Solution phase biopanning method using engineered decoy proteins

[0001] In the postgenomic era, the effort to create new drugs may be directed at seeking ways to specifically block the function of key proteins that have previously been identified by techniques such as microarray analysis of mRNA expression levels in disease states. In modern science, proteomics involves understanding protein interactions both through organized pathways and through the binding of other proteins. The association of protein structure and activity includes mapping known domains and determining three-dimensional (3D) conformations responsible for specific functions. Access to information on the three-dimensional structure of proteins has already become commonplace. For example, NCBI maintains open access to tools called VAST, which allows you to search for structures with similarity. It compares the coordinates of new 3D structures of proteins with those in the modeling database - MMDB ( Molecular Modeling Database) and Protein Database (PDB).

[0002] The phage display technique describes an in vitro selection method where the polynucleotide sequence encoding the peptide or protein is fused to bacteriophage coat proteins, which results in the display of fused proteins outside the virion while the introduced DNA remains inside the phage particle. Such a relationship between the presented protein and the DNA encoding it allows for screening analyzes of a huge number of different variants of the protein, each of which is associated with the corresponding DNA sequence. This in vitro selection method is called biopanning in English.

[0003] Phage, ribosome, yeast and bacterial libraries are tools enabling testing of numerous proteins or peptides. Ribosome display is a way of rewriting mRNA particles into their corresponding proteins, while this protein remains attached to the RNA. The nucleotide sequence coding for this protein is obtained by RTPCR (Mattheakis, LC et al. 1994. Proc. Natl. Acad. Sci. USA 91, 9022). Yeast display is based on the construction of proteins fused to membrane-bound yeast alpha-glutinin adhesive receptor, aga1 or aga2, which are part of the conjugation system (Broder, et al. 1997. Nature Biotechnology, 15: 553-7). Bacterial presentation involves fusion of protein with proteins transported outside the bacterial cell that remain bound to the membrane or cell wall (Chen and Georgiou. 2002. Biotechnol Bioeng, 79: 496-503).

[0004] Compared to hybridoma generation technology, phage antibody display methods allow manipulation of in vitro selection relative to the target antigen and overcome restrictions resulting from the host's influence on the antigen and vice versa. The main advantage of in vitro selection methods is the ability to manipulate selection so as to obtain antibodies that bind to various target protein sites.

[0005] While phage libraries allow easier recovery of genetic material associated with certain functional properties, multi-stage panning is required to obtain the best candidate from the library. From a different point of view, when the structure of the functional domain of a given polypeptide ligand is known, it is desirable to have a method for selecting antibodies or other binding agents such as peptides and proteins that bind the ligand to specific domains. Domain or epitope-targeted panning has become a routine method of selecting antibodies that bind to a target protein. This selection was initially obtained by gradual selection of antibodies using methods known as selective panning, ligand capture panning, subtractive panning or pioneer selection (Hoogenboom, HR et al (2000) supra).

[0006] In subtractive panning, the target (s) with overlapping but not identical binding sites can be used to reject unwanted binding particles. This approach is used to identify particles that bind even to unknown antigens, such as when using normal cells to remove particles that bind to cancer cells. Otherwise, naturally occurring proteins with frequently occurring domains or structure are used in sequential or competitive selection to obtain site-specific antibodies that distinguish related antigens or are common to them. Typically, naturally occurring proteins such as related chemokines or H-ras protein mutants are used in research (Horn, IR et al. 1999, FEBS Lett. 463: 115-120).

[0007] Panning aimed at capturing ligand is analogous to so-called sandwich variety of the ELISA test. Here, an immobilized antibody bound to an irrelevant and unrelated epitope is used to select and present the preferred binding surface of the ligand of interest in phage panning (US6376170). In another case, competitive antibodies were used to selectively block antigens in domains other than the subject of the study (Tsui, P. et al. 2002.

J. Immunol. Meth. 263: 123-132). Pioneer selection is based on the use of mono- and polyclonal antibodies as well as natural ligands directly or indirectly associated with horseradish peroxidase - HRP (horseradish peroxidase). In the presence of biotinated tyramine, these molecules catalyze biotinylation of the phage a short distance from the site where it binds to the antigen, enabling recovery of specific and labeled phage from the entire population using streptavidin. In this approach, the phage binding directly to the target or in its immediate vicinity is selectively recovered (Osbom, JK et al. 1998. Immunotechnol. 3: 293-302). The use of monoclonal antibodies to directly bind subsequent sites is termed "wandering after the epitope" (Osbom, JK et al. 1998. supra). Burioni et al. 1998. Research in virology 5: 327-330, Zhou et al. 2002, PNAS. 99: 5241-5246 and Parsons et al. 1996. Prot. Eng. 9: 1043-1049 describe negative selection methods.

[0008] A disadvantage of the above methods is that before obtaining the binding domain of the desired domain, undesirable binding particles are obtained and characterized. In addition, the target is not a specific epitope. The present invention relates to a new method of obtaining antibodies or ligand-binding particles that bind to a selected epitope through the use of a hybrid competing protein in the panning selection process, the so-called bait.

SUMMARY OF THE INVENTION [0009] The present invention relates to a new method of selecting ligand binding particles that bind to a preselected domain. To this end, a panning ligand is used in the panning process of the following claims. The ligand bait is designed to differ from the target protein only in the preselected domain that is the potential binding site. When designing a protein-bait, you can rely on information on the structure obtained from real measurements, e.g. data from X-ray crystallography, or use a model of three-dimensional structure of the molecule obtained in silico. When structure information is available, protein bait design is less complicated. However, when the structure of the bait is missing or incomplete, modifications of the safe sequence sites may be based on naturally occurring protein variants, such as homologues found in different species.

[0010] Herein, nucleic acids encoding proteins - baits of the invention that are useful for expressing bait proteins in cells or the host organism are provided.

[0011] After transfection of the host cell, the protein-bait can both be expressed on the cell surface and be a secretory protein that is recovered from the cell growth medium. Protein-bait can be purified or used in a heterogeneous environment such as cell surface. During the panning step, the molar ratio of target protein and bait protein is maintained to reject non-specific and low affinity binding particles, and then to recover only particles associated with the target protein. Thanks to this, the protein-binding particles - the target fused to the corresponding genetic material - are selected from the library based on the ability to specifically bind the protein - the target in a place that is modified in the bait and is therefore known to interact with desired domain.

[0012] In another aspect, the method of selecting antibodies described herein that bind to a predetermined epitope can be used to alter some of the properties of particles that bind to therapeutic antibodies or ligands that have been proven to act in one species, e.g., a model animal species, to directly obtain an analogous biotherapeutical effective in other species. Otherwise, human biological medicines can easily be changed into analogous, useful in the treatment of other mammals or with an analogous mechanism of action in the types and species of animals in which the specific species is intended to be used, e.g. cattle, pigs, poultry, dogs, cats or other farm or domestic animals. In one embodiment of the present invention, a process may be used to select for antibodies that interact with a homologous protein in the same, in three-dimensional structure, domain as the reference antibody. This is of particular use when, for example, a monoclonal antibody directed against a specific region or epitope of a human antigen is known and it is desirable to obtain a ligand for human antibodies directed against a human target and the same epitope. In another embodiment, this process is useful if there is an antibody that binds to the epitope of a human antigen, but there is a need to create for scientific purposes replacement antibodies that react with the same epitope of an analogous protein in another species, e.g., mice. In this way, anti-mouse antibodies can be obtained that have similar properties to the original anti-human antibodies.

[0013] Thus, in one aspect, the method of selecting from the library a polypeptide binding ligand blocking molecule, wherein a specific functional region of the ligand to be bound is defined, comprises the following steps: a) determining the functional protein domain to be blocked, b) analyzing the structural properties common to the ligand and at least one species or functional analogue of this ligand, c) creating a bait with the said structural properties common to the ligand and selected homologues, the bait having shared structural properties in regions other than the functional domain to be blocked, and d) using said bait in excess of the ligand binding particle to select particles that preferentially bind to the functional domain to be blocked

Burioni et al. 1998. Research in virology 5: 327-330, Zhou et al. 2002, PNAS. 99: 5241-5246 and Parsons et al. 1996. Prot. Eng. 9: 1043-1049 describe negative selection methods.

[0014] In another aspect, the present invention is directed to a method for identifying an antibody that binds to a preselected epitope of a target protein that comprises the following steps: a) construction of the library of phage particles on the surface of which the antibodies are expressed, b) creating a protein - baits that has an altered amino acid sequence corresponding to a pre-selected epitope of the target protein, whereby the bait protein differs from the target protein only within the previously selected epitope, c ) incubating the phage library with the target protein to select those phage particles from among phage particles with attached antibodies, which have antibodies that bind to the target protein, d) adding in a molar excess of the protein-bait as a competitor to perform negative selection of phage particles specific for the pre-selected epitope, e) separating the phage particles that bind to the target protein from those that they bind protein-bait, and f) recovering the target protein binding phage particles.

BRIEF DESCRIPTION OF THE DRAWING [0015]

Figure 1 is a graphic representation of the method according to the invention.

Figure 2 is a picture of the CDR sequence and contains skeletal assignment for a murine tissue factor (mTF) lead candidate binding Fab regions.

Figure 3 contains a graph of the effect of concentration on the binding of Fab regions selected by the method of the invention to a target protein (mTF, solid lines) and a protein bait that in the previously selected epitope has 2 altered amino acids (hu / mTF, dashed lines) .

Figure 4 is a graphical representation of the concentration-relative amount of fluorescence units for the two selected mTF binding Fabs.

Figure 5 contains a comparison of many IL-13 protein sequences from different species.

Figure 6A and B contains values ​​of energy and surface dimensions for human IL-4 (hIL-4) obtained by crystallography. Figures 7A and B contain energy and surface dimension values ​​for hIL-13 calculated from crystallography data for hIL-4.

Shortcuts [0016]

Abs bFGF

GM-CSF

LOAM

Has b

TF

FIIV

FIIVa

FX

FXa antibodies, polyclonal or monoclonal basic fibroblastic growth factor colony-stimulating factor granulocyte-macrophage colony interleukin monoclonal antibody tissue factor factor IIV (inactive) factor IIV (active) factor X (inactive) factor X (active)

DETAILED DESCRIPTION OF THE INVENTION

Definitions [0017] By the term "antibody" is meant an immunoglobulin or immunoglobulin binding fragment. Although not all immunoglobulins can bind antigen, it has been proven that antibody fragments can bind antigens, polypeptides or target proteins and other particles. Therefore, as used herein, the term "antigen binding fragments" includes, but is not limited to: (i) a Fab fragment consisting of the variable (V) domains of the heavy (H) and light (L) antibody chains simultaneously with the respective constant domains (C) (VL-CL and VH-CH1 domains); (ii) the Fd fragment consisting of the VH and CH1 domains; (iii) an Fv fragment consisting of the VL and VH domains of the same antibody; (iv) dAb fragment (Ward, ES et al., Nature 341: 544-546 (1989)), which consists of the VH domain; (v) isolated CDR regions; (vi) F (ab ') 2 fragments, a single chain of Fv particles (scFv) in which the VH and VL domains are joined by a peptide linker that allows the association of two domains to form an antigen binding site; (viii) bispecific dimers of single Fv chains and (ix) fusion junctions and proteins containing the aforementioned, including but not limited to diabodies, multivalent or multispecific fragments or other constructs constructed capable of binding the target peptide and containing an immunoglobulin derived fragment. [0018] The terms "chimera" or "chimeric protein" mean a protein containing residues or domains from at least one homologous protein derived from another species. For example, the chimeric antibody contains variable domains typically derived from mouse mAB fused to human immunoglobulin constant domains.

[0019] The terms "bait" or "protein-bait" mean a designed polypeptide containing a pre-selected or constructed domain that will find use in negative or positive selection of binding molecules binding a target ligand from a library of potential binding molecules.

[0020] An "epitope" is a three-dimensional region of a target ligand that corresponds to a structural unit bound by one antibody. Epitopes usually consist of clusters of chemically active particle surfaces, such as amino acid or sugar side chains, and usually have a specific three-dimensional structure, as well as specific characteristics of electrical charges. Conformational and non-conformational epitopes are distinguished in such a way that the first, unlike the second, loses its binding capacity in the presence of denaturing agents. The epitope may be inside the molecule or contain previously described functional units or protein domains with a characterized structure, such as receptor binding domains or fibronectin-like domains. Therefore, when the epitope is a functional domain of a protein, attachment by a particular binding molecule results in the expected modulation of the target ligand function that is antagonistic or agonist to the function of the target ligand.

[0021] The term "surrogate" means having an analogous biological function. The replacement antibody performs an analogous function, enhances or abolishes the activity of the target ligand in the context of animal species other than the one from which the antibody is derived.

[0022] By the term "human" or other species, an antibody, e.g. a human antibody, is meant an antibody whose variable regions, either variable or constant, are derived from, or are very similar in structure to, human or other species germline immunoglobulins. Antibodies of the invention may contain amino acid residues not encoded by germline immunoglobulin sequences (such as, but not limited to, mutations introduced by non-directional or site-specific mutagenesis in vitro or by somatic mutagenesis in vivo). Therefore, as used herein, the term "human antibody" refers to an antibody in which virtually any part of the protein (e.g., CDR, framework, CL and CH domains (e.g., CH1, CH2, CH3), the hinge region (VL, VH)) is much similar to antibodies encoded by human nucleotide germline sequences. Human antibodies have been classified based on similarities in their amino acid sequences, see e.g.<a href="http://people.cryst.bbk.ac.uk/-ubcg07s/">http://people.cryst.bbk.ac.uk/-ubcg07s/</a>. Therefore, by using sequence similarity search, a "human antibody" can be created based on an antibody with a similar linear sequence. Embryonic rodent sequences are also known and can be used in a similar manner. As the immunoglobulin data encoded by the germline sequences of other species are collected and ordered, they can also be used to generate non-human antibodies of the invention using phage expression libraries or other collections of antigen binding fragments using methods currently known to those skilled in the art.

[0023] In one aspect, the present invention requires the use of phage expression libraries and combinatorial peptide libraries. These libraries have become a useful and flexible tool for studying the interaction of peptides and proteins. The phage library can be created by inserting a library of random oligonucleotides or a library of polynucleotides containing interesting sequences, such as those from B cells of immunized animals or humans (Smith, GP 1985. Science 228: 1315-1317). Phage antibody libraries contain pairs of variable region fragments in one phage particle covering both heavy (H) and light (L) chains, which allows expression of single-chain Fv or Fab fragments (Hoogenboom, et al. 2000. Immunol. Today 21 (8) 371 libraries phagemid may be to increase and / or alter the monoclonal antibodies of a given library to generate and then identify additional, desired, human monoclonal antibodies. For example, genes encoding heavy (H) and light (L) chain antibody molecules can be randomly mixed (shuffled) to create new HL pairs in a complex immunoglobulin molecule. In addition, genes encoding one or both H and L chains can be mutagenized in the complementarity determining region (CDR). complementarity determining region) being part of the antibody variable region and then screened to find those with the desired affinity and neutralizing properties. Antibody libraries can be generated artificially by

8). Differentiation modified in immunospecificity by selecting one or more human antibody framework sequences and introducing a collection of CDR cassettes derived from the repertoire of human antibodies or by designed variations (Kretzschmar and von Ruden 2000, Current Opinion in Biotechnology, 13: 598-602). Positions of changes are not limited to CDRs only, but may also apply to variable regions framework regions.

[0024] Other libraries that are useful in the method of the invention include expression phage libraries obtained from non-human animals or artificially generated antibody libraries. An example of the former is the use of immunoglobulins derived from libraries of camelid family species that are naturally devoid of light chains (Hamers-Casterman et al., 1993, Nature 363: 446-448; Gahroudi et al., 1997, FEBS Lett.). Examples of second are single domain antibodies that are derived from the variable domain of one of the heavy or light chains with binding properties as described in US Pat. Well. 6248516.

[0025] Furthermore, in various attempts to understand certain mechanisms or to discover new drugs or drug targets, phage libraries of peptides or antibodies can be combined with different types of phages or other expression systems (ribosome, yeast, bacterial or animal cell based). For example, the phage expression peptide library can be used to screen the antibody phage library. By using the elimination process and the combination of phage expression and subtraction methods, differences in specificity between very closely related enzymes can be found, and this information can be used to create highly selective inhibitors (Ke, SH, et al.

1997. J. Biol. Chem. 272 (26): 16603-16609).

[0026] The binding between ligand and receptor such as antigen and antibody depends on hydrogen, hydrophobic bonds, electrostatic forces and van der Waals forces. All these bonds belong to weak, non-covalent bonds and yet the association of antigen and antibody is one of the strongest connections found in nature. Like antibodies, antigens can also be multivalent, either by having multiple copies of the same epitope, or by the presence of multiple epitopes that are recognized by different antibodies. Interactions using multivalence can lead to the formation of complexes with greater stability, however multivalence can also result in steric problems, and thus a reduction in binding capacity. However, all antigen-antibody bonds are reversible and occur according to the basic laws of thermodynamics for each reversible bimolecular interaction:

_K^k°ⁿ_L FAB-Ab] ^AND k_off [Ab] * [Ag] [0027] Where KA is an affinity constant, Ab and Ag are the molar concentrations of free binding sites on the antibody and antigen, respectively, and Ab-Ag is the molar concentration of antigen-antibody complexes. The forward reaction is called association, and the separation or backward reaction is called dissociation.

[0028] For an efficient reaction to occur between an antigen and an antibody, the epitope must be readily available for binding. As the antigen particles are spatial, the existence of an antibody-recognized epitope may depend on the presence of a particular three-dimensional conformation of the antigen (e.g., a unique site resulting from the interaction of two native protein subunits), or the epitope may correspond to a simple region of sequence. Such epitopes are called conformational or linear, respectively.

Method of the Invention [0029] The inventors of the present invention have disclosed a method of isolating antibodies or other ligands that bind a predetermined epitope, which method is based on the targeted selection of antibodies displayed on the surface of the phage using generated competition proteins (Fig. 1). The method is based on information on the structure of the target protein that is used to design the appropriate protein bait. This bait is used as a competitor in the selection process for antibodies with the desired epitope specificity from the antibody phage library (Fig. 1).

[0030] The specificity of the antibodies is obtained by traditional methods of immunizing animals, which is the result of the interaction of the protein antigen on the animal's immune system. Thus, immunized antibodies often interact with immunodominant epitopes that are different from the desired target epitopes. Existing antibody selection methods using antibody phage libraries cannot be targeted precisely at a specific epitope. The method described herein has the advantage of allowing precise and effective targeting of selection for specific epitope specific antibodies.

[0031] The method of selecting antibodies that bind a predetermined epitope can be used to modify specific properties of targeted therapeutic antibodies or ligand-binding particles (biotherapeutics) that have demonstrated efficacy in one species (e.g., animal model species) to directly obtain an analogous biotherapeutic agent that will be effective in other species. Otherwise, human biological medicines can be simply modified to be effective in the treatment of other mammals, e.g. cattle, pigs, dogs, cats, other important agricultural groups, e.g. poultry, domestic animals or rare species, or endangered species.

[0032] Of the 15 most common diseases affecting human animals (dogs, cats, horses), many have a hormonal background: canine and cat diabetes, canine and cat thyroid disease, hypothyroidism in dogs, hyperthyroid cat, Addison and Cushing disease and in dogs. Other common diseases among accompanying or other animals include osteoarthritis or various forms of cancer. Thus, there is potential for the effective use of human biological drug therapies, such as anti-cancer or anti-inflammatory therapies in the treatment of other species. For example, the drug REMICADE (infliximab), which binds a unique epitope of human TNF alpha, using the method of the invention can be modified to be effective in the treatment of animals accompanying human suffering from TNF alpha-mediated disorders, which are common in these animal species.

Selection of target binding site [0033] Each lymphocyte produces antibodies that are not specific for the whole antigen, but for a specific epitope. While the antigen is part of a foreign cell, particle or molecule that is recognized by the immune system and is the target for antibodies and / or cytotoxic T cells, the epitope is the binding site corresponding to the antigenic determinant of the protein. Polypeptides, lipids, nucleic acids and other particles can function as antigens. An immune response can also occur when smaller particles called haptens are present if they are chemically bound to larger carrier proteins, such as bovine serum albumin or hemocyanin or other synthetic matrices. Haptens can be many molecules such as drugs, sugars, amino acids, small peptides, phospholipids and triglycerides. Antigens that are able to strongly stimulate the immune system are referred to as strongly immunogenic. It has been experimentally proven that antigenic determinants that are able to elicit an immune response can be composed of up to 1 to 8 amino acids, or 1 to 6 monosaccharides. An epitope recognized by an immunoglobulin derived from one clone (monoclonal antibody) may include a larger, discontinuous sequence on the surface of the protein.

[0034] When the epitope lies in the functional region of the protein, the effect of binding the antibody to this protein will be to neutralize the function of the protein due to its structural properties. This thesis is proven and forms the basis for the functioning of therapeutic monoclonal antibodies. Therefore, the ability to repeat the selection of antibodies or other particles binding a specific epitope or protein domain will determine progress in the development of protein-based therapies.

[0035] Epitope mapping for an antibody is one method that allows identification of a functional domain. Epitope mapping can be done with low or high resolution depending on the target. Low resolution mapping requires exposure of the monoclonal antibody kit to the surface sequences of the native protein. It is important here to cover the entire surface of the target protein and identify the sequences relevant to the function of the protein being tested. Unlike the leading mAb candidates, the antibodies used in epitope mapping may have low affinity, should include neutralizing and non-neutralizing mAbs, and in the present method determination of a specific epitope is not always necessary. Once the epitope is identified at the desired resolution, competitive analysis should be performed using the antibody that produces the desired effect and usually has a neutralizing role to identify other particles that bind to this region, for example human antibodies that can compete with rodent antibodies directed against the human target protein .

[0036] Target protein binding antibody kits can also be used in other ways to identify epitopes. For example, the antigen can be digested using proteases and the binding of fragments resulting from this process to the antibody can be determined using ELISA or mass spectroscopy. The antigen-antibody complex may also be digested with proteases and the fragments thus obtained identified by mass spectroscopy. In this case, the antibody masking the sites for proteolysis identifies the epitope.

[0037] Also distinguished are methods that can be used to identify the epitope for a particular antibody. Peptides that correspond to overlapping fragments of the entire antigen sequence can be synthesized and antibody binding to these peptides can be determined by ELISA or surface plasmon resonance. NMR analysis using isotope-labeled antigens can also be used to identify which amino acids have changed with their magnetic environment due to antibody binding. Another technique is to measure changes in melting temperature. The crystal structure of the antigen-antibody complex can be determined and used to identify the epitope. Of these methods, the most effective is the use of crystallography supported by NMR analyzes.

[0038] The ELISA method uses a washing step to remove unbound particles before detection occurs. In case the epitope is a linear particle (the antibody only recognizes a single linear amino acid sequence), the affinity of the antibody for the peptide fragment containing the epitope may be sufficient to detect binding. When the epitope is conformational and consists of at least two discontinuous amino acid sequences within the protein, the affinity of the antibody for each sequence may be low and undetectable. Using surface plasmon resonance, binding of a peptide reflecting a conformational epitope may not be detected when the affinity of the antibody for each epitope peptide is low. Furthermore, binding may not be detected when rapid dissociation of a previously bound peptide occurs. [0039] Epitopes on proteins can also be identified by nuclear magnetic resonance (NMR). Parallel application from the inventors (US Ser. No. 10/393926) presents a technique by which specific atoms (usually H1, C13 and N15) are determined, hence amino acid residues, based on their local environment. The complete assignment of most or all resonances can be made for large proteins provided that there is sufficient time and apparatus with sufficient resolution. This method is based on the observations that when the antibody binds to the antigen, the local environment of some amino acids changes. The amino acids that undergo the greatest changes are most involved in contact with the antibody.

It is theoretically possible to identify an epitope by performing all NMR matches for antigen and antibody in both bound and unbound form and determining which amino acids exhibit atom changes. The complexity of NMR spectra for antigen-antibody complexes, however, means that these types of analyzes are extremely complex and not applicable in the routine identification of epitopes. However, the method proposed by the inventors allows the identification of protein epitopes using proteins enriched with any of the C13 or N15-labeled amino acids, where accurate NMR signal assignment is not always necessary. Multiple labeling of two or more different amino acids in the same protein can be used when signals for different amino acids are properly distinguishable. For example, alphaN15-alanine and epsilon-N15-lysine can be incorporated into one protein similar to epsilon-N15-histidine and alpha-N15-leucine. Epitopes can be both antibody and ligand binding regions. In addition, epitope identification can be assisted by using particle modeling or algorithms that predict sequences exposed on the surface of proteins. [0040] The epitope can also be designed based on the primary protein-target structure, i.e. its amino acid sequence if no measurements are made for this protein. For example, from 5 to 10 amino acid residues within a segment of about 5-15 amino acids in a protein may be altered to create a bait protein variant or chimeric target protein to then study binding and determine the epitope.

[0041] Homology among proteins is determined on the basis of similarities in the sequence of genes or amino acid sequences of proteins that indicate a common evolutionary origin. Basically, a similarity in the structure or function of proteins will occur when evolutionary origin is common. However, it may happen that divergent evolution and mutations cause that proteins with similar structure perform completely different functions, or vice versa, proteins with the same function differ significantly in sequence. They are called orthologs or paralogs, respectively. Mutagenesis to search for homologues is a known strategy for identifying receptor binding regions. This effect is obtained by substituting analogous homologous regions of proteins to preserve the native three-dimensional structure of the original protein, e.g. replacement of human growth hormone regions with regions from porcine growth hormone, human prolactin or human placental lactogen, after which the binding constants for the resulting constructs are determined. These natural structure variants can also be used to determine domains or epitopes within domains useful in constructing the appropriate bait proteins of the invention.

Bait Construction [0042] The term protein-bait is used herein to refer to a protein that differs from a protein cell by at least one structural property within a specific domain including a functional site or region that is bound. Therefore, the bait may be a chimeric target protein or the bait may be a naturally occurring protein, such as a homologous protein and a target ligand, may be a generated sequence that contains a preselected domain. In the method of the invention, the protein-bait binds particles with low affinity and non-specific binding as well as particles forming complexes with the target protein that will be retained and thus selected. In one aspect, a suitable structural homologue that can serve as the basis for selecting a target epitope may be a protein-target ortholog. The structural homologue can also be another representative of the same group of genes.

[0043] With the evolution and storage of large amounts of three-dimensional structure data obtained using X-ray crystallography, NMR and other techniques, information on the protein structure can be successfully obtained or modeled using various methods. The University of Glasgow's Bioinformatics Research Center provides access to a website that allows describing and comparing protein structures using TOPS ( Topology of Protein Structure) (TP Flores, DS Moss and JM Thornton. 1994. Protein Engineering, 7: 31-37). Protein coordinates - in a format such as in the PDB database can be uploaded to a server. The structure is presented in the form of an animation called TOPS animation, and then compared with a set of unique, known structures. The results are returned as a sorted list showing the degree of compression, the structure identification number and a commonly known zero-one description, where 1 means pairs of identical structures and 0 two structures that do not have common structural features.

[0044] The MASS algorithm (Multiple Alignment by Secondary Structures) is based on two-stage matching using both the secondary structure and the atom mapping. The premise for this approach is that proteins are actually composed of secondary structure elements (SSEs). These are the regions in the protein that provide framework stabilization for functional sites. Thus, SSEs are very conservatively evolutionarily, and mutations often appear in elastic loops, which are therefore difficult to fit. MASS is a highly efficient way to structurally match multiple protein particles and detect common structural motifs. Using secondary structure information helps reject noise and achieve performance and reliability. The advantage of MASS is its independence from sequence order and thus the ability to detect non-topological structure motifs in multiple matches or subsets. Using MASS, you can perform protein-protein docking, which is usually problematic. MASS is available for free on the site<a href="http://bioinfo3d.cs.tau.ac.il/MASS/">http://bioinfo3d.cs.tau.ac.il/MASS/</a>. (Dror, O. et al. Protein Science (2003), 12: 2492-250.) [0045] The present invention relates to a method of associating protein structure information with extensive protein and nucleic acid libraries in expression systems to select antibodies specific for unique epitope. For example, the target may be a complex and specific epitope on a rodent-derived human tissue factor (TF) homologue. Antibodies existing in the art either do not block mTF function or are not specific competitive inhibitors of TF binding factor X. The antibodies described herein possess these properties and, in addition, are currently unavailable tools for assessing the therapeutic potential of anti-TF antibodies that neutralize TF activity by inhibiting factor X activation. In addition, these antibodies are a valuable reagent in distinguishing TF functions in normal and thrombotic inflammation, angiogenesis, cancer and developmental processes.

Isolation of specific epitope-directed antibodies or other ligands [0046] To isolate the specific epitope-directed antibodies or other ligands of the invention, three principal approaches are used: (1) competitive selection using expression libraries of antibodies or other potential ligands; (2) non-competitive selection using expression libraries followed by a search for differential binding activity; and (3) immunizing the animal and then searching for differential binding activity.

[0047] In competitive selection using bait proteins, the expression library is selected based on binding to the target protein in the presence of a bait protein in molar excess relative to the target protein. The selectivity of the antibodies or other ligands thus obtained is confirmed by checking protein binding - target and bait binding.

[0048] Thus, one example of the use of the present method is a method for identifying polypeptides that bind to previously selected target protein epitopes, comprising the following steps: (a) creation of a library of phage particles on the surface of which the polypeptides are expressed, (b) preparation of a protein-bait that has an altered amino acid sequence corresponding to a previously selected target-protein epitope, (c) incubation of the library of phage particles with the target protein for selection phage particles containing target protein binding polypeptides, (d) addition of excess bait protein as competitor, to select negative phage particles specific for the previously selected epitope by negative selection, (e) separating the phage particles that bind the target protein from those that bind the protein-bait, and (f) recovering the phage particles bound to the protein-target rather than the bait.

[0049] It is less preferred to use native protein as bait to select for binding to a chimeric or mutated protein. In this case, the protein that contains the primary skeletal protein is used in excess of the chimeric or mutated protein. The selectivity of recovered antibodies or other ligands is confirmed by their binding to bait protein, target protein and skeletal protein binding.

[0050] In two-step selection using protein-bait, the expression library is screened against the target protein. The recovered antibodies or other ligands are then screened (usually separately) for selective binding of the target protein and binding of the bait protein.

[0051] In a variant using immunization and protein bait, an animal species suitable for isolation of stable hybridomas producing monoclonal antibodies is subjected to protein-target immunization. Hybridomas are produced which are then screened for expression of antibodies binding target non-binding bait protein. The immunization approach can be combined with any antigen and involving the above-mentioned presentation strategies. Therefore, mRNA from immune cells (e.g. spleen or peripheral blood lymphocytes) is used to generate a library of antibodies, which are then processed as described in any presentation approach. This approach is not limited to animals suitable for the isolation of stable hybridomas.

[0052] Peptide libraries may be designed in accordance with the method detailed in the present description and methods known to those skilled in the art (see, e.g., US Patent No. 5,723,286 issued March 3, 1998, Dower et al.). In another aspect, commercially available phage expression libraries (e.g., RAPIDLIB 'or GRABLIB', DGI BioTechnologies, Inc., Edison, NJ; Ph.D. C7C Disulfide Constrained Peptide Library, New England Biolabs) can be used. [0053] Antibody libraries are available, e.g., from Cambridge Antibody Technology, Morphosys, Affymax Research Institute, Palo Alto, CA. Many strategies have been devised to select the appropriate set of ligands for further analysis and affinity maturation. These include blocking immunodominant epitopes by competitive deselecting, obtaining a broader range of antibody specificities using epitope masking strategies, and "lifting with capture" search. capture lift), antibody-targeted selection using the capture-sandwich ELISA technique, neighborhood-based antibody selection (ProxiMol), isolation of human monoclonal antibodies using targeted selection of mouse monoclonal antibodies, selection of antibodies specific to cell surface antigens using techniques magnetic sorting, isolation of scFvs binding human surface antigens of cancer cells, subtractive isolation of single chain antibodies using tissue fragments, selection of antibodies based on binding kinetics, selection of functional antibodies based on valence (Antibody Phage Display. Methods and Protocols. IN: David WJ

Coomber, Ed. Methods in Molecular Biology. Humana Press. Vol.

178, December 2001 pps. 133-145).

[0054] Phage affinity enhancement is based on the low degree of dissociation of target ligands. Slow dissociation is usually a sign of high affinity. In this example of affinity enhancement, continuous incubation of the target phage and target binding phage is used in the presence of a known target binding particle at saturating concentration or by increasing the volume of the reaction mixture. In each case, rebinding of dissociated target binding phage is avoided, and higher affinity phages are recovered over time.

[0055] The preincubation time and conditions are optimized for each target binding particle of interest. To monitor the impact of changing conditions on affinity enhancement, pilot panning experiments are performed. After incubation of the target with the target binding phage and subsequent transformation of the host cells, these cells are transferred to selection media and evaluated. Determining the changes in survival colonies is a convenient tool for assessing the increase in affinity. As the number of surviving colonies decreases, the number of surviving weak binding particles decreases, leaving fewer target binding particles with higher affinity. For example, a decrease in the number of surviving colonies to levels of 1%, 0.1% or 0.001% indicates optimal conditions for selection of particles binding the target with higher affinity. In some circumstances, the number of surviving colonies may be limited to about 100 for further analysis by sequencing.

[0056] Depending on the variety of types of library used, the number of particles binding the target with greater affinity may be less than 10.

[0057] The use of the above affinity enhancement techniques allows it to be strengthened without using additional panning rounds. Affinity enhancement techniques can be used alone or in combinations. It should be understood that within the scope of the present invention it is possible to use multiple panning rounds to ensure an increase in affinity, if necessary.

[0058] Citations: All publications and patents cited herein represent the current state of the art: Ausubel, et al., Ed., Current Protocols in Molecular Biology, John Wiley &amp; Sons, Inc., NY, NY (1987-2004); Sambrook, et al., Molecular Cloning: A Laboratory Manual, 2nd Edition, Cold Spring Harbor, NY (1989); Harlow and Lane, antibodies, a Laboratory Manual, Cold Spring Harbor, NY (1989); Colligan, et al., Eds., Current Protocols in Immunology, John Wiley &amp; Sons, Inc., NY (1994-2004); Colligan et al., Current Protocols in Protein Science, John Wiley &amp; Sons, NY, NY, (1997-2004). [0059] The main principles of the invention have been described above. Embodiments of the invention are illustrated in the following examples.

EXAMPLE 1:

DESIGN AND PRODUCTION OF CHIMERIC HUMAN PROTEIN / MOUSE TISSUE FACTOR [0060] Mouse antibody designated no. TF8-5G9 recognizes and binds human tissue factor (TF) and prevents association

Factor X with TF or with the TF / Factor VIIa complex (Ruf, W. and

Edgington, TS 1991. Thromb. Haemost. 66: 529-539). Based on the analysis of the crystal structure of the Fab regions of TF8-5G9 in complex with TF, it was established that all amino acid residues that make up the Fab-recognized epitope are between residues 149 and 204 of human TF. The role of this protein region as an important TF interactor with the Factor Gla domain is also known (Ruf et al. 1992). Fifteen specific residues between 149 and 204 amino acids of human TF are in the right location to significantly affect binding energy (Huang, et al. J. Mol. Biol. 275, 873-894). The sequence of extracellular domains of human TF (GenPept Accession No. NP_001984), amino acids 149-204 and mouse TF (GenPept) are shown below.

Accession No. NP_034301), amino acids 152-207. Seven of the fifteen significant residues are identical (residues K149, K165, K166, T167, T170, N171, Q190 of human TF) while eight of the fifteen are different (residues of human TF converted to: Y156T, K169I, V192M, P194F,

V198T, R200Q, K201N and D204G). Residues in bold reflect the residues involved in stabilizing the TF8-5G9: huTF complex. These residues are characterized by a change in free binding energy of 1-4 kcal / mol or more.

Human

149KDLIYTLYYWKSSSSGKKTAKTNTNEFLIDVDKGENYCFSVQAVIPSRTVNRKSTD204

mousey

152KDLGYIITYRKGSSTGKKTNITNTNERSIDVEEGVSYCFFVQAMIFSRKTNQNSPG207 [0061] Based on this analysis and based on the coding sequence of the mouse Tissue Factor chimeric protein-bait can be constructed. To do this, a mutation must be introduced in murine TF, in residues that are in contact with TF8-5G9, in such a way that these residues correspond to the residues of human TF in positions according to the above matching. In addition, there are other positions that have amino acid residues that differentiate mouse and human tissue factors, but they do not affect the overall function or structure of the protein with respect to the target epitope. Using a mouse TF gene as a template, a chimeric protein was constructed that has mutations at eight sites corresponding to residues unique in TF8-5G9, which come into contact with mouse TF, where the changes introduced correspond to residues of human TF (SEQ ID NO.1). The protein anchoring regions in the membrane were removed to express only the soluble TF extracellular domain, and the C-terminal histidine tag was added to facilitate purification. Expression of soluble murine TF and chimeric protein occurred in HEK 293E cells and these proteins were purified from these cells. The purified protein was analyzed by SDSPAGE to check the molecular weight of chimeric (human mouse) TF (40kDa) and mouse TF (35 kDa).

[0062] Pannig in solution using the HuCAL expression phage library (Morphosys, Martinsreid, Germany) was made using biotinylated mouse TF protein. The chimeric human-mouse TF protein was used in a molar tenfold excess as bait to reject phages specific for epitopes other than the target TF mouse epitope. The biotinylated mouse TF phage was recovered using streptavidin-coated magnetic beads. All particles thus bound were sequenced and twenty-three unique Fab regions were obtained. At the concentration used in the tests, 9 of them recognized only mouse TF, 3 more recognized mouse TF rather than human-mouse TF, while 11 recognized both proteins to a similar degree (Table 1).

TABLE 1

<td>Experiment panning</td><td colspan="3">Binding of Fab clones</td>

<td></td><td>mTF &gt;&gt; h / mTF</td><td>mTF&gt; hu / mTF</td><td>mTF = h / mTF</td>

<td>The competition (m / hTG = 10X mTF)</td><td> 9/23</td><td> 3/23</td><td> 11/23</td>

<td>only mouse mTF</td><td> 1/7</td><td> 2/7</td><td> 4/7</td>

[0063] Panning of mouse TF without a chimeric protein as a competitor was performed to verify that the selected Fabs are the result of epitope-directed selection rather than the occurrence of so-called "Hot spots" (sites of accumulation of random mutations) in mouse TF. Panning conditions were identical in both experiments except for the use of competitive antigen, which was omitted in the control experiment. All bound molecules were sequenced and seven unique Fab regions were obtained. Only one Fab region selected by panning without a competitor specifically binds to murine TF, suggesting that the addition of competitive antigen allows selection of Fab regions that specifically recognize murine TF and not the chimeric human-murine TF protein with a modified TF8-5G9 epitope (Table 1).

[0064] Human anti-mouse TF-specific Fab regions were purified using affinity chromatography and binding to mouse TF or chimeric human-mouse TF was evaluated by ELISA. The CDR sequences of these Fab regions are listed in Fig. 2, the assignment of backbone sequences was made by comparison with the Morphosys HuCAL instruction. Skeletal sequences are listed at the bottom of Fig. 2. All nine mouse TF-specific Fab regions showed binding to mouse TF as a function of dose and minimal cross-reactivity with human-mouse TF (Figure 3). On the Fab scale, PHD127 showed the highest binding affinity for murine TF and PHD103 had the lowest affinity. Five Fab regions (PHD 103, 104, 126, 127 and 130) based on their affinity for mouse TF were selected for conversion to full immunoglobulins. The variable regions of the five Fab regions (PHD 103, 104, 126, 127 and 130) are shown in Figure 2. Sequences numbered 2 to 11 were cloned into vectors to express mlgG2 particles in HEK 293 cells.

Inhibition of coagulation [0065] Selected, murine anti-TF Fab surrogate regions were evaluated for the ability to inhibit coagulation in human plasma, using mouse brain extracts as a source of murine TF. Based on previous experiments, it was expected that the Fab regions that bound the TF8-5G9 epitope of the murine TF protein would interfere with coagulation and delay precipitation of the precipitate. In this analysis, the rate of precipitate formation in human plasma was measured. Four of the eight Fab regions tested delayed or blocked coagulation of human plasma in vitro: PHD 103, PHD 104, PHD 126 and PHD 127. PHD126 and PHD 127 significantly inhibited coagulation in human plasma. Based on the precipitation time curve, depending on the Fab concentration, measurable values ​​of the E50 parameter were obtained for concentrations in the range of 0.2 pg / ml to 63 μς / ml.

TABLE 2

<td>Fab</td><td>Conc. EC50 (μς / ml)</td>

<td>PHD102</td><td> &gt;200</td>

<td>PHD103</td><td> 63,3</td>

<td>PHD104</td><td> 23,8</td>

<td>PHD109</td><td> &gt;200</td>

<td>PHD126</td><td> 0,23</td>

<td>PHD127</td><td> 0,82</td>

<td>PHD128</td><td> &gt;200</td>

<td>PHD129</td><td> &gt;200</td>

Factor X Inhibition [0066] Factor X inhibition was measured in the presence of mouse brain extracts (TF source) by those anti-TF mouse Fab regions that exhibited coagulation inhibition (PHD 103, PHD 104, PHD 126 and PHD 127). The extracts were incubated with FVIIa, in the presence of Factor X anti-TF monoclonal replacement antibodies were added to the mice and inhibition of Factor X to Factor Xa conversion was measured. PHD 103, PHD 126 and PHD 127 inhibited the activation of Factor X (its cleavage) to Factor Xa. Inhibition of Factor X activation was then reassessed using mouse full-length anti-TF Ig. Clear inhibition was observed when PHD 103, PHD 126 and PHD 127 were used, whereas the presence of PHD 104 did not cause inhibition.

FACS Analysis [0067] PHD126 and PHD127, as the most active antibodies, were evaluated for the binding capacity of B6F10 melanoma cells that express murine TF at a high level. PHD126 and PHD127 bound mouse TF associated with the cells in a concentration dependent manner giving an EC50 result of 37.8 nM and 4.35 M, respectively (Fig. 4). Complete variable region sequences, both heavy and light chains, PHD126 and PHD 127 are attached as sequences numbered 6 to 9 and the individual components of the subdomains are shown in Figure 2.

Summary [0068] The experiments presented herein have shown that the selection of antibodies presented in the phage library, oriented to the epitope using the generated competition proteins is a feasible process. The method is based on information on the structure of the target protein, which information allows the design of an appropriate competitor. In addition, the method allows selection of antibodies reacting with a specific epitope of the protein of interest. Existing methods for selecting antibodies from expressive phage antibody libraries do not allow precise targeting of a particular epitope. The advantage of the method described herein is very accurate and effective targeting of the selection to the specific epitope specific antibody. The method used enabled the selection of antibodies binding a unique murine TF epitope.

[0069] TF is a complex molecule that functions as both a receptor and a ligand and is able to form a unique complex with Factor VIIa and Factor X. Therefore, monoclonal antibodies that prevent this interaction must be directed against the unique region of the molecule. Antibodies known in the art do not block the function of mouse TF or are not specific, competitive inhibitors of TF-binding Factor X. The antibodies described herein have such properties and are thus an example of previously unavailable tools for assessing the therapeutic potential of anti-TF antibodies that neutralize TF activity by inhibiting the activation of Factor X. In addition, these antibodies are a valuable reagent in distinguishing the role of TF in normal and thrombotic inflammation processes , angiogenesis, cancer or developmental processes.

EXAMPLE 2: CONSTRUCTION OF CHIMERIC PROTEIN-LURES FOR THE SELECTION OF BINDING PARTICULARS KNOWN DOMAINS ENABLE FOR ACTIVATION OF DIFFERENT RECEPTOR UNITS.

[0070] Interleukin 13 (IL-13) is a cytokine that occurs at elevated levels in the airways of asthmatic patients. IL-13 is produced by activated CD4 T cells⁺ and plays an important role in B cell proliferation, production of class E immunoglobulins (IgE), goblet cell hyperplasia, mucus hypersecretion, eosinophil infiltration and airway hyperresponsiveness observed in patients with asthma. Overexpression of IL-13 in transgenic mice resulted in a phenotype characteristic of asthma, while neutralization of IL-13 by means of antagonistic agents resulted in the reduction of asthma symptoms.

[0071] IL-13 binds to at least two receptors. One of them can be found on the surface of most cell types except T cells, and the other can act as a bait receptor. The receptor, which is involved in proinflammatory activity, is also an IL-4 receptor and consists of two subunits, IL4Ralpha1 and IL13Rbeta1. IL-13 belongs to the short chain cytokine family, which family includes IL-4, IL-2, IL-3 and GM-CSF. These proteins take the form of a bundle of four helices and contain two or three bonds in the form of disulfide bridges. The structure of IL-13 has been decoded and verified based on similarities to other proteins in this family (Eisenmesser, EZ, et al. J. Mol. Biol. (2001) 310: 231-241; Moy, FJ, et al., J Mol. Biol. (2001) 310: 219230). Although the IL-13 sequence is only 25% identical to the IL-4 sequence, the structures of both proteins are quite similar and the interaction of IL-13 with its receptor is expected to be similar to the recently described interaction of IL-4 with its receptor. Indeed, given the fact that the two cytokines share a common subunit at their receptors, it is likely that IL-13 and IL-4 will show structural similarities when interacting with ILRal. The three-dimensional structure of IL-13 in combination with mutation analyzes indicates that there are two sites in the cytokines that perform important functions in interacting with their receptors. This model suggests that the protein region composed of helix A and C interacts with the IL4Ralphal subunit of the receptor and the helix A and D interact with the ILl3Ralphal subunit.

[0072] Based on the IL-13 structure and receptor interaction model, an antibody is expected to block the interaction of R and A helices A and D or A and C helices

IL4Ralpha1 may be a suitable candidate for an IL-13 blocking drug. To direct the selection of antibodies to IL13 regions that interact with the receptor, it was decided to create a chimeric particle of this cytokine. In these chimeric proteins, the loop connecting C and D helixes has been replaced by the corresponding sequence from species used for immunization. In models, the loop between C and D helixes is the most surface fragment of the particle and does not interact with the IL-13 receptor. In addition, this loop is quite flexible and therefore the introduced mutations are unlikely to disturb the topology of the particle. In the resulting chimeric protein, this portion of the protein will be perceived by the host cells as self and probably will not trigger a significant immune response. However, the fragment of the molecule that retains the fully human sequence will be perceived by the host species as alien and will likely trigger an immune response. Antibodies selected from those generated by chimeric antigens are likely to be able to block activity in human receptor based assays.

[0073] There is a need to obtain an effective IL-13 antagonist to assess the benefits of inhibiting this interleukin in disease processes in humans, especially asthma, and thereby obtaining a drug. The new IL-13 variants described herein are useful as immunogens for enhancing the generation of antagonist antibodies, as factors for the screening and selection of neutralizing antibodies, and as direct antagonists of native IL-13. In addition, the development of effective and new IL-13 antagonists may be useful in targeting specific cancers that overexpress IL-13 receptor on the cell surface (Hussain,

SR and Puri, RK, Blood (2000) 95: 3506-351).

[0074] New IL-13 analogues were constructed. These particles can be considered as chimeras of human IL-13 and IL-13 of other species because they partly contain sequences from different species. These mutants were rationally designed by incorporating into the human IL-13 sequence amino acids from sequentially different regions of another species.

[0075] Based on the homology of structures between the two cytokines, a model of the IL-13: IL-13R1 complex has been proposed. Using the IL-13 model obtained in NMR (1 GA3 coordinate file) and the IL-13 sequence, IL-13 analogues were created that demonstrate usefulness as IL-13 agonists, IL-13 antagonists or immunogens or elements used in biopanning to generate antibodies against human IL-13.

[0076] File 1GA3 available on the site <a href="http://www.ncbi.nlm.nih.gov/">http://www.ncbi.nlm.nih.gov/</a> contains a comparison of 20 NMR structures of interleukin 13. Observations of the structure indicate that despite the preservation of 4 helices, the N- and C- terminal sequences and loops between C and D helices are variable, as evidenced by numerous conformations. The first structure from the file was used to analyze designed IL-13 mutants that retained both the structure and function of the original IL-13.

[0077] There is a large loop between C and D helixes, mostly clinging to the hidden B helix. This loop is where mutations can be accepted as this region is distant from A, C and D helixes. B loop is formed by amino acids from Met⁴³ to Asn⁵³ and the CD loop amino acids from Cys⁷¹ down

Thr⁸⁸. The end of the loop is difficult to determine, but it certainly ends before the beginning of the helix D that Glu creates⁹¹. In most structures, the amino acids involved in the interaction between the B helix and the CD loop are:

Helisa B:

Cys⁴⁵, Leu⁴⁸, Glu⁴⁹, and Val⁵⁴

CD loop:

Cys⁷¹, Val⁷⁵, Lys⁷⁴ (probably) How much⁹⁰

Leu⁵¹, probably Asn⁵³, (probably), Val⁸⁵, Arg⁸⁶ [0078] In addition, there are no hydrogen bonds in this region; Pro72 is not involved but essential for turning and there is significant interaction between Trp³⁵ and the rest of the loop from Arg⁸⁶ to Lys⁸⁹.

[0079] Residues of helix B that interact with the CD loop are Leu⁴⁸, Leu⁵¹ and Val⁵⁴.

[0080] Ala⁴⁷ fills the pocket and can be replaced. No hydrogen bonds were found between the CD loop and helix B. [0081] Residues of helix B that interact with the AB loop are: Met⁴³, Ala⁴⁷ and cheese⁵⁰.

[0082] To perform shared identify IL-13 other species search using the Blast algorithm by NCBI. The results are below (<a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Protein">http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Protein</a>):

Human IL-13

GPVPPSTALRELIEELVNITQNQKAPLCNGSMVWSINLTAGMYCAALESLINVSGCSA

IEKTQRMLSGFCPHKVSAGQFSSLHVRDTKIEVAQFVKDLLLHLKKLFREGRFN

Wilderness (Sus scrofa)

G PVP PH STALKELIEELVNTTQNQKTPLCNG SMVWS VNLTTSMQYCAALESLINISDC SAIQKTQRMLSALCSHKPPSEQVPGKHIKDTKIEVAQFVKDLLKHLRMIFRHG

Bovine (taurus sauce)

PVPSATALKELIEELVNITQNQKVPLCNGSMVWSLNLTSSMYCAALDSLISISNCSVI

QRTKKMLNALCPHKPSAKQVSSEYVRDTKIEVAQFLKDLLRHSRIVFRNERFN

Dog

PVTPSPTLKELIEELVNITQNQASLCNGSMVWSVNLTAGMYCAALESLINVSDCSAIQ

RTQRMLKALCSQKPAAGQISSERSRDTKIEVIQLVKNLLTYVRGVYRHGNF

rat

GPVRRSTS PPVALRELIEEL SNITQDQKTS LCNS SMVWSVDLT AGGFCAALESLTNIS SCNAIHRTQRILNGLCNQKASDVASSPPDTKIEVAQFISKLLNYSKQLFRYG

Mouse

GPVPRSVSLPLTLKELIEELSNITQDQTPLCNGSMVWS \ ZDLAAGGFCVALDSLTNISN

CNAIYRTQRILHGLCNRKAPTTVSSLPDTKIEVAHFITKLLSYTKQLFRHGPF [0083] The above sequences of interleukin 13 derived from human, cattle, pig, dog, rat and mouse were compared using the ClustalW algorithm, which is part of the Vector NTI package (InforMax, Inc.), see Bethesda, Fig. [0084] The helix B sequences are shown below. Amino acids that are different than in the human helix B sequence are underlined. Table 3 indicates those residues for which interactions between helix B and CD loop are predicted.

TABLE 3

M YCAALESL1NV

M YCAALDSLISI

MOYCAALESLINI M YCAALESLINV GFCAALESLTNI

GFCYALDSLTNI human bovine swine canine rat mouse interactions [0085] The comparison suggested suggest sites in the helix B and CD loop of human interleukin 13 as well as other species in which amino acids can be substituted, maintaining structural integrity and receptor binding activity. Using probably interacting residues from the B helix and CD loops, a strategy was developed to create chimeric proteins in which the residues of the human protein CD loops were replaced with corresponding residues of the B helix of the same species. In the case of bovine, porcine or mouse IL-13, the preferred implementation requires only one amino acid change, Val⁵⁴ on Ile⁵⁴while all other interacting residues are identical to residues of the human protein.

[0086] Additional implementations of the project include two substitutions in the helix B mouse protein: Glu⁴⁹-&gt; Asp⁴⁹ and Ala⁴⁶-&gt; Val⁴⁶. Tyr⁴⁴ can also be replaced by Phe a Leu⁵¹can be replaced by Val. However, the last substitution described is close to the C helix and may cause structural disorders. The CD loop sequences of the six species of proteins are shown in Table 4, where residue positions are likely to interact with the B loop indicated by asterisks in the last row.

TABLE 4

<img file="PL1747015T3_D0001.tif" />

human bovine porcine canine rat mouse interactions [0087] There are many preferred changes in the CD loop that can be made based on homology. Many of them will not change the final conformation of the loop, but Arg's substitution⁸⁶ proline in mouse and rat IL-13 will cause a significant change in the structure of the CD loop, such as a deletion of three amino acids in mouse and rat proteins. Similarly, the change of Val⁷⁵ on Pro in cattle, pigs and dogs suggests significant conformational changes. In addition, project implementation in this region includes changes to Ala⁴⁶-&gt; Val and Glu⁴⁹-&gt; Asp observed in helix B of mouse protein along with a change in Val⁸⁵ in Leu in the CD loop.

[0088] The amino acids in the B helix and CD loop of various species were substituted with their corresponding human protein amino acids and models of each were built using InsightII and significant optimization, but other programs could also be used for this.

[0089] Psi model: All five constructed models have similar energy. Examination of the models showed that they are quite similar, not only in the CD loop, but also taking into account the rest of the structure. Thus, substitution of canine IL-13 residues with human in the B helix and CD loop will likely result in the corresponding chimeric protein.

[0090] Bovine model: All five models have similar energy. There is a significant difference in the position of the side chains in amino acids 81-85 in all three models but the variability is not more significant than that observed for 20 NMR models. Addition of proline in the CD loop does not significantly affect conformation. Thus, these substitutions are likely to produce acceptable chimeras.

[0091] Pig model: All five models have the same energy. As with the bovine model, there are also differences in the position of the side chains of several amino acids in the loop, but they are not important for maintaining the structure. The addition of an amino acid at the beginning of helix B is desirable. So this variant will probably be a chimera with an acceptable structure.

[0092] Mouse model: After deleting three amino acids in a loop, all five models took on a conformation that was significantly different from the conformation of the human loop model. All are characterized by low energy; compared the energy of all chimer models, mouse models have the lowest energy. However, the final topology of the four helices is usually unchanged and this variant is probably an acceptable chimera.

[0093] All these chimeras have appropriate conformations and should have similar A, C and D helix structures. The sequences of recommended chimeras are listed below:

Human (native)

GPVPPSTALR ELIEELVNIT QNQKAPLCNG SMVWSINLTA GMYCAALESL human-bovine (SEQ ID No: 12) GPVPPSTALR ELIEELVNIT QNQKAPLCNG SMVWSINLTA GMYCAALESL human-pig (SEQ ID NoVSLNLQLPLSLNPLSLNPLSLVL GPL)

GPVPPSTALR ELIEELVNIT QNQKAPLCNG SMVWSINLTA GMYCAALESL human-mouse (SEQ ID No: 15)

GPVPPSTALR ELIEELVNIT QNQKAPLCNG SMVWSINLTA GMYCAALESL human (native)

INVSGCSAIE KTQRMLSGFC PHKVSAGQFS SLHVRDTKIE VAQFVKDLLL

Human-bovine

INISGCSAIE KTORMLSGFC PHKPSAKQVS SEYVRDTKIE VAQFVKDLLL

Human-pig • łJ ·

INISGCSAIE KTORMLSGFC SHKPPSEQVP GKHIRDTKIE VAQFVKDLLL

Human-dog

INVSGCSAIE KTORMLSGFC SQKPAAGQIS SERSRDTKIE VAQFVKDLLL

Human-murine

INISGCSAIE KTORMLSGFC NRKAPTTY S_SLP_DTKIE

VAQFVKDLLLHuman HLKKLFREGR FN

Human-bovine

Human-swine

Human-dog

Human-murine

HLKKLFREGR FN

HLKKLFREGR FN

HLKKLFREGR FN

HLKKLFREGR FN [0094] Chimeric bait proteins are shown in the sequence list as sequences numbered 12 (human-bovine), 13 (human-pig), 14 (human-canine), 15 (human-mouse).

One of their applications is the use of antibodies that neutralize the function of human IL-13 in the selection.

[0095] Antibodies can be recovered by using methods of screening and selecting antibody libraries, such as antibody expression libraries. In another aspect, chimeric IL-13 proteins are useful in the selection and / or screening of neutralizing antibodies. In one application, hybridomas obtained from animals immunized with IL-13 or one or more of these chimeras can be screened for binding to one or more chimeras that have not been used for immunization, thereby avoiding antibodies that recognize CD loop sites. In a second application, these chimeric proteins can be used in other combinations to screen and select combinatorial antibody libraries, essentially phage expression libraries. Based on the design aspects presented herein, selection and screening will stop the identification of antibodies that recognize the CD loop. Thus, both applications are expected to promote the isolation of neutralizing antibodies, essentially those recognizing helix A, C and D, which are fully conserved in species variants and produced chimeric proteins.

[0096] The second use of these mutants is to use them as antagonists of human IL-13. IL-13 is known to bind to two receptor subunits. Small changes in the structure of human IL-13 due to the introduction of non-human amino acids into two regions of the molecule can give an allosteric effect when bound to any subunit. A competitive antagonist was created by selective abolition of subunit binding.

EXAMPLE 3: CALCULATIONS IN PROTEIN DESIGN - LURE BY USING NMR DATA [0097] The present invention relates to new techniques for identifying epitopes on proteins by nuclear magnetic resonance (NMR). NMR is a technique by which specific atoms (usually H1, C13 and N15) are identified and thus amino acid residues based on their local environment. The NMR spectra of carbon and nitrogen are less complex than the spectra of protons, but naturally occurring a significant number of required nuclei may reduce sensitivity. In the case of large proteins, you can often get the effect of overlapping spectra of atoms with a similar environment. Complete assignment of most or all resonances can be done for large proteins if you have sufficient time and sufficiently high resolution equipment. [0098] When the antibody binds to the antigen, the local environment of some amino acids changes. The amino acids that undergo the greatest changes are also the most involved in contact with the antibody. The epitope is identified by performing all NMR spectra assignments for both antigen and antibody in bound and unbound state and determining which amino acids exhibit atom shifts. However, the complexity of NMR spectra for antigen-antibody complexes causes significant difficulties in analysis or completely prevents them using currently available tools and methods, so it is not routinely used to identify epitopes.

[0099] Protein epitopes can be identified using amino acids enriched in either C13 or N15 isotope, the use of which is not always required to precisely identify NMR signals. These epitopes can be binding regions for both antibodies and receptors.

[0100] Using recombinant protein technology, proteins are obtained in media in which one type of amino acid is replaced with an equivalent labeled N15 or C13.

The resulting protein has the same structure and activity as the unlabeled counterpart. The NMR spectra of N15 or C13 proteins are then read in the presence or absence of bound antibody. The natural rare occurrence of resonating nuclei of unlabeled amino acids simplifies the spectra so that the separated spectra will reveal singlets for N15 spectra and singlets, or simple C13 spectra depending on whether the amino acids were labeled equally or specifically. In the event that the labeled amino acid is involved in antibody binding, a resonance shift will occur. For example, if a 200 amino acid protein contains 10 N15-labeled alanines, there will be 10 singlets in the N15 spectrum. If two of them are shifted for the antibody-bound protein, the reason for this will be changes in the local environment. Hence, one can conclude about their placement in the epitope. The specific location of the two alanines in the sequence will not be known only by individual spectra. If the process is repeated twenty times and each time a different amino acid is labeled, it will be possible to know the epitope. If the protein is a recombinant protein, its sequence is known. The epitope location can be determined based on the epitope structure and protein sequence. Algorithms that determine surface-exposed sequences as well as particle modeling can be helpful in identifying the epitope. [0101] Preparation of 20 labeled proteins may not be necessary. If resonances for different amino acids are sufficiently well distinguishable, multiple labeling (i.e. 2 or more different amino acids labeled in one protein) can be used. For example, a-N15 alanine and e50

Nl5 lysine can be incorporated into one protein as well

3-Nl5 histidine and α-Nl5 leucine.

[0102] This technique has many advantages over current epitope identification procedures. Methods using synthetic peptides (PIN, SPOT or solution and ELISA or competition methods) or phages may omit conformational epitopes. NMR procedure, thanks to the analysis of whole proteins, enables detection of conformational and also linear epitopes. Variations of SPOT type synthesis (e.g. peptide matrix) are considered better for identifying conformational epitopes, but the number of peptides required in them increases exponentially with the number of amino acids in the protein, to an average of 2 million when analyzing a protein with a molecular weight of 40kDa. Proteolysis in combination with mass spectrometry can identify conformational epitopes, but causes protein destruction and requires more protein as its molecular weight increases. The NMR procedure does not destroy the molecule. If there is little labeled protein, it can be recovered after each experiment and reused to map another antibody. Point or so-called mutations "Alanine search" in a protein may well serve to identify both linear and conformational epitopes, but the difficulty in this case is that each protein requires its own DNA. In addition, not all molecules with specific mutations are secreted and its folding must be determined for each individual. The NMR procedure uses the same DNA for all labeled proteins and these are assembled and secreted identically to unlabelled proteins. In identifying epitopes 'Gold standard' is crystallography. Its disadvantages are high time expenditure, the required amount of protein, difficulties in obtaining adequate crystal diffraction and the fact that each antibody for the same antigen requires the use of a new crystal.

EXAMPLE 4: CREATION OF CHIMERIC PROTEIN-LURES WITH

APPLICATION OF CRYSTALLINE STRUCTURE [0103] Based on the crystal structure of IL-4 and its close analog, IL-13, specific receptor binding domains were determined that were targeted for the antibody. To select for particles binding to this region in both proteins, a chimeric antibody was created.

[0104] The crystal structure of IL-4 has been identified. The crystal structure of IL-13 has not been resolved, however, a theoretical model of this molecule has been created. Based on the biological function, it can be stated that both IL-4 and IL13 are therapeutically important proteins. IL-4 has been shown to inhibit autoimmune diseases, and both IL-4 and IL13 tend to enhance the anti-tumor immune response. In addition, as both cytokines are involved in the pathogenesis of allergies, their antagonism would provide therapeutic benefits in allergy and allergic asthma.

[0105] Some protein mutants (e.g. IL-4 protein antagonist Y124D and IL-13 protein agonist R112D, J. Biol. Chem (2000), 275, 14375-14380) have been described in the literature. Using particle modeling, the following new agonists for Il-4 and IL-13 were designed. Because their structure is predicted to be more stable than the structure of native proteins, they are expected to be biologically more strongly bound by cytokine receptors and may potentially find use as anti-neoplastic drugs. In addition, these proteins can be used as stable analogues of native cytokines in a liquid phase panning procedure and as intended by the invention as a selective domain binding agent, e.g. a receptor binding domain antagonist.

[0106] Using molecular modeling, the IL-4 crystal structure and the theoretical IL-13 model deposited at the Brookhaven Crystallographic Database were tested. Several amino acids have been identified within the structures that can be replaced without adversely affecting the structure. Indeed, the energy balance suggests that the resulting structures will be more stable than the native sequences.

The substitutions in IL-4 were Thr ^ Ser "',

Thr²²^ Ser²²,

phe⁴⁵^ Tyr⁴⁵,

^ Phe Tyr⁵⁵ and IL-13 I⁴⁸^ Val⁴⁸,

Gln⁹⁰^ Glu⁹⁰,

Leu&gt; How much⁹⁶, Leu⁹⁹=

How much

phe¹⁰³^ Tyr¹⁰³.

[0107] For IL-4, a database was created (Figures 6A and 6B) containing calculations for exposed amino acids. The first column contains data only for side chains and the second column contains data for both side chains and the backbone. Amino acids slightly exposed on the surface of the molecule, or not exposed at all, are marked in bold / blue.

[0108] Hidden residues were extracted from the table, after which cysteines that could not be substituted without

<td>disorders</td><td>structure.</td><td>Results</td><td>possible</td><td>substitution</td>

<td>shows</td><td>below:</td><td></td><td></td><td></td>

<td>FR</td><td>SIDE FR TOTAL</td><td>Possible substitution</td><td>Energy increase</td><td></td>

(Basic value 603.493 kcal / mol)

<td>7 LEU7</td><td> 0,04</td><td colspan="3"> 0,03</td>

<td>10 ILE10</td><td> 0</td><td> 0</td><td></td><td></td>

<td>11 ILE11</td><td> 0,09</td><td> 0,07</td><td></td><td></td>

<td>13 THR13</td><td> 0</td><td> 0</td><td>Cheese</td><td> 607,457/</td>

<td>14 LEU14</td><td> 0</td><td> 0</td><td></td><td></td>

<td>17 LEU17</td><td> 0,01</td><td> 0,04</td><td></td><td></td>

<td>22 THR22</td><td> 0,01</td><td> 0,1</td><td>Cheese</td><td> 606,487</td>

<td>25 THR25</td><td> 0,07</td><td> 0,05</td><td></td><td></td>

<td>29 VAL29</td><td> 0,01</td><td> 0,03</td><td></td><td></td>

<td>32 ILE32</td><td> 0,01</td><td> 0,06</td><td></td><td></td>

<td>45 PHE45</td><td> 0,04</td><td> 0,03</td><td>Tyr, His</td><td> 609,651/652,501</td>

<td>47 ARG47</td><td> 0,06</td><td> 0,04</td><td></td><td></td>

<td>48 ALA48</td><td> 0,02</td><td> 0,01</td><td>Cheese, Val (?)</td><td> 684,427/1355,443</td>

<td>49 ALA49</td><td> 0</td><td> 0</td><td>Cheese</td><td> 662,543</td>

<td>51 VAL51</td><td> 0,02</td><td> 0,03</td><td></td><td></td>

<td>52 LEU52</td><td> 0</td><td> 0</td><td>How much</td><td> 642,62</td>

<td>55 PHE55</td><td> 0,02</td><td> 0,04</td><td>Tyr</td><td> 612,756</td>

<td>56 TYR56</td><td> 0,07</td><td> 0,06</td><td></td><td></td>

<td>76 HIS76</td><td> 0,02</td><td> 0,02</td><td></td><td></td>

<td>79 LEU79</td><td> 0,03</td><td> 0,02</td><td></td><td></td>

<td>80 ILE80</td><td> 0,09</td><td> 0,07</td><td></td><td></td>

<td>83 LEU83</td><td> 0</td><td> 0</td><td></td><td></td>

<td>86 LEU86</td><td> 0</td><td> 0</td><td></td><td></td>

<td>87 ASP87</td><td> 0,04</td><td> 0,03</td><td>own</td><td> 1139,982</td>

<td>90 LEU90</td><td> 0</td><td> 0</td><td></td><td></td>

<td>93 LEU93</td><td> 0,02</td><td> 0,04</td><td></td><td></td>

<td>94 ALA94</td><td> 0</td><td> 0,03</td><td>Cheese</td><td> 660,925</td>

<td>109 LEU109</td><td> 0</td><td> 0</td><td></td><td></td>

<td>112 PHE112</td><td> 0</td><td> 0,01</td><td>Tyr</td><td> 855,581</td>

<td>113 LEU113</td><td> 0,04</td><td> 0,03</td><td></td><td></td>

<td>116 LEU116</td><td> 0,02</td><td> 0,02</td><td></td><td></td>

<td>120MET120</td><td> 0</td><td> 0</td><td></td><td></td>

[0109] IL-4 structure was minimized using 100 cycles of conjugated gradient, dielectric 100, with all hydrogen using Tripos force field and Kollman-Uni charges. The individual changes presented above were then made and energy calculated. Based on these calculations, the best substitutions were selected: Ser for Thr¹³, Ser for Thr²², Tire for Phe⁴⁵ and Tyr for Phe⁵⁵.

[0110] Then the IL-4 crystal structure was returned and the energy before minimization was calculated. After calculating the energy after making the above four changes, the following results were obtained:

Native crystal structure

<td>Tensile binding energy:</td><td> 231,645</td>

<td>Angle deformation energy:</td><td> 298,91</td>

<td>Torsion energy:</td><td> 453,633</td>

<td>Perpendicular strain energy:</td><td> 46,674</td>

<td>Energy 1-4 van der Waals:</td><td> 386,851</td>

<td>Energy van der Waals:</td><td> 1134,199</td>

<td>Energy 1-4 Electrostatic:</td><td> 30,608</td>

<td>Electrostatic energy:</td><td> 1043,112</td>

<td>Total energy:</td><td>3,625.631 kcal / mol</td>

<td>Structure Ser13, Ser22, Tyr45, Tyr55</td><td></td>

<td>Tensile binding energy:</td><td> 233,378</td>

<td>Angle deformation energy:</td><td> 299,147</td>

<td>Torsion energy:</td><td> 449,605</td>

<td>Perpendicular strain energy:</td><td> 46,468</td>

<td>Energy 1-4 van der Waals:</td><td> 384,334</td>

<td>Energy van der Waals:</td><td> 1125,512</td>

<td>Energy 1-4 Electrostatic:</td><td> 30,677</td>

<td>Electrostatic energy:</td><td> 1043,004</td>

<td>Total energy:</td><td>3612.126 kcal / mol</td>

[0111] A decrease in torsion energy, 1-4 van der Waals and van der Waals energy was observed. Based on energy calculations, this structure is expected to be more stable than the native sequence. The modified amino acids are inside the molecule, but since they have not shown the ability to change the structure of the molecule, the surface topology and therefore the activity of the molecule should not change. [0112] A similar table was created for IL-13 (Figures 7A and 7B). No crystal structure has been published, however a theoretical model of its structure is available. Ten minimization cycles were carried out for both structures - initial and modified.

[0113] Residues within the molecule that can be replaced:

Possible

<td></td><td>FR SIDE</td><td>FR TOTAL</td><td>substitution Increase in energy</td><td></td>

<td></td><td></td><td></td><td></td><td></td>

<td></td><td></td><td></td><td colspan="2">(Basic value 1725.421 kcal / mol)</td>

<td>6 LEU6</td><td> 0,06</td><td> 0,07</td><td></td><td></td>

<td>9 LEU9</td><td> 0,03</td><td> 0,02</td><td></td><td></td>

<td>13 LEU13</td><td> 0</td><td> 0,03</td><td></td><td></td>

<td>17 THR17</td><td> 0</td><td> 0,03</td><td>Cheese</td><td> 1987,449</td>

<td>44 LEU44</td><td> 0,07</td><td> 0,05</td><td></td><td></td>

<td>47 LEU47</td><td> 0,03</td><td> 0,02</td><td></td><td></td>

<td>48 ILE48</td><td> 0,08</td><td> 0,06</td><td>val</td><td> 1722,283</td>

<td>50 VAL50</td><td> 0,06</td><td> 0,05</td><td>How much</td><td> 1841,982</td>

<td>51 SER51</td><td> 0</td><td> 0</td><td>Thr</td><td> 2080,486</td>

<td>52 GLY52</td><td> -</td><td> 0</td><td></td><td></td>

<td>63 LEU63</td><td> 0</td><td> 0</td><td></td><td></td>

<td>66 PHE66</td><td> 0,04</td><td> 0,04</td><td>Tyr / His</td><td> 1995,898/1766,541</td>

<td>69 HIS69</td><td> 0,01</td><td> 0,01</td><td></td><td></td>

<td>72 SER72</td><td> 0,04</td><td> 0,03</td><td></td><td></td>

<td>73 ALA73</td><td> 0,02</td><td> 0,01</td><td></td><td></td>

<td>77 SER77</td><td> 0</td><td> 0,01</td><td></td><td></td>

<td>90 GLN90</td><td> 0,02</td><td> 0,04</td><td>Glu</td><td> 1730,812</td>

<td>92 VAL92</td><td> 0,01</td><td> 0,01</td><td>How much</td><td> 9159,549</td>

<td>95 LEU95</td><td> 0</td><td> 0,01</td><td>How much</td><td> 1631,393</td>

<td>96 LEU96</td><td> 0</td><td> 0</td><td>How much</td><td> 1619,286</td>

<td>99 LEU99</td><td> 0,03</td><td> 0,03</td><td>How much</td><td> 1628,907</td>

<td>103 PHE103</td><td> 0,01</td><td> 0,01</td><td>Tyr</td><td> 1624,121</td>

[0114] Although Tyr in place of Phe⁶⁶ shows high energy, seems to be a good substitution. Higher energy is caused by stronger van der Waals interactions of hydroxyl residues.

[0115] Phe⁶⁶ and His⁶⁹ interact with each other (π-π). Aromatic amino acids must remain in these places.

[0116] Substitution of Ile in place of Val⁹² results in higher energy, but a low degree of minimization significantly reduces it. This is probably an acceptable substitution.

[0117] Tire after Phe¹⁰³ results in an additional hydrogen bond to His⁶⁹ and will be a good substitution.

[0118] Returned to the original IL-13 structure, energy was calculated, substitutions were made, energy was re-calculated, which gave the following results:

Pre-modeled structure

Tensile energy of binding: 290,164

Angle deformation energy: 390,042

Torsion energy: 388.721

Perpendicular strain energy: 30.217

Energy 1-4 van der Waals: 286,329

Energy van der Waals: 210,384

Energy 1-4 Electrostatic: 27,906

Electrostatic energy: -1.547

<td>Total energy: Val48, Glu90, Ile95, Ile96, Ile99, Tyr103</td><td>1662.216 kcal / mol</td>

<td>Tensile binding energy:</td><td> 288,604</td>

<td>Angle deformation energy:</td><td> 383,273</td>

<td>Torsion energy:</td><td> 384,452</td>

<td>Perpendicular strain energy:</td><td> 29,889</td>

<td>Energy 1-4 van der Waals:</td><td> 284,92</td>

<td>Energy van der Waals:</td><td> 228,808</td>

<td>Energy 1-4 Electrostatic:</td><td> 27,989</td>

<td>Electrostatic energy:</td><td> -1,594</td>

Total energy:

1626.342 kcal / mol

<td>[0119] Noticed</td><td>decrease</td><td>stretching energy</td><td>bound by</td><td>and,</td><td>energy</td>

<td>angle deformations</td><td colspan="3">, 1-4 van der Waals, van energy</td><td>der</td><td>Waals</td>

<td>and energy increase</td><td>tensions</td><td colspan="2">torsion and stretching</td><td colspan="2">bonds. These</td>

<td colspan="3">recent interactions can be reduced</td><td>through</td><td colspan="2">removal</td>

<td colspan="3">side chains of the newly inserted Ile.</td><td></td><td></td><td></td>

<td colspan="2">Modified structure</td><td></td><td></td><td></td><td></td>

<td colspan="2">SiłaRMS</td><td>max strength Number</td><td>number</td><td></td><td>time</td>

<td colspan="2">Energy kcal / mol kcal / mol A</td><td>kcal / mol A reps</td><td>eval</td><td></td><td>CPU</td>

<td> 1626,342</td><td> 25,436</td><td> 243,087</td><td> 0</td><td> 1</td><td> 0</td>

<td> 0:00:00.33</td><td></td><td></td><td></td><td></td><td></td>

<td> 1326,768</td><td> 14,851</td><td> 134,040</td><td> 1</td><td> 8</td><td> 0</td>

<td> 0:00:01.11</td><td></td><td></td><td></td><td></td><td></td>

<td> 1207,892</td><td> 10,799</td><td> 137,468</td><td> 2</td><td> 14</td><td> 0</td>

<td> 0:00:01.78</td><td></td><td></td><td></td><td></td><td></td>

<td> 1121,012</td><td> 10,155</td><td> 171,991</td><td> 3</td><td> 20</td><td> 0</td>

<td> 0:00:02.44</td><td></td><td></td><td></td><td></td><td></td>

<td> 1065,839</td><td> 7,613</td><td> 105,188</td><td> 4</td><td> 26</td><td> 0</td>

<td> 0:00:03.11</td><td></td><td></td><td></td><td></td><td></td>

<td> 1031,513</td><td> 6,361</td><td> 73,260</td><td> 5</td><td> 32</td><td> 0</td>

<td> 0:00:03.76</td><td></td><td></td><td></td><td></td><td></td>

<td> 995,717</td><td> 6,643</td><td> 64,166</td><td> 6</td><td> 38</td><td> 0</td>

<td> 0:00:04.42</td><td></td><td></td><td></td><td></td><td></td>

<td> 965,486</td><td> 5,399</td><td> 54,345</td><td> 7</td><td> 44</td><td> 0</td>

<td> 0:00:05.09</td><td></td><td></td><td></td><td></td><td></td>

<td> 945,615</td><td> 5,429</td><td> 61,839</td><td> 8</td><td> 50</td><td> 0</td>

<td> 0:00:05.76</td><td></td><td></td><td></td><td></td><td></td>

<td> 924,480</td><td> 4,655</td><td> 50,214</td><td> 9</td><td> 56</td><td> 0</td>

<td> 0:00:06.42</td><td></td><td></td><td></td><td></td><td></td>

<td> 907,908</td><td> 4,448</td><td> 55,866 1</td><td> 0</td><td> 62</td><td> 0</td>

<td> 0:00:07.08</td><td></td><td></td><td></td><td></td><td></td>

NOTE: The maximum number of counts has been reached (10)

Particle energy: INTERLEUKINY-13 MODEL 1 (THEORETICAL MODEL)

Tensile energy of binding: 49.661 Angle deformation energy: 297.14

Torsion energy: 328.222

Perpendicular strain energy: 8,774

Energy 1-4 van der Waals: 189.249

Energy van der Waals: 8,139

Energy 1-4 Electrostatic: 28,271

Electrostatic energy: -1,556

Total energy:

907.908 kcal / mol

<td colspan="2">Time</td><td> %</td>

<td>number</td><td>CPU (sec)</td><td>the whole</td>

<td> 2</td><td> 0,08</td><td> 1,08</td>

<td> 64</td><td> 7,33</td><td> 98,92</td>

Unbound plays

Energy calculations

Average number of pairs van der Waals interaction + electrostatic = 5677 Average number of pairs van der Waals interaction + electrostatic = 3415 Average number of scaled pairs van der Waals interaction + electrostatic = 248

Initial structure [0120]

Particle energy: INTERLEUKINY-13 MODEL 1 (THEORETICAL MODEL)

<td>Energy</td><td>SiłaRMS</td><td>max strength</td><td colspan="3">number</td>

<td>kcal / mol</td><td>kcal / mol A</td><td>kcal / mol A</td><td>reps</td><td>Eval number</td><td>CPU time</td>

<td> 1622,189</td><td> 25,201</td><td> 243,087</td><td> 0</td><td> 1</td><td> 0</td>

<td> 0:00:00.34</td><td></td><td></td><td></td><td></td><td></td>

<td> 1328,927</td><td> 14,694</td><td> 132,377</td><td> 1</td><td> 8</td><td> 0</td>

<td> 0:00:01.12</td><td></td><td></td><td></td><td></td><td></td>

<td> 1212,237</td><td> 10,675</td><td> 135,392</td><td> 2</td><td> 14</td><td> 0</td>

<td> 0:00:01.79</td><td></td><td></td><td></td><td></td><td></td>

<td> 1127,031</td><td> 10,061</td><td> 169,922</td><td> 3</td><td> 20</td><td> 0</td>

<td> 0:00:02.47</td><td></td><td></td><td></td><td></td><td></td>

<td> 1072,925</td><td> 7,499</td><td> 103,477</td><td> 4</td><td> 26</td><td> 0</td>

<td> 0:00:03.13</td><td></td><td></td><td></td><td></td><td></td>

<td> 1039,540</td><td> 6,289</td><td> 73,212</td><td> 5</td><td> 32</td><td> 0</td>

<td> 0:00:03.81</td><td></td><td></td><td></td><td></td><td></td>

<td> 1004,102</td><td> 6,619</td><td> 64,355</td><td> 6</td><td> 38</td><td> 0</td>

<td> 0:00:04.49</td><td></td><td></td><td></td><td></td><td></td>

<td> 973,968</td><td> 5,370</td><td> 54,031</td><td> 7</td><td> 44</td><td> 0</td>

<td> 0:00:05.15</td><td></td><td></td><td></td><td></td><td></td>

<td> 954,110</td><td> 5,428</td><td> 64,203</td><td> 8</td><td> 50</td><td> 0</td>

<td> 0:00:05.82</td><td></td><td></td><td></td><td></td><td></td>

<td> 932,787</td><td> 4,648</td><td> 46,556</td><td> 9</td><td> 56</td><td> 0</td>

<td> 0:00:06.50</td><td></td><td></td><td></td><td></td><td></td>

<td> 915,954</td><td> 4,442</td><td> 52,618</td><td> 10</td><td> 62</td><td> 0</td>

<td> 0:00:07.16</td><td></td><td></td><td></td><td></td><td></td>

NOTE: The maximum number of counts has been reached (10)

Particle energy: INTERLEUKINY-13 MODEL 1 (THEORETICAL MODEL)

<td>Tensile binding energy:</td><td> 48,807</td>

<td>Angle deformation energy:</td><td> 300,15</td>

<td>Torsion energy:</td><td> 331,002</td>

<td>Perpendicular strain energy:</td><td></td>

<td>Energy 1-4 van der Waals:</td><td> 192,067</td>

<td>Energy van der Waals:</td><td></td>

<td>Energy 1-4 Electrostatic:</td><td> 28,178</td>

<td>Electrostatic energy:</td><td> -1,503</td>

Total energy:

915.954 kcal / mol

Time

Number of CPU (sec)% of total

Unrelated plays 2 0.08 1.07

Energy calculations 64 7.41 98.93

Average number of van der Waals + electrostatic pairs = 5717 Average number of van der Waals + 1- electrostatic pairs = 3426 Average number of scaled van der Waals + electrostatic pairs = 247 [0121] Using particle modeling, IL-4 and IL-13 structures were studied based on the IL4 crystal structure and the theoretical IL-13 model deposited in the Brookhaven Crystallographic Database. Several amino acids have been identified within the structures that can be replaced without adversely affecting the structure. Indeed, the energy balance suggests that the resulting structures will be more stable than the native sequences. Substitutions in IL-4 were

Thr¹³^ Ser¹³,

Thr

Cheese

phe

Tyr⁴⁵,

phe⁵

IL-13

How much⁴⁸^ Val⁴⁸,

Gln '^ Glu

Leu

How much

Leu

How much

phe¹⁰³^ Tyr¹⁰³.

[0122] The complete sequences are given below:

IL-4 construct (SEQ ID NO: 16)

HXCDITLQEI IKSLNSLTEQ KSLCTELTVT DIFAASKNTT EKETYCRAAT VLRQYYSHHE KDTRCLGATA QQFHRHKQLI RFLKRLDRNL WGLAGLNSCP VKEANQSTLE NFLERLKTIM REKYSKCSS

IL-13 construct (SEQ ID NO: 17)

PPSTALRELI EELVNITQNQ KAPLCNGSMV WSINLTAGMY CAALESLVNV SGCSAIEKTQ RMLSGFCPHK VSAGQFSSLH VRDTKIEVAE FVKDIILHIK KLYREGRFN [0123] The underlining of the amino acids means that the structure is smaller than the structure, which is more limited than the structure of the modified structure, which is minimized, the structure of which is more limited, the structure of which is less modified, the less structure is modified, In the method according to the invention these constructions or other constructions prepared in an analogous manner can be used.

22547 / PE / 1 EP 1 747 015 B1

SEQUENCE LIST [0125] &lt;110&gt; Centocor, Inc.

O'Neil, Karyn

Heavner, George Sweet,

Raymond &lt;120&gt; Liquid phase biopanning selection using prepared bait proteins

<td> &lt;130&gt;</td><td>CEN5055</td>

<td> &lt;140&gt;</td><td>TBD</td>

<td> &lt;141&gt;</td><td> 2005-03-21</td>

<td> &lt;150&gt;</td><td> 60/565,633</td>

<td> &lt;151&gt;</td><td> 2004-04-26</td>

<td> &lt;150&gt;</td><td> 60/565,674</td>

<td> &lt;151&gt;</td><td> 2004-04-26</td>

<td> &lt;160&gt;</td><td> 17</td>

<td> &lt;170&gt;</td><td>PatentIn version 3.3</td>

<td> &lt;210&gt;</td><td> 1</td>

<td> &lt;211&gt;</td><td> 221</td>

<td> &lt;212&gt;</td><td>PRT</td>

&lt;2l3&gt; Synthetic &lt;220&gt;

&lt;223&gt; human residues of mouse TF 8 &lt;400&gt; l

Gly Ile Pro Glu Lys Ala Phe Asn 1 5

Leu Thr Trp Ile Ser Thr Asp Phe 10 15

Lys Thr Ile Leu Glu Trp Gln Pro 20

Lys Pro Thr Asn Tyr Thr Tyr Thr 25 30

Val Gln Ile Ser Asp Arg Ser Arg 35 40

Asn Trp Lys Asn Lys Cys Phe Ser 45

Thr Thr Asp Thr Glu Cys Asp Leu 50 55

Thr Asp Glu Ile Val Lys Asp Val 60

Thr Trp Ala Tyr Glu Ala Lys Val 65 70

Leu Ser Val Pro Arg Arg Asn Ser 75 80

Val His Gly Asp Gly Asp Gln Leu 85

Val Ile His Gly Glu Glu Pro Pro 90 95

Phe Thr Asn Ala Pro Lys Phe Leu 100

Pro Tyr Arg Asp Thr Asn Leu Gly 105 110

Gln Pro Val Ile Gln Gln Phe Glu 115 120

Gln Asn Gly Arg Lys Leu Asn Val 125

Val Val Lys Asp Ser Leu Thr Leu 130 135

Val Arg Lys Asn Gly Thr Phe Leu 140

Thr Leu Arg Gln Val Phe Gly Lys 145 150

Asp Leu Gly Tyr Ile Ile Tyr Tyr 155 160

Arg Lys Gly Cheese Ser Thr Gly Lys 165

Lys Thr Asn Lys Thr Asn Thr Asn 170 175

Glu Phe Cheese Ile Asp Val Glu Glu 180

Gly Val Ser Tyr Cys Phe Phe Val 185 190

Gln Ala Val Ile Pro Ser Arg Lys 195 200

Val Asn Arg Lys Ser Pro Asp Ser 205

Cheese Thr Val Cys Thr Glu Gln Trp 210 215

Lys Ser Phe Leu Gly 220

<td> &lt;210&gt;</td><td> 2</td>

<td> &lt;211&gt;</td><td> 118</td>

<td> &lt;212&gt;</td><td>PRT</td>

<td> &lt;213&gt;</td><td>human</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (26)..(35)</td>

<td> &lt;223&gt;</td><td>CDR1</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (50)..(67)</td>

<td> &lt;223&gt;</td><td>CDR2</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (100)..(107</td>

<td> &lt;223&gt;</td><td>CDR3</td>

&lt;400&gt; 2

Gin Val Gin Leu Val Gin Cheese Gly Ala Glu Val Lys Lys Pro Gly Glu 15 10 15

Leu Lys Cheese Ile Cys Lys Gly Cheese Gly Tyr Cheese Phe Cheese Asn Cheese 20 25 30

Trp Ile Ala Trp Val Arg Gin Met Pro Gly Lys Gly Leu Glu Trp Met 35 40 45

Gly Gly Ile Ile Gly Pro Gly His Ser Tyr Thr Lys Tyr Ser Pro Ser 50 55 60

Phe Gin Gly Gin Val Thr Ile Ser Ala Asp Lys Ser Ile Ser Thr Ala

70 75 80

Tyr Leu Gin Trp Cheese Cheese Leu Lys Ala Cheese Asp Thr Ala Met Tyr Tyr

90 95

Cys Ala Arg Ile Asn Met Gly Tyr Phe Asp Tyr Trp Gly Gin Gly Thr 100 105 110

Leu Val Thr Val Ser Ser 115 &lt;2l0&gt; 3 &lt;2ll&gt; lll &lt;2l2&gt; PRT &lt;2l3&gt; Human &lt;220&gt;

&lt;22l&gt; Binding &lt;222&gt; (23) .. (34) &lt;223&gt; CDRl &lt;220&gt;

&lt;22l&gt; Binding &lt;222&gt; (50) .. (56) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (89) .. (99) &lt;223&gt; CDR3 &lt;400&gt; 3

<img file="PL1747015T3_D0002.tif" />

&lt;210&gt; 4 &lt;211&gt; 116 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (26) .. (34) &lt;223&gt; CDR1 &lt;220&gt;

&lt;22l&gt; Binding &lt;222&gt; (49) .. (65) &lt;223&gt; CDR2 &lt;220&gt;

&lt;22l&gt; Binding &lt;222&gt; (98) .. (l05) &lt;223&gt; CDR3 &lt;400&gt; 4

Gin Val Gin Leu Val Gin Cheese Gly Ala Glu Val Lys Lys Pro Gly Glu 15 10 15

Cheese Leu Lys Ile Cheese Cys Lys Gly Cheese Tyr Cheese Phe Thr Cheese Asn Trp 20 25 30

Ile Gly Trp Val Arg Gin Met Pro Gly Lys Gly Leu Glu Trp Met Gly 35 40 45

Trp Ile Tyr Pro Ser Asp Ser Met Thr Arg Tyr Ser Pro Ser Phe Gin 50 55 60

Gly Gin Val Thr Ile Ser Ala Asp Lys Ser Ile Ser Thr Ala Tyr Leu

70 75 80

Gin Trp Cheese Cheese Leu Lys Ala Cheese Asp Thr Ala Met Tyr Tyr Cys Ala

90 95

Arg Tyr Leu Phe Gly Leu Phe Asp Asn Trp Gly Gin Gly Thr Leu Val 100 105 110

Thr Val Ser Ser 115 &lt;2l0&gt; 5 &lt;2ll&gt; l07 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (23) .. (33) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (49) .. (55) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (88) .. (95) &lt;223&gt; CDR3 &lt;400&gt; 5

Asp Ile Glu Leu Thr Gln Pro Pro Cheese Val Cheese Val Ala Pro Gly Gln 15 10 15

Thr Ala Arg Ile Ser Cys Ser Gly Asp Asn Leu Gly Ser Tyr Tyr Val 20 25 30

Cheese Trp Tyr Gln Gln Lys Pro Gly Gln Ala Pro Val Leu Val Ile Tyr 35 40 45

Asn ASp Asn Asn Arg Pro Gly Cheese Ile Pro Glu Arg Phe Gly Cheese 50 55 60

Asn Ser Gly Asn Thr Ala Thr Leu Thr Ile Ser Gly Thr Gln Ala Glu

70 75 80

Asp Glu Ala Asp Tyr Tyr Cys Ala Thr Tyr Asp Cheese Ser Thr Asp Val

90 95

Phe Gly Gly Gly Thr Lys Leu Thr Val Leu Gly 100 105 &lt;210&gt; 6 &lt;211&gt; 119 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (26) .. (34) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (49) .. (65) &lt;223&gt; CDR2 &lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (98)..(</td>

<td> &lt;223&gt;</td><td>CDR3</td>

<td> &lt;400&gt;</td><td> 6</td>

Gin Val Gin Leu Val Gin Cheese Gly Ala Glu Val Lys Lys Pro Gly Glu 15 10 15

Cheese Leu Lys Ile Cheese Cys Lys Gly Cheese Tyr Cheese Phe Cheese Asn Tyr Trp 20 25 30

Ile Gly Trp Val Arg Gin Met Pro Gly Lys Gly Leu Glu Trp Met Gly 35 40 45

Phe Ile Asp Pro Asp Asp Asp Asp Thr Asn Tyr Ser Pro Ser Phe Gin 50 55 60

Gly Gin Val Thr Ile Ser Ala Asp Lys Cheese Ile Ser Thr Ala Tyr Leu 65 70 75 80

Gin Trp Cheese Cheese Leu Lys Ala Cheese Asp Thr Ala Met Tyr Tyr Cys Ala 85 90 95

Arg Ala Leu Tyr Met Gin Gly Gly Cheese Phe Asp Cheese Trp Gly Gin Gly 100 105 110

Thr Leu Val Thr Val Ser Ser 115 &lt;210&gt; 7 &lt;211&gt; 109 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (23) .. (33) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (49) .. (55) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (88) .. (97) &lt;223&gt; CDR3 &lt;400&gt; 7

Asp Ile Glu Leu Thr Gin Pro Pro Cheese Val Cheese Val Ala Pro Gly Gin 15 10 15

Thr Ala Arg Ile Ser Cys Ser Gly Asp Asn Leu Gly Ser Tyr Tyr Val 20 25 30

Cheese Trp Tyr Gin Gin Lys Pro Gly Gin Ala Pro Val Leu Val Ile Tyr 35 40 45

Arg Asp Thr Asp Arg Pro Ser Gly Ile Pro Glu Arg Phe Ser Gly Ser 50 55 60

Asn Ser Gly Asn Thr Ala Thr Leu Thr Ile Ser Gly Thr Gin Ala Glu

70 75 80

Asp Glu Ala Asp Tyr Tyr Cys Gin Cheese Tyr Asp Tyr Gly Val Ser Asn

90 95

Gin Val Phe Gly Gly Gly Thr Lys Leu Thr Val Leu Gly 100 105

<td> &lt;210&gt;</td><td> 8</td>

<td> &lt;211&gt;</td><td> 119</td>

<td> &lt;212&gt;</td><td>PRT</td>

<td> &lt;213&gt;</td><td>human</td>

&lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (26) .. (34) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (49) .. (65) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (98) .. (108) &lt;223&gt; CDR3 &lt;400&gt; 8

Gin Val Gin Leu Val Gin Cheese Gly Ala Glu Val Lys Lys Pro Gly Glu 15 10 15

Cheese Leu Lys Ile Cheese Cys Lys Gly Cheese Tyr Cheese Phe Thr Asn Ser Trp. 20 25 30

Ile Ser Trp Val Arg Gin Met Pro Gly Lys Gly Leu Glu Trp Met Gly

40 45

How much Asp Pro Asp Asp Asp Ser Tyr Thr Cheese Tyr Ser Pro Ser Phe Gin 50 55 60

Gly Gin. Val Thr Ile Ser Ala Asp Lys Ser Ile Ser Thr Ala Tyr Leu

70 75 80

Gin Trp Cheese Cheese Leu Lys Ala Cheese Asp Thr Ala Met Tyr Tyr Cys Ala

90 95

Arg Gly Ala Gly Tyr Gly Arg Met Phe Gly Asp Val Trp Gly Gin Gly 100 105 110

Thr Leu Val Thr Val Ser Ser 115

<td> &lt;210&gt;</td><td> 9</td>

<td> &lt;211&gt;</td><td> 108</td>

<td> &lt;212&gt;</td><td>PRT</td>

<td> &lt;213&gt;</td><td>human</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (23)..(33)</td>

<td> &lt;223&gt;</td><td>CDR1</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (49)..(55)</td>

<td> &lt;223&gt;</td><td>CDR2</td>

&lt;220&gt;

<td> &lt;221&gt;</td><td>binding</td>

<td> &lt;222&gt;</td><td> (88)..(96)</td>

<td> &lt;223&gt;</td><td>CDR3</td>

&lt;400&gt; 9

Asp ile Glu Leu Thr Gln Pro Pro Ser Val Ser Val Ala Pro Gly Gln

5 10 15 ·

Thr Ala Arg Ile Ser Cys Ser Gly Asp Asn Leu Gly Ser Tyr Tyr Ala

25 30

Cheese Trp Tyr Gln Gln Lys Pro Gly Gln Ala Pro Val Leu Val Ile Tyr 35 40 45

Gln Asp Asp Asn Arg Pro Gly Cheese Ile Pro Glu Arg Phe Cheese Gly Cheese 50 55 60

Asn Ser Gly Asn Thr Ala Thr Leu Thr Ile Ser Gly Thr Gln Ala Glu

70 75 80

Asp Glu Ala Asp Tyr Tyr Cys Gly Ala Tyr Thr Tyr Ser Thr Ser Trp

90 95

Val Phe Gly Gly Gly Thr Lys Leu Thr Val Leu Gly 100 105 &lt;210&gt; 10 &lt;211&gt; 118 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (26) .. (37) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (52) .. (67) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt;. (100) .. (107) &lt;223&gt; CDR3 &lt;400&gt; 10

<img file="PL1747015T3_D0003.tif" />

&lt;210&gt; 11 &lt;211&gt; 105 &lt;212&gt; PRT &lt;213&gt; Human &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (23) .. (33) &lt;223&gt; CDR1 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (49) .. (53) &lt;223&gt; CDR2 &lt;220&gt;

&lt;221&gt; Binding &lt;222&gt; (86) .. (93) &lt;223&gt; CDR3 &lt;400&gt; 11

Asp Ile Glu Leu Thr Gln Pro Pro Cheese Val Cheese Val Ala Pro Gly Gln 15 10 15

Thr Ala Arg Ile Ser Cys Ser Gly Asp Asn Leu Gly Glu Lys Tyr Ala 20 25 30

Tyr Trp Tyr Gln Gln Lys Pro Gly Gln Ala Pro Val Leu Val Ile Tyr 35 40 45

Asp Asp Asn Asn Arg Gly Ile Pro Glu Arg Phe Cheese Gly Cheese Asn Cheese 50 55 60

Gly Asn Thr Ala Thr Leu Thr Ile Ser Gly Thr Gln Ala Glu Asp Glu

70 75 80

Ala Asp Tyr Tyr Cys Gln Cheese Tyr Asp Ile Glu Ile Thr Val Phe Gly

90 95

Gly Gly Thr Lys Leu Thr Val Leu Gly 100 105 &lt;210&gt; 12 &lt;211&gt; 112 &lt;212&gt; PRT &lt;213&gt; Synthetic &lt;220&gt;

&lt;223&gt; A chimeric protein based on a human sequence with substitution of bovine homolog &lt;400&gt; 12 residues

<td>Gly 1</td><td>Pro</td><td>val</td><td>Pro</td><td>Pro 5</td><td>Cheese</td><td>Thr</td><td>ala</td><td>Leu</td><td>Arg 10</td><td>Glu</td><td>Leu</td><td>How much</td><td>Glu</td><td>Glu 15</td><td>Leu</td>

<td>val</td><td>own</td><td>How much</td><td>Thr twenty</td><td>Gin</td><td>own</td><td>Gin</td><td>lys</td><td>ala 25</td><td>Pro</td><td>Leu</td><td>Cys</td><td>own</td><td>Gly thirty</td><td>Cheese</td><td>Underworld</td>

<td>val</td><td>Trp</td><td>Cheese 35</td><td>How much</td><td>own</td><td>Leu</td><td>Thr</td><td>ala 40</td><td>Gly</td><td>Underworld</td><td>Tyr</td><td>Cys</td><td>ala 45</td><td>ala</td><td>Leu</td><td>Glu</td>

<td>Cheese</td><td>Leu 50</td><td>How much</td><td>own</td><td>How much</td><td>Cheese</td><td>Gly 55</td><td>Cys</td><td>Cheese</td><td>ala</td><td>How much</td><td>Glu 60</td><td>lys</td><td>Thr</td><td>Gin</td><td>Arg</td>

<td>Underworld 65</td><td>Leu</td><td>Cheese</td><td>Gly</td><td>phe</td><td>Cys 70</td><td>Pro</td><td>His</td><td>lys</td><td>Pro</td><td>Cheese 75</td><td>ala</td><td>lys</td><td>Gin</td><td>val</td><td>Cheese 80</td>

<td>Cheese</td><td>Glu</td><td>Tyr</td><td>val</td><td>Arg 85</td><td>Asp</td><td>Thr</td><td>lys</td><td>How much</td><td>Glu 90</td><td>val</td><td>ala</td><td>Gin</td><td>phe</td><td>val 95</td><td>lys</td>

<td>Asp</td><td>Leu</td><td>Leu</td><td>Leu 100</td><td>His</td><td>Leu</td><td>lys</td><td>lys</td><td>Leu 105</td><td>phe</td><td>Arg</td><td>Glu</td><td>Gly</td><td>Arg 110</td><td>phe</td><td>own</td>

<td> &lt;210&gt;</td><td> 13</td>

<td> &lt;211&gt;</td><td> 113</td>

<td> &lt;212&gt;</td><td>PRT</td>

<td> &lt;213&gt;</td><td>synthetic</td>

&lt;220&gt;

&lt;223&gt; A chimeric protein based on a human sequence with substitutions of porcine homolog &lt;400&gt; 13 residues

<img file="PL1747015T3_D0004.tif" />

&lt;210&gt; 14 &lt;211&gt; 112 &lt;212&gt; PRT &lt;213&gt; Synthetic &lt;220&gt;

&lt;223&gt; A chimeric protein based on a human sequence with substitutions of dog homologue &lt;400&gt; 14 residues

Gly Pro Val Pro Pro Ser Thr Ala Leu Arg Glu Leu Ile Glu Glu Leu

Val Asn Ile Thr Gin Asn Gin Lys Ala Pro Leu Cys Asn Gly Ser Met 20 25 30

Val Trp Ser Ile Asn Leu Thr Ala Gly Met Tyr Cys Ala Ala Leu Glu 35 40 45

Cheese Leu Ile Asn Val Cheese Gly Cys Cheese Ala Ile Glu Lys Thr Gin Arg 50 55 60

Met Leu Ser Gly Phe Cys Ser Gin Lys Pro Ala Ala Gly Gin Ile Ser 65 70 75 80

<td>Cheese</td><td>Glu</td><td>Arg</td><td>Cheese</td><td>Arg</td><td>Asp</td><td>Thr</td><td>lys</td><td>How much</td><td>Glu</td><td>val</td><td>ala</td><td>Gin</td><td>phe</td><td>val</td><td>lys</td>

<td></td><td></td><td></td><td></td><td> 85</td><td></td><td></td><td></td><td></td><td> 90</td><td></td><td></td><td></td><td></td><td> 95</td><td></td>

<td>Asp</td><td>Leu</td><td>Leu</td><td>Leu</td><td>His</td><td>Leu</td><td>lys</td><td>lys</td><td>Leu</td><td>phe</td><td>Arg</td><td>Glu</td><td>Gly</td><td>Arg</td><td>phe</td><td>own</td>

105

110

100 &lt;210&gt; 15 &lt;211&gt; 109 &lt;212&gt; PRT &lt;213&gt; Synthetic &lt;220&gt;

&lt;223&gt; A chimeric protein based on a human sequence with substitutions of mouse homologue &lt;400&gt; 15 residues

<img file="PL1747015T3_D0005.tif" />

&lt;210&gt; 16 &lt;211&gt; 129 &lt;212&gt; PRT &lt;213&gt; Synthetic &lt;220&gt;

&lt;223&gt; Human IL-4 with increased stability &lt;400&gt; 16

His Lys Cys Asp Ile Thr Leu Gin. Glu Ile Ile Lys Cheese Leu Asn Cheese 15 10 15

Leu Thr Glu Gin Lys Ser Leu Cys Thr Glu Leu Thr vąl Thr Asp Ile 20 25 30

Phe Ala Ala Ser Lys Asn Thr Thr Glu Lys Glu Thr Tyr Cys Arg Ala 35 40 45

Ala Thr Val Leu Arg Gin Tyr Tyr Ser His His Glu Lys Asp Thr Arg 50 55 60

Cys Leu Gly Ala Thr Ala Gin Gin Phe His Arg His Lys Gin Leu Ile 65 70 75 80

Arg Phe Leu Lys Arg Leu Asp Arg Asn Leu Trp Gly Leu Ala Gly Leu 85 90 95

Asn Ser Cys Pro Val Lys Glu Ala Asn Gin Ser Thr Leu Glu Asn Phe 100 105 110

Leu Glu Arg Leu Lys Thr Ile Met Arg Glu Lys Tyr Ser Lys Cys Ser 115 120 125

Cheese &lt;210&gt; 17 &lt;211&gt; 109 &lt;212&gt; PRT &lt;213&gt; Synthetic &lt;220&gt;

&lt;223&gt; Human IL-13 with increased stability &lt;400&gt; 17

<img file="PL1747015T3_D0006.tif" />

22547 / PE / 1 EP 1 747 015 B1