Synthetic immunoglobulin domains with binding properties engineered in regions of the molecule different from the complementarity determining regions

The present invention relates to a process for modifying and producing a modified immunoglobulin.

A general area is the modification of proteins in order to improve them with specific binding properties. In particular, the modified proteins in question are immunoglobulins (antibodies), and more particularly, single domains or pairs or combinations of single immunoglobulin domains. Specific immunoglobulin binding properties are important properties because they control the interaction with other molecules such as antigens, and allow the use of immunoglobulins for diagnostic and therapeutic applications.

The basic structure of the antibody will be explained here using intact IgG1 immunoglobulin as an example.

Two identical heavy (H) and two identical light (L) chains combine to form a Y-shaped antibody molecule. Each heavy chain has four domains. Amino terminal variable domains (VH) are present in types Y, followed by three constant domains: CH1, CH2 and carboxy terminal SNZ, at the base of the Y's root. A short stretch, the switch, connects the variable and constant areas of the heavy chain. The joint connects CH2 and SNZ (Fc fragment) with the rest of the antibody (Fab fragments). One Fc and two identical Fab fragments can be produced by proteolytic cleavage of the joint in an intact antibody molecule. Light chains are constructed of two domains, variable (VL) and constant (CL), separated by a switch.

Disulfide bonds in the joint area connect two heavy chains. Light chains are linked to heavy chains by additional disulfide bonds. Hydrocarbon groups connected via Asn are held at different positions in constant domains. depending on the class of immunoglobulin. For IgG 1, two disulfide bonds in the joint region, between the Cys235 and Cys238 pairs, unite the two heavy chains. The light chains are linked to the heavy chains by two additional disulfide bonds, between Cys229s in the CH1 domains and Cys214s in the CL domains. Hydrocarbon groups are linked to Asn306 of each CH2, creating a pronounced protrusion at the Y root.

50830 000 [0006] These traits have profound functional consequences. The variable regions of the heavy and light chains (VH) and (VL) lie in types of Y, where they are located so as to react with the antigen. This type of molecule is the side where the N-terminus of the amino acid sequence is located. The Y structure protrudes in a way that effectively mediates effector functions, such as complement activation and interaction with Fc receptors, or ADCC and ADCP. Its CH2 and SNZ domains protrude to facilitate interaction with effector proteins. The C-terminus of the amino acid sequence is located on the opposite side of the type, which can be called the lower part of Y. The structure of intact IgG1 is shown in Figure 1a.

Two types of light chain, called lambda (λ) and kappa (k), have been found in antibodies. A particular immunoglobulin has λ chains or k chains, never one at a time. No functional difference was found between antibodies having λ or k chains.

The structural organization of monomers of the major class of human immunoglobulin is shown in Figure 1b. The classes differ in the composition and sequence of their respective heavy chains. Both IgM and IgE are missing in the joint area, but each contains an extra heavy chain (CH4) domain. The numbers and locations of the disulfide bonds (lines) that connect the chains differ among the isotypes. They also differ in the distribution of N-linked hydrocarbon groups, symbolically represented as circles.

Each domain in the antibody molecule has a similar structure of two beta plates packed close to each other in a compressed antiparallel beta barrel. This preserved structure is called the immunoglobulin fold. The constant domain immunoglobulin set contains a 3-strand beaker packed opposite the 4-strand plate. The fold is stabilized with a hydrogen bond between the beta threads of each plate, with a hydrophobic bond between the remnants of opposite plates in the interior, and with a disulfide bond between the plates. The 3-strand plate contains strands C, F, and G, and the 4-strand plate has strands A, Β, E, and D. The letters A through G indicate the sequential positions of the beta strands along the amino acid sequence of the immunoglobulin fold.

The set of variable domains has 9 beta threads arranged in two plates of 4 and 5 threads each. The 5-thread board is structurally homologous to the 3-thread board of constant domains, but also contains additional threads C 'and C. The remaining threads (A, B, C, D, E, F, G) have the same topology and similar structure as their opposite parts in the constant domain of immunoglobulin folds, DisulfiDNA bonds connect strands B and F in opposite plates, as in constant domains. The immunoglobulin fold is shown in Figure 2 for the constant and variable immunoglobulin domains.

50830 00 The variable domain of both immunoglobulin chains, light and heavy, contains three hypervariable loops, or regions that determine complementarity (e. Complememarity-detemiining regions, CDRs), Three CDRs of the V domain (CDR1, CDR2, CDR3) collect be at one end of the beta barrel. CDRs are loops that connect the beta strands of the BC, C'-C, and FG immunoglobulin folds. Residues in the CDR vary from one immunoglobulin molecule to another, giving antigenic specificity for each antibody.

The VL and VH domains in antibody molecule types are tightly packed so that 6 CĐRs (3 in each domain) act together in the construction of an antigen-specific binding surface (or nest). The natural antigen binding site in the antibody is therefore composed of loops connecting the BC, S-S, and FG light chain variable domain strands and the BC strands. SS and FG heavy chain variable domain.

Using the 3D structure of the protein as an aid to construction, amino acid residues located on the surface of many proteins are randomized using the structure of the protein nucleus as a skeleton. Examples of this strategy are described or published in the following publications which are incorporated herein by reference: Nygren PA, Uhlen M., Curr Opin Struct Biol. (1997) 7: 463-9; Binz HK, Amstutz P, Kohl A, Stuinpp MT, Briand C, Forrer P, Grutter MG, Pluckthun A. Nat Biotechnol. (2004) 22: 575-82; Vogt M, Skerra A. Chembiochem. (2004) 5: 191-9; US 6,562,617.

The basic principle of this technique is based on the observation that many proteins have a stable nucleus, formed by specific distribution of secondary structural elements, such as beta plates or alpha coils, which are interconnected with structures such as loops, turns or random windings. Typically, these last three structural elements are less crucial to the overall structure of the protein, and the amino acid residues in these structural elements can be replaced frequently without destroying the overall protein fold. A naturally occurring example of this construction principle are CDRs in antibodies. Artificial examples include lipocalins, ankyrins and other protein skeletons.

The loops, which are not CDR-loops in the innate immunoglobulin, do not have an antigen binding or a specific binding epitope, but contribute to the proper folding of the whole immunoglobulin molecule and / or its effector or other functional groups and are therefore referred to as structural. loops.

U.S. Patent 6,294,654 shows that altered antibodies can be produced in which a peptide antigen can be incorporated into a non-CDR antibody loop (Ab) in the CH1 region between the articular region and the variable region, and the resulting Ab can be taken up in

50830 Β

ARS so that the peptide antigen is present on the surface of APC in the context of MHC II, and therefore causes an immune response. These inserted peptides are epitopes and the general structure of the carrier molecule is not important. It has been shown that a ras peptide can be placed on the (non-CDR) loop of an immunoglobulin and the immunoglobulin will still be secreted. There is an established quality control in the cells that prevents the immunoglobulin from being secreted if the op is not properly folded, and the altered amino acid sequence of the loop can cause the protein to fold in the structure that the cell will detect as irregular and therefore degrade. Thus, in addition to the examples presented, it was considered difficult to further modify the structural loops without altering the nature of the immunoglobulin.

U.S. Patent Application 2004/0101905 describes the binding of a molecule comprising a target binding site and an Fc effector peptide. An Fc effector peptide is a peptide that acts together with an effector molecule. Insertion of the effector peptide into the non-CDR loop of the CH1 domain of the immunoglobulin fragment has been shown.

Fc effector peptides are naturally occurring structures in non-CDR loops of antibodies and are therefore not expected to destroy the structure of the immunoglobulin if grafted to different equivalent sites in the immunoglobulin.

However, each peptide grafted into a non-CDR loop according to this publication has a high potential to become inactive due to the different structural environment that was selected.

In both of the above prior art documents, it is claimed that it is difficult to insert peptides into a loop that should retain its structure and function, as it is crucial not to disrupt the immunoglobulin structure of the folds, as this is important for function and secretion.

U.S. Patent Applications 2004/0132101 and 2005/0244403 describe immunoglobulin mutants with altered effector ligand binding affinity, which are natural ligands for antibody structural loops. These documents describe a number of mutations in various regions throughout the immunoglobulin molecule that affect the effector function of the whole antibody.

WO 01/83525 relates to proteins containing Fc domains fused to biologically active peptides, said peptides being fused to Fc domains at their N- or C-termini.

50830 US 2002/0106370 describes chimeric polypeptides comprising a binding moiety that exhibits specific binding affinity for the eukaryotic target cell surface and effector moiety.

WO 02/32925 describes proteins capable of similar antibody function. To achieve this, loop structures have also been introduced into said proteins that correspond in structure and position to CDR loops.

WO 2006/036834 describes molecules and methods in which biologically active peptides are incorporated into the loop region of an Fc domain. The biologically active peptide with the desired biological activity is first selected and then introduced into the Fc domains by binding the peptide to a protein or inserting the nucleic acid in question into the Fc domain encoded by that nucleic acid.

Other prior art documents show that an immunoglobulin-like skeleton has been used to the extent that it modifies an existing antigen binding site, thereby introducing new binding properties. However, if, only the CDR regions have been altered in terms of antigen binding, in other words, in the case of immunoglobulin folds, only the natural binding site of the antigen can be modified to determine its binding affinity or specificity. There is abundant literature describing different formats of such modified immunoglobulins, often expressed as single chain Fv fragments (scFv) or Fab fragments, displayed on the surface of phage particles or by solubility expressed in various prokaryotic or eukaryotic expression systems. Among the leading authors in this field are Greg Winter, Andreas Pliickthun and Hennie Hoogenboom, The aim of the present invention is to provide immunoglobulins with the introduction of new antigen binding sites, and methods for modifying and producing said immunoglobulins.

Therefore, the present invention relates to a method of modifying an immunoglobulin, which comprises modifying the structural region of the loop to obtain a pe-CDR binding site and determining binding of said immunoglobulin to an antigen epitope, wherein the unmodified structural region of the loop does not bind significantly to said epitope , and this process includes the following steps:

- providing a nucleic acid encoding an immunoglobulin that includes at least one structural region of the loop,

50830 Β

- modifying at least one nucleotide residue in at least one of the specified structural regions of the loop by mutagenesis, said modification being either:

a) by a method of mutagenesis selected from random, semi-random, randomization directly on site, mutagenesis scanning and combination approach, or

b) more than one structural loop to obtain on-site binding to said epitope, or

c) excluding the incorporation of a biologically active peptide of 2 to 40 amino acids into the Fc domain.

- transferring said modified nucleic acid to an expression system,

- expression of said modified immunoglobulin,

contacting the expressed modified immunoglobulin with the epilope, and

- determining whether said modified immunoglobulin binds to said epitope, In particular, said modification method involves binding an immunoglobulin specifically to an epitope of an antigen selected from the group consisting of allergens, tumor-associated antigens, self-antigens, enzymes, bacterial antigens antigens, protozoan antigens and viral antigens. By modification in the structural region of the loop, the immunoglobulin can be structurally bound to the epitope. In a preferred embodiment, the immunoglobulin binds specifically to at least two such epitopes, which differ from each other, and are of the same antigen or are of different antigens.

For example, a method of modifying the binding of an immunoglobulin specifically to at least one epitope and comprising at least one modification in at least one structural region of the loop of said immunoglobulin and determines the specific binding of said at least one region of the loop to at least one other epitope, wherein an epitope selected from the group of the above antigens, and wherein the unmodified loop structural region (non-CDR region) is already specific for said at least one other epitope, comprising the steps of:

50830 Β

- providing a nucleic acid encoding the specific binding of an immunoglobulin to at least one first epitope involving at least one structural region of the loop,

- modifying at least one nucleotide residue from at least one said region of the loop encoded by said nucleic acid,

- transferring said modified nucleic acid to an expression system,

- expression of said modified immunoglobulin,

- contacting the expressed modified immunoglobulin with said at least one other epitope, and

- determining whether said modified immunoglobulin binds specifically to another epitope.

The method according to the invention relates primarily to at least one modification in at least one structural region of the loop of said immunoglobulin and to determining the specific binding of said at least one region of the loop to at least one antigen selected from the group consisting of allergens, tumor-associated antigens, self- antigens, enzymes, bacterial antigens, fungal antigens, viral antigens and protozoan antigens, wherein the immunoglobulin containing the unmodified structural region of the loop does not bind specifically to said at least one antigen.

The term immunoglobulins according to the present invention to be modified (as used herein, the terms immunoglobulin and antibody may be interchanged herein) may exhibit mono- or multispecific or multivalent binding properties, at least two, preferably at least three specific binding sites for epitopes e.g. antigens, effector molecules / proteins. The immunoglobulins of the invention are also functional fragments accepted in the art, such as Fc, Fab, scFv, single chain dimers of the CH / CL domain, Fv or other immunoglobulin derivatives or combinations, variable and light chain domains of the variable region (such as Fd, VI, Vk, Vh) and the constant region of the intact antibody such as CH1, CH2, SNZ, CH4, Clambda and C-cap, as well as a minidomain consisting of two beta-strands of the immunoglobulin domain linked by a structural loop.

It is understood that the term immunoglobulin, modified immunoglobulin or immunoglobulin according to the invention also includes immunoglobulin derivatives. A derivative is any combination of one or more immunoglobulins according to the invention and / or fusion

50830 Β a protein in which any domain or minidomain of the immunoglobulin according to the invention can be fused to any position of one or more other proteins (such as other immunoglobulins, ligands, skeletal proteins, enzymes, toxins and the like). The immunoglobulin derivative according to the invention can also be obtained by binding to other substances by various chemical techniques such as covalent binding, electrostatic interaction, disulfide binding and the like.

Other substances associated with immunoglobulins may be lipids, hydrocarbons, nucleic acids, organic and inorganic molecules, or any combination thereof (e.g., PEG, prodrugs, or drugs). A derivative is also an immunoglobulin with the same amino acid sequence, but constructed in whole or in part from non-natural or chemically modified amino acids.

The modified molecules of the present invention will be used as stand-alone proteins, as well as fusion proteins or derivatives, most typically typically fused to become part of larger antibody structures or whole modeling antibodies. disulfide bonding, etc.

Other substances associated with immunoglobulins may be lipids, hydrocarbons, nucleic acids, organic and inorganic molecules, or any combination thereof (e.g., PEG, prodrugs, or drugs). The derivative is also an immunoglobulin with the same amino acid sequence, but constructed in whole or in part from non-natural or chemically modified amino acids.

The modified molecules of the present invention will be used as stand-alone proteins, as well as fusion proteins or derivatives, typically typically fused to become part of larger antibody structures or whole antibody molecules. or parts thereof such as Fab fragments, Fc fragments, Fv fragments and the like. Proteins that are the product of inodification will be able to be used to obtain molecules that are monospecific, bispecific, trispecific, and which may carry more specificity at the same time, and it will be possible to simultaneously control and pre-select binding valence according to the intended use of such molecules.

According to the present invention, antigen binding sites or antigen binding sites of all types of aphergens, tumor-related antigens, self-antigens, enzymes, bacterial antigens, fungal antigens, protozoan antigens and viral antigens can be introduced into the structural loop of a given antibody structures.

50830 The term antigen according to the present invention means molecules or structures that are known to interact with or be able to interact with the region of the CDR loop of an immunoglobulin. According to the prior art, the structural regions of the loops do not interact with antigens, but rather contribute to the overall structure and / or binding to effector molecules.

The terms allergens, lumor-associated antigens, self-antigens, enzymes, bacterial antigens, fungal antigens, protozoan antigen and viral antigens according to the present invention include all allergens and antigens recognizable by the structure of antibodies, and fragments of such molecules (especially substructures that generally sweat as epitopes (e.g., B-cell epitopes)), as long as they are immunologically relevant, i. while they can also be recognized by natural DNA or monoclonal antibodies.

The term epitope according to the present invention means a molecular structure that can completely comprise a specific binding participant or be part of a specific binding participant in the binding domain or immunoglobulin of the present invention.

Chemically, the epitope may or may include a hydrocarbon, peptide, fatty acid, inorganic substance, or derivatives thereof, and any combination thereof. If the epitope is a polypeptide, it will typically include at least 3 amino acids in the peptide, preferably 8 to 50 amino acids, and in particular between about 10-20 amino acids in the peptide. The upper limit of the length of the peptide is not DNA and it can cover almost the entire length of the polypeptide sequence. Epitopes can be linear or conformational epitopes. A linear epitope is composed of a single segment of the primary sequence of a polypeptide chain. Linear epitopes may be adjacent or may overlap. Conformational epitopes are composed of amino acids that are assembled together by folding the polypeptide to form a tertiary structure, and the amino acids are not necessarily adjacent to each other, that is, DNA to each other in a linear sequence.

Specifically, epitopes are at least part of the diagnostically relevant molecules, i. the absence or presence of epitopes in the sample is qualitatively or quantitatively correlated with the disease or health condition or with the state of the process in production or in the environment and in terms of the state of the food. Epitopes can also be at least part of the therapeutically relevant molecules, ie. molecules that can be targeted with specific binding domains that alter the course of the disease.

Allergens, tumor-associated antigens, self-antigens, enzymes, bacterial antigens, fungal antigens, protozoan antigens and preferred viral antigens

50830 Β are those allergens or antigens that have been shown to be or may be immunologically or therapeutically important, especially those for which clinical efficacy has been tested.

On the other hand, according to another embodiment of the present invention, other binding abilities can also be introduced into the structural areas of the loops, e.g. binding ability to malt molecules, such as drugs or enzymes, catalytic sites of enzymes or enzyme substrates, or to an analog of the transition state of an enzyme substrate.

Preferably, the novel antigen binding site in the structural loop is unknown to the unmodified immunoglobulin. Thus, targets such as effector niolecules or Fcreceptors are preferably excluded from the binding molecules and specificity of the immunoglobulins of the invention.

Preferably, novel antigen binding sites in the structural loop are introduced by substituting, deleting and / or inserting an immunoglobulin encoded with the selected nucleic acid.

According to another preferred embodiment of the present invention, the modification of the step of a nucleotide is the substitution, deletion and / or insertion of an imimoglobulin encoded by said nucleic acid.

Modification of at least one region of the loop may result in the substitution, deletion and / or insertion of one or more amino acids, preferably a point mutation, a change in the whole loop, and in particular a change in at least 2, 3, 4, 5, 6, 7 , 8, 9, 10 and up to 30 amino acids.

Preference is also given to a site-directed random mutation. In this method, one or more loop specific amino acid residues are replaced or introduced using randomly generated inserts into such loop structures. Alternatively, the use of combined approaches is preferred.

At least one region of the loop is preferably mutated or modified randomly, semi-randomly, or, in particular, by site-directed random mutagenesis methods. These methods can be used to obtain amino acid modifications at desired positions of the immunoglobulins of the present invention. In these cases, the positions are chosen randomly, or amino acid changes are made using simple rules. For example, all residues can be mutated into alariin, which is referred to as alanine screening. Such procedures may be associated with more complex engineering approaches that use selection methods to search for higher levels of sequence diversity. The preferred method of the invention relates to a randomly modified molecule

50830 Β a nucleic acid containing at least one nucleotide repeating unit having the sequence 5'-NNS-3 \ 5'-NNN-3 'or 5'-NNK-3'.

Randomly modified nucleic acid molecules may contain the above-mentioned repeating units encoding all known naturally occurring amino acids.

It is well known in the art that there are various selection technologies that can be used to identify and isolate proteins with certain binding characteristics and affinities, including, for example, display technologies, such as phage display, ribosome display, cell surface display, and similarly as described below. Methods for producing and screening for antibody variants are well known in the art. General methods of molecular biology for antibodies, expression, purification, and screening are described in Antibody Engineering, Duebel & Kontermann. Springer-Verlag, Heidelberg, 2001; and Hayhursl & Georgiou, 2001, Curr Opin Chem Biol 5: 683-689; Maynard & Georgiou, 2000, Annu Rev Biomed Eng 2: 339-76.

The structural loop or non-CDR loop of the present invention is understood as follows: immunoglobulins are composed of domains with so-called immunoglobulin folds. In essence, the anti-parallel beta plates are connected by a loop to form a compressed anti-parallel beta barrel. In the variable region, some domain loops significantly contribute to the specificity of antibodies, ie. antigen binding. These loops are called CDR loops. All other antibody domain loops contribute more to molecule structure and / or effector function. These loops are defined herein as structural loops or non-CDR loops.

Nucleic acid molecules encoding modified immunoglobulins (and which are always included below throughout the description: immunoglobulin fragments) can be cloned into host cells, expressed, and examined for their binding specificity. These particles are performed using well-known methods, and many methods that can be used in the present invention are described in the Molecuiar Cloning-A Laboratory Manual, 3.sup.rd Ed. (Maniatis, Cold Spring Harbor Laboratory Press, New York, 2001), and Current Protocols in Molecular Biology (John Wiley & Sons). Nucleic acids encoding the modified immunoglobulins of the present invention can be incorporated into an expression vector for expressing said immunoglobulins. Expression vectors typically include an immunoglobulin that is operably linked, i.e., placed in a functional relationship, with control or regulatory sequences, selective

50830 Β markers, any fusion participants and / or additional elements. The modified immunoglobulins of the present invention can be produced by culturing host cells transformed with a nucleic acid, preferably an expression vector, comprising a nucleic acid encoding the modified immunoglobulins, under appropriate conditions to induce or induce the expression of the modified immunoglobulins. Methods for introducing exogenous nucleic acid molecules into a host are well known in the art, and will depend on the host used. Of course. non-cellular or non-cellular expression systems can also be used to express the modified immunoglobulins.

In a preferred embodiment of the present invention, the modified immunoglobulins are purified or isolated after expression. Modified immunoglobulins can be isolated or purified in many ways known to those skilled in the art. Standard purification methods include chromatographic techniques, electrophoretic techniques, immunological techniques, precipitation, dialysis, filtration, concentration, and chromatofocusing. Cleaning can often be accomplished with a special fusion participant. For example, antibodies can be purified using glutathione resin if GST fusion, Ni, is used⁺² by affinity chromatography if His-tag is used or with immobilized anti-flag antibody if flag-tag is used. For general reference to suitable purification techniques, see Antibody Purification: Principles and Practice, 3rd ed., Ed., Scopcs, Springer-Verlag, NY, 1994, Of course, the modified immunoglobulins of the present invention may also be expressed on the host surface, especially on bacterial, insect or fungal cell surfaces or on phage or virus surfaces.

Modified immunoglobulins can be screened using a variety of methods, including, but not limited to, those used in in vitro, in vivo, and cell-based experiments and selection technologies. automated and high-throughput search technologies. A fusion or labeling agent, for example an enzyme, an immunolabeling agent, an isotope labeling or a small molecule labeling such as fluorescent or colorimetric dyes or luminogenic molecules, can be used for the search.

In a preferred embodiment, the functional and / or biophysical properties of the immunoglobulin are examined by in vitro analysis. In a preferred embodiment, the antibody is screened

50830 Β functionality, for example in terms of its ability to act categorically on a reaction or in terms of its affinity for its purpose.

Various detection methods can be used in the assays, including but not limited to chromogenic. fluorescent, Eminiscent or isotopic labeling.

As is known in the art, a subset of search methods are those selected for favorable library members. These methods are referred to herein as selection methods, and they are used in the present invention to search for modified immunoglobulins. When immunoglobulin libraries are searched using a selection method, only those preferred library members, that is, those that meet some of the selection criteria, are propagated, isolated and / or observed. As will be appreciated, since only most of the accepted variants are considered, such a method allows searching libraries that are larger than those that can be searched by the method of examining the consent of individual library members. Selection is made possible by any method, technique, or fusion participant that covalently or noncovalently associates an immunoglobulin phenotype with its genotype, i.e., the function of an antibody with a nucleic acid encoding it. For example, the use of phage display as a selection method allows the fusion of library members to gene III. protein. Thus, the selection or isolation of modified immunogiobulites that meet certain criteria , for example the affinity for binding to an immunoglobulin target, is also selected or isolated by a nucleic acid encoding it. Once isolated, the gene or genes encoding the modified immunoglobulins can then be increased. This process of isolation and augmentation, referred to as elution, can be repeated, allowing enrichment of the antibody variant in the library. Nucleic acid sequencing of the linked nucleic acid ultimately allows gene identification.

A variety of selection methods are known in the art that can be used in the present invention to search immunoglobulin libraries. These include, but are not limited to, Phage display of peptides and antibodies: a laboratory manual, Kau et al., 1996, Academic Press, San Diego, Calif., 1996; Lowman et al., 1991, Biochemistry 30: 10832 -10838; Smith, 1985, Science 228: 13151317) and its derivatives such as selective phage infection .. (Malmborg et al. 1997, J Mol Biol 273: 544-551), selectively infectious phage (Krebber et al., 1997, J Mol Biol 268: 619-630), and delayed infectious washout (Benhar et al., 2000, J Mol Biol 301: 893

50830 Β

904), showing the cell surface (Witrrup, 2001, Curr Opin Biotechnol, 12: 395-399) such as showing on bacteria (Georgiou et al „1997, Nat Biotechnol 15: 29-34; Georgiou et al„ 1993, Trends Biotechnol 11: 6-10; Lee et al., 2000, Nat Biotechnol 18: 645-648; Jun et al., 1998, Nat Biotechnol 16: 576-80), fungi (Boder & Wittrup, 2000, Metodas Enzymol 328: 430 -44; Boder & Wittrup, 1997, Nat Biotechnol 15: 553557), and mammalian cells (Whitehorn et al., 1995, Bio / technology 13: 1215-1219), as well as in vitro display technologies (Amstutz et al., 2001, Curr Opin Biotechnol 12: 400-405) such as polysome display (Mattheakis et al., 1994, Proc Natl Acad Sci IJSA 91: 9022-9026), ribosome display (Hanes et al., 1997, Proc Natl Acad Sci USA 94: 4937-4942), showing mRNA (Roberts & Szostak, 1997, Proc Natl Acad Sci USA 94: 12297-12302: Nemoto et al „1997, FEBS Lett 414: 405-408), and a system for showing ribosome inactivation (Zhou et al. 2002, J Am Chem Soc 124, 538-543).

Other selection methods that can be used in the present invention include methods that do not rely on demonstration, such as in vivo methods that include. but are not limited to periplasmic expression and cytometric screening (Chen et al., 2001, Nat Biotechnol 19: 537-542), antibody fragment complement analysis (Johnsson & Varshavsky, 1994, Proc Natl Acad Sci USA 91: 1034010344; Pelletier et al. ., 1998, Proc Natl Acad Sci USA 95: 12141-12146), and screening of two fungal hybrids (Fields & Song, 1989, Nature 340: 245-246) used in the selection method (Visintin et al., 1999, Proc Natl Acad Sci USA 96: 11723-11728). In an alternative embodiment, selection is possible by a fusion participant that binds to a specific sequence on the expression vector, and thus connects covalently or non-covalently to the fusion participant and the enhanced Fc variant of the library member with the nucleic acid encoding it. For example, PCT WO 00/22906; PCT WO 01/49058; PCT WO 02/04852; PCT WO 02/04853; PCT WO 02/08023; PCT WO 01/28702; and PCT WO 02/07466 describe such fusion participants and techniques that can be used in the present invention. Alternatively, in vivo selection may occur if antibody expression affects to some extent the growth, reproduction or survival of the cell.

A subset of selection methods referred to as directed evolution methods are those involving knitting or creating favorable sequences during selection; sometimes with the incorporation of new mutations. As the expert will appreciate, the method of directed evolution can facilitate the identification of the most favorable sequences in the library, and it can increase

50830 Β variety of sequences to be searched. Many methods of moderate evolution are known in the art that can be used in the present invention to search for antibody variants, which include, but are not limited to, DNA repositioning (PCT WO 00/42561 AZ; PCT WO 01/70947 AZ), exon repositioning. (U.S. Pat. No. 6,365,377; Kolkman & Steminer, 2001, Nat Biotechnol 19: 423-428), changing family position (Crameri et al., 1998, Nature 391: 288-291; U.S. Pat. 6,376,246), RACHITT.TM. (Coco et al., 2001, Nat Biotechno! 19: 354-359; PCT WO 02/06469), STEP and random priming of in vitro recombination (Zhao et al., 1998, Nat Biotechnol. 16: 258-261; Shao et al. al „1998, Nucleic acids Res 26: 681-683), exonuclease mediation (US Pat. No. 6,352,842; US Pat. No. 6,361,974), Gene Site Saturation Mutagenesis. TM. (U.S. Pat. No. 6,358,709), Gene Reassemblv. TM. (US Pat. br, 6,358,709), SCRATCHY (I.utz et al, 2001, Proc Natl Acad Sci USA 98: 11248-11253), DNA fragmentation method (Kikuchi et al., Gene 236: 159-167), changing the position of single-stranded DNA (Kikuchi et al „2000, Gene 243: 133-137), and the telinology of antibody evolution modification directed at the AMEsystem. TM. (Applied Molecular Evolution), (US Pat. No. 5,824,514; US Pat. No. 5,817,483; US Pat. No. 5,814,476; US Pat. No. 5,763,192: US Pat. 5,723,323), In a preferred embodiment the antibody variants can be precipitated using one or more cell-based assays or in vivo. For such assays, purified or crude modified immunoglobulins are typically added exogenously so that the cells are exposed to individual immunoglobulins or immunoglobulin precipitates belonging to the library. These assays are typically, but not always, based on immunoglobulin function; that is, the ability of an antibody to bind to its target and to mediate a biochemical event, for example effector function, inhibition of ligand / receptor binding, apoptosis, and the like. Such assays often involve positating the response of cells to an antibody, for example cell survival, cell death, altering cell morphology, or the transcriptional action of such cell expression from a natural gene or reporter gene. For example, such assays can measure the ability of an antibody variant to elicit ADCC, ADCP, or CDC. For some assays, it may be necessary to add additional cells or components, i.e., in addition to target cells, for example serum complement, or effector cells such as peripheral blood monocytes (PBMCs), NK cells, macrophages, and the like. Such additional cells can be from any organism, primarily human, from mice, rats, rabbits and monkeys. Immunoglobulins can cause apoptosis of certain cell lines that

50830 Β Express the target, or may mediate an attack on target cells with immune cells added to the assay. Methods for observing cell death or viability are known in the art and include the use of dyes, immunochemicals, cytochemicals and radioactive reagents. For example, caspase staining assays can allow the measurement of apoptosis, and the binding or release of radioactive substances or fluorescent dyes, such as alamar blue, which can allow monitoring of cell growth or activation. DELFIA can be used in the preferred solution. RTM. EuTDA-based cytotoxicity analysis (Perkin Elmer, MA). Alternatively, death or damage to the target cell can be monitored by measuring the release of one or more natural intracellular components, for example lactate dehydrogenase. Transcriptional activation also serves as a method for testing function in cell-based assays. In this case, the reaction can be monitored by examining natural genes or immunoglobulins that can be over-regulated, for example the release of certain interleukins can be measured, or alternatively it can be read via a reporter construct. Cell-based assays may also include the measurement of cell morphological changes in response to the presence of modified immunoglobulins. The cell types for such assays can be prokaryotic or eukaryotic, and it is known in the art that various cell lines can be used. Alternatively, cell-based screening can be performed using cells that have been transformed or transfected with nucleic acids encoding variants. That is, anlite antibody variants are not added exogenously to cells. For example, in one solution, cell-based search is used to show the area of ​​cells. A fusion participant that allows the display of modified immunoglobulins on the cell surface can be used (Witrrup, 2001; Curr Opin Biotechnol, 12: 395-399).

In a preferred embodiment, the immunogenicity of the modified immunoglobulins can be determined experimentally using one or more cell-based assays. In a preferred embodiment, ex vivo T-cell activation assays are used to experimentally quantify immunogenicity. By this method, antigen representing cells and native T cells from paired donors are excited one or more times with the peptide or with the whole antibody concerned. T cell activation can then be demonstrated using a number of methods, for example by monitoring cytokine production or by measuring the consumption of tritiated thymidine. In the most preferred solution, the production of interferon gamma

50830 Β is monitored using Elispot analysis (Schmittel et. Al., 2000, J. Immunol. Meth., 24: 1724).

The biological properties of the modified immunoglobulins of the present invention can be characterized in cell, tissue and whole body assays. As is known in the art, drugs are often tested on animals, which include, but are not limited to, mice, rats, rabbits, dogs, cats, pigs, and monkeys, to measure the efficacy of a drug for treating a disease or disease model, or for measuring pharmacokinetics, toxicity and other properties. Animals can be considered models of disease. Therapists are often tested on mice, including, but not limited to, hairless mice, SCID mice, xenograft mice, and transgenic mice (including crushed and erupted mice). Such analyzes can provide important data for determining the possibility of using antibodies as a therapeutic. Any organism, primarily a mammal, can be used for testing. For example, because of their genetic similarity to humans, monkeys may be suitable therapeutic models, and may therefore be used to test the efficacy, toxicity, pharmacokinetics, or other properties of the modified immunoglobulins of the present invention. Human tests are eventually required for drug approval, and therefore, of course, these experiments are also anticipated. Thus, the modified immunoglobulins of the present invention can be tested in humans to determine their therapeutic efficacy, toxicity, immunogenicity, pharmacokinetics and / or other clinical properties.

The modified immunoglobulins of the present invention can be used in a wide range of antibody products. In one embodiment, an antibody variant of the present invention is used for therapy or prophylaxis, for preparative or analytical use, as a diagnostic, industrial compound, or research reagent, preferably as a therapeutic. An antibody variant can be used in an antibody composition that is monoclonal or polyclonal. In a preferred embodiment, the modified immunoglobulins of the present invention are used to kill target cells carrying the target antigen, for example cancer cells. In an alternative embodiment, the modified immunoglobulins of the present invention are used to block, antagonize or agonize a target antigen, for example antagonizing a cytokine or cytokine receptor. In another preferred embodiment, the modified immunoglobulins of the present invention are used to block, antagonize or agonize a target antigen and to kill target cells carrying the target antigen. In another preferred embodiment,

50830 The modified immunoglobulins of the present invention are used to block, antagonize or agonize growth factor or growth factor receptor and to kill target cells that carry or need a target antigen. In an alternative embodiment, the modified immunoglobulins of the present invention are used to block, antagonize or agonize enzymes and enzyme substrates.

The modified immunoglobulins of the present invention can be used for a variety of therapeutic purposes. In a preferred embodiment, an antibody that binds the modified immunoglobulins is administered to a patient to treat a specific disorder. For the purpose of the present invention, the patient includes both humans and animals, preferably mammals, and preferably humans. By a specific disorder is meant a disorder that can be alleviated by the administration of a pharmaceutical composition comprising a modified immunoglobulin of the present invention.

In one embodiment, the modified immunoglobulin of the present invention is only a therapeutically active agent administered to a patient. Alternatively, the modified immunoglobulin of the present invention is administered in combination with one or more other therapeutic agents, which include, but are not limited to, cytotoxic agents, chemotherapeutic agents, cytokines, growth inhibitors, anti-hormonal agents, kinase inhibitors, anti- angiogenic agents, cardioprotectants or other therapeutic agents. Modified immunoglobulins may be co-administered with one or more other therapeutic regimens. For example, an antibody variant of the present invention may be administered to a patient with chemotherapy, radiation therapy alone, or both, with chemotherapy and radiation therapy. In one solution. the modified immunoglobulins of the present invention may be co-administered with one or more antibodies, which may or may not contain an antibody variant of the present invention. According to another embodiment of the invention, the modified immunoglobulins of the present invention and one or more other anticancer therapies are used to treat ex vivo cell cancer. It is conceivable that such ex vivo treatment may also be used specifically for autologous bone marrow transplantation. It is of course possible to use the antibodies of the invention in combination with other therapeutic techniques, such as surgery.

A variant of other therapeutic agents may be used for administration with the modified immunoglobulins of the present invention. In one embodiment, the modified immunoglobulin is administered with an anti-angiogenic agent, which is a compound that blocks or interferes to some extent with the development of blood vessels. Angio-angiogenic factor can

50830 Β be, for example, a small molecule or protein, for example an antibody, Fc fusion, or cytokine, that binds to a growth factor or to a growth factor receptor involved in the progression of angiogenesis. The preferred anti-angiogenic factor herein is an antibody that binds to vascular endothelial growth factor (VEGF). In an alternative embodiment, the modified immunoglobulin is administered with a therapeutic agent that elicits or enhances an adaptive immune response, for example an antibody targeting CTLA-4. In an alternative embodiment, the modified immunoglobulin is administered with a tyrosine kinase inhibitor, a molecule that inhibits to some extent the action of tyrosine kinase. In an alternative embodiment, the modified immunoglobulins of the present invention are administered with a cytokine. As used herein, the term cytokine refers to a generic term for proteins released by a population of cells that act on other cells as intercellular mediators involving chemokines.

Pharmaceutical compositions are provided in which modified immunoglobulins of the present invention and one or more therapeutically active agents are formulated. The antibody variant formulations of the present invention are prepared for storage by mixing said immunoglobulin having the desired degree of purity with the best pharmaceutically acceptable carriers, excipients or stabilizers (Remington's Pharmaceutical Sciences, 16, edition, Osol, A. Ed., 1980), in the form of lyophilized formulations or aqueous solutions. The formulations used for in vivo administration are preferably sterile. This is easily done by filtration through sterile filtration membranes or other methods. The modified immunoglobulins and other therapeutically active agents described herein may also be formulated as immunoliposomes and / or may be placed in microcapsules.

The administration of a pharmaceutical composition comprising a modified immunoglobulin of the present invention, preferably in the form of a sterile aqueous solution, can be performed in many ways including, but not limited to, oral, subcutaneous, intravenous, intranasal, intraotic, transdermal, topical (e.g. gays, ointments, lotions, creams, etc.), intraperitoneally, intramuscularly, intrapulmonarily (eg AERx 'inhalation technology^M which is commercially available from tm

Aradigm, or pulmonary delivery system commercially available from Inhale Thracpcutics, vaginally, parenterally, rectally, or intraocularly.

As used herein, the term specific binding refers to a coupling reaction by which the ligand in question is determined and recognized in a heterogeneous

50830 Β population of molecules. Thus, under established conditions (e.g., immune condition conditions in the case of immunoglobulins), a specific antibody binds to its specific target and does not bind in significant amounts to other molecules present in the sample. Modified regions of the structural loop of protein units that bind antigen or molecules can be compared with CDR antibodies, but not antigens as such.

The term expression system refers to nucleic acid molecules that contain the desired coding sequence and control sequences in working linkage, so that a host transformed or transfected with these sequences can produce encoded proteins. To encourage transformation, an expression system may be included in the vector; however, the relevant DNA may then also be incorporated into the host chromosome.

According to a preferred embodiment of the present invention, the expression system may include a vector. Any expression vector known in the art can be used for this purpose.

The modified immunoglobulin is primarily expressed in the host, preferably in a bacterial, fungal, plant cell, in an animal cell or in a plant or animal.

A wide variety of suitable host cells can be used to express the modified immunoglobulin, including, but not limited to, mammalian cells (animal cells), plant cells, bacteria (e.g., Bacillus subtilis, Escherichia coli), insect cells, and fungi ( eg Pichia pastoris, Saccharomvces cerevisiae). For example, a variety of cell lines that can be used in the present invention are described in the ATCC cell line catalog, which can be obtained from the American Type Culture Collection. In addition, plants and animals can also be used as hosts for the expression of immunoglobulins according to the present invention. Expression vectors or cassettes as well as transfections can be selected according to the host used.

Of course, cell-free expression systems or cell-free protein can also be used. Expression platforms for in vitro protein transcription / translation that produce sufficient amounts of protein offer many benefits of cell-free protein expression, eliminating the need for labor pre- and post-steps (e.g., host cell transformation, culture, or lysis) typically associated with systems expression on stem cells.

50830 Another aspect of the present invention relates to a method of producing an immunoglobulin or a pharmaceutical composition thereof comprising at least one modification in the structural region of a loop of said immunoglobulin and determining the binding of said immunoglobulin to an antigen epitope, wherein the unmodified immunoglobulin does not bind significantly to said epitope. and the procedure includes the steps of:

- generating a nucleic acid encoding an immunoglobulin containing at least one loop region,

- modification of at least one nucleotide residue of at least one specified loop region,

- transferring said modified nucleic acid to an expression system,

- expression of said modified immunoglobulin,

- contacting the expressed modified immunoglobulin with the epitope,

- determining whether said modified immunogiobulin binds to said epitope, and

- achieving the binding of the modified immunoglobulin to said epitope and, if necessary, its final processing into a pharmaceutical preparation.

In particular, the present invention relates to a method of multi-specific binding of an immunoglobulin, specifically to at least one first molecule or a pharmaceutical composition thereof comprising at least one modification in at least one structural region of a loop of said immunoglobulin and determining specific binding of said at least one loop region to at least another molecule selected from the group consisting of allergens, tumor-associated antigens, self-antigens, enzymes, bacterial antigens, fungal antigens, protozoan antigens and viral antigens, wherein the inunoglobulin containing the unmodified stricture region of the loop does not bind specifically to said at least one other molecule, and the method comprises the steps of:

- generating a nucleic acid encoding a specific binding immunoglobulin for at least one first molecule comprising at least one structural region of the loop,

- modifying at least one nucleotide residue from at least one said region of the loop encoded by said nucleic acid,

- transferring said modified nucleic acid to an expression system,

- expression of said modified immunoglobulin,

50830 Β

- contacting the expressed modified immunoglobulin with said at least one other molecule, and

- determining whether said modified immunoglobulin binds specifically to another molecule, and

- achieving a modified immunoglobulin specific binding to said at least one other molecule, and optionally

- its final processing into a pharmaceutical preparation.

Preference is given to modifying more than one specificity in a specific binding pair member (Kufer et al. (2004) Trends in Biotechnology Vol. 22, pp. 238-244).

Numerous attempts have been made to produce multi-specific, e.g. bispecific, monoclonal antibodies or antibody fragments. One problem in producing a bispecific antibody made up of two different polypeptide chains (heavy and light chains) requires the expression of four different chains (two heavy and two light chains) in one cell, and this results in a number of different combinations of molecules that must be separated from of the desired bispecific molecule in the mixture. Due to their similarity, disassembling these molecules is difficult and expensive. Numerous techniques have been used to minimize the occurrence of such unwanted pairs (Carter (2001) Journal of Imminological Methods, vol. 248, pages 7-15).

One solution to the problem is to produce a single polypeptide chain with two specificities, e.g. two interconnected scFvs iii the production of the so-called diatel. Such molecules have been shown to be very far from the folding of a natural molecule and are extremely difficult to produce (LeGall et al. (2004) Protein F.ngineering, Design & Selection, vol. 17 pages 357-366).

Another problem with the present construction of bispecific antibodies is the fact that even if the parent antibodies are bivalently bound to their respective binding participant (e.g., IgG), the resulting bispecific antibody is monovalent to each respective binding participant.

Preferred multi-specific molecules of the present invention solve these problems.

Expression of a bispecific molecule such as a polypeptide chain is possible (modified Ig domain with two specific linkages, see examples), and is easier to report.

50830 Β than the expression of two antibody polypeptide chains (Cabilly et al. Proc. Natl. Acad. Sci. USA 81: 3273-3277 (1984)).

An antibody-like molecule (i.e., constructed of 2 polypeptide chains) can also be produced, due to the fact that the second specificity is located in a non-variable part of the molecule and therefore requires two different heavy chains or two different light chains. Therefore, there is no possibility of mispairing two chains here.

The antibody of the present invention may consist of a heavy chain and a light chain, which together form a variable binding region for a specific binding participant. Another specificity can be created with a modified loop of any heavy-duty or light-chain structural loop. The binding site may also be formed by more than one non-CDR loop which may be structurally adjacent (on the heavy chain or on the light chain or on both chains).

The modified antibody or derivative thereof may be a complete antibody or antibody fragment (e.g., Fab, CH1-CH2, CH2-CH3).

Binding to binding participants is possible mono- or multivalently or even with different valences to different binding participants, depending on the construction.

There are a number of different loops that can be obtained for the selection and construction of a specific binding site in non-CDR heavy and light chain regions. Antibody derivatives that have even more than two specificities can be constructed without the above problems.

The specific binding domains within a single polypeptide chain may be linked to or without a peptide linker.

Some classes of antibodies can be considered multi-specific, especially bispecific in nature: they bind to an antigen (which is typically a foreign structure or cancer-related structure) with a variable region and bind to Fc-effector molecules with Fc in part (e.g., Fc receptors on various immune cells or complementary protein) thus enabling effects such as ADCC, AI9CP, or CDC.

Fc-effector molecules bind to the Fc portion of the immunoglobulin molecule (for IgG1 it consists of the CH2 and SNZ domains) and a number of methods have been described to optimize effector function by improving the binding of the Fc portion of the antibody molecule using glyco-engineering techniques. (US 6,602,684) either by protein engineering directly to Fc (US 2005/0054832) or indirectly by engineering outside Fc (US 2005/02444403). Both, binding of the Fc region to the Fc receptor and / or binding to

50830 Β Complementary proteins such as Cq1 were altered using these techniques. Typically, the binding affinity for such Fc-effector molecules tends to improve because it correlates with improved effector functions.

With the present invention it is possible to construct the binding of antibodies to Ffector molecules outside the natural Fc binding region. Modified loops in the antibody domain other than the loop involved in binding a natural Fc-effector molecule can be selected from the library or designed to bind to one or more Ffector molecules. An antibody with such an additional Fc-effector molecule binding site will have a stronger preference for a particular Fc-effector molecule or effector cell exhibiting an Fc-effector molecule and may therefore have an even stronger effect than antibodies subjected to glyco-engineering or otherwise enhanced Fc regions. However, for certain embodiments of the present invention, the effector characteristics of the antibody in question to be modified need not be directly altered. they should already be kept out of the scope of the modification in the structural loop according to the present invention.

Antibody fragments have certain advantages over cellular antibodies. Fragments usually have good biological distribution properties and can be more easily produced. However, most antibody fragment constructs lack effector functions and have a short duration in vivo (Holliger P, et al. Nat Biotechnol. (2005) 23: 1126-36.).

Neither the CH1 domain nor the Sk or SH domain mediate effector functions and this is why Fabs do not exhibit ADCC, ADCP or CDC. WO 02/44215 describes the binding of a molecule consisting of an antigen binding site of an antibody and a peptide that binds to the effector molecule. In this way, an antibody fragment that shows effector functions can be constructed. The peptide was incorporated into the binding molecule in a position that does not destroy antigen binding, nor the ability of the peptide to bind to the Fc-effector molecule.

However, according to the present invention, binding to an Fc effector molecule can be performed with modified immunoglobulin domains that have been selected to bind Fc effector molecules from a library of random loop sequences within a fixed skeleton of the immunoglobulin domain. Therefore, it is possible to select specific loop sequences that will not bind to Fc-effector molecules outside the Ig domain skeleton. The polypeptides obtained from the present invention may therefore consist preferably of more than 100 amino acids.

50830 To select the possible effector function of such domains according to the present invention, for binding to Fc receptors and / or complement factors such as C1 q, libraries of CH1, Sk or Ck mutant domains can be selected.

In order to increase the lifetime of in vivo molecules consisting of or containing such domains (e.g. CH1, CH2, SNZ, CH4, Sk or Ck), mutant libraries can be selected for binding to FcRn according to the present invention e.g. . CH1, CH2, SNZ, CH4, Sk or CXdomains.

FcRn receptors for selection can be provided on the surface of cells expressing said natural receptors or by expressing and purifying the extracellular portion of said receptor. For the purpose of the present invention, the first FcRn assay can be selected for dornene mutants that can be further tested in vitro and even further characterized in FACS assays by binding to FcRn receptor-expressing cells. It can be further characterized by ranking the binding affinity to various reconinbinant FcRn, isoform and αJotypes e.g. plasmon surface resonance techniques.

The immunoglobulin of the present invention is of human or murine origin.

Because the modified immunoglobulin can be used for a variety of purposes, particularly in pharmaceutical compositions, the immunogiobulin is primarily of human or murine origin. Of course, the modified immunoglobulin may also be a humanized or chimeric immunoglobulin.

According to another preferred embodiment, the human immunoglobidin is selected from the group consisting of IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4 and IgM.

Mouse immunoglobulin is preferably selected from the group consisting of IgA, IgD, JgE, IgG1, IgG2A, IgG2B, IgG2C, IgG3 and IgM.

The modified immunoglobulin can be obtained from one of the above classes of immunoglobulins.

The lininoglobulin preferably comprises a heavy and / or light chain immunoglobulin or a portion thereof.

The modified immunoglobulin may comprise a heavy and / or light chain, at least one variable and / or a constant domain.

The immunoglobulin of the present invention comprises at least one immunoglobulin constant doinen or a portion thereof that includes a minidomain.

50830 Konstant The constant domain of immunoglobulin fold units is a constant portion of an immutioglobulin molecule, also referred to as a constant region domain (e.g., CH1, CH2, SNZ, CH4, C-cap, C-lambda).

A variable domain is a unit of the immunoglobulin fold of a variable portion of an immunoglobulin molecule, also referred to as a variable region domain (e.g., Vh, Vk, VI, Vd).

The preferred immunoglobulin of the invention consists of a constant domain selected from the group consisting of CH1, CH2, SNZ, CH4, Ig-kappa-C, Ig-lambda-C. or a portion thereof comprising a minidomain, with at least one loop region, and characterized in that said at least one loop region includes at least one amino acid modification that forms at least one modified loop region, wherein said at least one modified loop region binds specifically to at least one antigen epitope.

The immunoglobulin may comprise at least one single domain or a portion thereof, including a minidomain consisting of two immunoglobulin beta-strands linked to a structural loop, to a variable single immunoglobulin domain, or to a single Fv chain.

Another preferred immunoglobulin according to the invention consists of a variable domain of a heavy or any chain, or part thereof including a minidomain, with at least one loop region, characterized in that said at least one loop region comprises at least one amino acid modification forming at least one inodified a loop region, wherein said at least one modified loop region binds specifically to at least one antigen epitope.

According to a preferred embodiment, the constant domain is selected from the group consisting of CH1, CH2, SNZ, CH4, ig-kappa-C, Ig-lambda-C and combinations thereof.

The modified immunoglobulin of the present invention may comprise one or more constant domains (e.g., at least two, three, four, five, six, ten domains). If more than one domain is present in the modified immunogiobulin, these domains may be of the same type or different types (e.g., CH1-CH1-CH2, SNZ-SNZ). Of course, the sequence of the single domain can also be of any kind (e.g. CH1-CH3-CH2, CH4CH1-CH3-CH2).

All amino acid sequence designations of immunoglobulins were performed according to the IMGT numbering scheme (IMGT, international ImmunoGene Tics information

50830 Β system@imgt.cines.fr; http://itngt.cines.fr; Lefranc et al., 1999, Nucleic Acids Res. 27: 209-212; Ruiz et al., 2000Nucieic Acids Res. 28: 219-221; Lefranc et al., 2001, Nucleic Acids Res. 29: 207-209; Lefranc et al., 2003, Nucleic Acids Res. 31: 307-310; Lefranc et al., 2005, Dev Comp Immunol 29: 185-203).

According to another preferred embodiment, the modified regions of the CH1, CH2, SNZ and CH4 strands comprise amino acids 7 to 21, amino acids 25 to 39, amino acids 41 to 81, amino acids 83 to 85, amino acids 89 to 103 and amino acids. acids 106 to 117, where the numbering is according to IMGT.

The loop regions of Ig-kappa-C and Ig-lambda-C of human origin include preferably amino acids 8 to 18, amino acids 27 to 35, amino acids 42 to 78, amino acids 83 to 85, amino acids 92 to 100, amino acids 108 to 117 and amino acids 123 to 126, wherein the numbering is according to IMGT.

The Ig-kappa-C and Ig-lambda-C loop regions of murine origin comprise preferably amino acids 8 to 20, amino acids 26 to 36, amino acids 43 to 79, amino acids 83 to 85, amino acids 90 to 101, amino acids 108 to 116 and amino acids 122 to 125, wherein the numbering is according to IMGT.

Areas of structural loops of the variable domain of immunogiobulins of human origin comprise preferably amino acids 8 to 20, amino acids 44 to 50, amino acids 67 to 76 and amino acids 89 to 101, wherein the numbering is according to 1MGT.

According to a preferred embodiment of the present invention, the regions of the structural loops of the variable domain of murine immunoglobulins comprise amino acids 6 to 20, amino acids 44 to 52, amino acids 67 to 76 and amino acids 92 to 101, wherein the numbering is according to IMGT.

The above amino acid regions of the immunogiobulins in question include the regions of the loops to be modified.

The immunoglobulin according to the invention is primarily of camel origin.

Camel antibodies contain only one heavy chain and have the same antigenic affinity as normal antibodies consisting of light and heavy chains. Therefore, camel antibodies are much smaller than e.g. Human antibodies, which allow them to penetrate the tissue to retrieve the antigen, are not able to make larger proteins. Moreover, comparative simplicity, high affinity and specificity, and the potential to achieve and interact with active sites, represent the advantages of heavy chain camel antibodies over conventional antibodies in the construction, production and application of compounds of clinical value.

50830 Kam Immunoglobulin of camel origin contains preferably at least one constant domain selected from the group consisting of CH1, CH2 and SNZ.

The loop regions of CH1, CH2 and SNZ of camel immunoglobulin contain amino acids 8 to 20, amino acids 24 to 39, amino acids 42 to 78, amino acids 82 to 85, amino acids 91 to 103 and amino acids 108 to 117, wherein the numbering is according to IMGT.

According to a preferred embodiment of the present invention, the specific binding of the modified immunoglobulin to the molecule is determined by a binding assay selected from the group consisting of immunoassays, primarily enzyme-linked immunosorbent assays (ELISA), plasmon surface resonance assays, nuclear spectroscopy magnetic resonance spectroscopy nuclear magnetic resonance NOE (trNOE), competition analysis, tissue binding analysis, living cell binding assays and cell extract assays.

Binding assays can be performed using a variety of methods known in the art that include, but are not limited to, assays based on FRET (Fluorescence Resonance Energy Transfer) and BRET (Bioluminescence Resonance Energy Transfer), AlfaScreen.TM . (Amplified Luminescent Proximity Homogeneous Assay), Scintillation Proximity Assay, ELISA (Enzyme-Linked Immunosorbent Assay), SPR (Surface Plasmon Resonance, also known as BIACORE), isothermal titration calorimetry, differential scanning calorimetry, and electrophoretic electrophoresis and gel filtration. These and other methods may have an advantage for some fusion or labeling participants.

The modified immunoglobulin is preferably conjugated to a labeling agent selected from the group consisting of organic molecules, labeling enzymes, radioactive labeling agents, colored labeling agents, fluorescent labeling agents, chromogenic labeling agents, luminescent labeling agents, hap , digoxigenin, biotin, metal complexes, metals, colloidal gold and mixtures thereof.

The modified immunoglobulin can be conjugated to other molecules that allow for easy detection of said conjugate, for example, in binding assays (e.g., ELISA) and binding studies.

A preferred embodiment of the present invention relates to a process for the production of immunogiobulin with increased half-life comprising the steps of:

50830 Β

- constructing a modified immunoglobulin by the method of the present invention, including a mutant domain selected from the group consisting of CH1, CH2, SNZ, CH4, C-cap and C-lambda, and having specificity for self-antigen binding through modification in the structural loop region, i

- production of a preparation that has an increased half-life.

A preferred embodiment of the present invention relates to a method according to the present invention, according to which the modified immunoglobulin has FcRn binding specificity.

A preferred embodiment of the present invention relates to a method of producing a constructed immunoglobulin with effector activity, comprising the steps of:

- constructing a modified immunoglobulin by the method of the present invention, comprising a mutant domain selected from the group consisting of CH1, CH2, SNZ, CH4, C-cap and C-lambda, and having the specificity of binding effector molecules through modification in the structural loop region, and

- production of preparations with effector action, wherein said fragment of the engineered antibody is the best Fab or Fc fragment.

The immunoglobulin of the present invention may consist of a constant domain selected from the group consisting of CH1, CH2, SNZ, CH4, Igk-C, Igl-C, or a portion thereof that includes a minidomain, or combinations thereof, with at least one region of the loop. , wherein said at least one loop region comprises at least one amino acid modification forming at least one modified loop region, wherein said at least one modified region of the loop binds specifically to at least one antigen epitope.

It is preferred to molecularly combine at least one modified antibody domain (= binding to a specific participant via non-variable sequences or a structural loop) with at least one other binding molecule which may be an antibody, an antibody fragment, a soluble receptor. ligand or other modified antibody domain.

The molecule is selected from the group consisting of protein molecules, nucleic acids and hydrocarbons.

The loop regions of the modified immunoglobulins can be specifically bound to any type of binding molecule, in particular to protein molecules, proteins, peptides,

50830 Β polypeptides, nucleic acids, glycans, hydrocarbons, lipids, small organic molecules, inorganic molecules. Of course, the modified immunoglobulins may contain at least two loop regions wherein each loop region may bind specifically to other molecules or epitopes.

According to a preferred embodiment of the present invention, the molecule that binds to the modified structural region of the loop is selected from the group consisting of tumor-associated antigens, in particular EpCAM, tumor-associated glycoprotein-72 (TAG-72), tumor-associated CA 125 antigen. , prostate membrane specific antigen (PSMA), high molecular weight melanoma antigen (HMW-MAA), tumor-associated antigen expressing hydrocarbons related to Lewis Y, embryonic carcinoma antigen (CEA), CEACAM5, HMFG PEM, mucin MUC1, MUCI8 and cytokeratin tumor-associated antigen, bacterial antigens, viral antigens, allergens, fluorescein, lysozyme, toll-like receptor 9, erythropoietin, CD2, CD3, CD3E, CD4, CD11, CD11a, CD14 , CD18, CD19, CD20, CD22, CD23, CD25, CD28, CD29, CD30, CD33 (protein p67), CD38, CD40, CD40L, CD52, CD54, CD56, CD80, CD147, GD3, IL-1, IL-IR , IL-2, IL-2R, IL-4, IL-5, 1L-6, IL-6R, IL-8, IL-12,1L-15, IL-18, IL23, interferon alpha, interferon beta, interferon gamma; TNF-alpha, TNFbeta2, TNF.alpha, TNFalphabeta, TNF-R1, TNF-RII, FasL, CD27L, CD30L, 4-1BBL, TRAIL, RANKL, TWEAK, APRIL, BAFF, LIGHT, VEGl, OX40L-1 TRAIL , A1 adenosine receptor, lymphotoxin beta receptor, TACI, BAFF-R, EPO; LFA-3, ICAM-1, ICAM-3, integrin betal, integrin beta2, integrin alpha4 / beta7, integrin alpha2, integrin alpha3, integrin alpha4, integrin alpha5, integrin alpha6, integrin alphaV, alphaVbeta3 integrin, FGFR-3, growth factor keratinocytes, VLA-1, VLA-4, L-selectin, anti-Id, E-sectin, HLA, HLA-DR, CTLA-4, T cell receptor, Β7-1, B7-2, VNRintegrin, TGFbetal, TGFbeta2, eotaxin, BLyS (B-imfocyte stimulator), complement C5, IgE, factor VII, CD64, CBL, NCA 90, EGFR (ErbB-1), Her2 / neu (ErbB-2), NegZ (ErbB-3), Her4 (ErbB4), tissue factor, VEGF, VEG-FR, endothelin receptor, VLA-4, hydrocarbons such as blood group antigens and related hydrocarbons, Galli-glycosylation, gastrin, gastrin receptors, with tumor associated hydrocarbons, hapten NP-cap or ΝΙΡ-sar, alpha / beta T cell receptor, eselectin, digoxin, placental alkaline phosphatase (PLAP) and testicular PLAP-like alkaline phosphatase, transferrin receptor. heparanase I, fed cardiac myosin, glycoprotein Ilb / Illa (GPIIb / lIla), human cytomegalovirus (HCMV) gH envelope glycoprotein, HIV gp 120, EICMV, respiratory syncytial virus RSV F, RSVF Fgp,

50830 Β

VNRintegrin, Hep V gpl20, CMV, gpllbllla, HIV SV gpl20 V3 loop, respiratory syncytial virus (R.SV) Fgp, herpes simplex virus (HSV) gD glycoprotein, HSV gB glycoprotein, HCMV gB glycoprotein envelope fragment and perfrostring .

The modified immunoglobulin of the present invention can bind primarily to one of the molecules described above. These molecules also include allergens.

According to another preferred embodiment of the present invention, the amino acid residues in the SNZ at positions 15 to 17, 29 to 34, 85.4 to 85.3, 92 to 94, 97 to 98 and / or 108 to 110 are modified, wherein is numbering according to IMGT.

The immunoglobulin modification of the present invention is preferably deletion, replacement or insertion.

According to the present invention at least 1, preferably at least 2, 3, 4, 5, 6, 7, 8. 9, 10 and 15, the amino acids are deleted, replaced with other amino acids (also with modified amino acids) or inserted into the area of ​​the immunoglobulin layer. However, the maximum number of amino acids inserted in the region of the immunoglobulin loop should not exceed 30, preferably 25, more preferably 20 amino acids. The replacement and insertion of amino acids is performed primarily by a random method known in the art and as described in the present patent application.

According to a specific embodiment, the immunoglobulin according to the invention is characterized in that the SNZ region comprises SEQ ID NO. 16 or SEQ 1D no. 18, if EpCam binds to said immunoglobulin, SEQ ID NO. 20, if fluorescein binds to said immunoglobulin, SEQ ID NO. 22, 24, 26, 28, 30 or 32, when lysozyme binds to said immunoglobulin, SEQ ID NO. 34, 36, 38 or 40, when TLR9 binds to said immunoglobulin, and SEQ ID NO. 42, when lysozyme and / or erythropoietin bind to said immunoglobulin, According to a specific embodiment of the invention, the immunoglobulin is characterized in that it comprises SEQ ID NO. 44 or SEQ ID NO. 46, when lysozyme and gp41 bind to said immunoglobulin.

The modified immunoglobulin is preferably conjugated to a labeling agent or to a reporter molecule selected from the group consisting of organic molecules, enzymatic labeling agents, radioactive labeling agents, colored labeling agents, fluorescent labeling agents, chromogenic agents.

50830 Β for coating, luminescent markers, haptens, digoxigenin, biotin, metal complexes, metals, coyoid gold and mixtures thereof.

Modified immunoglobulins conjugated to the labeling agent as indicated above can be used, for example, in diagnostic procedures.

The use of an immunoglobulin obtainable by the method of the present invention is used to prepare a vaccine for active immunization. In doing so, the immunoglobulin is used as an antigenic drug substance for formulating vaccines or is used to attract or capture antigenic structures used in the formulation of vaccines.

The immunoglobulin obtainable by the method of the present invention is also used to prepare an immunoglobulin protein library.

The immunoglobulins of the present invention can be used to isolate specific molecules from a sample. If multispecific immunoglobulins are used more than one, the molecules can be isolated from the sample. The use of modified immunoglobulins in such methods is particularly preferred because they allow, e.g. generating matrices with a homogeneous surface and with a certain amount of binding participants (i.e. modified immunoglobulins) that are thus immobilized, allowing binding to the molecules to be isolated. Otherwise, if monospecific binding participants are used, a homogeneous matrix cannot be produced, because single binding participants do not bind with the same efficiency to the matrix.

This can be achieved by a method for the specific binding and / or detection of molecules comprising the steps of:

(a) contacting a modified iminoglobulin according to the present invention or a modified immunoglobulin obtainable by the method of the present invention with a test sample which is expected to contain said molecule, and (b) detecting the possible formation of a specific complex of immunoglobulites and molecules, or by a method for specific isolation of molecules, comprising the steps of:

(a) contacting a modified immunoglobulin according to the present invention or a modified immunoglobulin obtainable by the method of the present invention with a sample containing said molecule, (b) separating the resulting complex of a specific immunoglobulin and molecule, and (c) optionally, isolating the molecule from said complex.

The modified immunogiobulins of the present invention can be used to deliver at least one compound linked to CDRs and / or a modified target loop region. Such immunoglobulins can be used to target therapeutic substances to sites desirable for action in the treatment of disease.

This can be accomplished by a method of targeting a compound to a target, comprising the steps of:

(a) contacting a modified immunoglobulin according to the present invention or a modified immunogiobulin obtained by the method of the present invention that can be specifically bound to said compound, (b) delivering the immunoglobulin / compound composition to the target.

The immunoglobulin of the present invention or obtainable by the method of the present invention may be part of a protein library.

Preferred methods for constructing said library can be found above and in the examples. The library of the present invention can be used to identify the binding of an immunoglobulin to a particular molecule.

The existing immunoglobulin can be modified to introduce an antigen binding site into any domain or minidornene using a protein library of said domain of at least 10, preferably 100, more preferably 1000, and even more preferably 10,000, and more preferably 100,000, preferably more than 1,000,000 variants. a domain with at least one modified loop. The library is then searched for binding to a specific antigen. After molecular characterization with respect to the desired property, the selected domain or minidomain is cloned into the original immunoglobulin genetic engineering techniques so that it replaces the wild-type region. Alternatively, to obtain an immunoglobulin with an additional binding site for a specific antigen, only the DNA encoding the loop or encoding the mutated amino acids can be altered. the purpose of the additional binding site. If, for example, the parent molecule is a complete immunoglobulin to which an additional antigen binding site needs to be inserted without affecting effector function, the loops to be modified should be chosen between domains away from CH2 and SNZ that are natural participants in binding to Fc-factor molecules. If the original immunoglobulin is Fab, modification

50830 Β Loops are possible in the constant domain of light chains or heavy chains or in the respective variable domain. Libraries derived from mutants of parent molecules with mutations in one or more structural loops of one or more domains can be prepared to create libraries. Selection with fully mutated parent molecules may have advantages, as the choice to bind antigen to a modified structural loop will yield spatially favorable modifications if also examined for other properties that the mutated immunoglobulin should exhibit.

The required size (i.e., number of protein variants) of the protein library of the mutated domain or minidomain or domain fusion molecule depends on the task. In general, the library for the new creation of the antigen binding site must be larger than the library used for further modification of the already constructed antigen binding site made of the modified structural loop (e.g. for enhancing affinity or altering fine antigen specificity), the immunoglobulin of the invention may be contained in an immunoglobulin library or a nucleic acid library comprising a plurality of immunoglobulins, e.g. a constant domain, a minidomain, and / or at least one structural region of the loop contained in the minidomain, or a nucleic acid molecule encoding it. The library contains members with different modifications, the plurality being defined with modifications in at least one structural region of the loop. The nucleic acid of the library includes primarily at least 10 different members (resulting in a single amino acid change) and even better includes at least 100, even better 1000 or 10000 different members (e.g., constructed using randomization strategies or combinatorics techniques). Preference is also given to an even greater variety of individual member numbers, such as at least 1,000,000 or at least 10,000,000.

Two different domains or minidomains selected from at least two libraries according to the invention can be used in combination to generate multispecific immunoglobulins. These selected specific immunoglobulins can be combined with each other and with other molecules, similar to building blocks, to construct an optimal arrangement of domains or minidomains to obtain the desired properties.

In addition, one or more modified immunoglobulins according to the invention can be introduced at different or all different sites of the protein, preferably without destroying the structure of the protein. With such domain drag techniques, new libraries are created that can be re-selected for the desired properties.

50830 The preferred library may comprise immunoglobulins according to the invention selected from the group consisting of immunoglobulin domains, minidomains or derivatives thereof.

A preferred embodiment of the present invention is the binding of an antigen molecule (antigen binding molecule) comprising at least one immunoglobulin domain and a structural region of a loop that has been modified according to the present invention for antigen binding, wherein said molecule binding does not include an antibody variable domain . It may contain other moieties that can be used for antibody activities (e.g. such as natural or modified effector regions (sequences); however, it lacks the natural binding site of the antibody, i. variable domains in the position of its natural origin, These antigen binding molecules according to the present invention have the advantages described above for the above molecules, but without the specific antibody binding activity, but with the newly introduced specific binding activity in the structural region of the loop.

Preferably, these antigen binding molecules of the present invention include CHI, CH2, SNZ, CH4, Ig-kappa-C, Ig-lambda-C, and combinations thereof; said combinations include at least two, preferably at least four, in particular at least six constant domains and at least one structural region of the fifth modified according to the present invention. Preferably, these loop loop regions are connected via a loop loop region that is modified according to the present invention or via a loop loop that is naturally present between such two constant domains. The solution of these antigen binding molecules of the present invention consists of the Fc region of an antibody with at least one modification in the structural loop of the present invention. It is also advantageous for the antigen binding molecules of the present invention to introduce new antigen binding sites into the structural loop by randomization techniques, i. by modifying one or more amino acid residues of the loop using randomization techniques or introducing randomly generated inserts into such structural loops. Alternatively, the use of cominous approaches is preferred.

The modified immunoglobulin may have an antigen binding site that is foreign to the unmodified immunoglobulin and that is incorporated into one or more structural loops. The term foreign means that the antigen binding site is not formed naturally by a specific immunoglobulin region, and a foreign binding participant, but the unnatural immunoglobulin binding participant is bound by the antigen binding site. This means

50830 Β that a binding participant, such as an Fc receptor or an immune system effector, is not considered to be bound to an antigen binding site that is foreign to unmodified immunoglobulin.

Preferably, the antigen is selected from the group consisting of a pathogenic antigen, a tumor-associated antigen, an enzyme, a substrate, a self-antigen, organic molecules or allergens. More preferably, the antigens are selected from the group consisting of viral antigens, bacterial antigens, or antigens from eukaryotic or phage pathogens. Preference is given to viral antigens which include HAV-, HBV-, HCV-, HIV i-, HIV II-, parvovirus-, influenza-, HSV-, hepatitis viruses, flaviviruses, Westnile virus, Ebola virus, Roh virus, small roh virus , Measles Virus, herpes virus, adenovirus, papilloma virus, polio virus, parvovirus, rhino virus, Coxsackie virus, polio virus, Echo virus, Japanese encephalitis virus, Dengue Virus, Tick Borne encephalitis virus, yellow fever virus, corona virus, respiratory syncytial virus, parainfluenza virus, La Crosse Virus, Lassa Virus, Rabies Viruses, rotavirus antigens; primarily bacterial antigens include antigens of pseudomonas, mycobacteria, staphylococci, salmonella, meningococci, borrelia, listeria, Neisseria, clostridia, Escherichia coli, legionella, bacilli, lactobacilli, streptococci, enterococci, Sogupe bacteria, Brucella, Carcinoma Francisella, Hclicobacter, Haemophilus, Klebsiella, Shigclla, Yersinia, Vibrio, Chlamydia, Leptospira. Rickettsia, mycobacteria, Treponema, Bartonella. Preference is given to eukaryotic antigens of pathogenic eukaryotes that include antigens Giardia, Toxoplasma, Cyclospora, Cryptosporidium, Trichinella, Yeasts, Candida, Aspergillus, Cryptococcus, Blastomyces, histoplasmas, coccidioids.

Preferred immunoglobulins of the present invention include at least two antigen binding sites, a first binding site for the first epitope, and a second binding site for the second epitope.

According to a preferred embodiment, the proposed immunoglobulin comprises at least two loop regions, a first region of the binding loop to the first epitope, and a second region of the binding loop to the second epilope. Barein the first or at least the second area of ​​the loop or both may be located in the structural loop. The immunoglobulins of the present invention include fragments thereof that are known in the art to be functional and to contain essential elements of the present invention: a structural region of a loop modified according to the present invention.

50830 Preferably, the immunoglobulin of the present invention is composed of at least two immunoglobulin domains, or portions thereof, that include minidomains, each domain containing at least one antigen binding site.

Also preferred is an immunoglobulin according to the invention, comprising at least one constant region domain or a portion thereof comprising a minidomain. Thus, a variable domain, which is for example modified in the C-terminal region, or a variable domain linked to a modified CH1 region, for example a modified CH1 minidomain, is one of the preferred solutions.

The preferred immunoglobulin of the invention comprises a domain having at least 50% homology to the unmodified domain.

The term homology means that the polypeptides have the same or preserved residues in an appropriate position in their primary, secondary or tertiary structure. The term also extends to two or more nucleotide sequences encoding homologous polypeptides. Homologous immunoglobulin domain means the immunoglobulin domain of the invention having at least about 50% amino acid sequence identity with respect to the innate sequence of the full-length sequence of the immunoglobulin domain or any other fragment of the full-length sequence of the immunoglobulin domain as described herein. Preferably, the homologous immunoglobulin domain will have at least about 50% amino acid sequence identity, preferably at least about 55% amino acid sequence identity, more preferably at least about 60% amino acid sequence identity. more preferably at least about 65% amino acid sequence identity, more preferably at least about 70% amino acid sequence identity, more preferably at least about 75% amino acid sequence identity, more preferably at least about 80% amino acid sequence identity, even better at least about 85% amino acid sequence identity, more preferably at least about 90% amino acid sequence identity, more preferably at least about 95% identity of the amino acid sequence to the innate immunoglobulin domain sequence, or any other specifically defined fragment of the full-length immunoglobulin domain sequence as described herein.

The percentage (%) of amino acid sequence identity with respect to the immunoglobulin domain sequences listed herein is defined as the percentage of amino acid residues in the candidate sequence that are identical to the amino acid residues in the specific immunoglobulin domain sequence, after sequence alignment.

50830 Β and inserting gaps, if necessary, to achieve the maximum percentage of sequence identity, without taking into account any conservative substitutions as part of the sequence identity. Alignment for the purpose of determining the percentage of amino acid sequence identity can be achieved in various ways known in the art, for example, by using publicly available computer programs such as BLAST, BLAST-2, ALIGN or Megalign (DNKSTAR). The person skilled in the art can determine the appropriate alignment measurement parameters, which include any algorithm needed to achieve maximum alignment along the entire length of the sequences to be compared.

Percentage values ​​for amino acid sequence identity can be obtained as described below using the WU-BLAST-2 computer program (Altschnl et al., Methods in Enzymology 266: 460-480 (1996)). Most WU-BLAST-2 search parameters are set as default values. Those that are not set as default values, ie. the parameters that can be set are set with the following values: overlap measure = 1, fraction overlap = 0.125, word threshold (T) = l 1, and addition matrix = BLOSUM62. When WGBLAST-2 is used,% amino acid sequence identity values ​​are determined by dividing (a) the number of pairs of identical amino acid residues between the amino acid sequence of the immunoglobulin domain in question having the sequence derived from the innate immunoglobulin domain and comparing the amino acid sequence in question. sequence comparing the respective immunoglobulin domain which may be the unmodified immunoglobulin domain) as determined by WU-BLAST-2 s (b) the total number of amino acid residues of the non-randomized portions of the immunoglobulin domain concerned. For example, in a statement, a polypeptide comprising the amino acid sequence A having or having at least 80% identity of the amino acid sequence to amino acid sequence B, amino acid sequence A is compared to said amino acid sequence and the amino acid sequence of immunogenic sequence to B is.

In a preferred embodiment of the present invention the immunoglobulin is a bispecific antibody or a single chain bispecific antibody. It is further preferred that the immunoglobulin comprises a bispecific domain or portion thereof including a minidomain.

The immunoglobulin of the present invention can be used for any purpose known in the art for immunoglobulins, but applications that depend on the combinations of specificities introduced with the present invention are also possible. In this regard,

50830 Β immunoglobulins according to the present invention are used primarily for therapeutic and prophylactic use (eg as active or passive immunotherapy); for preparative and analytical use and for diagnostic use.

The immunoglobulins of the present invention may be used in a set of bound participants comprising:

(a) a modified immunoglobulin having an antigen binding site of a foreign immunoglobulin incorporated into one or more structural loops, and (b) a binding molecule comprising an epitope of said antigen.

Such a binding molecule of this kit according to the present invention can be used to identify the binding specificity of the modified immunoglobulin according to the present invention. Using the binding molecule of this transfer kit of the present invention, the potential of the modified immunoglobulin according to the present invention can be determined.

The potential, as defined herein, is the binding property of a modified molecule to its antigen. Binding can be determined quantitatively and / or qualitatively in terms of specificity and / or affinity and / or suitability for use for quality control.

In addition, a binding molecule from a kit according to the present invention can be used to select a modified immunoglobulin according to the present invention from a library containing at least 10, preferably at least 100, more preferably at least 1000, more preferably at least 10000, especially at least 100000 immunoglobulins with different modifications in structural loops.

In accordance with the present invention, one key feature of the present invention is that the modification of the immunoglobulin domain takes place in regions not normally involved in antigen binding, in other words, in regions other than antibody CDRs. It has been observed that a specific set of immunoglobulin domains allows the introduction of random mutations into regions that are structurally analogous to CDRs but differ in sequence position. The regions of the present invention are, as CDRs, the regions of the loops connecting the beta bands of the immunoglobulin fold.

More specifically, it is described herein that by introducing random mutations into the loops connecting the beta strands of the AB and EF human IgG1 SNZ domains, mutated SNZ domains that bind specifically to the Toll-like 9-peptide receptor (TLR-9) are selected. or on „hen

50830 Β egg “lysozyme, which is a peptide and protein that is not normally recognized and binds to human SNZ doin in IgG1. Mutations that we have introduced include mutations in which selected amino acid residues in the wild-type sequence are replaced with randomly selected residues, and they also involve the insertion of specific amino acid residues into the above loops.

By analogy, immunoglobulin domains from any class of immunogiobulin and from immunoglobulins from any species are available for this type of modification. In addition, not only the specific loop targeted in the present invention can be manipulated. it can be manipulated in the same way with any loop that connects beta threads in the immunoglobulin domains.

Immunoglobulin domains after modification from any organism and from any class of immunoglobulins can be used according to the present invention as such (as single domains), or as part of a larger molecule. For example, they may be part of an intact immunoglobulin, which in this connection may have its normal antigen binding region formed from 6 CDRs and a new, modified antigen binding region. Similarly, a multi-specific, e.g. bispecific immunoglobulin. The immunoglobulin domain after modification may also be part of any fusion protein. The use of these structurally altered immunoglobulin domains generally falls within the scope of immunoglobulin use.

As immunoglobulin domains, the following immunoglobulin domains are meant herein:

for IgG, IgD and IgA: VL, CL, VH, CH1, CH2, SNZ for IgM and IgE: VL, CL, VH, CH1, CH2, SNZ, SI4

1. Single domains of immunoglobulins randomized on one side, i. in a loop connecting the beta threads BC, DE or FG (top, with the exception of the variable domains covered by many patents) or the beta threads L-V, CD, (C-C 'and CD in the case of variable domains) or EF (bottom). A single loop or any combination of loops can be randomized. Residues can be replaced, erased, or additional residues can be inserted.

2. Single immunoglobulin domains randomized on both sides, top and bottom.

50830 Β

3. Any protein containing one of the single randomized domains, such as:

a) SIZ (scCH3), scCH2, scCH1 / CL single chain dimers randomized on one or both sides,

b) a single chain Fv randomized at the bottom, ie. on the side opposite the CDRs,

c) Fab fragments randomized at the bottom, i. at the C-terminal end of the CH1 domain and the CL domain,

d) Fc fragments (ie proteins consisting of CH2-CH3) randomized on one or both sides,

e) complete immunoglobulins randomized at the bottom of Fc,

f) other suitable domains.

The main advantages of single domains are: they are similar for all reasons used to highlight camel VH molecules (nanobodies, see www.ablynx.com). Randomized immunoglobulin domains are very small proteins (molecular weights approx. 12-15 kDa, depending on the number of amino acid residues inserted) and will therefore have the following advantages over conventional antibodies or antibody fragments such as scFvs and Fabs: recognition of unusual or hidden epitopes, binding in nests or active sites of protein targets , light production and many others. In the case of an immunoglobulin domain that is randomized on both sides, a bivalent or bispecific molecule can be obtained. The main advantages of single domains as part of a fusion protein are the additional binding properties that can be obtained by modification on any other protein.

It was thought that any expression system could be used to produce the protein. Analogous to single domains, as described herein, camel antibodies can be found, which have only VH but not VL. In these proteins, only 3 CDRs are responsible for antigen binding (instead of 6 as in a normal antibody responsible for antigen binding).

The following patent documents are incorporated herein by reference in their entirety as follows:

50830 Β

US 6,294,654 A modified immunoglobulin molecule that incorporates an antigen into the non-CDR region of a loop,

US 5,844,094 Targeted binding of polypeptides,

U.S. Pat. No. 5,395,750 A process for the production of proteins that bind to predetermined antigens,

US 2004/0071690 Polyvalent and polyspecific high avidity reagents,

US 2004/0018508 Surrogate antibodies and methods for their preparation and use,

US 2003/0157091 Multi-functional proteins,

US 2003/0148372 A method for searching libraries showing phage with different ligands,

US 2002/0103345 Antigen-binding proteins similar to bispecific immunoglobulin and method for their production,

US 2004/0097711 Immunoglobulin superfamily of proteins,

US 2004/0082508 Excreted proteins,

US 2004/0063924 Excreted proteins,

US 2004/0043424 Immunoglobulin superfamily of proteins,

US 5,892,019 Production of immunoglobulin encoded by a single gene,

US 5,844,094 Targeted binding polypeptide.

The present invention will now be illustrated in the following figures and a non-limiting invention.

Figure 1a shows the structure of intact IgGI. Domains are marked with arrows.

Figure 1b shows the structural organization of the major isotypes of human immunoglobulin monomers. Disulfide bonds are shown as lines, N-linked hydrocarbon groups are shown as circles.

Figure 2 shows the immunoglobulin fold for the constant (left) and variable (right) domains of the immunoglobulin. Beta threads are marked with arrows.

Figure 3 shows a molecular model of a structurally modified SNZ domain according to the present invention, with a randomized portion marked with a solvent-accessible surface.

Figure 4 shows schematically the PCRs used to produce the fragments used to stack the mutated SNZ domain. PCR primers are denoted by arrows with their 5'-3 'orientation, and vertical lines indicate

50830 Β Approximate positions of the introduced restriction sites used to stack the mutated gene. The following restriction sites are found on primers for ligation of PCR fragments: CH3LNCO: Ncol; CH3LSAC and CH3CSAC: Sacl; CH3CHIN and CH3RHIN: HindIII; CH3RN0T: Notl.

Figure 5 shows some examples of how the immunoglobulin domains of this application can be used. Randomized areas are marked with a star symbol. The specifics of randomized regions in a single molecule may be the same or different.

Figure 6 shows a schematic construction of a bispecifically modified SNZ domain. The names of the primers are given in the fields and the arrows indicate the direction in which the primers extend. The slash fields indicate the relative positions of the regions randomized in this construct, the fields with vertical lines indicate the relative positions of the regions that were introduced to create clone C24, and the restriction sites used for the cloning process are shown.

Figure 7 shows schematically the construction of a bispecifically modified SNZ domain. The nucleotide sequence and its translation are presented based on the construction of a bispecifically modified SNZ domain. Red sequences indicate randomized regions to form a bispecific construct, while green fields indicate areas in which the sequence was randomized to form a C24 clone.

Figure 8 shows a list of sequences described herein.

DESCRIPTION OF SPECIFIC EXAMPLES;

Example 1: Construction of the SNZ library and showing the surface of the phage The crystal structure of the IgG1 Fc fragment, published in the Brookhaven Database as input 10Q0.pdb, was used to aid in the construction of the mutated SNZ domain.

The sequence used as the basis for the construction of the SNZ library is given in SEQ ID NO. 1. In this sequence, the first amino acid corresponds to proline 343 from the Brookhaven database entry loqo.pdb chain. The last residue found in loqo.pdb is serine 102 from SEQ ID NO. 1. After a detailed analysis of the structure of the loqo.pdb and visual control

50830 Β residues forming a loop connecting beta threads, it was decided to randomize residues 17, 18 and 19, which are part of the loop connecting beta strip AB, as well as 71, 72, 73, 76, and 77, which are part of the loop which connects the beta threads EF in SEQ ID no. 1. A molecular model of a modified SNZ domain, with a randomized portion shown by the surface available to the solvent is shown in Figure 3. The modified gene was produced by a series of PCR reactions, followed by ligation of the resulting PCR products. To facilitate ligation, some nucleotide sequence codons encoding SEQ ID NO. 1 have been modified to produce restriction sites without altering amino acid sequences (silent mutations). For insertion into the cloning vector pHEN1 (Nucleic Acids Res. 1991 Aug 11; 19 (15): 4133-7. Multi-subunit proteins on the surface of filamentous phage: methodologies fbr displaying antibodv (Fab) heavy and light chains. Hoogenboom HR, Griffiths AD, Johnson KS, Chiswell DJ, Hudson P, Winter G.) in a frame with a pelB secretion signal, extra nucleotide residues encoding Met-Ala were linked at the 5 'end of the sequence to form the Ncol restriction site. For randomized residues, the NNS codon (SRAS code, where S means C or G) is selected that encodes all 20 naturally occurring amino acids but avoids 2 out of 3 stop codons. The modified sequence is given as the nucleotide sequence in SEQ ID NO. 2 and as the amino acid sequence in SEQ 1D no. 3. Letter X in SEQ No. 1 3 denotes randomized amino acid residues. The PCR primer sequences used to match the mutated SNZ domain are given in SEQ ID NO. 4 to 9. Figure 4 shows schematically the PCR fragments created to match the mutated genes, and the primers used for this purpose.

Human monoclonal antibody heavy chain 3D6 cDNA (Felgenhauer M, Kohl J, Ruker F. Nucleotide sequences of cD-NAs encoding V-regions of H- and L-chains of human monoclonal antibody specific to HIV-1-gp41. Nucleic Acids Res. 1990 Aug 25:18 (16): 4927) was used as a template for PCR reactions. 3 PCR products were digested with Sacl and / or HindIII and interconnected. The coupling product was further digested with Ncol and NotI and coupled to the surface-showing phagemid vector pFIENI, which had previously been digested with Ncol and NotI. The number of selected clones was controlled by restriction analysis and DNA sequencing and found to contain the planned insert, including properly inserted randomized sequences. For the next steps in obtaining phage, standard protocols followed. Briefly, the binding mixture was transferred to E. coli TG1 cells by electroporation. Phage particles were then rescued from

50830 Β cell Ε. coli TGl with auxiliary phage Μ13-ΚΟ7. Phage particles were then precipitated from the culture supernatant with PEG / NaCl in 2 steps, dissolved in water and used for washing selection or, alternatively, they were placed at minus 80 ° C.

Example 2: Construction of SNZ + Z library [0202] This library was constructed and cloned in the same way as SNZ library. The amino acid sequence of the construct is given in SEQ 1D no. 10, which corresponds to the nucleotide sequence in SEQ ID NO. 11, and the primers used for the construction were SEQ ID NO. 4-7, SEQ 1D no. 9 and SEQ ID no. 12.

Example 3: Construction of SNZ + 5 library [0203] This library was constructed and cloned in the same way as SNZ library. The amino acid sequence of the construct is given in SEQ ID NO. 13, which corresponds to the nucleotide sequence in SEQ 1D no. 14, and the primers used for the construction were SEQ ID NO. 4-7, SEQ 1D no. 9 and SEQ 1D no. 15.

Example 4: Flushing of the SNZ - phage library on TLR-9 peptide [0204] 3 wash cycles were performed according to standard protocols. In short, the following method was applied. 96-well Maxisorp plates (Nunc) were coated with a synthetic peptide that is part of a Tollu-like receptor 9 sequence (TLR-9). 200 μΙ of sitting solution was added to each well: 0.1 M Na-carbonate buffer, pH 9.6, with the following concentrations of dissolved peptide:

1. wash cycle: 1 mg / ml TLR-9 peptide,

2. wash cycle: 500 pg / ml TLR-9 peptide,

3. wash cycle: 100 pg / ml TLR-9 peptide.

Incubation was 1 hour at 37 ° C, followed by bioquishing with 2% dry milk (M-PBS) at 200 μΙ per tocoin hole for 1 hour at room temperature.

The phage surface-indicating library was then allowed to react with bound peptide by the addition of 100 μΙ phage suspension and 100 μΐ 4% milk powder (M-PBS), and

50830 Β was then incubated for 45 minutes with shaking and 90 minutes without shaking at room temperature.

Unbound phage particles were washed as follows. After the 1st wash cycle 10 x 300 μΐ T-PBS, 5 x 300 μΐ PBS; after 2nd wash cycle: 15 x 300 μΙ T-PBS, 10 x 300 μΙ PBS; after 3rd wash cycle: 20 x 300 μΐ T-PBS, 20 x 300 μΐ PBS.

Purification of bound phage particles was performed by adding 200 μΙ per well of 0.1 M glycine, pH 2.2, and shaking incubation for 30 minutes at room temperature. The phage suspension was then neutralized by the addition of 60 μΐ 2M Tris base, and then infected into E. coli TG1 cells by mixing 10 ml of exponentially growing culture with 0.5 ml of washed phage and incubating for 30 minutes at 37 ° C. Finally, the infected bacteria were plated on ΤΥΕ medium with 1% glucose and 100 pg / ml ampicillin and incubated overnight at 30 ° C.

Table 1: Results of washing the SNZ - phage library on TLR-9 peptide (phage titers)

<td>j Flushing circuit</td><td>TLR-9 concentration during rinsing</td><td>Entrance (phage / ml)</td><td>Output (phage / ml)</td>

<td> ! 1.</td><td>1 mg / ml</td><td> 6,\10^TT'</td><td>2xIO¹⁰</td>

<td> ) 9</td><td>0.5 mg / m!</td><td>4xlO^IS</td><td>2xP) '°</td>

<td> ! -’·</td><td>0.1 mg / ml</td><td>4x10</td><td>6x10^I’</td>

Example 5: Cloning of selected clones of SNZ mutants selected according to TLR-9 for soluble expression Phageemid DNA from phage selected through 3 wash cycles was isolated with midiprep. DNA encoding mutated SNZ regions and batches was amplified by PCR and cloned NcoI-NotJ in the pNOTBAD / Myc-His vector, which is an E. coli expression vector pBAD / Myc-His (Invitrogen) with an inserted NotI restriction site to facilitate cloning. The bound constructs were transferred by electroporation of E.E cells. coli LMG194 (Invitrogen) and were grown overnight at 30 ° C on ΤΥΕ medium with 1% glucose and ampicillin. Selected clones were inoculated into 200 μΐ 2xYT medium with ampicillin, pellets overnight at 30 ° C, and induced by the addition of L-arabinose to a final concentration of 0.1%. After expression overnight at 16 ° C, the cells were harvested

50830 Β by centrifugation and treated with 100 μΐ Na-borate buffer, pH 8.0, overnight at 4 ° C to obtain periplasmic extracts. 50 μΙ of periplasmic extracts were used for ELISA (see below).

Example 6: ELISA of SNZ mutants selected according to TLR.-9

Selected clones were examined for specific binding to the TLR-9 peptide by ELISA.

[0210]

Coating: Microtiter plate (NUNC, Maxisorp), 100 μΐ per well, 20 pg TLR-9 peptide / ml 0.1 M Na-carbonate buffer, pH 9.6, 1h at 37 ° C.

Flushing: 3 x 200 μΙ PBS

Blocking: 1% BSA-PBS, 1 h at room temperature

Flushing: 3 x 200 μΐ PBS

Binding of periplasmic extract: 50 μΐ periplasmic extract, 50 μΙ 2% BSA-PBS, at room temperature overnight.

rinsing: 3 x 200 μΙ PBS

1. antibody: anti-His4 (Qiagen), 1: 1000 in 1% BSA-PBS, 90 min at room temperature 100 μΙ per well.

Rinse: 3 x 200 μΙ PBS

2. antibody: goat anti mouse * HRP (SIGMA), 1: 1000 in 1% BSA-PBS, 90 min at room temperature, 100 μ] per well.

Rinse: 3 x 200 μΙ PBS

Detection: 3 mg / inl OPD in Na-citrate / phosphate buffer, pH

4.5, 0.4 μΐ 30% H₂Oh₂.

Stopping: 100 ml ZM H2SO4.

Absorption reading: 492/620 nm Clones that gave a high signal in this first, preliminary ELISA assay were grown in a volume of 20 ml under the same conditions as described above. Theirs

50830 Β Periplasmic extracts were isolated in 1/20 of the culture volume as described above and tested by ELISA (as described above) for confirmation.

Table 2: ELISA confirmation results

<td></td><td>With antigen</td><td>No antigen</td>

<td>clone</td><td>A₄ 92/620 4 eyes</td><td>A492 / 620 1 reading</td>

<td>A67</td><td> 0.0435</td><td> 0.019</td>

<td>clone</td><td>A492 / 620 4 readings</td><td>A₄92/620 1 reading</td>

<td>B54</td><td> 0,0937</td><td> 0.051</td>

<td>C67</td><td> 0,0295</td><td> 0,013</td>

<td colspan="3">Background (antigen alone) (12 parallel readings): 0.0115</td>

Example 7: Rinsing of SNZ and SNZ + 5 - phage library on hen egg lysozyme 3 rinsing rounds were performed. Maxisc »rp 96-well plates (Nunc) were coated with hen egg lysozyme, with the addition of 200 μΐ of the following solution per hole:

PBS, the following concentrations of dissolved hen egg 1 isozyme:

1. rinse cycle: 2 mg / ml HEL

2. rinsing cycle: 1 mg / ml HEL

3. rinsing circuit; 1 mg / ml HEL The incubation was 1 hour at 37 ° C, then blocked with 2% milk powder (MPBS) at 200 μΙ per well for 1 hour at room temperature.

The phage surface showing library was then allowed to react with bound hen egg lysozyme by adding 100 μΙ phage suspension and 100 μΙ 4% milk powder (M-PBS), then incubated for 45 minutes with shaking and 90 minutes without shaking at room temperature. .

Unbound phage particles were washed as follows:

1. rinsing circuit: 10 x 300 μΐ T-PBS, 5 x 300 μΙ PBS

2. rinsing circuit: 15 x 300 μΙ T-PBS, 10 x 300 μΐ PBS

3. rinsing circuit: 20 x 300 μΐ T-PBS, 20 x 300 μ] PBS

50830 Is Elution of bound phage particles was performed by adding 200 μΐ per well of 0.1 M glycine, pH 2.2, and shaking incubation for 30 minutes at room temperature. The phage suspension was then neutralized by the addition of 60 μΐ 2M Tris-Base, followed by infection and E. coli TG1 cells with a 10 ml mixture for exponential growth of the culture with 0.5 ml of washed phage and incubation for 30 minutes at 37 ° C. Finally, infected bacteria were plated on ΤΥΕ medium with 1% glucose and 100 pg / ml ampicillin and incubated overnight at 30 ° C.

Table 3: Results of SNZ library phage washing on hen egg lysozyme (phage titers)

<td>Rinse circuit</td><td>HEL concentration during rinsing</td><td>IJIaz (phage / ml)</td><td>output (phage / ml)</td>

<td> 1.</td><td>2 mg / ml</td><td></td><td>4.7xlO^IU</td>

<td> 2.</td><td>1 mg / ml</td><td>1.2 9xl ()^JJ</td><td>8.0x1 ()⁹</td>

<td> 3.</td><td>1 mg / ml</td><td>5.71x10^Ζϋ</td><td>4.8x10¹⁰</td>

50830 Β

Table 4: Results of SNZ +5 phage library washing on hen egg lysozyme (HEL) (phage titers)

<td>Rinse circuit</td><td>HEL concentration during rinsing</td><td>IJIaz (phage / ml)</td><td>Iziaz (fag / ni 1)</td>

<td> 1.</td><td>2 mg / ml</td><td> 10</td><td>2.9x10^no</td>

<td> 2.</td><td>1 mg / ml</td><td>2.1x10^1U</td><td>2.6x10^U</td>

<td> 3.</td><td>1 mg / ml</td><td>5.4x10^{) U}</td><td>1.2Χ1Ο¹⁰</td>

Example 8: Cloning of Selected Clones from Example 7 for Soluble Expression Cloning of selected clones for soluble expression was performed as described above for SNZ mutants selected according to TLR-9.

Example 9: Soluble Expression of Selected Clones from Example 7 Soluble expression of selected clones was performed as described above for SNZ mutants selected according to TLR-9. Periplasmic extracts were examined in preliminary ELISA analyzes (see protocol in Example 10).

Clones that gave a high signal in this first, preliminary ELISA analysis were grown in a volume of 20 ml under the same conditions as described above. Their periplasmic extracts were isolated in 1/20 of the culture volume as described above and examined by ELISA (as described in Example 10) for confirmation.

Example 10: ELISA of SNZ mutants selected according to hen egg lysozyme [0220]

Coating: Microtiter plate (NUNC, Maxisorp), 100 μΐ per well, 100 pg hen egg lysozyme / ml in PBS, 1 h at 37 ° C.

Rinsing: 3 x 200 μ! PBS

Blocking: 1% BSA-PBS, 1 h at room temperature

Flushing: 3 x 200 μΐ PBS

Binding of periplasmic extract: 50 μΐ periplasmic extract 50 μΐ 2% BSAPBS, at room temperature overnight

Flushing: 3 x 200 μΙ PBS

50830 Β

1, antibody: anti-His4 (Qiagen), 1: 1000 in 1% BSA-PBS, 90 min at room temperature 100 μΐ per well

Flushing: 3 x 200 μΙ PBS

2. antibody: goat anti mouse * HRP (SIGMA), 1: 1000 in 1% BSA-

PBS, 90 min at room temperature (room temperature),

100 μΙ per hole rinsing: 3 x 200 μΐ PBS

Detection: 3 mg / ml OPD in Na-citrate / phosphate buffer, pH 4.5, 0.4 μΙ 30% H₂Oh₂

Stopping: 100 ml ZM H₂SO₄

Absorption reading: 492/620 nm

Table 5: Results of ELISA validation of SNZ mutants selected according to hen egg lysozyme

<td></td><td>With antigen</td><td>No antigens</td>

<td>clone</td><td>A₄92/620 4 readings</td><td>A492 / 620 1 reading</td>

<td>Β12</td><td> 0,396</td><td> 0,012</td>

<td>PART</td><td> 0,415</td><td> 0,026</td>

<td>D46</td><td> 0,398</td><td> 0,011</td>

<td colspan="3">Background (antigen alone) (12 parallel readings): 0.1763</td>

Table 6: Results of ELISA confirmation with dilutions of SNZ mutant antigens selected by hen egg lysozyme

<td>c (μ₈/ ιτι1) clone</td><td> 200</td><td> 100</td><td> 50</td><td> 25</td><td> 12,5</td><td> 6,25</td><td> 3,125</td><td> 1,55</td><td> 0.78</td><td> 0,39</td>

<td>Β12</td><td> 0,707</td><td> 0,532</td><td> 0,432</td><td> 0,297</td><td> 0,192</td><td> 0,150</td><td> 0,148</td><td> 0,049</td><td> 0,034</td><td> 0,015</td>

<td>D46</td><td> 0,713</td><td> 0,561</td><td> 0,342</td><td> 0,220</td><td> 0,133</td><td> 0,088</td><td> 0,047</td><td> 0,032</td><td> 0,021</td><td> 0,010</td>

<td>DI0</td><td> 0,715</td><td> 0,685</td><td> 0,571</td><td> 0,368</td><td> 0,231</td><td> 0,175</td><td> 0,171</td><td> 0.068</td><td> 0,047</td><td> 0,026</td>

<td>- (nc)</td><td> 0,449</td><td> 0,360</td><td> 0,165</td><td> 0,072</td><td> 0,038</td><td> 0,023</td><td> 0,017</td><td> 0,013</td><td> 0,009</td><td> 0,007</td>

nip: no periplasmic extract added.

Hen egg lysozyme was observed to react with the anti-his4 antibody, and therefore a relatively high background was observed.

Table 7: Confirmation results with ELISA SNZ + 5 mutants selected according to hen egg lysozyme

50830 Β

<td></td><td>With antigen</td><td>No antigen</td>

<td>clone</td><td>A492 / 620 4 readings</td><td>A492 / 620 1 reading</td>

<td>Α13</td><td> 0,197</td><td> 0,016</td>

<td>A66</td><td> 0,461</td><td> 0,019</td>

<td>B18</td><td>0.533 (5 readings)</td><td>not done</td>

<td>B20</td><td> 0,184</td><td> 0,016</td>

<td>B68</td><td> 0,535</td><td> 0.019</td>

<td>B40</td><td> 0,706</td><td> 0.051</td>

<td>C24</td><td> 0,352</td><td> 0,072</td>

<td>D22</td><td> 0,147</td><td> 0,019</td>

<td>Ч22</td><td> 0,439</td><td> 0,017</td>

<td>D37</td><td> 0,360</td><td> 0,026</td>

<td>D40</td><td> 0,559</td><td> 0,034</td>

<td>D56</td><td></td><td> 0,019</td>

<td colspan="3">Background (antigen alone) (12 parallel eyes): 0.1334 Note: hen egg lysozyme reacts with anti-his₄ antibody and therefore a relatively high background was observed.</td>

Example 11: CL library Visual control of the crystal structure of the Fab fragment (Fab structure of the human monoclonal antibody 3D6: RSCB Protein Data Bank http://www.rcsb.org/pdb/) was used Entry 1DFB.PDB (He HM, et Proc Natl Acad Sci USA, 1992 Aug 1; 89 (15): 7154-8) and computer-assisted analysis (e.g. The Protein Explorer was used for this purpose (http://molvis.sdsc.edu/protexpl/frntdoor.htm () the secondary and tertiary structures of this protein) allows the identification of residues located in the loop regions connecting the beta strands of the CL-domain skeleton. These residues include amino acids 8 to 18, amino acids 27 to 35, amino acids 42 to 78, amino acids 83 to 85, amino acids 92 to 100, amino acids 108 to 117 and amino acids 123 to 126 (numerical designations according to the IMGT system (Lefranc MP, et al. Nucleic Acids Res. 2005 Jan 1; 33 (Database issue): D593-7; Lelranc MP, et al. Dev Comp Imunol. 2005; 29 (3): 185203)).

More specifically, residues 11, 12, 14-18 and 92-95 were randomized within the human CL domain (SEQ ID NO: 48). Randomization was achieved by PCR amplification of coding sequences with PCR primers in which the positions of the relevant codons were encoded with the nucleotide sequence 5'-NNS-3 ', which potentially encodes all 20 amino acids, while avoiding 2 of the 3 stop codons. The library insert was amplified with two separate PCR reactions, and the two PCR fragments were linked together via an HpyCH4IV restriction site that was introduced as a silent mutation using a PCR primer. Primers

50830 Β further ensure the restriction of the endonuclease site Ncol and Notl for cloning into the pHEN vector showing phage (Hoogenboom HR, et al. Nuclcic Acids Res. 1991 Aug 11; 19 (15): 4133-7). The C-terminal cysteine ​​from the CL domain is not included for phage display, but can be added later when a modified CL clone is used e.g. for the construction of the Fab fragment.

A plasmid such as pRcCMV3D6LC (Riiker F, et al. Ann NY Acad Sci. 1991 Dec 27; 646: 212-9) was used as a template for PCR amplification, which contained a complete light chain of a human monoclonal antibody as an insert.

For CL + 3 (SEQ ID No. 50, 51) and CL + 5 (SEQ ID No. 52, 53) libraries, which contain additional residues inserted between positions 92 and 95 of the CL domain, primers CLRHPY3 and CLRHPY5 were used. instead of CLRHPY primer.

Below is the nucleotide DNA and amino acid sequence of the final product PCRs and linkages, cloned into the Ncol site of pHEN1, which leads to the binding of the pelB leader sequence to the N-terminus of the construct (SEQID Nos. 48, 49):

50830 Β + 3 Μ Κ Υ LLPT Α Α Α GLL LL Α Α

ATGAAATACC TATTGCCTAC GGCAGCCGCT GGATTGTTAT TACTCGCGGC

Ncol '+ Z Q Ρ Α Μ Α V AAPS VFI FPP

CCAGCCGGCC ATGGCCGTGG CTGCACCATC TGTCTTCATC TTCCCGCCAT + 3S Q ASVVCLLN

101 CTNNSNNSCA GNNSNNSNNS NNSNNSGCCT CTGTTGTGTG CCTGCTGAAT + 3 NF Υ PREA Κ VQW Κ V DNAL

151 AACTTCTATC CCAGAGAGGC CAAAGTACAG TGGAAGGTGG ATAACGCCCT + 3 QSG NSQ ESVT EQD SKD

201 CCAATCGGGT AACTCCCAGG AGAGTGTCAC AGAGCAGGAC AGCAAGGACA

HpyCH4IV + 3STY SLSSTLTL YE

251 GCACCTACAG CCTCAGCAGC ACCCTGACGT TGNNSNNSNN SNNSTACGAG +3 Κ Η Κ VYA S EVT Η QG LSSP

301 AAACACAAAG TCTACGCCTG CGAAGTCACC CATCAGGGCC TGAGCTCGCC

Notl + 3 VTK SFN RGEA AA

351 CGTCACAAAG AGCTTCAACA GGGGAGAGSC GGCCGCA

List of primers for the CL library:

[0204]

50830 Β cllnco: 5'-cttaccatgg ccgtggctgc accatctgtc ttcatcttcc cgccatctnn snnscagnns nnsnnsnnsn nsgcctctgt tgtgtgc-Z '(SEQ ID No. 56) gll: Z'-tgacaacgcc agg aacacaaagt s-3 '(SEQ ID No 58) clrhpy3: 5'-tcagaacgtt gnnsnnsnns nnsnnsnnsn nstacgagaa acacaaagtc-3' (SEQ ID No. 59) clrhpy5: 5'-tcagaacgtt gnnsnnsnn nsnnsnns 60) clrnot. · 5 -catcgcggcc gcctctcccc tgttgaagct c-3 '(SEQID No. 61) The number of selected library clones (mutated CL dotted cloned into the phagemid vector pHEN1) was controlled by restriction analysis and by DNA sequencing to see if they contained an insert as planned, including correctly inserted randomized sequences. For the next steps in obtaining phage, standard protocols follow. Briefly, the binding mixture is transferred by electroporation to TG1 cell E. inches. The phage particles are then rescued from cell E. coli TGl with helper phage Μ13-ΚΟ7. Phage particles are then precipitated from the culture supernatant with PEG / NaCl in 2 colognes, dissolved in water and used for washing by washing, alternatively, they can be stored at minus 80 ° C.

Example 12: CH1 library Visual control of the crystal structure of the Fab fragment (the structure of the Fab human monoclonal antibody 3D6: RSCB Protein Data Bank Entrv IDFB.PDB was used) and computer-assisted analysis (Protcin Explorer was used for this purpose) of secondary and tertiary structures these proteins allow the identification of residues located in the regions of the loops connecting the beta strands of the CH1 domain skeleton. These residues include amino acids 7 to 21, amino acids 25 to 39, amino acids 41 to 81, amino acids 83 to 85, amino acids 89 to 103 and amino acids 106 to 117 (numerical designations according to the IMGT numbering system).

More specifically, residues 12-19 and 93-100 were randomized within the human CH1 domain (SEQ ID Nos. 54, 55). Randomization was achieved by PCR code amplification

50830 Β Sequences with PCR primers in which the positions of the relevant codons are encoded by the nucleotide sequence 5'-NNS-3 ', which potentially encodes for all 20 amino acids, while avoiding 2 of the 3 stop codons. (The library insert is amplified by two separate PCR reactions, and the two PCR fragments are linked together via a BstEJI restriction site that occurs naturally in the CH1 domain. The primers further provide an Ncol and Notl endonuclease restriction site for cloning into a phage-showing pHEN vector. The C-terminal cysteine ​​in the CH1 domain is not included for phage display, but can be added later if a modified CH1 clone is used, e.g. for the construction of the Fab fragment.

A plasmid such as pRcCMV3D6HC was used as a template for PCR amplification, which contained a complete heavy chain of human monoclonal antibody as an insert. Below is the nucleotide DNA and amino acid sequence of the final product PCRs and linkages, cloned into the Mcol site in pHEN1, leading to the binding of the pelB leader sequence at the N-terminus of the construct (SEQ ID NOs 54, 55):

50830 Β

<td> + 3</td><td>MK. Υ</td><td>LLPT</td><td>AA Ά</td><td>GLL</td><td>LLAA</td>

<td> 1</td><td>ATGAAATACC</td><td>TATTGCCTAC</td><td>GGCAGCCGCT</td><td>GGATTGTTAT</td><td>TACTCGCGGC</td>

<td></td><td colspan="2">Ncol</td><td></td><td></td><td></td>

<td> + 3</td><td>QPA</td><td>MAA</td><td>STKG</td><td>PSV</td><td>FPL</td>

<td> 51</td><td colspan="2">CCAGCCGGCC ATGGCCGCCT</td><td>CCACCAAGGG</td><td>CCCATCGGTC</td><td>TTCCCCCTGG</td>

<td> + 3</td><td>APSS</td><td></td><td></td><td>AL</td><td>GCL</td>

<td> 101</td><td>CACCCTCCTC</td><td>CNNSNNSNNS</td><td>NNSNNSNNSN</td><td>NSNNSGCCCT</td><td>GGGCTGCCTG</td>

<td> + 3</td><td>VKD</td><td>YFPE</td><td>PV t</td><td>In sw</td><td>NSGA</td>

<td> 151</td><td>GTCAAGGACT</td><td>ACTTCCCCGA</td><td>ACCGGTGACG</td><td>GTGTCGTGGA</td><td>ACTCAGGCGC</td>

<td> + 3</td><td>LT 5</td><td>GVH</td><td>TFPA</td><td>VLQ</td><td>SSG</td>

<td> 201</td><td>CCTGACCAGC</td><td>GGCGTGCACA</td><td>CCTTCCCGGC</td><td>TGTCCTACAG</td><td>TCCTCAGGAC</td>

BstEII

<td> + 3</td><td>LYSL</td><td>SSV</td><td>VTV</td><td>P</td><td></td>

<td> 251</td><td>TCTACTCCCT</td><td>CAGCAGCGT &amp;</td><td>GTGACCGTGC</td><td>CCNNSNNSNN</td><td>SNNSNNSNNS</td>

<td> + 3</td><td>YOU</td><td>ICNV</td><td>N Η K</td><td>PSN</td><td>TKVD</td>

<td> 301</td><td>NNSACCTACA</td><td>TCTGCAACGT</td><td>GAATCACAAG</td><td>CCCAGCAACA</td><td>CCAAGGTGGA</td>

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

<td> + 3</td><td>KKV</td><td>EPC</td><td>SAAA</td><td></td><td></td>

<td> 351</td><td>CAAGAAAGTT</td><td>GAGCCCAAAT</td><td>CTGCGGCCGC</td><td>A</td><td></td>

Primer list for CH1 library [0232]

CHILNCO: 5'-acgtccatgg ccgcctccac caagggccca tcggtcttcc ccctggcacc ctcctccnns nnsnnsnnsn nsnnsnnsnn sgccctgggc tgcctggtc-Z '(SEQ ID No. 62)

CHILBST: 5 '> ggcacggtca ccacgctgct gag-3' (SEQID no. 63)

50830 Β

CHIRBST: 5'-agcgtggtga ccgtgcccnn snnsnnsnns nnsnnsnr.sa cctacatctg caacgtgaat c-3 '(SEQ ID No. 64)

CHIRNOT: 5'-catagcggcc gcagatttgg gctcaacttt cttgtc-3 '(SEQ ID NO: 65) [0233] The number of selected library clones (fused CH1 domains cloned into the phagemid vector pHEN1) is controlled by restriction analysis and by DNA sequencing according to the contents of the insert. plan, including properly inserted randomized sequences. The next steps for obtaining phage are followed by standard protocols. Briefly, the binding mixture is transferred by electroporation into E. coli TG1 cells. The phage particle is then rescued from the E. coli TG1 cell with the helper phage Μ13--7. Phage particles are then precipitated from the culture supernatant with PEG / NaCl in 2 steps, dissolved in water and used for selection by washing, alternatively, they can be stored at minus 80 ° C.

Example 13: Flushing of the CH1-phage library on hen egg lysozyme (HEL) 3 washings were performed with the CHI-phage library (see Example 12). Maxisorp 96-well plates (Nunc) were coated with hen egg lysozyme by adding 200 μΐ of the following solution per hole: PBS, with the following concentrations of dissolved hen egg lysozyme:

1. rinse cycle: 2 mg / ml HEL

2. rinsing cycle: 1 mg / ml HEL

3. wash cycle: 1 mg / ml HEL Incubation for 1 hour at 37 ° C, then blocked with 2% milk powder (M-PBS) at 200 μΐ per well for 1 hour at room temperature.

The surface showing the phage library was then allowed to react with bound hen egg lysozyme by the addition of 100 μ! phage suspensions and 100 μΐ 4% milk powder (MPBS), then incubated for 45 minutes with shaking and 90 minutes without shaking at room temperature.

Unbound phage particles are washed as follows:

1. rinsing circuit: 10 x 300 μΙ T-PBS, 5 x 300 μΐ PBS

50830 Β

2. rinsing circuit: 15 χ 300 μΐ T-PBS, 10 χ 300 μΙ PBS

3. wash cycle: 20 χ 300 μΙ T-PBS, 20 χ 300 μΐ PBS Wash the associated phage particles by adding 200 μΙ per well of 0.1 M glycine, pH 2.2, and shaking incubation for 30 minutes at room temperature. . The l'aga suspension was then neutralized by the addition of 60 μΙ 2M Tris-Base, then infected into E. coli TG1 cells using a 10 ml exponential growth culture mixture with 0.5 ml of washed phage and incubated for 30 minutes at 37 ° C. Finally, infected bacteria were plated on ΤΥΕ medium with 1% glucose and 100 pg / ml ampicillin, and incubated at 30 ° C overnight.

Cloning of Cloned CH1 Mutant Clones Selected by Lysozyme for Soluble Expression Phage resin DNA from phage selected through 3 wash cycles was isolated with midi-prep. The DNA encoding the mutated CH1 domain was amplified by batch PCR and cloned with Ncol-NotI into the pNOTBAD / Myc-His vector, which is an E. coli expression vector pBAD / MycHis (Invitrogen) with an inserted NotI restriction site to facilitate cloning. The bound constructs are transferred by electroporation and to E. coli cells LMG194 (Invitrogen) and grown at 30 ° C on medium ΤΥΕ with 1% glucose and ampicillin overnight. Selected clones were inoculated into 200 μΐ 2xYT medium with ampicillin, grown overnight at 30 ° C and induced by the addition of L-arabinose to a final concentration of 0.1%. After expression at 16 ° C overnight, cells were harvested by centrifugation and treated with 100 μΐ Na-borate buffer, pH 8.0, then 4 ° C overnight to obtain periplasmic extracts. 50 μΙ periplasmic extracts are used for ELISA.

Clones giving a high signal in the former. by preliminary ELISA analysis are grown in a volume of 20 ml under the same conditions as described above. Their periplasmic extracts were isolated in 1/20 of the culture volume as described above and examined by ELISA (as described below) for confirmation.

ELISA CH1 mutants selected according to hen egg lysozyme

50830 Β [0241]

Coating: Microtiter plate (NUNC, Maxisorp), 100 μΐ per well, 100 μg hen egg lysozyme / ml in PBS, 1 h at 37 ° C

Flushing: 3 x 200 μΐ PBS

Blocking: 1% BSA-PBS, 1 h at room temperature

Rinsing; 3 x 200 μΐ PBS

Binding of periplazine extract: 50 μΙ periplasmic extract 50 μΙ 2% BSA-PBS, at room temperature overnight

Flushing: 3 x 200 μΐ PBS

1. antibody: anti-His4 (Qiagen), 1: 1000 in 1% BSA-PBS, 90 min at room temperature 100 μΙ per well

Flushing: 3 x 200 μΙ PBS

2. antibody: goat anti mouse * HRP (SIGMA), 1: 1000 in 1% BSA-

PBS. 90 min at room temperature 100 μΐ per well

Flushing: 3 x 200 μΐ PBS

Detection: 3 mg / ml OPD in Na-citrate / phosphate buffer, pH 4.5, 0.4 μΙ 30% H₂Oh₂

Stopping: 100 ml ZM H2SO4

Absorption reading: 492/620 nm Clones are considered positive if their ELISA signal is at least three times higher than the background signal.

Example 14: Rinsing of the CL-fag library on hen egg lysozyme (HEL) 3 rinsing rounds are performed with the CL-fag library (see example 11). Maxisorp 96-well plates (Nunc) are coated with hen egg lysozyme by adding 200 μΐ of the following solution per hole: PBS, with the following concentrations of dissolved hen egg lysozyme:

1. rinse cycle: 2 mg / ml HEL

2. rinsing cycle: 1 mg / ml HEL

3. rinsing cycle: 1 mg / ml HEL

50830Β incubation for 1 hour and then 37 ° C, then blocking with 2% dry milk (MPBS) at 200 μΐ per well for 1 hour at room temperature.

The surface showing the phage library was then allowed to react with bound hen egg lysozyme by the addition of 100 μΐ of phage suspension and 100 μ! 4% milk powder (MPBS), zatiin was incubated for 45 minutes with shaking and 90 minutes without shaking at room temperature.

Unbound phage particles are washed as follows:

1. rinsing circuit: 10 x 300 μΙ T-PBS, 5x 300 μΐ PBS

2. rinsing circuit: 15 x 300 μΐ T-PBS, 10x 300 μΐ PBS

3. wash cycle: 20 x 300 μΙ T-PBS, 20 h 300 μΐ PBS The washings of bound phage particles were performed by adding 200 μΙ per rupee of 0.1 M glycine, pH 2.2, and shaking incubation for 30 minutes at room temperature. . The phage suspension was then neutralized by the addition of 60 μΐ 2M Tris base, followed by infection in TG1 E. coli cells using a 10 ml culture mixture for exponential growth with 0.5 ml of washed phage and incubation for 30 minutes and 37^QC. Finally, the infected bacteria were plated on ΤΥΕ medium with 1% glucose and 100 pg / ml ampicillin and incubated at 30 ° C overnight.

Cloning of selected clones of CL mulants selected according to lysozyme for soluble expression Phagemid DNA from phage selected through 3 wash cycles was isolated with midi-prep. DNA, encoding the mutated CL domain is amplified by batch PCR and cloned Ncol-Notl in the vector pNOTBAD / Myc-His, which is the expression vector of pBAD / Myc-His E. coli (Invitrogen) with inserted Notl restriction site to facilitate cloning . The bound constructs were electroporated into LMG194 E cells. coli (Invitrogen) and grown at 30 ° C on medium ΤΥΕ with 1% glucose and ampicillin overnight. Selected clones were inoculated into 200 μΙ 2xYT medium with ampicillin, grown overnight at 30 ° C and induced by the addition of L-arabinose to a final concentration of 0.1%. After expression at 16 ° C overnight, cells were harvested by centrifugation and treated with 100 μΐ Na-borate

50830 Β buffer, ρΗ 8.0, at 4 ° C overnight to obtain periplasmic extracts. 50 μΙ of periplasmic extracts are used for ELISA.

Clones that give a high signal in this first, preliminary ELISA assay were grown in a volume of 20 ml under the same conditions as described above. Their periplasmic extracts were isolated in 1/20 of the culture volume as described above and examined by ELISA (as described below) for confirmation.

ELISA for CL mutants selected according to hen egg lysozyme [0250]

Coating: Microtiter plate (NUNC, Maxisorp), 100 μΙ per well, 100 pg hen egg iisosima / ml in PBS, 1 h at 37 ° C

Flushing: 3 x 200 μΐ PBS

Blocking: 1% BSA-PBS, 1 h at room temperature

Flushing: 3 x 200 μΐ PBS

Binding of pcriplasmic extract: 50 μΙ periplasmic extract 50 μΙ 2% BSAPBS, at room temperature at night

Rinsing; 3 x 200 μΙ PBS

1. antibody: anti-His4 (Qiagen), 1: 1000 in 1% BSA-PBS, 90 min at room temperature 100 μΙ per well

Flushing: 3 x 200 μΐ PBS

2. antibody: goat anti mouse * HRP (SIGMA), 1: 1000 in 1% BSA-

PBS, 90 inin at room temperature 100 μΙ per well

Flushing: 3 x 200 μΙ PBS

Detection: 3 mg / ml OPD in Na-citrate / phosphate buffer, pH 4.5, 0.4 μ | 30% H₂Oh₂Stopping: 100 ml ZM H₂SO₄

Absorption reading: 492/620 nm Clones are considered positive if their ELISA signal is at least three times greater than the background signal.

50830 02 Example 15: Construction of an immunoglobulin domain randomized on both sides (bispecifically modified SNZ domain) This example describes a modified iminoglobulin domain with two binding specificities.

The construction of this modified immunoglobulin domain includes the following strategy:

• a modified SNZ domain, clone C24 (see example 10), obtained from a SNZ + 5 library bound specifically to lysozyme was used as starting material, • residues to be randomized are identified in that modified SNZ domain that binds the β-strands of the immunoglobulin folds, and located on the opposite side of the domain compared to the residues mutated to form clone C24, • PCR primers are designed to allow randomization of these residues and synthesis of this modified immunoglobulin domain by a procedure similar to the procedure described above for the SNZ, SNZ + Z and SNZ + 5 libraries.

The 4 PCR products at random positions were linked and the full-length inserts were amplified by PCR. They were then cloned into pHEN-1 via the NcolNotI site and transformed into E. coli TG-1 cells for a library construct of approximately 108 colonies. 20 randomly selected colonies were sequenced and the randomized sites were found to be independently mutated. Also, no wild-type (C24) sequences were observed. Library phage was generated according to standard protocols, and a phage titer of 6.32 x 1010 TU / ml was obtained.

For the bispecificity assay, recombinant human erythropoietin (rhEPO) was selected as the second antigen, while the construct was expected to retain its originally modified specificity for hen egg lysozyme. rhEPO-reactive phage were selected in 4 wash cycles. To preserve the population of C24 clones that should continue to bind hen egg lysozyme after mutagenesis, the first round of selection on rhEPO was followed by a round wash of the phage population on hen egg lysozyme (1 mg / ml in PBS). 200 μΙ rhEPO was coated on 5 wells of a microtiter plate (Maxisorp, Nunc) in 0.1 M Carbonate Buffer, pH 9.6, in decreasing concentrations in successive wash cycles (see table below). After blocking with 2% M-PBS, phage in the blocking agent were allowed to bind at room temperature for 2 h. After 20 washes with T-PBS and 20 with PBS, it was washed with 0.1 M glycine, pH 2.2, and neutralized with 2M

50830 Β

Tris. Washed phage was used directly to infect exponentially growing TG-I. infected cells were selected on ampicillin-containing medium. Phage particles were rescued from the culture supernatant after superinfection with helper phage Μ13-ΚΟ7, concentrated with PEG and used in the second wash round. Inlet and outlet phage numbers were determined as transfected E. coli units after each wash cycle (Table 8).

Table 8:

<td>Rinse circuit</td><td>Antigen</td><td>Inlet phage (TU / ml)</td><td>Output phage (TU / ml)</td>

<td> 1</td><td>rliEPO, 500 gg / ml</td><td>6.32 h! 0^š</td><td>1.9x10 '</td>

<td> 2</td><td>lysozyme, 1 mg / ml</td><td>6.16x10 * '</td><td>4.53x10¹¹’</td>

<td> 3</td><td>rhEPO, 100 pg / ml</td><td>6.07x10 * '</td><td>6.78x10 *^ύ</td>

<td>⁴</td><td>rhEPO, 50 gg / ml</td><td>8.42x10 * '</td><td>3.0x10 **</td>

<td> 5</td><td>rhEPO. 50 gg / ml</td><td>5.12h I θ '</td><td>4.28x10 '</td>

The resulting colonies were scraped from the plates, grown in 2xYT with ampicillin, and their plasmid DNA was isolated with midi-prep. The inserts were amplified by PCR, then subcloned into the vector pNOTBAD and transformed into E. coli species Ε104. 4x72 colonies were grown in 200 μΙ 2xYT with ampicillin and induced with 0.1% Larabinosis the next day. After 24 h expression at I6 ° C, they were lysed with 200 μΙ of Naborate buffer, pH 8.0 for 6 h at 4 ° C and periplasmic extract was used for ELISA.

For ELISA, Maxisorp plates were coated with hen egg lysozyme in PBS (20 pg / ml) or rhEPO in 0.1 M Na-carbonate buffer, pH 9.6, for 1 h at 37 ° C. After blocking with 1% BSA-PBS, the periplasmic extract was allowed to bind in the same blocking agent overnight. It has been shown to bind to anti-His- (4) antibody and to goat anti-mouse IgG antibody conjugated to HRP (for hen egg lysozyme detection) or AP (for rhEPO detection). The color of the OPD conversion reaction (HRP) was read at 492/620 nm after stopping with 1.25 M H2SO4, and pNPP conversion (AP) was read at 405/620 nm. 14 clones with promising absorption values ​​were selected for expression at a scale of 20 ml. After 24 h of induction with arabinose at 16 ° C, the cells were harvested and lysed overnight in 1 ml of Na-borate buffer at 4 ° C, and the lysate was used for ELISA. ELISA analysis was performed as above in 4 parallels and in wells without periplasmic extract, and the antigen-free sample was taken as a negative control. The results (Table 9) were obtained with a clone according to SEQ ID NO. 42, 43.

50830 Β

Table 9:

<td>Atitigen</td><td></td><td>Absorption on binding</td><td>Without peripiazmic extract</td><td>Without antigen</td>

<td>lysozyme</td><td>^ 492/620 p »</td><td> 0,299</td><td> 0,110</td><td> 0,018</td>

<td>rhEPO</td><td>A405 / 620 nm</td><td> 0.258</td><td> 0,095</td><td> 0.090</td>

Example 16: Modified SNZ domains provide bispecificity in Fab-like format In the construct used in this example, both chains, VL and VH, antibodies were fused to the modified SNZ domain.

VL and VH region of human monoclonal antibody 3D6 (He HM, et al. Proc Natl Acad Sci USA, 1992 89: 7154-8 .; Kohl J, et al. Ann NY Acad Sci. 1991 646: 106-

14 .; Felgenhauer M, et al. Nucleic Acids Res. 1990 18: 4927), which recognizes the epitope on gp41 in HIV-1 was used as a fusion participant to modify the SNZ domain of clone C24 that binds specifically to hen egg lysozyme.

In order to promote the formation of VL-CH3 / VH-CH3 dimers via a disulfide bond, Ser-Cys residues were added to the C-terminus of the C24 sequence.

The nucleotide and amino acid sequences of the two chains, 3D6VL-C24 and 3D6VI1-C24, are given in SEQ ID NO. 47, 46 and SEQ ID no. 45, 44.

[0263] Primers were constructed that allow the coded regions to be augmented, the introduction of constraint sites at the same time (silent mutations), which were used to interconnect the coded regions. The Pichia pastoris expression system was selected for gene expression. The constructs were cloned into suitable Pichia pastoris expression vectors: 3D6VL-C24 was cloned into pPIC9K (final name: pPIC9K3LC) and 3D6VH-C24 (final name: pPICZ3HC) was cloned into pPICZalfaA. The pPICZ3HC construct was linearized with Bgl P, transformed into Pichia pastoris GS1I5, and transformants were selected on zeocin-containing solid medium. One of the transformants was then used as a host cell for the Sal I-linearized construct pPIC9K3LC. Then, double transformants on RDB medium were selected.

Clones were inoculated into 30 ml of YPG tissue and grown to OD10 = 10, and then induced by the addition of 1% methanol to ΒΜΜΥ medium. Induction was continued for 36 hours at 16 ° C. The supernatants were removed by centrifugation and then concentrated to approximately 10

50830 Β times. The presence of recombinant protein was confirmed by Western blotting with antiHis (4) antibody, and a concentration of approximately 50-100 pg / l of the initial culture was determined.

The first functional analyzes were performed with a ten-fold concentrated supernatant. First, Maxisorp plate holes were coated with 20 pg / ml hen egg lysozyme in PBS or 20 pg / 'ml epitope of antibody 3D6 in 0.1 M Na-carbonate buffer, pH 9.6, for 1 h at 37 ° C. The 3D6 epitope was used in the form of a recombinantly produced GST fusion protein. After blocking with 1% BSA-PBS. concentrated supernatants were allowed to bind overnight in the same blocking agent. Binding to anti-His (4) antibody and goat anti-mouse antibody, conjugated to IIRP, was shown to be seen as a stained reaction obtained from OPD by conversion at 492/620 nm (Table 10).

Table 10

<td>Antigen</td><td>ELISA signal</td><td>Background</td><td>Background (without</td>

<td></td><td>(A.1Č2 / (, 2o)</td><td>(without antigen)</td><td>supernatant)</td>

<td>lysozyme</td><td> 0,198</td><td> 0.003</td><td> 0,043</td>

<td>3Ђ6 epitope</td><td> 0,061</td><td> 0,001</td><td> 0,007</td>