HK1179876B - Modified binding proteins inhibiting the vegf-a receptor interaction - Google Patents

Description

Modified binding proteins that inhibit VEGF-A receptor interactions

Technical Field

The present invention relates to modified recombinant binding proteins specific for VEGF-A, as well as pharmaceutical compositions comprising such proteins, and the use of such proteins for the treatment of tumors and ocular diseases.

Background

Angiogenesis, the growth of new blood vessels in the existing vasculature, is a key process in several pathological states, including tumor growth and ocular diseases, particularly ocular neovascular diseases such as age-related macular degeneration (AMD) or Diabetic Macular Edema (DME) (Carmeliet, p., Nature 438, 932-936, 2005). Vascular Endothelial Growth Factor (VEGF) stimulates angiogenesis and lymphangiogenesis (lymphangiogenesis) by activating VEGF receptor (VEGFR) tyrosine kinases in endothelial cells (Ferrara, n., Gerber, h. p. and lecuter, j., Nature med. 9,669-676, 2003).

The mammalian VEGF family consists of five glycoproteins, known as VEGF-A, VEGF-B, VEGF-C, VEGF-D (also known as fiff) and placental growth factor (PIGF, also known as PGF). VEGF-a has been shown to be an effective target for anti-angiogenic therapy (Ellis, l. m. and Hicklin, d. j., Nature rev. Cancer 8, 579-591, 2008). VEGF-A ligands bind to and activate three structurally similar type III receptor tyrosine kinases, designated VEGFR-1 (also known as FLT1), VEGFR-2 (also known as KDR), and VEGFR-3 (also known as FLT 4). VEGF ligands have different binding specificities for each of these tyrosine kinase receptors, which contributes to their functional diversity. In response to ligand binding, VEGFR tyrosine kinases activate different networks of downstream signaling pathways. VEGFR-1 and VEGFR-2 are found primarily in the vascular endothelium, while VEGFR-3 is found mostly in the lymphatic endothelium. These receptors all have an extracellular domain, a single transmembrane region, and a consensus tyrosine kinase sequence interrupted by a kinase insert. Recently neuropilin (NRP-1), originally identified as a receptor of the neuronal targeting mediator semaphorin/collapsin family, was shown to act as an isoform-specific receptor for VEGF-A.

Is known to passVEGF-AAlternative splicing Generation of eight exons within a GeneOf VEGF-A of (1). All isoforms contain exons 1-5 and the terminal exon, exon 8. Exons 6 and 7, which encode heparin-binding domains, may or may not be included. This results in a family of proteins named according to their amino acid number: VEGF-A165, VEGF-A121, VEGF-A189, etc. However, exon 8 contains two 3' splice sites in the nucleotide sequence that cells can use to generate two families of isoforms of the same length but different C-terminal amino acid sequences (Varey, a.h.r. et al, British j. cancer 98, 1366-1379, 2008). VEGF-Axxx ("xxx" indicates the number of amino acids of the mature protein), a family of pro-angiogenic isoforms, was produced by using the most proximal sequence of exon 8 (resulting in the inclusion of exon 8 a). The recently described anti-angiogenic VEGF-Axxxb isoform is produced by using a distal splice site 66 bp further along the gene from the proximal splice site. This resulted in the excision of exon 8a, producing an mRNA sequence encoding the VEGF-Axxxb family. VEGF-A165 is the dominant pro-angiogenic isoform and is often overexpressed in a variety of human solid tumors. VEGF-A165b was the first identified exon 8 b-encoded isoform shown to have anti-angiogenic effects (Varey et al, supra; Konopatskaya, O. et al, Molecular Vision 12, 626-632, 2006). It is an endogenous inhibitory form of VEGF-A, reducing VEGF-A-induced proliferation and migration of endothelial cells. Although VEGF-A165b binds to VEGFR-2, VEGF-A165b binding does not cause receptor phosphorylation or activation of downstream signaling pathways.

There are several ways to inhibit VEGF-a signaling, including neutralizing the ligand or receptor by antibodies, and blocking VEGF-a receptor activation and signaling with tyrosine kinase inhibitors. VEGF-A targeted therapy has been shown to be effective as a single agent for AMD, DME, renal cell carcinoma and hepatocellular carcinoma, while only beneficial in combination with chemotherapy in patients with metastatic colorectal, non-small cell lung and metastatic breast cancers (Narayanan, R. et al, Nat Rev. Drug Discov. 5, 815-816, 2005; Ellis and Hicklin, supra).

in addition to antibodies, other binding domains may be used to neutralise ligands or receptors (Skerra, A., J. mol. Recog. 13, 167-187, 2000; Binz, H. K., Amstutz, P. and Plu ckthun, A., nat. Biotechnol. 23, 1257-1268, 2005.) one such class of new binding domains is based on the design of repeat domains (WO02/20565; Binz, H. K., Amstutz, P., Kohl, A., Stumpp, M. T., Briand, C., Forrer, P., Gr ü tter, M. G., and Pl ü ckun, A., Nat. Biotechnol. 22, 575-582,2004) WO02/20565, describing how large a library of repeat proteins can be constructed and their use, but no specific VEGF binding motifs are disclosed for the repeat domains, WO02/20565, nor for specific VEGF binding motifs.

Targeting VEGF-a with currently available therapies is not effective in all patients or in all diseases (e.g., EGFR expressing cancers). It is even more and more evident that the concomitant therapeutic benefits of VEGF-a targeted therapy are complex, possibly involving multiple mechanisms (Ellis and Hicklin, supra). For example, anti-VEGF drugs on the market, such as bevacizumab (Avastin) or ranibizumab (Lucentis) (see WO 96/030046, WO 98/045331 and WO 98/045332) or drugs in clinical development, such as VEGF-Trap ® (WO 00/075319), are unable to distinguish between pro-angiogenic and anti-angiogenic forms of VEGF-A, so they are both inhibitory. As a result, they inhibit angiogenesis, but also deprive healthy tissue of the basic survival factor, VEGF-Axxxb, causing cytotoxicity and dose-limiting side effects, which in turn limit efficacy. Common side effects of current anti-VEGF-a therapies are gastrointestinal perforation, bleeding, hypertension, thromboembolic events, and proteinuria (Kamba, t. and McDonald, d.m., Br. j. Cancer 96, 1788-95, 2007). Another commercially available anti-VEGF drug for the treatment of AMD is Pegatanib (WO 98/018480; Macugen @, registered trademark of Pfizer). Pegaptanib is a pegylated anti-VEGF aptamer, i.e., a single-stranded nucleic acid that specifically binds to a target protein. Regarding the treatment of neovascular AMD, a great deal of evidence shows that the eyesight curative effect of Lucentis is better than that of Macugen, and no clear evidence indicates the difference of safety among the medicines. As a result, Macugen @ is an uncommon treatment for the disease.

In sum, there is a need for improved anti-angiogenic agents for the treatment of cancer and other pathological conditions.

The technical problem underlying the present invention is to identify novel anti-angiogenic agents, such as repeat domains with VEGF-Axxx binding specificity, for improved treatment of cancer and other pathological conditions, such as ocular diseases such as AMD or DME. This technical problem is solved by providing the embodiments characterized in the claims.

Summary of The Invention

The present invention relates to recombinant binding proteins comprising an ankyrin repeat domain and a polyethylene glycol moiety of at least 5kDa molecular weight, wherein the ankyrin domain is present at less than 10^-9The Kd of M binds to VEGF-Axxx and inhibits VEGF-Axxx binding to VEGFR-2.

In a preferred embodiment, the polyethylene glycol moiety is coupled to a single Cys residue of the binding domain.

The invention further relates to pharmaceutical compositions comprising one or more of the above-described binding proteins or nucleic acid molecules.

The invention further relates to methods of treating cancer and other pathological conditions, such as ocular diseases such as AMD or DME, with the binding proteins of the invention.

Brief Description of Drawings

FIG. 1 specific dog VEGF-A164 binding of selected designed ankyrin repeat proteins.

The interaction of the selected clones with dog VEGF-A164 (VEGF) and negative control protein (MBP, E.coli maltose binding protein) was shown by crude extraction ELISA. Biotinylated dog VEGF-A164 and MBP were immobilized with NeutrAvidin. Numbers refer to the single DARPin clone selected in ribosome display against dog VEGF-a164 or the corresponding human VEGF-a 165.

A = absorbance. White bars indicate binding to dog VEGF-A164 and black bars show non-specific background binding to MBP.

FIG. 2. inhibition of spheroid outgrowth by selected DARPin.

The length of shoots (sprouts) in the spheroid outgrowth inhibition assay was shown in the presence of various concentrations of (a) DARPin #30 (amino acids 1 to 126 of SEQ ID NO: 4) (DARPin specific for VEGF-Axxx), or (b) DARPin NC, a negative control DARPin that is not specific for VEGF-Axxx.

FIG. 3 specific recognition of VEGF-A isoforms.

Surface Plasmon Resonance (SPR) analysis of binding proteins for VEGF-A isoforms.

(a) And (b) SPR analysis of Avastin ®. 250 nM Avastin were applied to flow cells with fixed dog VEGF-A164 (a) or dog VEGF-A164b (b) for 100 seconds followed by washing with a buffer stream.

(c) And (d) SPR analysis of DARPin #27 (amino acids 1 to 159 of SEQ ID NO: 1). 250 nDMDARPin #27 was applied to flow cells with immobilized dog VEGF-A164 (c) or dog VEGF-A164b (d) for 100 seconds, followed by washing with a buffer stream.

RU = resonance unit.

FIG. 4. effective inhibition of human VEGF-A165 in rabbit eyes.

The vascular leakage rabbit model showed that DARPin had inhibited the efficacy of human VEGF-A165 in eyes compared to Lucentis @. PBS, DARPin #30 or Lucentis @ was applied by intravitreal injection to one eye (treated eye) of each rabbit on day 1. Both eyes of each rabbit were challenged by intravitreal injection of 500 ng human VEGF-A165 on day 4 or day 30. All eyes were evaluated 48 hours after VEGF-A165 injection by measuring the fluorescein content in the vitreous and retina 1 hour after intravenous injection of sodium fluorescein.

R = ratio of fluorescein measurements for treated/untreated eyes. Standard deviations are shown with error bars. 4-PBS = ratio 4 days after PBS injection (control); 4-D = ratio 4 days after DARPin #30 injection; 30-D = ratio 30 days after DARPin #30 injection; 4-L = rate of 4 days after injection of Lucentis ^ C; 30-L = rate of Lucentis ® injection after 30 days.

Detailed Description

Mammalian VEGF-a exists as two alternatively spliced isoform families: (i) a pro-angiogenic "VEGF-Axxx" isoform produced by proximal splicing of exon 8 and (ii) an anti-angiogenic "VEGF-Axxxb" isoform produced by distal splicing of exon 8. Preferably, the binding domains of the invention are specific for pro-angiogenic VEGF-Axxx of dog, rabbit, monkey or human origin. More preferably, the binding domains of the invention are specific for human derived pro-angiogenic VEGF-Axxx. Most preferably, the binding domains of the invention are specific for human VEGF-A165.

The term "protein" refers to a polypeptide, wherein at least part of the polypeptide has or is capable of attaining a defined three-dimensional arrangement within and/or between its polypeptide chains by forming a secondary, tertiary or quaternary structure. If the protein comprises two or more polypeptides, each polypeptide chain may be non-covalently linked or covalently linked, for example by a disulfide bond between the two polypeptides. Protein portions, each having or capable of obtaining a defined three-dimensional arrangement by forming secondary or tertiary structures, are referred to as "protein domains". Such protein domains are well known to those skilled in the art.

The term "recombinant" as used in recombinant proteins, recombinant protein domains, and the like, means that the polypeptide is prepared by using recombinant DNA techniques well known to those skilled in the relevant art. For example, a recombinant DNA molecule encoding a polypeptide (e.g., prepared by gene synthesis) can be cloned into a bacterial expression plasmid (e.g., pQE30, Qiagen). When the recombinant expression plasmid thus constructed is inserted into a bacterium (e.g., E.coli), the bacterium can produce a polypeptide encoded by such a recombinant DNA. The correspondingly produced polypeptide is referred to as recombinant polypeptide.

The term "polypeptide tag" refers to an amino acid sequence attached to a polypeptide/protein, wherein the amino acid sequence is useful for purifying, detecting or targeting the polypeptide/protein, or wherein the amino acid sequence improves the physicochemical behavior of the polypeptide/protein, or wherein the amino acid sequence has effector function. The various polypeptide tags, portions and/or domains of the binding protein may be linked to each other directly or via a polypeptide linker. These polypeptide tags are well known in the art and are fully available to those skilled in the art. Examples of polypeptide tags are small polypeptide sequences, e.g. His, myc, FLAG or Strep-tags or moieties such as enzymes (e.g. enzymes like alkaline phosphatase), or moieties that can be used for targeting (such as immunoglobulins or fragments thereof) and/or as effector molecules, allowing the detection of the polypeptide/protein.

The term "polypeptide linker" refers to an amino acid sequence that is capable of linking, for example, two protein domains, a polypeptide tag and a protein domain, a protein domain and a non-polypeptide moiety such as polyethylene glycol, or two sequence tags. Such additional domains, tags, non-polypeptide moieties and linkers are known to those of skill in the relevant art. A list of examples is provided in the description of patent application WO 02/20565. Specific examples of such linkers are glycine-serine-linkers and proline-threonine-linkers of various lengths; preferably the linker has a length of 2 to 24 amino acids; more preferably the linker has a length of 2 to 16 amino acids.

In the context of the present invention, the term "polypeptide" relates to a molecule consisting of one or more chains of multiple (i.e. two or more) amino acids linked via peptide bonds. Preferably the polypeptide consists of more than eight amino acids linked via peptide bonds.

The term "polymer moiety" refers to a proteinaceous polymer moiety or a non-proteinaceous polymer moiety. A "proteinaceous polymer moiety" is preferably a polypeptide which does not form a stable tertiary structure and at the same time does not form more than 10% (preferably not more than 5%; also preferably not more than 2%; even more preferably not more than 1%; most preferably no detectable amount, as determined by Size Exclusion Chromatography (SEC)) oligomers or aggregates when stored in PBS at a concentration of about 0.1 mM for 1 month at room temperature. When using globulin as a molecular weight standard for SEC, such proteinaceous polymer moieties operate in SEC at an apparent molecular weight that is higher than its effective molecular weight. Preferably the apparent molecular weight of the proteinaceous polymer moiety as determined by SEC is 1.5x, 2x or 2.5x of its effective molecular weight calculated from its amino acid sequence. It is also preferred that the apparent molecular weight of the non-proteinaceous polymer moiety, as determined by SEC, is 2x, 4x or 8x of its effective molecular weight, calculated from its molecular composition. Preferably, the proteinaceous polymer moiety does not form a stable secondary structure at room temperature in PBS at a concentration of about 0.1 mM over 50%, 70% or even 90% amino acids, as determined by Circular Dichroism (CD) measurements. Most preferably, the proteinaceous polymer exhibits a typical near-UV CD-profile of a random coil conformation. Such CD analysis is well known to those skilled in the art. Also preferred are proteinaceous polymer moieties consisting of more than 50, 100, 200, 300, 400, 500, 600, 700 or 800 amino acids. Examples of proteinaceous polymer moieties are XTEN (Amunix registered trademark; WO 07/103515) polypeptides, or polypeptides comprising proline, alanine and serine residues, as described in WO 08/155134. Such proteinaceous polymer moieties may be covalently linked, for example, to a binding domain of the invention by using standard DNA cloning techniques followed by standard expression and purification thereof to produce a gene fusion polypeptide. Examples of binding proteins comprising a repeat domain that binds VEGF-Axxx and such a proteinaceous polymer moiety are shown in SEQ ID NO 1 and SEQ ID NO 4. Amino acids 1 to 159 of SEQ ID NO. 1 correspond to the repeat domain and amino acids 161 to 1' 025 of SEQ ID NO. 1 correspond to the proteinaceous polymer moiety. Amino acids 1 to 126 of SEQ ID NO. 4 correspond to the repeat domain and amino acids 131 to 640 of SEQ ID NO. 4 correspond to the proteinaceous polymer moiety.

The polymer portion of the present invention can vary widely in molecular weight (i.e., from about 1 kDa to about 150 kDa). Preferably the polymer portion has a molecular weight of at least 2, 5, 10, 20, 30, 50, 70 or 100 kDa.

Preferably the polymer moiety is linked to the binding domain by a polypeptide linker. Examples of such polypeptide linkers are amino acids 1 to 8 of SEQ ID NO 8 and SEQ ID NO 9.

Examples of non-proteinaceous polymer moieties are hydroxyethyl starch (HES), polyethylene glycol (PEG), polypropylene glycol or polyalkylene oxide. The term "pegylated" means that a PEG moiety is covalently attached to, for example, a polypeptide of the invention. Examples of repeat proteins containing a polypeptide linker between the repeat domain and the C-terminal Cys residue for binding to the non-proteinaceous polymer moiety are SEQ ID NOs 2, 3, 5, 6 and 7.

In particular embodiments, the PEG moiety or any other non-proteinaceous polymer may be coupled to the thiol group of a cysteine, for example, via a maleimide linker, which is coupled to the N-or C-terminus of a binding domain as described herein (e.g., SEQ ID NO:3) via a peptide linker.

The term "binding protein" refers to a protein comprising one or more binding domains and one or more polymer moieties, as explained further below. Preferably the binding protein comprises up to four binding domains. More preferably the binding protein comprises at most two binding domains. Most preferably the binding protein comprises only one binding domain. Moreover, any such binding protein may comprise additional protein domains, multimerization moieties, polypeptide tags, polypeptide linkers, and/or a single Cys residue that is not a binding domain. Examples of multimerizing moieties are immunoglobulin heavy chain constant regions (which pair to provide a functional immunoglobulin Fc domain), and leucine zippers or polypeptides that contain free sulfhydryl groups that form intermolecular disulfide bonds between two such polypeptides. A single Cys residue may be used to conjugate other moieties to the polypeptide, for example, by using maleimide chemistry, which is well known to those skilled in the art.

Preferably the binding protein comprises up to four polymer moieties. More preferably, the binding protein comprises at most two polymer moieties. Most preferably, the binding protein comprises only one polymer moiety.

It is also preferred that the binding protein has an apparent molecular weight of at least 70, 100, 200, 300, 500 or 800 kDa when analyzed by SEC as a molecular weight standard in PBS at a concentration of 0.1 mM at room temperature.

The term "binding domain" means a protein domain that exhibits the same "folding" (three-dimensional arrangement) as the protein scaffold and has predetermined properties as defined below. Such binding domains can be obtained by rational, or most often combinatorial, protein engineering techniques (techniques known in the art) (Skerra, 2000, supra; Binz et al, 2005, supra). For example, a binding domain having a predetermined property may be obtained by a method comprising the steps of: (a) providing a collection of distinct protein domains exhibiting the same folding as the protein scaffold as defined further below; and (b) screening and/or selecting from said different repertoires to obtain at least one protein domain having said predetermined property. The collection of different protein domains may be provided by several methods depending on the screening and/or selection system used, and may comprise the use of methods well known to those skilled in the art, such as phage display or ribosome display.

The term "protein scaffold" means a protein having an exposed surface region in which amino acid insertions, substitutions or deletions are highly tolerated. Examples of protein scaffolds that can be used to generate binding domains of the invention are antibodies or fragments thereof such as single chain Fv or Fab fragments, from Staphylococcus aureus (S. aureus)Staphylococcus aureus) Protein A of (A), derived from Pieris brassicae (A)Pieris brassicae) The bile pigment binding protein or other lipocalin, ankyrin repeat protein or other repeat protein, and human fibronectin. Protein scaffolds are known to those skilled in the art (Binz et al, 2005, supra; Binz et al, 2004, supra).

The term "predetermined property" refers to properties such as binding to a target, blocking a target, activating a target-mediated reaction, enzymatic activity, and related other properties. Depending on the type of property desired, one of ordinary skill will be able to identify the format and necessary steps for performing the screening and/or selecting binding domains with the desired property. Preferably the predetermined property is binding to a target.

Preferably, the binding protein of the invention is not an antibody or a fragment thereof, such as a Fab or scFv fragment. Antibodies and fragments thereof are well known to those skilled in the art.

immunoglobulin folds are common full beta protein folds, consisting of a 2-layer sandwich of about 7 antiparallel beta-chains arranged in two β folds.

It is further preferred that the binding domains of the invention do not comprise an immunoglobulin-like domain as found in VEGFR-1 or VEGFR-2. Such binding domains are described in WO 00/075319.

Preferred binding domains are those with anti-angiogenic effects. The anti-angiogenic effect of the binding domain can be determined by assays well known to those skilled in the art, such as the HUVEC spheroid sprouting assay described in example 2.

Further preferred are binding domains comprising 70 to 300 amino acids, in particular 100 to 200 amino acids.

Further preferred is a binding domain lacking free Cys residues. Free Cys residues are not involved in the formation of disulfide bonds. Even more preferably a binding domain which does not contain any Cys residues.

Preferred binding domains of the invention are repeat domains or designed repeat domains, preferably as described in WO 02/20565.

Particularly preferred binding domains are designed ankyrin repeat domains (Binz, h. k. et al, 2004, supra), preferably designed ankyrin repeat domains as described in WO 02/20565. Examples of designed ankyrin repeat domains are shown in the examples.

The following definitions of the repeat proteins are based on patent application WO 02/20565. Patent application WO02/20565 further contains a general description of repetitive protein features, techniques and applications.

The term "repeat protein" refers to a protein comprising one or more repeat domains. Preferably each of said repeat proteins comprises up to four repeat domains. More preferably each of said repeat proteins comprises at most two repeat domains. Most preferably, each repeat protein comprises only one repeat domain. Furthermore, the repeat protein may comprise additional non-repeat protein domains, polypeptide tags and/or polypeptide linkers.

The term "repeat domain" refers to a protein domain comprising two or more consecutive repeat units (modules) as structural units, wherein the structural units have identical folds, closely stacked to create, for example, a supercoiled structure with a bonded hydrophobic core.

The terms "designed repeat protein" and "designed repeat domain" refer to the repeat protein or repeat domain, respectively, obtained by the inventive procedure explained in patent application WO 02/20565. Designed repeat proteins and designed repeat domains are synthetic and not natural. They are respectively artificial proteins or domains obtained by expression of correspondingly designed nucleic acids. Preferably, expression is carried out in eukaryotic or prokaryotic cells, such as bacterial cells, or by using a cell-free in vitro expression system.

The term "structural unit" refers to a partially ordered portion of a polypeptide formed by three-dimensional interaction between two or more secondary structural segments that are adjacent to each other along the polypeptide chain. Such building blocks represent structural motifs. The term "structural motif refers to the three-dimensional arrangement of secondary structural elements present in at least one structural unit. Structural motifs are well known to those skilled in the art. The individual building blocks do not allow a defined three-dimensional arrangement to be obtained; however, their sequential arrangement, for example as repeating modules of repeating domains, leads to mutual stabilization of adjacent units, resulting in a supercoiled structure.

The term "repeat unit" refers to an amino acid sequence comprising a repeat sequence motif of one or more naturally occurring repeat proteins, wherein said "repeat unit" is present in multiple copies, representing a defined folding topology common to all said motifs determining the folding of the protein. Examples of such repeat units are armadillo repeat units, leucine-rich repeat units, ankyrin repeat units, tetratricopeptide repeat units, HEAT repeat units and leucine-rich variant repeat units. Naturally occurring proteins containing two or more such repeat units are referred to as "naturally occurring repeat proteins". The amino acid sequences of the individual repeat units of a repeat protein may have a significant number of mutations, substitutions, additions and/or deletions when compared to one another, while still substantially retaining the general pattern or motif of the repeat unit.

Preferred repeat units for deriving repeat sequence motifs are homologous repeat units obtained from repeat domains selected on the target (as described in example 1) and having the same target specificity.

The term "repeat sequence motif refers to an amino acid sequence deduced from one or more repeat units. Preferably the repeat units are from repeat domains having binding specificity for the same target.

the folding topology is determined by amino acid extensions forming at least part of the α -helix or β fold, or amino acid extensions forming a linear polypeptide or loop, or any combination of alpha-helix, β fold and/or linear polypeptide/loop.

The term "continuous" refers to an arrangement in which repeating units or repeating modules are arranged in series. In the repeat protein designed, there are at least 2, usually about 2 to 6, especially at least about 6, often 20 or more repeat units. In most cases, the repeat units will exhibit a high degree of sequence identity (identical amino acid residues at corresponding positions) or sequence similarity (amino acid residues are different but have similar physicochemical properties), and some amino acid residues may be key residues that are well conserved among the different repeat units present in the naturally occurring protein. However, a high degree of sequence variability due to amino acid insertions and/or deletions and/or substitutions between different repeating units present in a naturally occurring protein will be possible as long as the common folding topology is maintained.

The person skilled in the art is familiar with methods for directly determining the folding topology of repetitive proteins by physicochemical means such as X-ray crystallography, NMR or CD spectroscopy. Methods for identifying and determining repeat units or repeat sequence motifs, or identifying related protein families comprising such repeat units or motifs, are well established in the field of bioinformatics and are well known to those skilled in the art, such as homology searches (BLAST et al). The step of refining the initial repeat sequence motif may comprise an iterative process.

The term "repeat module" refers to the repeating amino acid sequence of a designed repeat domain, originally derived from a repeat unit of a naturally occurring repeat protein. Each repeat module contained within a repeat domain is derived from one or more repeat units of a family or subfamily of naturally occurring repeat proteins (e.g., a family of armadillo repeat proteins or ankyrin repeat proteins).

A "repeat module" can comprise positions having amino acid residues present in all copies of the corresponding repeat module ("fixed positions") and positions having different or "randomized" amino acid residues ("randomized positions").

The term "capping module" refers to a polypeptide fused to an N-terminal or C-terminal repeat module of a repeat domain, wherein the capping module forms a tight tertiary interaction with the repeat module, thereby providing a cap that separates the hydrophobic core of the repeat module from the solvent on the side not in contact with consecutive repeat modules. The N-terminal and/or C-terminal capping module may be, or may be derived from, a capping unit adjacent to the repeat unit or other domain found in naturally occurring repeat proteins. The term "capping unit" refers to a naturally occurring, folded polypeptide, wherein the polypeptide defines a specific structural unit fused to a repeating unit at the N-or C-terminus, wherein the polypeptide forms an intimate tertiary interaction with the repeating unit, thereby providing a cap that separates the hydrophobic core of the repeating unit on one side from the solvent. Such capping units may have sequence similarity to the repeat sequence motif. The capping module and the capping are described repeatedly in WO 02/020565. For example, the N-terminal capping module of SEQ ID NO. 2 is encoded by amino acids 1 to 32. Such N-terminal capping modules with a glycine or aspartic acid residue at position 5 are also preferred.

The term "target" refers to an individual molecule such as a nucleic acid molecule, polypeptide or protein, carbohydrate, or any other naturally occurring molecule, including any portion of such individual molecule, or a complex of two or more such molecules. The target may be a whole cell or tissue sample, or may be any non-native molecule or moiety. Preferred targets are naturally occurring or non-naturally occurring polypeptides or polypeptides containing chemical modifications, for example by natural or non-natural phosphorylation, acetylation or methylation modifications. In particular applications of the invention, the target is VEGF-Axxx or VEGFR-2.

The term "consensus sequence" refers to an amino acid sequence, wherein the consensus sequence is obtained by structural and/or sequence alignment of multiple repeat units. Using two or more structurally and/or sequence aligned repeat units and allowing gaps in alignment, the most common amino acid residues at each position can be determined. A consensus sequence is a sequence comprising the most frequently occurring amino acids at each position. In the event that two or more amino acids occur at a single position in excess of an average, the consensus sequence may comprise a subset of those amino acids. The two or more repeat units may be taken from repeat units contained in a single repeat protein, or from two or more different repeat proteins.

Consensus sequences and methods for their determination are well known to those skilled in the art.

A "consensus amino acid residue" is an amino acid found at a position in a consensus sequence. If two or more, such as three, four or five, amino acid residues are found in the two or more repeat units with similar probability, the consensus amino acid may be one of the most common amino acids or a combination of the two or more amino acid residues.

Further preferred are non-naturally occurring binding proteins or binding domains.

The term "non-naturally occurring" means synthetic or not from natural sources, and more specifically the term means made by the human hand. The term "non-naturally occurring binding protein" or "non-naturally occurring binding domain" means that the binding protein or the binding domain is synthetic (i.e., prepared from amino acids by chemical synthesis) or recombinant and is not from nature. A "non-naturally occurring binding protein" or "non-naturally occurring binding domain" is an artificial protein or domain, respectively, obtained by expression of a correspondingly designed nucleic acid. Preferably, expression is carried out in eukaryotic or bacterial cells, or by using a cell-free in vitro expression system. Furthermore, the term means that the sequence of said binding protein or said binding domain is not presented as a non-artificial sequence entry in a sequence database, such as GenBank, EMBL-Bank or Swiss-Prot. These databases and other similar sequence databases are well known to those skilled in the art.

The binding domain inhibits binding of VEGF-Axxxx to VEGFR-2 by binding to VEGF-Axxx or by binding to VEGFR-2, resulting in an apparent dissociation constant (K) between VEGF-Axxx and VEGFR-2_d) Increase by more than 10²Times, preferably more than 10³Times, more preferably more than 10⁴Times, more preferably more than 10⁵Times, most preferably more than 10⁶And (4) doubling. Preferred binding domains are K that interact with VEGF-Axxx or VEGFR-2_dLess than 10^-7M, preferably less than 10^-8M, more preferably less than 10^-9M, more preferably less than 10^-10M, most preferably less than 10^-11And M. Methods for determining the dissociation constant for protein-protein interactions, such as Surface Plasmon Resonance (SPR) -based techniques, are well known to those skilled in the art.

Preferably the binding domain binds to VEGF-Axxx. Even more preferably binding domain of human VEGF-A165.

The term "PBS" means an aqueous phosphate buffered solution containing 137 mM NaCl, 10 mM phosphate, and 2.7 mM KCl, and having a pH of 7.4.

Preferably binding proteins and/or binding domains that do not lose their native three-dimensional structure after incubation in PBS containing 100 mM Dithiothreitol (DTT) for 1 or 10 hours at 37 ℃.

In a particular embodiment the invention relates to binding proteins comprising a binding domain that inhibits the binding of VEGF-Axxx to VEGFR-2 and has an indicated or preferred midpointDenaturation temperature and non-aggregating properties as defined above, wherein the binding protein has an IC of less than 100 nM₅₀Values inhibited sprouting of HUVEC spheroids.

The term "HUVEC" means human umbilical vein endothelial cells that can be isolated from normal human umbilical veins and that respond to VEGF-A stimulation. Assays for measuring the germination of HUVEC spheroids, such as the assay described in example 2, are well known to those skilled in the art.

IC₅₀Values are the concentration of a substance such as a binding protein or binding domain required for 50% in vitro inhibition of an experimentally determined parameter such as HUVEC spheroid sprouting. IC can be readily determined by one skilled in the art₅₀Values (Korff T. and AugustinH.G., J. Cell biol. 143(5), 1341-52, 1998).

Preferably at an IC of less than 10 nM, preferably less than 1 nM, more preferably less than 0.1 nM, most preferably less than 0.05 nM₅₀Binding proteins and/or binding domains that inhibit HUVEC spheroid germination.

Further preferred are monomer binding proteins and/or binding domains, IC's thereof, that inhibit the sprouting of HUVEC spheroids₅₀The values are lower than those of IC corresponding to Rankine monoclonal antibody (Lucentis, Genettech registered trademark), Bevacizumab (Avastin, Genetch registered trademark), Abutilizept (Aflibercept) (VEGF Trap, Regeneron registered trademark) or Pegamy He (Macugen, Pfizer registered trademark)₅₀The value is obtained.

Preferred binding domains are K interacting with VEGF-B, VEGF-C, VEGF-D, PIGF or PDGF_dGreater than 1 nM, preferably greater than 10 nM, more preferably greater than 10 nM²nM, even more preferably higher than 10³nM, most preferably higher than 10⁴nM。

Preferably, VEGF-Axxx is dog VEGF-A164 or simian VEGF-A165 or human VEGF-A165 and VEGF-Axxxb is dog VEGF-A164b or simian VEGF-A165b or human VEGF-A165 b.

Another preferred embodiment is a recombinant binding protein comprising a binding domain, wherein said binding domain inhibits binding of VEGF-Axxx to VEGFR-2, wherein said binding domain is a repeat domain or a designed repeat domain. Such repeat domains may comprise 1, 2, 3 or more internal repeat modules involved in binding VEGF-Axxx. Preferably such repetitive domains comprise an N-terminal capping module, 2 to 4 internal repeating modules and a C-terminal capping module. Preferably the binding domain is an ankyrin repeat domain or designed ankyrin repeat domain.

Preferably the recombinant binding protein comprises a binding domain as described herein conjugated to a polyethylene glycol (PEG) moiety, preferably wherein said PEG moiety is coupled to a single Cys residue of said binding domain. Preferably the Cys residue is genetically introduced at the C-terminus of the binding domain. The PEG moiety can then be coupled by chemical means (e.g., by using maleimide chemistry, which is well known to those skilled in the art). Examples of such binding proteins comprising a PEG moiety conjugated to a single Cys residue are given in the examples.

preferred embodiments of the present invention include recombinant binding proteins comprising a binding domain as described herein, wherein the binding domain is conjugated at its C-terminus via a peptide bond to SEQ ID NO:8, SEQ ID NO:8 in turn conjugated at the C-terminal cysteine thiol group to a maleimide-coupled PEG, such as α - [3- (3-maleimido-1-oxopropyl) amino ] propyl- ω -methoxy-polyoxyethylene (NOF, Sunbright ME-200MA (20kD) or Sunbright ME-400MA (40 kD)). in one embodiment α - [3- (3-maleimido-1-oxopropyl) amino ] propyl- ω -methoxy-polyoxyethylene has a molecular weight of at least about 2, 5, 10, 20, 30, 40, 50, 70 or 100 kD.

Another preferred embodiment is a recombinant binding protein as defined above comprising at least one repeat domain having binding specificity for VEGF-Axxx, wherein said repeat domain competes for binding to VEGF-Axxx with a repeat domain selected from SEQ ID NOs 1 to 7. Preferably, the repeat domain competes with the repeat domain of SEQ ID NO:1 or 3 for binding to VEGF-Axxx. More preferably, the repeat domain competes with the repeat domain of SEQ ID NO 3 for binding to VEGF-Axxx.

The term "competitive binding" means that two different binding domains of the invention are not capable of binding to the same target at the same time, but each is capable of binding to the same target. Thus, such two binding domains compete for binding to the target. Methods of determining whether two binding domains compete for binding to a target, such as competitive ELISA or competitive SPR measurements (e.g., by using the Proteon instrument from BioRad), are well known to those skilled in the art.

Recombinant binding proteins that compete with the selected repeat protein for binding to VEGF-Axxx can be determined by methods well known to those skilled in the art, such as competitive enzyme-linked immunosorbent assay (ELISA).

Another preferred embodiment is a recombinant binding protein comprising a repeat domain with binding specificity for VEGF-Axxx selected from the group consisting of the repeat domains of SEQ ID NOS: 1 to 7. Preferably, the repeat domain is selected from the repeat domains of SEQ ID NO 2 or 3. More preferably, the repeat domain is that of SEQ ID NO 3.

One or more polyethylene glycol moieties may be attached at various positions on the binding protein, and such attachment may be achieved by reaction with an amine, thiol, or other suitable reactive group. Attachment of the polyethylene glycol moiety (pegylation) can be site-directed, wherein a suitable reactive group is introduced into the protein to create a site where pegylation preferentially occurs, or is otherwise originally present in the binding protein. A sulfhydryl group may be present in a cysteine residue; the amine group may be, for example, a primary amine found at the N-terminus of a polypeptide or an amine group present in a side chain of an amino acid such as lysine or arginine. In a preferred embodiment, the binding protein is modified to have a cysteine residue at the desired position, allowing site-directed pegylation at the cysteine, e.g. by reaction with a polyethylene glycol derivative carrying a maleimide functional group. The molecular weight of the polyethylene glycol moiety can vary widely (i.e., from about 1 kDa to about 100 kDa), and the polyethylene glycol moiety can be branched or linear. Preferably, the polyethylene glycol has a molecular weight of from about 1 to about 50kDa, preferably from about 10 to about 40kDa, even more preferably from about 15 to about 30kDa, most preferably about 20 kDa.

In yet another embodiment, the invention relates to nucleic acid molecules encoding specific recombinant binding proteins. Moreover, vectors comprising said nucleic acid molecules are contemplated.

Furthermore, a pharmaceutical composition is contemplated comprising one or more of the above-mentioned binding proteins, in particular recombinant binding proteins comprising a repeat domain, or a nucleic acid molecule encoding a particular recombinant binding protein, and optionally a pharmaceutically acceptable carrier and/or diluent. Pharmaceutically acceptable carriers and/or diluents are known to those skilled in the art and are explained in more detail below. Even further, diagnostic compositions comprising one or more of the above-described recombinant binding proteins, in particular binding proteins comprising a repeat domain, are contemplated.

The binding proteins of the invention suppress or prevent VEGF-induced pathological angiogenesis, vascular leakage (edema), pulmonary hypertension, neoplasia and/or inflammatory disorders. By "repressed" is understood that the recombinant protein prevents the pathology to a certain extent, such as to 10% or 20%, more preferably 50%, in particular 70%, 80% or 90%, or even 95%.

The term "edema" means a condition caused by vascular leakage. Vasodilation and increased permeability during inflammation may be the major pathogenesis. For example, edema contributes to post-stroke infarct extension, which can cause life-threatening intracranial hypertension in cancer patients. Furthermore, extravasation of plasma proteins facilitates the metastatic spread of ocular tumors, and airway congestion can cause fatal asthma attacks. Increased vascular leakage that occurs during inflammation can lead to respiratory distress, ascites, peritoneal sclerosis (in dialysis patients), adhesion formation (abdominal surgery) and metastatic spread.

The term "angiogenesis" means the fundamental process by which new blood vessels are formed. The major angiogenic phase in humans occurs in the first three months of embryonic development, but angiogenesis also occurs as a normal physiological process during tissue growth, such as muscle or fat gain and the menstrual cycle and pregnancy.

The term "pathological angiogenesis" refers to the formation and growth of blood vessels during the maintenance and progression of a disease state. Specific examples of pathological angiogenesis are found in blood vessels (atherosclerosis, hemangioma, vascular endothelioma), bones and joints (rheumatoid arthritis, synovitis, bone and cartilage destruction, osteomyelitis, pannus growth, osteophyte formation, neoplasms and metastases), skin (warts, pyogenic granuloma, hair growth, kaposi's sarcoma, keloids, allergic edema, neoplasms), liver, kidney, lung, ear and other epithelia (inflammatory and infectious processes including hepatitis, glomerulonephritis, pneumonia; and asthma, nasal polyps, otitis, transplant disorders, liver regenerative disorders, neoplasms and metastases), uterus, ovaries and placenta (dysfunctional uterine bleeding due to intrauterine contraceptive devices, follicular cyst formation, ovarian hyperstimulation syndrome, endometriosis, neoplasms), brain, nerves and eyes (retinopathy of prematurity), premature retinopathy, Diabetic retinopathy, choroid and other intraocular disorders, leukomalacia (leukomalacia), neoplasms and metastases), overstrained heart and skeletal muscle, adipose tissue (obesity), endocrine organs (thyroiditis, thyromegaly, pancreatic transplantation disorders), hematopoiesis (AIDS's kaposi syndrome), hematopoietic malignancies (leukemia), and lymphatic vessels (tumor metastases, lymphoproliferative disorders).

The term "retinal ischemic disease" means that the blood and oxygen supply to the retina decreases, the peripheral part of the retina loses its nutrient source and stops functioning properly. A specific example of a retinal ischemic disease is retinopathy. Common diseases that cause retinopathy are diabetic retinopathy, central retinal vein occlusion, carotid artery stenosis, and sickle cell retinopathy. Diabetic retinopathy is the leading cause of vision loss in diabetic patients. In ischemic retina, the growth of new blood vessels (neovascularization) occurs. These blood vessels often grow on the surface of the retina, on the optic nerve, or on the iris in front of the eye. The new blood vessels cannot replace the essential nutrient streams and can cause a number of problems such as vitreous hemorrhage, retinal detachment and uncontrolled glaucoma. These problems arise because the new blood vessels are fragile and prone to bleeding. Proliferative diabetic retinopathy is sometimes prevented by panretinal photocoagulation if it is found early in the process. Sometimes, however, vitrectomy is the only option.

In addition to these retinopathies, vascular diseases of the eye include ocular neovascular diseases such as macular degeneration and Diabetic Macular Edema (DME). Macular degeneration results from the growth of new blood vessels in the choroidal tract beneath the macula. There are two types of macular degeneration: dry and wet. Although wet macular degeneration accounts for only 15% of all macular degeneration, almost all of it causes blindness. In addition, wet macular degeneration is almost always caused by dry macular degeneration. Once one eye is affected by wet macular degeneration, the condition almost always affects the other eye. Wet macular degeneration is often referred to as the age-related wet macular degeneration of wet-AMD, as it is common in the elderly.

the increasing number of diabetic individuals worldwide suggests that DR and DME will continue to be the major contributors to vision loss and related functional deficits in the coming years.

The term "pulmonary hypertension" means a condition in which the blood pressure of the pulmonary artery is abnormally high. In the absence of other cardiopulmonary disorders, it is referred to as primary pulmonary hypertension. Diffuse narrowing of the pulmonary arteries occurs as a result of pathological arteriogenesis, followed by pulmonary hypertension due to an increased response to resistance to blood flow. The incidence rate is 8 out of 100'000 individuals. However, pulmonary hypertension can also occur as a complication of Chronic Obstructive Pulmonary Disease (COPD), such as emphysema, chronic bronchitis or diffuse interstitial fibrosis, and in asthmatic COPD patients. The incidence of COPD is 5 out of 10'000 individuals.

In addition, the binding proteins of the invention are useful for treating inflammation, more particularly inflammatory disorders.

The term "inflammation" as used herein means a local reaction to a living tissue injury, in particular a local reaction of small blood vessels, their contents and their related structures. The passage of blood components through the vessel wall into the tissue is a hallmark of inflammation and the resulting collection of tissues is called exudate or edema. Any deleterious process that damages living tissue, such as bacterial infection, excessive heat, cold, mechanical injury such as crushing, acid, base, irradiation, or viral infection, can cause inflammation, whether involving an organ or tissue. It is clear that diseases classified as "inflammatory diseases" and tissue reactions ranging from burns to pneumonia, leprosy, tuberculosis and rheumatoid arthritis are all "inflammation".

The binding proteins of the invention are useful for treating neoplasia. The term "tumor" means a mass of abnormal tissue that arises from pre-existing somatic cells without obvious causes, has no meaningful function, and is characterized by a tendency to grow autonomously and unrestrained. Tumors are quite different from inflammatory or other swelling because of the abnormal appearance and other characteristics of the cells in the tumor. Abnormal cells, i.e., the cell types that collectively make up a tumor, differ from normal cells in that one or more of the following changes have occurred: (1) hypertrophy, or an increase in the size of individual cells; (2) hyperplasia or an increase in the number of cells in a given area; (3) or the physical characteristics of the cells degenerate to a more primitive or undifferentiated type. The tumor may be benign, such as lipoma, hemangioma, osteoma, chondroma, and adenoma. Examples of malignant tumors are carcinomas (such as breast, respiratory and gastrointestinal, endocrine and genitourinary), sarcomas (in connective tissue, including fibrous tissue, adipose tissue, muscle, blood vessels, bone and cartilage), carcinosarcomas (in both epithelial and connective tissue), leukemias and lymphomas, tumors of nervous tissue (including brain) and melanomas (cancers of pigmented skin cells). The use of the binding proteins of the invention against tumors may also be combined with any other tumor therapy known in the art, such as radiation therapy, photodynamic therapy, chemotherapy or surgery.

Pharmaceutical compositions comprise a binding protein as described above and a pharmaceutically acceptable carrier, excipient or stabilizer (Remington's Pharmaceutical Sciences 16 th edition, Osol, a. ed. [1980 ]). Suitable carriers, excipients or stabilizers known to the skilled worker are saline, ringer's solution, dextrose solution, Hank's solution, fixed oils, ethyl oleate, 5% dextrose in saline, substances which improve isotonicity and chemical stability, buffers and preservatives. Other suitable carriers include any carrier that does not itself cause the production of antibodies harmful to the individual receiving the composition, such as proteins, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids, and amino acid copolymers. The pharmaceutical composition may also be a combined preparation comprising additional active agents, such as an anti-cancer agent or an anti-angiogenic agent (e.g. human VEGF-Axxxb; preferably human VEGF-A165 b).

Preferred pharmaceutical compositions for treating ocular diseases comprise a binding protein as described above and a detergent such as a non-ionic detergent, including but not limited to polysorbate 20 (e.g., about 0.04%), a buffering agent such as histidine, phosphate or lactic acid, and a sugar such as sucrose or trehalose. Preferably such compositions comprise a binding protein as described above and PBS. The or any other pharmaceutical composition described herein may be administered topically, via topical administration to a portion of the eye or by injection into the eye, e.g., into the subconjunctival, anterior or interocular space, or directly into the eye. Alternatively, the or such other pharmaceutical compositions may be administered systemically by parenteral administration. Preferably, the or such other pharmaceutical composition is applied to the eye by intravitreal injection. It is also preferred that the pharmaceutical composition is applied topically to the eye and as eye drops. Eye drops can be applied to the cornea (the transparent portion of the center of the eye) to allow molecules to penetrate into the eye. For treatment of diseases involving the back of the eye, it may be most desirable for the binding protein to penetrate the sclera when injected under the conjunctiva or around the eyeball. Administration of the binding protein may be performed after a preparatory step to modulate the ocular surface to improve penetration of the molecule. Preferably, the epithelial layer, such as corneal epithelium, is modified by a penetration enhancer to allow sufficient and rapid penetration of the molecule as described, for example, above. The use of the binding proteins of the invention against ocular diseases may also be combined with any other therapy known in the art, such as photodynamic therapy.

Formulations to be used for in vivo administration must be sterile or aseptic. This is easily accomplished by filtration with sterile filtration membranes.

Sustained release formulations can be prepared. In one embodiment of the invention, intraocular implants may be used to provide a binding protein of the invention. Suitable examples of sustained-release preparations include semipermeable matrices of solid hydrophobic polymers containing the polypeptide of the invention, which matrices are in the form of shaped articles, e.g., films, or microcapsules. Examples of sustained release matrices include polyesters, hydrogels (e.g., poly (2-hydroxyethyl-methacrylate), or poly (vinyl alcohol)), polylactide, copolymers of L-glutamic acid and γ -ethyl-L-glutamate, non-degradable ethylene-vinyl acetate, degradable lactic acid-glycolic acid copolymers such as LUPRONDEPOT @ (injectable microspheres composed of lactic acid-glycolic acid copolymer and leuprolide acetate), and poly-D- (-) -3-hydroxybutyric acid.

The pharmaceutical composition may be administered by any suitable method within the knowledge of the skilled person. The preferred route of administration is parenteral. For parenteral administration, the medicaments of the invention are formulated in unit dose injectable forms, such as solutions, suspensions or emulsions, in combination with pharmaceutically acceptable excipients as defined above. The dosage and mode of administration will depend on the individual to be treated and the particular disease. In general, the pharmaceutical composition is administered such that the binding protein of the invention is administered at a dose of from 1. mu.g/kg to 20 mg/kg, more preferably from 10. mu.g/kg to 5mg/kg, most preferably from 0.1 to 2 mg/kg. Preferably it is administered as a bolus dose. Continuous infusion may also be used, including continuous subcutaneous delivery via osmotic mini-pumps. If so, the pharmaceutical composition may be infused at a dose of 5 to 20. mu.g/kg/min, more preferably 7 to 15. mu.g/kg/min. In particular, the pharmaceutical composition is administered intraocularly by injection, such that the binding protein of the invention is administered at a dose of 0.1 mg to 10 mg per injection, more preferably 0.3 to 6 mg per injection, most preferably 1 mg to 4 mg per injection. Further, the pharmaceutical composition is administered to the eye by eye drops so that a single drop solution containing the binding protein of the present invention at a concentration of 10 to 120 mg/ml, more preferably 20 to 100 mg/ml, most preferably 40 to 80 mg/ml is applied to the eye.

in another embodiment of the invention, a binding protein that inhibits VEGF-Axxxx activity as described above can be used in combination with a binding protein or small molecule that inhibits PIGF activity (with the same level of inhibition of PIGF as described above for VEGF-Axxxx.) this embodiment is based on the fact that PIGF is found to be angiogenic at sites where VEGF-Axxx levels are elevated.

The invention further provides methods of treatment. In one aspect, methods of treating retinopathy are provided, the methods comprising administering to a patient in need thereof a therapeutically effective amount of a binding protein of the invention, particularly a binding protein that inhibits the interaction between human VEGF-Axxxx and human VEGFR-2 but does not inhibit the interaction between human VEGF-Axxxb and human VEGFR-2, and a binding protein that inhibits VEGFR-2 mediated angiogenesis.

The invention further relates to methods of using the binding proteins to inhibit VEGF-A biological activity in a cell or to inhibit biological activity mediated by VEGFR-2. The cells may be in vivo or ex vivo and may be, for example, cells of a living organism, cultured cells, or cells of a tissue sample. The methods can include contacting the cell with any of the binding proteins disclosed herein that inhibit VEGF-a/VEGFR-2 interactions in an amount and for a time sufficient to inhibit such biological activity.

The present invention provides methods of treating a subject having a condition responsive to inhibition of VEGF-Axxx or VEGFR-2. Such methods comprise administering to the subject an effective amount of a binding protein described herein. The condition may be characterized by inappropriate angiogenesis. The condition may be a hyperproliferative condition. Examples of conditions (or disorders) suitable for treatment include autoimmune diseases, inflammatory disorders, retinopathies (particularly proliferative retinopathies) and cancer, particularly one of the diseases described above. Any of the binding proteins described herein may be used for the preparation of a medicament for the treatment of such disorders, in particular a disorder selected from the group consisting of: autoimmune diseases, inflammatory disorders, retinopathies and cancer. Preferred conditions (or disorders) suitable for treatment are first-line metastatic renal cell carcinoma (first-line metastatic renal cell carcinoma), recurrent polymorphic glioblastoma, adjuvant colon cancer (adjuvant colon cancer), adjuvant HER 2-negative breast cancer, adjuvant HER 2-positive breast cancer, adjuvant non-small cell lung cancer, diffuse large B-cell lymphoma, first-line advanced gastric cancer, first-line HER 2-negative metastatic breast cancer, first-line HER 2-positive metastatic breast cancer, first-line metastatic ovarian cancer, gastrointestinal stromal tumors, high-risk cancers, hormone-resistant prostate cancer, newly diagnosed polymorphic gliomas, metastatic head and neck cancer, recurrent platinum-sensitive ovarian cancer, second-line metastatic breast cancer, extensive small-cell lung cancer, non-squamous non-small-cell lung cancer with previously treated CNS metastasis and recurrent multiple myeloma, prostate cancer, non-small-cell lung cancer (NSCLC), Colorectal and pancreatic cancer, Advanced Ovarian Cancer (AOC), AOC patients with malignant ascites symptoms, and non-Hodgkin's lymphoma.

Recombinant binding proteins of the invention can be obtained and/or further evolved by several methods, such as display on the surface of bacteriophages (WO90/02809, WO 07/006665) or bacterial cells (WO 93/10214), ribosome display (WO 98/48008), plasmid display (WO 93/08278) or by using covalent RNA-repeat protein hybrid constructs (WO 00/32823), or intracellular expression and selection/screening, for example by protein complementation assays (WO 98/341120). Such methods are known to those skilled in the art.

The ankyrin repeat protein libraries used to select/screen for recombinant binding proteins of the invention may be obtained according to protocols known to those skilled in the art (WO 02/020565, Binz, H.K. et al, JMB, 332, 489-503, 2003, and Binz et al, 2004, supra). The selection of VEGF-Axxx specific DARPin with such a library is given in example 1. Similarly, ankyrin repeat sequence motifs as set out above can be used to establish ankyrin repeat protein libraries useful for selecting or screening for VEGF-Axxx specific darpins. Furthermore, the repetitive domains of the invention can be modularly assembled from the repetitive modules of the invention and appropriate capping modules (Forrer, P. et al, FEBS letters 539, 2-6, 2003) using standard recombinant DNA techniques (e.g., WO 02/020565, Binz et al, 2003, in the above citations and Binz et al, 2004, in the above citations).

The invention is not limited to the specific embodiments described in the examples. Other sources may be used and processed in accordance with the summary described below.

Examples

All of the starting materials and reagents disclosed below are known to those skilled in the art and are commercially available or can be prepared by well known techniques.

Material

Chemicals were purchased from Fluka (Switzerland). The oligonucleotides were from Microsynth (Switzerland). Unless otherwise stated, DNA polymerase, restriction enzymes and buffers were from New England Biolabs (USA) or Fermentas (Lithuania). The clone and protein preparation strain was E.coli XL1-blue (Stratagene, USA). VEGF variants from R&D Systems (Minneapolis, USA), or in Chinese hamster ovary cells or Pichia pastoris (Pichia pastoris) ((III))Pichia pastoris) Prepared and purified according to standard protocols (Rennel, E. S. et al, European J. Cancer 44,1883-94, 2008; Pichia pastoris (Pichia) expression System by Invitrogen). Biotinylated VEGF variants were obtained chemically by coupling the biotin moiety to the primary amine of a purified VEGF variant using standard biotinylation reagents and methods (Pierce, USA).

Molecular biology

Unless otherwise indicated, the methods were carried out according to the described protocol (Sambrook J., Fritsch E. F. and Maniatis T., Molecular Cloning: A Laboratory Manual, Cold Spring harbor Laboratory 1989, New York).

Designed ankyrin repeat protein library

Ankyrin repeat protein libraries designed by N2C and N3C are described (WO02/20565; Binz et al 2003, supra; Binz et al 2004, supra). The numbers in N2C and N3C describe the number of randomized repeat modules that exist between N-and C-terminal capping modules. The nomenclature used to define the positions within the repeat units and modules is based on Binz et al 2004, the modification being that the edges of the repeat modules and repeat units are shifted by one amino acid position in the above citations. For example, 1 bit of a Binz et al 2004 (in the above-referenced article) duplicate block corresponds to 2 bits of a duplicate block of the present disclosure, and thus 33 bits of a Binz et al 2004 (in the above-referenced article) duplicate block corresponds to 1 bit of a next duplicate block of the present disclosure.

All DNA sequences were confirmed by sequencing and the calculated molecular weights of all the proteins were confirmed by mass spectrometry.

Example 1 selection of binding proteins comprising a repeat Domain with binding specificity for VEGF-Axxx

using ribosome display (Hanes, j. and plukthun, a., PNAS 94, 4937-42, 1997), a number of designed ankyrin repeat proteins (darpins) with binding specificity for VEGF-Axxx were selected from the N2C or N3C DARPin library described by Binz et al 2004 (cited above), the binding of selected clones to specific (VEGF-Axxx) and non-specific (MBP, e.coli maltose binding protein) targets was assessed by crude extraction ELISA, suggesting that successful selection of VEGF-Axxx binding proteins (fig. 1) consists of the amino acid sequence of the selected binding protein, which contains repeat domains with binding specificity for VEGF-Axxx sequence analysis of the selected binding agents reveals specific ankyrin repeat sequence motifs inherent to certain selected binding agent families.

Selection of VEGF-Axxx specific ankyrin repeat proteins by ribosome display

the selection of VEGF-Axxx-specific ankyrin repeat proteins was performed by ribosome display (Hanes and Pl ü ckthun, in the above cited references) using either dog VEGF-a164 or human VEGF-a165 as target protein, as described designed ankyrin repeat protein library (WO 02/020565, Binz et al, 2003, in the above cited references and Binz et al, 2004, in the above cited references) and established protocols (Zahnd, c., Amstutz, p. and Pl ü ckthun, a., nat. Methods 4, 69-79, 2007), with the selection of VEGF-Axxx-specific ankyrin repeat proteins performed by ribosome display (Binz et al 2004, in the above cited references) using both the established protocols (Binz et al 2004), with biotinylation selection cycles of dog or human VEGF variants (including those immobilized with neutravidin or streptavidin variants) using both N2C and N3 dar 3C darran libraries, with a biotinylation selection cycle (RT) -Reverse Transcription (RT) -PCR cycles per selection cycle, reduced from 40 to 30, with further immobilization cycles of the rate of the resulting in the initial selection cycle of the biotinylated binding of the dog or the biotinylated binding of the initial binding to the biotin-binding of the dog, with the biotin-binding of the initial binding agent, followed by further amplification cycle with the selection cycle, with the subsequent acquisition of the selection cycle of the biotinylated selection of the biotinylated binding of the biotin-biotinylated binding of the initial binding of the biotin-binding to the dog or the monoclonal antibody binding of the monoclonal antibody binding to the monoclonal antibody.

Selected clones specifically bind VEGF-Axxx as shown by crude extraction ELISA

Using standard protocols, various selected DARPin that specifically bind VEGF-Axxx were identified by enzyme-linked immunosorbent assay (ELISA) using crude E.coli extracts of DARPin-expressing cells. Selected clones were cloned into pQE30 (Qiagen) expression vector, transformed into E.coli XL1-Blue (Stratagene), and then grown overnight at 37 ℃ in 96 deep well plates (each clone in a single well) filled with 1ml of growth medium (2YT, containing 1% glucose and 100. mu.g/ml ampicillin). In a fresh 96 deep well plate, 1ml of fresh 2YT containing 50 μ g/ml ampicillin was inoculated with 100 μ l of overnight culture. After incubation at 37 ℃ for 2 hours, expression was induced with IPTG (1 mM final concentration) for a further 3 hours. Collecting the cellsResuspended in 100 μ l B-PERII (Pierce) and incubated at room temperature for 15 minutes with shaking. Then 900 μ l of PBS-TB (PBS supplemented with 0.2% BSA, 0.1% Tween 20, pH 7.4) was added and the cell debris was removed by centrifugation. 100 μ l of each lysed clone were applied to wells of NeutrAvidin-coated MaxiSorp plates loaded with VEGF-Axxx variants or unrelated MBP immobilized by their biotin moieties and incubated for 1 hour at room temperature. After extensive washing with PBS-T (PBS supplemented with 0.1% Tween 20, pH 7.4), the plates were visualized using a standard ELISA procedure using monoclonal anti-RGS (His)₄Antibody (34650, Qiagen) as the primary antibody and polyclonal goat anti-mouse antibody conjugated to alkaline phosphatase (A3562, Sigma) as the secondary reagent. Binding was then detected using disodium 4-nitrophenylphosphate (4NPP, Fluka) as a substrate for alkaline phosphatase. Color development was measured at 405 nm. The results of the crude extraction ELISA of samples used to identify DARPin that bound VEGF-Axxx are shown in FIG. 1. Screening of several hundred clones by such crude cell extraction ELISA revealed more than one hundred different darpins with specificity for VEGF-Axxx. These binding proteins were selected for further analysis. Examples of amino acid sequences of selected ankyrin repeat domains that specifically bind VEGF-Axxx are provided in SEQ ID NOS: 1 to 7.

Inferring repeat sequence motifs from selected repeat domains with binding specificity for VEGF-Axxxx

the amino acid sequences of selected repeat domains with binding specificity for VEGF-Axxxx are further analyzed by sequence analysis tools known to those skilled in the art (WO 02/020565; Forrer et al, 2003, supra; Forrer, P., Binz, H.K., Stumpp, M.T. and Plouckthun, A., ChemBioChem, 5(2), 183-189, 2004.) however, in contrast to WO 02/020565 where naturally occurring repeat motifs are used to infer repeat sequence motifs, repeat sequence motifs are inferred here from repeat units of selected repeat domains with binding specificity for VEGF-Axxxx.

High level and soluble expression of darpins

For further analysis, selected clones that showed specific VEGF-Axxx binding in a crude cell extraction ELISA as described above were expressed in e.coli XL1-blue cells and purified using their His-tag using standard protocols. 1 l of the culture (same medium) was inoculated with 25 ml of a culture (LB, 1% glucose, 100 mg/l ampicillin; 37 ℃) which had been left to stand overnight. At a (600) = 0.7, the culture was induced with 0.5 mM IPTG and incubated at 37 ℃ for 4 hours. The culture was centrifuged and the resulting pellet was resuspended in 40 ml TBS500 (50 mM Tris-HCl, 500 mM NaCl, pH 8) and sonicated. The lysate was recentrifuged and glycerol (10% (v/v) final concentration) and imidazole (20 mM final concentration) were added to the resulting supernatant. The protein was purified using a Ni-nitrilotriacetic acid column (2.5 ml column volume) according to the manufacturer's instructions (QIAgen, Germany). Up to 200 mg of highly soluble DARPin with binding specificity for VEGF-Axxx can be purified from 1 liter of e.coli culture medium, e.g., >95% pure as estimated by SDS-15% PAGE. The darpins thus purified were used for further characterization.

Example 2 determination of selected DARPins with binding specificity for VEGF-Axxx in a spheroid outgrowth assay ₅₀IC value of

Addition of VEGF-Axxx to HUVEC spheroids embedded in a collagen matrix resulted in spheroid sprouting. Addition of a VEGF-Axxx inhibitor will block shoot formation, which can be statistically quantified by the number and length of shoots. IC can be determined by adding different concentrations of inhibitor and quantifying VEGF₅₀。

Inhibition of spheroid sprouting by VEGF-Axxx specific DARPin

Spheroid outgrowth measurements were performed according to standard protocols (Korff et al, supra). DARPin specific for VEGF-Axxx was selected and purified to >96% purity as described in example 1. Human umbilical vein cells were grown to confluence in monolayer culture. After trypsinization, the cell suspension was placed in hanging drops to form spheroids, i.e., about 500 HUVECs with tissue aggregation. Spheroids were embedded in a collagen matrix and stimulated with VEGF-A165 to initiate shoot outgrowth. A germination inhibitor was additionally added to observe its effect on germination inhibition. Shoot number and shoot length were quantified for each spheroid using mapping software.

The results of two example spheroid germination assays are shown in FIG. 2a (DARPin #30 with binding specificity for VEGF-Axxx) and FIG. 2b (DARPin NC, negative control DARPin with no binding specificity for VEGF-Axxx; e.g., DARPin E3_5 (Binz et al, 2005, supra.) the DARPin that performed best in this assay showed an IC of 10 to 50pM₅₀Value, and in parallel experiments, Avastin, Lucentis and Macugen show IC of 150 to 500 pM₅₀The value is obtained.

Example 3 determination of a target of DARPin #27 compared to Avastin by surface plasmon resonance analysis Specificity of

Dog VEGF-A164 or dog VEGF-A164b were immobilized in a flow cell and DARPin #27 (repeat domain of SEQ ID NO:1, corresponding to amino acids 1 to 159) and Avastin were analyzed for interaction with the immobilized targets.

Surface Plasmon Resonance (SPR) analysis

SPR was measured with a ProteOn instrument (BioRad). The working buffer was 20 mM HEPES, pH 7.4, 150 mM NaCl and 0.005% Tween 20. Approximately 1200 RU of dog VEGF-A164 or dog VEGF-A164b was immobilized on a GLC chip (BioRad). Measuring the interaction: the buffer stream was flowed at 60 μ l/min for 5 minutes, injected with 250 nM concentration of Avastin or DARPin #27 for 100 seconds, and the dissociation rate measurements were performed for several minutes with the buffer stream. The signal of the uncoated reference pool is subtracted from the measurement.

The results are shown in FIG. 3a (interaction of Avastin with dog VEGF-A164), FIG. 3b (interaction of Avastin with dog VEGF-A164 b), FIG. 3c (interaction of DARPin #27 with dog VEGF-A164), and FIG. 3d (interaction of DARPin #27 with dog VEGF-A164 b). Although Avastin clearly interacted with both fixed VEGF isoforms, DARPin #27 showed interaction with VEGF-A164 only and not VEGF-A164 b.

Example 4 in vivo efficacy of DARPin #30 in vascular leak rabbit model inhibiting VEGF-A165.

Pegylated DARPin #30 (repetitive domain of SEQ ID NO:4 corresponding to amino acids 1 to 126) or Lucentis @, were applied by intravitreal injection into rabbit eyes to test their efficacy in inhibiting vascular leakage induced by subsequent intravitreal injection of human VEGF-A165.

Vascular leakage inhibition measurement in rabbits

PBS, PEGylated DARPin #30 (125. mu.g) or an equimolar amount of Lucentis @ (162. mu.g) was applied by intravitreal injection into one eye of each rabbit (treated eyes). Treated eyes of each rabbit were challenged by intravitreal injection of 500 ng human VEGF-A165 on day 4 or day 30. Both eyes of all animals were evaluated 48 hours after VEGF-A165 injection by measuring the fluorescein content in all eyes 1 hour after intravenous injection of sodium fluorescein (50 mg/kg animal body weight, 10% (w/v) in 0.9% (w/v) saline solution). The ratio of the amount of fluorescence in the treated eye to the untreated eye was calculated for each animal. A ratio of 1 corresponds to no additional fluorescence leakage in the treated eye, with a ratio greater than 1 indicating more fluorescence leakage in the treated eye than in the untreated control eye.

Preparation of pegylated darpins

PEGylation of proteins by using a single Cys residue and maleimide chemistry is well known to those skilled in the art and can be performed according to established protocols (e.g., from Pierce). DARPin #30 containing an attached C-terminal linker (GGGSGGGSC, SEQ ID NO:8) was purified to near homogeneity using standard chromatographic methods. The protein was completely reduced with DTT, purified by gel filtration to remove DTT, and the buffer was replaced with PBS. PEG-maleimide (methoxy-poly (ethylene glycol) -oxopropylamino-propylmaleimide; NOF, No. Sunbright ME-200MA) dissolved in PBS was mixed with DARPin in PBS with a molar excess of PEG-maleimide of about 15% at room temperature for 2-4 hours. The pegylated darpins were then separated from the non-reactive darpins and non-reactive PEG moieties by using standard anion exchange chromatography.

The results are shown in FIG. 4. Both pegylated DARPin #30 and Lucentis @ were able to protect rabbit eyes from VEGF-A165 induced vascular leakage 4 days after their administration by intravitreal injection. However, only PEGylated DARPin #30, but not Lucentis @, was able to protect rabbit eyes from VEGF-A165 induced vascular leakage up to 30 days after intravitreal injection.

In other experiments the terminal intravitreal half-life of different binding proteins of the invention was measured after intravitreal injection into rabbit eyes. DARPin #30 containing an additional C-terminal linker (GGGSGGGSC, SEQ ID NO:8) was conjugated to 20 kDa and 40kDa non-proteinaceous PEG moieties with maleimide PEG from NOF, respectively (see example 5). The terminal half-lives of DARPin #30, DARPin #30 conjugated with a 20 kDa PEG moiety, and DARPin #30 conjugated with a 40kDa PEG moiety were determined to be 3.5 days (+/-0.3 days), 6.1 days (+/-1.0 days), and 5.4 days (+/-0.8 days). Surprisingly, increasing the molecular weight of the non-proteinaceous PEG moiety from 20 kDa to 40kDa did not result in an increase in terminal half-life. The same trend was observed in the corresponding experiments using binding proteins comprising the repeat domain of SEQ ID NO:1 (amino acids 1 to 159) or SEQ ID NO:3 (amino acids 1 to 126) in place of the repeat domain of SEQ ID NO: 4.

Example 5 recombinant binding proteins

Examples of recombinant binding proteins comprising a repetitive domain that binds VEGF-Axxx and a proteinaceous polymer moiety are SEQ ID Nos. 1 and 4. The repeat domain of SEQ ID NO. 1 corresponds to amino acids 1 to 159 and the proteinaceous polymer part of SEQ ID NO. 1 corresponds to amino acids 160 to 1' 024. The repeat domain of SEQ ID NO. 4 corresponds to amino acids 1 to 126 and the proteinaceous polymer part of SEQ ID NO. 4 corresponds to amino acids 127 to 536.

The binding proteins of SEQ ID NO 1 and 4 are expressed in the cytoplasm of E.coli using standard techniques known to those skilled in the art (see, e.g., Qiagen (Germany)) using the pQE expression system. The Met residue additionally encoded by the expression vector is efficiently cleaved from the expressed polypeptide in the E.coli cytoplasm, since the initial Met is followed by a small Gly residue (i.e.amino acid 1 of SEQ ID NO:1 and 4). The binding protein is purified from the crude cell extract to near homogeneity by standard chromatographic techniques known to those skilled in the art, by lysing the cells (e.g., by using a French press).

Examples of recombinant binding proteins comprising a repeat domain that binds VEGF-Axxx and a non-proteinaceous polymer moiety were prepared using the repeat proteins of SEQ ID Nos. 2, 3, 5, 6 and 7. These repeat proteins comprise an N-terminal repeat domain, followed by a polypeptide linker and a C-terminal Cys. The respective repeat domains correspond to amino acids 1 to 159 of SEQ ID NOS: 2 and 7, and amino acids 1 to 126 of SEQ ID NOS: 3 to 6. The repeat proteins of SEQ ID NO 2, 3, 5, 6 and 7 are expressed in the cytoplasm of E.coli using standard techniques known to the person skilled in the art (see e.g.expression by Qiagen (Germany)). The Met residue additionally encoded by the expression vector is efficiently cleaved from the expressed polypeptide in the E.coli cytoplasm, since the initial Met is followed by a small Gly residue (i.e.amino acid 1 of SEQ ID NOS: 2, 3, 5, 6 and 7). The cells are lysed (e.g., by using a French press), and the binding protein is purified to near homogeneity from the crude cell extract using standard chromatographic techniques known to those skilled in the art.

the purified repeat protein comprising a single Cys residue is then conjugated to the non-proteinaceous polymer moiety using standard maleimide chemistry, as outlined in example 4. thus, a binding protein of the invention is prepared comprising the repeat protein of SEQ ID NO:2 and a 40kDa non-proteinaceous PEG moiety (e.g.40 kDa maleimide-PEG of NOF (α - [3- (3-maleimido-1-oxopropyl) amino ] propyl-omega-methoxy-polyoxyethylene), product No. Sunbright ME-400MA), the repeat protein of SEQ ID NO:3 and a 20 kDa non-proteinaceous PEG moiety (e.g.20 kDa maleimide-PEG of NOF (α - [3- (3-maleimido-1-oxopropyl) amino ] propyl-omega-methoxy-polyoxyethylene), product No. Sunbright ME-200MA), the repeat protein of SEQ ID NO:5 and a12 kDa non-proteinaceous PEG moiety (e.g.12 kDa maleimide-PEG of NOF (α - [3- (3-maleimido-1-oxopropyl) amino ] propyl-omega-methoxy-polyoxyethylene), product No. Sunbright ME-200MA), the repeat protein of SEQ ID NO:5 and the non-proteinaceous PEG moiety of 12 kDa (e.g.g.g.12 kDa maleimide-PEG protein of NOF (e.g.g.3- (3-maleimide-1-oxopropyl) amino ] propyl-4), the non-polyethylene glycol moiety of the protein of the art), the protein of the amino-4 protein of the invention is further separated by standard PEG technology, and the protein of the art (e.5-maleimide-4-methoxy-4 protein of the invention, and the protein of the invention.

thus, SEQ ID NOs 2, 3, 5, 6 and 7 are conjugated with maleimide PEG (. alpha. - [3- (3-maleimido-1-oxopropyl) amino ] propyl-. omega. -methoxy-polyoxyethylene) at the thiol group of their C-terminal cysteines, resulting in the following structures:

wherein X is SEQ ID NO 2, 3, 5, 6 or 7; n is a positive integer.

Claims (16)

1. A recombinant binding protein consisting of: consisting of SEQ ID NO:3, a polypeptide linker consisting of 2 to 24 amino acids containing a C-terminal Cys residue, a maleimide-coupled polyethylene glycol having a molecular weight of at least 5 kDa; wherein said ankyrin repeat domain is coupled at its C-terminus to said polypeptide linker by a peptide bond, and wherein the thiol group of said C-terminal Cys is conjugated to said maleimide-coupled polyethylene glycol.

2. The binding protein of claim 1, consisting of the amino acid sequence of SEQ ID NO 3 and maleimide-coupled conjugated polyethylene glycol of at least 5kDa molecular weight, wherein the amino acid sequence of SEQ ID NO: the thiol group of the C-terminal Cys of 3 is conjugated to the maleimide-coupled polyethylene glycol.

3. The binding protein of claim 1 or 2, wherein said polyethylene glycol has a molecular weight of at least 10 kDa.

4. The binding protein of claim 1 or 2, wherein said polyethylene glycol has a molecular weight of about 20 kDa.

5. the binding protein of claim 1 or 2, wherein said maleimide-conjugated polyethylene glycol is α - [3- (3-maleimido-1-oxopropyl) amino ] propyl- ω -methoxy-polyoxyethylene.

6. the binding protein of claim 1 or 2, wherein said maleimide-coupled polyethylene glycol is α - [3- (3-maleimido-1-oxopropyl) amino ] propyl- ω -methoxy-polyoxyethylene, wherein said polyoxyethylene has a molecular weight of at least 10 kDa.

7. the binding protein according to claim 1 or 2, wherein said maleimide-conjugated polyethylene glycol is α - [3- (3-maleimido-1-oxopropyl) amino ] propyl- ω -methoxy-polyoxyethylene, wherein said polyoxyethylene has a molecular weight of about 20 kDa.

8. The binding protein of claim 1 or 2, wherein said polyethylene glycol has a molecular weight of 10-40 kDa.

9. The binding protein of claim 1 or 2, wherein said polyethylene glycol has a molecular weight of 15-30 kDa.

10. A pharmaceutical composition comprising a binding protein according to any one of claims 1 to 9 and optionally a pharmaceutically acceptable carrier and/or diluent.

11. The pharmaceutical composition of claim 10 for use in the treatment of an eye disease.

12. The pharmaceutical composition of claim 10 for use in the treatment of an eye disease by intravitreal injection.

13. Use of a binding protein according to any one of claims 1 to 9 in the manufacture of a medicament for the treatment of pathological angiogenesis.

14. A recombinant binding protein comprising an ankyrin repeat domain consisting of amino acid sequence positions 1-126 of SEQ ID No. 3 and a maleimide-coupled polyethylene glycol of at least 5kDa molecular weight, wherein said ankyrin repeat domain is coupled at its C-terminus by a peptide bond to a polypeptide linker consisting of 2-24 amino acid sequences and comprising a C-terminal Cys residue, and wherein the thiol group of said C-terminal Cys is conjugated to said maleimide-coupled polyethylene glycol.

15. The binding protein of claim 14, wherein said polyethylene glycol has a molecular weight of 10-40 kDa.

16. The binding protein of claim 14, wherein said polyethylene glycol has a molecular weight of 15-30 kDa.