RU2454427C2 - ANTI-panNrpA ANTIBODY (VERSIONS) - Google Patents

Abstract

FIELD: chemistry.

SUBSTANCE: invention relates to biotechnology and obtaining an anti-neuropilin antibody which binds with both Nrp1 and Nrp2.

EFFECT: obtained antibody is capable of inhibiting biological activity of semaphorin and can be used for therapeutic purposes.

15 cl, 20 dwg, 2 ex

Description

FIELD OF THE INVENTION

The present invention relates to the crystal structures of fragments of neuropilin 1 (Nrp1) and neuropilin 2 (Nrp2) both in the form of single fragments and in combination with antibodies to neuropilin, as well as their use. Additionally, the invention provides antibodies that bind to Nrp1 and / or Nrp2, and methods for their use.

BACKGROUND OF THE INVENTION

The development of the vascular system is a fundamental need for many physiological and pathological processes. Actively growing tissues, such as embryonic and tumor, require sufficient blood supply. They satisfy this need by producing pro-angiogenic factors that stimulate the formation and maintenance of new blood vessels through a process commonly called angiogenesis. Vascular formation is a complex but ordered biological phenomenon that includes all or many of the following steps: a) endothelial cells (EC) proliferate from existing ECs or differentiate from precursor cells; b) ECs migrate and coalesce to form heavy structures; c) then vascular cords undergo tubulogenesis, forming vessels with a central lumen; d) existing strands or vessels release processes, forming secondary vessels; e) the primitive vascular plexus undergoes further remodeling and structural adjustment; and f) periendothelial cells are enclosed, enclosing the endothelial tubes, which provides the vessels with a supportive and modulating function: such cells include pericytes for small capillaries, smooth muscle cells for larger vessels, and myocardial cells in the heart. Hanahan, D. Science 277: 48-50 (1997); Hogan, B. L. & Kolodziej, P. A. Nature Reviews Genetics. 3: 513-23 (2002); Lubarsky, B. & Krasnow, M. A. Cell. 112: 19-28 (2003).

It has now been established that angiogenesis is involved in the pathogenesis of many disorders. Such disorders include solid tumors and metastases, atherosclerosis, retrolental fibroplasia, hemangiomas, chronic inflammation, intraocular neovascular diseases such as proliferative retinopathies, such as diabetic retinopathy, age-related macular degeneration (AMD), neovascular glaucoma and other tissue tissues, rheumatoid arthritis and psoriasis. Folkman et al., J. Biol. Chem., 267: 10931-10934 (1992); Klagsbrun et al., Annu. Rev. Physiol. 53: 217-239 (1991); and Garner A., "Vascular diseases", In: Pathobiology of Ocular Disease. A Dynamic Approach, Garner A., Klintworth GK, eds., 2nd Edition (Marcel Dekker, NY, 1994), pp. 1625-1710.

In the case of tumor growth, angiogenesis is critical for the transition from hyperplasia to neoplasia, as well as for providing the nutrition necessary for tumor growth and metastasis. Folkman et al, Nature 339: 58 (1989). Neovascularization allows tumor cells to achieve growth efficiency and proliferative autonomy comparable to normal cells. The tumor usually begins with one aberrant cell, which can proliferate only to the size of several cubic millimeters due to the distance from the available capillary bed and can remain in a “sleeping” state for a long time without further growth and dissemination. The same tumor cells subsequently switch to an angiogenic phenotype, activating endothelial cells that proliferate and mature into new capillary blood vessels. Such newly formed blood vessels make it possible not only to continue the growth of the primary tumor, but also to disperse and recolonize metastatic tumor cells. In accordance with this, there is a correlation between the density of microvessels in sections of tumors and the survival of patients with breast cancer and a number of other tumors. Weidner et al, N. Engl. J. Med 324: 1-6 (1991); Horak et al, Lancet 340: 1120-1124 (1992); Macchiarini et al., Lancet 340: 145-146 (1992). The exact mechanisms governing the switch to angiogenesis are not well understood, but it is believed that neovascularization of the tumor mass is due to a direct numerical balance between stimulators and angiogenesis inhibitors (Folkman, 1995, Nat Med 1 (1): 27-31).

The process of vascular development is strictly regulated. To date, a large number of molecules have been identified, mainly secreted by surrounding cells, which regulate the differentiation, proliferation, migration, and EC fusion into heavy structures. For example, vascular endothelial growth factor (VEGF) has been identified as a key factor involved in stimulating angiogenesis and in inducing vascular permeability. Ferrara et al, Endocr. Rev. 18: 4-25 (1997). The discovery of the fact that the loss of even one VEGF allele leads to the lethal outcome of the embryo indicates an irreplaceable role of this factor in the development and differentiation of the vascular system. In addition, VEGF has been demonstrated to be a key mediator of neovascularization associated with tumors and intraocular diseases. Ferrara et al., Endocr. Rev. above. Studies have shown that most human tumors overexpress VEGF mRNA. Berkman et al., J. Clin. Invest. 91: 153-159 (1993); Brown et al, Human Pathol. 26: 86-91 (1995); Brown et al, Cancer Res. 53: 4727-4735 (1993); Mattern et al, Brit. J. Cancer 73: 931-934 (1996); Dvorak et al, Am. J. Pathol. 146: 1029-1039 (1995).

VEGF concentrations in the ocular fluid are highly correlated with the presence of active proliferation of blood vessels in patients with diabetic and other ischemia-related retinopathies. Aiello et al, N. Engl. J. Med. 331: 1480-1487 (1994). In addition, studies have shown that in patients affected by AMD, VEGF is localized in choroidal neovascular membranes. Lopez et al, Invest. Ophthalmol Vis. Sci. 37: 855-868 (1996).

Anti-VEGF antibodies inhibit the growth of many human tumor cell lines in nude mice (Kim et al, Nature 362: 841-844 (1993); Warren et al, J. Clin. Invest. 95: 1789-1797 (1995); Borgström et al, Cancer Res. 56: 4032-4039 (1996); Melnyk et al, Cancer Res. 56: 921-924 (1996)), and also suppress intraocular angiogenesis in models of ischemic retinal disorders. Adamis et al, Arch. Ophthalmol. 114: 66-71 (1996). Therefore, monoclonal antibodies against VEGF or other inhibitors of the action of VEGF are promising candidates for the role of agents for treating tumors and various intraocular neovascular disorders. Such antibodies are described, for example, in documents EP 817648, published January 14, 1998, and WO98 / 45331, and WO98 / 45332, published October 15, 1998. One of the antibodies against VEGF, bevacizumab (bevacizumab), was approved by the FDA for use in combination with chemotherapy for the treatment of metastatic colorectal cancer (CRC) and non-small cell lung cancer (NSCLC). Bevacizumab is also currently being investigated in many ongoing clinical trials for the treatment of various cancers.

During the development of the nervous system, neurons release axons (such as cables) that migrate long distances to reach their targets. See review of Carmeliet and Tessier-Lavigne (2005) Nature 436: 193-200. At the front end of the growing axon, there is a highly mobile sensory structure called the growth cone. Through dynamic cycles of stretching and retracting filopodial lengthenings, the growth cone constantly perceives and evaluates signals from myriad directing stimuli in the space surrounding it, choosing the right route for lengthening towards its final target.

Over the past decade, significant progress has been made in understanding the mechanism of directional movement of axons. See Dickson (2002) Science 298: 1959-64 Review. Directing stimuli exist in four varieties: attractants and repellents, which can act either at a short distance (that is, associated with cells or a matrix) or at a longer distance (that is, capable of diffusion). So far, four major families of axon-guiding molecules have been identified: netrins, semaphorins, esfrins, and proteins called slit proteins. See review of Huber et al (2003) Annu Rev Neurosci 26: 509-63.

Semaphorins (Sema), also called collapsins, belong to a large family of phylogenetically conserved, secreted and membrane-bound proteins. Members of the semaphorin family are capable of mediating both repulsive and attracting impulses in axon guidance during the development of the nervous system. Raper (2000) Curr Opin Neurobiol 10: 88-94. To date, more than thirty semaphorins have been identified, all of which have a common conserved N-terminal Sema domain of approximately 500 amino acids. Members of the semaphorin family are classified into eight subfamilies depending on their structural similarity and biological (species) origin. For more information on the unified nomenclature of semaphores, see Semaphorin Nomenclature Committee (1999) Cell 97: 551-552.

The neuropilin family (NRP) consists of two homologous proteins, neuropilin-1 (NRP1) and neuropilin-2 (NRP2). NRP1 was first identified as a Type I transmembrane glycoprotein with a molecular weight of 130 kDa expressed on growth cones of growing axons. Later, based on expression cloning, NRP2 was identified. Fujisawa and Kitsukawa (1998) Curr Opin Neurobiol 8: 587-592. NRPs have been found to be receptors in a subgroup of semaphorins (class 3 semaphorins). It has been suggested that NRPs function as non-signal coreceptors, along with another family of semaphorin receptors, plexins.

Although NRPs were originally described as axon direction mediators, they have also been found to play a critical role in vascular development. Carmeliet and Tessier-Lavigne (2005). They are identified as isoform-specific VEGF receptors expressed on tumor and endothelial cells, which indicates the need to direct significant efforts to clarify the role of NRP in vascular and tumor biology. Soker et al (1998) Cell 92: 735-745; Klagsbrun et al (2002) Adv Exp Med Biol 515: 33-48. Genetic studies have provided strong evidence that Nrp1 is required for vascular morphogenesis. The loss of Nrp1 function leads to vascular remodeling and branching defects, a phenotype that is further enhanced by the loss of Nrp2 function. Kawasaki et al. (1999) Development 126: 4895- 4902; Takashima et al. (2002) Proc Natl Acad Sci USA 99: 3657-3662. These results suggest that in the early stages of development, Nrp1 and Nrp2 may have overlapping functions. However, at later stages of development, the expression of each Nrp is separated, with Nrp1 expressed mainly in the arteries, and Nrp2 in the veins and lymphatic vessels. Yuan et al (2002) Development 129: 4797-4806; Herzog et al. (2001) Mech Dev 109: 115-119. It is noteworthy that the isolated loss of Nrp2 function in a specific way disrupts the development of the lymphatic system.

Since Nrp1 is expressed in many other cell types during development, the role of vascular Nrp1 was studied by generating an EC-specific gene knockout that led to the development of similar vascular defects with a zero allele. Gu et al. (2003) Dev Cell 5: 45-57. Interestingly, this study also showed that binding of Sema3A to NRP1 is not necessary for the development of the vascular system. In another study, defects in the direction of the upper part of endothelial cells were observed during the development of the hindbrain in Nrp1 KO embryos. Gerhardt et al. (2004) Dev Dyn 231: 503-509.

Despite extensive research on the role of NRP1 in the development of the vascular system, it remains unclear whether NRP1 realizes its vascular function exclusively through the metabolic pathway of the VEGF-VEGF 2 receptor (VEGFR2), acting as an enhancer of VEGF binding to VEGFR2 and, therefore, as a signaling factor for VEGFR2, either through a signaling pathway independent of VEGFR2, or through a combination of both of these metabolic pathways.

Monoclonal antibodies can be obtained using recombinant DNA technology. Monoclonal antibodies, especially those derived from rodents, have found widespread use, but antibodies of a non-human origin often turn out to be antigenic to humans. Attempts have been made in the art to overcome this problem associated with the construction of “chimeric” antibodies in which an antigen-binding domain of non-human origin is coupled to a human constant domain (Cabilly et al., US Patent No. 4816567). The human constant domain isotype can be selected so as to give the chimeric antibody the ability to participate in antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity. In further attempts to resolve the antigen-binding functions of antibodies and to minimize the use of heterologous sequences in human antibodies, humanized antibodies were generated for various antigens, in which a much smaller part of the sequence than the intact variable domain of a person was replaced in some regions by the corresponding sequence from biological species, different from human. For example, residues from rodent sequences were substituted into the corresponding segments of human antibodies. In a practical aspect, humanized antibodies are typical human antibodies in which some residues of the hypervariable region (CDR) and, possibly, some residues of the framework region (FR) are replaced by residues from similar sites of rodent antibodies. Jones et al. (1986) Nature 321: 522-525; Riechmann et al. (1988) Nature 332: 323-327; Verhoeyen et al. (1988) Science 239: 1534-1536.

Before a therapeutic antibody will be administered to humans, it is desirable to conduct preclinical studies on mammals other than humans in order to evaluate the effectiveness and / or toxicity of the antibody. In the ideal case, the antibodies that are the subject of such studies are able to recognize the target antigen endogenous to the animal host, such as a mouse or non-human primate, and interact with it with high efficiency.

Phage display technology has provided a powerful tool for generating and selecting new proteins capable of binding to a ligand such as an antigen. Using the phage display technique, it is possible to create large libraries of protein variants and quickly sort them to select sequences that are capable of binding to the target antigen with high affinity. Nucleic acids encoding variant polypeptides are fused to sequences of nucleic acids encoding an enveloped protein of a virus, for example, gene III protein or gene VIII protein. Monovalent phage display systems have been developed in which a nucleic acid sequence encoding a protein or polypeptide is fused to a nucleic acid sequence encoding a portion of a gene III protein. (Bass, S. (1990) Proteins 8: 309; Lowman and Wells (1991) Methods: A Companion to Methods in Enzymology 3: 205). In a monovalent phage display system, gene fusion is expressed at a low level, and wild-type III gene proteins are also expressed in such a way that the invasive ability of the particles is maintained. Methods for creating peptide libraries and screening these libraries have been disclosed in many patents (for example, US Patent No. 5,723,286, US Patent No. 5,432,018, US Patent No. 5,580,717, US Patent No. 5,427,908 and US Patent No. 5,498,530).

The demonstration of peptide expression on the surface of the filamentous phage and the expression of functional antibody fragments in the periplasm of E. Coli was important for the development of phage antibody display libraries. (Smith et al. (1985) Science 228: 1315; Skerra and Pluckthun (1988) Science 240: 1038). Libraries of antibodies or antigen binding polypeptides have been prepared in a variety of ways, including altering a single gene by inserting random DNA sequences or by cloning a family of related genes. Methods for demonstrating antibodies or antigen binding fragments using phage display have been described in US Pat. Nos. 5,750,373, 5,733,743, 5,837,242, 5,969,108, 617,2177, 5580717 and 5658727. The library is then screened for expression of antibodies or antigen binding proteins having the desired properties.

Phage display technology has several advantages over traditional hybridoma and recombinant methods in obtaining antibodies with the desired characteristics. This technology allows you to create large libraries of antibodies with different sequences in less time and without the use of animals. For preparatory work in order to obtain hybridomas or humanized antibodies, it can undoubtedly take several months. In addition, since immunization is not required, phage libraries of antibodies can be created for those antigens that are toxic or have low antigenicity (Hogenboom (1988) Immunotechniques 4: 1-20). Antibody phage libraries can also be used to create and identify new therapeutic antibodies.

Phage display libraries were used to create human antibodies from immunized and non-immunized human germline sequences or from an Ig repertoire not exposed to any B-cell exposure (Barbas & Burton (1996) Trends Biotech 14: 230; Griffiths et al. (1994) EMBO J. 13: 3245; Vaughan et al. (1996) Nat. Biotech. 14: 309; Winter EP 0368 684 B1). Unexposed antigen-binding libraries were created using a variety of lymphoid tissues. Some of these libraries are commercially available, such as those developed by Cambridge Antibody Technology and Morphosys (Vaughan et al. (1996) Nature Biotech 14: 309; Knappik et al. (1999) J. Mol. Biol. 296: 57). However, many of these libraries have a limited variety.

The ability to identify and isolate high affinity antibodies from the phage display library is important for the production of new antibodies for their therapeutic use. The ability to isolate high affinity antibodies from a library depends on the size of the library, the efficiency of production in bacterial cells, and the variety of the library. See, for example, Knappik et al. (1999) J. Mol. Biol. 296: 57. The size of the library decreases with production inefficiencies due to improper coagulation of the antibody or antigen binding protein or due to the presence of a stop codon. Expression in bacterial cells can be suppressed if the antibody or antigen binding domain is not folded correctly. Expression can be improved by mutating residues at the boundary of the variable and constant domains or selected CDR residues. (Deng et al. (1994) J. Biol. Chem. 269: 9533, Ulrich et al. (1995) PNAS, 92: 11907-11911; Forsberg et al. (1997) J Biol. Chem. 272: 12430). If phage libraries of antibodies are produced in bacterial cells, the sequence of the framework region is an important factor in ensuring proper coagulation.

To isolate high affinity antibodies, it is also important to create a variety of libraries of antibodies or antigen binding proteins. Diversity libraries with limited CDRs have been created using a variety of approaches. See, for example, Tomlinson (2000) Nature Biotech. 18: 989-994. The CDR3 regions are partly of interest because their involvement in antigen binding is often found. The heavy chain CDR3 regions vary significantly in size, sequence and structural conformation.

Other specialists also generated diversity through randomization of the CDRs of the variable regions of the heavy and light chains using all 20 amino acids at each position. It was believed that the use of all 20 amino acids could lead to a wide variety of variant antibody sequences and to an increased chance of identifying new antibodies. (Barbas (1994) PNAS 91: 3809; Yelton, DE (1995) J. Immunology 155: 1994; Jackson, JR (1995) J. Immunology 154: 3310 and Hawkins, RE (1992) J. Mol. Biology 226: 889 )

SUMMARY OF THE INVENTION

The present invention relates to the crystal structures of fragments of neuropilin-1 and neuropilin-2 (Nrp1 and Nrp2) as single fragments or in complex in antibodies that selectively block the binding of either semaphorin or VEGF. Nrp perceive an unexpected arrangement of domains, in which domains a2, b1 and b2 form a tightly packed core. Variants of the arrangement of antibody epitopes, along with in vitro experiments, show that VEGF and semaphorins do not directly compete for binding to Nrp. Based on the crystallographic Nrp dimer mediated by the a1 domain, the present invention further provides models for receptor dimerization and ligand binding.

Based on these results, in one aspect, the present invention provides a crystal formed by a Nrp1 _b1b2 fragment that diffracts X-rays to produce a diffraction pattern that reflects the three-dimensional structure of said fragment having approximately the following unit cell parameters: a = 65.9 Å, b = 66.7 Å, c = 74.7 Å and space group P2 ₁ 2 ₁ 2 ₁ .

In a further aspect, the invention relates to a crystal formed by an Nrp1 _a2b1b2 fragment that diffracts X-ray radiation to produce a diffraction pattern that reflects the three-dimensional structure of said fragment having approximately the following unit cell parameters: a = 53.2 Å, b = 68.2 Å , c = 66.6 Å and the space group P2 ₁ .

In yet another aspect, the invention relates to a crystal of a complex formed by an Nrp1 _b1 fragment and a Fab fragment of an anti-Nrp1 ^B antibody (YW107.4.87) that inhibits the binding of vascular endothelial growth factor (VEGF) to Nrp1, wherein said crystal diffracts X-rays to produce such a diffraction pattern that reflects the three-dimensional structure of the indicated complex, having approximately the following unit cell parameters: a = 213 Å, b = 213 Å, c = 45.3 Å and space group H3.

In a further aspect, the invention relates to a crystal formed by an Nrp2 _b1b2 fragment that diffracts X-rays to produce a diffraction pattern that reflects a three-dimensional structure of said fragment having approximately the following unit cell parameters: a = 36.5 Å, b = 70.5 Å , c = 122 Å and the space group P2 ₁ 2 ₁ 2 ₁ .

In yet another aspect, the invention relates to a crystal formed by a Nrp2 _a2b1b2 fragment that diffracts X-rays to produce a diffraction pattern that reflects a three-dimensional structure of said fragment having approximately the following unit cell parameters: a = 50.1 Å, b = 193 Å, c = 66.2 Å and space group P2 ₁ .

In yet another aspect, the invention relates to a crystal of a complex formed by an Nrp2 fragment _a1a2b1b2 and a Fab fragment of an anti panNrp ^A antibody that inhibits the binding of semaphorin to Nrp2, said crystal diffracting X-ray radiation to produce a diffraction pattern that reflects the three-dimensional structure of said complex having approximately the following unit cell parameters: a = 148 Å, b = 106 Å, c = 92.4 Å, and space group C2.

The invention further relates to a crystal of a complex formed by an Nrp2 fragment _a1a2b1b2 and a Fab fragment of an anti-panNrp ^A antibody that suppresses the binding of semaphorin to Nrp2, said crystal diffracting X-ray radiation to produce a diffraction pattern that reflects a three-dimensional structure of said complex having approximately the following parameters unit cell: a = 121 Å, b = 121 Å, c = 203 Å and space group P3 ₂ 21.

In yet another aspect, the invention relates to an anti-panNrp ^A antibody comprising the light chain variable domain sequence shown in FIG. 7 and / or the heavy chain variable domain sequence shown in FIG. 8 or a fragment thereof.

In yet another aspect, the invention relates to a Fab fragment of an anti panNrp ^A antibody YW68.11, comprising the sequence shown in FIG. 9A.

In a further aspect, the invention relates to a Fab fragment of an anti panNrp ^A antibody YW68.11.26 comprising the sequence shown in FIG. 9B.

In yet a further aspect, the invention relates to an anti-Nrp antibody that competes for binding of Nrp to an anti-panNrp ^A antibody.

In one embodiment, the anti-Nrp antibody binds essentially to the same epitope as the anti-panNrp ^A antibody.

In another embodiment, the anti-Nrp antibody binds to an epitope containing at least a portion of the interaction surface defined by amino acid residues Y39, Y45, P46, Q47, F72, N73 P74, H75, F76, A133 and R138 of the amino acid sequence Nrp2 _a1a2b1b2 .

In yet another embodiment, the anti-Nrp antibody binds to both Nrp1 and Nrp2.

In further embodiments, the anti-Nrp antibody has a binding affinity of at least about 0.10 nM with both Nrp1 and Nrp2, or at least about 0.15 nM with both Nrp1 and Nrp2, or at least about 0.20 nm with both Nrp1 and Nrp2 or at least about 0.25 nm with both Nrp1 and Nrp2 or at least about 0.30 nm with Nrp1, so with Nrp2.

In yet another embodiment, the anti-Nrp antibody blocks the binding of Sema3 to both Nrp1 and Nrp2.

In yet another embodiment, the anti-Nrp antibody does not block the binding of VEGF to Nrp1 or Nrp2.

In a particular embodiment, the anti-Nrp antibody inhibits the biological activity of semaphorin in vitro.

In a further embodiment of the invention, the anti-Nrp antibody inhibits the biological activity of semaphorin in vivo.

In another aspect, the invention relates to a method for producing a semaphorin antagonist, comprising constructing a molecule that binds to a site containing at least a portion of the interaction surface defined by amino acid residues Y39, Y45, P46, Q47, F72, N73 P74, H75, F76, A133 and R138 of the amino acid sequence of Nrp2 _a1a2b1b2 , synthesis of the compound and confirmation that the compound blocks the binding of semaphorine to Nrp1 and Nrp2.

In one embodiment, the antagonist does not interfere with the binding of VEGF to Nrp1 or Nrp2, and the method may include an additional step of confirming that the antagonist does not interfere with the binding of VEGF to Nrp1 or Nrp2.

In yet another embodiment of the invention, the method further includes the step of confirming that the antagonist does not interfere with the biological activity of VEGF.

The antagonist may, for example, be selected from the group consisting of antibodies, antibody fragments, binding polypeptides, peptides, and low molecular weight non-peptide molecules, and preferably is an antibody or antibody fragment, the antibody fragments including (without limitation) Fab, Fab ' , F (ab ') ₂ , scFv, (scFv) ₂ , dAb, linear antibodies, single-chain antibody molecules, minodies, dimeric antibodies (diabodies) and multispecific antibodies formed from antibody fragments.

In yet another aspect, the invention relates to a method for producing a VEGF antagonist, comprising the use of a three-dimensional structure obtained from a crystal of a complex formed by an Nrp1 _b1 fragment and a Fab fragment of an anti-Nrpl ^B antibody (YW107.4.87), in which the crystal diffracts X-rays to produce such a pattern diffraction, which displays the three-dimensional structure of the indicated complex, having approximately the following unit cell parameters: a = 213 Å, b = 213 Å, c = 45.3 Å and space group H3, to construct a molecule that binds A site containing at least a portion of the epitope associated with the indicated Nrpl ^B antibody synthesizes the compound and confirms that the compound inhibits the binding of VEGF to Nrp1.

In one embodiment, the antagonist does not interfere with the binding of semaphorin to Nrp1, and the method may further include the step of confirming this fact.

In another embodiment, the antagonist inhibits the biological activity of VEGF.

In yet another embodiment, the method further includes the step of confirming that the antagonist inhibits the biological activity of VEGF.

In a further embodiment, the antagonist inhibits vascular remodeling.

In yet a further embodiment, the antagonist is selected from the group consisting of antibodies, antibody fragments, binding polypeptides, peptides, and low molecular weight non-peptide molecules, and is preferably an antibody or antibody fragment, wherein the antibody fragment may be, for example, Fab, Fab ', F (ab') ₂ , scFv, (scFv) ₂ , dAb, a linear antibody, a single chain antibody molecule, an antibody, a dimer antibody, or a multispecific antibody, which are formed from antibody fragments.

The invention further relates to a method for treating cancer, comprising administering to a mammal in need thereof an effective amount of a VEGF antagonist obtained by the method described above, which is proposed by the present invention. The cancer may, for example, be selected from the group consisting of breast cancer, colorectal cancer, non-small cell lung cancer, non-Hodgkin lymphoma (NHL), kidney cancer, prostate cancer, liver cancer, head and neck cancer, melanoma, ovarian cancer, mesothelioma and multiple myeloma.

In another embodiment, the treatment further includes a second therapeutic agent, wherein the second therapeutic agent can be selected (without limitation) from the group consisting of an antiangiogenic agent, an antineoplastic composition, a chemotherapeutic agent and a cytotoxic agent, for example, such as an additional VEGF antagonist.

In a further embodiment, the additional VEGF antagonist is an anti-hVEGF antibody or fragment thereof.

An anti-hVEGF antibody may, for example, be able to bind to the same VEGF epitope as antibody A4.6.1, and more specifically bevacizumab or ranibizumab.

In other embodiments, the second therapeutic agent is a tyrosine kinase receptor inhibitor selected from the group consisting of vatalanib (PTK787), erlotinib (TARCEVA ^® ), OSI-7904, ZD6474 (ZACTIMA ^® ), ZD6126 (ANG453), ZDl839, sunin SUTENT ^® ), semaxanib (SU5416), AMG706, AG013736, imatinib (GLEEVEC ^® ), MLN-518, CEP-701, PKC-412, lapatinib (GSK572016), VELCADE ^® , AZD2171, sorafenib (NEXAVAR ^® ), XL880 and CHIR -265.

BRIEF DESCRIPTION OF THE DRAWINGS

Table 1 - Data collection and detail statistics

Figure 1 - VEGF does not block Sema3A-induced DRG neuron cone collapse

A) Images of axon growth cones. Untreated DRGs have large and actin-rich growth cones (arrowheads), which decrease significantly when Sema3A is added. Antibodies against Nrp were added at a dose of 50 μg / ml.

B) Quantification of the collapse of growth cones induced by Sema3A. The percentage of decayed growth cones was calculated by counting collapsed and not collapsed growth cones (N = 4 explants per state). Error bars represent the standard error of the mean.

Figure 2 - Summary of the structure of the neuropilin crystal.

A) The Nrp ectodomain consists of the tandem CUB (a1a2), the tandem of coagulation factor V / VIII (b1b2) and one MAM domain (c1). A simplified schematic representation of the seven crystal structures presented in this post with the resolution limits listed below. The orange spheres show the bound calcium ion in the a2 domain.

B) Alignment of the sequence of domains a1a2b1b2 in Nrp1 and Nrp2 person. Elements of the secondary structure belong to the structure of Nrp2-a1a2b1b2 (blue - a1, green - a2, yellow - b1, red - b2) and are named according to the arrangement adopted for the spermadhesin CUB domain (Romero, A. et al., Nat Struct Biol 4, 783-8 (1997)) and the C2 domain of coanulation factor V (Macedo-Ribeiro, S. et al., Nature 402, 434-9 (1999)). Residues bounded by blue and yellow frames depict antibody epitopes for anti-panNrp ^A and anti-Nrp1 ^B, respectively. Amino acids highlighted in orange indicate the Ca ²⁺ binding site in a2, and residues highlighted in red represent the putative Ca ²⁺ binding site in the a1 domain. The residues shaded in green highlight the positions of amino acid substitutions that break the interaction between Sema3A and Nrp1 (Gu, C. et al., J Biol Chem 277, 18069-76 (2002)). This alignment was performed using the EsPript computer program (Gouet, P. et al., Nuclei Acids Res 3l, 3320-3 (2003)).

Figure 3 - the general domain architecture of neuropilins.

A) Nrp2 domain organization (blue - a1, green - a2, yellow - b1, red - b2) in combination with the Fab fragment of anti-panNrp ^A antibody (light orange - heavy chain, gray - light chain). N-glycosylated residues are highlighted in magenta (magenta).

B) Tape mapping of structures a2b1b2 Nrp1 and Nrp2; orange spheres emit bound calcium ion.

C) Combination of the Nrp2 / Fab complex from two different crystalline forms based on a2b1b2 domains. Note the poor alignment of a1 domains compared to the a2b1b2 domain. All patterns of structures were obtained using the PyMol computer program (http://www.pymol.org).

Figure 4 - molecular details of the domains of the CUB neuropilin.

A) In a2b1b2 structures, the a2 domain includes bound calcium ion (orange). The Nrp domains a1 (see panel C) and a2 lose the thread b1 found in spermadgesin (Romero et al, supra).

B) The calcium ion in the a2 Nrp1 domain is coordinated by three negatively charged amino acids, two main chain carbonyl oxides and a water molecule. These interactions are highly conserved for Nrp (see FIGS. S2-S3).

C) (Left panel) Amino acids are colored according to the percentage of solvent accessible surface that is hidden at the Nrp2 / Fab interface (red - 75-100%, orange - 50-74%, yellow - 25-49%). The anti-panNrp ^A antibody is cross-reactive with Nrp1 and Nrp2, and eleven of the fourteen residues in the structural epitope are identical (black text is identical, white text is not conservative, asterisks indicate the residues in which the side chain is directed toward the protein core a1). The residues highlighted in green show the positions of amino acid substitutions that are necessary for the interactions between Nrp1 and Sema3A ²³ . (Right panel) The Ca atoms in these amino acid substitutions are shown as green spheres. Magenta Ca atoms show the putative calcium binding site.

D) Partition surface against panNrp ^A / Nrp2. Nrp2 is shown as a molecular surface in accordance with the electrostatic potential (red - acid, blue - alkaline). Antibody contact residues are driven by aromatic residues from CDR H2, H3 and L3.

Figure 5 - Signs of the Nrp domains binding to VEGF and heparin.

A) The combination of the crystal structures of Nrp b1b2 (Nrp1 is yellow (b1) and red (b2), Nrp2 is gray). Blue (Nrp1) and green (Nrp2) residues emphasize conformational differences in the “peaks” of b2, but not b1.

B) Molecular surfaces of rat (accession number PDB 2ORZ) (Vancer Kooi, C. W., et al., Proc Natl Acad Sci USA (2007)) and human b1b2 crystal structures are stained according to electrostatic potential. The yellow arrows indicate an acid groove, which is formed by "peaks" in the b1 domain and represents the tuftsin binding site in the rat structure (Vander Kooi et al., Supra). Green arrows indicate the approximate location of the heparin binding site.

C) The relative conservatism of the sequence (green — 100%, yellow — ≥75%) of the b1b2 domains among twelve Nrp (FIG. S3) was mapped onto the surface of the human Nrp1 b1b2 structure. Two highly conserved areas are highlighted in orange. Residues highlighted in blue (cyan) mean those residues that are in contact with Fab in the anti-Nrp1 ^B- Fab / b1 complex. Domain a2 (wheat) is shown using the combination of structures b1b2 and a2b1b2 from Nrp1.

D) Tape presentation of the anti-Nrp1 ^B- Fab / b1 complex (yellow - b1, orange - heavy chain, gray - light chain).

E) Anti-Nrp1 ^B / b1 interface surface. Domain b1 is simplified as a molecular surface according to electrostatic potential: a yellow arrow indicates a groove for binding the VEGF tail. Only CDR H3 and Ll are in contact with b1.

6 - Crystallographic dimer Nrp2 involves models of binding of VEGF and semaphorin

A) Nrp2 forms a saddle-shaped dimer in both crystalline forms of the Nrp2 / Fab complex. This figure isolates the Nrp2 a1a2b1b2 domains from the monosymmetric form of the Fab complex. The alleged binding sites of the tail of VEGF, heparin and semaphorin are indicated.

B) Potential models of the VEGF / Nrp and semaphorin / Nrp complexes based on the Nrpa1-mediated dimer in crystal structures.

7 - Alignment of the variable domain sequence of the light chain of the antibody against panNrp ^A  (clones YW68.11 and YW68.11.26).

Fig - Alignment of the sequence of the variable domain of the heavy chain of the antibodies against panNrp ^A  (clones YW68.11 and YW68.11.26).

Figa - The complete sequence of human IgG1 antibodies against panNrp ^A  YW68.11.

Figv - The complete sequence of human IgG1 antibodies against panNrp ^A  YW68.11.26.

Table S1 - Conditions used for crystallization and cryoprotection.

Figure S1 - Analysis of the kinetics of binding of antibodies against Nrp.

Kinetic analysis of antibodies against panNrp ^A by BIAcore. Injection sensing patterns of 500 nM of each human NRP protein at 25 ° C over the BIAcore immobilized IgG sensor chip demonstrate binding specificity: the anti-panNrp ^A antibody binds the a1a2b1b2 Nrp1 and Nrp2 domains, but not the b1b2 Nrp2 domains.

Fig. S2 - CUB Nrp domains include a conserved calcium binding site.

A) The electron density around the calcium ion in the structure of Nrp1 a2b1b2 (the final map 2F _O -F _C map is drawn at 1.5σ).

B) Calcium ion of the a2 Nrp2 domain.

C) The three negatively charged amino acids (shaded in orange) that outline the calcium binding site are conserved in the CUB domains of Nrp1 and Nrp2.

Fig S3 - Alignment of the sequences of neuropilins.

Alignment of the full-sized sequence of the a1a2b1b2 domains from Nrp1 and Nrp2 of humans, mice, rats, zebrafish (BRARE), frogs (XENLA) and chicken. This figure is colored in the same way as FIG. 2B, and was obtained using the EsPript computer program (Gouet P. et al., Above).

Fig. S4 - Comparison of the structures of Nrp1 / tuftsin and b1 / Fab

A) In the Nrp1-b1 / anti-Nrp1 ^B- Fab complex, the C-terminal tail (purple) of a heavy chain of a symmetrically linked molecule occupies the tuftsin binding site (Nrp2, the molecular surface according to the electrostatic potential: orange - antibody heavy chain, gray - light antibody chain).

B) Localization of tuftsin (green) in combination with rat Nrp1 b1b2 (accession number PDB 2ORZ) (Vander Kooi, et al., Above).

C) The combination of the two structures highlighted in panel A. The antibody and tuftsin peptide are colored in the same way as in the previous panels: human Nrp1 b1 is colored yellow, and rat Nrp1 b1b2 is colored blue (cyan).

Fig. S5 - Section Surface of Nrp2 Dimer

A) The tape representation of the Nrp2-a1a2b1b2 / Fab dimer, as it appears in the monosymmetric form of the structure.

B) The superposition of the crystal structure of Nrp1-b1 / against Nrpl ^B- Fab on the crystal structure of Nrp2 / Fab.

C) Amino acids hidden at the a1 / a1 interface are colored according to the percentage of solvent accessible surface that is hidden by dimerization (red - 75-100%, orange - 50-74%, yellow - 25-49%) .

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to new crystalline structures, compositions, methods for modulating NRP-mediated biological activity, and screening assays to identify candidates capable of modulating NRP-mediated biological activity.

Definitions

"Neuropilin" or NRP is a collective term referring to neuropilin-1 (NRP1), neuropilin-2 (NRP2), as well as their isoforms and variants, as described in Rossignol et al. (2000) Genomics 70: 211-222. Neuropilins are nontyrosine kinase receptors with a molecular weight of 120 to 130 kDa. There are many options for splicing and soluble isoforms of NRP-1 and NRP-2. The basic structure of neuropilins contains five domains: three extracellular domains (a1a2, b1b2 and c), a transmembrane domain, and a cytoplasmic domain. The a1a2 domain is homologous to the complement components C1r and C1s (CUB), which usually contains four cysteine residues forming two disulfide bridges. The b1b2 domain is homologous to coagulation factors V and VIII. The central part of the domain is designated as MAM due to homology with meprin, A5, and tyrosine phosphatase receptor μ-proteins. Domains a1a2 and b1b2 are responsible for ligand binding, while domain c is critical for homodimerization or heterodimerization. Gu et al. (2002) J. Biol. Chem. 277: 18069-76; He and Tessier-Lavigne (1997) Cell 90: 739-51.

The terminological definition of "neuropilin-mediated biological activity" refers generally to physiological or pathological phenomena in which neuropilin-1 and / or neuropilin-2 play an essential role. Non-limiting examples of such activity are the direction of axons in the embryonic period of the development of the nervous system or during the regeneration of neurons, angiogenesis (including vascular modeling), tumorigenesis and tumor metastasis.

The term “antibody” is used in its broadest sense and specifically encompasses monoclonal antibodies (including full-size monoclonal antibodies), polyclonal antibodies, multispecific antibodies (eg, bispecific antibodies), and antibody fragments, provided that they exhibit the desired biological activity.

The term "monoclonal antibody" as used herein refers to an antibody derived from a combination of substantially homogeneous antibodies, i.e. individual antibodies contained in the aggregate are identical with the exception of possible spontaneous mutations, which can be represented in a minimal amount. Monoclonal antibodies are highly specific, being directed against a single antigenic site. In addition, in contrast to traditional (polyclonal) antibody preparations, which typically include different antibodies directed against different determinants (epitopes), each monoclonal antibody is directed against one determinant or antigen. The "monoclonal (s)" modifier indicates the nature of the antibody, which is obtained from a substantially uniform set of antibodies and should not be constructed, that is, not require development in any special way. For example, monoclonal antibodies used in accordance with the present invention can be obtained by the hybridoma method, first described in reference Kohler et al. (1975) Nature 256: 495, or can be obtained using recombinant DNA methods (see, for example, US patent No. 4816567). "Monoclonal antibodies" can also be isolated from phage libraries of antibodies, for example, using the techniques described in references Clackson et al. (1991) Nature 352: 624-628 and Marks et al. (1991) J Mol. Biol. 222: 581-597.

In this specification, monoclonal antibodies specifically include “chimeric” antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical or homologous to the corresponding sequences of antibodies obtained from a particular species or belonging to a particular class or subclass of antibodies, while the rest of the chain (s) is identical or homologous to the corresponding sequences of antibodies obtained from another species or belonging to another class or subclass of antibodies, and fragments of such antibodies, provided that they exhibit the desired biological activity (US patent No. 4816567 and reference Morrison et al. (1984) Proc. Natl. Acad. Sci. USA 81: 6851-6855).

“Humanized” forms of non-human (eg, murine) antibodies are chimeric antibodies that contain a minimal sequence extracted from non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (recipient antibodies) in which residues from the reciprocal region of the recipient are replaced by residues from a hypervariable region of a non-human species (donor antibodies), for example, a mouse, rat, rabbit or non-human primate, provided that they possess the desired specificity, affinity and performance. In some cases, residues of the framework region (FR) of the human immunoglobulin Fv are replaced by corresponding non-human residues. In addition, humanized antibodies may contain residues that are not found in the recipient antibody or in the donor antibody. Such modifications are carried out in order to further increase the performance of the antibody. Typically, a humanized antibody contains essentially all or at least one, and typically two variable domains in which all or substantially all of the hypervariable loops correspond to those in non-human immunoglobulins, and all or essentially all FR regions are regions of a human immunoglobulin sequence. A humanized antibody may also optionally contain at least a portion of an immunoglobulin constant region (Fc), typically derived from a human immunoglobulin. For more information, see Jones et al. (1986) Nature 321: 522-525; Riechmann et al. (1988) Nature 332: 323-329; and Presta (1992) Curr. Op. Struct. Biol. 2: 593-596.

A "species-dependent antibody" is one that has a stronger binding affinity for an antigen of a first mammalian species than the homolog of this antigen from a second mammalian species. Normally, the species-dependent antibody "specifically binds" to the human antigen (i.e., has a binding affinity (K _d ) of not more than about 1 × 10 ^-7 M, preferably not more than about 1 × 10 ^-8 M, most preferably not more than approximately 1 × 10 ^-9 M), but has an affinity for binding to an antigen homolog from a second, non-human, mammalian species that is at least 50 times or at least 500 times or, at least 1000 times weaker than its binding affinity for antigen th person. The species-specific antibody may be any of the various types of antibodies defined above, but preferably it may be a humanized antibody or a human antibody.

As used herein, the term “antibody mutant” or “antibody variant” refers to a variant amino acid sequence of a species-dependent antibody in which one or more amino acid residues of the species-dependent antibody have been modified. Such mutants necessarily have less than 100% sequence identity or similarity with a species-dependent antibody. In a preferred embodiment, the antibody mutant has an amino acid sequence that is identical to the amino acid sequence of the variable domain or heavy or light chain of a species-dependent antibody (or similar to it) by at least 75%, more preferably at least 80%, more preferably at least 85%, more preferably at least 90%, and most preferably at least 95%. Identity or similarity with respect to this sequence is defined here as the percentage of amino acid residues in a candidate sequence that are identical (i.e., exactly match) or similar (i.e. are amino acid residues from the same group based on the general properties of the side chain, see below) with the remains of a species-specific antibody after alignment of the sequences and, if necessary, the introduction of “gaps” to achieve the maximum percentage sequence identity. None of the N-terminal, C-terminal or internal extensions, deletions or insertions in the sequence of antibodies outside the variable domain should not be construed as affecting the identity or similarity of the sequence.

An “isolated” antibody is one that has been identified and separated from and / or recovered from components of its natural environment. Contaminating components from the natural environment of the antibody are materials that interfere with its diagnostic or therapeutic use: they may include enzymes, hormones, and other protein or non-protein dissolved substances. In preferred embodiments, the antibody will be purified (1) to more than 95% by weight of the antibody, as determined by the Lowry method, or, most preferably, to more than 99% by weight, (2) to the extent that it is sufficient to produce, at least 15 residues of the N-terminal or internal amino acid sequence, using a rotary beaker, or (3) homogeneity by the SDS-PAGE method under reducing or non-reducing conditions using Coomassie blue staining or, preferably, silver m An isolated antibody includes the antibody in situ within recombinant cells, since in this case at least one component of the natural environment of the antibody will not be present. However, usually an isolated antibody will be prepared using at least one purification step.

As used herein, the terminological definition of an "antibody variable domain" refers to parts of the light and heavy chains of antibody molecules that include amino acid sequences of the hypervariable regions (CDR, i.e., CDR1, CDR2, and CDR3) and framework regions (FR). The abbreviation V _H refers to the variable domain of the heavy chain. The abbreviation V _L refers to the variable domain of the light chain. In accordance with the methods used in this invention, amino acid positions attributed to CDR and FR can be determined by Kabat (Sequences of Proteins of Immunological Interest (National Institutes of Health, Bethesda, Md., 1987 and 1991). The numbering of amino acids in antibodies or antigen binding fragments is also carried out in accordance with the Kabat classification.

As used herein, the term “hypervariable regions” (CDRs, i.e., CDR1, CDR2, and CDR3) refers to the amino acid residues of an antibody variable domain, the presence of which is necessary for antigen binding. Each typical variable domain has three CDR regions identified as CDR1, CDR2, and CDR3. Each hypervariable region may contain amino acid residues from the "complementarity determining region" according to Kabat (i.e., approximately residues 24-34 (L1), 50-56 (L2) and 89-97 (L3) in the variable domain of the light chain and 31-35 ( H1), 50-65 (H2) and 95-102 (H3) in the variable domain of the heavy chain. Kabat et al, Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD. ( 1991)) and / or residues from the "hypervariable loop" (that is, approximately residues 26-32 (L1), 50-52 (L2) and 91-96 (L3) in the variable domain of the light chain and 26-32 (H1), 53-55 (H2) and 96-101 (H3) in the variable domain of the heavy chain. Chothia and Le sk (1987) J. Mol. Biol. 196: 901-917). In some cases, the hypervariable region may include amino acids from both the CDR region as defined by Kabat and the hypervariable loop. For example, the 4D5 antibody heavy chain CDRH1 includes amino acids 26 through 35.

"Frame regions" (hereinafter referred to as FR) are represented by other variable domain residues compared to CDR residues. Each typical variable domain has four FRs identified as FR1, FR2, FR3, and FR4. When determining Kabat CDRs, light chain FR residues are approximately positioned as residues 1-23 (LCFR1), 35-49 (LCFR2), 57-88 (LCFR3) and 98-107 (LCFR4) in the light chain, and heavy chain FR residues are approximately positioned as residues 1-30 (HCFR1), 36-49 (HCFR2), 66-94 (HCFR3) and 103-113 (HCFR4) in the heavy chain. If the CDRs contain amino acid residues from hypervariable loops, the light chain FR residues are approximately positioned as residues 1-25 (LCFR1), 33-49 (LCFR2), 53-90 (LCFR3) and 97-107 (LCFR4) in the light chain, and residues of the heavy chain FR are approximately positioned as residues 1-25 (HCFR1), 33-52 (HCFR2), 56-95 (HCFR3) and 102-113 (HCFR4) in the heavy chain. In some cases, when the CDR contains amino acids from both the CDR as defined by Kabat and the hypervariable loop, the FR residues will be adjusted accordingly. For example, when CDRH1 includes amino acids H26-H35, the heavy chain residues FR1 are at positions 1-25, and the FR2 residues are at positions 36-49.

As used herein, the terminological definition of “codon set” refers to a set of different nucleotide triplet sequences encoding desired amino acid variants. The oligonucleotide set can be synthesized, for example, by solid phase synthesis, including those sequences that represent all kinds of combinations of nucleotide triplets contained in the codon set and will encode the desired group of amino acids. The standard codon designation is the IUB (International Biochemical Union) code, which is known in the art and will be described here. A typical codon set is represented by 3 uppercase letters in italics, for example, NNK, NNS, XYZ, DVK , etc. Thus, as used herein, the term “non-random codon set” refers to a codon set that encodes selected amino acids that partially or, preferably, completely satisfy amino acid selection criteria, as described herein. The synthesis of oligonucleotides with selected “degeneracy” at certain positions is well known in the art, for example, the TRIM approach (Knappek et al. (1999) J. Mol. Biol. 296: 57-86); Garrard & Henner (1993) Gene 128: 103). Such oligonucleotide kits containing specific codon kits can be synthesized using industrial nucleic acid synthesizers (e.g., from Applied Biosystems, Foster City, California), or pre-made (e.g., from Life Technologies, Rockville, Maryland) . Therefore, the synthesized oligonucleotide set containing a particular codon set will typically include a plurality of oligonucleotides with different sequences, the differences being set by the codon set within the general sequence. When used in accordance with the present invention, oligonucleotides have sequences that allow them to hybridize with the variable domain nucleic acid template, and may, although not necessarily, include restriction enzyme sites useful, for example, for cloning purposes.

The Fv fragment is an antibody fragment that contains a complete antigen recognition and binding site. This region consists of a dimer of the variable domains of one heavy and one light chain, which are in close association, which by its nature can be covalent, for example, in scFv. For this configuration, it is typical that the three CDRs of each variable domain interact to determine the antigen binding site on the surface of the V _H -V _L dimer. All together, six CDRs or a subset thereof give the antibody specificity for antigen binding. However, even a single variable domain (or half of an Fv containing only three CDRs specific for an antigen) has the ability to recognize and bind antigen, although, as a rule, with less affinity than the full binding site.

The Fab fragment contains the variable and constant domain of the light chain, as well as the variable domain and the first constant domain (CH1) of the heavy chain. Antibody fragments F (ab ') ₂ contain a pair of Fab fragments, which, as a rule, are covalently linked to each other in the region of their carboxyl ends by hinge cysteines. Other chemical bonds between antibody fragments are also known in the art.

Fragments of the antibody "single-chain Fv" or "scFv" contain the antibody domains V _H and V _L , and these domains are presented in the same polypeptide chain. Typically, the Fv polypeptide further comprises between the V _H and V _L domains a polypeptide linker that allows scFv to form the desired structure for antigen binding. For an overview of scFv, see Pluckthun in The Pharmacology of Monoclonal Antibodies, Vol 113, Rosenburg and Moore eds. Springer-Verlag, New York, pp. 269-315 (1994).

The term "dimeric antibodies" refers to small fragments of antibodies with two antigen-binding sites, and these fragments contain the variable domain of the heavy chain (V _H ) associated with the variable domain of the light chain (V _L ) in the same polypeptide chain (V _H and V _L ). By using a linker that is too short to allow pairing between two domains on the same chain, the domains are forced to pair with the complementary domains of the other chain, which creates two antigen-binding sites. Dimeric antibodies are described in more detail, for example, in references EP 404097, WO 93/11161 and Hollinger et al. (1993) Proc. Natl. Acad. Sci. USA 90: 6444-6448.

The expression "linear antibodies" refers to antibodies described in the link Zapata et al. (1995 Protein Eng, 8 (10): 1057-1062). Briefly, these antibodies comprise a pair of tandem segments (V _H —C _H 1 -V _H —C _H 1), which, together with complementary light chain polypeptides, form a pair of antigen binding regions. Linear antibodies may be bispecific or monospecific.

As used herein, the term “library” refers to a plurality of antibody sequences or antibody fragments (eg, polypeptides of the invention) or nucleic acids that encode these sequences, said sequences differing in the combination of variant amino acids introduced into these sequences in accordance with methods proposed invention.

A “phage display” is a technique by which variant polypeptides are demonstrated as fusion proteins fused to at least a portion of a membrane protein on the surface of a phage, such as filamentous phage, phage particles. The potential use of phage display is based on the fact that large libraries of randomized protein variants can be quickly and efficiently sorted into sequences that bind to the target antigen at a high level of affinity. Demonstration of peptide and protein libraries on the phage was used to screen millions of polypeptides for specific binding properties. Multivalent phage display methods have been used to demonstrate small random peptides and small proteins through their fusion with either gene III or gene VIII of the filamentous phage. Wells and Lowman (1992) Curr. Opin. Struct. Biol. 3: 355-362 and references cited there. In a monovalent phage display, a protein or peptide library hybridized with gene III or part thereof is expressed at a low level in the presence of a wild-type protein encoded by gene III, so that phage particles show one copy of the hybridized proteins or do not show a single copy. The avidity effects are limited compared to the polyvalent phage so that sorting based on the natural affinity for the ligand and phagemid vectors, which simplify DNA manipulation, can be used. Lowman and Wells (1991) Methods: A companion to Methods in Enzymology 3: 205-0216.

A “phagemid” is a plasmid vector having a bacterial origin of replication, for example ColE1, and a copy of the intergenic region of the bacteriophage. The phagemid can be used on any known bacteriophage, including a filamentous bacteriophage and a lambdoid bacteriophage. The plasmid also typically contains a selectable antibiotic resistance marker. DNA segments cloned in these vectors can be propagated as plasmids. When cells containing these vectors are equipped with all the genes necessary for the production of phage particles, the plasmid replication method changes to “rolling ring” replication, which creates copies of one strand of plasmid DNA followed by packaging of phage particles. The phagemid may form infectious or non-infectious phage particles. This term includes phagemids that contain the phage coat protein gene or its fragment linked to the gene of a heterologous peptide by the type of gene fusion, due to which the heterologous polypeptide is displayed on the surface of the phage particle.

The term "phage vector" means a double-stranded replicative form of a bacteriophage containing a heterologous gene and capable of replication. The phage vector has phage origin of replication, which allows the phage to replicate and form phage particles. A preferred phage is a filamentous bacteriophage, for example, phage M13, fl, fd, Pf3, or derivatives thereof, or a lambdoid phage, for example, lambda, 21, phi80, phi81, 82, 424, 434, etc., or derivatives thereof.

As used herein, the terminology definition “solvent accessible position” refers to the position of an amino acid residue in the variable regions of the heavy and light chains of the parent antibody or antigen binding fragment to be determined, which is based on the structure, ensemble of structures and / or the simulated structure of the antibody or antigen binding fragment and is potentially available for solvent and / or for contracting with a molecule, for example, an antibody-specific antigen. Typically, such positions are in the CDR and on the outer surface of the protein. The positions of the antibody or antigen binding fragment available to the solvent, as indicated herein, can be determined using any of a variety of algorithms known in the art. Preferably, the positions accessible to the solvent are determined using coordinates from a 3-dimensional model of the antibody, as well as using a computer program, for example, InsightII (Accelrys, San Diego, California). Solvent accessible positions can also be determined using algorithms known in the art (e.g., Lee and Richards (1971) J. Mol. Biol. 55, 379 and Connolly (1983) J. Appl. Cryst. 16, 548). Determination of positions available for the solvent can be carried out using software suitable for modeling proteins and 3D structural information obtained for the antibody. Software that can be used for this purpose includes the SYBYL Biopolymer Module (Tripos Associates) program. Usually and preferably, if the algorithm (program) requires the user to input size parameters, then the "size" of the probe used in the calculations is set approximately at a radius of 1.4 angstroms or less. In addition to this, methods for determining solvent accessible sites and areas using personal computer software have been described in Pacios (1994) Comput. Chem. 18 (4): 377-386.

An “angiogenic factor or agent” is a growth factor that stimulates the development of blood vessels, for example, promotes angiogenesis, growth of endothelial cells, stability of blood vessels and / or vasculogenesis, etc. For example, angiogenic factors include, but are not limited to, VEGF and members of the VEGF family, PIGF family, PDGF family, fibroblast growth factor family (FGF), TIE ligands (angiopoietins), ethers, Del-1, fibroblast growth factors: acidic (aFGF ) and alkaline (bFGF), follistatin, colony stimulating granulocyte factor (G-CSF), hepatocyte growth factor (HGF) / scattering factor (SF), interleukin-8 (IL-8), leptin, midkin, neuropilins, placental growth factor, platelet endothelial cell growth factor (PD-ECGF), platelet growth factor, especially PDGF-BB or beta-PDGFR, pleiotrophin (PTN), progranulin, proliferin transforming alpha growth factor (TGF-alpha), transforming beta growth factor (TGF-beta), tumor necrosis alpha-factor (TNF-alpha), etc. They also include factors that accelerate wound healing, such as growth hormone, insulin-like growth factor I (IGF-I), VIGF, epidermal growth factor (EGF), CTGF and its family members, as well as alpha-TGF and beta-TGF. See, for example, Klagsbrun and D'Amore (1991) Annu. Rev. Physiol. 53: 217-39; Streit and Detmar (2003) Oncogene 22: 3172-3179; Ferrara & Alitalo (1999) Nature Medicine 5 (12): 1359-1364; Tonini et al. (2003) Oncogene 22: 6549-6556 (e.g., Table 1, which lists known angiogenic factors), and Sato (2003) Int. J. Clin. Oncol. 8: 200-206.

The term “anti-angiogenesis agent” or “angiogenesis inhibitor” refers to low molecular weight substances (polynucleotide, polypeptide, isolated protein, recombinant protein, antibody), as well as their conjugates or fusion proteins that inhibit angiogenesis, vasculogenesis or undesired direct vascular permeability or indirectly. It should be understood that the term "anti-angiogenesis agent" encompasses those agents that bind and block the angiogenic activity of an angiogenic factor or its receptor. For example, an anti-angiogenesis agent is an antibody or other antagonist of an angiogenic agent as defined above, for example, antibodies to VEGF-A or to a VEGF-A receptor (e.g., KDR receptor or Flt-I receptor), anti-PDGFR inhibitors, e.g. Gleevec ™ (imatinib mesylate). Anti-angiogenesis agents also include natural angiogenesis inhibitors, such as angiostatin, endostatin, etc. See, for example, Klagsbrun and D'Amore (1991) Annu. Rev. Physiol. 53: 217-39, Streit and Detmar (2003) Oncogene 22: 3172-3179 (e.g., Table 3, which lists methods of antiangiogenic therapy for malignant melanoma), Ferrara & Alitalo (1999) Nature Medicine 5 (12): 1359- 1364, Tonini et al. (2003) Oncogene 22: 6549-6556 (e.g., Table 2, which lists known anti-angiogenic factors), as well as Sato (2003) Int. J. Clin. Oncol. 8: 200-206 (for example, Table 1, which lists anti-angiogenesis agents evaluated in clinical trials).

The term “VEGF” or “VEGF-A” as used herein refers to a human vascular endothelial growth factor cell composed of 165 amino acids, as well as related human vascular endothelial growth factor cells consisting of 121, 189 and 206 amino acids, as described in references Leung et al. (1989) Science 246: 1306 and Houck et al. (1991) Mol. Endocrin, 5: 1806, along with their naturally occurring allelic and processed forms. The term "VEGF" also refers to VEGF variants from other species, such as a mouse, rat or primate. Sometimes VEGF from a particular species is denoted by terms such as hVEGF for human VEGF, mVEGF for mouse VEGF, etc. The term “VEGF” is also used to refer to truncated forms of a polypeptide containing amino acids 8 through 109 or from 1 through 109 of human vascular endothelial growth factor consisting of 165 amino acids. Reference to any such forms of VEGF in the present application can be identified, e.g., abbreviations of "VEGF (8-109)", " VEGF (1-109)" or "VEGF _165." The amino acid positions in the "truncated" natural VEGF are numbered as indicated in the natural VEGF sequence. For example, the position of the 17th amino acid (methionine) in truncated natural VEGF is also the position 17 (methionine) in natural (full-sized) VEGF. Truncated natural VEGF has binding affinity for the KDR and Flt-1 receptors comparable to natural (full-sized) VEGF.

An “anti-VEGF antibody” is an antibody that binds to VEGF with a sufficiently high affinity and specificity. In a preferred embodiment, the anti-VEGF antibody of the invention is used as a therapeutic agent targeted to interfere with diseases or conditions in which VEGF activity is involved. An anti-VEGF antibody generally does not bind to other VEGF homologues, such as VEGF-B or VEGF-C, or to other growth factors, such as P1GF, PDGF or bFGF. A preferred anti-VEGF antibody is a monoclonal antibody that binds to the same epitope as an anti-VEGF A4.6.1 monoclonal antibody produced by ATCC HB 10709. The more preferred anti-VEGF antibody is a recombinant humanized anti-VEGF monoclonal antibody generated in accordance with reference Presta et al. (1997) Cancer Res. 57: 4593-4599, including, but not limited to, an antibody known as bevacizumab (BV; Avastin ™).

The anti-VEGF antibody "bevacizumab (BV)", also known as the "rhuMAb VEGF" or "Avastin ^® , is a recombinant humanized monoclonal anti-VEGF antibody generated in accordance with Presta et al. (1997) Cancer Res. 57: 4593- 4599. It contains mutant human IgG1 framework regions and antigen-binding hypervariable regions from a mouse anti-hVEGF A4.6.1 monoclonal antibody that blocks the binding of human VEGF to its receptors. Approximately 93% of the bevacizumab amino acid sequence, including most of the framework regions obtained from human IgG1, and approximately 7% of this sequence comes from mouse antibody A4.6.1 Bevacizumab has a molecular weight of approximately 149,000 Da and is glycosylated.

The term "VEGF antagonist" refers to a molecule capable of neutralizing, blocking, inhibiting, nullifying, reducing the manifestations of VEGF activities or interfering with manifestations of VEGF activity, including, but not limited to, binding of VEGF to one or more receptors. VEGF antagonists include, but are not limited to, anti-VEGF antibodies and antigen binding fragments thereof, receptor molecules and derivatives that specifically bind to VEGF, thereby blocking its binding to one or more receptors, anti-VEGF receptor antibodies and VEGF receptor antagonists, such as small molecules VEGFR tyrosine kinase inhibitors. The term “VEGF antagonist” as used herein specifically includes molecules, including antibodies, antibody fragments, other binding polypeptides, peptides, and low molecular weight non-peptide molecules that bind to neuropilin-1 and / or neuropilin-2 (Nrp-1 and / or Nrp -2) and are capable of neutralizing, blocking, inhibiting, nullifying, decreasing manifestations of VEGF activity or inhibiting manifestations of VEGF activity, including, but not limited to, anti-Nrp1 and anti-Nrp2 antibodies and antibodies cross-reacting with Nrp1 and Nrp2, while Wii that they are able to neutralize, block, inhibit, abrogate, reduce manifestation of VEGF activity or inhibit VEGF activity manifestations. Thus, the term “manifestations of VEGF activity” specifically includes neuropilin-mediated manifestations of the biological activity of VEGF (as defined here above).

The term "semaphorin antagonists" refers to a molecule capable of neutralizing, blocking, inhibiting, nullifying, reducing the manifestations of semaphorin activity or inhibiting manifestations of semaphorin activity, including, but not limited to, the binding of semaphorine to one or more of its receptors. Semaphorin antagonists include, but are not limited to, anti-semaphorin antibodies and antigen binding fragments thereof, receptor molecules and derivatives that specifically bind to VEGF, thereby blocking its binding to one or more receptors, anti-semaphorin receptor antibodies and semaphorin receptor antagonists, such as small molecules -inhibitors of semaphorins. The term “semaphorin antagonist” as used herein specifically includes molecules, including antibodies, antibody fragments, other binding polypeptides, peptides and low molecular weight non-peptide molecules that bind to neuropilin-1 and / or neuropilin-2 (Nrp-1 and / or Nrp-2 ) and are capable of neutralizing, blocking, inhibiting, nullifying, decreasing manifestations of semaphorin activity or inhibiting manifestations of semaphorin activity, including, but not limited to, antibodies against Nrp1 and against Nrp2 and antibodies, cross-reacting e with Nrp1 and Nrp2, with the proviso that they are able to neutralize, block, inhibit, abrogate, reduce manifestation of activity semaphorin or semaphorin activity prevent manifestations. Thus, the term “manifestations of semaphorin activity” specifically includes neuropilin-mediated manifestations of the biological activity of class 3 semaphorins (as defined here above). Such manifestations of biological activity include, for example, an inhibitory effect on the growth of neurites (axons) in the embryonic period of the development of the nervous system, as well as during the regeneration of neurons.

The term "treatment" refers to both therapeutic effects and prophylactic or preventative measures. Subjects in need of treatment include those who are already affected by the disease, and those who need to prevent the development of the disease.

“Disruption” is any condition that can be improved as a result of treatment. For example, abnormalities are observed in mammals that suffer from abnormalities of angiogenesis (excessive, inadequate or uncontrolled angiogenesis) or abnormalities of vascular permeability, or require the prevention of such abnormalities. This group includes chronic and acute disorders or diseases, including those pathological conditions that predispose the mammal to the disorders / diseases in question. Non-limiting examples of disorders to be treated in accordance with this invention include malignant and benign tumors, non-leukemic and lymphoid malignancies, neuronal, glial, astrocytic, hypothalamic glandular, macrophage, epithelial, stromal and blastocele related disorders, as well as inflammatory, angiogenic and immunological disorders.

Abnormal angiogenesis occurs when new blood vessels grow excessively, insufficiently or inadequately (for example, localization, the onset of angiogenesis is undesirable from a medical point of view), which occurs either in a disease state or in a state of predisposition to the disease. Excessive, inadequate, or uncontrolled angiogenesis occurs when new blood vessels grow, contributing to worsening of a disease state or leading to a disease state such as cancer, especially vascular solid tumors and metastatic tumors (including colon cancer, lung cancer (especially small cell lung cancer) or prostate cancer), diseases caused by ocular neovascularization, especially diabetic blindness, retinopathy, primary diabetic retin patia or age-related macular degeneration (AMD), psoriasis, psoriatic arthritis, hemangioblastoma, such as hemangioma, inflammatory kidney diseases, such as glomerulonephritis, especially mesangioproliferative glomerulonephritis, hemolytic uremic syndrome, diabetic nephropathy or hypertensive diseases such as arthritis, nephrosis, especially rheumatoid arthritis, inflammatory bowel disease, psoriasis, sarcoidosis, arteriosclerosis and post-transplant diseases AI, endometriosis, chronic asthma, and more than 70 other pathological conditions. New blood vessels can nourish painful (altered by pathology) tissues, destroy normal tissues, and in the case of cancer, new blood vessels enable tumor cells to evacuate to the circulatory system and enter other organs (tumor metastases). Insufficient angiogenesis occurs when inadequately reduced blood vessel growth occurs, contributing to worsening of the disease state, for example, in pathological conditions such as coronary artery disease, stroke, and slow wound healing. In addition, ulcers, strokes and heart attacks can develop as a result of the lack of angiogenesis, which is necessary in the norm for the natural healing of lesions. The present invention contemplates the treatment of such patients who are at increased risk of developing the aforementioned diseases / pathological conditions.

Other patients who are candidates for antibodies or other molecules proposed by this invention are already sick or have an increased risk of developing diseases / disorders such as abnormal proliferation of vascular fibrous tissue, redheads, acquired immunodeficiency syndrome, artery blockage, atopic keratitis, bacterial ulcers, Behcet's disease, hematogenous tumors, obstructive carotid artery disease, choroidal neovascularization, chronic inflammation, chronic retinal detachment, chron Critical uveitis, chronic vitritis, eye strain when wearing contact lenses, corneal transplant rejection, corneal neovascularization, corneal neovascularization, Crohn’s disease, Ilza’s disease, epidemic keratoconjunctivitis, fungal ulcers, herpes simplex virus virus infection, typhoid fever, typhoid infection blood), Kaposi's sarcoma, leukemia, fatty degeneration, Lyme disease, marginal keratolysis, Murena ulcer, mycobacterial infections other than leprosy, myopia, neovascular disease eye disease, fossa of the optic disc, Osler-Weber syndrome (Osder-Weber-Rendu), osteoarthritis, Paget's disease, pars planitis (chronic cyclitis or peripheral uveitis), pemphigoid, conlict conjunctivitis, polyarteritis, complications from laser procedures, proto elastic pseudoxanthoma, pterygium with dry keratitis, radial keratotomy, neovascularization of the retina, retinopathy of premature infants, retrolental fibrodysplasias, sarcoid, scleritis, sickle cell anemia, Sjögren's (Sjögren's) syndrome, solid tumors whether, Stargardt’s disease, Steven-Johnson’s disease, upper limbic keratitis, syphilis, systemic lupus erythematosus, marginal degeneration of Terjen, toxoplasmosis, trauma, tumors in Ewing's sarcoma, tumors in neuroblastoma, tumors in osteosarcoma, tumors in retinoblastoma, , ulcerative colitis, venous obstruction, vitamin A deficiency and Wegener's sarcoidosis (granulomatosis), unwanted angiogenesis in diabetes, parasitic diseases, abnormal wound healing, hypertrophy after surgery, damaged e or trauma, inhibition of hair growth, inhibition of ovulation and corpus luteum formation, inhibition of implantation and inhibition of embryo development in the uterus.

Anti-angiogenesis therapies are useful in the general treatment of transplant rejection, pneumonia, nephrotic syndrome, preeclampsia, pericardial effusion, for example, pericardial effusion, and pleural effusion, diseases and disorders characterized by undesirable vascular permeability, such as edema associated with brain tumors, ascites in malignant diseases, Meigs syndrome, pneumonia, nephrotic syndrome, pericardial effusion, pleural effusion, permeability, associated Noah with cardiovascular diseases, for example, with a condition after myocardial infarction, strokes, etc.

Other angiogenesis-dependent diseases in accordance with the present invention include angiofibroma (abnormal accumulation of blood from vessels prone to bleeding), neovascular glaucoma (growth of blood vessels in the eye), arteriovenous malformations (abnormal communication between arteries and veins), non-healing fractures (fractures that do not heal), atherosclerotic plaques (hardening of the arteries), pyogenic granuloma (frequent damage to the skin, consisting of blood vessels), scleroderma (a form of joint disease tissue), hemangioma (a tumor made up of blood vessels), trachoma (the main cause of blindness in third world countries), hemophilic arthropathy, vascular adhesions and hypertrophic scars (abnormal scar formation).

The terms “cancer” and “cancerous” refer to a physiological state or describe a physiological state in mammals, which is typically characterized by uncontrolled cell growth. Examples of cancer include, but are not limited to, carcinoma, lymphoma, blastoma, sarcoma, and leukemia. More specific examples of such cancers include squamous cell carcinoma, lung cancer (including small cell lung cancer, non-small cell lung cancer, lung adenocarcinoma and squamous cell carcinoma), peritoneal cancer, hepatocellular cancer, gastric cancer or gastric cancer (including gastrointestinal cancer), pancreatic cancer glands, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladder cancer, hepatoma, breast cancer, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary carcinoma, kidney cancer, liver cancer, prostate cancer, vulvar cancer, thyroid cancer, liver carcinoma and various types of head and neck cancer, as well as B-cell lymphoma (including low-grade / follicular non-Hodgkin's lymphoma (NHL), small-cell lymphocytic (SL) NHL, follicular / medium malignancy NHL, diffuse NHL medium malignancy, high malignant lymphoblastic NHL, high malignant small cell NHL with undigested cell nuclei, NHL with massive lesion, mantle cell lymphoma, AIDS-related lymphoma, and Waldenstrom acroglobulinemia), chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), nappy cell leukemia, chronic myeloid leukemia and post-transplant lymphoproliferative disease (PTLD), as well as abnormal vascular proliferation associated with tumor-related tumors associated with fakomat ) and Meigs syndrome.

The term "anti-neoplastic composition" refers to a composition useful for treating cancer and containing at least one active therapeutic agent, for example, an "anti-cancer agent". Examples of therapeutic agents (anti-cancer agents) include, but are not limited to, chemotherapeutic agents, growth inhibitors (cells), cytotoxic agents, agents used in radiation therapy, anti-angiogenesis agents, apoptotic agents, anti-tubulin agents, and other cancer treatments, such as anti-HER-2 antibodies, anti-CD20 antibodies, epidermal growth factor receptor (EGFR) antagonist (e.g., tyrosine kinase inhibitor), HER1 / EGFR inhibitor (e.g., erlotinib (Tarceva ™), platelet inhibitors growth factor (e.g., Gleevec ™ (imatinib mesylate)), a COX-2 inhibitor (e.g., celecoxib), interferons, cytokines, antagonists (e.g., neutralizing antibodies) that bind to one or more of the following targets: ErbB2, ErbB3, ErbB4 , PDGFR-beta, BlyS, APRIL, BCMA or VEGF receptor (s), TRAIL / Apo2 and other biologically active and organic chemicals, etc. Combinations of them are also included in the invention.

The term "cytotoxic agent" as used herein refers to a substance that inhibits or prevents the functioning of cells and / or causes destruction of cells. For its intended purpose, the term covers radioactive isotopes (for example, I ¹³¹ , I ¹²⁵ , Y ⁹⁰ and Re ¹⁸⁶ ), chemotherapeutic agents and toxins, in particular enzymatically active toxins of bacterial, fungal, plant or animal origin or their fragments.

A “chemotherapeutic agent” is a chemical compound useful in the treatment of cancer. Examples of chemotherapeutic agents include chemical compounds useful in treating cancer. Examples of chemotherapeutic agents include alkylating agents such as thiotepa and CYTOXAN ^® cyclosphosphamide such alkylsulfonates as busulfan, improsulfan and pirosulfan such aziridines as benzodopa, carboquone, meturedopa, and uredopa, ethylenimines and methylamelamines including altretamine, triethylenemelamine, trietylenephosphoramide, triethylenethiophosphoramide and trimethylolomelamine, acetogenins (especially bullatacin and bullatacinon), camptothecin (including the synthetic analogue of topotecan), bryostatin, callistatin, CC-1065 (including its synthetic analogues adozelezin, carzelesin and biselezin), cryptophycins (especially cryptophycin 1 and cryptophycin 8), dolastatin, duocarmycin (including synthetic analogs, KW-2189 and CB1-TM1), eleutherobin, pankratistatin, sarcodictiin, such as chloristrin, azothistrin, chlorothiamine, chlorophosphamide, estramustine, ifosfamide, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, novembikhine, fenesterin, prednimustine, trophosphamide, uracil mustard gas, such nitrosoureas as carmustine, chlorozotocin, fotemustinstinum, luminom, tics such as enediin antibiotics, e.g. calicheamicin, especially calicheamicin gamma II and calicheamicin omega II (see, for example, Agnew (1994) Chem Intl. Ed. Engl. 33: 183-186), dynemycin, including dynemycin A, bisphosphonates such as clodronate, esperamycin, as well as neocarcinostatin chromophore and related chromoprotein enediin antibiotic chromophores), aclacinomycin, actinomycin, autramycin, azaserin, carlein, blemin, blemin, chromomycin, dactinomycin, daunorubicin, detorubicin, 6-diazo-5-oxo-L-norleucine, ADRIAMYCIN ^® doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin eoxoxide, deoxorubicin eoxox) m, idarubicin, marcellomycin, mitomycin such as mitomycin C, mycophenolic acid, nogalamycin, olivomycin, peplomycin, potfiromycin, puromycin, quelamycin, rhodorubicin, streptonigrin, streptozocin, tubercidin, ubenimex, zinotorubin, 5-FU), folic acid analogs such as denopterin, methotrexate, pteropterin, trimerexate, purine analogues such as fludarabine, 6-mercaptopurine, thiamiprine, thioguanine, pyrimidine analogs such as ancitabine, azacitidine, 6-azauridine, carm cytarabine, dideoxyuridine, doxyfluuridine, enocytabine, phloxuridine, androgens such as calusterone, dromostanolone propionate, epithiostanol, mepitiostan, testolactone, antiadrenal agents such as aminoglutetol amide acid, such as aminoglutinolide acid, such as eniluracil, amsacrine, bestrabucil, bisanthrene, edatraxate, defofamine, demecolcin, diaziquon, elfornithine, elliptinium acetate, epothilone, etoglucid, gallium nitrate, hygroxyurea, flax Inan, lonidainin such maytansinoids both maytansine and ansamitocins, mitoguazone, mitoxantrone, mopidanmol, nitraerin, pentostatin, losoxantrone, pirarubicin, losoxantrone, podophyllinic acid 2-ethylhydrazide, procarbazine, complex polysaccharides PSK ^® (JHS Natural Products, Eugene, Oregon) , razoxane, rhizoxin, sisofiran, spiro germanium, tenuazonic acid, triaziquon, 2,2 ', 2-trichlorotriethylamine, trichothecenes (especially T-2 toxin, verracurin A, roridine A and anguidine), urethane, vindesine, dacarbazine, mitobron, mannustol, mannustol , mitolactol, pipobromane, gacitosin, ar abinoside ("Ara-C"), cyclophosphamide, thiotepa, taxoids, for example TAXOL ^® paclitaxel (Bristol-Myers Squibb Oncology, Princeton, NJ), cremophore-free ABRAXANE ™, albumin-engineering design of paclitaxel nanoparticles (American Pharmaceutical Partners , Schomberg, Illinois) and TAXOTERE ^® doxetaxel (Rhone-Poulenc Rorer, Anthony, France), chloranbucil, GEMZAR ^® gemcitabine, 6-thioguanine, mercaptopurine, methotrexate, platinum analogues such as cisplatin and carboplatin, vinblastine, platinum, platin, -16), ifosfamide, mitoxantrone, vincristine, NAVELBINE ^® vinorelbine, novantrone, teniposide, edatre Sat, daunomycin, aminopterin, xeloda, ibandronate, irinotecan, (campomstar, CPT-11) (including treatment regimen with irinotecan with 5-FU and leucovorin), topoisomerase inhibitor RFS 2000, difluoromethylornithine (DMFO), such retinoids as retinoic acid, combretastatin, leucovorin (LV), oxaliplatin, including the oxaliplatin treatment regimen (FOLFOX), PKC-alpha, Raf, H-Ras, EGFR inhibitors (e.g., erlotinib (Tarceva ™)) and VEGF-A, which reduce cell proliferation, as well as pharmaceutically acceptable salts, acids or derivatives of any of the above substances.

This definition also includes anti-hormonal agents that act to regulate or suppress the effects of hormones on tumors, in particular antiestrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen (including the drug tamoxifen NOLVADEX ^® ), raloxifene, droloxifene, 4 -hydroxytamoxifen, trioxifene, keoxifen, LY117018, onapriston and FARESTON-toremifene, aromatase inhibitors that inhibit the aromatase enzyme that regulates the production of estrogen in the adrenal glands, for example, 4 (5) -imidazoles, aminoglutethimide, MEGASE ^® megestrol acetate, AROMASIN ^® exemestane, formestane, fadrozole, RIVISOR ^® Vorozol, FEMARA ^® letrozole and ARIMIDEX ^® anastrozole, as well as antiandrogens such as flutamide, nilutamide, bicalutamide, leuprolide and goserelin, as well as troxacitoxidine (an analogue of troxacytoxidine (nucleoside diacidoxide) cytosine), antisense oligonucleotides, especially those that suppress gene expression in signaling pathways involved in impaired cell proliferation, for example PKC alpha, Raf and H-Ras, such ribozymes as a VEGF expression inhibitor (e.g. ANGIOZYME ^® ribozyme) and expression inhibitor HER2, such a vac ins as vaccines for gene therapy, vaccine e.g. ALLOVECTIN ^®, LEUVECTIN ^® vaccine, and VAXID ^® vaccine, PROLEUKIN ^® rIL-2 inhibitor, a topoisomerase 1 LURTOTECAN ^®, ABARELIX ^® rmRH, vinorelbine and esperamicins (see. US patent No. 4675187), as well as pharmaceutically acceptable salts, acids or derivatives of any of the above substances.

The term "prodrug" as used herein refers to a precursor or derivative form of a pharmaceutically active substance that is less cytotoxic to tumor cells compared to the parent drug and is capable of enzymatic activation or conversion to a more active parent dosage form. See, for example, Wilman (1986) "Prodrugs in Cancer Chemotherapy" Biochemical Society Transactions, 14, pp. 375-382, 615th Meeting Belfast and Stella et al. (1985). "Prodrugs: A Chemical Approach to Targeted Drug Delivery," Directed Drug Delivery, Borchardt et al, (ed.), Pp. 247-267, Humana Press. The products of this invention include, but are not limited to, phosphate-containing prodrugs, thiophosphate-containing prodrugs, sulfate-containing prodrugs, peptide-containing prodrugs, D-amino acid modified prodrugs, glycosylated prodrugs, β-lactam-containing prodrugs, optionally substituted phenoxyacetaminoacetamide -fluorocytosine and other 5-fluoruridine prodrugs that can be converted Vanir into the more active cytotoxic free drug. Examples of cytotoxic drugs that can be derivatized into a prodrug form for use in this invention include, but are not limited to, those chemotherapeutic agents described above.

Diseases and conditions that can be treated with semaphorin antagonists include, without limitation, neurological diseases and diseases requiring neuroregeneration or ameliorating from neuroregeneration.

A “low molecular weight molecule” is defined herein as a molecule having an approximate molecular weight of less than 500 daltons.

An "isolated" nucleic acid molecule is a nucleic acid molecule that is identified and separated from at least one contaminating nucleic acid molecule with which it is usually associated in the natural source of the antibody nucleic acid. An isolated nucleic acid molecule differs from the form or from the environment with which it can be found in nature. Therefore, the isolated nucleic acid molecules are different from the nucleic acid molecules that exist in natural cells. However, the isolated nucleic acid molecule includes a nucleic acid molecule contained in cells that typically expresses an antibody, for example, the nucleic acid molecule is in a different chromosomal position than the molecule found in natural cells.

The expression "regulatory sequences" refers to DNA sequences necessary for the expression of a functionally linked coding sequence in a particular host organism. Regulatory sequences that are suitable for prokaryotes include, for example, a promoter, optionally an operator sequence, as well as a ribosome binding site. Eukaryotic cells are known to use promoters, polyadenylation signals, and enhancers.

A nucleic acid is “operably linked” if it is in functional relationship with another nucleic acid sequence. For example, DNA for a subsequence or secretory leader is operably linked to DNA for a polypeptide if it is expressed as a precursor protein involved in the secretion of the polypeptide. A promoter or enhancer is functionally linked to a coding sequence if it affects the transcription of the sequence, and the ribosome binding site is functionally linked to a coding sequence if it is positioned so that it facilitates translation. Typically, a “functional link” means that the linked DNA sequences are adjacent, and in the case of a secretory leader, are adjacent and are in the reading phase. However, enhancers should not be contiguous. Adhesion is accomplished by crosslinking at suitable restriction sites. If there are no such sites, then, in accordance with generally accepted practice, synthetic oligonucleotide adapters or linkers are used.

As used herein, the terminology “cell,” “cell line,” and “cell culture” are used interchangeably, all of which mean progeny. Thus, the concepts of “transformants” and “transformed cells” include primary cell-objects and cultures obtained from them without taking into account the number of transfers. It should also be understood that all offspring may not be completely identical in the informational content of DNA due to intentional or unintentional mutations. Mutant progeny are included that have the same function or biological activity as those that were detected by screening for the initial transformed cells. In those places where clear designations are planned, this will be clear from the context.

MODES FOR CARRYING OUT THE INVENTION

Based on the architecture and functional properties of the domains, the extracellular Nrps domains can typically be divided into three functional units: a1a2, b1b2, and c, representing their primary regions of semaphorin binding, VEGF binding, and dimerization, respectively (Ellis, LM, Mol Cencer Ther 5, 1099-107 (2006)). Although structural and functional studies allow us to outline the general domain requirements for neuropilin function, the molecular details of these receptor / ligand complexes remain obscure. Currently, structural information on Nrp is limited to the b1b2 domain of Nrp1 (Vander Kooi et al., Supra; Lee, C. et al., J Biol Chem 281, 5702-10 (2006)). The crystal structure of the b1b2 domain of Nrp1 in complex with tuftsin, a tetrapeptide homolog of the C-terminus of VEGF ₁₆₅ , provided the first structural information on the interaction of Nrp with its VEGF binding partner (Vander Kooi, et al., Supra; von Wronski, MA, et al, J. Biol Chem 281, 5702-10 (2006)).

The present invention provides such a crystal structure of the a1a2b1b2 Nrp2 domains in complex with a Fab fragment of an antibody that is capable of blocking Sema3 binding to both Nrp1 and Nrp2 in vitro. The structures of fragments a2b1b2 and b1b2 of both Nrp isoforms are also compared and contrasted. In addition, the structure of the b1 domain of Nrp1, combined with the Fab fragment of the recently described antibody derived from phage, which specifically inhibits the binding of VEGF ₁₆₅ to Nrp1 in vivo (Liang, WC, et al., J Mol Biol 366, was also proposed and evaluated. 815-29 (2007); Pan, Q et al., Cancer Cell 11, 53-67 (2007). Together, these structures provide a detailed picture of a significant portion of the extracellular domain of Nrp and suggest models for binding of VEGF and semaphorin. Based on the Nrp2 dimer, presented in two different crystalline forms, a new mechanism of Nrp dimerization and binding ligand.

Details of crystallographic studies are presented in the examples below. General methods for producing antibodies against NRP are described here below and in Example 1.

Production of antibodies against NRP

The invention described herein includes the production and use of antibodies against NRP. Typical methods for generating antibodies are described in more detail in the following sections.

Antibodies against NRP were selected using the NRP antigen (e.g., NRP1 and / or NRP2) obtained from different species of mammals. The preferred antigen is human NRP (hNRP). However, NRPs from other species, for example murine NRP (mNRP), can be used as a target antigen. NRP antigens of various biological species of mammals can be isolated from natural sources. In other embodiments, the antigen is produced recombinantly or made by other synthetic methods known in the art.

The selected antibody usually has a sufficiently strong binding affinity for the NRP antigen. For example, an antibody can bind hNRP with a K _d value of not more than about 5 nM, preferably not more than about 2 nM, and more preferably not more than about 500 nM. The affinity of antibodies can be determined by analysis based on surface plasmon resonance, for example, such as the BIAcore analysis described in the examples, enzyme-linked immunosorbent assay (ELISA) and competition based assays (e.g., RIA).

The antibody can also be subjected to other tests for biological activity, for example, to evaluate its effectiveness as a therapeutic agent. Such assays are known in the art and depend on the target antigen, as well as on the purpose of the antibody. Examples include an HUVEC inhibitory assay (described in the examples below), tumor cell growth suppression assays (eg, described in WO 89/06692), antibody-dependent cell cytotoxicity (ADCC) assays, and complement-mediated cytotoxicity (CDC) (US Pat. No. 5,500,362) ), as well as tests for agonistic activity or hematopoiesis (see document WO 95/27062).

For screening for antibodies that bind to a particular epitope on an antigen of interest, a standard epitope cross-competitive analysis can be performed, such as described in Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory, Ed Harlow and David Lane (1988). Alternatively, epitope mapping can be performed, for example, as described in reference by Champe et al. (1995) J. Biol. Chem. 270: 1388-1394 to determine whether an antibody binds an epitope of interest.

Generation of anti-NRP antibodies from synthetic phage antibody libraries

In a preferred embodiment, anti-NRP antibodies are selected using a unique phage display approach. This approach involves generating synthetic phage antibody libraries based on a single framework region template, constructing sufficient diversity within the variable domains, selecting candidate antibodies with high affinity for the NRP target antigen, and isolating selected antibodies.

Methodological details of phage display methods can be found, for example, in document WO03 / 102157, published December 11, 2003.

In one aspect, antibody libraries can be generated by mutating solvent accessible and / or highly dissimilar positions in at least one CDR of an antibody variable domain. Some or all of the CDRs can be mutated using the methods described here. In some embodiments, it may be preferable to generate a variety of antibody libraries by mutating the positions in CDRH1, CDRH2 and CDRH3 to form a single library or by mutating the positions in CDRL3 and CDRH3 to form a single library, or by mutating the positions in CDRL3 and CDRH1, CDRH2 and CDRH3 to form a single library.

A library of variable domains of antibodies can be generated, for example, by mutating at CDRH1, CDRH2, and CDRH3 accessible to the solvent and / or highly dissimilar. Another library of this kind can be generated by mutations in CDRL1, CDRL2 and CDRL3. These libraries can also be used in combination with each other to generate binders with desired affinity. For example, after one or several heavy chain library selection cycles, binding to the target antigen can substitute the light chain library into the aggregate of heavy chain binding elements for subsequent selection cycles to increase the affinity of the connecting elements.

Preferably, the library is created by replacing the original amino acids with other (variant) amino acids in the CDRH3 region of the variable region of the heavy chain sequence. The resulting library may contain multiple antibody sequences, and the sequence diversity mainly affects the CDRH3 region in the heavy chain sequence.

In one aspect, a library is created in the context of the sequence of a humanized 4D5 antibody or the amino acid sequence of a framework region of a humanized 4D5 antibody. Preferably, to create the library, substitution of at least residues 95-100a in the amino acid sequence of the heavy chain encoded by the DVK codon set is used, the DVK codon set being used to encode variant amino acids at each of these positions. In an exemplary embodiment of an oligonucleotide kit useful for some embodiments of the invention, a library is created by substituting residues 95-100a for amino acids encoded by both DVK and NNK codon sets. An example of an oligonucleotide kit that is useful for making such substitutions is the sequence (DVK) ₆ (NNK). In yet another embodiment, a library is created by substituting at least 95-100a residues with amino acids encoded by both DVK and NNK codon sets. An example of an oligonucleotide kit that is useful for making such substitutions is the sequence (DVK) ₅ (NNK). Another example of an oligonucleotide kit that is useful for making such substitutions includes the sequence (NNK) ₆ . Other examples of suitable oligonucleotide sequences may well be determined by a person skilled in the art based on the criteria described herein.

In yet another embodiment, various CDRH3 constructs are used to isolate high affinity binding elements, as well as to isolate binding elements to a plurality of epitopes. The length range of the CDRH3 generated in this library is usually 11 to 13 amino acids, although constructs of other lengths can be generated. By using NNK, DVK, and NVK codon sets, H3 diversity can be increased, as well as more limited diversity at the N- and / or C-terminal regions.

You can also generate diversity in CDRH1 and CDRH2. Diversity constructs CDR-H1 and -H2 follow a strategy to target a mock natural antibody repertoire, as described, with a modification that focuses on diversity more closely matching natural diversity than the previous design.

For CDRH3 diversity, it is possible to individually construct multiple libraries with different H3 lengths, and then combine them to select binding elements with target antigens. Multiple libraries can be combined and sorted using methods of selection on a solid-phase substrate and sorting solutions, as described previously and will be described here below. You can use multiple sorting strategies. For example, one variation includes sorting on a target associated with a solid support, and subsequent sorting on a label that can be represented on a hybrid polypeptide (e.g., a label against gD), which is accompanied by another sorting on a target associated with a solid support. In an alternative embodiment, the libraries can first be sorted for the purpose associated with the solid support, then the eluted binders are sorted using liquid-phase binding to lower the concentration of the target antigen. The use of combinations of different sorting methods minimizes the selection, limited only to highly expressed sequences, and allows the selection of a significant number of different clones with high affinity.

High affinity binders targeting the NRP antigen can be isolated from libraries. Limited diversity in the H1 / H2 region reduces degeneracy by about 10 ⁴ to 10 ⁵ times, and a possible greater diversity of H3 provides for higher affinity binders. The use of libraries with different types of diversity in CDRH3 (for example, the use of DVK or NVT) involves the isolation of binding elements that can bind to different epitopes of the target antigen.

By connecting elements isolated from the combined libraries, as described above, it was found that affinity can be further improved due to the limited diversity of the light chain. In this embodiment, light chain diversity is generated as follows, in CDRL1: amino acid position 28 is encoded by RDT, amino acid position 29 is encoded by RKT, amino acid position 30 is encoded by RVW, amino acid position 31 is encoded by ANW, amino acid position 32 is encoded by THT, optionally amino acid position 33 is encoded CTG, in CDRL2: amino acid position 50 is encoded by KBG, amino acid position 53 is encoded by AVC, and optionally amino acid position 55 is encoded by GMA, in CDRL3: amino acid pol 91 ix encoded TMT or SRT or both, amino acid position 92 is encoded by DMC, amino acid position 93 is encoded by RVT, amino acid position 94 is encoded by NHT, and amino acid position 96 is encoded by TWT or YKG or both.

In yet another embodiment, a library or libraries with a variety of CDRH1, CDRH2, and CDRH3 are generated. In this embodiment of the invention, diversity in CDRH3 is generated using different lengths of H3 regions, as well as mainly codon sets XYZ and NNK or NNS. Libraries can be formed using individual oligonucleotides, followed by combination, or oligonucleotides can be combined to form a subset of the libraries. In this embodiment, the libraries can be sorted by the target associated with the solid support. Clones isolated from multiple sortings can be screened for specificity and affinity using ELISA assays. Clones can be screened for specificity for the desired target antigens, as well as for other non-target antigens. Then, these binding elements with the target NRP1 antigen can be screened for affinity in a competitive ELISA with binding in the liquid phase or in a competitive spot analysis. High affinity binders can be isolated from the library using XYZ codon sets made as described above. These binding elements can easily be obtained in high yield in cell culture as antibodies or antigen binding fragments.

In some embodiments, it may be desirable to generate libraries with a greater diversity of CDRH3 region in length. For example, it may be desirable to generate libraries with CDRH3 regions varying in length in an approximate range of 7 to 19 amino acids.

In these embodiments, high affinity binders isolated from libraries are readily produced in high yield in bacterial or eukaryotic cell culture. Vectors can be constructed to easily remove sequences such as gD tails, the sequence of the viral envelope protein component, and / or sequences are added to the constant region to ensure the production of full length antibodies or antigen binding fragments in high yield.

A library with mutations in CDRH3 can be combined with a library containing various variants of other CDRs, for example CDRL1, CDRL2, CDRL3, CDRH1 and / or CDRH2. For example, in one embodiment, the CDRH3 library is combined with a CDRL3 library created in the context of the sequence of a humanized 4D5 antibody with variant amino acids at positions 28, 29, 30, 31, and / or 32 using predetermined codon sets. In yet another embodiment, the CDRH3 mutation library can be combined with a library containing varieties of the heavy chain variable domains CDRH1 and / or CDRH2. In one embodiment, the CDRH1 library is created based on the sequence of the humanized 4D5 antibody with variant amino acids at positions 28, 30, 31, 32 and 33. The CDRH2 library can be created based on the sequence of the humanized 4D5 antibody with variant amino acids at positions 50, 52, 53 , 54, 56, and 58 using predetermined codon sets.

Anti-NRP Antibody Mutants

An anti-NRP antibody generated from phage libraries can be further modified by creating antibody mutants with improved physical, chemical and / or biological properties compared to the parent antibody. If the analysis used is a biological activity analysis, then the preferred antibody mutant has a biological activity in the selection analysis that is at least about 10 times greater than the original antibody in the same analysis, more preferably at least about 20 times even more preferably at least about 50 times, and sometimes at least about 100 times, or 200 times. For example, a preferred anti-NRP1 antibody mutant has an NRP binding affinity that exceeds the binding affinity of the parent anti-NRP antibody by at least about 10 times, more preferably at least about 20 times, more preferably at least about 50 times, and sometimes at least about 100 times or 200 times.

In order to generate an antibody mutant, one or more amino acid changes (e.g., substitutions) are introduced into one or more hypervariable regions of the original antibody. Alternatively, or in addition to this, one or more changes (for example, substitutions) of framework region residues can be introduced into the parent antibody if this leads to an improvement in the binding affinity of the mutant antibody to an antigen from a mammal of another species. Examples of frame region residues for modification include those that non-covalently bind antigen directly (Amit et al. (1986) Science 233: 747-753), interact with CDR / affect the conformation of CDR (Chothia et al. (1987) J. Mol. Biol. 196: 901-917) and / or participate in the VL-VH interface (EP 239 400B1). In some embodiments, the modification of one or more such framework region residues results in an increase in the binding affinity of the antibody to the antigen from a mammal of another species. For example, in this embodiment, about one to five residues of the frame region can be changed. Sometimes this may be enough to obtain a mutant antibody suitable for use in preclinical trials, even if not one of the residues of the hypervariable region has been changed. However, usually an antibody mutant will contain additional alteration (s) of the hypervariable region.

Residual hypervariable region residues can be changed randomly, especially when the initial binding affinity of the original antibody is such that randomly obtained antibody mutants of this kind can be easily screened.

One useful procedure for generating such antibody mutants is called "alanine scanning mutagenesis" (Cunningham and Wells (1989) Science 244: 1081-1085). In this case, one or more residues of the hypervariable region are replaced by the residue (s) of alanine or polyalanine in order to affect the interaction of amino acids with the antigen from a mammal of another biological species. Those hypervariable region residues that exhibit functional sensitivity to substitutions are later improved by introducing additional or other mutations at the substitution sites. Thus, although the site for introducing variations in the amino acid sequence is predetermined, the nature of the mutation, as such, does not need to be preliminarily determined. The alanine mutants (ala mutants) thus obtained are screened for their biological activity, as described herein.

You can usually start with a conservative substitution, for example, such as shown below under the heading of "preferred substitutions". If such substitutions result in a change in biological activity (for example, binding affinity), then more substantial changes are made, referred to as “typical substitutions” in the table below or hereinafter in reference to amino acid classes.

Preferred Substitutions:

Even more significant modifications in the biological properties of antibodies are achieved through selection of substitutions, which differ significantly in their effect on maintaining (a) the skeletal structure of the polypeptide in the substitution region, (b) the charge or hydrophobicity of the molecule at the target site, or (c) the mass of the side chain. Naturally occurring (amino acid) residues are divided into groups based on the general properties of the side chain:

(1) hydrophobic: norleucine (norleucine), met, ala, val, leu, ile;

(2) neutral hydrophilic: cys, ser, thr, asn, gln;

(3) acidic: asp, glu;

(4) basic: his, lys, arg;

(5) residues that affect chain orientation: gly, pro; and

(6) aromatic: trp, tyr, phe.

Non-conservative substitutions will entail the replacement of a member of one of these classes with another class.

In yet another embodiment, sites selected for modification mature by affinity using a phage display (see above).

Nucleic acid molecules encoding mutants of amino acid sequences are prepared by various methods known in the art. These methods include, but are not limited to, oligonucleotide-mediated (or site-specific) mutagenesis, PCR mutagenesis, and cassette mutagenesis of a previously obtained mutant or non-mutant variant of the parent antibody. A preferred method for producing mutants is site-specific mutagenesis (see, for example, Kunkel (1985) Proc. Natl. Acad. Sci. USA 82: 488).

In some embodiments, an antibody mutant will have only one residue substitution in the hypervariable region. In other embodiments, two or more residues of the hypervariable region of the parent antibody will be substituted, for example, approximately two to ten substitutions in the hypervariable region may occur.

Typically, an antibody mutant with improved biological properties will have an amino acid sequence that is identical to or identical to the amino acid sequence of the variable domain or heavy or light chain of the parent antibody, at least 75%, more preferably at least 80% more preferably at least 85%, more preferably at least 90%, and most preferably at least 95%. Identity or similarity with respect to this sequence is defined here as the percentage of amino acid residues in the sequence of candidates that are identical (i.e., exactly match) or similar (i.e. are amino acid residues from the same group based on the general properties of the side chain, see above) with the remains of the original antibody after alignment of the sequences and, if necessary, the introduction of gaps to achieve maximum percent sequence identity. None of the N-terminal, C-terminal or internal extensions, deletions or insertions in the sequence of antibodies outside the variable domain should not be construed as affecting the identity or similarity of the sequence.

After the production of an antibody mutant, the biological activity of this molecule relative to the parent antibody is determined. As noted above, this may include determining the binding affinity and / or other manifestations of the biological activity of the antibody. In a preferred embodiment, a panel of antibody mutants is prepared and the mutants are screened for binding affinity for an antigen such as NRP1 or a fragment thereof. One or more antibody mutants selected from this initial screening is optionally subjected to one or more additional biological activity tests to confirm that antibody mutants (mutants) with enhanced binding affinity are indeed useful, for example, for preclinical studies.

The mutant (s) of the antibody selected in this way can be subjected to further modifications, which often depends on the purpose of the antibody. Such modifications may include further changes in the amino acid sequence, fusion with a heterologous polypeptide (s) and / or covalent modifications, for example, those that will be discussed in detail below. As for changes in the amino acid sequence, typical modifications have been discussed above. For example, any cysteine residue that is not involved in maintaining the correct conformation of the antibody mutant can also be replaced, as a rule, with serine to improve the oxidative stability of the molecule and prevent aberrant crosslinking. Conversely, to improve the stability of the antibody, a cysteine bridge (bridges) can be added to it, especially if the antibody is represented by an antibody fragment, for example, an Fv fragment. Another type of amino acid mutant has an altered glycosylation pattern. This can be achieved by deletion of one or more carbohydrate components found in the antibody and / or the addition of one or more glycosylation sites that are not present in the antibody. Glycosylation of antibodies is usually either N-linked or O-linked. N-bonding refers to the attachment of a carbohydrate component to the side chain of an aspartic moiety. Tripeptide sequences such as asparagine-X-serine and asparagine-X-threonine, where X can be represented by any amino acid except proline, are recognition sequences for the enzymatic attachment of a carbohydrate component to the asparagine side chain. Thus, the presence of either of these two tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the addition of one of the sugars (N-acetylgalactosamine, galactose or xylose) to a hydroxy amino acid, most often serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used. Adding glycosylation sites to an antibody is easily accomplished by such a change in amino acid sequence that consists in introducing into it one or more of the tripeptide sequences described above (for N-linked glycosylation sites). The change can also be made by additional insertion or substitutional substitution in the sequence of the original antibody of one or more serine or threonine residues (for O-linked glycosylation sites).

Vectors, host cells and recombinant methods

Antibodies against Nrp proposed by the invention can be produced recombinantly using readily available techniques and materials.

For recombinant production, anti-NRP antibodies secrete the coding nucleic acid and insert the nucleic acid into the replicated vector for subsequent cloning (DNA amplification) or for expression. DNA encoding an antibody can be easily isolated or synthesized using conventional procedures (for example, using oligonucleotide probes capable of specifically binding to DNA molecules encoding the antibody heavy and light chains). Many vectors are available. Vector components typically include, but are not limited to, one or more of the following: a signal sequence, the start of replication (replicator), one or more marker genes, an enhancer element, a promoter, and a transcription termination sequence.

(i) Component of the signal sequence

The antibody of this invention can be produced recombinantly, not only directly, but also as a hybrid polypeptide with a heterologous polypeptide, which is preferably the signal sequence of another polypeptide having a specific cleavage site at the N-terminus of a mature protein or polypeptide. Preferably, the selected heterologous signal sequence is recognized and amenable to processing (i.e., signal peptidase cleavage) of the host cell. For prokaryotic cells that do not recognize and process the signal sequence of a natural antibody, the signal sequence is replaced by a selected prokaryotic signal sequence, for example, from the group of leaders of alkaline phosphatase, penicillinase, lpp or thermostable enterotoxin II. For yeast secretion, the natural signal sequence can be replaced, for example, by a yeast invertase leader, an α-factor leader (including α-factor leaders Saccharomyces and Kluyveromyces) or an acid phosphatase leader, glucose amylase leader C. albicans or the signal described in WO 90/13646 . For expression in mammalian cells, mammalian signal sequences are available, as well as viral secretory leaders, for example, the herpes simplex virus gD signal.

DNA for such a precursor region is ligated in reading frame with DNA encoding the antibody.

(ii) Replication Start Component

Both expression vectors and cloning vectors contain a nucleic acid sequence that allows the vector to replicate in one or more selected host cells. Typically, in the cloning vectors, this sequence is such that it allows the vector to replicate independently of the host chromosomal DNA and includes initial replication elements (replicators) or autonomously replicating sequences. Such sequences are well known for a variety of bacteria, yeast, and viruses. The plasmid replicator pBR322 is suitable for most gram-negative bacteria, the 2μ plasmid replicator is suitable for yeast, and various viral replicators (SV40, polyomas, adenovirus, VSV or BPV) are useful for cloning vectors in bacterial cells. Typically, the replicator component is not needed for mammalian expression vectors (the SV40 replicator can typically be used only because it contains an early promoter).

(iii) Component of the selection gene

The expression and cloning vectors may contain a selection gene, also called a selectable marker. Typical selection genes encode proteins that (a) confer resistance to antibiotics or other toxins, such as ampicillin, neomycin, methotrexate or tetracycline, (b) complement auxotrophic deficiency, or (c) deliver critical nutrients that are inaccessible from compound nutrient media, for example, a gene encoding D-alanine racemase for bacilli.

In one example of a selection scheme, a drug is used to inhibit the growth of a host cell. Those cells that are successfully transformed by a heterologous gene produce a protein that confers drug resistance, that is, the ability to survive in the selection scheme. Examples of this dominant selection use drugs such as neomycin, mycophenolic acid, and hygromycin.

Another example of suitable breeding markers for mammalian cells are genes that allow the identification of cells that are capable of accepting nucleic acid antibodies, for example genes DHFR, thymidine kinase, metallothionein-I and -II, preferably genes metallothionein primates, adenosine deaminase, ornithine decarboxylase, etc.

For example, cells transformed with the DHFR selection gene are primarily identified by culturing all transformants in a culture medium containing methotrexate (Mtx), a competitive DHFR antagonist. The corresponding host cell in which wild-type DHFR is involved is the Chinese Hamster Ovary (CHO) cell line with deficiency in DHFR activity.

Alternatively, host cells (especially wild-type hosts containing endogenous DHFR) that are transformed or cotransformed with DNA sequences encoding an antibody, wild-type DHFR protein and another selectable marker, for example, aminoglycoside 3'-phosphotransferase (APH), can be selected by growing in a culture medium containing an appropriate selection agent for a selectable marker, for example an aminoglycoside antibiotic (kanamycin, neomycin or G418). See US Patent No. 4965199.

A suitable selection gene for use in yeast is the trp1 gene present in the yeast plasmid YRp7 (Stinchcomb et al. (1979) Nature 282: 39). The trp1 gene provides a selectable marker for a mutant strain of yeast that loses its growth ability in the presence of tryptophan, for example, ATCC strain No. 44076 or PEP4-1. Jones (1977) Genetics 85:12. The presence of trp1 damage in the genome of the yeast host cell provides an effective environment for detecting transformation when grown in a culture medium lacking tryptophan. Similarly, Leu2 deficient yeast strains (ATCC 20622 or 38626) are complemented with known plasmids carrying the Leu2 gene.

In addition to this, vectors derived from a 1.6 μm pKD1 ring plasmid can be used to transform Kluyveromyces yeast. An alternative expression system for the large-scale production of recombinant calf chymosin has been described for K.lactis. Van den Berg (1990) Bio / Technology 8: 135. Multiple expression vectors have also been disclosed for secretion of mature recombinant human serum albumin by industrial Kluyveromyces strains. Fleer et al. (1991) Bio / Technology 9: 968-975.

(iv) Component promoter

The expression and cloning vectors typically contain a promoter that is recognized by the host organism and is operably linked to the nucleic acid of the antibody. Promoters suitable for use with prokaryotic hosts include the phoA promoter, β-lactamase and lactose, alkaline phosphatase promoter systems, tryptophan (trp) promoter system, and hybrid promoters such as the tac promoter. However, other known bacterial promoters are also suitable. Promoters for use in bacterial systems will also contain a Shine-Dalgarno (S.D.) sequence functionally linked to the DNA encoding the antibody.

Promoter sequences for eukaryotes are known. Almost all eukaryotic genes have an abundant AT region located approximately 25-30 bases above the site where transcription is initiated. Another sequence located 70-80 bases above the start of transcription of many genes is the CNCAAT region, where the symbol N can be represented by any nucleotide. At the 3'-end of most eukaryotic genes, there is an AATAAA sequence that can serve as a signal to add a poly-A tail to the 3'-end of the coding sequence. All of these sequences can conveniently be inserted into eukaryotic expression vectors.

Examples of suitable promoter sequences for use with yeast include promoters for 3-phosphoglycerate kinase or other glycolytic enzymes such as enolase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, pyruvate decarboxylase, phosphofructose isomerase, glucose-6-phosphate isomerase, 3-phosphatose isomerase, and glucokinase.

Other yeast promoters, which are inducible promoters having the additional advantage of growth-driven transcription, are the promoter regions of alcohol dehydrogenase 2, isocytochrome C, acid phosphatase, dehydrating enzymes associated with nitrogen metabolism, metallothionein, glyceraldehyde-3-phosphate dehydrogen dehydrogenase for the disposal of maltose and galactose. Suitable vectors and promoters for use in yeast expression are further described in EP 73657. Yeast enhancers are also predominantly used with yeast promoters.

Transcription of antibodies from vectors in mammalian host cells is controlled, for example, by promoters derived from the genomes of viruses such as polyoma virus, chickenpox virus, adenovirus (e.g. adenovirus 2), bovine papilloma virus, avian sarcoma virus, cytomegalovirus, retrovirus, hepatitis virus -B and, most preferably, monkey virus 40 (SV40), heterologous mammalian promoters, such as the actin promoter or immunoglobulin promoter, heat shock promoters, provided that such promoters are compatible with host cell system.

The early and late promoters of the SV40 virus are conveniently obtained as an SV40 restriction fragment, which also contains the SV40 viral replicator. The immediate early promoter of human cytomegalovirus is conveniently obtained as a HindIII E. restriction fragment. A system for expressing DNA in mammalian host cells using bovine papilloma virus as a vector is disclosed in US Pat. No. 4,419,446. A modification of this system is described in US Pat. No. 4,601,978. See also reference Reyes et al. (1982) Nature 297: 598-601 on the expression of human β-interferon cDNA in mouse cells under the control of the herpes simplex thymidine kinase promoter. Alternatively, a long terminal repeat of Routh sarcoma virus can be used as a promoter.

(v) Component of an enhancer element

Transcription of the DNA encoding the antibody of this invention in higher eukaryotic cells often increases when an enhancer sequence is inserted into the vector. Many enhancer sequences from mammalian genes (globin, elastase, albumin, alpha-fetoprotein and insulin) are now known. However, it is typical to use an enhancer from a eukaryotic cell virus. Examples include a late-side replicator enhancer SV40 (bp 100-270), an early cytomegalovirus promoter enhancer, a polyoma late-side replicator enhancer, and adenovirus enhancers. See also Yaniv (1982) Nature 297: 17-18 link on enhancer elements for activation of eukaryotic promoters. An enhancer can be spliced into a vector at position 5 ′ or 3 ′ of the sequence encoding the antibody, but is preferably located on the 5 ′ side of the promoter.

(vi) Transcription Termination Component

The expression vectors used in eukaryotic host cells (yeast, fungal, insects, plant, animal, human, or nuclear cells from other multicellular organisms) will also contain the sequences needed to terminate transcription and to stabilize mRNA. Such sequences are usually available from 5 'and sometimes from 3' untranslated regions of eukaryotic or viral DNA or cDNA. These regions contain nucleotide segments transcribed as polyadenylated fragments in the untranslated portion of the mRNA encoding the antibody. One useful component of transcription termination is the polyadenylated region of bovine growth hormone. See document WO94 / 11026 and expression vectors disclosed therein.

(vii) Selection and transformation of host cells

Suitable host cells for cloning and expression of DNA in the vectors discussed herein are prokaryotic, yeast, or higher eukaryotic cells described above. Suitable prokaryotes for this purpose include eubacteria such as gram-negative or gram-positive organisms, for example, intestinal bacteria such as Escherichia, e.g. E. coli, Enterobacter, Erwinia, Klebsiella, Proteus, Salmonella, e.g. Salmonella typhimurium, Serratia, e.g. Serratia marcescans and Shigella, as well as bacilli such as B. subtilis and B. licheniformis (e.g. B. licheniformis 41P, disclosed in DD 266710 published April 12, 1989), representatives of the genus Pseudomonas such as P. aeruginosa, and Streptomyces. A preferred host for cloning from E. coli is E. coli 294 strain (ATCC 31446), although other strains are suitable, for example, E. coli B, E. coli X1776 (ATCC 31537) and E. coli W3110 (ATCC 27325). These examples are illustrative rather than restrictive.

In addition to prokaryotes, eukaryotic microbes such as filamentous fungi or yeast are suitable hosts for cloning or expressing vectors encoding antibodies. Among the lower eukaryotic microorganisms, Saccharomyces cerevisiae or ordinary baker's yeast are most often used as hosts. However, it is worth mentioning many other generally available and useful genera, species, and host strains, such as Schizosaccharomyces pombe, Kluyveromyces such as K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATCC 16045), K. wickeramii (ATCC 24178), K. waltii (ATCC 56500), K. drosophilarum (ATCC 36906), K. thermotolerans and K. marxianus, yarrowia (EP 402226), Pichia pastoris (EP 183070), Candida, Trichoderma reesia (EP 244234) , Neurospora crassa, Schwanniomyces such as Schwanniomyces occidentalis and filamentous fungi such as Neurospora, Penicillium, Tolypocladium and Aspergillus such as A. nidulans and A. niger.

Suitable host cells for the expression of glycosylated antibodies are obtained from multicellular organisms. Examples of cells of invertebrate organisms include plant cells and insect cells. Numerous baculovirus strains and variants have been identified, as well as corresponding permissive host cells from insects such as Spodoptera frugiperda (leafworm), Aedes aegypti (mosquito species), Aedes albopictus (mosquito species), Drosophila melanogaster (fruit fly) and Bombyx mori (silkworm ) Many viral strains for transfection are freely available, for example, L-1 NPV Autographa californica variant and Bm-5 NPV Bombyx mori strain, and such viruses can be used in the present invention, especially for transfection of Spodoptera frugiperda cells. Cultures of plant cells such as cotton, cereal, potato, soy, petunia, tomato and tobacco can also be used as hosts.

However, vertebrate cells have long been of greatest interest, and the reproduction of such cells in culture (obtaining tissue culture) has become a routine. Examples of useful mammalian cell lines for use as hosts are monkey CV1 kidney cell line transformed with SV40 (COS-7, ATCC CRL 1651), human embryonic kidney cell line (293 or 293 cells subcloned for suspension culture, Graham et al . (1977) J. Gen Virol. 36:59), infant hamster kidney cells (BHK, ATCC CCL 10), Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al. (1980) Proc. Natl. Acad. Sci USA 77: 4216), Sertoli mouse cells (TM4, Mather (1980) Biol. Reprod. 23: 243-251), monkey kidney cells (CVCC ATL CCL 70 CV1), African kidney cells green monkey (VERO-76, ATCC CRL-1587), human cervical carcinoma cells (HELA, ATCC CCL 2), dog kidney cells (MDCK, ATCC CCL 34), buffalo rat liver cells (BRL 3A, ATCC CRL 1442), human lung cells (W138, ATCC CCL 75), human liver cells (Hep G2, HB 8065), mouse breast tumor cells (MMT 060562, ATCC CCL51), TR1 cells (Mather et al. (1982) Annals N. Y. Acad. Sci. 383: 44-68), MRC 5 cells, FS4 cells and human hepatoma cell line (Hep G2).

Host cells are transformed with the above expression or cloning vectors to generate antibodies and cultured in traditional culture media suitably modified to induce promoters, select transformants, or amplify genes encoding the desired sequences.

(viii) Cultivation of host cells

The host cells used in the present invention for the production of antibodies can be cultured in many different environments. For cultivation of host cells, such commercially available media as Ham F10 medium (Sigma), Minimum Support Medium (MEM), (Sigma), RPMI-1640 (Sigma) and Needle medium as modified by Dulbecco ((DMEM), Sigma) are suitable. In addition, any medium described in references Ham et al. Can be used to cultivate host cells. (1979) Meth. Enz. 58:44, Barnes et al. (1980) Anal. Biochem. 102: 255, U.S. Patent Nos. 4,767,704, 4,657,866, 4,927,762, 4,560,655 and 5,222,069, WO 90/03430, WO 87/00195 or U.S. Pat. Re. 30985. If necessary, any of these media can be supplemented with hormones and / or other growth factors (eg, insulin, transferrin, or epidermal growth factor), salts (eg, sodium chloride, calcium, magnesium, and phosphate salts), buffers (eg, HEPES) , nucleotides (e.g., adenosine and thymidine), antibiotics (e.g., GENTAMYCIN ™), trace elements (defined as inorganic compounds typically present in the final micromolar range), and glucose or an equivalent energy source. In addition, any other necessary additives that may be known to the competent person may be administered in an appropriate concentration. Cultivation conditions such as temperature, pH and the like, which are commonly used in the cultivation of host cells selected for expression, should be apparent to those of ordinary skill in the art.

(ix) Purification of antibodies

Using recombinant techniques, an antibody can be produced intracellularly, in the periplasmic space, or directly secreted into the culture medium. If the antibody is developed intracellularly, then at the first stage of purification, debris particles (host cells or lysed fragments) are removed, for example, by centrifugation or ultrafiltration. Carter et al. (1992) Bio / Technology 10: 163-167 describe a procedure for isolating antibodies that are secreted into the periplasmic space of E. coli. Briefly, cell paste is thawed in the presence of sodium acetate (pH 3.5), EDTA and phenylmethylsulfonyl fluoride (PMSF) for approximately 30 minutes. Cell debris can be removed by centrifugation. If the antibody is secreted into the culture medium, the supernatants from such expression systems are usually primarily concentrated using a commercially available protein concentration filter, for example, Amicon or a Millipore Pellicon ultrafiltration unit. At any of the previous steps, you can connect a protease inhibitor such as PMSF to inhibit proteolysis, as well as antibiotics to prevent the growth of random contaminating microbiological impurities.

An antibody composition prepared from cells can be purified using, for example, hydroxyapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography, with affinity chromatography being the preferred purification technique. The suitability of protein A as an affinity ligand depends on the species and isotype of any immunoglobulin Fc domain that is present in the antibody. Protein A can be used to purify antibodies that are based on the human heavy chains of γl, γ2 or γ4 (Lindmark et al. (1983) J. Immunol. Meth. 62: 1-13). Protein G is recommended for all murine isotypes and for human γ3 (Guss et al. (1986) EMBO J. 5: 15671575). The matrix to which the affinity ligand is attached is most often agarose, but other matrices are also available. Mechanically stable matrices, such as glass with a controlled pore size or poly (styrene divinyl) benzene, provide faster flows and reduce processing time compared to agarose. If the antibody contains a CH3 domain, then Bakerbond ABX ™ resin (J. T. Baker, Philipsburg, NJ) is useful for purification. Other protein purification techniques are available, depending on the antibody to be extracted, such as fractionation on an ion exchange column, ethanol precipitation, reverse phase HPLC, silica gel chromatography, heparin sepharose chromatography (SEPHAROSE ™), anion exchange chromatography or a cation exchange resin (for example, on a column with polyaspartic acid), chromatofocusing, SDS-PAGE and precipitation with ammonium sulfate.

After carrying out any pre-treatment step (s), the mixture containing the antibody of interest and contaminants can be subjected to low pH hydrophobic chromatography using an elution buffer with an approximate pH of 2.5-4.5, preferably at low salt concentration (e.g., approximately 0-0.25 M).

Pharmaceutical Formulations

Therapeutic drug formulations with antibodies are prepared for storage by mixing an antibody of the desired purity with optional physiologically acceptable carriers, excipients or stabilizers (Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980)) in the form of lyophilized formulations or aqueous solutions . Acceptable carriers, excipients, or stabilizers are nontoxic to recipients at the applicable dosages and concentrations and include buffers such as phosphate, citrate and other organic acids, antioxidants including ascorbic acid and methionine, preservatives (such as octadecyldimethylbenzylammonium chloride, hexamethonium chloride, benzalkonium chloride phenol, butyl or benzyl alcohol, such as alkyl parabens such as methyl paraben or propyl paraben, catechin, resorcinol, cyclohexanol, 3-pentanol and m-cresol), a low molar polypeptide weight (less than about 10 residues), proteins such as serum albumin, gelatin or immunoglobulins, hydrophilic polymers such as polyvinylpyrrolidone, amino acids such as glycine, glutamine, asparagine, histidine, arginine or lysine, monosaccharides, disaccharides and other carbohydrates, including glucose, mannose or dextrins, chelating agents such as EDTA, sugars such as sucrose, mannitol, trehalose or sorbitol, salt-forming counterions such as sodium, metal complexes (e.g. Zn-protein complexes) and / or non-ionic surfactants such as TWEEN ™, PLURONICS ™ or polyethylene glycol (PEG).

The pharmaceutical composition contemplated herein may also contain more than one active compound, if necessary for appropriate medical indications, and it is preferred that these compounds exhibit complementary activity and do not adversely affect each other. For example, it may be desirable to introduce an additional immunosuppressant in the formulation. Such molecules are conveniently combined in amounts that are effective against a given target.

The active ingredients can also be enclosed in a microcapsule made, for example, by coacervation or phase boundary polymerization, for example a microcapsule of hydroxymethyl cellulose, gelatin or a microcapsule of (poly) methyl methacrylate, respectively, or introduced into colloidal drug delivery systems (e.g. in liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or in macroemulsions. Such techniques are disclosed in Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980).

Medicinal formulations intended for in vivo administration must be sterile. This is easily achieved by filtration through sterile filter membranes.

Slow release formulations can be made. Suitable examples of sustained release formulations include semipermeable matrices or solid hydrophobic polymers containing an antibody, said matrices being in the form of shaped articles, such as films or microcapsules. Examples of sustained release matrices include polyesters, hydrogels (e.g., (poly) 2-hydroxyethyl methacrylate or (poly) vinyl alcohol), polylactides (US Pat. No. 3,773,919), copolymers of L-glutamic acid and γ-ethyl-L-glutamate, non-degradable ethylene vinyl acetate disintegrating copolymers of lactic acid with glycolic acid such as LUPRON DEPOT ™ (injection microspheres consisting of a copolymer of lactic acid with glycolic acid and leuprolide acetate), as well as poly-D - (-) - 3-hydroxybutyric acid. While polymers such as ethylene vinyl acetate and lactic acid-glycolic acid provide release of molecules for more than 100 days, some hydrogels release proteins for shorter periods of time. When encapsulated antibodies are inside the body for a long time, they can denature or aggregate as a result of contact with a humid environment at 37 ° C, which leads to a loss of biological activity and possible changes in immunogenicity. Rational stabilization strategies can be developed depending on the mechanism involved. For example, if the revealed aggregation mechanism is based on the formation of S-S intermolecular bonds through thio-disulfide exchange, stabilization can be achieved by modifying sulfhydryl residues, lyophilization from acidic solutions, controlling moisture content, using appropriate additives, and developing specific polymer matrix compositions.

Therapeutic options

The use of antibodies of the present invention for the treatment of mammals is contemplated. In one embodiment, the antibody is administered to a mammal, an unrelated human, for example, in order to obtain preclinical data. Typical non-human animals used in treatment studies include non-human primates, dogs, cats, rodents, and other mammals that are undergoing preclinical studies. Such mammals may be representatives of biological species, which are recognized models of diseases to be treated with antibodies, in addition, mammals can be used to study the toxicity of antibodies of interest. In each of these embodiments of the invention, mammalian dose-enhancing studies can be performed. For example, if the test antibody is an anti-NRP1 antibody, then it can be introduced into the body of a rodent host, which is a model of a solid tumor.

In addition or in an alternative to this, the antibody is used to treat a person, for example, a patient suffering from a disease or disorder that may improve as a result of administration of the antibody.

The present invention encompasses anti-angiogenic cancer therapy, a new cancer treatment strategy aimed at suppressing the development of blood vessels in a tumor necessary for the delivery of nutrients that support tumor growth. Since angiogenesis is involved in the growth of both the primary tumor and metastases, the anti-angiogenic treatment proposed by the invention is able to suppress neoplastic tumor growth in the primary focus, as well as prevent tumor metastases in the secondary foci, and therefore help other drugs attack the tumors. Examples of cancer to be treated in accordance with the present invention include, but are not limited to, carcinoma, lymphoma, blastoma, sarcoma, and leukemia. More specific examples of such cancers include squamous cell carcinoma, lung cancer (including small cell lung cancer, non-small cell lung cancer, lung adenocarcinoma and squamous cell carcinoma), peritoneal cancer, hepatocellular cancer, gastric cancer or gastric cancer (including gastrointestinal cancer), pancreatic cancer glands, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladder cancer, hepatoma, breast cancer, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary carcinoma, kidney cancer, liver cancer, prostate cancer, vulvar cancer, thyroid cancer, liver carcinoma and various types of head and neck cancer, as well as B-cell lymphoma (including low-grade / follicular non-Hodgkin's lymphoma (NHL), small cell lymphocytic (SL) NHL, follicular / medium malignancy NHL, diffuse NHL medium malignancy, high-grade lymphoblastic NHL, high-grade small-cell NHL with undigested cell nuclei, NHL with massive lesion, mantle cell lymphoma, AIDS-related lymphoma and m Waldenstrom acroglobulinemia), chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), nappy cell leukemia, chronic myeloid leukemia and post-transplant lymphoproliferative disease (PTLD), as well as abnormal vascular proliferation associated with tumor-related tumors associated with fakomat ) and Meigs syndrome. More specifically, cancers to be treated with the antibodies of the invention include breast cancer, colorectal cancer, colorectal cancer, non-small cell lung cancer, non-Hodgkin’s lymphoma (NHL), renal cell carcinoma, prostate cancer, liver cancer, pancreatic cancer, soft tissue sarcoma, Kaposi's sarcoma, carcinoid carcinoma, head and neck cancer, melanoma, ovarian cancer, mesothelioma and multiple myeloma.

It is contemplated that when used to treat various diseases, such as tumors, the antibodies of the invention can be combined with other drugs suitable for treating the same disease or similar diseases. For cancer treatment, the antibodies of the present invention can be used in combination with traditional cancer treatments such as surgery, radiation therapy, chemotherapy, or combinations thereof.

In some aspects, other drugs useful for combination anti-cancer therapy in combination with the antibodies of the invention include other anti-angiogenic agents. Many antiangiogenic agents have been identified and are currently known in the art, including those listed in Carmeliet and Jain (2000).

In one aspect, the antibody of the invention is used in combination with a VEGF antagonist or with a VEGF receptor antagonist, for example, anti-VEGF antibodies, VEGF variants, soluble VEGF receptor fragments, aptamers capable of blocking VEGF or VEGFR, neutralizing anti-VEGFrzin antibodies, VEGFR and any combination of these funds. In an alternative or in addition to this, two or more anti-NRP1 antibodies can be co-administered to a patient. In a more preferred embodiment, the anti-NRP1 ^A or anti-NRP ^B antibody of the invention is used in combination with an anti-VEGF antibody to produce additive or synergistic effects. Preferred anti-VEGF antibodies include those that bind to the same epitope as the anti-hVEGF antibody A4.6.1. A more preferred anti-VEGF antibody is bevacizumab or ranibizumab.

In some other aspects, therapeutic agents useful for combination antitumor therapy in combination with the antibodies of the invention include an antagonist of other factors involved in tumor growth, such as EGFR, ErbB2 (also known as Her2) ErbB3, ErbB4 or TNF. Preferably, the anti-NRP1 antibody of the invention can be used in combination with low molecular weight tyrosine kinase receptor inhibitors (RTKIs) that target one or more tyrosine kinase receptors, such as VEGF receptors, FGF receptors, EGF receptors, and PDGF receptors. In the art are known many low molecular RTKI, including, but not limited to, vatalanib (PTK787), erlotinib (TARCEVA ^{®), OSI-7904, ZD6474 (ZACTIMA ®), ZD6126} (ANG453), ZDl839, sunitinib (SUTENT ^®), semaxanib (SU5416), AMG706, AG013736, imatinib (GLEEVEC ^® ), MLN-518, CEP-701, PKC-412, lapatinib (GSK572016), VELCADE ^® , AZD2171, sorafenib (NEXAVAR ^® ), XL880 and CHIR-265.

An anti-Nrp antibody of the invention alone or in combination with a second therapeutic agent (such as an anti-VEGF antibody) can be further used in combination with one or more chemotherapeutic agents. In the combination treatment methods contemplated by the invention, a wide variety of chemotherapeutic agents can be used. An exemplary and non-limiting list of chemotherapeutic agents proposed for use in accordance with the invention is presented here in the "Definition" section.

If the anti-Nrp antibody is used in conjunction with another therapeutic agent, then the other therapeutic agent can be administered first in turn, after which the anti-Nrp antibody is administered to the patient. However, it is also possible to administer both drugs simultaneously or administer an anti-Nrp antibody in the first place. Suitable dosages of the second therapeutic agent are consistent with its standard dosages currently recommended, but may be reduced due to the combined effect (synergy) of this agent and anti-Nrp antibody.

The appropriate dose of antibody for the prevention or treatment of the disease will depend on the type of disease to be treated, on the severity and nature of the course of the disease, on whether the antibody is administered for therapeutic or prophylactic purposes, on previous therapy, on the patient’s clinical history, on the patient’s response to antibody and from the decisions of the attending physician. The antibody is appropriately administered to the patient once or as a series of therapeutic doses.

Depending on the type and severity of the disease, the possible initial dose of the antibody for administration to the patient is approximately 1 μg / kg to 50 mg / kg (e.g. 0.1-20 mg / kg), regardless of whether a single administration is carried out, several fractional injections or continuous infusion. A typical daily dose may vary in the approximate range of 1 μg / kg to 100 mg / kg or more, depending on the above factors. When re-administered for several days or more, depending on the clinical condition, treatment is continued until the desired suppression of disease symptoms is achieved. However, other dosing regimens may be useful. In a preferred aspect, the antibody of the invention is administered regularly at intervals of two to three weeks in a dose ranging from about 5 mg / kg to 15 mg / kg. More preferably, such a dosage regimen should be used in combination with a chemotherapy regimen (such as first-line therapy for the treatment of metastatic colorectal cancer). In some aspects, the chemotherapy regimen includes the traditional intermittent administration of high doses of drugs. In some other aspects, chemotherapeutic agents are administered using reduced and more frequent doses without scheduled breaks (“uniform chemotherapy”). The progress of therapy carried out in accordance with the present invention can be easily monitored using traditional techniques and analyzes.

The antibody composition can be formulated, dosed and administered in accordance with the principles of good medical practice. Factors worthy of consideration in this aspect include the particular disease to be treated, the specific mammal to be treated, the clinical condition of the patient, the cause of the disease, the place of delivery of the drug to the body, route of administration, timing of administration, and other factors known to practitioners. A “therapeutically effective amount” of antibody for administration to a patient is determined by the above considerations and is the minimum amount necessary to prevent, alleviate, or treat a disease / disorder. An antibody is not necessary, but optionally combined in a prescription with one or more agents currently used to prevent or treat the diseases in question. The effective amount of such adjuvants depends on the amount of antibody present in the formulation, on the type of disease or treatment, and also on the other factors discussed above. Usually they are used in the same doses and with the same methods of administration, which were indicated here above, or approximately in an amount of from 1 to 99% of the previously used dosages. Generally, alleviating or treating a disease / disorder involves reducing one or more symptoms or medical problems associated with the disease / disorder. In the case of cancer, a therapeutically effective amount of the drug can lead to one or more of the following goals: reducing the number of cancer cells, reducing the size of the tumor, suppressing (that is, decreasing and / or stopping) peripheral organ infiltration with cancer cells, suppressing tumor metastasis, suppressing certain tumor growth rates and / or some relief of one or more symptoms associated with the cancer. To the extent that the drug is able to prevent tumor growth and / or kill existing cancer cells, it can be cytostatic and / or cytotoxic. In some embodiments, a composition of his invention can be used to prevent an initial or recurrent outbreak of a disease / disorder in a subject or mammal.

The possibilities of non-therapeutic use

The antibodies of the invention can be used as affinity purification agents. In this process, antibodies are immobilized on a solid phase, for example on Sephadex resin or on filter paper, using methods well known in the art. The immobilized antibody is brought into contact with a sample containing the purified antigen, after which the support is washed with a suitable solvent, which removes substantially all of the material of the sample except for the purified antigen bound to the immobilized antibody. Finally, the support is washed with another suitable solvent, for example glycine buffer, pH 5.0, which separates the antigen and the antibody.

Antibodies of the invention may also be useful in diagnostic assays, for example, to detect the expression of an antigen of interest in specific cells, tissues or serum.

In typical diagnostic applications, the antibody is labeled with a detectable component. Numerous tags are available, which are usually grouped into the following categories.

(a) Radioisotopes such as ³⁵ S, ¹⁴ C, ¹²⁵ I, ³ H and ¹³¹ I. An antibody can be labeled with a radioisotope using the techniques described, for example, in Current Protocols in Immunology, Volumes 1 and 2, Coligen et al . (1991) Ed. Wiley-Interscience, New York, New York, Pubs, and radioactivity can be measured using a scintillation counter.

(b) Fluorescent labels are also available, such as rare earth chelates (europium chelates) or fluorescein and its derivatives, rhodamine and its derivatives, dansil, lissamine, phycoerythrin and Texas red. Fluorescent labels can be conjugated to an antibody using the techniques disclosed, for example, in Current Protocols in Immunology, supra. Fluorescence can be quantified using a fluorimeter.

(c) Various enzyme-substrate type labels are available, and a review of some of them is presented in US Pat. No. 4,275,149. An enzyme typically catalyzes a chemical change in a chromogenic substrate that can be measured by various methods. For example, an enzyme can catalyze a color change in a substrate, which can be measured by spectrophotometry. Alternatively, the enzyme may alter the fluorescence or chemiluminescence of the substrate. Techniques for quantifying fluorescence changes are described above. A chemiluminescent substrate becomes electronically excited as a result of a chemical reaction, after which it can emit light that can be measured (using, for example, a chemiluminometer), or transfers energy to a fluorescent acceptor. Examples of enzymatic labels include luciferase (e.g., firefly luciferase and bacterial luciferase, US Patent No. 4,737,456), luciferin, 2,3-dihydrophthalazinediones, malate dehydrogenase, urease, such as horseradish peroxidase (HRPO), alkaline galactose, β-glucose, lysozyme, saccharide oxidases (e.g. glucose oxidase, galactose oxidase and glucose-6-phosphate dehydrogenase), heterocyclic oxidases (e.g. uricase and xanthine oxidase), lactoperoxidase, microperoxidase, etc. Antibody conjugation techniques are described in O'Sullivan et al. (1981) Methods for the Preparation of Enzyme-Antibody Conjugates for use in Enzyme Immunoassay, in Methods in Enzym. (ed J. Langone & H. Van Vunakis), Academic press, New York 73: 147-166.

Examples of enzyme-substrate combinations include:

(i) horseradish peroxidase (HRPO) with hydrogen peroxide as a substrate, where hydrogen peroxide oxidizes the pigment precursor (e.g., orthophenylenediamine (OPD) or 3.3 ', 5.5'-tetramethylbenzidine hydrochloride (TMB)),

(ii) alkaline phosphatase (AP) with paranitrophenyl phosphate as the chromogenic substrate; and

(iii) β-D-galactosidase (β-D-Gal) with a chromogenic substrate (e.g. p-nitrophenyl-β-D-galactosidase) or with a fluorogenic substrate (4-methylumbelliferyl-β-D-galactosidase).

Those skilled in the art will recognize many other combinations of enzymes and substrates. For a general overview, see US Patent Nos. 4,275,149 and 4,318,980.

Sometimes the label is conjugated to the antibody indirectly. A competent specialist will be aware of the various techniques to achieve this result. For example, an antibody can be conjugated to biotin, and any of the labels of the three broad categories mentioned above can be conjugated to avidin, or vice versa. Biotin selectively binds to avidin, and thus, the label can be conjugated to the antibody in this indirect way. Alternatively, to achieve indirect conjugation of the label with the antibody, the antibody is conjugated to small hapten (e.g., digoxin), and one of the labels of the various types mentioned above is conjugated to an anti-hapten antibody (e.g., antidigoxin). Thus, indirect conjugation of the label to the antibody can be achieved.

In yet another embodiment, the antibody does not need to be labeled, and its presence can be detected using a labeled antibody that binds to the first antibody.

Antibodies of the present invention can be used in any of the known assay methods, for example, competitive binding assays, direct and indirect sandwich assays, and immunoprecipitation assays. Zola, Monoclonal Antibodies: A Manual of Techniques, pp. 147-158 (CRC Press, Inc. 1987).

Competitive binding assays are based on the ability of the labeled standard to compete with a test sample for binding to a limited amount of antibody. The amount of antigen in the test sample is inversely proportional to the amount of standard that binds to antibodies. In order to facilitate the determination of the amount of a standard that enters into an immune connection, antibodies are usually solubilized (transferred to an insoluble state) before or after competition so that the standard and the analyzed sample associated with the antibodies can be easily separated from the standard and the analyzed sample, remaining unconnected.

Sandwich assays include the use of two antibodies, each of which is capable of binding to a particular immunogenic part or epitope of the protein to be detected. In a sandwich analysis, the test sample binds to the first antibody that is immobilized on a solid support, and then the second antibody binds to the analyzed sample, thus forming an insoluble three-part complex. See, for example, US Patent No. 4,376,110. The second antibody alone can be labeled with a detectable component (direct sandwich assays) or can be measured using an anti-immunoglobulin antibody that is labeled with a detectable component (indirect sandwich assays). For example, one type of sandwich analysis is an ELISA in which the detectable component is an enzyme.

For immunohistochemical analysis, the tumor sample can be fresh, frozen or embedded in paraffin and fixed, for example, with a preservative such as formalin.

Antibodies can also be used for in vivo diagnostic assays. Typically, an antibody is labeled with a radionuclide (such as ¹¹¹ In, ⁹⁹ Tc, ¹⁴ C, ¹³¹ I, ¹²⁵ I, ³ H, ³² P or ³⁵ S) or pigment so that the tumor can be localized using immunoscintigraphy.

In one embodiment of the invention, a method for detecting NRP1 in a native biological sample (e.g., tissue, blood, serum, cerebrospinal fluid) or in a prepared biological sample may include the step of contacting the antibody of this invention and the sample, followed by observing the antibody binding against NRP1 with NRP1 in the sample or determining the amount of anti-NRP1 antibody bound to NRP1 in the sample. In another embodiment, a method for detecting NRP1 in a subject comprises the step of administering an antibody of the invention to a subject and observing the association of the anti-NRP1 antibody with NRP1 in the subject or determining the amount of anti-NRP1 antibody bound to NRP1 in the subject (e.g., human, mouse, rabbit , rats, etc.).

Diagnostic kits

For convenience, the antibody of the invention can be delivered in a kit, i.e. in a packaged combination of reagents with a given amount and with instructions for conducting a diagnostic analysis. When an antibody is labeled with an enzyme, the kit will include the substrates and cofactors necessary for the enzyme (for example, a substrate precursor that serves as a source of a detectable chromophore or fluorophore). In addition to this, other additives may be included in the kit, such as stabilizers, buffers (e.g., blocking buffer or lysis buffer) and the like. The relative amount of different reagents can vary widely, so that it is possible to envisage the concentration in the solution of those reagents that significantly optimize the sensitivity of the analysis. In particular, the reagents can be presented in the form of dry powders, usually lyophilized, including fillers, which upon dissolution will create a reaction solution having the desired concentration.

Products

In yet another embodiment of the invention, there is provided an article comprising materials useful for treating the diseases described above. The product implies the presence of a container and a label. Suitable containers include, for example, bottles, vials, syringes, and test tubes. Containers can be made of various materials, such as glass or plastic. The container holds a composition that is effective in treating a pathological condition and may have a sterile access port (for example, the container may be an intravenous solution bag or a vial with a stopper that can be pierced with a hypodermic needle). The active agent in the composition is an antibody. A label affixed to or attached to the container indicates that the composition is useful for treating a particular disease. The product may further comprise a second container containing a pharmaceutically acceptable buffer, such as phosphate buffered saline, Ringer's solution, and dextrose solution. It may further include other materials desirable from the point of view of the buyer and user, including other buffers, diluents, filters, needles, syringes and packaging inserts with instructions for use.

The following examples are intended only to illustrate the practical implementation of the present invention and should in no way be construed as limitations. The disclosures of all patents and scientific publications mentioned herein in their entirety are expressly incorporated by reference.

EXAMPLES

EXAMPLE 1

Design and function of antibodies against Nrp1 ^B  and against panNrp ^A

A strategy for developing phage antibodies capable of selectively blocking the binding of either semaphorin or VEGF to Nrp1 has been described by Liang et al., J Mol Biol 366, 815-29 (2007) and Pan et al., Cancer Cell 11, 53-67 (2007 ) These monoclonal antibodies were designed as a tool for distinguishing between Nrp1-mediated responses to one or another ligand, as well as a means of assessing their therapeutic potential in murine tumor models.

In short, a phage library of synthetic human antibodies was constructed that was built on a single matched framework with a variety of VH / VL. Details of the construction and selection of this antibody library are described in Liang et al., Supra, and can also be found in WO 03/102157 published December 11, 2003, the entire disclosures of which are incorporated herein by reference. Functional blocking antibodies to human and mouse NRP1 were generated from this library. Antibodies mapping the b1b2 domain of NRP1 were able to block the binding of VEGF and NRP1, as well as VEGF-induced migration of HUVEC cells. Identified antibodies included an antibody blocking VEGF (clone YW107.4.87, against Nrp1 ^B ), which binds Nrp1 with an affinity of 0.2 nM (Liang et al., Supra, Pan et al., Supra). Although this antibody blocks the interaction between VEGF and Nrp1, it does not counteract the function of Sema3A. Under in vivo conditions, the anti-Nrp1 ^B antibody not only reduces vascular reconstruction in the mouse retina, but also acts additively with anti-VEGF therapy, slowing tumor growth (Liang et al., Supra and Pan et al., Supra).

Following the same approach using a phage library of synthetic human antibodies built on a single matched framework, an antibody was developed (clone YW68.1 1.26, against panNrp ^A ), which cross-reacts with both Nrp1 and Nrp2 with an affinity of 0.21 and 0.15 nM, respectively (Fig. Sl). The sequences of the variable domain of the light and heavy chains of YW68.11.26 and the related YW68.11 are shown in FIGS. 7 and 8, respectively. The amino acid sequences of Fab fragments of anti-panNrp ^A IgG1 antibodies YW68.11 and YW68.11.26 are shown in FIGS. 9A and 9B, respectively.

Unlike the anti-Nrp1 ^B antibody, the anti-panNrp ^A antibody does not affect the binding of VEGF ₁₆₅ or VEGF-C (data not shown). Evaluation of the ability of both antibodies to suppress Sema3A-mediated collapse of axon growth cones from the ganglia of the posterior roots (DRG) of the mouse spinal cord (Fig. 1). The addition of Sema3A leads to a reduction in actin processes in DRG growth cones: this anti-panNrp ^A antibody but not anti-Nrp1 ^B completely resists this action.

EXAMPLE 2

Structural studies of neuropilin / antibody complexes

Materials and methods

Functional assays and antibody binding affinity

Collapse assays and determination of the binding affinity of the anti-Nrp antibody were performed as previously described (He, Z and Tessier-Lavigne, M., Cell 90, 739-51 (1997); Liang et al., Supra and Pan et al ., above).

Protein expression and purification for crystallographic studies

Nrp1-b1, Nrp1-b1b2 and Nrp2-b1b2 (see table S1 for domain boundaries) were cloned into pET15b (Novagen) and expressed in E. coli after induction at 37 ° C (Nrp1-b1 and -b1b2) or at 16 ° C (Nrp2-b1b2). All type b neuropilin fragments are expressed as soluble proteins without the need for a refolding protocol. After cell lysis, the proteins were purified using a nickel-nitrile tri-acetic acid (Ni-NTA) resin in 50 mM Tris (pH 8.0), 300-500 mM NaCl and 20 mM imidazole, and eluted in the same buffer with 250 mM imidazole. His6 tails were removed by thrombin, and samples were further purified by gel filtration chromatography using a Superdex-75 column equilibrated in 25 mM Tris (pH 8.0) and 150 mM NaCl.

To facilitate the secretion of Nrp1-a2b1b2, Nrp2-a2b1b2, Nrp2-a1a2b1b2 and full-sized Nrp2-ECD (Table S1), recombinant baculoviruses were generated from Hi5 cells. Nrp2-a1a2b1b2 and full-sized Nrp2-ECD were subcloned with the Nrp2 native secretion signal and His ₆ C-terminal tail into pENTR / D-TOPO (Invitrogen) and recombined into pDEST8 (Invitrogen) to generate a viral bacmid. Nrp1-a2b1b2 and Nrp2-a2b1b2 were cloned into pAcGP67B (Clonetech). After infection, the culture medium was collected and 50 mM Tris (pH 8.0), 5 mM CaCl ₂ and 1 mM NiCl _{2 were} added to it. Proteins were purified using Ni-NTA and gel filtration chromatography, as described for neuropilin constructs expressed by bacteria.

Fab fragments of antibodies against Nrp1 ^B (YW107.4.87) and against panNrp ^A (YW68.11.26) were expressed in E. coli, captured on a column with Protein G carrier balanced in PBS, and eluted with 0.58% acetic acid. The protein fractions were further purified by ion exchange chromatography (SP-Sepharose) in 20 mM MES (pH 5.5) and eluted with a NaCl gradient from 0 to 250 mM. The Fab / Nrp complexes were typically mixed in a 1: 1 molar ratio and further purified using a Superdex-200 column equilibrated in 25 mM Tris-HCl (pH 7.5) and 200 mM NaCl. For crystallization, all unbound neuropilin and the Nrp / Fab complex were concentrated, as detailed in Table S1.

Crystallization, determination and refinement of the structure

All crystals were obtained by vapor diffusion at 19 ° C by mixing equal volumes of protein and solution from the wells (see table 1 for details). For cryoprotection, crystals were usually transferred to the mother liquor with the addition of 20% glycerol or ethylene glycol (Table S1). Crystals of the Nrp1-b1 / Fab complex were transferred to 10 mM Hepes (pH 7.2), 25% PEG 1500 and 10% ethylene glycol, allowing them to dehydrate overnight in the vicinity of 10 mM Hepes (pH 7.2), 25% PEG 1500 and 20% ethylene glycol, after which it was quickly frozen in liquid nitrogen. Data arrays were collected at Advanced Light Source (ray lines 5.0.1 or 5.0.2) or at the Stanford Synchrotron Radiation Laboratory (ray lines 9-2 or 11-1). The data were processed using Denzo and Scalepack computer programs from the HKL Suite (Otwinowski, Z & Minor, W., Methods in Enzymology 276,307-326 (1997)). Cell parameters and data statistics are summarized in table 1.

All crystal structures were elucidated by molecular substitution using a phaser (McCoy et al., Acta Crystallogr D Biol Crystallogr 61, 458-64 (2005)). The structure of Nrp1-b1b2 was determined using the crystal structure of Nrp1-b1 (pdb, accession number 1KEX) (Lee, et al., Structure 11, 99-108 (2003) as a study model for each coagulation factor domain. Refined coordinates of the Nrp1 structure -b1b2 served as a search probe for the Nrp2-b1b2 structure.The monosymmetric form of the Nrp2 a1a2b1b2 / anti panNrp ^A -Fab complex was determined using the Nrp2-b1b2 structure and Fab fragments containing either variable domains (V _H / V _L ) or constant domains (C _H1 / C _L ) from the B20-4 / VEGF complex (pdb, accession number 2FJH). Domain a2 was identified by by molecular substitution using the N-terminal CUB domain of MASP-2. The a1 domain could not be localized by molecular substitution and its position was manually determined by electron density. The Nrp1-b1 / Fab complex was clarified using the crystal structure Nrp1-b1 and Fab B20-4 fragments described above All other structures were elucidated using the adjusted coordinates of the b1b2 structures and the CUB a2 domain from the Nrp2-a1a2b1b2 / Fab complex. Atomic models were constructed using the Coot computer program (Emsley, P. & Cowtan, K. Coot, Acta Crystallogr D Biol Crystallogr 60, 2126-32 (2004)) and refined using the Refmac computer program (Murshudov, et al., Acta Crystallographica D53, 240-255 (1997)).

Crystallization of Nrp1, Nrp2 and Neuropilin / Fab Complexes

Two different strategies were used to obtain protein for structural research. Smaller neuropilin fragments, which include only type b domains, were expressed in E. Coli, while larger Nrp constructs required the production of secreted protein from insect cells infected with baculovirus. An overview of the seven structures is presented in FIG. Briefly, crystals of the VEGF binding part (b1b2) Nrp1 and Nrp2 were diffracted to a maximum resolution of 1.8 and 1.95 Å, respectively. The crystals of the a2b1b2 domains Nrp1 and Nrp2 were improved to a resolution of 2.0 and 2.3 Å, respectively, and the b1 domain of Nrp1 in combination with the VEGF-blocking Fab against Nrp1 ^{B was} diffracted to a resolution of 2.2 Å. Finally, the crystal structure of the a1a2b1b2 Nrp2 domains has been elucidated in combination with a semaphorin-blocking Fab against panNrp ^A. Two crystalline forms of this complex were identified that diffracted to 2.75 and 3.1 Å. All structures were clarified by molecular substitution and described with final values of R _work / R _free less than 20/25% with good stereochemistry (table 1). The a2b1b2 Nrp1 structures had two N-linked glycosylation sites at opposite ends of the a2 domain (FIG. 3B). The monosymmetric form of the Nrp2 / Fab complex also shows the presence of two glycosylation sites within the a2 domain (Fig. 3A), however, these sugar components are poorly defined in the electron density of the Nrp2-a2b1b2 structure and, therefore, are not modeled (Fig. 3B).

Common domain architecture of the neuropilin ectodomain

The extracellular region of neuropilin is often subdivided into three blocks that describe the domains of primary semaphorin binding (a1a2), VEGF binding (b1b2), and dimerization (c), respectively (Ellis, LM, Mol Cancer Ther 5, 1099-107 (2006)). Indeed, in the recently described crystal structure b1b2 of rat Nrp1 (Vander Kooi, CW, et al., Proc Natl Acad Sci USA 104, 6152-6157 (2007)), a large interface was identified between the domains b1 and b2, and, since the residues, hidden between domains, locked up in sequence, this configuration of the b1b2 domain was recognized as a common property of the Nrp family. Four crystal structures of Nrp were found here, which include the domains a2, b1, and b2 (Figs. 2, 3), based on different crystallization conditions (Table S1). Two models include a1 domains associated with Fab fragments, while the other two models include calcium ions in a2 domains. Despite these differences, the three domains (a2, b1, and b2) combine the same configuration in all crystal structures and are tightly packed around the pseudo triple axis (Fig. 3). The structures a2b1b2 Nrp1 and Nrp2 are very similar and overlap at the rmsd (standard deviation) 1.2 Å level at 317 Ca atoms. The interface between b1 and b2 is preserved in both neuropilins, being hidden approximately 1200-1500 Å ² from the surface accessible to the solvent. Interestingly, the interface between the a2 and b1b2 domains is also conserved in all structures and is hidden for more than 2000 Å ² from the surface accessible to the solvent, which indicates that all three domains form a rigid structural block (Fig. 3).

In contrast, the domain arrangement between a1 and a2b1b2 is not fixed between the two crystalline forms of the Nrp2 / anti panNrp ^A- Fab complex. When a2b1b2 overlay disrupts the core of the receptors, the a1 domains shift by about 7 Å relative to each other. The interface between a1 and a2b1b2 is small (hidden surface area ≈800 Å ² ) and not preserved. The loss of strong interactions means the flexibility of the a1 domain, which suggests that it may undergo conformational changes with respect to the rest of the Nrp in solution or upon binding to the receptor (Figure 3C).

CUB Nrp domains include calcium and semaphorin binding site

Domains a1 and a2 Nrp1 and Nrp2 are CUB domains (Ellis, L. M, Mol Cencer Ther 5, 1099-107 (2006)). In the CUB domain prototype, spermadgesin (Romero, A., et al., Nat Struct Biol 4, 783-8 (1997)), the fold contains two 5-strand β-leaves that form a β-sandwich. The yarns β2, β4, β9, β6 and β7 form one β-leaf with another leaf containing the yarns βl, β3, β10, β5 and β8. However, β1 and β2 filaments are often absent in the CUB domains of several proteins of the complement family (Feinberg, H. et al., EMBO J 22, 2348-59 (2003), Gregory, LA et al., J Biol Chem 278, 32157-64 ( 2003), Gregory, LA et al., J Biol Chem 279, 29391-7 (2004)). Like these proteins, the a1 and a2 Nrp domains lose the β1 strand (Fig. 4). In general, CUB domains show a high degree of similarity. For example, the a2 Nrp domain is structurally similar to the a2 Nrp1 domain (rmsd - 0.9 Å), the a1 Nrp2 domain (rmsd - 0.9 Å), and the spermadhesin CUB domain (Romero, A. et al., Above) (rmsd - 1 , 4 Å).

The crystal structures of the CUB domains in the proteins of the complement family C1s and MASP-2 ^36.37 contain a calcium binding site at one end of the sandwich. The a2b1b2 structures in Nrp1 and Nrp2 also contain a bound ion at this position within the a2 domain (Figs. 3B, 4A). In the Nrp1 structure, the ion is clearly observed in electron density (Fig. S2), although divalent cations were not included during crystallization (Table S1). In Nrp1, the calcium ion is coordinated by three negatively charged side chains (Glu ¹⁹⁵ , Asp ²⁰⁹ and Asp ²⁵⁰ ), two carbonyl oxygen (Ala ²⁵² and Ile ²⁵³ ), and also a water molecule (Fig. 4B). Similarly, the coordination of calcium in Nrp2 affects three negatively charged amino acids (Glu ¹⁹⁷ , Asp ²¹¹ and Asp ²⁵² , see FIG. S2). These three residues are completely conserved in the a2 domains of twelve biological species with distant affinity (Fig. S3).

Sequence alignment in the a1 and a2 Nrp domains (FIG. S2) shows that this triad of charged residues is also strictly conserved within the a1 Nrp domain. However, the complex Nrp2 / against Nrp ^A -Fab does not indicate the presence of a bound ion in the a1 domain. The loops contributing to the residues that determine the putative Ca ²⁺ binding site in a1 are poorly ordered, which suggests that Ca ²⁺ plays a role in the coagulation of the domain. Based on the high degree of conservatism of the sequence, we can assume that they are probably domains.

Neuropilins function as coreceptors for selected members of class 3 of the semaphorin family. The N-terminus contains the signature domain "Sema", a 7-lobed β-propeller (Antipenko, A., et al., Neuron 39, 589-98 (2003)), which is necessary for the binding of the a1a2 domains of neuropilins. The anti-panNrp ^A antibody blocks Sema3 binding to both Nrp1 and Nrp2 (FIG. 1), but does not affect VEGF binding (data not shown). In the structure of the Nrp2 / anti-panNrp ^A- Fab complex (Fig. 4C), the Fab fragment contacts only the a1 Nrp2 domain on the β8-5-10-3 sandwich plane, hidden from the surface accessible to the solvent by about 1400 Å ² . The interface recognized by this antibody is well conserved between Nrp1 and Nrp2, since 11 of the 14 Nrp2 residues that have more than 25% of the accessible surface hidden at the interface are identical in both receptors (Figure 4C). Recognition of the epitope that is conserved between Nrp1 and Nrp2 explains the ability of an antibody to bind both receptors with affinity in the subnanomolar range (FIG. S1).

On the surface of the antibody, 11 side chains contact Nrp2 from the CDRs of L3, H2 and H3, 7 of which are aromatic in nature (Fig. 4D). Previous studies have identified the semaphorin binding site in Nrp1 using site-specific mutagenesis (Gu, C. et al., J Biol Chem 111, 18069-76 (2002)). Based on the semaphorin model, residues were selected for amino acid substitutions on the surface of the a1 domain, which are supposedly available for the solvent. Using this approach, many mutants were identified that break the interactions between Sema3A and Nrp1 (Gu et al., Supra), and when mapping on model a1 Npr1, they turned out to be located on the loops of one of the poles of β-sandwich a1 (Fig. 4C). Substitutions at the other end of the domain do not affect Sema3A binding (Gu et al., Supra). Mutations that disrupt Sema3A binding are adjacent to the epitope recognized by the Fab fragment blocking Sema3A (Fig. 4C), and this strictly suggests that the sema domain binds loops and face of the 8-5-10-3 sandwich within the a1 Nrp2 domain. The location of the semaphorin binding site is also adjacent to the putative calcium binding site in the a1 domain (Figure 4C), suggesting that calcium binding may play a role in the interactions between the ligand and the receptor.

Type b Nrp domains contain heparin and VEGF binding sites

Domains b from structures b1b2 of human neuropilin (Fig. 5A and references Vander Koo et al., Supra, Lee et al., Supra) combine significant homology with phospholipid binding modules (type C2) of coagulation factors V and VIII (F5 / 8) (Macedo-Ribeiro et al., Nature 402, 434-9 (1999), Pratt, KP, et al., Nature 402, 439-42 (1999)) and with the galactose binding domain of bacterial sialidase (Gaskell, A. et al ., Structure 3, 1197-205 (1995)). Together, these domains define a fold of discoidin, which is topologically classified as a curved β-barrel according to the type of roll with jam, consisting of eight core β-threads. One pole of the domain contains three elongated hairpins or loops that typically constitute a ligand binding site for members of the discoidin family (Macedo-Ribeiro et al., Supra, Pratt et al., Supra, Gaskell et al., Supra). Fragment b1b2 Nrp1 and Nrp2 combines 50% sequence identity and overlay with r.m.s.d. 2.3 Å at 307 Ca atoms (Fig. 5A). While the b1 domains Nrp1 and Nrp2 are almost indistinguishable (rmsd = 0.6 Å), the b2 domains have not so good overlap (rmsd = 2.7 Å), and their differences are largely associated with different conformation of the hairpins ( figa). Because these hairpins often define binding sites within the discoidin family (Vander Kooi et al., Supra, Lee et al., Supra, Macedo-Ribeiro et al., Supra, Pratt et al., Supra, Gaskell et al., Supra) , the dissimilarity between Nrp1 and Nrp2 may be one of the ways in which different neuropilins recognize different binding partners.

Despite the differences between the b2 domains of Nrp1 and Nrp2, the domains of b are densely packed and form a rigid platform (Fig. 5). The interdomain connection forms a deep cleft, which runs almost perpendicular to the long axis of two domains b (Fig. 5). This cleft generated by the β4: β5 loop of the b1 domain and the β5: β6 loop of the b2 domain is surrounded by many positively charged residues, and, based on mutagenesis experiments with rat Nrp1 (Vander Koo et al., Above), represents heparin binding sites in Nrp . In addition, the electropositive region is represented in the structure of both human Nrp (Fig. 5B). The C-terminal domain of VEGF ₁₆₅ (also known as VEGF ₅₅ ) (Fairbrother et al., Structure 6, 637-48 (1998)) also contains a heparin binding site. Since heparin increases the affinity of b1b2 in VEGF ₁₆₅ up to 100 times (Mamluk, R. et al., J Biol Chem 277, 24818-25 (2002), Fuh, G. et al., J Biol Chem 275, 26690-5 (2000 )), it is possible that Nrp use heparin to complement VEGF ₁₆₅ in this area.

In addition to this heparin-mediated interaction between exon 7, VEGF _{165 the} VEGF C-terminal tail (CDKPRR _COOH ) encoded by exon 8 can directly bind to Nrp. In the crystal structure of rat Nrp1-b1b2 in complex with tuftsin (Vander Kooi et al., Supra), tetrapeptide mimetic (TKPR _COOH ) of VEGF tail (von Wronski, MA et al., J Biol Chem 281, 5702-10 (2006)) C-terminal arginine is tightly hidden in the acid groove generated by residues from the conserved hairpins of domain b1. In several of our structures, including all three Fab complexes, the C-terminal residue (histidine) from a symmetrically linked molecule occupies the same acid pocket in the tuftsin peptide (Fig. S4).

To further refine the residues involved in VEGF binding, the surface conservation of Nrp residues on the b1b2 surface was investigated (FIG. 5C). Between the twelve proteins Nrp1 and Nrp2, two contiguous sites were preserved (Fig. S3). One of these sites maps directly to the tuftsin binding site, while the second site includes residues at the border of the heparin binding site, which suggests the presence of an additional region for interactions between Nrp and VEGF.

In the crystal structure of the Nrp1-b1 / anti-Nrp1 ^B- Fab complex (Fig. 5D), the Fab fragment contacts an epitope located between these two putative VEGF binding sites and partially overlaps with a tuftsin binding cleft. The epitope of this VEGF blocking antibody is unusual because it includes only residues from CDR L1 and H3. On average, the Fab / antigen interfaces are hidden at 1680 Å ² from the surface accessible to the solvent (Lo Conte et al., J Mol Biol 285.2177-98 (1999)), however, only Nrp1-b1 / Fab are protected 900 Å ² . Despite the small interface, the anti-Nrp1 ^B antibody binds strongly to Nrp1 at an affinity level of 0.2 nM (Liang et al., Supra, Pan et al., Supra).

VEGF and Sema3A do not compete for Nrp binding

In the Nrp / Fab crystal structures contemplated by the present invention, the binding epitopes that block the binding of VEGF and semaphorin are 65 Å apart and located on opposite sides of the Nrp (Fig. S5). Several studies have found that VEGF and semaphorins compete for binding on the cell surface, and this competition affects a partially overlapping binding site in domain b1 (Gu et al., Supra, Miao, HQ et al., J Cell Biol 146, 2177- 98 (1999), Narazaki, M. & Tosato, G., Blood 107, 3892-901 (2006)). Since the carboxyl tails of VEGF ₁₆₅ and class 3 semaphores are rich in basic (alkaline) residues, it has been suggested that these tails can compete for the electronegative groove created by the hairpins in domain b1 (Vander Kooi et al., Supra, Lee et al., Supra ) In the recently published b1b2 / tuftsin crystal structure, it has been identified that the tail of VEGF ₁₆₅ appears to occupy this binding site. We previously showed that the anti-NRP1 ^B antibody does not counteract Sema3A-mediated axon collapse from the posterior root ganglia (DRG) (Liang et al., Supra, Pan et al., Supra), suggesting that Sema3 C-terminal tails do not contact the same groove.

In previous competition experiments (Gu et al., Supra, Miao et al., Supra, Narazaki et al., Supra), VEGF and semaphorins that contained a heterologous label (e.g., alkaline phosphatase) at the C-terminus of the protein were involved. It is possible that the observed competition is more likely the result of steric collisions of the free ends than the direct competition between the tails of VEGF and Sema3. We examined the ability of VEGF _{165 to} counteract Sema3-mediated axonal growth collapse from DRG. Even at a concentration of 100 mm VEGF ₁₆₅ does not affect the ability of Sema3A to cause a decrease in actin processes (figure 1). These data demonstrate that Sema3 does not compete with VEGF for binding to the b1 domain, and the tail of semaphorins contacts another site within this domain.

Neuropilin Dimerization Models

Nrp1 and Nrp2 can form homo- or heteromultimers even in the absence of a ligand (Takahashi, L. et al., Cell 99, 59-69 (1999), Chen, H. et al., Neuron 21, 1283-90 (1998), Giger, RJ et al., Neuron 21, 1079-92 (1998), Takahashi et al., Nat Neurosci 1, 487-93 (1998)), and although the exact stoichiometry of neuropilin complexes has not yet been established, it is widely accepted that Nrp form homodimers upon ligand binding. Current evidence suggests that domain c is necessary but not sufficient for Nrp oligomerization because truncated mutants that have lost this domain exhibit less multimerization compared to full-sized ECD (Takahashi et al., Cell 99, 59-69 (1999), Chen et al ., supra, Giger et al., supra, Nakamura et al, Neuron 21, 1093-100 (1998)). Neuropilin constructs that include the a1a2 and b1b2 domains have greater affinity for certain VEGF isoforms than constructs that span only the b1b2 domains (Mamluk, R. et al., J. Biol. Chem. 277, 24818-25 (2002), Karpanen, T. et al., FASEB J 20, 1462-72 (2006), Giger, RJ et al., Neuron 25, 29-41 (2000)), although the a1a2 domains do not bind VEGF. Therefore, it is quite acceptable that the a1a2 domains directly contribute to the Nrp dimerization, i.e., enhance the interactions between Nrp and the VEGF dimer by stabilizing the 2: 2 complex. Interestingly, the crystal structures of Nrp2-a1a2b1b2 of two different crystal forms (Table 1) contain a conserved crystallographic interface, which is mediated by a1. In this dimer, the a1 domains are arranged in the form of a rough antiparallel structure along the β7 and β8 filaments. The interface is hidden approximately 1200 Å ² from the surface area accessible to the solvent and is controlled by hydrophobic interactions (Fig. S5). Interestingly, similar dimers were observed in other members of the CUB domain family (Romero et al., Supra). However, a study of the molecular weights of the three Nrp2 constructs (a2b1b2, a1a2b1b2, and a1a2b1b2c) by polygonal light scattering showed that all three Nrp2 constructs are monomeric in solution (data not shown). These data mean that Nrp homodimerization is very weak in solution, even in the presence of the MAM domain, and receptor dimerization can occur only upon ligand binding. The orientation of the Nrp2 crystallographic dimer suggests a VEGF binding model. The a1 interface creates a saddle-shaped dimer approximately 70 Å wide (FIG. 6), large enough to accommodate a VEGF ₁₀₉ dimer having an approximate size of 35 × 60 Å. It is important that tuftsin binding sites and heparin binding sites are located on the inner surface of the saddle. Heparin can stabilize the interactions between the heparin binding sites of Nrp and VEGF and further facilitate the association of the VEGF tail with b1 hairpins. Such a structure is also capable of adapting VEGFR binding through the VEGF receptor binding domain to send signals down. Therefore, although the a1 domains do not directly interact with VEGF, they can enhance the binding of VEGF to Nrp, since they facilitate the dimerization of Nrp. This dimer may further provide Sema3 binding (FIG. 6). In the crystal structure of Sema3A (Antipenko, A. et al., Neuron 39, 589-98 (2003)), two “sema” domains are densely packed together at the interface. Upon binding to Nrp, the Sema domains of semaphorin 3A dissociate, which makes it possible to interact with the primary binding site a1a2 (Antipenko, et al., Supra). In the a1 a-mediated Nrp dimer model presented here, the two putative Sema3A binding sites are located on opposite sides and are sufficiently far apart to accommodate the large β-propellers of the two bound Sema3 molecules (FIGS. 6 and S5).

The structural analysis presented here provides a first detailed picture of the two parts of the ECD Nrp designed to bind VEGF (b1b2) and semaphorine (a1a2). Antibody complexes (FIGS. 3-5) along with previous mutagenesis studies (Gu et al., Supra, Vander Kooi et al., Supra) outline the Nrp binding sites for the Sema domain of semaphorins and the VEGF domain for heparin binding. These structures form the basis for further mutagenesis experiments and outline strategies for elucidating Nrp structures in combination with their ligands and signal receptors VEGF, VEGFR, semaphorins, and plexins. In addition, the proposed crystal structures and other information contained in the present invention provide opportunities for constructing VEGF and / or semaphorin antagonists and agonists, and can also be used in screening assays to identify such antagonists or agonists.

Although the invention has been illustrated in the foregoing description with respect to certain embodiments, this description is not limiting. Indeed, a competent specialist, based on the foregoing description, in addition to the embodiments of the invention that are shown and described in this application, various modifications of this invention will be apparent that relate to its scope and correspond to the attached claims.

All bibliographic sources cited in the description of this invention are clearly incorporated herein by reference in their entirety.

Claims (15)

1. Antibody against panNrp ^A , containing:
(a) the sequence of the variable domain of the light chain of SEQ ID NO: 4 or SEQ ID NO: 5 and the sequence of the variable domain of the heavy chain of SEQ ID NO: 7, or
(b) the sequence CDRL1 SEQ ID NO: 28, the sequence CDRL2 SEQ ID NO: 30 and the sequence CDRL3 SEQ ID NO: 32 or 33 and the sequence CDRH1 SEQ ID NO: 35, the sequence CDRH2 SEQ ID NO: 37 and the sequence CDRH3 SEQ ID NO : 39; or
its antigen binding fragment. 2. The anti-panNrp ^A antibody of claim 1, which is an YW68.11 antibody containing the sequence shown in Figure 9A (SEQ ID NO: 9), or an antigen binding fragment thereof. 3. The anti-panNrp ^A antibody of claim 1, which is an YW68.11.26 antibody comprising the sequence shown in Figure 9B (SEQ ID NO: 10), or an antigen-binding fragment thereof. 4. The anti-panNrp ^A antibody of claim 1, which binds to both Nrp1 and Nrp2. 5. The anti-panNrp ^A antibody of claim 4, having a binding affinity for both Nrp1 and Nrp2 of at least about 0.10 nM. 6. The anti-panNrp ^A antibody of claim 4, having a binding affinity for both Nrp1 and Nrp2 of at least about 0.15 nM. 7. The anti-panNrp ^A antibody of claim 4, having a binding affinity for both Nrp1 and Nrp2 of at least about 0.20 nM. 8. The anti-panNrp ^A antibody of claim 4, having a binding affinity for both Nrp1 and Nrp2 of at least about 0.25 nM. 9. The anti-panNrp ^A antibody of claim 4, having a binding affinity for both Nrp1 and Nrp2 of at least about 0.30 nM. 10. The anti-panNrp ^A antibody according to claim 4, blocking the binding of Sema3 to both Nrp1 and Nrp2. 11. The anti-panNrp ^A antibody of claim 4, which does not block the binding of VEGF to Nrp1 or Nrp2. 12. The anti-panNrp ^A antibody of claim 10 or 11, inhibiting the biological activity of semaphorin in vitro. 13. The anti-panNrp ^A antibody of claim 10 or 11, inhibiting the biological activity of semaphorin in vivo. 14. Antibody against panNrp ^A according to claims 1, 4, 5 or 6, comprising additional binding specificity. 15. The anti-panNrp ^A antibody of claim 14, wherein said additional binding specificity is VEGF binding ability.

Images