MC200088A1 - Identification and characterization of anti-fibronectin antibodies ED-B with blocking function - Google Patents

Description

002516

2 0 0 0 8 8

2 a nv. m f S" 3*0

,o*>u

\C

DESCRIPTIVE MEMOIR

Filed in support of an application for:

PATENT FOR INVENTION

Trained by

SCHERING AKTIENGESELLSCHAFT

And

MorphoSys AG

For :

Identification and characterization of anti-fibronectin ED-B antibodies with a function-blocking effect.

Temporary protection (Art. 11 — Paris Convention on Union)

Publication made on March 17, 2005 at the "IBC Drug Discovering Technology 2005" congress

Inventors:

Andréas MENRAD, Allertstrasse 7, 16515 Oranienburg / Germany

Josef PRASSLER, Sandstrasse 20, 82110 Germering / Germany

Armin WEIDMANN, Vorhoelzerstrasse 3, 86911 Diesser am Ammersee / Germany

-1 -

Identification and characterization of anti-fibronectin ED-B antibodies with a function-blocking effect

Description

The present invention relates to recombinant polypeptides, in particular antibodies or antibody fragments, which bind to the ED-B isoform of fibronectin and can block its function. Furthermore, the invention describes the diagnostic and pharmaceutical use of the peptides according to the invention.

During tumor formation and progression, the extracellular matrix (ECM) in which the tumor grows undergoes transformation through proteolytic degradation of the existing ECM. This process generates a tumor-induced matrix, which differs from the ECM found in normal tissue. Tumor-induced ECM appears to be the optimal environment for tumor growth and angiogenesis (1-4).

15 In tumor angiogenesis, new blood vessels are formed from already existing vessels. This process presupposes ECM proteolysis, targeted growth and differentiation of endothelial cells into new vascular structures, which are essential for continued tumor growth (5).

Fibronectins represent an important class of matrix glycoproteins. Their primary role is to allow cell adhesion to a wide variety of extracellular matrices. The presence of fibronectins on the surface of cultured, untransformed cells, as well as their absence in transformed cells, has led to the identification of fibronectins as important adhesion proteins. They interact with many other molecules, such as collagen, heparanesulfate proteoglycans, and fibrin, thereby regulating cell shape and cytoskeletal structure. Furthermore, they

-2-

are responsible for cell migration and differentiation during embryogenesis. They are also important for wound healing, during which they allow the migration of macrophages and other immune cells to the area concerned, and during the formation of blood clots, by allowing blood platelets to adhere to damaged areas of blood vessels.

Fibronectins are dimers of 2 analogous peptides, each chain having a length of approximately 60 to 70 nm. At least 20 different chains of fibronectins have been identified, all of these chains being produced by alternative splicing of the RNA transcript of a single fibronectin gene.

Fibronectins are high molecular weight adhesion-promoting glycoproteins that play an important role in the development of the vascular system. Furthermore, fibronectins have a chemotactic effect on endothelial cells, modulating the action of growth factors and promoting the longitudinal growth of endothelial cells during angiogenesis (6-9). Analysis of the proteolytic dissociation products of fibronectin shows that the polypeptides are composed of six highly folded domains, each containing repeat sequences whose amino acid sequence similarities allow for classification into three types (Type I, II, III). The central region of the two chains of the dimer consists of a segment of Type III repeats, which average 90 amino acids in length (10). Structural studies have shown that each Type III repeat consists of seven chains (3, which are folded into two antiparallel sheets, with short loop regions being exposed, as potential protein-protein interaction sites (11).

Type III repeats allow fibronectins to act as adhesion molecules, interacting with surface molecules.

30

of the cell, what are called "integrins." The concept of integrin was first used in 1987 (12) to describe a related group of cell surface heterodimeric molecules that act as promoters between the extracellular matrix and the intracellular cytoskeleton, thereby inducing cell adhesion and migration. These heterodimeric receptors "integrate," or thus carry out, signals from the extracellular environment that have specific cellular functions. Currently, 17 β subunits are known, which can interact, specifically and non-covalently, with more than 20 α subunits, thus forming more than 20 different families (13). In particular, the RGDS sequence, which is found in the tenth repeat of fibronectin Type III (111-10), causes fibronectin to interact with at least 8 different integrins. Furthermore, it has been shown that at least four integrins can, independently of the RGDS, interact specifically with fibronectin (13). The Type III repeat sequence group also includes, in addition to sequences III7, III8, III9 and 11110, the EIIIB and EIIIA repeats (ED-B and ED-A).

Fibronectin ED-B is not detectable in normal adult tissue (with the sole exception of proliferative endometrium), whereas it is highly expressed in fetal tissue and during tumor growth, in addition to local stromal expression of ED-B. Furthermore, fibronectin ED-B is localized to a perivascular site around vessels that have newly formed during angiogenesis. This is why fibronectin ED-B represents a specific marker protein for the (tumor) angiogenesis process (14).

The ED-B domain is a complete, highly conserved Type III homology constituent composed of 91 amino acids. The degree of homology between humans and rats is 100%, and between humans and chickens is 96%. Little is said in the literature regarding the function of the ED-B domain. A small number of publications (15-17) begin to explore

-4-

Hypotheses exist regarding a general effect, promoting adhesion, for different cells. A specific effect on endothelial cells has not yet been demonstrated.

5 II It was shown in application WO 02/20563 that recombinant ED-B exhibits a specific pro-angiogenic effect in vitro: (i) proliferation of human dermal microvascular endothelial cells stimulated by bFGF (HDMVEc) is enhanced by recombinant PED-B, (ii) the protein causes the adhesion of HDMVEc and (iii) recombinant ED-B stimulates 10 invasion and differentiation (tube formation) of HDMVEc in collagen gels.

Furthermore, these effects induced by ED-B can be specifically blocked by synthetic peptides derived from the ED-B domain. The peptide sequences thus represent the binding region for a specific ED-B receptor on the surface of endothelial cells, a receptor which, by affinity chromatography, has been identified as the integrin CC2P1. The specific interaction between the integrin CC2P1 and the ED-B domain has not yet been described in the literature.

'Furthermore, application WO 02/20563 describes proteins which undergo specific regulation by the ED-B domain of fibronectin, which includes the integrin a2Pi, the focal adhesion kinase, the CD6 ligand (ALCAM), the alpha chain of the fibronectin receptor, the integrated alpha 25 8 subunit, or the precursor of the follistatin-related process.

A large number of integrin receptors, exhibiting partially overlapping properties, are expressed by endothelial cells (Ref.: DG Stupack and DA Cheresh, SciSTKE, February 12, 2002; 2002). These 30 expression patterns show that different integrins (e.g., αvβ3 and α5β1) lead to analogous biological phenomena (adhesion, migration, and survival) and thus represent systems

-5-

redundant for endothelial cells, systems that guarantee their behavior and survival. To date, it was known that Palpha2beta1 interacts with these natural ligands, collagens. Blocking this interaction can lead to an anti-angiogenic effect (Y Funahashi et al. Cancer Res.

5 62:6116-6123,2002).

One object of the present invention is therefore to provide binding molecules with function-blocking effects, such as antibodies, that specifically block the binding sites of the 10 receptors in the ED-B domain. These binding molecules have an anti-angiogenic effect on endothelial (tumor) cells. Unlike the CC2P1 integrin, which is expressed relatively broadly, the ED-B domain represents an ideal and specific target molecule for Hp np tvr>A binding molecules

The structure of ED-B makes the development of monoclonal and polyclonal antibodies difficult, as ED-B exhibits low immunogenicity in vivo. For example, the BC-1 antibody (J. Cell Biol. 108 (1989), 1139-1148) reacts with a triptych epitope of fibronectin 20, which is present only in the presence of ED-B, and therefore does not react directly with ED-B. The L19 antibody (Tarli et al. Blood 94 (1999), 192-198 and WO 01/62800), which was produced as an immunogen using recombinant ED-B, is biologically inactive; that is, it cannot meaningfully recognize cell adhesion to ED-B.

Surprisingly, it has now been found that it is possible to manufacture ED-B binding molecules with function-blocking effects, such as the ED-B function-blocking antibody MOR03255, which strongly inhibit cell adhesion to ED-B and, in vivo, lead to a significant decrease in tumor growth. The ED-B blockade caused by the binding molecules according to the invention clearly cannot be compensated for by the mechanisms

15

compensatory mechanisms of the collagen-integrin interaction.

Using the HuCAL®-GOLD antibody library, a library containing approximately 1.6 x 10¹⁰ different antibodies in Fab fragment format, which was generated from the HuCAL consensus sequences (WO 97/08320; Knappik A, Ge L, Honegger A, Pack P, Fischer M, Wellnhofer G, Hoess A, Wolle J, Plunckthun A, and Virnekas B (2000) J. Mol. Biol. 296:57-86) by diversification based on the diversity of human antibodies in the six CDR domains, and which was modeled using CysDisplay (WO 01/05950), a variant of the phage surface presentation method, Fab antibody fragments with function-blocking effects, and which bind selectively to the ED-B domain, were identified in the context of the tests leading to the present invention. The efficacy of these antibody fragments was demonstrated in an in vitro adhesion assay, reflecting the specific interaction between recombinant ED-B and isolated HDMVEc. Targeted modification of the binding molecules improved the binding affinity to the ED-B domain. Furthermore, in vivo, the binding molecules were shown to inhibit tumor growth in an animal model.

A first aspect of the invention thus relates to a polypeptide which

(i) binds in a specific manner to the ED-B domain of fibronectin, and

(ii) inhibits the interaction between the ED-B domain and its receptor.

The polypeptide according to the invention is preferably an antibody or an antibody fragment. The term "antibody", as used in the present invention, includes polyclonal, monoclonal, chimeric, humanized or human antibodies, as well as recombinant antibodies, for example single-stranded antibodies or antigen-binding antibody fragments, for example monovalent antibody fragments such as Fab fragments or scFv fragments, or divalent antibody fragments such as F(ab')2 fragments.

-7-

In this regard, it is important, according to the invention, that the antibodies contain one or more antigen-binding sites that satisfy the prerequisites mentioned above, namely specific binding to the ED-B domain and inhibition of the interaction between the ED-B domain and its receptor, in particular its receptor on endothelial cells. These antigen-binding properties are preferably achieved by combining a VH region and a VL region, these regions being constituted from what are called framework regions (FR1, FR2, FR3 and FR4), as well as CDR regions ensuring antigen binding (H-CDR1, H-CDR2, H-CDR3 for the VH region, and L-CDR1, L-CDR2, L-CDR3 for the VL region).

The polypeptide according to the invention preferably has a certain affinity 15 for the ED-B domain, which corresponds to a Ko value < 1 pM, preferably <100 nM, particularly preferred <10 nM, even more preferred <1 nM and especially preferred < 0.1 nM, the affinity being able to be determined as indicated in the examples, nar nn <=>«t<5ai <51 ir If» <5\/<:tème BIAcore®.

The invention is characterized by the fact that it binds specifically to the ED-B domain of fibronectin, that is, it binds with a significantly lower affinity to other fibronectin domains, in particular to fibronectin domain 6 (FN6) and/or to fibronectin domains 7-8-9 (7-8-9). In a preferred embodiment, the polypeptide according to the invention and the ED-B domain of fibronectin bind with an affinity at least twice as high, in particular at least five times as high, and in a particularly preferred manner at least ten times as high as to FN6 and/or 7-8-9, as can be determined, for example, by the binding assay presented in the examples.

-8-

In another preferred embodiment, the polypeptide according to the invention exhibits in vitro an inhibition of the adhesion of recombinant ED-B to HDMVEc cells, and preferably an inhibition of at least 50%, and in a particularly preferred manner of at least 75%, at a concentration which in each case is 10 pg/ml.

It is further preferred that the polypeptide according to the invention exhibit in vivo inhibition of a tumor - produced by implantation of F9 teratocarcinoma cells (ATCC CRL-1720) in test animals, for example naked mice.

Y

The polypeptide according to the invention is preferably selected from antibodies and antibody fragments comprising:

(a) a VH region

15 (i) encoded by a nucleic acid sequence SEQ ID NO. 1 (MOR

20

02610), SEQ ID NO. 5 (MOR 02611), SEQ ID NO. 9 (MOR 02613), SEQ ID NO. 13 (MOR 02614), SEQ ID NO. 17 (MOR 02616), SEQ ID NO. 21 (MOR 02618), SEQ ID NO. 25 (MOR 02619), SEQ ID NO. 29 (MOR 02622), SEQ ID NO. 33 (MOR 02715), SEQ ID NO. 37 (MOR 02718), SEQ ID NO. 41 (MOR 02721), SEQ ID NO. 45 (MOR 02722), or at least one H-CDR1, H-CDR2 and/or H-CDR3 region of one of the VH regions mentioned, or

(ii) which derives from a VH region according to (i) by a modification of at least one H-CDR region, and/or

25

(b) a VL region

30

(i) encoded by a nucleic acid sequence SEQ ID NO. 3 (MOR 02610), SEQ ID NO. 7 (MOR 02611), SEQ ID NO. 11 (MOR 02613), SEQ ID NO. 15 (MOR 02614), SEQ ID NO. 19 (MOR 02616), SEQ ID NO. 23 (MOR 02618), SEQ ID NO. 27 (MOR 02619), SEQ ID NO. 31 (MOR 02622), SEQ ID NO. 35 (MOR 02715), SEQ ID NO. 39 (MOR 02718), SEQ ID NO. 43 (MOR 02721), SEQ ID NO. 47 (MOR 02722), or at least one L-CDR1, L-CDR2 and/or L-CDR3 region, of

-9-

one of the VL regions mentioned, or (ii) which derives from a VL region according to (i) by modification of at least one L-CDR region.

5 In region VL, modifications of region L-CDR3 are preferred and/or in region VH, modifications of region H-CDR2 are preferred.

v.

For example, the polypeptide according to the invention has a VL region derived from the VL region encoded by the nucleic acid sequence SEQ ID NO. 10 27 (MOR 02619) or SEQ ID NO. 35 (MOR 02715). A VH region is particularly preferred.

(i) encoded by the nucleic acid sequence SEQ ID NO. 25 (MOR 02619) or SEQ ID NO. 33 (MOR 02715), or at least one H-CDR1, H-CDR2 and/or H-CDR3 region of one of the VH regions mentioned, or (ii) derived from a VH region according to (i) by modification of at least one H-CDR region.

has

Preferably, the polypeptide has a VH region obtained by modifying the H-CDR2 region of a VH region encoded by nucleic acid sequence 20 of SEQ ID NO. 25 (MOR 02619) or SEQ ID NO. 33 (MOR 02715). A polypeptide comprising

(a) a VH region

(i) encoded by the nucleic acid sequence SEQ ID NO. 63 (MOR 03243), SEQ ID NO. 67 (MOR 03245), SEQ ID NO. 71 (MOR 25 03246), SEQ ID NO. 75 (MOR 03251), SEQ ID NO. 79 (MOR

03252), SEQ ID NO. 81 (MOR 03253), SEQ ID NO. 83 (MOR 03255), SEQ ID NO. 85 (MOR 03257), SEQ ID NO. 87 (MOR 03258), or at least the H-CDR2 region of one of the VH regions mentioned, or 30 (ii) which is derived from a VH region according to (i) by modification of at least one H-CDR region, and/or

(b) a VL region »

-10-

20

(i) encoded by the nucleic acid sequence SEQ ID NO. 65 (MOR

03243), SEQ ID NO. 69 (MOR 03245), SEQ ID NO. 73 (MOR 03246), SEQ ID NO. 77 (MOR 03251 as well as MOR03252, MOR03253, MOR03255 and MOR03257), SEQ ID NO. 89 (MOR 03258), or at least region L-CDR1, L-CDR2 or L-CDR3, of one of the VL regions mentioned, or

(ii) which derives from a VL region according to (i) by modification of at least one L-CDR region.

10 Specific examples of polypeptides according to the invention are presented below:

fcy

A polypeptide comprising the VH region encoded by SEQ ID NO. 1 and the VL region encoded by SEQ ID NO. 3 (MOR 02610), or at least one H-15 region CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3.

A polypeptide comprising the VH region encoded by SEQ ID NO. 5 and the VL region encoded by SEQ ID NO. 7 (MOR 02611), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 9 and the VL region encoded by SEQ ID NO. 11 (MOR 02613), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

25 A polypeptide comprising the VH region encoded by SEQ ID NO. 13 and the VL region encoded by SEQ ID NO. 15 (MOR 02614), or at least one region

H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3.

^—_ -—.

A polypeptide comprising the VH region encoded by SEQ ID NO. 17 and the VL region encoded by SEQ ID NO. 19 (MOR 02616), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

-11 -

A polypeptide comprising the VH region encoded by SEQ ID NO. 21 and the VL region encoded by SEQ ID NO. 23 (MOR 02618), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 27 (MOR 02619) or at least one region

H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3.

^

A polypeptide comprising the VH region encoded by SEQ ID NO. 29 and the 10 VL region encoded by SEQ ID NO. 31 (MOR 02622), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 35 (MOR 02715), or at least one 15 H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

Y

A polypeptide comprising the VH region encoded by SEQ ID NO. 37 and the VL region encoded by SEQ ID NO. 39 (MOR 02718), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

20

A polypeptide comprising the VH region encoded by SEQ ID NO. 41 and the VL region encoded by SEQ ID NO. 43 (MOR 02721), or at least one region

H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3.

>—

25 ' A polypeptide comprising the VH region encoded by SEQ ID NO. 45 and the VL region encoded by SEQ ID NO. 47 (MOR 02722), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 49 (MOR 03055), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 51 (MOR 03066), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 53 (MOR 03075), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 55 (MOR 03069), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 57 (MOR 03071), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

Y

A polypeptide comprising the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 59 (MOR 03064), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 61 (MOR 03062), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 63 and the VL region encoded by SEQ ID NO. 65 (MOR 03243), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 67 and the VL region encoded by SEQ ID NO. 69 (MOR 03245), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

-13-

A polypeptide comprising the VH region encoded by SEQ ID NO. 71 and the VL region encoded by SEQ ID NO. 73 (MOR 03246), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

^ A polypeptide comprising the VH region encoded by SEQ ID NO. 75 and the VL region encoded by SEQ ID NO. 77 (MOR 03251), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 79 and the VL region encoded by SEQ ID NO. 77 (MOR 03252), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

\

A polypeptide comprising the VH region encoded by SEQ ID NO. 81 and the VL region encoded by SEQ ID NO. 77 (MOR 03253), or at least one region

H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3.

>

/

A polypeptide comprising the VH region encoded by SEQ ID NO. 83 and the VR region encoded by SEQ ID NO. 77 (MOR 03255), or at least one region

H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3. >

A polypeptide comprising the VH region encoded by SEQ ID NO. 85 and the VR region encoded by SEQ ID NO. 77 (MOR 03257), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

A polypeptide comprising the VH region encoded by SEQ ID NO. 87 and the VR region encoded by SEQ ID NO. 89 (MOR 03258), or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.

The structure of the VH and VR chains of the polypeptides according to the invention is as follows:

>——-

VH Channel:

?

- 14-

- The framework region 1 goes from nt 1 to 78 (that is 26 aa), the last codon (aa) of FR1 is always: TCC (Cys).

- The "CDR1 region" can have two different lengths, either 30 nt (10 aa) (possible in families VH1A, VH1B, 3, 4 and 5); or 36 nt (12 aa)

5 (possible in families VH2, VH4 and VH6).

- The framework region 2 always has 33 nt (that is, 11 aa), the first codon (aa) of FR2 is always: TGG (Trp), the last two codons are always CTC.GAG (LeuGlu).

- The "CDR2 region" can have three different lengths, either 57 nt (19 10 aa) (possible in the VH2 and VH4 families); 60 nt (20 aa) (possible in the VH1A, VH1B, 3 and 5 families) or 63 nt (21 aa) (possible in the VH6 family).

- The framework region 3 still has 96 nt (that is, 32 aa) the third codon (aa) of FR3 is still: ACC (Thr), the last five codons are still TAT.TAT.TGC.GCG.CGT (Tyr Tyr Cys Ala Arg).

15 - The "CDR3 region" can have several different lengths, from 12 to 69 nt (4 to 23 aa) (in all VH2 and VH4 families); 60 nt (20 aa) (possible in VH1A, VH1B, 3 and 5 families) or 63 nt (21 aa) (possible in the VH6 family).

- The framework region 4 always has 33 nt (that is 11 aa) and is 20 identical in the 7 Families: TGG.GGC.CAA.GGC.ACC.CTG.GTG.ACG.

GTT.AGC.TCA

VL kappa channel:

- The framework region 1 goes from nt 1-69 (that is 23 aa), the last codon 25 (aa) of FR1 is always: TGC (Cys).

- The "CDR1 region" can have four different lengths, either 24 nt (8 aa) (possible in the Vk1 and Vk3 families); 27 nt (9 aa) (possible in the Vk3 family); 39 nt (13 aa) (possible in the Vk2 family); or 42 nt (14 aa) (possible in the Vk4 family); the first codon (aa) is always AGA (Arg),

30 and the penultimate codon (aa) of the CDR1 region is still: CTG (Leu).

- The framework region 2 always has 33 nt (that is, 11 aa); the first two codons (aa) of FR2 are always: TGG.TAC (Trp Tyr), the front

-15-

last codon is always CCG (Pro).

- The "CDR2 region" still has 33 nt (i.e. 11 aa) the first three codons (aa) still being CTA.TTA.ATT (Leu Leu Ile).

- The framework region 3 still has 96 nt (that is, 32 aa) the first three codons (aa) of FR3 are still: GGG.GTC.CCG (Gly Val Pro), and the last three codons are still TAT.TAT.TGC (Tyr Tyr Cys).

- The "CDR3 region" always has 24 nt (i.e. 8 aa) the second codon (aa) always being CAG (Gin).

- The timber frame region 4 always has 39 nt (that is 13 aa), and is identical in four families: ACC.TTT.GGC.CAG.GGT.ACG.AAA.GTT. GAA.ATT.AAA.CGT.ACG.

VL lambda chain:

- The framework region 1 goes from nt 1 to 66 (that is 22 aa), the last codon (aa) of FR1 is always: TGT (Cys).

- The "CDR1 region" can have three different lengths, 33 nt (11 aa) (possible in the VI3 family); 39 nt (13 aa) (possible in the VU family); or 42 nt (14 aa) (possible in the VU and VI2 families).

- The framework region 2 always has 33 nt (that is, 11 aa), the first two codons (aa) of FR2 are always: TGG.TAC (Trp Tyr), and the antepenultimate codon is always GCG (Ala).

- The "CDR2 region" still has 33 nt (i.e. 11 aa) the third codon (aa) still being ATT (Ile), and the last three codons still being CGT.CCC.TCA (Arg Pro Ser).

- The framework region 3 always has 96 nt (that is, 32 aa) the first codon (aa) of FR3 always being: GGC (Gly) and the last four codons always being GAT.TAT.TAT.TGC (Asp Tyr Tyr Cys).

- The "CDR3 region" can have three different lengths, 24 nt (8 ay); 27 nt (9 ay); or 30 nt (10 ay).

- The framework region 4 always has 39 nt (that is, 13 aa) and is identical in the three families: GTG.TTT.GGC.GGC.GGC.ACG.AAG.TTA.

ACC.GTT.CTT.GGC.CAG.

>

- 16-

The polypeptide according to the invention can be used for therapeutic or diagnostic purposes, for example for in vitro or in vivo diagnosis.

5. For therapeutic applications, the polypeptide according to the invention can be in the form of a conjugate with a therapeutic active ingredient, chosen, for example, from radiotherapeutic or chemotherapeutic products, such as low molecular weight cytostatic or cytotoxic active ingredients, or biological agents. The conjugation of the therapeutic active ingredient and the polypeptide can be achieved by known methods, preferably by covalent coupling to reactive amino, carboxy, hydroxy, and/or thiol groups of the protein, possibly using homo- or heterobifunctional linkers.

y ■

15 Furthermore, the polypeptide can also exist as a fusion protein, which, in addition to the antibody (for example, an IgG molecule or a fragment thereof), contains a cytokine fused to it, such as IL-2, IL-11,12, or the TNF-α polypeptide. Moreover, the fusion protein can exist as a bispecific antibody, in which case, in addition to binding to the ED-B domain, binding to another antigen is preferred. Other specificities of bispecific antibodies include binding domains against chelating agents for radionuclides of diagnostic and/or therapeutic importance, for example a, p or y radiation sources, notably 90Y, diagnostic NIR 25 (near-infrared) dyes, dyes with therapeutic effect, surface molecules on immunological effector cells (e.g. NK cells, cytotoxic T cells or NKT cells), integrins of importance in angiogenesis, in particular binding domains that block their functions (e.g. ai, P3, cciP3, 30 C12P1, CX2P2), inactivating anti-VEGF binding domains and inactivating binding domains against VEGF receptors 1, 2 and 3.

- 17-

For diagnostic applications, the polypeptide may be presented as a conjugate to a diagnostically detectable labeling group, for example, a labeling group for in vitro or in vivo diagnosis. Examples of labeling groups include the five radioactive labeling groups, NMR labeling groups, dye labeling groups, enzymatic labeling groups, and fluorescence labeling groups (e.g., near-infrared fluorescence).

For therapeutic applications, the polypeptide is preferably formulated as a pharmaceutical composition containing, as the active ingredient, the polypeptide according to the invention, optionally other active ingredients, and excipients, adjuvants, and/or diluents commonly used in pharmacology. The pharmaceutical composition contains the active ingredient in a dose with a therapeutic effect, which can be easily determined by a person skilled in the art through in vitro tests, for example, on suitable cell cultures or animal models. The composition is preferably administered by injection or infusion, but other methods of administration are possible. Preferably, intravenous and/or subcutaneous administration is used. The dose of the active ingredient administered depends on the nature and severity of the disease, as well as the condition of the patient being treated. Preferably, the therapeutic composition is administered several times over a long period of time, for example, at least 2 to 4 weeks. In this regard, reference is made to known methods of administering antibodies or antibody conjugates, such as those described by example in Ferrarra, N. et al., Nature Reviews Drug Discovery, Volume 3, May 2004, 391-400 and Salgaller, M.L., Current Opinion in Molecular Therapeutics 2003 5(6):657-667, or to existing protocols for administering pharmaceutical antibodies, including the Rituximab, CAMPATH, Remicade protocols, etc.

The invention also relates to a diagnostic composition comprising, as a diagnostic reagent, a polypeptide according to the invention. Furthermore, the

30

- 18-

The diagnostic composition may also contain other reagents, excipients, adjuvants, and/or diluents commonly used in diagnostics. The diagnostic composition contains the polypeptide in sufficient quantity to allow for diagnostic detection according to the chosen format, for example, an in vivo or in vitro diagnostic format. In this regard, reference is made to known in vivo and in vitro diagnostic methods using labeled antibodies.

The pharmaceutical and diagnostic formulation can be used for the treatment and diagnosis of all diseases related to ED-B expression, for example, stromal and/or perivascular ED-B expression. Examples of such diseases include hyperproliferative disorders, such as angiogenesis-associated diseases, particularly cancer, retinal diseases, hypertrophic scars, etc.; diseases associated with myofibroblast dysfunction, such as endometriosis, arteriosclerotic plaques, etc.; and inflammatory diseases, such as psoriasis, Crohn's disease, rheumatoid arthritis, or multiple sclerosis.

The pharmaceutical or diagnostic composition according to the invention may contain one or more effective polypeptides, for example a combination of polypeptides that bind to different sectors of the ED-B domains. The compositions are suitable for use in human and veterinary medicine.

The invention also relates to a nucleic acid that codes for a polypeptide or a fusion polypeptide according to the invention. This nucleic acid may be single-stranded or double-stranded DNA, or RNA. The nucleic acid is preferably in operational linkage with an expression regulatory sequence, which allows expression in a suitable host cell or in a suitable host organism. The nucleic acid may be

- 19-

The vector is presented on a suitable vector for introduction into a host cell or host organism. The vector may be, for example, a prokaryotic vector, suitable for introducing prokaryotic cells, such as a plasmid or bacteriophage. Alternatively, the vector may be a eukaryotic vector, suitable for introduction into eukaryotic host cells or eukaryotic host organisms, such as a plasmid, an artificial chromosome, or a viral vector. Suitable vectors are known to those skilled in the art, for example, from Sambrook et al. (1989), Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, and Ausubel et al. (1989), Current Protocols in Molecular Biology, John Wiley and Sons.

Another object of the invention relates to a cell, for example a prokaryotic cell or a eukaryotic cell, in particular, for example, a human cell, which is transformed with a nucleic acid according to the invention or with a vector according to the invention. Another object of the invention relates to a non-human organic, for example, a transgenic animal, which is transformed with a nucleic acid according to the invention or with a vector according to the invention. In the context of the present invention, the term "transformation" encompasses all possibilities of introducing foreign nucleic acids into a cell or an organic, including transfection or infection.

The polypeptide according to the invention can be obtained by culturing a cell according to the invention or a non-human organic according to the invention, under conditions in which polypeptide expression occurs, the operation being followed by obtaining the expressed polypeptide, for example from the cell, the culture medium, the organism or the excretory products of the organism.

The invention will be explained in more detail with the help of the examples and figures below:

-20-

Figure 1 presents the schematic implementation of an inhibition assay designed to identify antibodies with functional activity.

5. Human dermal microvessel endothelial cells (HDMVEc) obtained from skin microvessels are incubated alongside ED-B-specific Fab antibody fragments. The number of bound cells is visualized by crystal violet staining and measured with a photometer. High color intensity (10) represents a large number of adherent cells and the absence of binding-blocking antibodies. Low color intensity represents weak adhesion and the presence of a binding-blocking antibody.

Figure 2 presents a screening scheme to identify antibodies 15 blocking the function of fibronectin ED-B, by presentation of the Fab fragment.

S

In the first step, phage selection is performed using recombinant ED-B. The antibody fragments obtained in this way are subjected to a specificity assay (ELISA), a functional adhesion assay, and immunohistochemical examination. The affinity of the antibodies that pass these assays is then determined. Next, to improve affinity, modifications are made to the CDR domains: an affinity maturation of 1 corresponds to a modification of the L-CDR3 domain, and an affinity maturation of 25 corresponds to a modification of the H-CDR2 domain.

After the antibodies obtained by affinity maturation have performed the tests mentioned above, an examination of the in vivo activity is carried out.

30 V)

Figure 3 shows the results of a specificity assay in ELISA format, with Fab antibody fragments produced by phage selection with

recombinant ED-B.

ED-B denotes binding to recombinant ED-B, 6-FN represents binding to the recombinant fibronectin domain 6, domain 7-8-9 denotes binding to the 7-8-9 domains of recombinant fibronectin (without ED-B), and domain 7-EDB-8-9 denotes binding to the 7-8-9 domains of recombinant fibronectin with ED-B. The ratio of ED-B to 6-FN or 7-8-9 gives the specificity of the antibody being studied. AP-39 denotes a biologically inactive covalent dimer of the L19 anti-ED-B scFv antibody.

Figure 4 shows the result of an adhesion test with the antibodies studied.

The inhibition of HDMVEc cell adhesion to ED-B coated plates by the indicated Fab fragments at Fab concentrations of 25 and 0.4 pg/ml, respectively, is studied. The values indicated are the results of a triple determination.

Figure 5 presents the immunohistochemical reaction model of function-blocking anti-ED-B Fab fragments on the example of MOR 02616. Figures 5A and 5C present the results obtained with a negative control (HuCAL-Fab negative), and Figures 5B and 5D present the results obtained with the MOR 02616 antibody on MRI xenografts of human neuroblastoma cells (Figures 5A and 5B), and on murine F9 teratocarcinoma cells (Figures 5C and 5D).

Cryosections of the cells (10 µm thick) are air-dried and then fixed for 10 minutes in ice-cooled acetone. The sections are then washed in PBS and incubated for 60 minutes at room temperature with 2 pg/ml of MOR 02616 or a negative control. As a secondary antibody, a peroxidase-labeled polyclonal goat anti-human F(ab)2 antiserum (diluted 1:100 in PBS) is used.

-22-

Dianova) in combination with diaminobenzidine (Sigma Chemicals), as a chromogenic substrate.

Figure 6 shows the results of a specificity assay in ELISA format, 5 with maturation-optimized Fab antibody fragments.

v

The test is implemented as described in Figure 3.

Figure 7 shows the immunohistochemical reaction model of the 10th fragment of the Fab antibody MOR 03255 optimized by maturation on cryostat sections of F9 teratocarcinoma cells (Figure 7A), and on SKMEL-28 human melanoma cell xenografts on naked mice

(Figure 7B).

> —

15 The test is carried out as described in Figure 5.

Figure 8 shows the result of an adhesion test with the Fab antibody fragments according to the invention, by comparison with the comparator antibodies L19 and Ly60.3.

20 ^ —-— *

The test is performed in a manner similar to that described in Figure 4. The antibody concentrations are respectively 1, 10 and 20 µl/ml.

25 Figure 9 shows the stability of the Fab antibody fragments according to the invention after 4 hours of incubation in human plasma or PBS at 37 °C.

Immunoreactivity is determined in an ELISA assay, with

30 concentrations ranging from 0.039 to 5 pg/ml. The signal is defined as 0.62 pg/ml in unincubated human plasma as the 100% value.

-23-

Figure 10 presents the therapeutic activity of anti-ED-B Fab antibody fragments with function-blocking effect on the example of MOR 03255.

—

Figure 11 shows the nucleotide sequences encoding the 5 VH and VL regions of the antibodies according to the invention.

b—

Example

1. Materials and methods

10 ^ o —

1.1 Proteins and antibodies

Y •

We prepare according to standard operating procedures (e.g. Sambrook et al., Molecular Cloning. A Laboratory Manual, Cold Spring Harbour Press) 15 the ED-B (His)6 domain, the 6 (6FN) domain, the 7-8-9 domain of recombinant and purified fibronectin, and the 7-ED-B-8-9 domain.

Mouse antibodies against the purified ED-B receptor are also prepared using standard procedures. This involves administering 20 µg of the purified receptor to a mouse in Freund's complete adjuvant, followed three weeks later by 25 µg of the purified receptor in Freund's incomplete adjuvant, and then, three weeks after that, all administrations are performed intraperitoneally in the incomplete adjuvant. Fourteen days after the third immunization, blood is collected via the retroorbital sinus. Fibronectin from the plasma is obtained from Sigma.

1.2 Selection of ED-B specific phages

y

Selection is performed using the HuCAL Gold Phage selection protocol. The HuCAL® Gold library is subdivided into six specific framework pools. ED-B specific phages are enriched in three successive selection steps on a recombinant protein immobilized in

24

96-well plates (Maxisorb, Nunc, Rochester, NY, USA). For selection A, the plate was coated for 2 hours at 37°C with 1 µg per well of ED-B (His6) (10 µg/ml in Ampuwa®). For selection B, the plate was coated overnight at 4°C with ED-B at the same concentration in PBS (pH 7.2 with 1 mM MgCh/1 mM CaCh). The plates were blocked with 5% skimmed milk powder in PBS, 0.05% Tween20 (Sigma, St. Louis, MO, USA) (blocking buffer), and incubated with 1 x 10¹³ HuCAL® Gold phages, which had been pre-incubated with a volume of the blocking buffer. For selection B, the phages were pre-incubated with an additional 0.5 pg/ml of fibronectin domain 6 to reduce The selection of binding molecules, which are specific to the conserved structure of a fibronectin Type 3 domain repeat, is carried out. After 2 hours of incubation with the phages at room temperature, the wells are washed five times with PBS, 0.05% of 15 Tween20, and five times with PBS. The remaining phages are eluted for 10 minutes at room temperature with 20 mM dithiothreitol (DTT) in 10 mM Tris-HCl at pH 8.0. Incubation with E. coli TG-1 (Stratagene, DOeoo = 0.5) is then performed to ensure infection. The wells are then incubated with E. coli TG-1 as a further elution step.

wmc.L.w.. w m., jhaqémide, phage amplification and purification

HuCAL® Gold phages are amplified in 2x TY medium with 34 pg/ml chloramphenicol and 1% glucose (2x TY-CG). After infection with a 25-phage assistant (VCSM 13) at 37°C, with an OD600 of approximately 0.5, followed by centrifugation and resuspension in 2x TY-CG/50 pg/ml kanamycin, the cells are induced with 0.25 mM IPTG and cultured overnight at 22°C. The phages are precipitated with polyethylene glycol from the supernatant (Ausubel et al. (1998), Current 30 Protocols of Microbiology), resuspended in PBS, and used for subsequent selection steps.

20

-25-

1.4 Subcloning of selected Fab fragments and expression of soluble Fab fragments

The Fab-encoding insertions of selected HuCAL® Gold phages are subcloned into the pMORPHx9-FS expression vector. Plasmid DNA from the selected HuCAL® Gold clones is cleaved with the restriction enzymes Xball/EcoRI, with extraction of the Fab-encoding insertion (ompA-VL and phoA-Fd). The Fab molecules expressed in this vector contain two C-terminal markers (FLAG™ and Strep-tag1) for detection and purification.

^1.5 Expression and purification of HuCAL® Gold antibodies in E. coli

—

Expression of pMORPHx9-FS-encoded Fab fragments in E. coli TG-1 F cells is performed in shake-flask cultures with 0.75 µL of 2x TY medium and 34 pg/ml of chloramphenicol. After induction with 0.75 mM IPTG, the cells are cultured for 16 h at 30°C. Alternatively, Fab clones obtained from the second maturation pool 2 are induced with 0.1 mM IPTG before being cultured at 22°C. Periplasmic extracts of cell pellets are prepared by osmotic shock, and Fab fragments are isolated by Strep-Tactin® chromatography (IBA, Göttingen, Germany). Apparent molecular weights are determined by size-exclusion chromatography (SEC) with calibration standards. Concentrations are determined by

25 UV spectrophotometry.

1.6 Identification of Fab fragments binding to ED-B by ELISA

v> ~

96-well Maxisorb ELISA plates are coated overnight at 30°C with 100 µl of an ED-B solution (10 µg/ml in coating buffer (PBS pH 7.4, 1 mM CaCl₂, 1 mM MgCl₂)). Crude lysates or purified Fab molecules are added; non-binding Fab molecules are then

The fragments were removed by five washes with a wash buffer (PBS pH 7.4, 0.05% Tween20). Fab fragments were detected by incubation with anti-human Fab-peroxidase antibody conjugates (Dianova), followed by development with a soluble peroxidase substrate (Roche) and measurement at 370 nm. Clones expressing ED-B-specific Fab molecules were identified by a positive ELISA signal on immobilized ED-B, compared to a zero or only a weak signal in the case of 6FN.

1.7 Determination of plasma stability under physiological temperature conditions by ELISA

The 2.5 pg/ml ED-B coating is carried out essentially as described above. The Fab fragments are incubated for 4 hours at 37°C and a concentration of 50 pg/ml in human plasma (Deutsches Rotes Kreuz; Lot 9985550). After incubation, the Fab molecules are diluted to concentrations of 5.0, 2.5, 1.25, 0.62, 0.31, 0.156, 0.078, and 0.039 pg/ml in PBS with 1% bovine serum albumin for the ELISA assay, to determine the linear signal intensity zone. Functional Fab molecules are determined with the anti-Flag M2 antibody (Sigma F3165) and with an anti-mouse alkaline phosphatase secondary antibody conjugate, then developed with Attophos (Roche) and measured at 535 nm.

1.8 HDMVEc cell culture and cell adhesion assay

Human dermal microvascular endothelial cells (HDMVEc), isolated from the juvenile foreskin, are cultured in EGM-MV medium (Clonetics Inc.) with the addition of 10% fetal calf serum, 2 mM glutamine, 20 µg/ml heparin, and 3 ng/ml bFGF in 0.1% gelatin-coated vessels (Sigma). For adhesion assays, cells that have not been cultured beyond passage 8 are used. 10 µg/ml ED-B or 2.5 pg/ml plasmid fibronectin are immobilized overnight at 4°C in PBS pH 7.4, 0.9 mM CaCk, and 0.5 mM MgCb in [unclear - possibly "unclear"].

-27-

96-well ELISA plates are prepared and blocked for one hour at 37°C with 1% RSA in PBS pH 7.4, 0.9 mM CaCl2, and 0.5 mM MgCl2. Cells are labeled with 0.15 pg/ml calcein for 30 minutes at 37°C and washed once in adhesion medium (MCDB 131, 2 mM glutamine, 0.1% RSA, and 20 pg/ml heparin). One × 10⁵ cells containing the antibody concentration specified for each case, diluted in adhesion medium (final volume 100 µl), are incubated for 20 minutes at 37°C. The cell suspension is then placed in ED-B coated plates for two hours at 37°C. After two washes of the cells in PBS pH 7.4, CaCl2 0.9 mM, MgCl2 0.5 mM, adherent cells are detected in a plate fluorometer (excitation 485 nm, emission 530 nm).

As a control, (A) cell binding to ligand-coated plates without the addition of specific antibodies is determined, and (B) cell adhesion to wells without ligand coating is determined, with the value obtained in this latter control assay being defined as the minimum cell adhesion.

1.9 Affinity maturation of selected Fab molecules, by step-by-step exchange of CDR cassettes

To increase the affinity and biological activity of selected antibody fragments, CDR regions are optimized by cassette mutagenesis, using trinucleotide-directed mutagenesis.

To this end, Fab fragments from the pMORPHx9 expression vector are cloned into the pMORPH-23 phagimide vector using EcoRI/Xbal restriction sites. To optimize the affinity of the selected Fab fragments, two successive maturation steps are performed for each maturation pool. In the first step, a phage library of antibody fragments is generated, in which the L-CDR3 region of the starting clone is varied using a 7 x 10⁸ repertoire (Pool).

-28-

1) and 3.8 x 10⁸ (Pool 2) individual short-chain CDR sequences. The derivatives with improved affinity, obtained in the first maturation step, are then subjected to a second maturation step by replacing the H-CDR2 region with a library of diverse elements. In this way, two phage libraries are produced, each containing 1 x 10⁸ individual clones.

Affinity maturation libraries are produced by transformation in E. coli TOPIOF. Phages are prepared as described above. Selection for Fab fragments with improved affinity is performed under stringent conditions for three operations in the first maturation step, and for two operations in the second step.

1.10 Determination of affinity by surface resonance of plasmon 15 (BIAcore®)

v——

To determine the Ko, monomeric fractions (monomer content at least 80%, determined by size-exclusion chromatography; Superdex75, Amersham Pharmacia) from 20 Fab fragments are used. F1 chips (Biacore, Sweden) containing approximately 200 UR of ED-B (30 pg/ml, 10 mM acetate buffer, pH 4.0) and reference flow cells are coated with a corresponding amount of human serum albumin (20 pg/ml, 10 mM acetate buffer, pH 4.5) using a standard EDC-NHS amine coupling technique. The antigen density is reduced to approximately 100 UR for characterization of the second-maturation Fab. Regeneration is performed with 10 mM HCl. All kinetic measurements were performed in PBS buffer (NaCl 136 mM, KCl 2.7 mM, Na₂HPO₄ 10 mM, KH₂PO₄ 1.76 mM, pH 7.4) at a flow rate of 20 µL/min using a 30 Fab concentration range from 1.5 to 500 nm. The injection duration was 1 minute in all cases. All sensograms were evaluated using BIA 3.1 evaluation software.

-29-

Abbreviations: EDC: 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, NHS = N-hydroxysuccinimide, UR = resonance units.

2.1 Preparation of antibodies against human ED-B

Two different phage selection operations are performed. The ED-B coating for the first operation (selection A) is carried out according to conditions known in biological assays, to present the ED-B in a conformation that allows HDMVEc cells to adhere to the immobilized antigen. During the second selection (B), the phages are further blocked with an excess of 6FN to prevent the selection of binding molecules that would cross-react with fibronectin domain 6.

Overall, 23 different HuCAL-Fab molecules were identified from 1315 primary candidates that met the criteria for binding to ED-B and not to 6FN. Furthermore, the antibodies were subjected to effects to determine their ability to bind to fibronectin expressed under recombinant conditions, comprising domains 7, 8, and 9, but lacking the ED-B domain in its natural arrangement (see Figure 3). All the antibodies studied exhibited specific recognition for the 7-ED-B-8-9 construct, but did not react with the corresponding construct lacking the ED-B domain.

2.2 Functional characterization of ED-B specific binding molecules

The lamented anorectal cells, identified as specific by the ELISA technique, are purified, and their ability to block cell adhesion is determined.

5.2. Results

-30-

HDMVEc to ED-B coated plates. The test focuses on the adhesion of HDMVEc cells in the presence of ED-B specific Fab fragments, at two concentrations, by comparison with a non-specific HuCAL control Fab molecule and positive mouse control sera.

Twelve of the 23 Fab molecules described in section 2.1 exhibit inhibition of cell adhesion to immobilized ED-B. Figure 4 shows a representative selection of the six most effective Fab molecules. At a concentration of 25 pg/ml, cell adhesion is inhibited by 50% to nearly 100%. At a concentration of 0.4 pg/ml, only a few Fab molecules, for example MOR02610, MOR02616, MOR02715, and MOR02718, exhibit reproducible inhibition of cell adhesion.

Function-blocking Fab antibody fragments were characterized immunohistochemically for their localization on F9 mouse teratocarcinoma cells and cryostat sections of human IRM30 neroblastoma xenografts. All antibodies studied showed vascular localization on both tissue sections, but no vessel staining on normal tissues (e.g., normal liver and skin, and sclera). Figures 5B and 5D show perivascular abluminal staining of vessels, proliferating endothelium, and tumor stroma by MOR02616. ED-B accumulation was observed near neovascular structures in exponentially growing tumors, but not on normal vessels. With a lysozyme-negative control antibody, at the same concentration, no staining was visible (Figures 5A and 5D).

The properties of twelve selected linker molecules were determined by BIAcore analysis (see Table 1). Affinities (Kon) ranged from 15 to 500 nm. The affinity of the divalent derivative L19AP39 (a

5C).

-31 -

The comparative anti-ED-B antibody (biologically inactive) was found to be 2.4 nM. Table 1 also shows the CDR sequences (only L-CDR3 and H-CDR3) and scaffold combinations. At least two independent measurements were performed with Fab molecules from different reaction masses for protein expression and purification. The binding ratio (Kon) and dissociation ratio (Koff) of the Fab molecules are shown in separate columns.

Table 1

Clone

VH

H-CDR3

VL

L-CDR3

kon [1/Ms]

koff [1/ s]

KD [nM]

AP 39

VH3

PFPY--FDY

K3

QQTGRI- - PP

6.6E+05

1.5E-03

2.4 ± 0.6

MOR02610

; VH1B- .

SPVYYKYDY

• "1, *

QSYDKTSSTY-. .y

-.•Î.3E+Ô6

, 9.1E-02

75 ± 19;/

MOR02611

VH1B

GLYYR-FAS

L2

AAATG GW

1.5E+05

5.1E-02

332 + 23

MOR02613

VH3

YVNG--FDI

L2

QTYAKKDYSL

7.8E+05

1.3E-01

164 + 14

MOR02614

VH3

A YDV

L1

AMFSP EG

2.8E+05

4.2E-02

160 + 6

MÔR02616

VH3

<

VIVL--FDY

K3

ii'

LQKYSX-'-PF

-S^E+OS"

Z8È-k)2

59 ±2 5'

MORÛ2618

VH3

NYWV--FAY

L3

QSYDNFNDSV

4.3E+05

2.0E-01

477+177

MORÔ2619:

: . VH4

F -FDV '

~ Lti.

QSWDGAS-TG:..Aè

' 7.8E+05

**' A ^

J1,2Em

154±06 "

MOR02622

. VH3

GLVT--FDN >

L2

SSWTHSETDY.. S

2.4E+05

1 5,1E-03^

r '2T,3±-1.7

MOR02715

VH3; :

NKVG--FDV.

• L2,„

QAWDNQGMKY

8.9E+05

4.7E-02,

53.7 tî, 7

MORÛ2718

: . VH3

GWF---FAH

Ka-

FQYSSV--PL

4.5E+05s

•. 3v9E-02

8i5;±i2;5v;

MOR02721

VH5

YH GAF

L1

QAYTTG-- SI

4.3E+05

2.4E-02

56 ±11.6

MORQ2722

VH3

F IAS

K2

QQYSNF--PF

1.5E+05

1.4E-02

94 + 2.3

2.3

Affinity maturation of function-blocking anti-ED-B antibodies by optimization of L-CDR3 and H-CDR2 regions

The affinity maturation of Fab molecules is carried out in two steps. First, the L-CDR3 sequence is diversified, then the H-CDR2 sequence of the improved Fab molecule, obtained in the first step, is replaced.

Two affinity maturation libraries are generated for each step. Optimization is performed in two separate pools. The L-CDR3 1 library, with a diversity degree of 7 x 10⁸ elements, contains the four starting Fab molecules MOR02610, MOR02616, MOR02619, and MOR02622. The L-CDR3 2 library, with a diversity degree of 3.8 x 10⁸ elements, contains only derivatives of MOR02715 and MOR02718. The Fab molecule is then grouped in each case according to its affinity and biological activity in the adhesion assay.

The two Dioiiotneques are kept separate during the selection procedure. Two phage selection strategies are then employed. Selection 1 uses ED-B immobilized on Maxisorb plates for three operations, as described above. For selection 2, selection is performed on biotinylated ED-B in solution. The phage-antigen complex is then recaptured by streptavidin-coated particles in the first operation, and by neutravidin-coated plates in the following two operations. Stringent conditions are used in the selection procedure.

Screening of optimized Fab molecules is performed by determining K0ff values in the BIAcore system. A total of 11 derivatives from library 1 were characterized. The five derivatives with the highest affinity are listed in Table 2. For all improved clones, these are derivatives of MOR02619, with the affinity multiplied by a coefficient of up to 7. With clone MOR03075, an affinity of

-33-

of 2.4 nM.

V

10

■>

Two derivatives of MORÛ2715 are analyzed in detail from library 2. For clone MOR03062, a 15-fold increase in affinity is found, resulting in an affinity for monomers of 2.4 nM. The variations in the amino acid sequence of the L-CDR3 region are recorded in Tables 2a and 2b.

Table 2a

Pool deMORO

selection

1

Clone LCDR3 kon strain [1/Ms]

koff [1/s] KD [nm]

multiplication coefficient relative to the Kon Koff KD clone

AP

Biotin

39-

QQTGRI - - PP

6.6E + 05

1.49E

-003 2.4 ±0.6

2619

QSWDGAS-TG

7.8E + 05

1.2E -

02

15.4 ±0.6

solid

3055

2619

QAWTRAHRYP

1.8E +06

5.1E-

03

3.0 ± 0.5

2.2

2.6

5.4

dissolved

3066

2619

SSYD-TQVTR

1.1E +06

4.7E-

03

4.2 ± 0.6

1.4

2.8

3.8

dissolved

3075

2619

QSWDP-RSFT

1.1E +06

2.6E -

03

2.4 ± 0.3

1.4

5.0

6.8

dissolved

3069

2619

WTGM - -SYHF

1.2E +06

4.1E-

03

3.3 ± 0.2

1.5

3.2

4.8

dissolved

3071

2619

LAYIQS-KGH

1.9E +06

5.7E -

03

3.1 ± 0.8

2.3

2.3

5.1

w. Table 2b

Pool deMORO

Clone

LCDR3

kon [1/Ms] koff

[1/S] KD

coefficient

selection 2

strain

[nm]

multiplier the improvement by

relation to the clone strain

kon koff

KD

AP

39-

QQTGRI - - PP

6.75E + 05

1.49E

-03

2.4 ± 0.6

Biotin

2715

QAWDNQGMKY

9.9E + 05

4.7E -

02

53.7 ±1.7

solid

3064

2715

QSWDLLAPSV

1.5E + 06

1.4E -

02

10 ±3.5

1.7 3.5

5.3

solid

3062

2715

QSWDLSVHQV

3.5E + 0.6

1.1E -

02

3.4 ± 0.8

4.0 4.2

15.8

15 _ The specificity for ED-B (determined by ELISA, immunohistochemistry and cell adhesion assay) of all derivatives from the first maturation operation remains unchanged.

The Fab molecules presented in Tables 2a and 2b were selected for H-CDR2 maturation, which was subsequently carried out in two pools. In pool 1, five Fab molecules, having similar activity but different L-CDR3 regions, were subjected to diversification in

-34-

H-CDR2 (library size 7 x 10⁷ elements). For pool 2 (library size 8.5 x 10⁷ elements), two Fab molecules are selected. The libraries are selected separately, as described above. Binding molecules with increased affinity are enriched by two phage selection operations in solution under stringent conditions.

From pool 1, three Fab molecules are selected. For derivatives MOR03243 and MOR03245, affinities of approximately 0.7 nM are found. For derivative MOR03246, an affinity of 0.6 nM is found.

The H-CDR2 sequences of the corresponding clones, as well as the affinity data, are shown in Table 3a.

15 - Table 3a

_ ^ MORO

Parental clone;

" .HCDR2-Sequence '

f _ "•» w -,

^ -, 'v i kon [1/Ms]

£ t 'F -J

V- i

* koff,[1/s]

KD [nM] - *

MORÛ3243

3055

EIHRGGYTQYNP S LKS

2.9E+06

1.8E-03

0.7 ± 0.2

MONCI3245

3069

VIHKUGF "INYWPS LKS

3.5E+06

1.6E-03

0.7 ±0.3

MORQ3246

3075

YIHKYGUTKYNP S LKS

4.8E+06

3.0E-03

0.6 ±0.1

^ We can also select derivatives from pool 2 that have

- improved properties. Compared to the original clones, we have a

- improvement, corresponding to a multiplier of 26 to 250, of the 20 - affinities of these derivatives. The affinities of clones MOR03255 and MORÛ3258

are found to be 30 and 40 pM respectively.

^ Table 3b presents the amino acid sequences of H-CDR2 from selected clones of pool 2, along with the corresponding affinity data 25.

-35

Table 3b

'Clone' .parental^

;^^nce^CDR2^

pUJ&.Cf-as"A •kon [1/Ms]v

î>. • >-

fikoff;[17s]i

-j-l'"!5: 'f gKDj[nMl

STTDEW

MOR03251

3062

VISNMSYTIYYADSVKG

3.3E+06

2.0E-04

0.07

0.04

MORC13252

3062

VISNYSWHIYYADSVKG

3.1E+06

6.1E-05

0.03

0.03

MOR03253

3062 Mut

VISNMGFEIYYADSVKG

1.6E+06

2.0E-04

0.13

0.05

MOR03255

3062

VISNRSSYIYYADSVKG

4.9E+06

8.6E-05

0.03

0.03

MOR03257

3062

VISNRGNYIYYADSVKG

5.3E+06

3.0E-04

0.08

0.06

MORQ3258

3064

VISNQSNYIYYADSVKG

2.7E+06

8.8E-05

0.04

0.02

10

15

20

S

25

2.4 Characterization of anti-ED-B antibodies with function-blocking and high affinity

To verify whether the antibodies obtained by affinity maturation remain specific for ED-B, specificity tests are performed using ELISA (Figure 6), immunohistochemical analyses (Figure 7), and adhesion tests (Figure 8). In all three tests, the specificity of the affinity-matured Fab molecules for ED-B remains unchanged. Furthermore, Figure 8 shows that both the stem clones, such as MOR 02715, and the clones obtained by affinity maturation, such as MOR 03062, MOR 03255, MOR 03064, and MOR 03259, exhibit significantly higher biological activity (inhibition of adhesion) than the known antibody L19.

In addition, the thermal stability of the Fab molecule in human plasma is studied. To this end, the Fab molecules are incubated with human plasma for 4 hours at 37°C, without observing any significant loss of immune reactivity, as determined by ELISA (Figure 9).

I ^ **""* " ~ ""

All the selected Fab molecules are able to block the interaction of a receptor on human microvascular endothelial cells with ED-B in a dose-dependent manner, a property not exhibited by the known anti-ED-B antibody L19. As the adhesion of

-36-

The binding of HDMVEc cells to the extracellular matrix is one of the early stages of neoangiogenesis; blocking this process by the binding molecules according to the invention represents a therapeutic means to inhibit the formation of new vessels and to inhibit tumor growth.

F9 teratocarcinoma cells (10⁶/mouse in Matrigel with/without 100 pg of Fab MORÛ3255) were implanted subcutaneously in the flank of naked mice. After three days, each animal was treated intravenously for seven consecutive days with 100 pg MORÛ3255/mouse. Compared to the solvent-treated control, both treatment groups showed a 42% to 64% inhibition of tumor growth. Figure 10 summarizes the results of the 15 trials.

2.5 In vivo activity

Bibliography

-37-

(1)Brown LF, Guidi AJ, Schnitt SJ, Van De Water L, Iruela-Arispe ML, Yeo T-K, Tognazzi K, Dvorak HF (1999) Clin. Cancer Res. 5: 1041-1056

5 W

(2)Folkman J (1995) Nature (Med.) 1:27-31

(3)Risau W and Lemmon V (1988) Develop. Biol. 125:441–450 10(4)Van Den Hoff A (1988) Adv. Cancer Res. 50:159-196

15

(5) Folkman J and Shing Y (1992) J. Biol. Chem. 267: 10931-10934

(6)George EL, Baldwin HS, Hynes RO (1997) Blood 90: 3073-3081

(7) Bowersox JC and Sorgente N (1982) Cancer Res. 42: 2547-2551

(8) Madri JA, Pratt BM, Tucker AM (1988) J. Cell Biol. 106: 1375-1384

■ V-

20 (9)Nicosia RF, Bonnano E, Smith M (1993) J. Cell. Physiol. 154:654-661

^ —

(10)Kornblihtt AR, Vibe-Pedersen K and Baralle FE, 1983. Isolation and characterization of cDNA clones for human and bovine fibronectins. Proc. Natl. Acad. Sci. UNITED STATES, 80, 3218-22

25

(11)Leahy DJ, Hendrickson WA, Aukhil I and Erickson HP. 1992. Structure of fibronectin type III domain from tenascin phased by MAS analysis of the selenomethionyl protein. Science, 258, 987-91

30 (12)Hynes R.O., 1987, Cell 48, 549-550

(13)Plow E.F. et al. 2000, J. Biol. Chem., 275, 21785-21788

-38-

(14)Castellani P, Viale G, Dorcaratto A, Nicolo G, Kaczmarek J, Querze G, Zardi L (1994) Int. J Cancer 59:612-618

5(15)Hashimoto-Uoshima M, Yan YZ, Schneider G, Aukhil I (1997) J. Cell Science 110:227-2280

V

(16) Chen W and Culp LA (1996) Exp. Cell Res. 223:9-19

S-—

10 (17) Chen W and Culp LA (1998) Clin. Exp. Metastasis 16: 30-42

-39 -

Claims (1)

Demands Polypeptide, characterized in that, (i) it binds in a specific manner to the ED-B domains of fibronectin, And (ii) it inhibits the interaction between the ED-B domain and its receptor. Polypeptide according to claim 1, characterized in that it is an antibody or an antibody fragment. Polypeptide according to claim 1 or 2, characterized in that it has an affinity for the ED-B domain corresponding to a Kd value <100 nM. Polypeptide according to claim 3, characterized in that it has an affinity for the ED-B domain corresponding to a KD value <10 nM. Polypeptide according to claim 3, characterized in that it has an affinity for the ED-B domain corresponding to a K0 value <1 nM. Polypeptide according to any one of claims 1 to 5, characterized in that it has a specificity for the ED-B domain of fibronectin at least 5 times greater than its specificity for the 6 domain of fibronectin and/or for the 7-8-9 domains of fibronectin. Polypeptide according to any one of claims 1 to 6, characterized in that it exhibits in vitro an inhibition of the adhesion of ED-B to HDMVEc cells, tumor cells and/or stromal cells. Polypeptide according to any one of claims 1 to 6, characterized in that it exhibits in vivo inhibition of tumor growth produced by -40- implantation of F9 teratocarcinoma cells in experimental animals. 9. Polypeptide according to any one of claims 1 to 8, characterized in that it is selected from antibodies or antibody fragments, comprising (a) a VH region (i) encoded by a nucleic acid sequence SEQ ID NO. 1 (MOR 02610), SEQ ID NO. 5 (MOR 02611), SEQ ID NO. 9 (MOR 02613), SEQ ID NO. 13 (MOR 02614), SEQ ID NO. 17 (MOR 02616), SEQ ID NO. 21 (MOR 02618), SEQ ID NO. 25 (MOR 02619), SEQ ID NO. 29 (MOR 02622), SEQ ID NO. 33 (MOR 02715), SEQ ID NO. 37 (MOR 02718), SEQ ID NO. 41 (MOR 02721), SEQ ID NO. 45 (MOR 02722) or at least one H-CDR1, H-CDR2 and/or H-CDR3 region selected from one of the VH regions mentioned, or (ii) which is derived from a VH region according to (i) by modification of at least one H-CDR region, and/or (b) a VL region (i) encoded by a nucleic acid sequence SEQ ID NO. 3 (MOR 02610), SEQ ID NO. 7 (MOR 02611), SEQ ID NO. 11 (MOR 02613), SEQ ID NO. 15 (MOR 02614), SEQ ID NO. 19 (MOR 02616), SEQ ID NO. 23 (MOR 02618), SEQ ID NO. 27 (MOR 02619), SEQ ID NO. 31 (MOR 02622), SEQ ID NO. 35 (MOR 02715), SEQ ID NO. 39 (MOR 02718), SEQ ID NO. 43 (MOR 02721), SEQ ID NO. 47 (MOR 02722) or at least one L-CDR1, L-CDR2 and/or L-CDR3 region of one of the VL regions mentioned, or (ii) which derives from a VL region according to (i) by modification of at least one L-CDR region. 10. Polypeptide according to claim 9, characterized in that it comprises a VL region derived from a VL region according to (b)(i) by modification of -41 - the L-CDR3 region. > 11. Polypeptide according to claim 10, characterized in that it comprises a VL region which is derived from a VL region encoded by the nucleic acid sequence SEQ ID NO. 27 (MOR 02619) or SEQ ID NO. 35 (MOR 02715). 12. Polypeptide according to any one of claims 9 to 11, characterized in that it comprises (a) a VH region (i) encoded by the nucleic acid sequence SEQ ID NO. 25 (MOR 02619) or SEQ ID NO. 33 (MOR 02715) or at least one H-CDR1, H-CDR2 and/or H-CDR3 region of one of the VH regions mentioned, or (ii) which is derived from a VH region according to (i) by modification of at least one H-CDR region and/or (b) a VL region (i) encoded by the nucleic acid sequence SEQ ID NO. 49 (MOR 03055), SEQ ID NO. 51 (MOR 03066), SEQ ID NO. 53 (MOR 03075), SEQ ID NO. 55 (MOR 03069), SEQ ID NO. 57 (MOR 03071), SEQ ID NO. 59 (MOR 03064), SEQ ID NO. 61 (MOR 03062), or at least the L-CDR3 region of one of the VL regions mentioned, or (ii) which derives from a VL region according to (i) by modification of at least one L-CDR region. includes a VH region which derives from a VH region according to (a)(i) by modification of the H-CDR2 region. it. ruiyjjcjjnuc aciun ici i evci luxation 13, characterized in that it comprises a VH region which derives from the VH region encoded by the acid sequence 13. Polypeptide according to any one of claims 9 to 12, characterized in that it -42- nucleic acid SEQ ID NO. 25 (MOR 02619) or SEQ ID NO. 33 (MOR 027151 15. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises (a) a VH region (i) encoded by the nucleic acid sequence SEQ ID NO. 63 (MOR 03243), SEQ ID NO. 67 (MOR 03245), SEQ ID NO. 71 (MOR 03246), SEQ ID NO. 75 (MOR 03251), SEQ ID NO. 79 (MOR 03252), SEQ ID NO. 81 (MOR 03253), SEQ ID NO. 83 (MOR 03255), SEQ ID NO. 85 (MOR 03257), SEQ ID NO. 87 (MOR 03258) or at least the H-CDR2 region of this VH region, or (ii) which is derived from a VH region according to (i) by modification of at least one H-CDR region and/or (b) a VL region (i) encoded by the nucleic acid sequence SEQ ID NO. 65 (MOR 03243), SEQ ID NO. 69 (MOR 03245), SEQ ID NO. 73 (MOR 03246), SEQ ID NO. 77 (MOR 03251), SEQ ID NO. 89 (MOR 03258) or at least the L-CDR1, L-CDR2 or L-CDR3 region of one of the VL regions mentioned, or (ii) which derives from a VL region according to (i) by modification of at least iinf» rpninn I -P.nR 16. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 1 and the VL region encoded by SEQ ID NO. 3 (MOR 02610) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 17. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 5 and the VL region encoded by SEQ ID NO. 7 (MOR 02611) or at least one H-CDR1, H- -43- CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 18. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 9 and the VL region encoded 5 by SEQ ID NO. 11 (MOR 02613) or at least one H-CDR1, H- region CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 19. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 13 and the VL region encoded 10 by SEQ ID NO. 15 (MOR 02614) or at least one H-CDR1, H- region CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 20. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 17 and the VL region encoded 15 by SEQ ID NO. 19 (MOR 02616) or at least one H-CDR1, H- region ^ CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 21. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 21 and the VL region encoded 20 by SEQ ID NO. 23 (MOR 02618) or at least one H-CDR1, H- region CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 22. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 27 (MOR 02619) or at least one H-CDR1, H- CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. 23. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 29 and the VL region encoded by SEQ ID NO. 31 (MOR 02622) or at least one H-CDR1, H- CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 of this VH region. -44- 10 15 25 30 24. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 35 (MOR 02715) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 25. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 37 and the VL region encoded by SEQ ID NO. 39 (MOR 02718) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 26. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 41 and the VL region encoded by SEQ ID NO. 43 (MOR 02721) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 27. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 45 and the VL region encoded by SEQ ID NO. 47 (MOR 02722) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 20 V> 28. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 49 (MOR 03055) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 29. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 51 (MOR 03066) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. ^ — . 30. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded *>;-45-;by SEQ ID NO. 53 (MOR 03075) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;31. Polypeptide according to any one of claims 9 to 14, characterized in that it 5 comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 55 (MOR 03069) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;S;32. Polypeptide according to any one of claims 9 to 14, characterized in that it 10 comprises the VH region encoded by SEQ ID NO. 25 and the VL region encoded by SEQ ID NO. 57 (MOR 03071) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;33. Polypeptide according to any one of claims 9 to 14, characterized in that it 15 comprises the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 59 (MOR 03064) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;v> ,;34. Polypeptide according to any one of claims 9 to 14, characterized in that it 20 comprises the VH region encoded by SEQ ID NO. 33 and the VL region encoded by SEQ ID NO. 61 (MOR 03062) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 35. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 63 and the VR region encoded by SEQ ID NO. 65 (MOR 03243) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 36. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 67 and the VR region encoded by SEQ ID NO. 69 (MOR 03245) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;-46-;37. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 71 and the VL region encoded by SEQ ID NO. 73 (MOR 03246) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;38. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 75 and the VL region encoded by SEQ ID NO. 77 (MOR 03251) or at least one H-CDR1, H-;10 CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;39. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 79 and the VL region encoded by SEQ ID NO. 77 (MOR 03252) or at least one H-CDR1, H-;15 CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region.;40. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 81 and the VL region encoded by SEQ ID NO. 77 (MOR 03253) or at least one H-CDR1, H-;20 CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;41. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 83 and the VL region encoded by SEQ ID NO. 77 (MOR 03255) or at least one H-CDR1, H-;25 CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region.;42. Polypeptide according to any one of claims 9 to 14, characterized in that it comprises the VH region encoded by SEQ ID NO. 85 and the VL region encoded by SEQ ID NO. 77 (MOR 03257) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. 43. Polypeptide according to any one of claims 9 to 14, characterized in that it * -47- includes the VH region coded by SEQ ID NO. 87 and the VL region coded by SEQ ID NO. 89 (MOR 03258) or at least one H-CDR1, H-CDR2, H-CDR3, L-CDR1, L-CDR2 or L-CDR3 region of this VH region. S — 44. Polypeptide according to any one of claims 1 to 43, characterized in that it is in the form of a conjugate with a therapeutic active ingredient. Y, 45. Polypeptide according to claim 44, characterized in that the therapeutic active principle is selected from radiotherapeutic and chemotherapeutic products. V) 46. Polypeptide according to any one of claims 1 to 43, characterized in that it is in the form of a fusion protein. 47. Polypeptide according to claim 46, characterized in that it is in the form of a fusion protein with a cytokine, or in the form of a bispecific antibody. N> _ 48. Polypeptide according to any one of claims 1 to 43, characterized in that it is in the form of a conjugate with a diagnostically detectable labeling group. v, 49. Polypeptide according to claim 46, characterized in that the diagnostically detectable labeling group is selected from radiographic, NMR, dye, enzymatic and fluorescence labeling groups. 50. Pharmaceutical composition, characterized in that it comprises as an active ingredient a polypeptide according to one of claims 1 to 47, as well as adjuvants, excipients and/or diluents commonly used in pharmacology. -48- 51. Composition according to claim 50, characterized in that it is intended for the prevention or treatment of hyperproliferative diseases. 52. Composition according to claim 50 or 51, characterized in that it is intended for the prevention or treatment of cancer. v, - 53. Diagnostic composition, characterized in that it comprises as a diagnostic reagent a polypeptide according to any one of claims 1 to 43, 48 or 49. 54. Composition according to claim 53, characterized in that it is intended for the diagnosis of hyperproliferative diseases, or a predisposition to them. 55. Composition according to claim 53 or 54, characterized in that it is intended for the diagnosis of cancer or a predisposition to cancer. S 56. Composition according to any one of claims 50 to 55, characterized in that it is intended for use in human medicine. ^ —— 57. Nucleic acid characterized in that it codes for a polypeptide according to any one of claims 1 to 43 or 46 to 47. 58. Nucleic acid according to claim 57, characterized in that it is in operational linkage with an expression regulatory sequence. 59. Vector, characterized in that it contains a nucleic acid according to any one of claims 57 to 58. 60. Cell, characterized in that it is transformed by a nucleic acid according to any one of claims 57 to 58 or a vector according to the -49- Claim 59. 61. Non-human organism, characterized in that it is transformed by a nucleic acid according to any one of claims 57 to 58 or a vector according to claim 59. 62. A method for preparing a polypeptide according to any one of claims 1 to 43, characterized in that a cell according to claim 60 or a non-human organism according to claim 61 is cultured under conditions in which polypeptide expression takes place, and the expressed polypeptide is obtained. 63. Use of a polypeptide according to any one of claims 1 to 47, to prepare a medicinal product for the prevention or treatment of diseases associated with ED-B. 64. Use according to claim 63, for preparing a medicinal product for the prevention or treatment of cancer. 65. Use of a polypeptide according to any one of claims 1 to 43, 48 or 49 to prepare a detection reagent for the diagnosis of diseases associated with ED-B or a predisposition to such diseases. 66. Use according to claim 65 for preparing a detection reagent for the diagnosis of cancer or a predisposition to cancer.