Modular antigen transportation molecules and uses therof

Abstract

The present invention relates to (isolated) recombinant proteins, also referred to as improved MAT (iMAT) molecules, comprising at least one translocation module, at least one targeting module and at least one antigen module, wherein at least one cysteine residue is substituted with a different amino acid residue. Such iMAT molecules are useful specifically as vaccines, e.g., for therapy and/or prevention of allergies and/or infectious diseases and/or prevention of transmission of infectious diseases in equines. The present invention further relates to nucleic acids encoding such iMAT molecules, corresponding vectors and primary cells or cell lines.

Claims

1. A method of preventing and/or treatment of insect bite hypersensitivity (IBH), urticaria or combinations thereof in equines, comprising administering a pharmaceutically effective amount of an improved Modular Antigen Transportation (iMAT) molecule comprising: (a) at least one first module being an amino acid sequence allowing the translocation of the iMAT molecule from the extracellular space into the interior of cells, wherein the at least one first module is selected from: (i) the amino acid sequence of HIV-tat, VP22, and/or Antennapedia, or (ii) the amino acid sequence according to SEQ ID NO: 1; (b) at least one second module being an amino acid sequence allowing species-specific intracellular targeting of the iMAT molecule to the cell organelles which are involved in the processing of antigens and/or the loading of MHC molecules with antigens, wherein such at least one second module comprises one of the amino acid sequences selected from: SEQ ID NOS: 2, 15, 16; and (c) at least one third module as antigen module being an amino acid sequence derived from at least one epitope of at least one antigen, wherein such at least one antigen is an allergen derived from blood feeding insects from the genus Culicoides, characterized in that in the entire iMAT molecule all cysteine residues are substituted with a different amino acid residue.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1: shows examples of allergens of the genus culicoides identifying the species, the allergen and the uniprot accession number (database release is UniProt release 2013_09)

(2) FIGS. 2A and 2B show an SDS-PAGE of MAT-Fel d1 (IVN201) under reducing conditions; FIG. 2A) Original SDS-PAGE (10 g protein per lane, Coomassie staining) and FIG. 2B) SDS-PAGE with bands that have been cut out of gel A and reloaded on SDS-PAGE (silver staining)

(3) FIG. 3: shows an SDS-PAGE of Fel d1 under reducing conditions; NUPAGE 4-12% Bis-Tris Gel. Lanes: 1) Marker: SeeBlue Plus2 Pre-Stained Standard 2) 5 g Fel d1; 3) 5 g Fel d1

(4) FIG. 4: shows an RP-HPLC chromatogram 0.1% TFA/Acetonitril gradient a) native protein (MAT-Fel d1) without additives (left graph); b) MAT-Fel d1 denaturated by addition of guanidinium chloride plus DTT (right graph)

(5) FIG. 5: depicts MAT-Fel d1 molecule and its Kyte-Doolittle hydrophobicity plot

(6) FIG. 6: shows a NUPAGE SDS-PAGE-System (4-12% Bis-Tris Gels, 1MES-Running buffer, 35 min, 200 V). Lanes: 1) Lysozyme, 1 g Protein ox. 2) PageRuler Prestained Protein Ladder; 3) MAT-Fel d1 (5 g) oxidized with Iodacetamide (ox.); 4) iMAT-Cul o4 (ox.); 5) PageRuler Prestained Protein Ladder; 6) MAT-Fel d1 (5 g) reduced; 7) iMAT-Cul o4 reduced; 8) Lysozyme reduced

(7) FIG. 7: shows an RP-HPLC chromatogram 0.1% TFA/Acetonitril gradient. The peak reflects native (oxidized) protein (iMAT-Cul o4) without additives

(8) FIG. 8: depicts an example of display of counts of local alignments hits of selected Culicoides allergens in comparison to sensitizations measured (see description for details). The figure depicts the comparison between the antigenic prediction (light gray and right Y-axis) and the outcome (dark gray and left Y-axis) in a HRT test in terms of severity (total score values per allergen) in a cohort of IBH horses. The severity describes the degree of a reaction in the test in a semi-quantitative scale (left Y axis). The data from the prediction is scaled by setting the highest occurrence score equal to the highest severity score.

(9) FIG. 9: shows the results of the following experiment: Proteins and ADJU-PHOS that were incubated at RT for 30 min while mixing gently. After incubation, the samples were centrifuged for 3 min. and subsequently analyzed by SDS-Page Lane 1) pageRuler Prestained Protein Ladder; 2) iMAT-Cul o3 Supernatant; 3) iMAT-Cul o3 Pellet in Urea 4) iMAT-Cul o3 Supernatant; 5) iMAT-Cul o3 Pellet in Urea; 6) Empty; 7) iMAT-Cul o2 Supernatant 8) iMAT-Cul o2 Pellet in Urea 9) iMAT-Cul o2 Supernatant 10) iMAT-Cul o2 Pellet in Urea; 11) Empty; 12) iMAT-Cul o4 Supernatant; 13) iMAT-Cul o4 Pellet in Urea 14) iMAT-Cul o4 Supernatant; 15) iMAT-Cul o4 Pellet in Urea; (2, 3, 7, 8, 12, 13 w/o freeze thaw); (4, 5, 9, 10, 14, 15 after two times freeze thaw process)

(10) FIG. 10: shows Histamine Release Test (Details of assay in Example 2) of 5 concentrations of iMAT-Cul o2 and iMAT-Cul o3 and the respective allergens in a polysensitized horse (Horse 1)

(11) FIG. 11: shows the Kyte-Doolittle hydrophobicity plot of MAT molecules versus iMAT molecules in the generic part of the respective fusion proteins [i.e. without the respective antigen module(s)]. The hydrophobicity index is shown on the Y-axis versus the amino acid position on the X-axis. Positive numbers in the index indicate hydrophobicity. The shift of the graph of the iMAT molecule at the beginning of the hydrophobicity plot as compared to the MAT molecule is due to the additional N-terminal presence of a His-tag and one methionine residue in the iMAT molecule as well as the absence of any spacer module(s).

(12) FIG. 12: shows the number of non-random hits for each peptide with a length of 13 amino acids for Cul o3 (M4X062, SEQ ID NO: 6) in the canonical database and their position within the corresponding precursor protein. The y-axis depicts the number of hits and the x-axis the amino acid position in reference to the precursor protein. Vertical dotted lines indicate cysteine residues, the dotted-dashed line depicts the moving average of the number of non-random found homologies and horizontal solid lines correspond to in silico cleaved peptides.

EXAMPLES

(13) The following examples serve to further illustrate the present invention; but the same should not be construed as a limitation of the scope of the invention disclosed herein.

Example 1Surrogate Marker for Immunity/Duration of Immunity

(14) Administration of an (isolated) recombinant protein, as disclosed and claimed herein, to an equine produces an immunological response to the allergen and/or epitope present in the antigen module. Additionally, the C- or N-terminal tag module together with the adjacent amino acid residues from the adjacent module can be used in order to detect a unique product-specific immunological signal (e.g. an antibodyor a T-cell response) in a target subject that can be used as a surrogate marker for immunity or duration of immunity. This surrogate marker, as a treatment-specific immunological parameter, enables the assessment of immunity or immune modulation or the duration of immunity or the duration of immune modulation after the administration. Thus, a specific indicator for an immune response triggered by the (isolated) recombinant protein according to the invention is the induction of terminal-tag (optionally with the adjacent amino acid residues) specific antibodies, such as IgG antibodies. Alternatively or additionally, the indicator is the induction of antibodies specific to the junction of a spacer and a module as described and claimed herein or the junction between two modules.

Example 2Hypoallergenicity

(15) Freshly withdrawn blood of horses suffering from IBH can be prepared to test the basophil reactivity to (isolated) recombinant proteins/iMAT molecules, as disclosed and claimed herein. This can be determined by a modified method of Kaul [Kaul, S., 1998. Type I allergies in the horse: Basic development of a histamine release test (HRT). Doctoral thesis. Veterinary School Hannover, Germany]. Briefly, a 10-fold dilution series (e.g. ranging from 10 nM to 0.001 nM final allergen concentration) of iMAT molecules and/or recombinant allergens is prepared in PIPES buffer (AppliChem, Darmstadt, Germany), pH 7.4. Washed red and white blood cells obtained from Na-EDTA coagulation-inhibited blood are incubated with individual dilutions for 1 h at 37 C. The reaction is stopped by incubation on ice for 20 min and the supernatant containing the released histamine is collected from each sample after centrifugation. Maximal histamine content is obtained by boiling the blood cells for 10 min in a water bath (maximal release). The incubation of releasing buffer with washed blood cells served as a negative control (spontaneous release). Histamine concentrations are determined using a competitive RIA (LDN Nordhorn, Germany) as per the manufacturer's instructions.

(16) Alternatively, another basophil activation test is the Cellular Antigen Stimulation Test CAST ELISA which can also be considered as an in vitro allergy provocation test. This assay is done according to the manufacturer's instructions (Bhlmann Laboratories AG, Allschwil, Switzerland). In the CAST, sedimented leukocytes from allergic subjects' blood are simultaneously primed with the cytokine IL-3 and stimulated with iMAT molecules and/or recombinant allergens. Basophilic cells among others generate the allergic mediator, sulfidoleukotriene LTC4, and its metabolites LTD4 and LTE4. These freshly synthesized sulfidoleukotrienes (sLT) are subsequently measured in an ELISA test (Enzyme Linked Immunosorbent Assay).

(17) The potential of iMAT molecules to induce adverse events, e.g., provoke anaphylaxis as a side effect of administration can be evaluated in vitro with these assays by comparing the effects of the iMAT molecules (containing an allergen) to the respective recombinant allergen alone.

(18) A reduced basophil degranulation, e.g., histamine and/or sulfidoleukotriene release by iMAT molecules as compared to the recombinant allergen indicates a lower potential for adverse effects, i.e., a better safety profile of the iMAT molecules.

(19) Said HRT (or CAST) can be used as an in vitro provocation test for type 1 allergic reactions in horses. Allergen specific histamine release indicates the relevance of the respective allergen for the basophilic cell activation and thus can be used as a quantitative parameter for the allergen specific sensitization of a subject.

(20) With MAT molecules described in the prior art, hypoallergenicity could be demonstrated in comparison to the corresponding native allergen (Fel d1) [Senti G et al., J Allergy Clin Immunol. 2012, 129(5): 1290-1296] in the Cellular Antigen Stimulation Test (CAST) assay as well as in the intradermal and in the intracutaneous test. The quantitative difference in sensitivity between the allergen and the MAT molecule comprising the Fel d1 was 100-, 23- and 16-fold, respectively. Though MAT-Fel d1 was clearly hypoallergenic, some allergenicity remained.

(21) With regard to the present invention, the safety of 2 iMAT molecules manufactured according to the present invention comprising Cul o2 and Cul o3 in the antigen module, respectively, was tested. Freshly withdrawn blood of a poly-sensitized horse suffering from IBH was employed in the histamine release test (HRT) as described above.

(22) As shown in FIG. 10, the native allergens elicited a strong histamine release whereas surprisingly the two different iMAT molecules (iMAT-Cul o3 and iMAT-Cul o2) showed virtually no response at all.

(23) Thus, iMAT molecules showed clear superiority in respect to safety as compared with the MAT molecule as described in the prior art (see above).

(24) It can be expected that no allergen related adverse reactions occur if horses suffering from IBH are treated with said iMAT molecules.

(25) The consequence of this surprising safety property of iMAT molecules in contrast to MAT molecules is, that iMAT molecules used as desensitizing proteins can be used similar to vaccines against pathogens. No up-dosing as with classical therapeutic allergens is needed, since vaccines comprising iMAT molecules do not show allergen properties with respect to allergic adverse events. Already the dose of the first injection of the iMAT molecule in a treatment course may be selected based on efficacy considerations only, and one does not have to consider potential allergic adverse reactions. This could not be performed using MAT molecules described in the prior art since the allergenicity of MAT, compared to the native allergen, was only reduced to a certain level. However, MAT molecules still are allergens; iMAT molecules, in contrast, are not. The advantage of these improved properties renders a more efficacious treatment regime possible with e.g., three subcutaneous or intralymphatic injections with a high biopharmaceutical content (e.g., 3 times 20 g to 50 g iMAT protein).

(26) The lack of allergenicity of the two iMAT molecules tested can be explained by the fact that in contrast to the MAT molecules described in the prior art no linker amino acid residues [i.e. spacer module(s) between the first, second and/or third module(s)] were used to separate the different modules in such iMAT molecules. It is known in the prior art that engineered fusion proteins containing two or more functional polypeptides joined by a peptide or protein linker are important for the function (e.g. epitope recognition by the immune system) of the proteins [Klein J S et al., Protein Eng Des Sel. 2014, 27(10): 325-330]. The separation distance between functional units can impact epitope access and the ability to bind with avidity. If the missing amino acid residue linkers between the modules, in particular between the targeting domain and the antigen module, lead to a more rigid structure, conformational epitopes of the allergen module might not be formed due to incorrect folding. A cross linking of antibodies bound on the surface of basophils (e.g., IgE) by its high affinity receptors is necessary to induce activation and histamine release. However, misfolded allergens might not be able to induce such cross linking. Thus, an iMAT molecule without linker may not form conformational IgE epitopes which renders the iMAT molecules non-allergenic.

Example 3in Silica Detection of Allergens/Biomarkers

(27) As described in detail in the description of the invention in IBH the major allergens eliciting IBH have not yet been describedthe bioinformatical engineering of iMAT antigen modules for targeting IBH may be employed to evaluate in silico the relevance of a certain allergens (see Example 6 herein). For example, as shown in FIG. 8 the in silico as well as the ex vivo functional HRT test results of Cul o3 seem to indicate that Cul o3 is a major allergen. Most of the horses tested, though not all, showed a clear hypersensitivity against Cul o3.

(28) This is surprising, as in other experiments, e.g., by van der Meide et al. (Vet J 2014, 200: 31-37), the different recombinant allergens could not be discriminated in their relevance for the disease: around 80% to 90% of horses suffering from IBH showed a sensitization to each of the allergens tested.

(29) Thus, surprisingly, in the course of the present invention it could be shown that Cul o3 (i) in fact is a major allergen (i.e. >50% of the animals are sensitized) in IBH and (ii) is of superior relevance as compared to other allergens.

(30) Apart from being employed in the iMAT antigen module, the respective allergens can also serve to optimize the treatment success by investigating prior to initiation of the treatment the strength of hypersensitivity against said allergen. And throughout the course of the treatment the allergen specific modulation of the different components of the immune system can be monitored, e.g., changes in allergen specific IgE and IgG antibody titers can indicate therapy and/or prevention effects.

(31) Thus, the in silico detected allergens and/or the respective allergen-specific immunoglobulins may be employed as diagnostic and/or theranostic biomarkers.

Example 4Therapeutic Vaccine/Prophylaxis of IBH in Equines

(32) A single iMAT molecule or a combination of iMAT molecules containing different antigen modules according to the present invention may be employed for treating prophylactically or therapeutically a horse suffering from or to be at risk of insect bites inducing IBH. In horses, iMAT molecules according to the present invention can be administered into a submandibular lymph node (Landholt et al., Vet Rec 2010, 167: 302-304). Briefly, the hair over the lymph node is clipped and surgically prepared. Using palpation and/or ultrasound for guidance, a 25 G needle is inserted into the lymph node. The injected iMAT molecule is adsorbed to an adjuvant. The adjuvant consists e.g., of aluminum phosphate (ADJU-PHOS, Brenntag Biosector, Denmark). The iMAT molecule stock is a frozen solution of 375 g/mL protein concentration in vials, each containing 500 L to be thawed before use.

(33) After thawing the iMAT molecule solution, 400 L of the solution are mixed with e.g., 200 L of the aluminum phosphate gel suspension. This final formulation is left at room temperature for 60 minutes prior to the intralymphatic injection to allow for absorption of the iMAT molecule to the ADJU-PHOS. e.g., 200 L of the mixture containing 50 g iMAT molecule is removed into a 500 L syringe for lymph node injection. This preparation is first administered typically on day 0, day 28 and day 56 in a dose between 20 g and 50 g (referring to the weight of solely the one or more antigen modules) per injection and iMAT molecule.

(34) Throughout the treatment period and/or thereafter the efficacy of a therapy or the prevention of IBH can be investigated clinically by quantitative, semi-quantitative or qualitative assessment of skin lesions (e.g., broken hairs, alopecia, crusts/scrubs and wet areas). The intensity of scrubbing/itching can be scored.

(35) These clinical parameters can be compared to clinical signs of the individual horse in previous IBH seasons and/or to the severity prior to the start of a therapeutic intervention. Alternatively, a comparison to IBH affected horses that are not treated or treated with placebo can demonstrate the efficacy of the iMAT molecule-mediated treatment and/or prevention of clinical signs of IBH.

(36) Fresh blood of said horses can be used in an in vitro provocation test for type 1 allergic reactions in horses, e.g., with HRT or CAST (for details see Example 3 supra). A reduced basophil degranulation in response to a challenge with certain culicoides allergens, e.g., histamine and/or sulfidoleukotriene release after the iMAT molecule administration, as compared to before, indicate a therapy and/or prevention effect.

(37) Alternatively or in addition an intradermal provocation test [Langner et al., Vet Immunol Immunopathol 2008, 122(1-2):126-37] with certain culicoides allergens can be employed in said horses. A reduced response (immediate and/or late phase reactivity) indicates a therapy and/or prevention effect of the iMAT molecule administration.

(38) Furthermore, the modulation of the different components of the immune system can be monitored, e.g., changes in allergen specific IgE and IgG antibody titers can indicate therapy and/or prevention effects.

(39) Apart from changes in IgE levels, an increase in allergen-specific IgG can be surprisingly found when treating a horse for IBH with such iMAT molecules. These antibodies can block IgE-mediated anaphylaxis in vivo and seem to inhibit not only the allergen-induced release of inflammatory mediators from basophils and mast cells, but also IgE-mediated allergen presentation to T cells. Among the iMAT-induced IgG antibodies specifically binding to the allergen, some allergen-specific subtypes have been suggested to play an important protective role, as they compete with allergen-specific IgE antibodies and can prevent the activation of CD4+ T cells, by inhibiting the IgE-mediated antigen presentation. Furthermore, the IgG subset which is secreted promotes a significant reduction in mast cells and eosinophils, accompanied by a diminished release of inflammatory mediators [Senna G et al., Curr Opin Allergy Clin Immunol. 2011, 11(4): 375-380].

(40) The functional assay of serum blocking IgG activity as determined by suppression of FcRI-dependent basophil histamine release [Niederberger V et al., Proc Natl Acad Sci USA 2004, 101: 14677-14682] and/or inhibition of FcRII (CD23)-dependent binding of IgE-allergen complexes to B cells seems to offer a better prediction of the individual clinical response (Wacholz P A et al., J Allergy Clin Immunol 2003; 112: 915-922).

(41) Allergen-specific immunotherapy can modulate different components of the immune system. Cellular modifications consist of a reduction in allergen-induced T-cell proliferation, indicating the induction of peripheral tolerance in allergen-specific T cells and a decrease in antigen-specific Th2-dominated immune response in favor of a Th1 reaction with increased IFN- production (Durham S R et al., J Allergy Clin Immunol 1996, 97: 1356-1365). The key cell type responsible for coordinating this immunological switch is a heterogeneous T-cell population, called regulatory T cells (T.sub.reg). At the cellular level, the crucial factor for successful allergen immunotherapy is the peripheral induction of type 1 T.sub.reg cells. Functional studies on type 1 T.sub.reg cells, specific in recognizing antigens, revealed that the modulation of Th1 and Th2 responses by type 1 T.sub.reg cells mostly depends on the secretion of the cytokine IL-10, which has immunosuppressive properties. In fact, IL-10 inhibits the proliferative response of peripheral T cells against specific allergens and plays a central role in the induction of T-cell anergy (James L K, Durham S R, Clin Exp Allergy 2008, 38: 1074-1088). In vitro, IL-10 enhances the expression of the regulatory factor FoxP3, modulates eosinophilic function and reduces pro-inflammatory mediators released by mast cells (Mobs C, et al., Int Arch Allergy Immunol 2008, 147: 171-178).

(42) Another possible marker of the outstanding clinical efficacy of said iMAT molecules-mediated immunotherapy is the detection of changes in the number or the nature of allergen-specific T cells. On the basis of, for example, Bet v1 tetramer staining studies, the levels and characteristics of circulating birch pollen-specific CD4.sup.+ T cells can potentially be compared before and after SIT (van Overtvelt L et. al., J Immunol 2008, 180: 4514-4522). Recently, transforming growth factor (TGF)- has also been identified as a key cytokine in successful SIT. Many actions may account for its relevance, such as the suppression of specific Th1, Th2 and Th17 cells, the induction of FoxP3 and the suppressive function of Tregs. In addition, TGF- down regulates FcRI expression on Langherhans cells and suppresses IgE synthesis (Akdis C A, Akdis M J, Allergy Clin Immunol 2009, 123: 735-746).

Example 5Comparison of iMAT Molecules According to the Present Invention with Prior Art MAT Molecules According to WO 2004/035793 (US Equivalent US 2005/0281816)

(43) For assessment of purity of a MAT protein, a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) test procedure has been established (Thompson J et al., J Biol Chem 2002, 277: 34310-34331). The method, including sample preparation with a reducing agent, Lithium Dodecylsulfate (LDS) and heating at 75 C., resulted in reproducible multiple sharp bands after electrophoretic separation. Staining with Coomassie blue gives linear quantitative (densitometry) features in gels loaded with 200 to 1000 g protein. Using a monoclonal antibody, detecting the allergen module in a MAT molecule with Fel d 1 as allergen module (MAT-Fel d1) it has been shown, that the main band and 13 minor bands all contain the MAT-Fel d1 protein. The small bands migrate at the same position as on the original gel also after re-loading on the gel (FIGS. 2A and 2B). Several different methods, such as different temperature and buffer composition protocols, for sample preparation prior to PAGE generated the same band pattern.

(44) In all of these bands the presence of the full length (complete) MAT-Fel d1 protein and only traces of host cell proteins could be demonstrated after each band was cut out of the gel, digested by trypsin and subsequently analyzed by mass spectrometry (nanoLC/ESI-MS). From these experiments an anomalous feature (gel shifting), e.g., of different folding variants, of MAT-Fel d1 in the SDS-PAGE can be concluded. This means that MAT-Fel d1 in the analyzed preparation is not suitable as biopharmaceutical molecule, in particular for clinical and/or commercial biopharmaceutical manufacturing, since its purity could not be determined with standard methods (e.g. SDS-PAGE), but only with the modified procedure explained above.

(45) In contrast to this gel shifting phenomenon of the MAT-Fel d1 molecule, the Fel d1 as such does not show such anomalous feature in SDS PAGE (FIG. 3, lanes 2 and 3). The Fel d1 displays a single sharp band at the expected molecular weight (19.6 kD).

(46) A further anomalous feature could be observed in RP-HPLC analysis. No single peak of the MAT-Fel d1 was seen in this analytical method (FIG. 4), neither in the native conformation of the protein nor in the denatured form induced by chaotropic and reducing conditions. However, for GMP certified biopharmaceutical manufacturing a single isoform of the biomolecule in marketed pharmaceutical preparations is mandatory.

(47) These observations in SDS-PAGE and RP-HPLC analysis may be explained by the physicochemical properties based on the amino acid sequence. Analysis in the Kyte & Doolittle hydrophobicity plot [Kyte J, Doolittle R F, Journal of Molecular Biology 1982, 157(1), 105-132] revealed adjacent extreme hydrophobic and hydrophilic domains (FIG. 5) which may be responsible for this anomalous behavior.

(48) In particular the hydrophobic region of the targeting domain of the fusion protein is similar to the transmembrane segments of membrane proteins which are known in the art to cause such anomalous feature [Rath A et al., Proc Natl Acad Sci USA. 2009, 106(6): 1760-1765].

(49) Migration on SDS-PAGE, that does not correlate with formula's molecular weights, termed gel shifting appears to be common for membrane proteins. This means, that the prerequisite of the SDS-PAGE method, which is a separation of molecules solely according to their molecular weight, independent on their native 2D- or 3D-structure does not apply in these cases. In the above cited work (PNAS article), the authors investigate the anomalous gel mobility of helical membrane proteins using a library of wild-type and mutant helix-loop-helix (hairpin) sequences derived from transmembrane segments 3 and 4 of the human cystic fibrosis transmembrane conductance regulator (CFTR), including disease-phenotypic residue substitutions. They found that these hairpins migrate at rates of minus 10% to plus 30% vs. their actual formula's molecular weights on SDS-PAGE and load detergent at ratios ranging from 3.4-10 g SDS/g protein. They additionally demonstrated that mutant gel shifts strongly correlate with changes in hairpin SDS loading capacity, and with hairpin helicity, indicating that gel shift behavior originates in altered detergent binding. In some cases, this differential solvation by SDS may result from replacing protein-detergent contacts with protein-protein contacts, implying that detergent binding and folding are intimately linked.

(50) The SDS PAGE (FIG. 6) as well as the RP-HPLC analysis (FIG. 4 and FIG. 7) of MAT and iMAT proteins revealed substantial differences in the migration pattern or elution, respectively. The oxidized form of MAT-Fel d1 did not show a single sharp band on the SDS-PAGE gel (Lane 3) but several diffuse bands with larger and smaller apparent molecular weights than the actual 32.2 kD of MAT-Fel d1. In contrast iMAT-Cul o4 exhibited a single sharp band (M=41.6 kD) under oxidized conditions. Also the RP-HPLC chromatogram showed a single peak (FIG. 7).

(51) Under reducing conditions the MAT-Fel d1 in the SDS-PAGE reveals a main band migrating approximately at the known molecular weight but in addition some of the minor bands described in FIGS. 2A and 2B known to contain the complete sequence of MAT-Fel d1 emerge again (FIG. 6, lane 6). This is an attribute, which is characteristic for the anomalous feature of MAT-Fel d1. Additionally the RP-HPLC chromatogram of MAT-Fel d1 under reducing conditions (FIG. 4, right graph) exhibit at least 3 different isotypes of the MAT-Fel d1. In contrast the iMAT molecules reveal characteristics evidently indicative for a single isotype in the SDS-PAGE (FIG. 6, lane 7) as well as in the RP-HPLC (FIG. 7).

(52) The reducing conditions lead to a cleavage of the disulfide bridges in the MAT molecule, thus the MAT and the iMAT molecules should behave alike under reducing conditions if the disulfide bridges are solely responsible for the anomalous feature of MAT. However, this is not the case, since the anomalous gel shifting and the occurrence of isoforms in RP-HPLC of MAT molecules is still present under reducing conditions.

(53) However, the iMAT molecule does not show such gel shifting and exhibits a peak in RP-HPLC chromatogram in the native (oxidized) form of the protein. Furthermore the Kyte-Doolittle plots (Kyte, J. and Doolittle, R. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157: 105-132) of MAT and iMAT molecules are nearly identical at the N-terminus covering the sequence of His-tag, TAT and targeting domain (FIG. 11). Consequently, a person skilled in the art would not be motivated to construct a MAT molecule according to the prior art with cysteine residues substituted with other amino acid residues in order to overcome the disadvantages of the prior art.

(54) The three yielded iMAT molecules have been constructed by the bioinformatical engineering procedures according to Example 6 below and produced by recombinant expression technology in E. coli. All three iMAT molecules were stable in buffer (20 mM citrate, 1 M arginine, pH 6.0) after freezing and thawing twice and could be adsorbed to ADJU-PHOS (Brenntag, Denmark) as adjuvant, so that the iMAT molecules can be used as a vaccine. The proteins could be desorbed from ADJU-PHOS without degradation in the same buffer system (FIG. 9).

Example 6Bioinformatical Engineering of iMAT Molecules for Targeting IBH in Equines

(55) In order to, for example, treat horses with allergic insect bite hypersensitivity (IBH) using the iMAT technology effectively, it is further mandatory: a. to select a culicoides protein as allergen module in iMAT molecules that is a major allergen and thus has a high prevalence to cause hypersensitivity in affected horses and thus can also be the target for tolerance induction, and b. to construct an iMAT molecule with said major allergens that is thermodynamically stable and can be produced efficiently by protein engineering and can additionally be analyzed with standard methods to ensure sufficient enough quality (i.e., identity, purity and potency).

(56) In order to fulfil these requirements, a bioinformatics approach was chosen for the selection of the allergen to be included into the iMAT molecules according to the invention. The objective of the selection was (i) to choose one or more allergens to be expected to be of relevance in IBH, i.e., that the majority of horses suffering from IBH are sensitized to the respective allergen, and (ii) to choose the allergen with the highest probability of comprising linear epitopes of allergen characteristics, i.e., comprising high numbers of short peptide sequences (7 to 13 amino acid residues) homologue to those in published allergens.

(57) Currently approximately 230 proteins from culicoides (Culicoides sonorensis, Culicoides nubeculosus, Culicoides obsoletus) found in saliva are known that might potentially elicit allergies in subjects, e.g., equines. To select appropriate antigens for pharmaceutical preparations, a homology comparison based on local sequence alignments to known non-culicoides allergens was chosen. Often epitope detection for antibody recognition (mostly conformational epitopes) is achieved by functional analysis (e.g., peptide microarrays) or for T-cells epitopes (linear epitopes) by calculation of peptide binding probabilities to MHC molecules. The therapeutic principle of the iMAT technology inter alia is based on endocytosis and degradation by acid-dependent proteases in endosomes followed by MHC Class II binding and antigen presentation. In equines, the characteristics of such systems are currently not fully explored since the allelic diversity of the equine MHC genes is far less investigated as compared to the human system and thus experimentally not readily available. Furthermore the research on IBH as a Type I allergy and the causal allergens eliciting the symptoms has only recently started. The first protein characterized as allergen active in effector cells (i.e., basophil granulocytes) was not described until 2008 [Langner et al., Vet Immunol Immunopathol 2008, 122(1-2):126-137].

(58) Thus a differentnon experimental but bioinformaticsapproach for allergen selection was chosen that is based on local homology searches of peptides derived from culicoides proteins to known non-culicoides allergenic proteins, most of which are known to raise allergies in humans. Amino acid sequences of proteins suspected to have allergenic properties were exported from publicly available databases (e.g., UNIPROT) and redundancies were determined by analysis of sequences homologies within the exported dataset. Highly homologue sequence counterparts were eliminated and the resulting remainder of sequences served as the canonical sequence database of probable valid antigens for subsequent analyses. To determine culicoides proteins with putative high allergenic potential, proteins were in silico cleaved into peptides with lengths of 6 to 15 amino acids with a one amino acid shifting. Next local-pairwise alignments of each culicoides protein and the corresponding peptides to the canonical sequence database were performed. Following this, a scaling of obtained alignment hits was conducted by setting the self-alignment score of a given culicoides protein to one and alignment hits of the corresponding peptides accordingly. Next the number of alignment hits exceeding a given threshold were counted for each peptide and compared by local-pairwise alignment to a randomly generated database of protein sequences with no known allergenic properties, and subsequently scaled and counted. The resulting non allergic protein counts were subtracted from those of the allergen results and cumulative hit scores for each culicoides protein based on the number of hits for all corresponding peptides were calculated (FIG. 12). Proteins with highest counts were selected as iMAT antigen module candidates.

(59) For validation of the bioinformatic antigen selection method the occurrence of sensitization against five culicoides allergens in 13 horses suffering from clinically diagnosed IBH was measured. All selected antigens were tested in a semi-quantitative functional ex vivo experiment based on the allergen specific histamine release from equine basophiles isolated from blood samples (Histamine Release Test=HRT, details of the assay: see Example 2). The recombinant allergens were tested in 3 concentrations and the score value for each allergen ranged between 0 (no sensitization) to 6 (highest sensitization). The data from all 13 horses were compared to the results of the bioinformatics calculations (FIG. 8). This analysis shows an excellent overall correlation (r=0.97) between the sensitization (total score values per allergen of the cohort) and the prediction score. From this correlation analysis we conclude that the bioinformatics antigen selection method is suitable to select relevant culicoides allergens for the antigen module in iMAT molecules. Thus, the IBH affected horses can be treated with iMAT molecules according to the present invention comprising the selected allergens in order to induce tolerance against these allergens with high probability in a large proportion of the subgroup of diseased subjects.

(60) Three allergens have been selected based on the bioinformatics analysis to be included in three different iMAT molecules, i.e., Cul o3, Cul o2 and Cul o4.

(61) Each of the selected allergens were integrated into separate iMAT molecules as the antigen module and subsequently optimized for thermodynamic stability by iterative modeling of three-dimensional protein structures and calculation of changes of free energies after substitution of single amino acids. Physicochemical properties and stability is influenced by substituting different amino acid residue(s) within the primary amino acid sequence.

(62) The results of the herein described analyses (antigen search and modeling) were transformed into an iMAT amino acid sequence suitable for pharmacological production and application in equines in order to treat e.g., IBH.

Example 7Construction of Mosaic-Like iMAT Molecules According to the Invention

(63) It is expected that iMAT molecules according to the invention can further be improved if components of more than one allergen are included into the antigen module. For this purpose it is possible to apply the basic principle of the above described bioinformatics selection approach (Example 6) in a different way. Instead of selecting complete allergens based on the hit count of allergen peptides found in the allergen data base, only the most abundant peptides of several of such allergens are used to engineer an iMAT antigen module. Thus, such an iMAT molecule consists of an antigen module of peptides that stem from several allergens. This allows broadening of the spectrum of a single iMAT molecule with respect to its targeted allergic profile and is thus beneficial for pharmacological drug development.

(64) In order to find short peptide sequences that qualify for such mosaic-like iMAT protein 230 proteins from Culicoides species (Culicoides sonorensis, Culicoides nubeculosus, Culicoides obsoletus) found in saliva were analyzed by homology comparison as described above (see Example 6). Briefly in silico cleaved Culicoides proteins with peptide lengths of 6-15 amino acid residues were locally aligned to a canonical sequence database of allergen-related proteins and a random database of non-allergy-related proteins. The number of differences of significant homologies for each peptide found within the canonical database was determined. Subsequently, each peptide was locally aligned to a random database triple the size of the canonical database to reduce false positive hits. The top (e.g., tenth percentile) of remaining homologies for each peptide length is especially suitable to serve as a base for construction of a mosaic-like or hybrid allergen carrying iMAT molecule. To construct a mosaic-like iMAT molecule a protein precursor is chosen (for example, from the list of precursor proteins corresponding to top ranking peptides) as a scaffold protein for embedding top ranking peptides. The signal peptide sequence is removed from the scaffold protein and top ranking peptides with additional adjacent N- or C-terminal amino acids may be inserted within the original sequence of the scaffold protein or may replace parts of the original sequence of the scaffold protein. The position for insertion or replacement is determined using similarity alignments or the reference position of the peptide in the corresponding precursor protein. As a next step, His-Tag, the TAT and targeting domain are added. Finally cysteine residues are replaced by most stabilizing residues as described above.

(65) In an exemplary instance of this approach a salivary protein of Culicoides sonorensis (UniProt database entry Q66U13) was used as a scaffold protein. The most allergenic structure within the protein was determined in order to conserve this part of the sequence. The signal peptide sequence is removed and subsequently sequences from UniProt database entries Q5QBI9, Q5QBK6, Q5QBL6, Q5QBJ4 and Q5QBI2 are added. Afterwards, the mosaic-like allergen is integrated into the iMAT general molecule structure. Finally, cysteine residues are replaced by other amino acid residues, preferably serine, leucine, isoleucine and/or aspartic acid that do not derogate the stability of the constructed iMAT molecule (fusion protein).

(66) SEQ ID NO: 17 depicts the whole sequence of an exemplary mosaic-like (hybrid) iMAT molecule according to the present invention [Amino acid residue 1: N-terminal methionine, amino acid residues 2-7: His-tag, amino acid residues 8-18: TAT sequence, amino acid residues 19-128: targeting domain, amino acid residues 129-333: hybrid allergen (scaffold protein Q66U13 without signal peptide and inserted/replaced sequences; inserted/replaced sequences: amino acid residues 137-151: Q5QBI9; amino acid residues 155-166: Q66U13; amino acid residues 185-200: Q5QBK6; amino acid residues 215-229: Q5QBL6: amino acid residues 235-250; Q5QBL6; amino acid residues 265-278: Q5QBJ4; amino acid residues 318-333: Q5QBI2; substituted cysteines (serine): 46, 146, 158, 173, 222, 225, 246, 273, 283, 329; substituted cysteines (leucine): 295; substituted cysteines (isoleucine): 299; substituted cysteines (aspartic acid): 181].

Example 8Therapeutic Vaccine/Prophylaxis of Recurrent Airways Obstruction (RAO) in Equines

(67) A single iMAT molecule or a combination of iMAT molecules containing different antigen modules according to the present invention can be employed for treating prophylactically or therapeutically a horse suffering from or being at risk of RAO and/or summer pasture associated RAO. In horses iMAT molecules according to the present invention may be administered as described in Example 4.

(68) Adult horses with a known history of RAO will be employed. Prior to the treatment, horses are conditioned e.g., to stand in stocks wearing a face-mask to enable a complete physical airway examination, including pneumotachograph measurements, upper airway endoscopy and bronchoalveolar lavage (BAL) for analysis of e.g., percent of neutrophils in BAL fluid. Complete blood count and biochemistry profile are performed to exclude the presence of concomitant medical conditions.

(69) The following parameters may be investigated to examine the respiratory system and to measure efficacy of a therapeutic and/or prophylactic treatment with iMAT molecules: change in lung function variables e.g., respiratory rate, maximal transpulmonary pressure, lung resistance, and/or lung elastance. Additionally, the parameter may be incorporated into a weighted clinical scoring system e.g. also evaluating clinical scores of breathing effort/abdominal lift, nasal discharge and/or flaring, cough and/or abnormal lung, bronchial and/or tracheal sounds.

(70) Thus, throughout the treatment period and/or thereafter the efficacy of a therapy or the prevention of RAO can be investigated clinically by quantitative, semi-quantitative or qualitative assessments.

(71) These clinical parameters may be compared to clinical signs of the individual horse in previous seasons and/or to the severity prior to the start of a therapeutic intervention. Alternatively, a comparison to RAO affected horses that are not treated or treated with placebo may demonstrate the efficacy of the iMAT molecule-mediated treatment and/or prevention of clinical signs of RAO.

(72) Bronchoalveolar lavage (BAL) fluid and/or fresh blood of said horses may be used in an in vitro provocation test for type 1 allergic reactions in horses, e.g., with HRT or CAST (for details, see Example 2 and/or Hare et al Can J Vet Res 1998; 62: 133-139). A reduced pulmonary mast cell and/or basophil degranulation in response to a challenge with certain allergens, e.g., histamine and/or sulfidoleukotriene release after the iMAT molecule treatment as compared to before, indicate a therapy and/or prevention effect.

(73) Alternatively or in addition, employing certain recombinant allergens an intradermal provocation test, skin prick test or also allergen specific IgE and/or IgG determination in BAL fluid or serum may be monitored in said horses (Tilley P et al., J Equine Vet Sci 2012, 32: 719-727). A reduced response (immediate and/or late phase reactivity) and/or changes of the antibody titers indicate a therapy and/or prevention effects of the iMAT molecule treatment.

REFERENCES

(74) (1) Akdis C A, Akdis M J, Allergy Clin Immunol 2009, 123: 735-746 (2) Allergome (www.allergome.org) (3) Durham S R et al., J Allergy Clin Immunol 1996, 97: 1356-1365 (4) Gadermaier G et al., Molecular Immunology 2010, 47: 1292-1298 (5) Ginel P J et al., Vet. Dermatol. 2014, 25(1): 29-e10 (6) Hare J E et al., Can J Vet Res 1998, 62: 133-139 (7) James L K, Durham S R, Clin Exp Allergy 2008, 38: 1074-1088 (8) Kaul, S., Type I allergies in the horse: Basic development of a histamine release test (HRT). Doctoral thesis. Veterinary School Hannover 1998, Germany (9) Kehrli D et al, J Vet Intern Med 2015, 29(1): 320-326 (10) Kelley J et al., Immunogenetics 2005, 56: 683-695 (11) Klein J S et al., Protein Eng Des Sel 2014, 27(10): 325-330 (12) Kyte J, Doolittle R F, Journal of Molecular Biology 1982, 157(1): 105-132 (13) Landholt G A et al., Vet Rec 2010, 167: 302-304 (14) Langner K et al., Vet Immunol Immunopathol 2008, 122(1-2):126-137 (15) Leclere Metal., Respirology 2011, 16: 1027-1046 (16) Martnez-Gmez J M et al., Allergy 2009, 64(1): 172-178 (17) Mobs C, et al., Int Arch Allergy Immunol 2008, 147: 171-178 (18) Niederberger V et al., Proc Natl Acad Sci USA 2004, 101: 14677-14682 (19) Olsen L et al., Vet. J. 2011, 187: 347-351 (20) Pires D E et al. Bioinformatics 2014, 30(3): 335-342 (21) Pirie R S, Equine Vet J 2014, 46: 276-288 (22) Rath A et al., PNAS 2009, 106(6): 1760-1765 (23) Report of the 3.sup.rd Havemeyer workshop on allergenic diseases of the Horse, Hlar, Iceland, June 2007, Veterinary Immunology and Immunotherapy 2008, 126: 351-361 (24) Rose H, Arb Paul Ehrlich Inst Bundesinstitut Impfstoffe Biomed Arzneim Langen Hess 2009, 96: 319-327 (25) Schaffartzik A et al., Veterinary Immunology and Immunopathology 2012, 147: 113-126 (26) Senna G et al., Curr Opin Allergy Clin Immunol. 2011, 11(4): 375-380 (27) Senti G et al., J Allergy Clin Immunol. 2012, 129(5): 1290-1296 (28) SIAF Annual Report 2010 (29) SIAF Annual Report 2011 (30) Tilley P et al., J Equine Vet Sci 2012, 32: 719-727 (31) US 2005/0281816 (32) US 2014/0105906 (33) U.S. Pat. No. 7,653,866 (34) van Bargen, K and Haas, A. FEMS Microbiol Rev 2009, 33: 870-891 (35) van der Meide N et al., Vet J 2014, 200: 31-37 (36) van Overtvelt L et al., J Immunol 2008, 180: 4514-4522 (37) Vazquez-Boland, J A et al., Vet Microbiol 2013, 167: 9-33 (38) Wacholz P A et al., J Allergy Clin Immunol 2003; 112: 915-922 (39) WO 2004/035793

In the Sequence Listing

(75) SEQ ID NO: 1 relates to a suitable minimal amino acid sequence for one translocation module according to the present invention which is still being functional, i.e., capable of effectively promoting cell entry;

(76) SEQ ID NO: 2 relates to the full equine invariant chain amino acid sequence;

(77) SEQ ID NO: 3 relates to the full Cul n4 allergen amino acid sequence;

(78) SEQ ID NO: 4 relates to the full Cul of allergen amino acid sequence;

(79) SEQ ID NO: 5 relates to the full Cul o2 allergen amino acid sequence;

(80) SEQ ID NO: 6 relates to the full Cul o3 allergen amino acid sequence;

(81) SEQ ID NO: 7 relates to a Cul n4 iMAT molecule according to the present invention;

(82) SEQ ID NO: 8 relates to a Cul n4 iMAT molecule according to the present invention;

(83) SEQ ID NO: 9 relates to a Cul of iMAT molecule according to the present invention;

(84) SEQ ID NO: 10 relates to a Cul of iMAT molecule according to the present invention;

(85) SEQ ID NO: 11 relates to a Cul o2 iMAT molecule according to the present invention;

(86) SEQ ID NO: 12 relates to a Cul o2 iMAT molecule according to the present invention;

(87) SEQ ID NO: 13 relates to a Cul o3 iMAT molecule according to the present invention;

(88) SEQ ID NO: 14 relates to a Cul o3 iMAT molecule according to the present invention;

(89) SEQ ID NO: 15 relates to a specific fragment of the equine invariant chain amino acid sequence with maintained intracellular transport function;

(90) SEQ ID NO: 16 relates to a specific fragment of the equine invariant chain amino acid sequence with maintained intracellular transport function;

(91) SEQ ID NO: 17 relates to the whole sequence of an exemplary mosaic-like iMAT molecule according to the present invention;

(92) SEQ ID NO: 18 relates to the full Cul o4 allergen amino acid sequence;

(93) SEQ ID NO: 19 relates to the full Cul o7 allergen amino acid sequence;

(94) SEQ ID NO: 20 relates to a Cul o4 iMAT molecule according to the present invention;

(95) SEQ ID NO: 21 relates to a Cul o4 iMAT molecule according to the present invention;

(96) SEQ ID NO: 22 relates to a Cul o7 iMAT molecule according to the present invention;

(97) SEQ ID NO: 23 relates to a Cul o7 iMAT molecule according to the present invention.

(98) All prior art references cited herein as well as the sequence listing accompanying this application are hereby independently from each other incorporated by reference in their entirety.