCP, (d) optionally, formulating the TCP of step (c) as a pharmaceutically acceptable composition.

16. A transgenic, preferably, non-human animal comprising the nucleic acid according to claim 13.

17. A pharmaceutical composition comprising the TCP according to claim 1, and, optionally, a pharmaceutically acceptable carrier and/or excipient.

18. The pharmaceutical composition according to claim 15 for use in modulating an immune response, preferably, for suppressing an immune response or inducing tolerance in a subject, wherein, optionally, said immune response is an immune response to an agent with which the TCP is co-administered, e.g., in covalently linked form.

19. The pharmaceutical composition according to claim 17 for use in the prevention or treatment of an autoimmune related disorder, allergy, viral infection, or transplantation-related immune reaction or disorder in a subject, preferably, for use in the treatment of an autoimmune disorder.

Description

FIGURE LEGENDS

[0525] FIG. 1: Ig domain structure according to Kuby, Immunology, Seventh Edition, W. H. Freeman & Co., New York, 2013, with an indication of the origin of tregitope sequences.

[0526] FIG. 2: Sequence structure of the constant part of the human IgG heavy chain (P01857; SEQ ID NO: 1) and the preferred Fc-part sub-sequence of positions 104-330 of SEQ ID NO: 1 (SEQ ID NO: 60) as a carrier molecule sequence. Tregitope substitution frames identified by multiple sequence alignment are highlighted.

[0527] FIG. 3: The Fc-part sub-sequence (SEQ ID NO: 60) with intramolecular disulfide bonds, domain boundaries, and substitution frames with local residue numbering.

[0528] FIG. 4: Excerpt from the ClustaIX alignment of the tregitope sequences Treg289, Treg167, Treg009A, Treg029B, Treg084, and Treg134 (SEQ ID Nos: 10, 5, 2, 3, 7 and 8) with the full carrier molecule sequence (P01857, pos. 150-220 of SEQ ID NO: 1 are shown, SEQ ID NO: 104). Additional alignment of trimmed Treg088x-v1 against P01857, pos. 104-330 was performed and added. The overall alignment with the full carrier molecule sequence defines frame A of 30 residues.

[0529] FIG. 5: Excerpt from the ClustaiX alignment of the tregitope sequences Treg289, Treg167, Treg009A, Treg029B, Treg084, and Treg134 (SEQ ID Nos: 10, 5, 2, 3, 7 and 8) with the partial carrier molecule sequence (P01857, pos. 241-310 of SEQ ID NO: 1 are shown, SEQ ID NO: 105). Additionally, an alignment of trimmed Treg088x-v1 (SEQ ID NO: 15) against P01857 was performed and added. The overall alignment with this partial carrier molecule sequence defines frame B of 28 residues.

[0530] FIG. 6: Excerpt from the ClustaIX alignment of the tregitope sequences Treg289, Treg167, Treg009A, Treg029B, Treg084, and Treg134 (SEQ ID Nos: 10, 5, 2, 3, 7 and 8) with the partial carrier molecule sequence (P01857, pos. 205-250 of SEQ ID NO: 1 are shown, SEQ ID NO: 106). The flanking cysteines have not been included into the alignment target. Additionally, an alignment of trimmed Treg088x-v1 (SEQ ID NO: 15) against P01857 was performed and added. The overall alignment with this partial carrier molecule sequence defines frame C of 32 residues.

[0531] FIG. 7: Predicted binding energies of single-substituted Fc homo-dimers into frame A (left bar, black), frame B (middle bar, grey), or frame C (right bar, light coloured). Covalent contributions from intermolecular disulfide bonds are ignored.

[0532] FIG. 8: Predicted binding energies of triple-substituted Fc homo-dimers. 1: tgp0084fa, tgp0167fb, tgp009Afc; 2: tgp0134fa, tgp029Bfb, tgp0167fc; 3: tgp029Bfa, tgp0289fb, tgp0134fc; 4: tgp0167fa, tgp009Afb, tgp0084fc; 5: tgp0289fa, tgp0134fb, tgp0084fc; 6: tgp0084fa, tgp0167fb, tgp009Afc; 7: P01857 Pos. 104-330, unsubstituted carrier (tgp: tregitope, fa: frame A, fb: frame B, fc: frame C. Covalent contributions from intermolecular disulfide bonds are ignored. the single tregitopes are designated as tgp0084 etc.

[0533] FIG. 9: Binding Energies of Selected Hetero-Dimers. [0534] 1: P01857 #104-330 (glycosylated), unsubstituted carrier [0535] 2: P01857 #104-330 with tgp0289fa-tgp0134fb-tgp0084fc and P01857 #104-330 with tgp029B-tgp0289fb-tgp0134fc [0536] 3: P01857 #104-330 with tgp0289fa-tgp0134fb-tgp0084fc and P01857 #104-330 with tgp009Afa-tgp029Bfb-tgp0167fc [0537] 4: P01857 #104-330 with tgp0084fa-tgp0134fb-tgp029Bfc and P01857 #104-330 with tgp0167fa-tgp0289fb-tgp009Afc [0538] 5: P01857 #104-330 with tgp0289fa-tgp0134fb-tgp0084fc and P01857 #104-330 with tgp029Bfa-tgp0167fb-tgp009Afc [0539] 6: P01857 #104-330 with tgp029Bfa-tgp0289fb-tgp0134fc and P01857 #104-330 with tgp0289fa-tgp0134fb-tgp0084fc [0540] 7: P01857 #104-330 with tgp0134fa-tgp0289fb-tgp0084fc and P01857 #104-330 with tgp0167fa-tgp009Afb-tgp029Bfc [0541] 8: P01857 #104-330 with tgp009Afa-tgp0084fb-tgp0289fc and P01857 #104-330 with tgp0134fa-tgp029Bfb-tgp0167fc [0542] 9: P01857 #104-330 with tgp0134fa-tgp0167fb-tgp009Afc and P01857 #104-330 with tgp029Bfa-tgp0084fb-tgp0289fc [0543] 10: P01857 #104-330 with tgp029Bfa-tgp009Afb-tgp0167fc and P01857 #104-330 with tgp0084fa-tgp0134fb-tgp0289fc [0544] 11: P01857 #104-330 with tgp0084fa-tgp0167fb-tgp0134fc and P01857 #104-330 with tgp009Afa-tgp029Bfb-tgp0289fc

[0545] FIG. 10: Fc dimer as a tregitope carrying polypeptide molecule in the sense of the protein of the invention. Correlation of total energies of hetero-dimeric complexes [E(A:B)] (lower graph) and monomers [E(A), E(B)] (upper graphs) with binding energies.

[0546] FIG. 11: Expression analysis of construct V32 and three direct tregitopes (Dir-Treg). Plasmid DNA carrying sequence information either for construct V32 or Dir-Treg-01-FLAG or Dir-Treg-02-FLAG or Dir-Treg-03-FLAG (SEQ ID NO: 101-103), whereas each Dir-Treg describes three successively cloned tregitope sequences C-terminally followed by a FLAG-Tag, were nucleofected under identical conditions into CAP-T cells and protein expression was performed for 4 days. Cell supernatants were harvested by centrifugation. Construct V32 was diluted 1:10, Dir-Treg-Ox-FLAG supernatants were diluted 1:2 and samples were loaded onto a SDS-PAGE gel. Carboxy-terminal FLAG-BAP Fusion Protein (Sigma-Aldrich, P7457-.1MG) was used in a serial dilution as control. Precision Plus Protein All Blue Standard (Bio-Rad, 161-0373) was used for size identification. Subsequently after the SDS-PAGE run, proteins from the gel were blotted onto a PVDF membrane and fluorescent detection was carried out using anti-FLAG and anti-Fc antibodies. Expression of construct V32 resulted in dramatically higher protein amounts compared to Dir-Treg-Ox-FLAG variants.

[0547] FIG. 12: Western Blot analysis of FcTregV1, V3, V13, V14, V20, V23, V32 and V34 and the corresponding unmodified Fc-part (SEQ ID 60). CAP-T cells were transiently transfected with plasmids encoding the respective constructs and 4-days cell culture supernatants were loaded and separated by reduced SDS-PAGE. Western Blot analysis was carried out using an AffiniPure Mouse Anti-Human IgG, Fcγ fragment specific primary antibody and IRDye 800CW Donkey anti-Mouse secondary antibody. All tregitope carrying polypeptides are well expressed and secreted. Protein sizes obtained by appling the Precision Plus Protein All Blue Prestained Protein Standards are indicated.

[0548] List of Sequences [0549] SEQ ID NO: 1: human wt IgG constant regions [0550] SEQ ID NO: 2: Treg009A [0551] SEQ ID NO: 3: Treg029B [0552] SEQ ID NO: 4: Treg088 [0553] SEQ ID NO: 5: Treg167 [0554] SEQ ID NO: 6: Treg289n-native [0555] SEQ ID NO: 7: Treg084 [0556] SEQ ID NO: 8: Treg134 [0557] SEQ ID NO: 9: Treg088x [0558] SEQ ID NO: 10: Treg289 [0559] SEQ ID NO: 11: trimmed Treg009A [0560] SEQ ID NO: 12: trimmed Treg029B-v1 [0561] SEQ ID NO: 13: trimmed Treg029B-v2 [0562] SEQ ID NO: 14: trimmed Treg088 [0563] SEQ ID NO: 15: trimmed Treg088x-v1 [0564] SEQ ID NO: 16: trimmed Treg088x-v2 [0565] SEQ ID NO: 17: trimmed Treg167 [0566] SEQ ID NO: 18: trimmed Treg289n [0567] SEQ ID NO: 19: trimmed Treg289 [0568] SEQ ID NO: 20: trimmed Treg084 [0569] SEQ ID NO: 21: trimmed Treg134 [0570] SEQ ID NO: 22: signal peptide [0571] SEQ ID NO: 23-44 and 46-58: FcTregV1-V22 and V24-V36 [0572] SEQ ID NO: 59: na encoding signal peptide [0573] SEQ ID NO: 60: Fc-part sub-sequence [0574] SEQ ID NO: 61-96: na encoding FcTregV1-V22 and V24-V36 [0575] SEQ ID NO: 97: Dir-Treg01-FLAG [0576] SEQ ID NO: 98: Dir-Treg02-FLAG [0577] SEQ ID NO: 99: Dir-Treg03-FLAG [0578] SEQ ID NO: 100: FLAG sequence [0579] SEQ ID NO: 101: Dir-Treg01-FLAG DNA [0580] SEQ ID NO: 102: Dir-Treg02-FLAG DNA [0581] SEQ ID NO: 103: Dir-Treg03-FLAG DNA [0582] SEQ ID NO: 104: partial sequence of SEQ ID NO: 1 shown in FIG. 4 [0583] SEQ ID NO: 105: partial sequence of SEQ ID NO: 1 shown in FIG. 5 [0584] SEQ ID NO: 106: partial sequence of SEQ ID NO: 1 shown in FIG. 6 [0585] SEQ ID NO: 107: linker1 [0586] SEQ ID NO: 108: linker2 [0587] SEQ ID NO: 109: linker3 [0588] SEQ ID NO: 110: sequence or partial sequence of GS linker [0589] SEQ ID NO: 111: FcTregV32_variant

Examples

[0590] Molecular Modeling Study to Identify a Carrier Platform for Tregitopes

[0591] Tregitopes are peptides originally found in the constant region of human and primate type G immunoglobulins (IgGs) that are able to activate regulatory T cells. Recombinant production of these peptides, however, is extremely difficult. In accordance with their natural origin, the Fc-part of human IgG was selected as a cloning framework candidate for a set of different tregitopes (SEQ ID NOs: 2, 3, 5, 6, 7, 8). These sequences were originally derived from different domains of immunoglobulins as shown in FIG. 1. The aim of the present experiment was to identify suitable sequence frames for tregitope cloning and expression. The exact basic sequence used in this study was the UNIPROT sequence P01857 (SEQ ID NO: 1), positions 104 through 330 (FIG. 2), which comprises the CH2 and CH3 domains of the constant region of the heavy chain of human IgG1, starting with part of the hinge region. Sequence details are given in FIG. 3.

[0592] Potential substitution frames within the carrier molecule sequence were identified by CLUSTALX multiple sequence alignment. To maintain the epitope character of the tregitopes, an alignment without gaps in the tregitope sequences was generated. This was achieved by using the highest possible gap penalty value (=100). FIGS. 4, 5, and 6 show the alignment of the tregitope sequences with the carrier molecule sequence. Relative to the full carrier molecule sequence, the alignments in frame A (FIG. 4) are obtained. To avoid perturbing the intramolecular disulfide bonds, the full sequence was divided into sections, which do not cover the cysteine residues. Those sections were additionally aligned with the tregitopes. This leads to the alignments shown in FIG. 5 (frame B) and FIG. 6 (frame C). It should be noted that the homology for these other two sections is significantly lower than for the full sequence.

[0593] In its biologically active form, the Fc-part is a homo-dimer comprising CH2 and CH3 domains, which is covalently connected by intermolecular disulfide bonds in the hinge region (FIG. 1). A model based on homology to similar sequences was built by a computational method called homology modeling using the YASARA software suite. The energetically most favorable structure was selected as the resulting model structure. Poor convergence and failure to form a dimer predicted by the software were taken as indications for real-world folding problems and interpreted as forecast of instability. Based on the calculated structure models, it could be shown that frame A entirely belongs to domain CH2, while frame B is more or less in the middle of domain CH3. Frame C overlaps with domain CH2 and comprises the domain boundary between CH2 and CH3. As a consequence, frame A was supposed to be the least critical area for tregitope substitution, while frames B and C may have more pronounced impact on the structures of the monomers, of the dimer, and the dimer binding energy.

[0594] The fold of the substitution frames was predicted to always comprise a δ-strand and a loop or short helix at either or both ends. In each case, the δ-strand is paired with other strands in the same domain, underlining a close coupling with the other secondary structure elements of the carrier. However, none of the frames is directly involved in intermolecular interactions with another Fc molecule chain. A correctly folded carrier should display a binding energy comparable to the existing carrier structure (P01857). Formation of the intermolecular disulfide bonds in the hinge region can only be expected if a stable dimer structure is formed. It has been shown that the interaction between the CH3 domains is the dominant contribution to this dimer formation. Thus, the CH3-CH3 interaction energy of the model is a useful criterion for a first validation of a predicted structure. For prediction of annealing and minimization, no water molecules were used. Instead, a special force field, the YASARA NOVA force field, which has been parametrized to reproduce crystal structures as close as possible, has turned out to be a reasonable compromise between computing effort and precision for dimerization energy assessment.

[0595] Practically, in the case of Fc-part variants (approximately 7500 atoms), a number of 36 cycles of simulated annealing and steepest descent minimization following the homology modeling process, turned out to be a useful strategy to achieve convergence in structures, total energies, monomer energies, and binding energies. Nevertheless, classical force-field based models are limited in their precision and should only be interpreted for comparison of similar structures and the qualitative deduction of trends therefrom. Reference to reliable experimental data has been done in order to increase the reliability of the results.

[0596] The following structure variants have been analyzed: [0597] a) Single-tregitope equipped Fc homo-dimers (1 tregitope per Fc molecule for each frame) [0598] b) Triple-tregitope equipped Fc homo-dimers (3 tregitopes per Fc molecule, 1 tregitope in each frame) [0599] c) Sextuple-tregitope equipped Fc hetero-dimers (3 tregitopes per Fc molecule, 1 tregitope in each frame, different tregitopes in each Fc monomer) [0600] d)C-terminal attachment of tregitopes [0601] e) Combinations of a), b), and d)

[0602] FIG. 7 shows the dimer binding energies of a TCP (for case a). It has to be noted that these and all the other results ignore the covalent contribution from the two disulfide bonds in the hinge region. Their contribution is assumed to be identical for all tregitope insertion variants. As expected, any substitution of the basic sequence with tregitopes leads to a reduction of binding energy. Modified, respectively missing glycosylation are likely to contribute significantly to this difference. It should be noted that none of the variants modeled did have the same glycosylation pattern as the original Fc fragment. The carrier model structure bears the glycosylation pattern from its leading template, the PDB crystal structure 3SGK. This structure is glycosylated at Asn77 of SEQ ID NO: 60 (cf. FIG. 3). This is position Asn180 in SEQ ID NO: 1 (cf. FIG. 4). It is clear that glycosylation could at most be expected with tgp084, tgp0134 and with tgp088x, which also have a Asn in this very position. Insertion of all other tregitopes in frame A leads to a loss of this glycosylation site. From the data, it is not clear to which extent glycosylation affects the dimer binding energy. On the one hand, tgp0289, which differs from the original Fc sequence only by the substitution N180Q (see FIG. 4), is closest to the Fc dimer, however, a difference in binding energy of 135 kJ/mol is surprising. It mainly originates from differences in dihedral and electrostatic energy. In particular, the difference in electrostatic energy is a hint towards solvation phenomena, which are not taken into account explicitly with the force filed used. The difference in binding energy may hence be overestimated by the force field. With the exception of tgp0084 and tgp0134, this also applies to frame C. The diagram also confirms the critical role of frame B, which directly affects the dimer binding energy. Tgp009A should be advantageously introduced into frame A.

[0603] FIG. 8 shows the results for case b), the triple substituted Fc-part variants as homodimers. First of all, there seems to be a certain synergistic effect on the binding energy from the triple substitution of tregitopes, comparing the data with those from FIG. 7. The error bars show the standard deviation (fluctuations) of the binding energy during the optimization process.

[0604] Examples of case c) substitutions are given in FIG. 9. These are no longer homo-dimers but hetero-dimers, which do not have any intrinsic symmetry. With the exception of variant 2, all hetero-dimer variants are significantly less stable than the unsubstituted Fc dimer. However, in comparison with the homo-dimers as of FIG. 8, FIG. 9 shows that only variant 11 is significantly less stable than those homo-dimers. In principle, it is possible to generate hetero-dimers containing six different tregitopes experimentally. Variants 7, 8, and 9 could also have at least one glycosylated monomer.

[0605] FIG. 10 gives an overview of the energy situation of hetero-dimers. The majority of variants shows binding energies between −250 kJ/mol and −320 kJ/mol with total complex energies between −8740 kJ/mol and −11000 kJ/mol. Exceptions in this representation are variant 11, as defined in FIG. 9, with a very weak binding energy (−154 kJ/mol binding energy; −8214 kJ/mol total complex energy)), the very stable variant 2 (−424 kJ/mol binding energy; −11481 kJ/mol total complex energy), and the unsubstituted carrier variant 1 (−515 kJ/mol binding energy; −10123 kJ/mol total complex energy). Variant 2 appears to be unique in that there is a fortuitous cancellation of stability problems caused by the tregitopes. A certain indication of consistency is the weak correlation found between total energies and binding energies. Furthermore, as the number of atoms does not differ dramatically between the variants (7200-7600 atoms) the vertical spread of the energies gives the order of magnitude for both, force field precision and entropic contributions.

[0606] A preliminary analysis of direct tregitope attachment to the C-terminus of the Fc-parts revealed a destabilization of the respective dimers (data not shown). The reason may be related to the fact that the carboxy group of the C-terminal arginine is not surface accessible, but hidden in the internal of the dimer structure. Any modification, which leads to a change of position of the C-termini may lead to a deformation of the CH3 domain and reduces the dimer binding energy. Three linker versions (SEQ ID NO: 107-109) have been analyzed with three different tregitope constructs without frame B substitution. One of the linkers, linker 3 (SEQ ID NO: 109), has also been used with a truncated Fc molecule missing the normal C-terminal lysine residue. It was found that linker 2 (PTGSG; SEQ ID NO: 108) gives an improvement of the binding energy, which is more pronounced with tgp029B as C-terminal attachment (e.g., variant 3; carrier sequence with tgp009Afa (Treg009A in frame a), no fb (no tregitope in frame B), tgp0084fc (Treg084 in frame C, and tgp029B (Treg029B)C-terminal attachment—Homodimer) than with tgp0289 (variant 1; carrier sequence with tgp009Afa, no fb, tgp0084fc, and tgp0289 C-terminal attachment—Homodimer).

[0607] Expression of Different TCPs

[0608] 36 different expression constructs for TCPs (also designated FcTreg herein), constructs FcTregV1 up to FcTregV22 and FcTregV24-FcTregV36, were prepared. FcTregV23 was also prepared (SEQ ID NO: 45). The amino acid sequences (SEQ ID NO: 23-44 and 46-58) and nucleic acid sequences (SEQ ID NO: 61-82 and 84-96) of the respective TCP variants are provided in the sequence listing of the present disclosure. For secretion of all constructs, a Fc signal peptide was used, e.g., Fc-Signal_AA (SEQ ID NO: 22): METDTLLLWVLLLWVPGSTG.

TABLE-US-00010 This signal sequence was encoded by Fc-Signal_DNA  (SEQ ID NO: 59): ATGGAAACCGACACACTGCTGCTGTGGGTGCTGCTTTTGTGGGTGCCAGG CAGCACCGGC.

[0609] The signal sequence was added at 5′ terminus of the DNA respectively N-terminus of the protein. The signal peptide is cleaved off during transport and secretion of the protein.

[0610] For analysis of the expression and dimer formation, HEK293F cells and CAP-T cells have been used for transient expression of the constructs. CAP-T cells are an immortalized cell line based on primary human amniocytes and grow in suspension in PEM medium (Life Technologies) supplemented with 4 mM L-Glu. Compared to CAP Go cells, CAP-T cells additionally express the large T antigen of simian virus 40. The HEK 293-f cell line is derived from the original HEK 293 cell line and is adapted to suspension growth in serum-free medium. Transient transfection was done by electroporation using the commercially available Nucleofector™ system.

[0611] During the exponential growth phase of the culture, the CAP-T cells were counted by Cedex XS (Roche Applied Science, Innovatis) and viable cell density and viability were determined. For each nucleofection reaction, 1.Math.10.sup.7CAP-T cells were harvested by centrifugation (150×g for 5 min). The cells were resuspended in 100 μL complete nucleofector solution SE (Lonza, Switzerland) and mixed with the respective Fc-Treg construct (plasmid encoding the tregitope carrier molecule). The DNA/cell suspension was transferred into a cuvette and the nucleofection was performed using the X001 program on a Nucleofector II. After the pulse, cells were recovered by adding 500 μL prewarmed complete PEM medium (=supplemented with 4 mM L-alanyl-L-glutamine) to the cuvette and gently transferred into 11.5 mL complete PEM medium in a 125 mL shaking flask. The cuvette was washed once with 500 μL fresh medium to recover residual cells. The final cultivation volume was 12.5 ml. Electroporation was similarly performed with 7.Math.10.sup.6 HEK293-F cells and 7 μg plasmid. After transfection the cells were incubated for 4 days. Cell pellets and the supernatant were subsequently tested by Western Blot. Reference was transfection with Fc monomer. All tested constructs showed expression in the pellet, but differences in secretion were observed. There was a good correlation in between observed expression in HEK293F and CAP-T cells.

[0612] Molecules V1, V3, V13, and V14 gave good results in secretion and expression in HEK293F. V7, V9 and V12 resulted also in secretion and expression, although to a slightly lesser extent. The TCP performing best under these aspects were V1, V3, V13 and V14 (V13 and V14 were only tested in CAP-T cells). Further tests with supernatants of V15-V36 in CAP-T cells showed particularly good results for V20, V23, V32 and V34.

[0613] Thus, preferred TCP of the invention have the following structure, wherein frames not noted do not comprise a heterologous tregitope: [0614] (a) Treg289 in frame A, Treg084 in frame C, Treg134 C-terminal, e.g., V1 [0615] (b) Treg289 in frame A, Treg084 in frame C, Treg134 in frame B, e.g., V3 [0616] (c) Treg289 in frame A, Treg084 in frame C, Treg88 C-terminal, e.g., V13 [0617] (d) Treg084 in frame C, Treg134 C-terminal, e.g., V14 [0618] (e) Treg009A in frame C, Treg088x in frame B, e.g., V20 [0619] (f) Treg084 in frame A, e.g., V23 [0620] (g) Treg009A in frame A, Treg084 in frame C, Treg009A in frame B, Treg088x C-terminal, e.g., V32 [0621] (h) Treg289 in frame B, Treg088x C-terminal, e.g., V34.

TABLE-US-00011 Dimer Name Frame A Frame C Frame B C-terminal Expression formation Construct V1 Treg289 Treg084 Treg134 ++ ++ Construct V3 Treg289 Treg084 Treg134 + +/− Construct V13 Treg289 Treg084 Treg088x ++ +++ Construct V14 Treg084 Treg134 ++ + Construct V20 Treg009A Treg088x ++ +++ Construct V23 Treg084 ++ ++ Construct V32 Treg009A Treg084 Treg009A Treg088x ++ ++ Construct V34 Treg289 Treg088x ++ ++

[0622] An exemplary Western Blot of FcTregsV1, V3, V13, V14, V20, V23, V32 and V34 and the corresponding unmodified Fc-part (SEQ ID NO:60) is demonstrated in FIG. 12. Cell culture supernatants of transiently transfected CAP-T cells as described above, were prepared by mixing 20 μL of the respective supernatant with 10 μL NuPage LDS Sample Buffer (4×, Thermo Fisher), 4 μL NuPage Sample Reducing Agent (10×, Thermo Fisher) and 6 μl Aqua Dest. (B. Braun, Germany). Samples were denatured for 10 min at 70° C. and 10 μL of the prepared samples were loaded onto a NuPAGE gradient 4-12% BisTris gel. Precision Plus Protein All Blue Prestained Protein Standards (Bio-Rad) was applied as size marker. Reduced sodium dodecyl sulphate—polyacrylamide gel electrophoresis (SDS-PAGE) was performed at 200V and using MOPS buffer (40 mL NuPAGE MOPS SDS running buffer+760 mL Aqua dest. +500 μL NuPAGE Antioxidant). Subsequent Western Blotting was performed onto a nitrocellulose membrane at 30V for 60 min using a transfer buffer (50 mL NuPAGE Transfer Buffer (20×, Thermo Fisher)+1 mL NuPAGE Antioxidant+100 mL methanol+849 mL Aqua Dest.). In order to block unspecific antibody binding for detection, the membrane was first blocked with Odyssey Blocking Buffer (PBS) at 4° C. over night. Subsequently, AffiniPure Mouse Anti-Human IgG, Fcγ fragment specific (Jackson ImmunoResearch, 209-005-098) was used as primary antibody and incubated for 1 h at room temperature under gentle shaking, diluted 1:100 in a solution prepared by mixing 30 mL blocking buffer+0.05% Tween 20. The secondary antibody IRDye 800CW Donkey anti-Mouse (diluted 1:15000, Li-Cor) was incubated for 1 h at 4° C. Between and after the primary and secondary antibody incubation steps, the membrane was washed four times each with approx. 25 mL of PBS-T (500 mL PBS+0.1% of a 10% Tween 20 solution). After rinsing the membrane twice for 5 min with PBS, it was dried over night protected from light. The membrane was scanned using an Odyssey CLx Imager (Li-Cor) for antibody-marked protein band detection.

[0623] Western Blot results based on a reduced SDS-PAGE as shown in FIG. 12 clearly demonstrate the particular good expression and secretion of FcTregV1, V3, V13, V14, V20, V23, V32, and V34. TCPs are mostly visible as monomers in size ranges of approx. 26 to 36 kDa due to reduced conditions during the SDS-PAGE run. Multiple bands within individual FcTreg variants within this size range might be different post-translational modification species (e.g glycosylation) as proteins result from transient transfected pools and not from clonal expression. A small portion of dimerized TCP molecules remained as bands between approx. 55 and 70 kDa are visible.

[0624] These results show that it is possible to effectively express and produce tregitopes by integration into an immunoglobulin Fc-part according to the inventive approach.

[0625] The molecules V1, V3, V13, V14, V20, V23, V32 and V34 were thus chosen to generate CAP Go basic cell lines stably expressing the recombinant proteins. Transfection of CAP Go cells was carried out as described above for CAP-T cells, but using solution V instead of solution SE and running the transfection program X001 on a Nucleofector II. In addition, selection with blasticidin was started 72 h after nucleofection.

[0626] Several attempts to purify the recombinant TCP variants from cell culture supernatants by common affinity purification protocols via protein A/G or Thermo Scientific's FcXL column failed. The TCPs did not properly bind to the resin and were found in the flow through. A specific affinity purification strategy was developed by applying a polyclonal mouse anti-human IgG, Fc-gamma fragment specific antibody which has been shown to bind the recombinant protein variants in Western blot detections. This antibody (AffiniPure polyclonal mouse anti-human IgG, Fc-gamma fragment specific antibody (Jackson ImmunoResearch, Cat 209-005-098)) was used as capturing antibody for affinity chromatography.

[0627] The commercially available AffiniPure polyclonal mouse anti-human IgG, Fc-gamma fragment specific antibody (Jackson ImmunoResearch, Cat 209-005-098) was covalently conjugated to NHS-activated Sepharose 4 Fast Flow resin (GE Healthcare, Cat. 17-0906) by applying the following steps: [0628] (a) Anti-human IgG, Fc-gamma fragment specific antibody buffer was exchanged to coupling buffer (0.2 M NaHCO.sub.3, 0.5 M NaCl, pH 8.3). [0629] (b) NHS-activated Sepharose 4 Fast Flow matrix was washed 6 times with 1×matrix volume of 1 mM HCl and once with 1×matrix volume of coupling buffer. [0630] (c) 12.5 mL anti-human IgG, Fc-gamma fragment specific antibody (˜1 mg/mL) was added to 25 ml prepared NHS-activated Sepharose 4 Fast Flow resin. Conjugation was carried out at 2-8° C. overnight using a rotator. [0631] (d) Non-reacted groups of the matrix were blocked by incubation for ˜4 h with 0.1 M Tris-HCl, pH ˜8.5. [0632] (e) Washing of resin was carried out using 0.1 M Tris-HCl buffer, pH 8 to 9, and 0.1 M acetate buffer, 0.5 M NaCl, pH 4 to 5. The washing procedure was: 3×1 matrix volumes Tris buffer followed by 3×1 matrix volumes acetate buffer. This cycle was repeated 3 to 6 times. Finally, the resin was kept in 20% ethanol. [0633] (f) The resin was packed into a XK 16/40 or Tricorn 10/300 column.

[0634] For purification of the recombinant protein variants from cell culture supernatants of molecule expression preparations, the cell culture supernatants were firstly adjusted to pH 7.4. The supernatants were loaded onto the prepared affinity column with flow rates of 2-6 ml/min and pressure of 0.15-0.2 MPa. The column was then washed with DPBS (Dulbecco's phosphate-buffered saline). The recombinant protein variants were eluted using 100 mM glycine-HCl, pH 2.7. Flow rates and pressures were identical to the loading step. About 10% of the final fraction volumes was used for neutralization with 1 M Tris-HCl pH 8.8. The recombinant protein variants were rebuffered to PBS (phosphate-buffered saline) and concentrated (˜30×) using Pierce Protein Concentrators (Thermo, Cat: 88535). Optionally, Amicon ultrafiltration filters (Merck, Cat: ufc901024) were used for further concentration.

[0635] Bystander Suppression Assay

[0636] A bystander suppression assay, based on ex vivo stimulation of PBMC (peripheral blood mononuclear cells) of healthy donors with the corresponding antigen leading to a proliferative response with tetanus toxoid (TT assay), was used to assess the molecule constructs with respect to their inhibitory capacity on proliferation/activation of effector CD4 cells. When selecting donors, consideration was given to obtain an as wide as possible coverage of the human population (covering the main HLA-DR B1 supertypes), covering more than 95% of the allelic variability in the human population. Analysis of the incubated cells was performed with immunostaining with intracellular and cell surface markers and analyzed by flow cytometry. Inhibitory effect of the molecule constructs was observable as diminished proliferation and activation of effector CD4 T cells. Particularly effective molecule constructs were identified by statistical analysis.

[0637] The assay was performed by plating 3×10.sup.5 cells/well in 96-wells plates at day 0, each data point performed in duplicate. All subsequent operations including addition of stimuli, tregitopes, antibodies for immunostaining and flow cytometry set up were done without removing the cells from the plates. Stimulation of the PBMC was carried out at day 1 with 0.5 μg/mL tetanus toxoid (TT) in the presence of either 0, 10, 20, 40 or 80 μg/mL of the TCP constructs. Controls receiving only TCP constructs or only TT, as well as controls receiving none of these were included as well. Readout was carried out at day 7 following effector (proliferation, CD25), memory (CCR7, CD45RA) and regulatory (FoxP3, CD25) T cell markers.

[0638] The TCP variant V20 (containing tregitopes 009A and 088x) was tested in PBMC from two healthy donors using the TT suppression assay (see above). Native Fc was used as control. The suppressive response varied by donor, and according to the stimulation parameter measured. V20 at sub-micromolar to low micromolar concentrations suppressed the TT effector response when assessing CD69 (in both donors) or HLA-DR (in one donor) by more than 75% (once the background was subtracted). The order of susceptibility to suppression of the parameters measured as response to stimulation by TT was CD69>HLA-DR>proliferation>CD25. For one of the donors (EV0156), V20 suppressed all four stimulatory parameters tested in this study; strongly with regard to CD69 and HLA-DR, and more weakly with regard to proliferation and CD25. In the second donor (EV0159), V20 showed a strong effect on CD69, a weaker effect on HLA-DR, and no appreciable effect on proliferation or CD25.

[0639] Comparison of the Expression of a Recombinant Protein of the Invention with Directly Fused Tregitope Peptides

[0640] The goal of this experiment was to compare the expression of a TCP according to the invention with expression of three sequentially cloned tregitopes (direct tregitopes, Dir-Treg-FLAG) N-terminally fused to murine IgG1 signal peptide and C-terminally fused to a FLAG-Tag for detection or potential purification (see below Table and indicated SEQ ID NOs).

TABLE-US-00012 SEQ Tregi- Tregi- Tregi- Tregi-   ID Con- tope tope tope tope FLAG NO struct 1 2 3 4 Seq.  97 &  Dir- 289Q 084 134 — DYKDDDDK 101 Treg- (SEQ ID  01-FLAG NO: 100)  98 &  Dir- 009a 088x 084 — DYKDDDDK 102 Treg- (SEQ ID  02-FLAG NO: 100)  99 &  Dir- 088x 289Q 009a — DYKDDDDK 103 Treg- (SEQ ID  03-FLAG NO: 100)  92 Fc_Treg_ 009a 084 009a 088x — V32

[0641] CAP-T cells were cultured in PEM medium supplemented with 4 mM GlutaMAX (Thermo Fisher Scientific, 35050038) and 5 μg/ml blasticidin (Thermo Fisher Scientific, R21001; complete PEM medium). In order to thaw the cells, the required amount of frozen vials were transferred to a 37° C. water bath. After thawing, each vial was transferred to 10 mL of chilled, complete PEM medium. The cell suspension was centrifuged at 150×g for 5 minutes. During this washing step, the DMSO was removed. The pellet was resuspended in 15 mL warm, complete PEM medium and transferred to a 125 mL shaker flask. The cells were incubated at 37° C. in a humidified incubator with an atmosphere containing 5% CO.sub.2. The flasks were set on a shaking platform, rotating at 185 rpm with an orbit of 50 mm.

[0642] Subculturing of the cells was performed every 3 to 4 days. The fresh culture was set to 0.5×10.sup.6 cells/ml by transferring the required amount of cultured cell suspension to a new flask and adding complete PEM medium. In the case that the transferred cell suspension would exceed 20% of the total volume, the suspension was centrifuged at 150×g for 5 minutes and the pellet was resuspended in fresh complete PEM medium. The volume of cell suspension per shaking flask was 20% of the total flask volume. A minimum of three subcultures were performed after thawing before transfection experiments were performed.

[0643] The CAP-T cells were transfected using the 4D-Nucleofector. For each transfection, 10×10.sup.6 CAP-T cells were centrifuged at 150×g for 5 minutes in 15 ml conical tubes. The cells were resuspended in 95 μL supplemented SE Buffer, taking into account the volume of the pellet and the volume of the plasmid solution. Afterwards, 5 μg of the respective plasmid were added to the cell suspension followed by gentle mixing. The solution was transferred to 100 μL Nucleocuvettes. The used transfection program was ED-100. After the transfection, the cells from one Nucleocuvette were transferred to 125 mL shaker flasks, containing 12.5 mL complete PEM medium. The cells were cultivated for 4 days as described above. At day 4 the cells were harvested by centrifugation at 150×g for 5 minutes.

[0644] Supernatants of the protein of the invention were diluted 1:10 and supernatants of Dir-Treg-FLAG were diluted 1:2 with reducing sample buffer. Carboxy-terminal FLAG-BAP Fusion Protein (Sigma-Aldrich, P7457-1MG) was used in a serial dilution (final amount load to the gel: 640 ng, 320 ng, 160 ng, 80 ng, 40 ng, 20 ng) as control. Reducing sample buffer was produced by combining 2.5 parts of NuPAGE LDS Sample Buffer (4×, Thermo Fisher Scientific, NP0007) with 1 part of NuPAGE Sample Reducing Agent (10×, Thermo Fisher Scientific, NP0004). 20 μL of each sample were mixed with 20 μL of reducing sample buffer in a 1.5 mL vial and heated for 10 min at 70° C. using a thermoshaker (Eppendorf). A NuPAGE 4-12% Bis-Tris Protein Gel (Thermo Fisher Scientific) was inserted into the XCell SureLock Mini-Cell Electrophoresis System (Thermo Fisher Scientific) and inner and outer chambers were filled with 1×NuPAGE MES SDS Running buffer (Thermo Fisher Scientific, NP000202). 500 μL of NuPAGE Antioxidant (Thermo Fisher Scientific) was added to the inner chamber. 10 μL of the each prepared sample and 4 μL of Precision Plus Protein All Blue Standard (Bio-Rad, 161-0373) diluted 1/10 in 1×LDS Sample Buffer were loaded onto the gel. The sample separation was achieved by running the gel at a constant voltage of 200 V for 50-60 min.

[0645] To investigate the separated proteins by immunofluorescence detection, they were transferred onto an Amersham Hybond Low Fluorescence 0.2 μm polyvinylidene fluoride (PVDF) membrane (GE Healthcare Life Sciences) by using the XCell II Blot module (Thermo Fisher Scientific) for semi-wet protein transfer. The PVDF membrane was directly applied to the SDS gel and the system was filled with NuPAGE Transfer Buffer (20×, Thermo Fisher Scientfic) according to the manufacturer's instructions. Protein blotting was performed for 1 h at 30 V. After protein transfer, the membrane was blocked over night at 4° C. in Odyssey Blocking buffer (Licor) and incubated afterwards simultaneously with 2 μg/mL Monoclonal ANTI-FLAG M2 antibody (Sigma Aldrich, F1804-200UG) and 17 μg/mL AffiniPure Mouse Anti-Human IgG, Fcγ Fragment Specific (Jackson Immuno Research, 209-005-098) diluted in Odyssey Blocking buffer containing 0.05% Tween 20 for 1 h at room temperature. After incubation, the PVDF membrane was washed four times for 5 min in 0.1% PBST. For detection of proteins the membrane was cut into two pieces and the membrane part for FLAG-detection was incubated for 1 h with 0.067 μg/ml of IRDye 800CW Donkey Anti-Mouse (Licor). The other part of the membrane containing the protein of the invention was incubated with IRDye 680RD Donkey Anti-Mouse (Licor). Finally, the PVDF membrane was washed four times for 5 min in 0.1% PBST, two times for 5 min in PBS and rinsed in water. The membrane was visualized using the Licor Odyssey Imager. Band intensities were quantified using the Phoretix 1D software and expression rates between the protein of the invention and Dir-Treg-FLAG were compared.

[0646] A concentration-dependent immunofluorescence signal of the latter control protein was observed, demonstrating the quality of the anti-FLAG antibody detection (FIG. 11). Dir-Treg-01-FLAG was hardly expressed. Dir-Treg-03-FLAG was expressed in minimal amounts, while Dir-Treg-02-FLAG demonstrated the best expression of the sequential tregitope peptides. However, the expression of the protein of the invention (construct V32) was 9-times higher compared to Dir-Treg-02-FLAG and 20-times higher compared to Dir-Treg-03-FLAG.

[0647] This experiment clearly demonstrated the favorable expression of the TCP according to the invention compared to the expression of three different versions of three sequential tregitope peptides fused to a FLAG-tag.