WO2012017069A1

WO2012017069A1 - Linker peptide and its use in fusion proteins

Info

Publication number: WO2012017069A1
Application number: PCT/EP2011/063516
Authority: WO
Inventors: Michael Bachmann
Original assignee: Technische Universität Dresden
Priority date: 2010-08-06
Filing date: 2011-08-05
Publication date: 2012-02-09
Also published as: DE102010039015A1

Abstract

The invention relates to linker peptides, fusion proteins and fusion protein complexes containing them, to processes for the preparation of fusion proteins and to their use. The object of the invention is to provide a flexible linker peptide which, as the result of being used in fusion proteins, yields stable and functionalizable fusion proteins and which can be employed in a modular fashion. To this end, the invention comprises a linker peptide containing an amino acid sequence with at least 10 and not more than 50 amino acids and at least 90%, preferably at least 95%, especially preferably at least 99% amino acid sequence identity to the amino acid sequence according to SEQ ID No. 1. The invention also comprises the use of a linker peptide according to the invention as part of a fusion protein and fusion proteins which contain at least one linker peptide according to the invention and at least one further protein or peptide as fusion partner. Also comprised are fusion protein complexes which contain a fusion protein according to the invention and further molecules bound via domains of the linker peptide. The invention furthermore comprises antibodies or antibody fragments which specifically bind a linker peptide according to the invention, and the use of these antibodies or antibody fragments in fusion proteins.

Description

Linker peptide and its use in fusion proteins

The invention relates to a linker peptide, fusion proteins and fusion protein complexes containing the same, processes for the preparation of fusion proteins and their use. The invention finds application in the field of molecular biology, medicine, pharmacy and biomedical research.

Recombinant fusion proteins, in particular bifunctional recombinant fusion proteins are increasingly being used as therapeutics. In order to prepare the recombinant fusion proteins (also referred to herein more simply as fusion proteins), a fusion gene is generated by fusing one gene sequence after removal of the stop codon with another gene sequence. In the translation of this fusion gene, a fusion protein is generated. Fusion proteins consist of a protein domain fused to another domain. In this case, at least one peptide domain (as a linker peptide) is included, possibly further protein domains are included.

The functionality of the individual protein domains remains intact in fusion proteins, since the functional domains of proteins are usually of modular design, so that the amino acid sequence of a protein domain, which is assigned a specific function, can be separated from the rest of the protein without impairing its functionality.

Fusion proteins are used on the one hand in molecular biology, e.g. to facilitate the experimental effort in the purification and detection of proteins. Since the production of a protein-specific antibody which is suitable for this purpose requires a great deal of time and labor and, if appropriate, is disadvantageous due to cross-reactivities or different affinities for different epitopes, fusion proteins are preferred. In particular, fusion proteins from a protein to be expressed (target protein) with protein markers or affinity tags have gained great importance for molecular biology, since they allow specific detection and facilitate the purification of the target proteins, without the production of a specific antibody against the target protein in question is necessary.

Many other biochemical and molecular biology methods rely on the synthesis of fusion proteins, such as the yeast two-hybrid assay and coimmunoprecipitation. For all of these procedures, various fusion partners have been developed and described, many of which have become firmly established in molecular biology. Examples of common fusions with complete proteins or protein domains are β-galactosidase, protein A, avidin, Streptavidin, ubiquitin, GFP (green fluorescent protein), GST (glutathione transferase), MBP (maitosis binding protein), calmodulin or thioredoxin.

Bi- and multifunctional fusion proteins have proven to be particularly advantageous for immunological applications. Examples of bifunctional fusion proteins are bispecific antibodies consisting of two linker-linked variable regions of antibodies.

However, because large proteins in fusion proteins may affect the functionality of their fusion partner under certain circumstances, a large number of short peptides or peptide tags has now been developed, which are used for the production of fusion proteins. Many of these are recognized by specific antibodies (eg, HA tag, myc tag, flag tag). Others also have specific binding properties that can be used for purification of the fusion protein, such as the binding of Ni ²⁺ or Zn ²⁺ ions by the His tag. However, even short fusion partners can, in the worst case, affect the functionality of the desired target protein.

Therefore, peptide domains are preferably used as linkers, which have a flexible structure and separate the fusion partners so that no interaction between the fusion partners occurs. The functionality of the fusion partners should not be changed and, in particular, should not be impaired.

US 5,525,491 discloses linker peptides of the structure (Ser, Ser, Ser, Ser, Gly) _y , with y> 1 (so-called glycine-serine linkers) used to link two or more protein domains to form recombinant fusion proteins. The fused protein domains retain their respective biological activity and are stable to proteolysis.

Further, linker peptides of structure A (EAAAK) _n A (where n = 2-5) are known which have an alpha-helical structure. Corresponding methods for designing corresponding linker peptides which minimize the interaction of the fusion partners are described in Arai R, Protein Engineering (2001).

Other known linker peptides with a helical structure are peptides of the structure (EAAAR) ₄ . One application of the linker peptides in artificially prepared zinc finger peptides is disclosed in Yan W, Biochemistry (2007).

In the event that individual fusion partners are to be separated after the preparation of the fusion proteins, preferably functionalized linker peptides are used. These either contain a specific sequence (cleavage site), which is recognized by proteolytic enzymes, so that an enzymatic cleavage takes place within the linker peptide (Arnau J, Protein Expr Purif (2006)). Alternatively, linker peptides containing self-cleaving or autoproteolytic sequences are used.

Disadvantages of previously used linker peptides are limited stability in the use of large protein domains as fusion partners. In order to use fusion proteins for immunological application, it is usually necessary that, in addition to the transport of antigen, further co-stimulatory signals, e.g. in the form of nucleic acids, peptides or proteins, in particular signal molecules are transported.

The object of the invention is to provide a flexible linker peptide which, by its use in fusion proteins, provides stable and functionalizable fusion proteins; and which is modular.

The object is achieved according to the invention by a linker peptide containing an amino acid sequence with at least 10 and at most 50 amino acids and at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity, to the amino acid sequence according to SEQ ID no. 1. The amino acid sequence according to SEQ ID no. 1 comprises 18 amino acids, therefore preferably a linker peptide according to the invention comprises at least 18 and a maximum of 50 amino acids. In this case, preferably the 18-amino-acid-long part of the linker peptide according to the invention, which has a homology to the amino acid sequence of SEQ ID NO. 1, at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity to SEQ ID NO. 1 on. Further preferred linker peptides comprise an amino acid sequence with a maximum of 40, more preferably a maximum of 30, amino acids.

The invention is based on the discovery of the inventors that peptide structures having an amino acid sequence corresponding to those of amino acids No. 311 - No. 328 of the human La protein (also referred to in the literature as SS-B) are useful for linking protein domains , The amino acid sequence of the human La protein is shown in SEQ ID no. 3 shown.

The human La protein (hLa) was originally described as an autoantigen in patients with systemic lupus erythematosus and Sjögren's syndrome and has the alternative name SS-B. It comprises 408 amino acids according to SEQ ID no. 3. The primary structure of the hLa protein can be divided into three regions that form independent spatial domains (Figure 1). N-terminal is the so-called La motif, followed by a central RNA recognition motif (RNA recognition motif, RRM), also referred to as RRM1. These two domains form the N-terminal half of the protein and are collectively called LaN. The second half, LaC, contains the C-terminal RRM (RRM2, amino acids 225-334) and a subsequent long, flexible element of about 80 aa that has no secondary structure features. The human La protein is predominantly localized in the cell nucleus, where it binds via the RRMs to newly synthesized RNA polymerase III transcripts. The human La protein also has several regulatory elements, such as a nuclear retention element (NRE), which inhibits nuclear export and retains the protein in the nucleus. The NRE extends via amino acids 316-332 of SEQ ID NO. 3. Furthermore, a previously not further restricted dimerization domain extends at amino acids 293-348 of SEQ ID no. Third

Peptides are organic compounds that result from a combination of several amino acids, the amino acids in a defined sequence (sequence) via peptide bonds are interconnected (primary structure). Peptides in the sense of the invention are understood to mean amino acid sequences having a maximum length of 100 amino acids. Polypeptides longer than 100 amino acids in length are called proteins. The secondary structure of peptides and proteins may include alpha-helices, beta-sheets, beta-loops, and / or random coil structures. Linker peptides are peptides used for association with at least one other peptide or protein, preferably in the form of fusion proteins.

The linker peptide according to the invention contains an alpha-helical structure. This ensures the flexibility of the linker peptide and advantageously allows application with a variety of possible fusion partners.

The linker peptide according to the invention comprises an amino acid sequence which has at least 90%, preferably 95%, particularly preferably at least 99% amino acid sequence identity to an amino acid sequence of at least 10, preferably at least 10 and at most 50, amino acids from the RRM2 region of the human La protein , especially from the long C-terminal alpha helix of RRM2.

Under amino acid sequence identity according to the invention the percentage of amino acids of a candidate sequence (at this point: the amino acid sequence of the central RNA-binding region of the human La protein) understood that are identical to the amino acid sequence of a linker peptide according to the invention.

A preferred linker peptide comprises an amino acid sequence according to SEQ ID no. 1 (herein this particular linker peptide is also referred to as "E7B6".) The amino acid sequence of SEQ ID No. 1 comprises amino acid # 311 - # 328 of SEQ ID No. 3. Preferably, a linker peptide according to the invention contains a nucleic acid binding domain and / or a protein dimerization domain. Possibly. For example, the nucleic acid binding domain and the protein dimerization domain are arranged so that their amino acid sequences overlap.

The nucleic acid binding domain is part of the linker peptide to which nucleic acids can bind. The nucleic acid binding domain can be based on a linear or spatial peptide domain.

The protein dimerization domain is part of the linker peptide through which several, especially two, linker peptides interact with each other to form a dimer. As a result, fusion proteins according to the invention can advantageously be dimerized with other fusion proteins according to the invention via the protein dimerization domain located in the linker peptide, thereby forming a fusion protein complex. A preferred linker peptide having a protein dimerization domain comprises the amino acid sequence of SEQ ID NO. 2 (herein this particular linker peptide is also referred to as "E7B6L").

In a linker peptide of the invention, the nucleic acid binding domain and the protein dimerization domain may overlap, i. that a part of the amino acid sequence of the linker peptide can be assigned to both the nucleic acid binding domain and the protein dimerization domain.

A further preferred linker peptide comprises an amino acid sequence with at least 90%, preferably at least 95%, particularly preferably at least 99% amino acid sequence identity to the amino acid sequence according to SEQ ID no. 2. The amino acid sequence according to SEQ ID no. Figure 2 comprises amino acids # 303 - # 344 of the human La protein (SEQ ID NO: 3). Particularly preferred is a linker peptide having an amino acid sequence which is identical to the amino acid sequence according to SEQ ID no. 2 is.

The linker peptide having the amino acid sequence of at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity to the amino acid sequence of SEQ ID NO. 2 preferably contains a nucleic acid binding domain.

In a linker peptide which has the amino acid sequence according to SEQ ID no. 2, a nucleic acid binding domain is included.

The linker peptide having the amino acid sequence of at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity to the amino acid sequence of SEQ ID NO. 2 preferably contains a protein dimerization domain. In a linker peptide which has the amino acid sequence according to SEQ ID no. 2, a protein dimerization domain is included.

Preferably, linker peptides according to the invention contain, in addition to any nucleic acid binding domains and / or protein dimerization domains, a peptide epitope. This is understood to mean a peptide structure (linear or three-dimensional) to which a specific antibody binds. The amino acids of the peptide epitope can overlap with the other domains (nucleic acid binding domain and / or protein dimerization domain).

A preferred amino acid sequence of the peptide epitope corresponds to an amino acid sequence with at least 90%, preferably at least 95%, particularly preferably at least 99% amino acid sequence identity to the amino acid sequence according to SEQ ID NO. 1.

The invention further includes the use of a linker peptide of the invention as part of a fusion protein.

Preferred fusion proteins, as part of which a linker peptide according to the invention having a nucleic acid binding domain is used, are transport molecules for nucleic acids, in particular siRNA. Likewise preferred fusion proteins, of which part a linker peptide according to the invention having a nucleic acid binding domain is used, are transport molecules for ligands of pattern recognition receptors (PRR), in particular for ligands of toll-like receptors (TLR), particularly preferred for CpG- rich DNA, dsRNA, ssRNA or derived derivatives, in particular imiquimod or resiquimod.

The term "nucleic acids" in the sense of the invention encompasses not only deoxyribonucleic acids (DNA) and ribonucleic acids (RNA) but also all other linear polymers in which the bases adenine (A), cytosine (C), guanine (G) and thymine (T) or uracil (U ) are arranged in a corresponding sequence, in particular nucleic acids with altered backbone, such as. A phosphothioate, phosphoramidate or O-methyl derivatized backbone, peptide nucleic acids (PNA), and locked nucleic acids (LNA).

The invention also encompasses the corresponding RNA sequences in which thymine is replaced by uracil, complementary sequences and sequences with modified nucleic acid backbone or 3 'or 5' terminus. The term modified 3 'or 5' terminus includes both modifications that serve to stabilize as well as the attachment of markers. Examples of markers are enzymes, dyes or fluorescent dyes, radionucleotides, and haptens, such as. B. digoxigenin or biotin. In the case of the enzymes, detection is usually carried out via a color-forming reaction catalyzed by the enzyme. Colored or fluorescent molecules can be directly detected photometrically or fluorometrically. Hapten ligands are usually given a corresponding enzyme- or dye-labeled receptor, which has an affinity for the ligand, brought into contact and detected over it.

Transport molecules are fusion proteins which are used to transport molecules, in this case nucleic acids, in particular selected from siRNA, innate immune ligands, antigens of the adaptive immune system, in particular TLR ligands, to target cells by binding specifically to target cells and followed by a recording of the transported molecules in the target cell. Particularly preferred molecules that are transported by transport molecules is siRNA. Further preferred molecules that are transported are ligands of TLR7, TLR8 and TLR9.

Also included in the invention are fusion proteins which contain at least one linker peptide according to the invention and at least one further protein or peptide as fusion partner. For this purpose, a fusion gene is generated which fuses the nucleic acid sequence of a linker peptide according to the invention with the gene sequence of a protein or peptide (herein: fusion partner) after removal of the stop codon. The translation of such a fusion gene leads to the formation of a fusion protein according to the invention.

Preferred fusion proteins according to the invention contain as fusion partner an antibody or an antibody fragment. A particularly preferred fusion partner in fusion proteins of the invention is a variable region of an antibody.

The term "antibody" in the sense of the invention encompasses all antibodies, fragments and derivatives thereof which are capable of binding specifically to an antigen, including the complete monoclonal antibodies and also the epitope-binding fragments of these antibodies, in which the epitope-binding fragments ( also referred to herein as antibody fragments or antibody derivatives) include all portions of the antibody capable of binding to the antigen, but include, but are not limited to, Fab, Fab ', F (ab') _2, examples of antibody fragments encompassed by the invention , Fd, single-chain variable fragments (scFv), single-chain antibodies, disulfide-linked variable fragments (sdFv), and fragments containing either a light chain variable region (VL) or a heavy chain variable region (VH). Also included are recombinantly produced antibodies, such as diabodies, triabodies, tetrabodies.

Preferably, the antibody carries a marker molecule, such as. Biotin, dioxygenin, a radionuclide or a fluorescent dye.

Antibody fragments contain the variable regions, either alone or in combination with other regions selected from the hinge region, and the first, second, and third Range of constant region (CHI, CH2, CH3). Also encompassed by the term "antibody" are chimeric antibodies in which different portions of the antibody are from different species, such as antibodies having a murine variable region combined with a human constant region.

The term "variable region" is defined according to the invention as the parts of the heavy and light chains of the antibodies, which differ in their sequence between antibodies and which determine the specificity of the antibody and the binding to its antigen.The variability is not uniform in the variable It is usually concentrated within three defined segments of the variable region, called complementarity determining regions (CDRs) or hypervariable regions, present in the variable regions of both the light and heavy chains variable regions are called framework regions The heavy and light chain variable regions contain four framework regions that predominantly adopt a beta-sheet structure, each framework region being linked to three CDRs that form loops that form the beta-fold connect leaf structures and in some cases are part of the beta-sheet structure. The CDRs of the respective chain are brought into close proximity by the framework regions and, together with the CDRs of the other chain, contribute to the formation of the antigen-binding region of the antibodies.

Further preferred fusion proteins contain an antigen as fusion partner. A particularly preferred fusion partner is a bacterial, viral or eukaryotic antigen.

Further preferred fusion proteins according to the invention contain at least two fusion partners, preferably selected from antibodies, parts of antibodies, in particular variable regions of an antibody, antigens, in particular bacterial, viral or eukaryotic antigens, as well as combinations thereof.

If the linker peptide is designed such that it contains a nucleic acid binding domain, a protein dimerization domain, a peptide epitope or combinations thereof, a fusion protein complex can be formed by binding a further molecule to the abovementioned domains of a linker peptide of a fusion protein according to the invention. Nucleic acids are suitable as further molecules for binding to the nucleic acid binding domain and peptides or proteins as further molecules for binding to the protein dimerization domain. For binding to the peptide epitope, specific antibodies or antibody fragments which bind the sequence or structure of the peptide epitope are suitable.

For binding to the protein dimerization domain, fusion proteins according to the invention which contain a linker peptide with a protein dimerization domain are suitable. The two can Fusion proteins according to the invention, thereby forming a dimer as a fusion protein complex, be the same or be constructed differently. Preferably, the fusion proteins of the invention are different in a dimer.

The bonds between the domains and the other molecules is not covalent, but preferably formed from non-covalent bonds between the functional groups of the other molecules with the amino acids of the domains.

Therefore, the invention also encompasses fusion protein complexes containing a fusion protein according to the invention, wherein a further protein or peptide is bound via the protein dimerization domain of the linker peptide and / or a nucleic acid is bound and / or bound via the nucleic acid binding domain of the linker peptide another fusion protein according to the invention is bound to the protein dimerization domain. Furthermore, if appropriate, further molecules, in particular glycolipids or glycoproteins, are indirectly bound via a fusion partner.

Preferred fusion protein complexes according to the invention are characterized in that siRNA is bound to the nucleic acid binding domain.

Preferably, said fusion protein contains an inventive linker peptide having an amino acid sequence of at least 90%, preferably at least 95%, particularly preferably at least 99% amino acid sequence identity to the amino acid sequence according to SEQ ID no. 2, preferably identical to the amino acid sequence according to SEQ ID no. 2. In the fusion protein complex according to the invention formed by means of this fusion protein, siRNA is bound to the fusion protein via the nucleic acid binding domain. Such fusion protein complexes are suitable for transporting siRNA to CD33-positive cells, in particular to CD33-positive tumor cells.

Further preferred fusion protein complexes according to the invention are characterized in that a further fusion protein according to the invention is bound to the protein dimerization domain.

Further preferred fusion protein complexes according to the invention are characterized in that a specific antibody directed against a peptide epitope of the linker peptide according to the invention is bound to the peptide epitope of the linker peptide. This antibody may in turn be part of a fusion protein without inventive linker peptide or a fusion protein according to the invention. Thus, a fusion protein complex according to the invention as a single component of the complex contains at least one fusion protein according to the invention. Particularly preferred antibodies contained in fusion protein complexes of the invention are bispecific antibodies, wherein a variable region of the bispecific antibody is directed against a peptide epitope of the linker peptide of the invention and the other variable region against another antigen, in particular a surface antigen of immune cells, in particular of T cells, dendritic cells, B cells, NK cells or macrophages.

Preferred fusion protein complexes according to the invention comprise a fusion protein according to the invention comprising a linker peptide according to the invention with at least 95%, preferably at least 99%, amino acid sequence identity to SEQ ID no. 1, more preferably SEQ ID NO. Second

Via noncovalent binding, an antibody or antibody derivative is bound to the linker peptide of the invention, wherein the antibody or antibody derivative preferably has variable regions which prefer the following CDR regions or CDR regions having an amino acid sequence of at least 90%, preferably at least 95% at least 99% amino acid sequence identity to the following CDR regions include: a) heavy chain variable region

CDR1 SEQ ID no. 4, CDR2 SEQ ID NO. 5, CDR3 SEQ ID NO. 6, and b) variable region of the light chain

CDR1 SEQ ID no. 7, CDR2 SEQ ID NO. 8, CDR3 SEQ ID NO. 9th

The antibodies or antibody fragments thus defined specifically bind the following sequence of the linker peptide according to the invention: SEQ ID no. 1 .

In principle, all antibodies which specifically bind a linker peptide according to the invention are suitable for the formation of fusion protein complexes with the aid of antibodies. Therefore, the invention also encompasses antibodies or antibody fragments which specifically bind a linker peptide of the invention and the use of these antibodies or antibody fragments in fusion proteins. The invention also encompasses the use of antibodies or antibody fragments according to the invention for the preparation of fusion protein complexes according to the invention.

Particularly preferred antibodies or antibody fragments of the invention have variable regions whose CDR regions have at least 90%, preferably at least 95%, preferably at least 99% amino acid sequence identity to the following CDR regions: heavy chain variable region: CDR1 SEQ ID NO. 4, CDR2 SEQ ID NO. 5, CDR3 SEQ ID NO. 6 light chain variable region: CDR1 SEQ ID NO. 7, CDR2 SEQ ID NO. 8, CDR3 SEQ ID NO. 9th

A particularly preferred antibody or a preferred antibody fragment which specifically binds to a linker peptide according to the invention comprises the CDR regions according to SEQ ID no. 4 to SEQ ID no. 9th

The invention also encompasses a kit for producing fusion protein complexes according to the invention, which comprises the following constituents: a) at least one fusion protein according to the invention, preferably comprising a linker peptide according to the invention having a peptide epitope and / or a nucleic acid binding domain, and b) at least one according to the invention Antibody and / or at least one antibody fragment according to the invention or at least one nucleic acid, preferably siRNA, or at least one further fusion protein according to the invention.

In the kit according to the invention, the components from a) and b) are packaged separately.

Preferred kits according to the invention comprise a fusion protein according to the invention which contains a linker peptide with a nucleic acid binding domain and a nucleic acid, in particular siRNA. Further preferred kits according to the invention comprise a fusion protein according to the invention which contains a linker peptide with a peptide epitope and an antibody according to the invention or an antibody fragment according to the invention. Further preferred kits according to the invention contain two fusion proteins according to the invention which contain a linker peptide with a protein dimerization domain and of which at least one contains a nucleic acid binding domain and a nucleic acid.

Fusion proteins or fusion protein complexes according to the invention are suitable for targeting target cells. Accordingly, fusion proteins or fusion protein complexes according to the invention are suitable for the targeted therapeutic treatment of diseases.

Target cells in the sense of the invention are understood as meaning cells, in particular cells of the human body, to which a fusion protein according to the invention or a fusion protein complex according to the invention can bind. As a result, it is preferable to initiate uptake of the fusion proteins or fusion protein complexes according to the invention into the target cells and subsequently to process the fusion proteins or fusion protein complexes of the target cells. Therefore, the invention also encompasses the use of fusion proteins according to the invention or fusion protein complexes according to the invention for targeting target cells. In preferred uses, the target cells are tumor cells, particularly tumor cells expressing CD33. In further preferred uses, the target cells are immune cells, in particular selected from T cells, natural killer cells (NK cells) and dendritic cells (DCs).

The fusion proteins or fusion protein complexes according to the invention are preferably used for the therapeutic treatment of neoplasms, in particular tumors. More preferably, the inventive fusion proteins or fusion protein complexes are used for the therapeutic treatment of diseases of the immune system, in particular autoimmune diseases or allergies.

Fusion proteins according to the invention are produced recombinantly by the fusion of genes, ie nucleic acid sequences, wherein at least one nucleic acid sequence encodes a linker peptide according to the invention.

Therefore, the invention also encompasses a, preferably isolated, nucleic acid sequence which codes for a linker peptide according to the invention. The invention also encompasses the use of a preferably isolated nucleic acid sequence which codes for a linker peptide according to the invention for the recombinant production of fusion proteins. Preferred nucleic acid sequences according to the invention comprise a nucleic acid sequence with at least 90%, preferably at least 95%, particularly preferably at least 99% nucleic acid sequence identity to SEQ ID no. 10th

Further preferred nucleic acid sequences according to the invention comprise a nucleic acid sequence with at least 90%, preferably at least 95%, particularly preferably at least 99% nucleic acid sequence identity to SEQ ID no. 11th

Further encompassed by the invention is a, preferably isolated, nucleic acid sequence which codes for a fusion protein according to the invention. The invention also encompasses the use of a preferably isolated nucleic acid sequence which encodes a fusion protein according to the invention for the recombinant production of fusion proteins.

For expression of fusion proteins according to the invention, the nucleic acid sequences according to the invention are preferably cloned into an expression vector.

Therefore, the invention further comprises an expression vector containing a nucleic acid sequence according to the invention. Preferably, at least the following components are included: a. at least one nucleic acid sequence according to the invention which codes for a linker peptide according to the invention; and b. optionally at least one nucleic acid sequence encoding a further protein or peptide as a fusion partner.

For the purposes of the invention, an expression vector means a plasmid, virus or other carrier which contains a nucleic acid sequence according to the invention recombinantly by insertion or incorporation. The expression vector typically contains an origin of replication, a promoter, as well as specific gene sequences that allow phenotypic selection of the expression vector-containing host cells.

Further encompassed by the invention is a host cell which recombinantly contains a nucleic acid sequence according to the invention or an expression vector according to the invention.

A host cell is a naturally occurring cell or a transformed or genetically altered cell line or a (multicellular) host organism which contains at least one expression vector according to the invention. The invention encompasses transient transfectants (eg by mRNA injection), host cells or organisms in which at least one expression vector according to the invention is contained as a plasmid or artificial chromosome, and host cells or organisms in which an expression vector according to the invention stably in the genome the host (or individual cells of the host) is integrated. The host cell is preferably selected from prokaryotes or eukaryotes. Preferred prokaryotes are selected from Escherichia coli, Bacillus subtilis. Preferred eukaryotes are yeast cells (eg Saccharomyces cerevisiae, Pichia pastoris), insect cells, amphibious cells or mammalian cells, such as e.g. CHO, HeLa, HEK293. Preferred host organisms are plants, such. B. Maize or tobacco, invertebrates or vertebrates, in particular Bovidae, Drosophila melanogaster, Caenorhabditis elegans, Xenopus laevis, Medaka, Zebrafish or Mus musculus, or cells or embryos of said organisms.

To produce a fusion protein according to the invention, the invention comprises a method which is characterized by the following steps: a. Providing a host cell according to the invention; b. Culturing the host cell under conditions under which expression and optionally secretion of the fusion protein occurs; and c. optionally at least partially purifying the fusion protein. Preferably, host cells are thereby grown (cultured) in a selective medium which selects for the growth of cells containing an expression vector. The expression of the gene sequences of the expression vector results in the production of the fusion protein according to the invention. The expressed fusion proteins are then preferably isolated and purified by any conventional method, including extraction, precipitation, chromatography, electrophoresis or by affinity chromatography.

Since fusion proteins according to the invention or fusion protein complexes according to the invention are suitable for targeting target cells, they are furthermore suitable for therapeutic applications. Particularly suitable therapeutic applications include, but are not limited to, the therapy of malignant neoplasms or diseases of the immune system, preferably each targeting tumor cells or immune cells.

Therefore, the invention also encompasses a pharmaceutical composition comprising a fusion protein or a fusion protein complex according to the invention or a combination thereof in combination with a pharmaceutically acceptable diluent or carrier.

The pharmaceutical compositions of the invention comprise various dosage forms. The pharmaceutical compositions are preferably administered parenterally, more preferably intravenously. In one embodiment of the invention, the parenteral pharmaceutical composition is in a dosage form suitable for injection. Particularly preferred compositions are therefore solutions, emulsions or suspensions of the fusion protein or fusion protein complex of the invention which is present in a pharmaceutically acceptable diluent or carrier.

As pharmaceutically acceptable carriers are preferably used water, buffered water, 0.4% saline, 0.3% glycine and similar solvents. The solutions are sterile. The pharmaceutical compositions are sterilized by conventional, well known techniques. The compositions preferably contain pharmaceutically acceptable excipients, such as those required to give approximately physiological conditions, and / or to increase the stability of the fusion protein or fusion protein complex of the invention, such as pH adjusters and buffering agents, means for Adjustment of toxicity and the like, preferably selected from sodium acetate, sodium chloride, potassium chloride, calcium chloride and sodium lactate. The concentrations of the fusion proteins or fusion protein complexes according to the invention in these formulations are variable depending on the application, they are preferably less than 0.01% by weight, preferably at least 0.1% by weight, more preferably between 1 and 5% by weight and they are primarily based on Fluid volumes, viscosity, etc. selected or in accordance with the respective mode of administration.

The fusion proteins, fusion protein complexes of the invention or the individual molecules constituting the fusion protein complex of the present invention are preferably incorporated into a composition suitable for parenteral administration. Preferably, the pharmaceutical composition is an injectable buffered solution containing between 0.1 to 500 mg / ml of antibody, more preferably between 0.1 to 250 mg / ml of antibody, especially together with 1 to 500 mmol / 1 of buffer. The injectable solution may be in both a liquid and a lyophilized dosage form. The buffer may preferably be histidine (preferably 1 to 50 mM, more preferably 5 to 10 mM) at a pH of 5.0 to 7.0 (most preferably at a pH of 6.0).

Other suitable buffers include, but are not limited to, sodium succinate, sodium citrate, sodium phosphate or potassium phosphate. Preferably, sodium chloride is used between 0 to 300 mM, more preferably 150 mM for a liquid dosage form. For a lyophilized dosage form, the pharmaceutical composition preferably contains cryoprotectants, preferably 0-10% sucrose (more preferably 0.5-1.0%). Other cryoprotectants include trehalose and lactose. For a lyophilized dosage form, the pharmaceutical composition preferably contains swelling agents, preferably 1 to 10% mannitol. Other swelling agents include glycine and arginine. Both in liquid and in lyophilized administration forms stabilizers are preferably used, more preferably between 1 to 50 mM L-methionine (particularly preferably between 5 and 10 mM).

In a preferred embodiment, the pharmaceutical composition comprises inventive fusion proteins or fusion protein complexes at a dose level of 0.1 mg / kg to 10 mg / kg per application. Particularly preferred dose levels include 1 mg / kg body weight.

Pharmaceutical compositions must be sterile and stable under the conditions of manufacture and storage. The composition may be formulated as a solution, microemulsion, dispersion, in liposomes, or other ordered structure suitable for high concentrations of fusion proteins or fusion protein complexes of the invention. Sterile injectable solutions can be prepared by taking up the fusion protein or fusion protein complex in the required amount in a suitable solvent, with or with a combination of the ingredients enumerated above, as needed, followed by filter sterilization. For sterile, lyophilized powders for the preparation of sterile injectable solutions, the preferred methods of vacuum drying are Spray-drying, resulting in a powder of the fusion protein or fusion protein complex of the invention plus any additional desired ingredients from a previously sterile-filtered solution thereof.

The invention also encompasses methods of therapeutic treatment in which a patient is administered a pharmaceutical composition according to the invention. Diseases that are treated by the method according to the invention are preferably malignant neoplasms, diseases of the immune system, infectious diseases or rejection reactions as a result of transplantations; especially preferred are malignant neoplasias. Preferred diseases are: autoimmune diseases, preferably selected from multiple sclerosis, lupus erythematosus, amyotrophic lateral sclerosis (ALS), glomerulonephritis, rheumatoid arthritis, polyarthritis, collagenosis, Sjögren's syndrome, Sharp syndrome, progressive systemic sclerosis (systemic scleroderma), polymyositis, vasculitis , hemolytic anemia, immune thrombocytopenia (ITP), immune neutropenia, type 1 diabetes, Crohn's disease, ulcerative colitis, alopecia, celiac disease, chronic fatigue syndrome, psoriasis, uveitis and thyroiditis; Allergies, in particular selected from allergic asthma, allergic rhinitis, conjunctivitis, neurodermatitis, contact dermatitis, urticaria or anaphylactic shock; Infections, especially viral or bacterial infections; Malignant neoplasms, in particular selected from carcinomas, preferably squamous cell carcinoma, adenocarcinoma, signet-ring cell carcinoma or urothelial carcinoma, sarcomas, neuroendocrine tumors, leukemias and lymphomas or graft versus host disease (GvhD).

Linker peptides according to the invention are flexible and advantageously allow increased stability of fusion proteins, especially when large protein domains are used as fusion partners. As a result, fusion proteins of the invention are stable even with large fusion partners.

The property of linker proteins according to the invention that allows the binding of specific antibodies or else the binding of nucleic acids makes possible the modular construction of fusion protein complexes. With fusion proteins according to the invention it is advantageously possible to have further costimulatory signals, e.g. in the form of nucleic acids, peptides or proteins, in particular signal molecules to transport target cells.

By means of inventive fusion protein complexes, it is advantageously possible, for example, to ensure the transport of different antigens to target cells with the use of a single, recombinant, bispecific antibody. This must be able to bind a linker peptide of the invention on the one hand and an antigen on target cells on the other hand. The production and optimization effort for the transport system is therefore limited to the optimization of the bispecific antibody. By preparation of fusion proteins from a According to the invention, the linker peptide of any antigen can be used to form any number of different fusion protein complexes as transport molecules with a single bispecific antibody. This modular design is superior to previously described systems for transporting molecules to target cells by reducing the manufacturing overhead. The clinical effort involved in the approval of therapeutics must be done only once for a module, which can be used in a variety of ways.

By means of fusion proteins according to the invention, which siRNA is bound to the nucleic acid binding domain of the linker peptide, advantageously siRNA can be transported specifically to desired target cells. The linker peptide mediates the uptake of the siRNA into the target cells, which was previously not possible. The targeted transport of the siRNA by means of the fusion protein complexes according to the invention also advantageously brings about a significant reduction in the amount of siRNA to be used.

Reference to the following figures and embodiments, the invention will be explained in more detail, without limiting the invention to these.

The following abbreviations are used in sequence: E - linker peptide according to the invention, E7B6 - linker peptide according to the invention having an amino acid sequence according to SEQ ID no. 1, E7B6L - linker peptide according to the invention having an amino acid sequence according to SEQ ID no. 2, TZ target cell, TA target antigen, EZ effector cell, GFP green fluorescent protein,

FIG. 1. Three-dimensional structure of the human La protein (A: La motif, B: RRM1, C: RRM2, A:

La motif, B: RRM1, C: RRM2, D: La domain and detailed enlargement of the alpha-helical region containing the sequence regions according to SEQ ID no. 1).

FIG. 2. Schematic representation of the mechanisms underlying a therapeutic application of the linker peptide according to the invention in fusion protein complexes. Fusion proteins contain a linker peptide according to the invention and antibodies or antibody fragments which bind specifically to a target cell (TZ), from left to right: fusion protein complex in which cytostatics are covalently bound to the linker peptide; Fusion protein complex in which nucleic acids, eg siRNA, or TLR ligands are bound to the linker peptide; Fusion protein complex in which toxins are linked to the linker peptide; Fusion protein complex in which radioisotopes are bound to the linker peptide. Schematic representation of the mechanism of action of the invention fusion proteins or fusion protein complexes in the therapy of tumor diseases or diseases of the immune system. Linker peptides according to the invention are either contained once or twice in fusion proteins according to the invention.

A) Scheme of fusion proteins according to the invention. Left: bispecific antibody containing a linker peptide according to the invention, wherein the antibody is directed against effector cell (EZ) and target cell (TZ). Middle: bispecific antibody containing two linker peptides according to the invention and one central GFP, wherein the antibody is directed against effector cell (EZ) and target cell (TZ). Right: Fusion protein complex comprising an effector module and a targeting module. The effector module comprises a bispecific antibody directed against the effector cell and the linker peptide of the invention. The targeting module is an antibody fragment which is directed against a target cell and contains a linker peptide according to the invention.

B) Scheme of the mechanism of action in the therapeutic use of fusion proteins or fusion protein complexes according to the invention by targeting T cells to target cells. a) fusion protein complex comprising a monospecific antibody directed against CD3 and containing a linker peptide of the invention and further comprising a bispecific antibody directed against a target antigen (TA) and the linker peptide according to the invention.

b) fusion protein complex comprising three proteins, one of which contains two linker peptides according to the invention. A monospecific antibody contains two linker peptides E of the invention. Further, the fusion protein complex contains two bispecific antibodies directed against the linker peptide and against a target antigen (TA).

c) fusion protein complex comprising a monospecific antibody directed against a target antigen (TA) on a target cell (TZ) and containing a linker peptide of the invention and further comprising a bispecific antibody directed against a CD3 and the inventive linker peptide.

d) fusion protein complex comprising an antibody directed against a target antigen (TA) on a target cell (TZ) containing a central linker peptide of the invention and further comprising a bispecific antibody directed against a CD3 and the linker peptide of the invention.

Schematic representation of inventive fusion proteins with peptide antigens and their mechanism of action. Left: Fusion protein containing an antibody fragment which is directed against a surface antigen on dendritic cells, a linker peptide according to the invention, a peptide antigen, in particular tetanus toxin. Middle: fusion protein complex consisting of a fusion protein containing a central peptide antigen and two linker peptides; and two bispecific antibodies, wherein the bispecific antibodies each specifically bind the linker peptide of the invention and a surface antigen on dendritic cells. Right: Fusion protein complex which is composed of a fusion protein with three linker peptides according to the invention and three bispecific antibodies. The fusion protein contains between the linker peptides each a GFP and a peptide antigen. The bispecific antibodies each specifically bind the linker peptide of the invention and a surface antigen on dendritic cells.

A and B) Sequences and schematic representation of the structure of fusion proteins according to the invention, containing at least one linker peptide according to the invention according to SEQ ID no. 1 (E7B6) or SEQ ID no. 2 (E7B6L) according to SEQ ID no. 25 - SEQ ID no. 54, as used in the embodiments; C) (a, b) Sequences and schematic representation of the structure of fusion proteins according to the invention, comprising at least one linker peptide according to the invention according to SEQ ID no. 1 (E7B6) according to SEQ ID no. 55 - SEQ ID no. 58, as used in the embodiments, (c, d) sequences and schematic representation of the structure of bispecific antibodies without inventive linker peptide according to SEQ ID NO. 59 - SEQ ID no. 62 for the formation of inventive fusion protein complexes with fusion proteins according to the invention, (e-g) sequences and schematic representation of the structure of bispecific control antibodies according to SEQ ID no. 63 - SEQ ID no. 68 as a comparative example to fusion proteins according to the invention in the form of bispecific antibodies.

Determination of the stability and function of fusion proteins with a linker peptide according to the invention in comparison to fusion proteins with glycine-serine linker (Gly-Ser-linker). A) Schematic representation of the constructs used. Top: Bispecific antibody to CD3 and CD33 with Gly-Ser linker (comparative example). Bottom: Bispecific antibody to CD3 and CD33 with a linker peptide of the invention (E7B6). B) SDS-PAGE / immunoblot of the purified antibodies. C) Specific lysis in the chromium release test with the fusion protein with glycine-serine linker (comparative example) and the fusion protein according to the invention.

Comparison of the function of fusion proteins according to the invention with linker peptide according to the invention and additionally incorporated large protein domains (here: GFP). A) Schematic representation of the construct used, a bispecific antibody against CD3 and CD33 with two linker peptides according to the invention (E7B6) and a central GFP. B) Comparison of the Specific Lysis in the Chromium Release Test with the Fusion Protein According to the Invention from FIG. 6A and the Fusion Protein According to the Invention from FIG. 7A. No difference in functionality was detected. C) (a, c, e) phase contrast exposure; (b, d, f) fluorescence image. (a, b) Effector cells (T cells) and target cells (CD33 positive tumor cells) randomly disperse in the absence of the fusion protein according to the invention. (c, d, e, f) In the presence of the According to the invention of FIG. 7A, the T cells interact with the tumor cells and structures similar to synapses (f, arrowheads) are formed.

(A-F) complex formation with nucleic acids by fusion proteins according to the invention with a linker peptide according to SEQ ID no. 1 determined by complex formation on SPR chips. Complex formation with (A) dsDNA, (B) dsRNA, (C) ssRNA, (D) ssDNA. (E) binding of a specific antibody after complex formation with nucleic acid. (F) Comparative example: Binding of a fusion protein with Gly-Ser-linker, without inventive linker peptide, to ssRNA. (G-H) Protein dimerization of fusion proteins according to the invention. Fusion proteins containing a linker peptide according to SEQ ID no. 1, bind to proteins which also contain a linker peptide according to SEQ ID no. 1 included (G). Equally constructed fusion proteins without inventive linker peptide do not interact with each other (H).

Inhibition of luciferase expression after transport of siRNA to target cells by erfmdungsgemäße fusion proteins after different incubation period. Fusion protein complexes: control siRNA (non-luciferase-specific) with CD33-E7B6L-CD33 (shown as "control siRNA + CD33-E7B6L-CD33", open rectangles); luciferase-specific siRNA with CD33-E7B6L-CD33 (shown as "Luc siRNA + CD33-E7B6L-CD33", black rectangles); Comparative Examples: Transfection of luciferase-specific siRNA (shown as "Luc siRNA transfected"); Addition of luciferase-specific siRNA without transfection (shown as "Luc siRNA").

FACS analysis of the binding of fusion proteins of the invention to: PSCA-positive tumor cells. Binding of fusion proteins scFv PSCA-E7B6 (A) or E7B6L (B) to PSCA-positive tumor cells compared to bispecific antibodies without inventive linker peptide (Comparative Examples CE) CD3xPSCA in tandem (C) or scBsDB (D) format or to anti PSCA mAb (E).

FACS analysis of the binding of fusion proteins of the invention to: CD3-positive T cells. Binding of fusion proteins scBsDB CD3xE7B6 (A) to CD3-positive human T cells in comparison to bispecific antibodies without linker peptide according to the invention (comparative examples) CD3xPSCA in tandem (C) or scBsDB (D) format or to anti-CD3 mAb (B) ,

FACS analysis of the binding of fusion proteins of the invention to: PSCA-positive Hek293T cells whose surface has an amino acid sequence according to SEQ ID no. 1 was decorated. For this, the Hek293T (A) with the fusion protein scFv PSCA-E7B6, (B) with the fusion protein scFv PSCA-E7B6L and the binding of the fusion protein scBsDB CD3xE7B6 according to the invention were determined. The protein can be detected there with an anti-La antibody (D). Also the anti-E7B6 domain of the fusion protein scBsDB CD3xE7B6 binds to the cells decorated with the linker peptide (C) under these conditions.

Fig. 13. Chromium release assay to demonstrate the functionality of fusion protein complexes of the invention. PSCA-positive T cells were isolated from effector T cells in the presence of fusion protein complexes from the effector module scBsDB CD3x7B6 and the target modules anti-PSCA E7B6 (scFvPSCA-E7B6, diagonally hatched) and anti-PSCA E7B6L (scFvPSCA-E7B6L, black-white -dotted) specifically and efficiently lysed. The lysis efficiency of the fusion protein complexes is comparable to conventionally constructed bispecific antibodies (scBsDB CD3xPSCA, vertically hatched). None of the individual components causes unspecific lysis at the effector (E) to target (T) cell ratios of 5: 1 and 20: 1 used.

FIG. 14. Targeting of antigens to dendritic cells by means of inventive fusion protein complexes. Schematic representation of the constructs: (A) bispecific antibody scBsDB DD2x7B6. (C) SDS-side of the purified fusion protein from (A). (D) FACS analysis of PBMCs: (a) slan DCs as a subclass of CD16 positive cells (staining with anti-slan IgM (DD2); (b) binding of monovalent scBsDB DD2x7B6; (c) binding of fusion protein E7B6-TTp- E7B6-GFP-E7B 6 (d) Binding of the fusion protein complex [E7B6-TTp-E7B6-GFP-E7B6] / [DD2x7B6] according to the invention.

Fig. 15. (E) Fluorescence microscopic image of the binding of the fusion protein complex [E7B6-TTp-E7B6-GFP-E7B6] / [DD2x7B6] a n s l a n D C. (a fluorescence; b transmitted light). Since the fluorescence is also intracellular, the fusion protein was picked up from the DC shown. (F) Proliferation of TTp-specific CD4 T cells after incubation of PBMCs of a TTp immunized subject with fusion protein. (F, a) fusion protein E7B6-E7B6-GFP-E7B6 without antigen (comparative example) at 37 ° C, no proliferation; (F, b) E7B6-TTp-E7B6-GFP-E7B6 fusion protein at 37 ° C, TTp specific proliferation; Proliferation of TTp-specific CD4 T cells after incubation with TTp-containing fusion protein complexes. (F, c) fusion protein complex [E7B6-E7B6-GFP-E7B6] / [DD2x7B6] at 4 ° C, no proliferation; (F, d) fusion protein complex [E7B6-TTp-E7B6-GFP-E7B6] / [DD2x7B6] at 4 ° C, proliferation; (F, e) E7B6-TTp-E7B6-GFP-E7B6 fusion protein at 4 ° C, low proliferation.

Example 1: Preparation and Use of Fusion Proteins According to the Invention

For the preparation of a fusion protein according to the invention in accordance with the schematic structure according to FIG. 3A (left), the conventional glycine-serine linker in a bispecific antibody derivative according to FIG. 6A (top) by a linker peptide according to the invention according to SEQ ID No. 1 (E7B6, Fig. 6A, bottom). This construct is shown schematically in Figure 5A and is also referred to herein as "scBsTaFv CD33-E7B6-CD3".

Antibody fragments in the form of variable regions, whose CDRs have the following sequences, were used as fusion partners for the anti-CD33 specificity: heavy chain variable region: CDR1 SEQ ID no. 13, CDR2 SEQ ID NO. 14, CDR3 SEQ ID NO. 15, light chain variable region: CDR1 SEQ ID NO. 16, CDR2 SEQ ID NO. 17, CDR3 SEQ ID no. 18.

The construct according to the invention scBsTaFv CD33-E7B6-CD3 has an amino acid sequence according to SEQ ID no. 30 (Nucleic acid sequence: SEQ ID No. 29). The construct of the comparative example has an amino acid sequence according to SEQ ID no. 68 (Nucleic acid sequence: SEQ ID No. 67).

For expression, the reading frames of the corresponding fusion proteins according to the invention were stably transduced into Hek293T, HeLa or CHO cells or transiently transfected. The cells were cultured at 37 ° C or at 30 ° C in various media (RPMI1640, DMEM) containing different concentrations of FCS. Although the optimal time of harvest of the supernatants was different, but the harvesting of the expression cultures was carried out at the optimal time, then the yields differed only by about a factor of 1.5. Data shown are based on expression in Hek293T cells in 10% FCS. The supernatants of the cells were harvested, optionally concentrated by means of a fractionated ammonium sulfate precipitation (45-75%) and pre-purified (fraction 1 to 45% saturation was discarded, fraction 45 to 75% saturation usually contained the corresponding fusion protein according to the invention) and finally by means of a nickel Purified affinity chromatography. The loading of the affinity columns was carried out by means of PBS buffer containing 10 mM imidazole. After loading, several column volumes of PBS (50 mM imidazole) were washed and the bound fusion proteins were isolated with PBS buffer containing 350 mM imidazole. The isolated proteins were dialyzed overnight at 4 ° C.

After cloning, expression and purification of the bispecific CD33-CD3 antibody (Figure 5C, scBsDB CD33xCD3) with glycine-serine linker, strong fragmentation of the recombinant protein was detected by SDS gel electrophoresis / immunoblots (Figure 6B, scBsDB CD3xCD33). In the fusion protein according to the invention increased stability was found (Fig. 6B).

To determine the cytotoxicity, a chromium release experiment was carried out. Target cells are loaded with radioactive chromium for this purpose. When effector cells (EZ) destroy the target cells (TZ), the chromium is released and can be measured in the supernatant. The fusion protein without linker peptide according to the invention was only weakly cytotoxic in effect (FIG. 6C, truncated rectangles). The fusion protein according to the invention showed a significantly higher cytotoxic effect (FIG. 6C, black rectangles).

The linker peptide according to the invention was multiply incorporated into a fusion protein without this having a negative effect on the stability and functionality. Further advantageously, a linker peptide according to the invention also allows the incorporation of large protein domains, e.g. from CFP. This is not possible when using Gly-Ser linkers.

To demonstrate stability and functionality, a construct according to Fig. 3A (center) was made. The construct is a bispecific tandem antibody directed against CD33 and CD3 and is also referred to herein as "scBsTaFv CD33-E7B6-GFP-E7B6-CD3" (shown in Figure 5A).

The inventive construction scB sTaFv CD33-E7B6-GFP-E7B6-CD3 has an amino acid sequence according to SEQ ID no. 34 (Nucleic acid sequence: SEQ ID No. 33).

Figure 7B shows the cytotoxicity of this construct in the chromium release assay. The incorporation of GFP has no negative effect on functionality. As shown in Figure 7C, the anti-CD33-E7B6-GFP-E7B6-CD33 fusion construct of the present invention can visualize the interaction and formation of immune synapses between effector and target cells.

Example 2: Targeted transport of siRNA to target cells with fusion proteins according to the invention

A fusion protein was prepared which is suitable for the transport of nucleic acids to tumor cells. For this, the nucleic acid sequence according to SEQ ID no. 37 introduced into an expression vector. The fusion protein is a monospecific antibody directed against CD33 which comprises a linker peptide according to the invention of the amino acid sequence according to SEQ ID no. 2 contains. This fusion protein is shown schematically in Figures 2 and 5B and is also referred to herein as "scTaFv CD33-E7B6L CD33." For anti-CD33 specificity, the fusion partner had variable regions whose CDRs have the following sequences: variable region heavy chain: CDR1 SEQ ID No. 13, CDR2 SEQ ID No. 14, CDR3 SEQ ID No. 15, light chain variable region: CDR1 SEQ ID No. 16, CDR2 SEQ ID No. 17, CDR3 SEQ ID No. 18.

The amino acid sequence of the fusion protein scTaFv CD33-E7B6L-CD33 is SEQ ID no. 38th The fusion protein was expressed by transfection or transduction of the expression vectors in host cells analogously to the fusion proteins described in Example 1 and subsequently purified.

In the following, experiments were carried out which demonstrate the nucleic acid binding properties and dimerization properties of the linker peptide according to the invention.

The nucleic acid binding was determined by surface plasmon resonance (SPR) (Figure 8A-E). For this purpose, various nucleic acids (dsDNA, ssDNA, dsRNA, ssRNA) were coupled to a commercially available SPR-CHIP and contacted with fusion proteins according to the invention (containing linker peptide according to SEQ ID No. 2) and fusion proteins without linker peptide according to the invention (comparative example). Fusion proteins of the invention bind with high affinity to nucleic acids of the dsDNA type (FIG. 8A), ssDNA (FIG. 8B), dsRNA (FIG. 8C) and with lower affinity ssRNA (FIG. 8D). Fusion proteins without linker peptide according to the invention (FIG. 8F) do not bind to nucleic acids. Binding to nucleic acids does not hinder the binding of specific antibodies to the linker peptide. Figure 8E shows the binding of an antibody (CDR regions according to SEQ ID Nos. 4-9) that specifically binds to the linker peptide after binding to nucleic acids. The shortest bound nucleic acid was a 15 mer ssDNA.

To demonstrate the dimerization properties of fusion proteins according to the invention, a fusion protein according to the invention (with linker peptide according to SEQ ID No. 2) was coupled to a CHIP and contacted with fusion proteins according to the invention (FIG. 8G) and for a comparative example with a fusion protein without linker peptide according to the invention (FIG. The SPR data show that dimerization occurs only in the presence of the linker peptide of the present invention.

To demonstrate the successful transport of siRNA to target cells, HeLa cells were permanently transduced with the genes encoding the surface receptor CD33 as well as the enzyme luciferase (as expression marker). The fusion protein provided was a fusion protein scTaFv CD33-E7B6L-CD33 according to the invention.

The HeLa cells were transfected with a commercial siRNA specific for the luciferase (Figure 9, "Luc siRNA transfected"). Depending on the time, an inhibition of the luciferase expression is found The same amount of siRNA was complexed via the invention Linker peptide of the fusion protein scTaFv CD33-E7B6L-CD33 transported to the CD33 positive target cells (Figure 9, "Luc siRNA + CD33-E7B6L-CD33"). This also causes an inhibition of the luciferase expression in the target cells, which is even stronger than in the transfection with siRNA. A control siRNA has no effect under the same conditions (Figure 9, "Control siRNA + CD33-E7B6L-CD33"), as well as a simple addition of Luc siRNA without transport vehicle and without transfection (Figure 9, "Luc siRNA"). Thus, a therapeutic application of the fusion proteins of the invention by siRNA is possible, which can now be administered selectively.

Example 3 Targeted Targeting of CD3 T Cells to Tumor Cells Using Fusion Protein Complexes According to the Invention

In the following example, the functionality of fusion protein complexes for the recruitment of effector cells to target cells was investigated. As target antigens (TA), the tumor antigens CD33 and PSCA were selected by way of example. In the exemplary embodiment, CD3 T cells were transported as effector cells to CD33-positive or PSCA-positive target cells (tumor cells). In therapeutic application, the tumor cells are so selectively destroyed by effector cells that have been transported by the fusion protein complex according to the invention to the target cell. The fusion protein complexes used herein contain at least one linker peptide according to SEQ ID no. Second

Here, the fusion protein complexes include an effector module and a targeting module. The effector module serves to bind to the effector cell. The targeting module binds to the target cell.

The following fusion protein complexes were investigated: Targeting module: scFv PSCA-E7B6: Nucleic acid sequence according to SEQ ID no. 49, amino acid sequence according to SEQ ID no. 50, scFv PSCA-E7B6L: Nucleic acid sequence according to SEQ ID no. 51, amino acid sequence according to SEQ ID no. 52

scTaFv PSCA-E7B6L-PSCA: Nucleic acid sequence according to SEQ ID no. 53, amino acid sequence according to SEQ ID no. 54, scFv CD33-E7B6 Nucleic acid sequence according to SEQ ID no. 25, amino acid sequence according to SEQ ID no. 26, scTaFv CD33-E7B6-CD33: Nucleic acid sequence according to SEQ ID no. 27, amino acid sequence according to SEQ ID no. 28

Effector module: scBsDB CD3x7B6: Nucleic acid sequence according to SEQ ID no. 61, amino acid sequence according to SEQ ID no. 62nd

All modules were individually cloned into expression vectors. The modules were recombinantly expressed and purified analogously to Example 1.

To demonstrate the specific binding of the individual modules of the fusion protein complexes to target cells, effector cells and the linker peptide, the binding of the modules to PSCA-positive HEK293T cells (FIG. 10) or primary PBMCs containing CD3-positive T cells (Figure 11), and against the linker peptide on cells (Figure 12).

The binding of the targeting modules scFv PSCA-E7B6 (with amino acid sequence according to SEQ ID No. 50, with a linker peptide according to the invention according to SEQ ID No. 1, FIG. 10A), scFv PSCA-E7B6L (with amino acid sequence according to SEQ ID No. 52, with a linker peptide according to the invention according to SEQ ID No. 2, Fig. 10 B) to PSCA-positive cells was detected by means of FACS analysis. This is comparable to bispecific antibodies without a linker peptide according to the invention (FIG. 10C as a tandem antibody, FIG. 10D as a diabody) and monospecific antibodies against PSCA (FIG. 10E).

The binding of the effector module scBsDB CD3x7B6 (with amino acid sequence according to SEQ ID No. 62) to CD3 T cells (Figure IIA) is comparable to commercially available anti-CD3 (Figure IIB) and bispecific antibodies without linker peptide according to the invention (Figure 11C as Tandem antibody, Fig. 11 D as a diabody).

In addition, the binding of a bispecific antibody directed against the linker peptide of the invention to the fusion proteins scFv PSCA-E7B6 (Figure 12A) and scFv PSCA-E7B6L (Figure 12B) was shown.

Fusion protein complexes were generated by incubation of the two fusion partners (effector module and targeting module). The specific lysis of PSCA-positive tumor cells targeted recruitment of CD3 T cells using inventive fusion protein complexes was determined in the chromium release assay. The results of the chromium release experiment are shown in FIG.

Inventive fusion protein complexes consisting of the effector module scBsDB CD3x7B6 and either the targeting module anti-PSCA E7B6 (Figure 13 scFvPSCA-E7B6, diagonally hatched) or anti-PSCA E7B6L (scFvPSCA-E7B6L, black and white dotted) led to a specific and efficient lysis of PSCA-positive cells. The efficiency of lysis by the fusion protein complexes is comparable to that of conventional bispecific antibodies (Figure 13 scBsDB CD3xPSCA, vertically hatched). None of the individual components, neither the anti-PSCA single chain antibody-linker peptide fusion proteins nor the scBsDB CD3x7B6 alone, resulted in specific lysis of the PSCA-positive cells.

Example 4: Transport of antigen to dendritic cells by fusion protein complexes according to the invention

Fusion protein complexes have been provided which are suitable for transport of the peptide antigen (PA) from tetanus toxin (TT _P ) to dendritic cells (DC). The fusion protein according to the invention contained three linker peptides according to the invention of the sequence according to SEQ ID no. 1. The schematic structure of the fusion protein also referred to herein as "E7B6-TTp-E7B6-GFP-E7B6" is shown in Fig. 14A (amino acid sequence according to SEQ ID No. 56, nucleic acid sequence according to SEQ ID No. 55) a bispecific antibody (Scheme in Figure 14B) directed against the linker peptide and slan, a surface antigen on dendritic cells, is also referred to herein as "7B6xDD2" (amino acid sequence of SEQ ID No. 60, nucleic acid sequence according to SEQ ID No. 59).

Fusion protein (Figure 14C) and fusion partners were cloned, expressed and affinity purified analogously to the fusion protein in Example 1. Fusion protein complexes were generated by incubation of the two fusion partners.

To demonstrate binding to DC, the fusion protein complexes were incubated with PBMCs at 4 ° C. Incubation was performed at 4 ° C to avoid endocytosis of the fusion protein complexes also from other antigen-presenting phagocytes (APCs) in PBMCs.

Fusion protein complexes bind to slan DCs (Figure 14D, d). None of the individual components alone is capable of binding to slanDC (shown for scBsDb DD2xE7B6 in Figure 14D, b); E7B6-TTp-E7B6-GFP-E7B6 in Figure 14D, c). The data show that fusion protein complexes of the invention are suitable for the transport of TTp to slan DCs.

This is confirmed by epifluorescence microscopy of the GFP-tagged fusion protein complexes (Figure 15 E, a). Since fluorescence is also intracellular, these data also show that after binding, the complexes are taken up by the slan DCs.

In order to show that processing and presentation of the PA take place via this transport and the uptake and that an antigen-specific immune reaction (in the case of TTp as an example a reactivation / proliferation of anti-tetanus memory cells) can be triggered, fusion protein Complexes from bispecific scBsDB DD2x7B6 and that shown in FIG. 15 (D) previously described fusion protein (E7B6-TTp-E7B6-GFP-E7B6), thus containing the PA TTp produced and transported to Slan DCs in PBMCs at 4 ° C. In parallel, fusion protein complexes from the bispecific scBsDB DD2x7B6 but also with the fusion protein E7B6-E7B6-GFP-E7B6 (FIG. 5C, amino acid sequence according to SEQ ID No. 58, nucleic acid sequence according to SEQ ID No. 57), which does not contain the PA TTp , but otherwise not different from the fusion protein E7B6-TTp-E7B6-GFP-E7B6, prepared and also transported to Slan DCs in PBMCs at 4 ° C. Unbound complexes were then removed by washing and allowed to adhere the bound fusion protein complexes into the slan DCs by subsequent incubation at 37 ° C. Positive or negative controls were the fusion proteins with and without PA, which were added directly to PBMCs at 37 ° C. As a further negative control, the fusion protein was used with PA, which was first incubated at 4 ° C. with the PBMCs, washed and then likewise incubated at 37 ° C. The data is summarized in FIG. 15F. They show that if a fusion protein such as the fusion protein E7B6-TTp-E7B6-GFP-E7B6 is taken up at 37 ° C by APCs in PBMCs, then the PA contained therein (here TTp) is processed and presented MHC-dependent (here tetanus-specific memory cells ), which then proliferate antigenically and specifically (Figure 15 F, b). Overall, about 7% of its CD4-positive cells proliferate in the donor shown. This proliferation is TTp specific since the same fusion protein lacking only the TTp portion does not result in any specific proliferation (Figure 15F, a). If the multivalent anti-DD2 fusion protein complexes are bound to the slanDCs at 4 ° C and, after removal of the unbound material, the inclusion of the complexes in the slan DCs at 37 ° C is allowed, this also leads to a proliferation of CD4-positive T cells. Cells (Figure 15 F, d). Also, this proliferation is TTp specific since the comparable complexes lacking only the TTp peptide antigen do not cause proliferation (Figure 15F, c). The proliferation of TTp-specific memory cells in Figure 15 (F, d) is not only PA-specific, but also depends on the binding / transport of the fusion protein complexes to the slan DCs. Because fusion proteins, even if they contain the peptide antigen TTp, in contrast to the multivalent anti-DD2 fusion protein complexes bound only marginally at 4 ° C, the largest part can then be washed away and thus does not lead to a significant proliferation of memory cells (Fig 15 F, e).

Overall, these data indicate that fusion protein complexes are suitable for transporting PA to DCs, these complexes are taken up by DCs, processed, and the PAs are presented to MHC-dependent T cells and elicit an antigen-specific immune response.

Quoted Non-Patent Literature: Arai R, Ueda H, Kitayama A, Kamiya N, Nagamune T. Design of the linkers which effectively separate domains of a bifunctional fusion protein. Protein Eng. 2001 Aug; 14 (8): 529-32.

Arnau J, Lauritzen C, Petersen GE, Pedersen J. Current strategies for the use of affinity tags and tag removal for the purification of recombinant proteins. Protein Expr Purif. 2006 Jul; 48 (l): l-13.

Yan W, Imanishi M, Futaki S, Sugiura Y. Alpha-helical linker of an artificial 6-zinc finger. Peptide binding to a discontinuous recognition sequence. Biochemistry. 2007 Jul 24; 46 (29): 8517-24.

Claims

claims

1. Linker peptide containing an amino acid sequence having at least 10 and a maximum of 50 amino acids and at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity, to the amino acid sequence according to SEQ ID NO. 1.

2. Linker peptide according to claim 1, characterized in that the linker peptide has an alpha helical structure.

3. Linker peptide according to claim 1 or 2, characterized in that the linker peptide has a nucleic acid binding domain and / or a protein dimerization domain, the amino acid sequences of the nucleic acid binding domain and the protein dimerization domain possibly overlapping.

4. Linker peptide according to claim 3, characterized in that the linker peptide has an amino acid sequence with at least 90%, preferably at least 95%, more preferably at least 99% amino acid sequence identity to the amino acid sequence according to SEQ ID NO. 2 contains.

Use of a linker peptide as defined in any one of claims 1 to 4 as part of a fusion protein or fusion protein complex.

6. Use according to claim 5, wherein the fusion protein is a transport molecule for nucleic acids, in particular siRNA.

7. fusion protein containing

a. at least one linker peptide as defined in any one of claims 1 to 4; and b. at least one other protein or peptide as fusion partner.

8. fusion protein according to claim 7, characterized in that at least one fusion partner a. ) an antibody or an antibody fragment, in particular a variable region of a

Antibody, and / or

b. ) is an antigen, in particular a bacterial, viral or eukaryotic antigen.

9. fusion protein complex containing a fusion protein according to claim 7 or 8, wherein a further protein or peptide is bound to the linker peptide and / or a nucleic acid is bound via the nucleic acid binding domain of the linker peptide.

An antibody or antibody fragment which specifically binds a linker peptide according to any one of claims 1 to 4 and in which the CDR regions comprise at least 95%, preferably at least 99% amino acid sequence identity to the following CDR regions:

a) variable region of the heavy chain

CDR1 SEQ ID no. 4, CDR2 SEQ ID NO. 5, CDR3 SEQ ID NO. 6, and

b) variable region of the light chain

CDR1 SEQ ID no. 7, CDR2 SEQ ID NO. 8, CDR3 SEQ ID NO. 9th

11. Kit for the preparation of a fusion protein complex, comprising:

a. at least one fusion protein according to claim 7 or 8 and

b. at least one antibody or antibody fragment according to claim 10 or at least one nucleic acid, preferably siRNA or at least one further fusion protein according to claim 7 or 8.

12. Use of a fusion protein according to claim 7 or 8 or a fusion protein complex according to claim 9 for targeting target cells, in particular tumor cells or immune cells, preferably selected from T cells, NK cells and dendritic cells, in the therapeutic treatment of diseases.

A nucleic acid sequence encoding a linker peptide as defined in any one of claims 1 to 4 or a fusion protein as defined in claim 7 or 8 for use in the recombinant production of fusion proteins.

14. An expression vector comprising at least one nucleic acid sequence according to claim 14.

15. A host cell containing a nucleic acid sequence according to claim 13 or an expression vector according to claim 14.

16. A process for the preparation of a fusion protein as defined in claim 7 or 8, characterized by the following steps:

a. Providing a host cell as defined in claim 15;

b. Culturing the host cell under conditions under which expression and optionally secretion of the fusion protein occurs; and

c. optionally at least partially purifying the fusion protein.

A pharmaceutical composition containing a fusion protein as defined in claim 7 or 8 or a fusion protein complex as defined in claim 9 in combination with a pharmaceutically acceptable carrier.