HK1188230A - Human oncostatin m antibodies and methods of use - Google Patents

Description

Human oncostatin M antibodies and methods of use

Technical Field

The present invention relates to human antibodies and uses capable of neutralizing the biological activity generated by oncostatin M binding to membrane receptors on human cells.

Background

Oncostatin M (OSM) is a 28kDa multifunctional member of the IL-6 family of cytokines secreted by monocytes, macrophages, neutrophils and activated T-lymphocytes (Tanaka & Miyajima, Rev Physiol Biochem Pharmacol (review of physiology, biochemistry and pharmacology) 149: 39-53, 2003). Proteolytic cleavage near the carboxy terminus of secreted OSM produces the fully active form of OSM, which is 209 amino acids in length, with two N-linked glycosylation sites. OSM belongs to the IL-6 family of cytokines, including (IL-6, IL-11, Leukemia Inhibitory Factor (LIF), cardiotrophin-1, ciliary neurotrophic factor (CNTF) and cardiotrophin-like cytokine (CLC)), which share a common receptor subunit gp130 protein. In humans, OSM signals through a receptor heterodimer consisting of gp130 and the LIFR α subunit or gp130 and the OSMR β subunit. OSM binds directly to gp130 in the absence of any additional membrane-bound co-receptor, as compared to other cytokines of the other IL-6 family (Gearing et al, Science 255: 1434-1437, 1992). After binding of OSM to gp130, OSMR β or LIFR α is recruited to form a high affinity signaling complex (Mosley et al, J Biol Chem (J. Biol. Chem.) 271: 32635-32643, 1996). Activation of either receptor results in signaling through the JAK/STAT pathway (August et al, Jbiol Chem (J. Biochem.) 272: 15760-15764, 1997).

OSM is produced primarily by cells of immune system origin and, due to its widespread signaling receptors, has been associated with a variety of biological activities, including regulation of cell growth, neural development, and regulation of extracellular matrix composition.

As its name implies, oncostatin M is associated with a carcinogenic process. However, OSM is also involved in early events in inflammation and the hypertrophic pathway leading to deleterious conditions such as pulmonary fibrosis. Thus, there is a need to provide human antibodies specific for human OSM that are capable of blocking receptor signaling (gp130 signaling) events, which signal blocking antibodies can exert clinically useful cytotoxic, cytostatic, or immunomodulatory effects on gp130 expressing cells.

Disclosure of Invention

The present invention provides OSM binding monoclonal antibodies capable of blocking activities associated with one or more biological activities associated with OSM-to-OSM binding receptor interaction on a cell, tissue or organ of a host subject. Amino acid sequences of exemplary OSM binding monoclonal antibodies, which may be encoded by nucleic acids for expression in a host cell, are provided. One or more OSM monoclonal antibodies of the invention define epitopes on the surface of OSM that, when tightly bound by an antibody of the invention, are prevented from interacting with the receptor components of the signaling complex, i.e., gp130 and lifra or gp130 and OSMR β, thereby preventing ligand-ligation driven signaling and downstream biological activity.

One aspect of the present invention is an isolated antibody reactive with human OSM protein, said antibody having the antigen binding ability of a monoclonal antibody comprising a heavy chain variable region having the amino acid sequence set forth in SEQ id no: 13-18 or in the amino acid sequence as set forth in SEQ ID NO: 1-3, an antigen binding domain at the designated position of FR1-CDR1-FR2-CDR2-FR3 as set forth in SEQ ID NO: 27-29 and 47, CDR 3; or comprises a polypeptide having the sequence shown in SEQ ID NO: 23-26 or in the amino acid sequence as set forth in SEQ ID NO: 5-8 and variants thereof and antigen binding domains at the indicated positions as set forth in SEQ ID NOs: 19-22, CDR 3. In a specific embodiment, the human OSM-binding antibody comprises a heavy chain variable region selected from the group consisting of SEQ ID NOs: 49-55.

In another embodiment of the invention, the monoclonal antibody binding domain used as a full-length IgG structure has a constant domain derived from a human IgG constant domain or a specific variant thereof and is used as a therapeutic molecule in a pharmaceutical formulation to prevent OSM binding to cells displaying OSM receptor components. In another embodiment, the binding domain is configured as an antibody fragment for use as a therapeutic molecule capable of preventing OSM binding to a cell displaying an OSM receptor component. In one aspect of the invention, there is provided a pharmaceutically acceptable formulation, delivery system, or kit or method for treating an oncostatin M-associated disorder comprising one or more OSM binding domains of the invention, such as, but not limited to, one or more OSM binding domains as set forth in SEQ ID NO: 29 and 47, 13-28 and 30-46 and variants.

Drawings

Figure 1 shows germline gene sequences for the construction of Fab libraries displayed on pIX coat protein, where each HV domain is represented by a sequence according to SEQ ID NO: 1-3(1 ═ IGHV1-69, 2 ═ IGHV3-23, and 3 ═ IGHV5-51) of differential FR1-CDR1-FR2-CDR2-FR3 followed by variable length differential H-CDR3 region and J region (FR4, SEQ ID NO: 4); and each LV domain consists of a sequence according to SEQ ID NO: the differentiated FR1-CDR1-FR2-CDR2-FR3 of 5-8(5 ═ IGKV1-39(O12), 6 ═ IGKV3-11(L6), 7-IGKV3-20(a27), and 8 ═ IGKV4-1(B3)) was then according to SEQ ID NO: CDR3 of 9 was composed next to the J region (FR4, SEQ ID NO: 10).

Fig. 2A is a dose-response plot of CHO cell-derived recombinant human and cynomolgus monkey oncostatin M inhibition of a375-S2 cell proliferation, measured by BrdU incorporation and normalized to a control in the presence of vehicle only.

FIG. 2B is a graph showing the ability of antibody M71, consisting of the variable domains L180(SEQ ID NO: 53) and H17(SEQ ID NO: 54), to mitigate the inhibition of proliferation of A375-S2 by OSM, wherein OSM is present at a concentration of 2 ng/ml.

FIG. 3 shows that M64, M71, M55 and M69 at 20. mu.g/ml will³⁵SO₄Bar graph of the effect of increased absorption (measure of increased proteoglycan synthesis) to levels above that observed in the absence of antibody, where the non-specific isotype antibody is a control in a co-culture of human chondrocytes and human macrophages capable of secreting OSM in alginate beads.

Fig. 4A is a graph showing dose response of human OSM-stimulated pSTAT3 in NHLF cells, where EC50 was found to be about 1 ng/ml.

Figure 4B is a graph showing the ability of antibody M71 to neutralize pSTAT3 signaling in the presence of 2ng/ml OSM.

FIGS. 5A and 5B are scatter plots showing the amount of IP-10(A) and MCP-1(B) detected in the serum of individual mice after stimulation with OSM and with or without pretreatment with M71 antibody at the indicated concentration.

Fig. 6A and 6B are graphs showing serum concentrations over time in cynomolgus monkeys after intravenous (a) or subcutaneous (B) administration of 3mg/Kg of the Fc variant of M71 or M71.

Detailed Description

All publications (including but not limited to patents and patent applications) cited in this specification are herein incorporated by reference as if fully set forth herein.

Abbreviations

BSA ═ bovine serum albumin; CDR is a complementarity determining region; cyno (cynomolgus monkey (Macaca fascicularis)); DN ═ diabetic nephropathy; ECD ═ extracellular domain; FR ═ framework; h ═ heavy chain; IPF ═ interstitial pulmonary arthritis; l ═ light chain; ig ═ immunoglobulins; mab ═ monoclonal antibody; OSM ═ oncostatin M; OA is osteoarthritis; PBS ═ phosphate buffered saline; RA ═ rheumatoid arthritis; VL ═ variable light chain; VH is a variable heavy chain.

Definition of

As used herein, "antibody" includes whole antibodies and any antigen-binding fragment or single chain thereof. Thus, an antibody includes any protein or peptide comprising a molecule that comprises at least a portion of an immunoglobulin molecule, such as, but not limited to, at least one Complementarity Determining Region (CDR) of a heavy or light chain or a ligand binding portion thereof, a heavy or light chain variable region, a heavy or light chain constant region, a Framework Region (FR), or any portion thereof, or at least a portion of a binding protein that may be incorporated into an antibody of the invention. The term "antibody" is also intended to encompass antibodies, digested fragments, specified portions, or variants thereof, including antibody mimetics or antibody portions that comprise structures and/or functions that mimic antibodies or specified fragments or portions thereof, including single chain antibodies and single domain antibodies and fragments thereof. Functional fragments include antigen-binding fragments directed against a preselected target. Examples of binding fragments encompassed within the term "antigen-binding portion" of an antibody include (i) Fab fragments, i.e., monovalent fragments consisting of VL, VH, CL and CH domains; (ii) a F (ab') 2 fragment, i.e., a bivalent fragment comprising two Fab fragments linked by a hinge region disulfide bond; (iii) an Fd fragment consisting of VH and CH domains; (iv) (iv) Fv fragments consisting of the VL and VH domains of a single arm of an antibody, (v) dAb fragments consisting of the VH domains (Ward et al, (1989) Nature 341: 544-546); and (vi) an isolated Complementarity Determining Region (CDR). Furthermore, although the two domains of the Fv fragment, VL and VH, are encoded by separate genes, they can be made into a single protein chain using recombinant methods by ligating them through synthetic linker peptides that allow them to be made into a single protein chain in which the VL and VH regions pair to form a monovalent molecule (known as single chain Fv (scFv); see, e.g., Bird et al (1988) Science 242: 423 + 426 and Huston et al (1988) Proc. Natl. Acad. Sci. USA 85: 5879 + 5883). Such single chain antibodies are also intended to be encompassed within the term "antigen-binding portion" of an antibody. These antibody fragments are obtained using conventional techniques known to those skilled in the art, and fragments are screened for utility in the same manner as intact antibodies. Conversely, scFv construction libraries can be used to screen for antigen binding and then spliced together with other DNA encoding human germline gene sequences using conventional techniques. An example of such a library is "HuCAL: human Combinatorial antibody library "(Knappik, A. et al, J Mol Biol (journal of molecular biology) (2000)296 (1): 57-86).

The term "CDR" refers to the complementarity determining region or hypervariable region amino acid residues of an antibody which are involved in or responsible for antigen binding. The hypervariable regions or CDRs of the human IgG antibody subclasses comprise amino acid residues from residues 24-34(L-CDR1), 50-56(L-CDR2) and 89-97(L-CDR3) in the light chain variable domain and residues 31-35(H-CDR1), 50-65(H-CDR2) and 95-102(H-CDR3) in the heavy chain variable domain (e.g., Kabat et al (1991Sequences of proteins of Immunological Interest), 5 th edition, 1991sequence of proteins of Immunological Interest)Published Health Service, National Institutes of Health, Bethesda, Md.) and/or those residues from the hypervariable loops in the light chain variable domain (i.e., residues 26-32(L1), 50-52(L2), and 91-96(L3)) and 26-32(H1), 53-55(H2) in the heavy chain variable domain or 52-57 and 96-101(H3) as currently defined by H2Chothia (e.g., (Chothia and Lesk, j.mol.biol.) (journal of molecular biology) 196: 901-917 (1987)). Chothia and Lesk refer to structurally conserved HV as "canonical structures". Framework or FR1-4 residues are those variable domain residues other than and excluding the hypervariable region. The numbering system of Chothia and Lesk takes into account the numbering differences of residues in the loops by showing extensions at the designated residues in lower case notation, e.g., 30a, 30b, 30c, etc. Recently, a universal numbering system, i.e., the International ImmunoGeneTiCs information System (International ImmunoGeneTiCs information) has been developed and widely adopted) (IMGT) (LaFranc et al, 2005.Nucl Acids Res. ("nucleic Acids research") 33: D593-D597).

Herein, CDRs are represented by sequence numbering in amino acid sequence and position in the heavy or light chain. Since the "position" of a CDR in the structure of an immunoglobulin variable domain is conserved across species and is present in a structure called a loop, the CDR and framework residues are readily identified by using a numbering system that aligns the variable domain sequences according to structural features. This information is used to graft and replace CDR residues from an immunoglobulin of one species into the acceptor framework from a typically human antibody.

The term "maturation" applies to directed changes made in the variable regions of antibodies for the purpose of altering the performance of the polypeptide. As known in the art and described herein, many positions in the V region sequence can affect antigen recognition. In fact, antibodies achieve high affinity and specificity through a progressive process of somatic mutation. This process can be simulated in vitro to allow parallel selection and targeting of variations while maintaining the sequence integrity of each antibody chain so that they reflect species (in this case human) antibodies while enhancing affinity or biophysical parameters such as solubility or antioxidant properties. The process of making directed changes or "maturation" is typically done at the coding sequence level and can be achieved by making sub-libraries for selecting enhanced performance.

As used herein, "OSM" refers to an oncostatin M polypeptide or a polynucleotide comprising a coding sequence that encodes an OSM polypeptide. Human OSM is the product of the human OSM gene (gene 5008).

The term "epitope" refers to a protein determinant capable of specific binding to an antibody. Epitopes usually consist of chemically active surface groups of molecules such as amino acids or sugar side chains and usually have specific three-dimensional structural characteristics as well as specific charge characteristics. Conformational and non-conformational epitopes are distinguished in that binding to the former, but not the latter, is lost in the presence of denaturing solvents.

As used herein, K_DRefers to the dissociation constant, specifically, the K of the antibody to the predetermined antigen_DAnd is a measure of the affinity of the antibody for a particular target. K of high affinity antibodies to predetermined antigens_DIs 10^-8M or less, more preferably 10^-9M or less, and more preferably 10^-10M or less. K_DIs reciprocal of K_AAnd is the association constant. As used herein, the term "k_dis"or" k₂"or" k_dBy "is meant the off-rate of a particular antibody-antigen interaction. ' K_D"is the dissociation rate (k)₂Also known as the "off-rate" (k)_off) ") and association rate (k)₁) Or "on-rate" (k)_on) "is used in the following description. Thus, K_DIs equal to k₂/k₁Or k_off/k_onAnd is expressed in molar concentration (M). It follows the principle that K_DThe smaller the binding, the stronger. Thus, and 10^-9M (or 1nM) comparison, 10^-6K of M (or 1. mu. mol)_DIndicating a weak binding.

As used herein, the term "monoclonal antibody" or "monoclonal antibody composition" refers to a preparation of antibody molecules of a single molecular composition. Monoclonal antibody compositions exhibit a single binding specificity and affinity for a particular epitope. The term also includes "recombinant antibodies" and "recombinant monoclonal antibodies" because all antibodies are made, expressed, produced or isolated by recombinant means, e.g., (a) antibodies isolated from an animal or hybridoma, which are made by antibody secreting animal cells and fusion partners; (b) antibodies isolated from host cells transformed to express the antibodies, e.g., from transfectomas; (c) antibodies isolated from recombinant, combinatorial human or other species antibody libraries and (d) antibodies prepared, expressed, produced or isolated by any other method that involves splicing immunoglobulin gene sequences to other DNA sequences. As used herein, "isolated antibody" means an antibody that: substantially free of other antibodies having different antigen specificities. However, an isolated antibody that specifically binds to a human OSM epitope, isoform or variant may be cross-reactive to other related antigens, such as antigens from other species (e.g., OSM species homologs). Furthermore, the isolated antibody may be substantially free of other cellular material and/or chemicals. In one embodiment of the invention, compositions of "isolated" monoclonal antibodies with different specificities are mixed in a well-defined composition.

As used herein, "specifically binds," "immunospecific binding," and "immunospecifically binds" refer to the binding of an antibody to a predetermined antigen. Typically, the antibody is present in 10^-7Dissociation constant (K) of M or less_D) K binding to and binding to a predetermined antigen_DK for its binding to a non-specific antigen other than the predetermined antigen (e.g. BSA, casein or any other specific polypeptide)_DAt least 1/2. The phrases "antibody recognizing an antigen" and "antibody specific for an antigen" are used interchangeably herein with the term "antibody that specifically binds to an antigen". As used herein, "highly specific" binding refers to the relative K of an antibody to a particular target epitope_DAt least less than the K of the antibody bound to other ligands_D1/10 of (1).

As used herein, "class" refers to the class of antibodies (e.g., IgA, IgE, IgM, or IgG) encoded by the heavy chain constant region gene. Some antibody classes also encompass subclasses or "isotypes" that are also encoded by heavy chain constant regions (e.g., IgG1, IgG2, IgG3, and IgG 4). The antibody may be further modified by oligosaccharides attached to the protein at specific residues in the constant region that further enhance the biological function of the antibody. For example, among human antibody isotypes, IgG1, IgG3, and the lesser existing lgG2 exhibit the same effector functions as the murine lgG2a antibody.

By "effector" function or "effector positive" is meant that the antibody comprises a domain that is distinguishable from the antigen-specific binding domain, which is capable of interacting with a receptor or other blood component, such as complement, resulting in, for example, macrophage recruitment, and causing the destruction of cells bound by the antigen-binding domain of the antibody. Antibodies have several effector functions mediated by the binding of effector molecules. For example, binding of the C1 component of complement to antibodies activates the complement system. Activation of complement is important for the opsonization and lysis of cellular pathogens. Activation of complement stimulates an inflammatory response and may also involve autoimmune hypersensitivity. In addition, the antibody binds to the cell via the Fc region, i.e., the Fc receptor site on the Fc region of the antibody binds to an Fc receptor (FcR) on the cell. There are many classes of Fc receptors that are specific for different classes of antibodies, including IgG (gamma receptor), IgE (eta receptor), IgA (alpha receptor), and IgM (mu receptor). Binding of antibodies to Fc receptors on cell surfaces triggers many important and diverse biological responses, including phagocytosis and destruction of antibody-coated particles, clearance of immune complexes, lysis of antibody-coated target cells by killer cells (known as antibody-dependent cell-mediated cytotoxicity, or ADCC), release of inflammatory mediators, placental transfer, and modulation of immunoglobulin production.

The term "polypeptide" refers to a molecule comprising amino acid residues joined by peptide bonds to form a polypeptide. Small polypeptides of less than 50 amino acids may be referred to as "peptides". Polypeptides may also be referred to as "proteins".

Overview

The present invention provides isolated mabs capable of binding and neutralizing the biological activity of OSM proteins and cleavage products. Specifically, OSM binding to a mAb of the invention can block OSM binding to gp130 or prevent gp130 binding of OSM recruiting LIFRa or OSMRb. In either case, the OSM mabs of the invention are capable of blocking OSM-driven gp130 receptor signaling.

Phosphorylation of STAT3 has been reported to lead to excessive collagen production by fibroblasts in a variety of pathological settings (Lim et al, Oncogene (23) (39): 5416-25, 2006; Huang et al, J Cell Biochem (J. Cell. Biochem.) 81 (1): 102-13, 2001). These properties demonstrate the potential therapeutic value of these antibodies for RA, OA and for fibrotic indications such as Idiopathic Pulmonary Fibrosis (IPF) and Diabetic Nephropathy (DN).

The human OSM gene product OSM (NCBI accession NP-065391) is a pre-pro-polypeptide of 252 amino acids in length (SEQ ID NO: 11) with a signal peptide of 25 amino acids in length and a proteolytic cleavage site between residues 234 and 235. It is a secreted protein with five cysteine residues forming two internal disulfide bonds between residues 31 to 152 and 74 to 192 (Kallestad JC et al, J Biol Chem. ("J. Biol. Chem.) (5.15.1991; 266 (14): 8940-5). There are two potential N-linked glycosylation sites at residues 100 and 217, and when produced in eukaryotic cells, proteins are glycosylated. Human OSM has a free thiol group at residue 105.

The sequence of cyno OSM protein is not available in the public domain, but there are recordings generated by automated calculation of 1867bp mRNA (NCBI No. xm _001110148) derived from annotated gene sequences (NW _ 001095169). To obtain cyno OSM sequences, RNA was isolated from cyno PBMC, and then genes were amplified from the cDNA by RT-PCR and sequenced. As disclosed in applicants' co-pending application (U.S. Ser. No. 12/648430), the predicted translation of the cloned sequence (SEQ ID NO: 12) was found to be 99.6% identical to the predicted Rhesus monkey (Macaca mulatta) (Rhesus) sequence, 92% identical to the human OSM protein sequence, and 41% identical to the mouse OSM protein sequence.

Thus, the present invention is directed to the identification of human OSM-binding mabs capable of inhibiting downstream biological activity caused by OSM-bound gp130 signaling, and wherein the mabs exhibit the following capabilities:

restoring cell proliferation in the presence of OSM, inhibiting chondrocyte degeneration of OSM-driven inter-articular (joint) matrix in tissue explants,

Effectively neutralize OSM-dependent STAT3 phosphorylation in human lung fibroblasts and prevent OSM-induced cytokine release.

1. Antibody compositions of the invention

OSM neutralizing antibodies of the invention are antibodies that inhibit, block or interfere with at least one OSM activity or OSM receptor binding in vitro, in situ and/or in vivo and do not promote, stimulate, induce or compete OSM activity or ligand binding, and antibody binding does not mimic downstream effects of OSM driven OSM receptor ligation (specifically gp130 interacting with OSM), such as signal transduction in a host cell. Suitable OSM neutralizing antibodies, specified portions or variants may also optionally affect at least one OSM activity or function, such as, but not limited to: RNA, DNA or protein synthesis; protein release; cell activation, proliferation or differentiation; antibody secretion; OSM receptor signaling; OSM cracking; OSM binding, OSM or gp130 induction, synthesis or secretion.

The present invention is based on the discovery of anti-human OSM monoclonal antibodies capable of inhibiting gp130 signaling following OSM binding or recruitment of LIFR by OSM. The antibody binding domains in the form of Fab libraries displayed on filamentous phage particles linked to pIX coat protein (see WO29085462a 1and described further below) were selected for their ability to bind OSM. Competition assays using gp130 were used to distinguish those fabs that, when bound to OSM, prevent OSM from binding to gp 130. Alternatively, Fab, when bound to OSM, prevents recruitment of LIFR to gp130 binding. Cell-based (a375, human melanoma cells) assays were used to identify a number of candidate antibodies capable of inhibiting gp 130-mediated activation of pSTAT3 in host cells expressing OSM.

OSM binding antibodies described herein recognize at least two different regions on the active form of human OSM protein, indicating that multiple compounds on OSM were additionally found that are suitable for targeting antibodies or other compounds with similar blocking function. Thus, expression and purification of the antibody binding domains provided herein as amino acid sequences also provides tools that provide methods for selecting novel molecules that exhibit OSM neutralizing activity.

In one embodiment, an anti-human OSM antibody has a binding region comprising a polypeptide having the sequence set forth in SEQ ID NO: 49-55, and the antibody or binding portion thereof immunospecifically binds to OSM. In another embodiment of the invention, a polypeptide having the sequence of SEQ ID NO: 54 or 55, and the antigen binding portion thereof binds to OSM protein and additionally has the specific functional properties of the antibody of the invention, for example:

1. at a K of less than 100pM_DBinding to human OSM

2. At a K of less than 500pM_DBinding to cyno OSM

3. Can restore the proliferation of A375-S2 cells to 90% level in the absence of human OSM in the presence of 2ng/ml human OSM,

4. can restore the proliferation of A375-S2 cells to 90% level in the absence of OSM in the presence of 2ng/ml cyno OSM,

5. inhibit OSM-driven chondrocyte degeneration of the matrix between joints (joints) in tissue explants,

6. effectively neutralize OSM-dependent STAT3 phosphorylation in Normal Human Lung Fibroblasts (NHLF), or

7. Cytokine release following systemic stimulation with OSM in mice was blocked.

In another aspect of the invention, the structural features of antibodies exhibiting some or all of the above-mentioned biological activities as described herein, in particular mabs designated M55 and M71 binding domains, are used to prepare structurally related human anti-OSM antibodies that retain at least one functional property of the antibodies of the invention, e.g., binding to OSM. More specifically, one or more CDR regions of M55 and M71 (e.g., particular residues of SEQ ID NOs: 1and 8) can be aligned with known human framework regions and CDRs such as those of SEQ ID NOs: 13-28, 30-46, to produce additional recombinantly engineered human anti-OSM antibodies of the invention.

In one embodiment, the antibody of the invention has a sequence comprising FR1, 2 and/or 3 of IGVH1-69(SEQ ID NO: 1) or IGVH5-51(SEQ ID NO: 3), wherein the amino acid sequence is derived from a sequence selected from SEQ ID NO: 13-22 is present in SEQ ID NO: 1 or3 while still retaining the ability of the antibody to bind OSM (e.g., conservative substitutions). Thus, in another embodiment, the engineered antibody may consist of one or more CDRs that, for example, differ from SEQ ID NO: 13-22 or as set forth in SEQ ID NO: 29 or 47 give variants of L-CDR3 that are 90%, 95%, 98%, or 99.5% identical.

In addition to binding only OSM, engineered antibodies such as those described above may also be selected for their retention of other functional properties of the antibodies of the invention, e.g., the ability to inhibit OSM protein or its cleavage products from binding to GP130 positive cells, which binding would result in inhibition of GP130 positive cell proliferation in vivo.

The human monoclonal antibodies of the invention can be tested for binding to OSM by, for example, standard ELISA.

Generation of OSM neutralizing antibodies

OSM neutralizing antibodies exhibiting a desired spectrum of biological activity can be produced by a variety of techniques, as exemplified herein by M5, M6, M9, M10, M42, M45, M53, M54, M55, M62, M63, M65, M66, M67, M68, M69, M71, and M83, examples of which include antibodies having the library framework SEQ ID NO: 1. 3, 8 and having the specified heavy and light chain sequences of SEQ ID NO: 13-28, 30-46.

In another embodiment, the epitope bound by the antibody of the invention comprises SEQ ID NO: 11, or a nucleic acid coding sequence derived therefrom, can be used to immunize a subject to directly produce an antibody of the invention in a host for the purpose of treating, preventing or ameliorating a disease or disease symptoms associated with OSM production.

In one embodiment and as exemplified herein, the human antibody is selected from a phage library, wherein the phage comprises human immunoglobulin genes, and the library expresses the human antibody binding domain as, for example, a single chain antibody (scFv), as Fab, or some other construct that exhibits paired or unpaired antibody variable regions (Vaughan et al, Nature Biotechnology (Nature Biotechnology) 14: 309-314 (1996); Sheets et al, PITAS (USA) 95: 6157-6162 (1998); Hoogenboom and Winter, J.mol.biol. ("J.Mol.Biol.) (227: 381 (1991); Marks et al, J.mol.biol. (" J.Mol.Biol.) ("molecular biology J.581 (1991)). The human monoclonal antibodies of the invention can also be prepared by phage display methods for screening libraries of human immunoglobulin genes. Such phage display methods for isolating human antibodies are well established in the art. See, e.g., U.S. Pat. Nos. 5,223,409 to Ladner et al; no.5,403,484; and No.5,571,698; U.S. Pat. Nos. 5,427,908 and 5,580,717 to Dower et al; U.S. patent nos. 5,969,108 and 6,172,197 to McCafferty et al; and U.S. patent No.5,885,793 to Griffiths et al; no.6,521,404; no.6,544,731; no.6,555,313; no.6,582,915 and No.6,593,081.

Phage clones were selected and identified by a multi-step procedure called biopanning (biopanning). Biopanning is performed by: phages displaying protein ligand variants (phage display libraries) are incubated with the target, unbound phages are removed by washing techniques, and then bound phages are specifically eluted. The eluted phage are optionally amplified and then passed through additional cycles of binding and optional amplification that enriches the pool of specific sequences that favor those phage clones that carry antibody fragments that display optimal binding to the target. After several rounds, each phage clone was characterized and the sequence of the clonally displayed peptide was determined by determining the sequence of the corresponding DNA of the phage virion.

Fab phage-pIX libraries

In a specific example of phage display technology, Shi et al, J Mol Biol (journal of molecular biology) 397: 385-396, 2010; the synthetic Fab library displayed on pIX phage coat protein, described in WO29085462a 1and U.S. serial No. 12/546850, and further detailed herein, was used to select linkers from all the components of human IgG sequences derived from human germline genes. Libraries were constructed of four VL and three VH domains encoded by known IGV and IGJ germline sequences selected based on the frequency with which these sequences have been observed to be present in human antibodies isolated from natural sources. Selected VH, IMGT were designated IGHV1-69(SEQ ID NO: 1), IGHV3-23(SEQ ID NO: 2) or IGHV5-51(SEQ ID NO: 3). The diversity of VH design results in heavy chains with variable length sequences in the CDR3 regions, with positions of limited diversity in H-CDR 1and H-CDR2 that are still of constant length. Framework 4(H-FR4) remained unchanged among all members of the library (SEQ ID NO: 4).

VH169, IGHV1-69, length 98, CDR1 31-35, CDR2 50-66

QVQLVQSGAE VKKPGSSVKV SCKASGGTFS SYX ₁ ISWVRQAPGQGLEWMGX ₂ IX ₃ X ₄ X ₅ X ₆ GTANY AQKFQGRVTI TADESTSTAYMELSSLRSED TAVYYCAR(SEQ ID NO：1)

Wherein, in the 169 library, X₁Is A or G, X₂Is G or W, X₃Can be I or S, X₄Can be P or A, X₅Can be I or Y and X₆And may be F or N.

VH323, IGHV3-23 × 01 98, CDR1 31-35, CDR2 50-66

EVQLLESGGG LVQPGGSLRL SCAASGFTFS X ₁ YX ₂ MX ₃WVRQAPGKGLEWVSX₄ IX ₅ X ₆ X ₇ GX ₈ STYY ADSVKGRFTI SRDNSKNTLYLQMNSLRAED TAVYYCAK(SEQ ID NO：2)

Wherein, in the 323 library, X₁May be S, D, N or T; x₂May be A, G or W; x₃Can be S or H; x₄May be V, A, N or G; x₅May be S, N, K or W; x₆May be Y, S, G or Q; x₇Can be S or D; and X₈Can be S or G.

VH551, IGHV5-51, 03 length 98, CDR1 31-35, CDR2 50-66

EVQLVQSGAE VKKPGESLKI SCKGSGYSFT X ₁ YWIX ₂WVRQMPGKGLEWMGX ₃ IX ₄ PX ₅ DSX ₆ TRY SPSFQGQVTI SADKSISTAYLQWSSLKASD TAMYYCAR(SEQ ID NO：3)

Wherein, in the 551 library, X₁May be S, N or T; x₂Can be S or G; x₃Can be I or R; x₄Can be D or Y; x₅Can be G or S; x₆Can be D or Y.

The FR4 or JH region having (11 residues) WGQGTLVTVSS (SEQ ID NO: 4) has been linked to the above sequence to form a complete heavy chain variable region.

The H-CDR 1and H-CDR2 positions targeted for diversification were determined by: 1) the diversity of germline genes; and 2) the frequency of contact with antigen found in antibody-antigen complexes of known structure (Almagro, J Mol Recognit. ("journal of molecular recognition) 17: 132-143, 2004). The amino acid diversity at the selected position is determined by: 1) use in a germline; 2) the amino acids most frequently observed in human rearranged V genes; 3) predicting an amino acid to be mutated from a monobasic matrix cell; and 4) the biochemical and biophysical properties of the amino acids that contribute to antigen recognition.

The library integrates the diversity of CDRs 3 of VH (H3), mimicking all the components of human antibodies (Shi et al, 2010, supra), as shown below (formula I), with a final length between 7 and 14 residues. In CDR3, which is more than 5000 human variable regions, the amino acids glycine (G) and alanine (a) are often used in all positions. In addition, aspartic acid (D) is often used in position 95, and tyrosine (Y) is often encoded at a position prior to the canonical region of the J segment. The amino acids phenylalanine (F), aspartic acid (D) and tyrosine (Y) predominate at positions 99-101, which are used at these positions in IgG. Since these positions generally serve as structural supports for the H-CDR3 and are less accessible to the surface of antigen and/or IgG, the amino acids phenylalanine highlight the amino acid 99 (ratio 50/50), aspartic acid 100 and tyrosine 101 are fixed. Thus, the sequence of formula I is inserted in SEQ ID NO: 1. 2 or3 and SEQ ID NO: 4 form a complete VH.

-(D)-(N)n(N+O)m(F)DY-(I)

Wherein:

(D) positions asp (d) and gly (g) rich.

(N) N-positions rich in ala (a) and gly (g), N-3-7.

(O) m ═ enriched with ala (a), gly (g) and y (tyr), and m ═ 1-4.

(F) Phe (f) dominant position.

The multiform library encompasses pairing with a fixed or diversified light chain that is also derived from the human germline repertoire. In the present invention, four light chainsLibrary VL_κGenes (Kawasaki et al, 2001, Eur J Immunol (J European Immunol) 31: 1017-&Zachau, 1993Biol Chem hopper Seyler (hopper Seyler biochemistry) 374: 1001-1022) a27(IGKV3-20 x 01), B3(IGKV4-1 x 01), L6(IGKV3-11 x 01) and O12(IGKV1-39 x 01), wherein the names of the genes in parentheses are the putative corresponding IMGT genes. Fab is displayed on pIX by expression of a dicistronic vector in which the VH-CH1 domain is fused to the coat protein sequence and VL-CL κ or VL-CL λ is expressed as a free polypeptide self-associated with VH-CH 1. CDR regions are underlined.

Variable library of light chains based on the germline gene of Vkappa (Vk)

012; IGKV1-39 × 01, IGKV1D-39 × 01 length 88; 24-34 in CDR 1and 50-56DIQMTQSPSS LSASVGDRVT ITCRAS in CDR2QSIS X ₁ X ₂ X ₃ NWYQQKPGKAPKLLIYX ₄ ASSLQSGVPS RFSGSGSGTD FTLTISSLQP EDFATYYC(SEQ ID NO：5)

L6, IGKV3-11 × 01 length 88, CDR 1-24, CDR 2-50-56

EIVLTQSPAT LSLSPGERAT LSCRASQSV X ₁ X ₂ X ₃ LAWYQQKPGQAPRLLIYX ₄ ASNRATGIPA RFSGSGSGTD FTLTISSLEP EDFAVYYC(SEQ ID NO：6)

A27, IGKV3-20 × 01, length 89, CDR 1-24-35, CDR 2-51-57

EIVLTQSPGT LSLSPGERAT LSCRASQSVX ₁ X ₂ X ₃ X ₄ LAWYQQKPGQAPRLLIY X ₅ ASSRATGIP DRFSGSGSGT DFTLTISRLE PEDFAVYYC(SEQ ID NO：7)

B3, DPK24, vkiklobeck; IGKV4-1 × 01 length 94, CDR1 length 24-40, CDR2 length 56-62

DIVMTQSPDS LAVSLGERAT INCKSSQSVL X ₁ SSNNX ₂ NX ₃ LAWYQQKPGQPP KLLIYX ₄ ASTR ESGVPDRFSG SGSGTDFTLTISSLQAEDVA VYYC(SEQ ID NO：8)

The diversity at specific positions for each variable region scaffold is summarized in Table 1 below, where the amino acid one letter codes are used and appear alternately at specific positions as shown in FIG. 1.

TABLE 1：

The VL CDR3 in all libraries had seven residues, of which the first two residues were glutamine (Gln, Q) and the residue corresponding to Kabat residue 95 was proline (Pro, P). For the L-CDR3, the sequence conforms to QQX₁X₂X₃X₄PX₅T (SEQ ID NO: 9), wherein the variations are as in the table below, with residue positions according to Kabat.

TABLE 2：

Because of the differences in variable sequence length between genes, the diversity of specific residue positions in hypervariable loops or CDRs can be described as follows using residue numbering as defined in Al-Lazikani B, LeskAM, Chothia C, 1997(Standard protocols for the structural oligonucleotides of immunoglobulins) J Mol Biol (journal of molecular biology) 273: 927-948). In this system, the length change of the hypervariable loops is accommodated by assigning sub-positions a, b, c, etc. to a given residue.

A framework 4(FR4) segment, such as JK4, FGQGTKVEIK (SEQ ID NO: 10), is used to form a fully human light chain variable region.

The affinity of Fab to various protein targets ranging from 0.2 to 20nM has been shown in the initial selection.

Methods for improving the integrated maturation process of binding parameters consisting of shuffling VL or VH diversity or optionally targeted or limited VL modifications are accomplished using vectors and primers designed and used for libraries as described in the cited publications, as taught herein, and combined with methods known in the art.

Alternative sources of OSM-binding immunoglobulin domains

OSM-binding antibodies having the characteristics of the human mabs disclosed herein can be prepared or binding fragments derived from immunoglobulin domains formed by a variety of methods, including Kohler and Milstein (1975) Nature (Nature) 256: 495 standard somatic hybridization technique (hybridoma method). In a hybridoma method, a mouse or other suitable host animal, e.g., a hamster or cynomolgus monkey, is immunized as described herein to elicit lymphocytes that produce or are capable of producing antibodies that will specifically bind to the protein used for immunization. Alternatively, lymphocytes can be immunized in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusing agent, such as polyethylene glycol, to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp 59-103 (Academic Press, 1986)).

OSM neutralizing antibodies can also optionally be generated by immunizing transgenic animals (e.g., mice, rats, hamsters, non-human primates, etc.) that produce a full repertoire of human antibodies, as described herein and/or as known in the art. Cells producing human anti-OSM antibodies can be isolated from these animals and immortalized using suitable methods, such as those described herein. Alternatively, antibody coding sequences can be cloned, introduced into a suitable vector, and used to transfect host cells for expression and isolation of antibodies by methods taught herein and those known in the art.

The use of transgenic mice carrying the human immunoglobulin (Ig) locus in their germline gene structure provides for the isolation of high affinity fully human monoclonal antibodies against a variety of targets, including human autoantigens to which the normal human immune system is tolerant (Lonberg, N.et al, US5569825, US6300129 and 1994, Nature (Nature) 368: 856-9; Green, L. et al, 1994, Nature Genet. ("Nature genetics") 7: 13-21; Green, L. & Jakobovits, 1998, exp. Med. ("Experimental medicine") 188: 483-95; Lonberg, N and Huszaider, D.1995, int.Rev.Immunol. ("Immunol.) (13: 65-93; Chelapahal et al, US 3610; Brugmann, J. Immunol.) (Nature J. 19921; Eur. biol. Immunol., Nature J. Immunol., Australia., USA., 18; Nature J., Nature J. 1996; Nature J., Nature biol., Nature J. Immunol.),851; Eur., Nature J., Nature, 1998, m. et al, 1997, nat. genet. ("nature genetics") 15: 146-156; green, l., 1999, j.immunol.methods (journal of immunological methods) 231: 11-23; yang, x et al, 1999, Cancer Res. ("Cancer research") 59: 1236-1243; maggomann, m. and tausig, M j., curr. opin. biotechnol. ("current review of biotechnology) 8: 455-458, 1997; tomizuka et al, WO 02043478). The endogenous immunoglobulin locus in such mice can be disrupted or deleted to eliminate the ability of the animal to produce antibodies encoded by the endogenous gene. In addition, companies such as Abgenix, Inc (Freemont, Calif.) and Medarex (San Jose, Calif.) may be engaged in providing human antibodies to selected antigens using techniques as described above.

Polypeptides for use as target ligands in panning strategies and as immunogenic antigens can be prepared using any suitable technique, such as recombinant protein preparation techniques. The target ligand or fragment thereof, or antigen in the case of immunization, in the form of a purified protein or protein mixture (including whole cells or cell extracts or tissue extracts) may be formed de novo in an animal from a nucleic acid encoding the antigen or a portion of the antigen.

The isolated nucleic acids of the present invention can be prepared using (a) recombinant methods, (b) synthetic techniques, (c) purification techniques, or a combination thereof, as is well known in the art. DNA encoding the monoclonal antibody is readily isolated and sequenced using methods known in the art, for example, by using oligonucleotide probes that bind specifically to genes encoding the heavy and light chains of the human antibody region. In the case of hybridoma production, such cells may be used as a source of such DNA. Alternatively, the use of display technologies in which the coding sequence and translation product are associated, such as phage or ribosome display libraries, simplifies the selection of linkers and nucleic acids. Following phage selection, the antibody coding regions from the phage can be isolated and used to produce whole antibodies, including human antibodies or any other desired antigen binding fragments, and expressed in any desired host, including mammalian cells, insect cells, plant cells, yeast and bacteria.

Human antibodies

The invention also provides human immunoglobulins (or antibodies) that bind to human OSM. These antibodies may also be characterized as engineered or adapted. The immunoglobulins have one or more variable regions substantially from human germline immunoglobulins and include directed variations of residues known to be involved in antigen recognition, such as residues of CDRs or structurally defined hypervariable loops. One or more constant regions, if present, are also substantially derived from a human immunoglobulin. K exhibited by human antibodies to OSM_DIs at least about 10^-6M (1. mu.M), about 10^-7M(100nM)、10^-9M (1nM), or less. To influence the change in affinity, for example to improve the affinity of human antibodies for OSM or to reduce the K of human antibodies for OSM_DCDR residues or other residues may be substituted.

The source for generating human antibodies that bind OSM is preferably the sequences provided herein, which are variable regions, frameworks and/or CDRs, labeled SEQ ID NO: 13-55, identified as capable of binding to human OSM and cross-reacting with cynomolgus monkey OSM using a full set of human fabs displayed on filamentous phage particles.

If a human variable domain framework assumes the same or similar conformation as the parent variable framework from which the CDRs were generated, then replacing any of the CDRs into any human variable domain framework is likely to result in maintaining their correct spatial orientation. The heavy and light chain variable framework regions that will be paired in the final Mab may be derived from the same or different human antibody sequences. The human antibody sequences may be naturally occurring human antibody sequences, derived from human germline immunoglobulin sequences, or may be a consensus sequence of several human antibodies and/or germline sequences.

Suitable human antibody sequences can be identified by computer comparison of the amino acid sequences of mouse variable regions to the sequences of known human antibodies. Comparisons of heavy and light chains were made separately, but the principle of each comparison was the same.

In terms of experimental methods, it has been found particularly convenient to construct variable sequence libraries, which can be screened for a desired activity, binding affinity or specificity. One form for constructing such variant libraries is a phage display vector. Alternatively, variants can be generated using alternative known methods for randomizing or diversifying nucleic acid sequences encoding target residues in the variable domains.

Another method of determining whether further substitutions are required and selecting amino acid residues for substitution can be accomplished using computer modeling. Computer hardware and software for generating three-dimensional images of immunoglobulin molecules are widely available. Typically, generation of molecular models begins with the solved structure of an immunoglobulin chain or domain thereof. The chains to be modeled are compared for amino acid sequence similarity to chains or domains of resolved three-dimensional structures, and the chain or domain exhibiting the highest sequence similarity is selected as the starting point for molecular model construction. The solved starting structures are modified to allow for differences between the actual amino acids in the immunoglobulin chains or domains being modeled and those in the starting structures. The modified structures are then assembled into complex immunoglobulins. Finally, the model was refined by energy minimization and by verifying that all atoms are at the proper distance from each other and that the bond length and bond angle are within chemically acceptable limits.

Due to the degeneracy of the code, multiple nucleic acid sequences will encode each immunoglobulin amino acid sequence. The desired nucleic acid sequence may be generated by solid phase de novo DNA synthesis or by PCR mutagenesis of an earlier prepared variant of the desired polynucleotide. All antibody-encoding nucleic acids described in the present application are specifically included in the present invention.

The variable segments of human antibodies produced as described herein are typically linked to at least a portion of a human immunoglobulin constant region. The antibody will comprise light and heavy chain constant regions. The heavy chain constant region typically comprises a CH1, hinge, CH2, CH3 domain, and sometimes also a CH4 domain.

The human antibody may comprise any type of constant region from any antibody class, including IgM, IgG, IgD, IgA, and IgE, and from any subclass (isotype), including IgG1, IgG2, IgG3, and IgG 4. When it is desired that the humanized antibody exhibit cytotoxic activity, the constant region is typically a complement-binding constant region and the species is typically an IgG₁. When such cytotoxic activity is not required, the constant region may be an IgG₂And (4) class. Humanized antibodies may comprise sequences from more than one species or isotype.

Nucleic acids encoding the humanized light chain variable region and the heavy chain variable region, optionally linked to a constant region, are inserted into an expression vector. The light and heavy chains may be cloned in the same or different expression vectors. The DNA segment encoding the immunoglobulin chain is operably linked to control sequences in one or more expression vectors that ensure expression of the immunoglobulin polypeptide. Such control sequences include signal sequences, promoters, enhancers, and transcription termination sequences (see Queen et al, Proc. Natl. Acad. Sci. ("Proc. Acad. Sci., USA86, 10029 (1989)), WO 90/07861; Co et al, J. Immunol. (" J. Immunol.) 148, 1149(1992), which is incorporated herein by reference in its entirety for all purposes).

3. Methods of using antibodies against OSM

As described in detail below, the present invention shows that isolated monoclonal antibodies having variable domains of M5, M6, M9, M10, M42, M45, M53, M54, M55, M62, M63, M65, M66, M67, M68, M69, M71, and M83 bind to overlapping epitopes on OSM and exhibit in vitro and/or in vitro OSM inhibitory activity. Importantly, the reactivity of selected mabs includes the ability to: blocking OSM interaction with gp130 dose-dependently, reducing OSM signaling in the presence of gp130, reducing OSM-stimulated a375 cell proliferation, preventing macrophage-stimulated chondrocyte collagen production, or reducing OSM-induced cytokine release in vivo.

In view of the properties of the monoclonal antibody described in the present invention, the antibody or antigen-binding fragment thereof is suitable as a therapeutic and prophylactic agent for treating or preventing OSM-related disorders in humans and animals.

Generally, use will involve administering a therapeutically or prophylactically effective amount of one or more monoclonal antibodies or antigen-binding fragments of the invention, or selected antibodies or molecules with similar binding profiles and biological activities, to a sensitive subject or a subject exhibiting a condition in which OSM activity is known to have pathological consequences (e.g., immune disorders or tumor growth and cancer cell metastasis). Any active form of the antibody may be administered, including Fab and F (ab') 2 fragments.

Preferably, the antibody used is compatible with the recipient species such that an immune response to the MAb in the subject does not result in an unacceptably short circulating half-life or elicit an immune response to the MAb. The administered MAb may exhibit some ancillary functions, such as binding to the Fc receptor of the subject and activating ADCC mechanisms, so that the target cell population is depleted using cytolytic or cytotoxic mechanisms, or they may be engineered to limit or eliminate these ancillary functions, thereby preserving the target cell population.

Treatment of an individual may comprise administering a therapeutically effective amount of an antibody of the invention. The antibodies may be provided in kits as described below. The antibodies may be used or administered, for example, as a mixture of equal amounts, or may be provided separately, sequentially or administered together. In providing the patient with an antibody or a fragment thereof capable of binding to OSM or an antibody capable of blocking OSM in the subject patient, the dose of the administered agent may vary depending on factors such as the age, weight, height, sex, general physical condition, past medical history of the patient, and the like.

In a similar manner, another therapeutic use of the monoclonal antibodies of the invention is the active immunization of patients with anti-idiotype antibodies raised against a monoclonal antibody of the invention. Immunization with Anti-idiotypes mimicking the epitope structure can elicit active Anti-OSM responses (Linthicum, D.S. and Farid, N.R., Anti-idiotypes, Receptors, and Molecular mix (Anti-idiotypes, Receptors and Molecular Mimicry) (1988), pages 1-5 and 285-300).

Likewise, active immunity may be elicited by administration of one or more antigenic and/or immunogenic epitopes as a component of a vaccine. The vaccine can be administered orally or parenterally in an amount sufficient to prophylactically or therapeutically effect production of protective antibodies by the subject against the biological domain. The host may be actively immunized with the antigenic/immunogenic peptides, fragments of the peptides, or modified forms of the peptides in purified form. One or more amino acids that do not correspond to the original protein sequence may be added to the amino terminus or the carboxy terminus of the original peptide or truncated form of the peptide. Such additional amino acids may be used to couple the peptide to another peptide, to a large carrier protein, or to a support. Amino acids useful for these purposes include: tyrosine, lysine, glutamic acid, aspartic acid, cysteine and their derivatives. Alternative protein modification techniques, such as NH2 terminal acetylation or COOH terminal amidation, may be used to provide additional means of coupling or fusing the peptide to another protein or peptide molecule or support.

Antibodies capable of blocking unwanted OSM biological activity are intended to be provided to a subject in an amount sufficient to achieve alleviation, cure or amelioration of OSM-related symptoms or lesions. An amount is considered to be a sufficient amount or "therapeutically effective amount" to "achieve" symptom relief if the dose, route of administration, etc., of the agent is sufficient to affect such a response. The response to antibody administration can be determined by analyzing the subject's affected tissues, organs, or cells, such as by imaging techniques, or by analyzing tissue samples in vitro. An agent is physiologically significant if its presence results in a detectable change in the physiology of the subject patient.

Therapeutic applications

The OSM neutralizing antibodies, antigen binding fragments or specific variants thereof of the present invention may be used to measure or elicit effects in cells, tissues, organs or animals (including mammals and humans) to diagnose, monitor, regulate, treat, ameliorate, help prevent the incidence of or alleviate symptoms of a disorder mediated, affected or regulated by OSM or cells expressing OSM. Accordingly, the present invention provides methods of modulating or treating at least one OSM-associated disease in a cell, tissue, organ, animal or patient using at least one OSM antibody of the invention, as known in the art or as described herein.

OSM is known to be upregulated in a variety of disease states involving inflammation, and has been implicated in a variety of biological actions, including bone formation, cartilage degradation, cholesterol absorption, pain, and inflammation. Specific indications are discussed below.

Indications of

The present inventors have demonstrated that OSM mediates cartilage destruction, and have shown that OSM causes chondrocyte degeneration in the inter-articular matrix from tissue explants. OSM also promotes cytokine release, such as TNF α, which promotes collagen release from cartilage, as shown by t.cawston et al (1998, Arthritis and rheumatism, 41(10) 1760-: antibodies currently exemplified as the M71 binding domain are capable of blocking systemic cytokine release. The antibodies M55, M64, M69 and M71 of the present invention showed the following abilities: proteoglycan synthesis in macrophage-chondrocyte co-culture systems is increased above levels seen in the absence of OSM-specific antibodies.

The inventors have further demonstrated that administration of neutralizing antibodies against OSM of the invention will inhibit OSM-driven cytokine and chemokine release in vivo, such as IL-6, IP-10 and KC. IP-10, interferon gamma inducible protein 10kDa or inducible small cytokine B10, is a protein encoded by the CXCL10 gene (C-X-C motif chemokine 10(CXCL10)) in humans. Various effects have been attributed to CXCL10, such as chemotropism of monocytes/macrophages, T cells, NK cells and dendritic cells, as well as promotion of T cell adhesion to endothelial cells. KC, now called chemokine (C-X-C motif) ligand 1(CXCL1), is a small cytokine belonging to the CXC chemokine family, previously known as GRO1 oncogene, GRO α, neutrophil activating protein 3(NAP-3) and melanoma growth stimulating activity α (MSGA- α). In humans, this protein is encoded by the CXCL1 gene. CXCL1 is expressed by macrophages, neutrophils and epithelial cells and has neutrophil chemotactic activity.

According to the present invention there is therefore provided the use of an antibody or antibody fragment selected from M5, M6, M9, M10, M42, M45, M53, M54, M55, M62, M63, M65, M66, M67, M68, M69, M71and M83 in the manufacture of a medicament for the treatment or prevention of a proteoglycan degrading disease, such as osteoarthritis, inflammatory joint disease or an inflammatory condition. One particular use of antagonists of OSM is in the manufacture of a medicament for preventing or reducing collagen release from cartilage. The invention also provides methods for treating or preventing inflammatory joint disease or an inflammatory condition comprising administering to a patient having such a condition an effective amount of such an antibody that blocks OSM binding to gp 130.

The antibodies of the invention are useful in formulations for the treatment of pro-inflammatory processes in which OSM causes pathogenesis in tissues or organs, especially skin, lungs and joints, either directly or indirectly (e.g., through the release of inflammatory cytokines). Such conditions include osteoarthritis, rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, neuropathic joint disease, reactive arthritis, rotator cuff tear arthropathy, rheumatic fever, reiter's syndrome, progressive systemic sclerosis, primary biliary cirrhosis, pemphigus, pemphigoid, necrotizing vasculitis, myasthenia gravis, multiple sclerosis, lupus erythematosus, polymyositis, sarcoidosis, granulomatosis, vasculitis, pernicious anemia, CNS inflammatory disorders, diseases mediated by antigen-antibody complexes, autoimmune hemolytic anemia, hashimoto's thyroiditis, graves disease, raynaud's syndrome, glomerulonephritis, dermatomyositis, chronic active hepatitis, celiac disease, autoimmune complications of AIDS, atrophic gastritis, and addison's disease, endotoxemia or septic shock (septicemia), Or one or more of the symptoms of sepsis, as well as other types of acute and chronic inflammation. Those patients that can particularly benefit from the method of the invention are those suffering from infections caused by e.coli (e.coli), Haemophilus influenzae B (Haemophilus influenzae B), neisseria meningitidis (neisserial meningitides), staphylococcus or pneumococcus. Patients at risk of sepsis include those suffering from burns, wounds, renal or liver failure, trauma, burns, immune insufficiency (HIV), hematopoietic tumors, multiple myeloma, Castleman's disease, or cardiac myxoma.

Other conditions associated with OSM and amenable to therapeutic or prophylactic treatment with the antibodies of the invention include fibrotic diseases such as pulmonary fibrosis, diabetic nephropathy, idiopathic pulmonary fibrosis, systemic sclerosis, and cirrhosis of the liver. Another indication for treatment or prevention using the antibodies of the invention is nociceptive pain in neurons that involve the dorsal root ganglion.

Administration and dosing

The present invention provides stable formulations of OSM neutralizing antibodies, preferably aqueous phosphate buffered saline or mixed salt solutions, as well as preserved solutions and formulations and multipurpose preserved formulations suitable for pharmaceutical and veterinary use, comprising at least one OSM neutralizing antibody in a pharmaceutically acceptable formulation. Suitable vehicles and formulations thereof, including other human proteins such as human serum albumin, are described, for example, in Remington: the Science and Practice of Pharmacy, 21 st edition, Troy, D.B. ed, Lipincott Williams and Wilkins, Philadelphia, PA2006, part 5, Pharmaceutical Manufacturing (Pharmaceutical Manufacturing) page 691-.

OSM neutralizing antibodies in the stable or preserved formulations or solutions described herein may be administered to a patient according to the present invention by a variety of delivery methods, including intravenous (i.v.); intramuscular (I.M.); subcutaneous (s.c.); percutaneous; through the lung; transmucosal; formulations for use in implants, osmotic pumps, cartridges, micropumps; or in other ways as understood by the skilled person and well known in the art.

For example, specific administration into a body compartment or cavity can be by: intraarticular, intrabronchial, intraabdominal, intracapsular, intracartilaginous, intracavitary, intracerebroventricular, intracolic, intracervical, intragastric, intrahepatic, intramyocardial, intraosteal, intrapelvic, intrapericardiac, intraperitoneal, intrapleural, intraprostatic, intrapulmonary, intrarectal, intrarenal, intraretinal, intraspinal, intrasynovial, intrathoracic, intrauterine, intravesical, intralesional, vaginal, rectal, buccal, sublingual, intranasal, or transdermal means.

In general, if a systemic dose of antibody is administered, it is desirable to provide the recipient with a dose in the range of about 1ng/kg-100ng/kg, 100ng/kg-500ng/kg, 500 ng/kg-1. mu.g/kg, 1. mu.g/kg-100. mu.g/kg, 100. mu.g/kg-500. mu.g/kg, 500. mu.g/kg-1 mg/kg, 1mg/kg-50mg/kg, 50mg/kg-100mg/kg, 100mg/kg-500mg/kg (body weight of the recipient), although lower or higher doses may also be administered. Of course, the appropriate dosage of the antagonist of the invention will vary depending upon factors such as the disease or condition to be treated, the route of administration and the age and weight of the individual to be treated, as well as the nature of the antagonist. While not being bound by any particular dose, it is believed that, for example, for parenteral administration, a daily dose of 0.01 to 20mg/kg of an antibody (or other macromolecule) of the invention (typically present as part of a pharmaceutical composition as described above) may be suitable for treating a typical adult human.

Treatment may be carried out as a single dose therapy, or preferably as a multiple dose therapy, in which the initial course of treatment may be 1-10 separate doses administered, followed by administration of further doses at subsequent intervals, for example a second dose at 1-4 months, and if necessary one or more subsequent doses after several months, in order to maintain and/or enhance the response. Examples of suitable treatment sessions include: (i)0, 1 month, and 6 months, (ii)0, 7 days, and 1 month, (iii)0 and 1 month, (iv)0 and 6 months, or other courses of treatment sufficient to elicit a desired response expected to reduce the symptoms or severity of the disease.

The antibodies of the invention may be used alone or in combination with immunosuppressive agents such as steroids (prednisone, etc.), cyclophosphamide, cyclosporine a, or purine analogs (e.g., methotrexate, 6-mercaptopurine, etc.), or antibodies such as anti-lymphocyte antigen antibodies, anti-leukocyte antigen antibodies, TNF antagonists such as anti-TNF antibodies, or TNF inhibitors such as soluble TNF receptors, or agents such as NSAIDs or other cytokine inhibitors.

Sequence listing

The invention will now be described with reference to the following specific, non-limiting examples.

Example 1: reagents and assays

To select and characterize OSM binding antibodies, constructs of human and Cyno OSM were formed for mammalian cell expression. Human OSM (NP-065391, encoded by NM-020530) is a 252 amino acid precursor that is processed into a full-length secreted protein of 227 amino acids (SEQ ID NO: 11), which is a proprotein that is further processed into a more fully active mature form, amino acids 1-184. Human OSM cDNA was ordered from OriGene (Cat. No. SC121421), the ORF of human OSM from OriGene clones was amplified by PCR, and a signal peptide (murine IgG1) was introduced together with a six-His tag for protein purification and an Avi tag for site-directed protein biotinylation (SEQ ID NO: 56). The latter is chosen to avoid randomized biotin acylation of lysine residues present in the vicinity of the OSM interacting with the receptor.

Cynomolgus monkey OSM was cloned from cyno PBMC's RNA using Superscript III first strand synthesis system (InVitrogen) to obtain cDNA, followed by PCR amplification using UTR primers designed from human OSM sequences, as described in U.S. patent application serial No. 12/648430. The expressed full-length protein is shown in SEQ ID NO: 12, wherein the cleaved active form is represented by 184 residues, i.e., 51-227.

The precursor and mature forms of human and Cyno OSM were expressed and purified in HEK293 using standard methods. The functional activity of the proteins was tested in the A375-S2 cell proliferation and pSTAT signaling assays using E.coli-derived, commercially available human OSM (R & D Systems, Cat. No.295-OM) as a control.

Mab

When a control antibody was used, the human IgG1 isotype antibody designated CNTO6234 was used.

Acylation of chemical biotins

Recombinant human OSM was biotinylated using NHS-ester chemistry targeting amine residues on cytokines (EZ-linked sulfo-NHS-LC-biotinylation kit, Pierce, # 21435). The biotin coupling reaction was optimized to have a target labeling efficiency of one mole of biotin per mole of antigen. The latter minimizes the loss of binding and functional activity while ensuring near complete labeling of the protein population. Upon completion of the reaction, the protein was purified from free biotin reagent and residual leaving groups using a Zeba desalting spin column included in the EZ-linked sulfo-NHS-LC-biotinylation kit (Pierce). About 80% of the starting material was recovered. The HABA assay (Pierce biotin quantification kit, #28005) was used to measure the biotin binding level, which indicated approximately one mole of biotin per mole of human OSM. Binding of streptavidin coupling and GP130-Fc (R & D Systems, Cat. No.671-GP) to biotinylated proteins was verified using an Octet instrument (Fort BIO). Octet measurements show that the profile of biotinylated human OSM binding to gp130 is essentially identical to that of the unlabeled starting material.

In vitro targeted biotinylation

The 15-residue Avi tag (GLNDIFEAQKIEWHE) (SEQ ID NO: 56) has biotin-receptor kinetics similar to that of the endogenous BirA substrate BCCP (Beckett et al, 1999, Protein Science (Protein sciences)). When linked to a protein of interest, an Avi tag with one acceptor lysine residue will be biotinylated at only one position. Recombinant Cyno OSM was site-specifically biotinylated in vitro using biotin-protein ligase and reagents commercially available from Avidity. Biotinylated Cyno OSM was purified using a monovalent streptavidin affinity resin. The quality of the resulting protein was evaluated by SDS-PAGE and SEC-HPLC. The biotin binding level, indicated as approximately one mole of biotin per mole of cyno OSM, was measured using the HABA assay (Pierce biotin quantification kit, # 28005). Binding of streptavidin coupling and gp130-Fc chimera to biotinylated proteins was verified using an Octet instrument (Fort BIO). Octet measurements show that the profile of biotinylated Cyno OSM binding to gp130 is essentially identical to that of the unlabeled starting material.

Solid phase immunoassay

Primary phage-Fab panning Using NEUTRAVIDIDIN coated with Biotinylated human OSM or cyno OSM in TBS at 2. mu.g/ml^TMELISA plates (Pierce). After overnight incubation at 4 ℃, blocking and washing, a 1: 100 dilution from the polyclonal phage pool from each round of panning was added. Bound phage were detected with HR-conjugated monoclonals specific for pVIII, M13 phage major coat protein (GE Healthcare, Cat. No.27-9421-01), followed by addition of the chemiluminescent substrate POD (Roche, Cat. No.11582950001) and reading on a PerkinElmer instrument.

The secreted soluble Fab-His protein from E.coli supernatants was used for a primary screen of individual clones. Binding in ELISA format was performed using bacterial supernatants containing soluble Fab-His proteins. Black MaxiSorp plates (Nunc, Cat. No.437111) were coated with 1. mu.g/ml goat anti-human Fd (CH1) antibody (The Binding Site, Cat. No. PC075) and incubated overnight at 4 ℃. After washing and blocking the plates, 50 μ l of undiluted bacterial supernatant (containing Fab-His protein) was added and allowed to incubate for 1 hour at room temperature with gentle shaking. Plates were washed and biotinylated human or Cyno OSM (20nM) was added to the captured fabs. After 1 hour at room temperature, SA-HRP (Invitrogen, Cat. No.43-4323) was added and chemiluminescence detection was performed as above. By calculation, a preliminary binding ELISA screen with human OSM at a concentration of 20nM would allow the detection of clones with affinities in the nanomolar range.

Epitope box (epicpe Binning)

Epitope binning is a competition assay performed to classify MABs based on binding characteristics using human IgG1 transformed MABs and a commercial antibody MAb29, known as OSMR β/lifra recruitment blocker (R blocker).

To each well of a 384-well multi-array plate (Meso Scale Discovery (MSD), L25XA-4), 2.5. mu.g/ml of anti-human Fc (Jackson Immuno, 709-. The plates were incubated overnight at 4 ℃ and then blocked with 50. mu.l of MSD blocker A for 1 hour at room temperature. The 384 multi-array plates were washed three times (PBS pH7.4, 0.05% Tween20(Scytek, PBT010), and 1.0. mu.g/ml test mAb solution was added to each well, followed by shaking at level 6 on a titer plate shaker for 1 hour at room temperature.

In parallel, 5. mu.g/ml of the competitor mAb solution was mixed with 1. mu.g/ml of biotinylated OSM (R & D Systems, 295-OM-010/CF) in MSD assay buffer (1: 3 blocking buffer with PBS pH7.4, 0.05% Tween20) in separate 96-well plates (COSTAR, 3357) and shaken at level 3 on a titer shaker for 1 hour at room temperature. The pre-complex of competing antibody and biotinylated OSM was added to each well of 384 multi-array plates and shaken (at level 6 on titer plate shaker) at room temperature for 1 hour. The plates were then washed three times and streptavidin sulfo TAG (MSD, R32Ad-S) was added to each well and shaken (at level 6 on titer plate shaker) for 30 minutes at room temperature. The plates were washed three times and a read buffer T (MSD, R92TC-1) diluted 1: 4 with distilled water was added to each well. The plate was read on an MSD S6000 instrument.

Data were interpreted based on the signals obtained for the test mAb in the presence of biotinylated OSM and in the absence of competing mAb (maximal signal) and the signals from self-competition (2-4 fold reduction in maximal signal). Epitope competition is assigned when the value in the presence of competing mabs is within three standard deviations of the value from self-competition.

A375 cells

A375 cells (ATCC: CRL-1619) are human epithelial cells derived from malignant melanoma. A375-S2 cells (ATCC: CRL-1872, IL-1 sensitive CRL-1619 subline) were cultured in T-175 flasks (Corning; Cat. No.431080) in complete growth medium (DMEM/GlutaMax-I, Gibco) supplemented with 10% FBS (Gibco) and subcultured at a 1: 20 subculture ratio at approximately 80% confluence every three to four days.

A375-S2 cell proliferation (BrdU incorporation) assay

The reduction of a375-S2 cell proliferation by human or cynomolgus OSM was measured by BrdU incorporation using a chemiluminescent ELISA. Oncostatin M reduced proliferation of the A375-S2 human melanoma cell line (Zarling et al, PNAS 83: 9739-9743). This cell line was used to evaluate anti-oncostatin M antibodies at all stages of antibody discovery. A375-S2 cells were maintained in DMEM (Gibco11995) + 10% FBS (Gibco16140) + 1% penicillin/streptomycin (Gibco15140) in T-150 tissue culture flasks (BD Falcon35-5001) at 37 ℃ in 95% O2-5% CO2 and separated twice a week 1: 10.

For proliferation assays, cells were trypsinized (0.25%, Gibco25200) to remove them from T150 flasks and then cultured at 2000 cells/well density in 200 μ L DMEM/FBS/penicillin-streptomycin in the inner 48 wells of a black TC-coated plate (BD Falcon 353948). After overnight incubation at 37 ℃/95% O2-5% CO2, the media was removed and replaced with 180 μ Ι _ of fresh media. A single plate was prepared containing all test solutions at 10X final concentration. Transfer 20 μ L from the plate to the corresponding wells in the cell plate to create the appropriate experimental conditions. Each experimental condition was tested in triplicate. In any experiment in which neutralization was evaluated, the antibody and oncostatin M were incubated for at least 1 hour and then added to the cells. The plates were then incubated at 37 ℃/95% O2-5% CO2 for an additional 72 hours. At this time, a chemiluminescent BrdU cell proliferation ELISA (Roche11669915001) was performed. BrdU labeling reagent was added to the culture and held for 4 hours. The medium was then removed and 100 μ Ι _ of fixative solution was added to each well. After 30 minutes at room temperature, the solution was removed and 100. mu.L/well of anti-BrdU-POD solution was added. After 2 hours at room temperature, plates were washed with PBS-Tween and 100. mu.L of Super Signal Pico (Thermoscientific37069) was added to each well. Luminescence was then read using a Perkin Elmer Victor3 instrument.

Initial experiments were performed to determine the dose-response of self-generated CHO cell-derived recombinant human and cynomolgus oncostatin M to inhibit proliferation. Osm was tested from a starting concentration of 100ng/ml to a final concentration of 0.0244ng/ml obtained by dilution at 1: 5. Percent proliferation inhibition was calculated compared to cells treated with vehicle control. The results of this experiment are shown in fig. 2A. As the concentration of human or cyno oncostatin M increases, the incorporation of BrdU decreases accordingly. The antiproliferative activity of human and cynomolgus oncostatin M was indistinguishable based on the degree and effect of the antiproliferative activity of EC 50.

Based on these experiments, anti-oncostatin M antibodies were evaluated and compared using an oncostatin M concentration of 2ng/ml, as this concentration inhibited proliferation by approximately 80%, giving a larger window in which antibody dose responses were determined. The full dose range of the antibody was evaluated in the presence of 2ng/ml human or cyno oncostatin M. Furthermore, isotype controls were included in the experiments when the highest concentration of anti-oncostatin M antibody was used. The assay window is defined by the difference between untreated control wells (maximum proliferation) and wells incubated with 2ng/ml oncostatin M alone (minimum proliferation), and the percent neutralization at any antibody concentration is defined within this window.

pSTAT3 signaling using A375-S2 phospho-STAT 3 assay

OSM inhibits the growth of A375-S2 cells by binding to the cell surface receptor gp130 and inducing heterodimerization of the receptor with OSMR β to initiate the intracellular signaling cascade, which involves activation (phosphorylation) of the signaling molecule STAT3 (Kortylewski et al, Oncogene 18: 3742-3753, 1999). Disruption of STAT3 signaling abrogates OSM growth inhibition in a375-S2 cells, indicating that STAT3 activation is a critical step in OSM signaling (Heinrich et al, biochem. j. (journal of biochemistry) 374: 1-20, 2003). STAT3 phosphorylation in a375-S2 cells has been shown to be OSM concentration dependent and commercial kits are available to measure STAT3 phosphorylation in stimulated cells. Neutralization of OSM-induced STAT3 phosphorylation in a375-S2 cells was selected as a primary screening assay for Mab candidates.

For antibody neutralization of OSM-induced STAT3 phosphorylation, A375-S2 cells were seeded at 25,000 cells/well into 200. mu.l of complete growth medium in 96-well tissue culture plates (Corning; Cat. No.3596) and incubated for 24 hours. Cells were treated with a solution containing 5ng/ml human OSM (recombinant, mammalian cell-derived OSMN1-1) which had been incubated with 1: 5 serial dilutions of experimental mAb (starting at 10. mu.g/m) for 3 hours at room temperature. Controls included untreated cells, stimulated cells (only 5ng/ml OSM), and hIgG1 isotype control mAb. All treatments were performed in triplicate unless otherwise indicated.

Analysis of pSTAT3 content was performed using phospho-STAT 3 whole cell lysate kit (MSD; Cat. No. K150DID-1, Lot No. K0010570) according to the manufacturer's protocol. Briefly, cells were treated in a 200. mu.l/well volume for 10 minutes; the treatment solution was removed and 50 μ Ι of cell MSD lysis buffer was added by multichannel pipette. The plate was placed on an orbital shaker (300RPM) for 5 minutes. Then, each lysate was transferred to a MSD phospho-STAT 396 well plate at 30 l. The plates were sealed and placed on an orbital shaker (300RPM) for 1 hour at room temperature, washed three times with 150. mu.l of MSD wash buffer, then 25. mu.l of a second detection antibody conjugate (anti-pSTAT 3-Ru (bpy)32+) was added to each well, resealed and incubated on an orbital shaker (300RPM) for 1 hour at room temperature. The plate was washed as before and 150 μ l of MSD reading buffer (tripropylamine solution) was added to each well. Plates were read on an MSD SECTOR Imager6000 instrument.

Complete EC50 dose-response curves for mature, parental and control mabs within the row were obtained and plotted as normalized percent pSTAT3 signal.

Affinity measurements by surface plasmon resonance (Biacore)

Binding affinity was measured using Surface Plasmon Resonance (SPR) with a Biacore3000 optical biosensor (Biacore) using the human or Cyno OSM constructs described. The biosensor surface was prepared by: a mixture of anti-mouse (Jackson, Cat. No.315-005-046) and anti-human (Jackson, Cat. No.109-005-098) anti-IgG Fc antibodies was coupled to the carboxymethylated dextran surface of a CM-5 chip (Biacore, Cat. No. BR-1000-14) using the manufacturer's instructions for amine coupling chemistry. Approximately 19,000RU (reaction unit) of anti-OSM antibody was immobilized in each of the four flow cells. Kinetic experiments were performed in running buffer (DPBS + 0.005% P20+3mM EDTA) at 25 ℃. Serial dilutions of human and Cyno OSM ECD were made at 100nM to 0.412nM in running buffer. Approximately 200RU of mAb was captured on flow cells 2 to 4 of the sensor chip. The flow cell 1 serves as a reference surface. After capture of the mAb, the antigen was injected at 50uL/min for three minutes (association phase) followed by a 10 minute buffer flow (dissociation phase). The chip surface was regenerated by two 18 second pulses injecting 100mM H3PO4(Sigma, Cat. No.7961) at 50. mu.l/min.

The collected data was processed using BIAevaluation software (Biacore, version 3.2). The data is first subtracted by two reference values as follows: the reference value was subtracted from the curve obtained from the analyte injection, and the curve generated from the buffer injection was subtracted from the curve. Kinetic analysis of the data was performed using a globally fitted 1: 1 binding model. Results for each mAb were expressed as Ka (binding), Kd (dissociation) and K_D(affinity constant) format.

Example 2: selection of OSM binding to FAB

The de novo Fab-pIX library has been described in Shi et al, J Mol Biol (journal of molecular biology) 397: 385-396, 2010; WO09085462a 1; U.S. serial No. 12/546850; and designated 169, 323 and 551 herein above, with reference to the heavy chain human germline framework used: IGHV1-69(SEQ ID NO: 1), IGHV3-23(SEQ ID N) in IMGT nomenclatureO: 2) or IGHV5-51(SEQ ID NO: 3). The three heavy chain library frameworks were aligned with four light chain library VL as follows_κFrame combination: a27(IGKV 3-20X 01(SEQ ID NO: 5)), B3(IGKV 4-1X 01(SEQ ID NO: 6)), L6(IGKV 3-11X 01(SEQ ID NO: 7)) and O12(IGKV 1-39X 01(SEQ ID NO: 8)). In the library, the heavy chain is replaced by the addition of a peptide comprising SEQ ID NO: 4 and SEQ ID NO: 10 (FR4) and the Fab V region is intact. Heavy chain CDR3 has a variable length of 7-14 residues. Examples of the complete V regions of each library are shown in figure 1and numbered, and the CDR regions are shown according to Kabat.

The initial set of anti-OSM phage display hits were identified using commercially available glycosylated human OSMs. Fab-pIX phage display libraries were panned using biotinylated human OSM (R & D Systems, Cat. No.295-OM) trapped on paramagnetic Streptavidin (SA) beads (Invitrogen, Cat. No.112.05D) according to published protocols for phage selection (Marks and Bradbury, Antibody Engineering, Vol. 248: 161-176, Humana Press, 2004). Briefly, biotinylated human OSM was added to a phage library that had been pre-adsorbed on unconjugated beads to a final concentration of 100nM and incubated for 1 hour with gentle rotation. End-capped SA beads were added and incubated for 15 minutes to capture biotinylated OSM and bound phage. The magnetically trapped phage/antigen/bead complexes were washed 5 times with 1ml of TBST and 1 time with 1ml of TBS. After removal of the final TBS wash, 1ml of exponentially growing TG1 cells (Stratagene, cat. No.200123) were added and incubated at 37 ℃ for 30 minutes without shaking. The infected bacteria were spread on LB/agar (1% glucose/100. mu.g/ml carbenicillin) plates (Teknova, Cat. No. L5804) and incubated overnight at 37 ℃. The bacterial lawn was scraped and a glycerol stock [ 15% glycerol/carbenicillin (100. mu.g/ml)/2 XYT ] was prepared and stored at-80 ℃. To prepare the phage for the second round of panning, 25ml of 2 XYT/carbenicillin (100. mu.g/ml) was inoculated with 25. mu.l of bacterial glycerol stock and incubated at 37 ℃ until the OD600 was approximately 0.5. The helper phage VCSM13(Stratagene, Cat. No.200251) was added to the culture at a multiplicity of infection of approximately 10: 1and incubated for 30 minutes at 37 ℃ without shaking. The bacteria were spun down and the pellet resuspended in induction medium (2 XYT/carbenicillin/kanamycin/IPTG) and cultured overnight at 30 ℃. Phage were pelleted with 2% PEG/0.25M NaCl (final concentration) and resuspended in 2ml of PBS. The first round of phage was stored at 4 ℃ and used to perform a second round of panning. The elutriation parameters are: round 1, 100nM antigen, incubated at room temperature for 1 hour, washed 5 times with TBST and then 1 time with TBS; run 2, 10nM antigen, incubated 1 hour at room temperature, washed 10 times with TBST/1 time with TBS; and round 3, 1nM antigen, incubated at 4 ℃ for 16 hours (overnight), washed 10 times with TBST/1 time with TBS.

Success was monitored using ELISA, in which Fab was captured by anti-human Fd (CH1) antibody, 20nM biotinylated human OSM was added and bound OSM was detected with SA-HR.

Thirty (30) unique heavy-light chain pairings of Fab as displayed were identified and shown by ELISA to cross-react with Cyno OSM. The heavy chains represent sequences from 169(IGHV1-69 derived) and 551(IGHV5-51 derived) libraries, combined with light chain variable regions representing all four library germline sources (a27, B3, L6 and O12).

Example 3: characterization of osm-binding MAB

The four-helix bundle configuration OSM is characterized by four alpha-helical segments connected by relatively loose loops, these helical segments being designated A, B, C and D. OSM interacts with gp130 through a surface located in helices a and C (site II) determined to include the amino acid sequence of SEQ ID NO: 1 (Deller et al, Structure 8 (8): 863-. It is believed that the surface responsible for the interaction of OSM with OSMR β and LIFR α (site III) is defined primarily by residues located in helix D (Deller et al, supra).

The objective is to select high affinity linkers for OSM that can prevent OSM driven gp130 signaling by preventing OSM binding to gp130 (site II or B blockers) or gp130 recruitment of LIFRa or OSMRb (site III or R blockers) that prevent OSM binding.

29 of 30 initially selected OSM binding fabs were cloned into the vector for conversion to full-length human IgG1 Mab. Characterized by being determined as

(1) Competitive binding for identification of epitope clusters or "bins" (2) affinity measurement by surface plasmon resonance (Biacore), and (3) the ability to block pSTAT3 signaling. All screens and assays have been performed using mammalian cell-produced (glycosylated) human and cyno proteins as described in example 1.

Results

Data for affinity measurements for a subset of MABs selected based on ranking of MABs in pSTAT3 and ELISA binding assays relative to control mAb295(R & DSystems) are shown in table 3.

Table 3: selected affinities against the first round mAb subset of OSM。

The results of the binning assay showed that M5, M6, and M9 compete with each other but not with MAB 295. M10 competes with MAB 295. These four mabs also appear to be functional neutralizers of gp130 signaling as determined by pSTAT 3. Therefore, the following conclusions are drawn: m10 is an R blocker, M5, M6 and M9 block OSM binding to gp130 (B blocker).

The target affinity of the therapeutic leads is determined by the need to specifically compete for the OSM-gp130 interaction, the affinity of the OSM-gp130 interaction (K)_D) Has been measured at approximately 1 nM. Thus, a target affinity for OSM of 100pM or less K is desired_D。

Neutralizing MAbs M5, M6 and M9, found to be blockers of OSM-gp130 interactions,plus M10 was selected for in-line maturation in order to generate derivatives that meet the 100pM affinity requirements for therapeutic candidates. An additional desirable property is that the mAb binds cynomolgus OSM (K) with an affinity within five times that of human OSM_D500pM or less).

The composition of the binding domains of these four mabs was found to represent four unique heavy and light chain pairs, as identified in table 4.

TABLE 4：

Mab	HC libraries	HC ID	LC ID
				M10	169	H2	L2
M6	551	H14	L12
				M9	551	H17	IGKV4-1(B3)
M5	551	H135	L111

The complete variable region sequences comprise germline sequences used to form the phage libraries described herein above, and thus, the fixed residues match the unmutated parent germline residues. In this regard, each V region consists of a sequence designated SEQ ID NO: 1-3 or 5-8, designated CDR 1and CDR2 within the scaffold; followed by CDR3 (tables 3 and 4) and light and heavy chain variable regions (tables 5LC and 6HC), respectively, followed by J regions, the J region of the heavy chain being SEQ ID NO: 4, the J region of the light chain is SEQ ID NO: 10.

TABLE 5：

HC variable region H2 comprises a sequence derived from SEQ ID NO: FR1-CDR1-FR2-CDR2-FR3 of 1, wherein X₁＝A，X₂＝G，X₃＝I，X₄P, X5 ═ I, and X₆F, with an additional mutation in H-CDR 2. The HCs from library 551 (H14, H17, and H135) comprise the amino acid sequences derived from seq id NOs: FR1-CDR1-FR2-CDR2-FR3 of 3, wherein X₁＝S，X₂Or G, X₃＝I，X₄＝Y，X₅G and X₆Y or D. Each of the four HCs consists of a unique CDR3(SEQ ID NOS: 19-22).

TABLE 6：

The LC variable regions of selected fabs were all derived from the B3 library and included germline sequences denoted IGKV4-1(B3) in the IMGT database, IGKV4-1(B3) being used as starting sequences for library diversification as described herein and in the cited publications. Three of the four light chains differed in CDR 1and had the same H-CDR 2. The consensus sequence of the four selected LC variable FR1-CDR1-FR2-CDR2-FR3 was derived from SEQ ID NO: 8, wherein X₁Y, S or A; x₂K, E or N; x₃Y, W or F; and X₄Always W.

Two unique CDR3 sequences were identified. L-CDR3 can be represented by a consensus sequence (SEQ ID NO: 29) represented by the formula Q-Q- (S, Y) - (F, Y) -S- (F, T) -PLT.

Example 4: affinity reselection

The four V-region pairings described in example 3, OSMM5, OSMM6, OSMM9, and OSMM10, were selected for light chain reselection to improve affinity. To mature the large number of antibody affinities from the primary selection in an efficient and rapid manner, the methods described in Shi et al, JMol Biol (journal of molecular biology) 397: 385, 396, 2010 and WO09085462A 1and the "inline" maturation method of U.S. Ser. No. 12/546850. Briefly, V of antigen-specific clones obtained in the first few rounds of selection_HZone and corresponding V_LThe library combination of the scaffold, in this case based on the V region of B3 (SEQ ID NO: 8).

Three new libraries were formed, one using V for primary selection_LLibraries as multiple V_LThe source of the chains, while two other libraries were designed based on recent analysis of antigen-antibody complexes of known structure (Raghuanthan et al, j.mol. recognit (journal of molecular recognition), 2010). Residues are selected for diversification based on those most likely to be involved in binding of the target protein, also referred to as Specificity Determining Residue Usage (SDRU). In the case of the V.kappa.light chain,this has been identified as three contact regions centered on the hypervariable loops defined by Chothia and Lesk, which differ slightly depending on whether the target is a protein, peptide or small hapten. Table 5 shows V for affinity maturation_LLibrary, where B3 is the same library used during the discovery phase, library 2 "SDRM focused (SDRM focused)" is based on SDRU residues and focuses diversity, while NNK is a randomized library. In the table, "X" means any amino acid formed by NNK mixing plus a stop codon.

Table 7 summarizes the diversity in libraries 1, 2 and 3. During the analysis of the final library, some amino acids were identified which were not part of the original design and were therefore introduced as a result of the synthetic method. For library 2, synthesis was performed using dinucleotides, these amino acids being: s (position 30c), T (position 30D), EK (position 30F), IW (position 32), TV (position 50), I (position 92), D (position 93), and F (position 96).

TABLE 7：

Fab-His proteins were prepared from the third round of panning output and a monoclonal Fab-His ELISA was used to identify individual hits with higher binding signals than the corresponding parental Fab. Two concentrations of biotinylated human antigen were used in these fractionation ELISAs: 2nM and 0.2 nM. The binding signal for each parental clone was set to 100%. There were 22 hits from affinity maturation of M6 and M9, showing up to 9-fold (900%) improved binding when compared to the parent Fab comprising the original heavy and light chain V regions. Some of these new pairs of heavy and light chains were selected for further evaluation.

Affinity of the in-line mature mAbs of human and cynomolgus OSM compared to the affinity of MAB295 control and parent mAbs M6 and M9 (K)_D) The CDR compositions for those mabs are shown in table 9 and summarized in table 8. In some cases, LCVs were paired with both H14 and H17. Certain mabs designated as candidate therapeutic leads based on their biophysical properties and functional potency are highlighted in gray.

TABLE 8：

Table 9: LC-CDR tables for higher affinity mabs

The LC-CDR3 diversity Q-Q- (S, Y) - (F, Y) -S- (F, T) -PLT (SEQ ID NO: 29) in the selected candidates was reduced, and the consensus sequence of the reselected high affinity LC-CDR3 was denoted QQY- (F, Y) -STP- (L, I) -T (SEQ ID NO: 47).

For the reselected mabs, the VH of H14 and H17 both have a common FR1-CDR1-FR2-CDR2-FR3, consisting of SEQ ID NO: 48 and comprises the sequence given in SEQ ID NO: 14(CDR1) and SEQ ID NO: 17(CDR 2).

Reducing the ability of human and Cyno A375-S2 cells to proliferate

To evaluate the in-line matured antibody, a dose-response was performed, starting with 5. mu.g/ml or 1. mu.g/ml and decreasing to 0.0016 or 0.00032. mu.g, respectively, at a 1: 5 dilutiong/ml. The neutralization resulting from M71 is shown in fig. 2B. As antibody concentrations increased, the antiproliferative effects of both human and cynomolgus oncostatin M were neutralized. Dose-response curves were calculated using data from three separate experiments performed with human (open symbols) and cynomolgus monkey (closed symbols) oncostatin M. Table 10 summarizes the IC's of M55, M64, M69 and M71 against human and cynomolgus monkey oncostatin M₅₀(95% confidence interval).

Watch 10：

Human gp130 competition

Competition experiments were performed between anti-OSM mabs selected for in-line maturation plus selected new linkers from the above table and human gp130 using Surface Plasmon Resonance (SPR) as described in example 1 with a Biacore3000 optical biosensor (Biacore).

Biosensor surfaces were prepared by coupling each test mAb to the carboxymethylated dextran surface of a CM-5 chip (Biacore, Cat # BR-1000-14) according to the manufacturer's instructions. Approximately 4,000 and 15,000RU (reaction units) for each test mAb were immobilized in each of the four flow-through cells of the instrument. The competition experiments were performed in running buffer (DPBS + 0.005% P20+3mM EDTA) at 25 ℃. Human OSM (self OSMN1-1) was diluted to 30nM with running buffer and injected at3 μ Ι/min for 3 min on each flow cell with immobilized mAb. After trapping human OSM, 300nM of competing mAb or human gp130-Fc (associated phase) was injected for three minutes followed by 3 minutes of buffer flow (dissociation phase). 100mMH was injected at 50 μ L/min by two 12 second pulses₃PO₄(Sigma, Cat #7961) to regenerate the chip surface. The collected data was processed using BIAevaluation software version 3.2 (Biacore). First, the sensorgrams are aligned when injecting human OSM. The level of binding (RU) of human OSM, competitor mAb or gp130-Fc was then recorded. Binding when competing mAb or gp130-Fc is injected on the surface of OSMAn increase in level (RU) indicates no competition with the immobilized test mAb and vice versa. Flow cell (Fc1, etc.) immobilized mabs appeared along the transverse sample rows of table 10.

M2 was previously shown to compete with the commercial antibody MAB295, with MAB295 not known to compete with gp130 binding to OSM. The results of the competitive binding assay showed that M54, M55, M64, M69 and M71 compete with human gp130-Fc for OSM antigen (table 11).

Table 11: human gp130 competitive maturation of candidate leads mAb。

Ability to block pSTAT3 in A375 cells

Inhibition of pSTAT3 by M6 and M9, as well as by the in-line mature variants of these mabs, was performed in a375 cells, with or without 5ng/ml hOSM, with calculated EC50 values shown in table 12.

Table 12: pSTAT3 inhibition

mAb ID	EC50(ng/ml)	95% confidence interval	Curve fitting R 2
				M65	29.5	24.35 to 35.82	0.9921
M67	37.7	30.93 to 45.92	0.9905
				M45	50.0	42.45 to 58.91	0.9931
M83	71.2	59.59 to 85.05	0.9908
				MAB295	72.1	46.91 to 110.7	0.9848
M54	78.5	66.11. To 93.23	0.9930
				M63	88.2	58.55 to 132.9	0.9819
M64	97.1	77.01 to 122.4	0.9902
				M68	126.9	107.2 to 150.1	0.9924
M69	141.7	115.1 to 174.5	0.9914
				M66	152.7	111.1 to 209.8	0.9833
M71	159.6	128.3 to 198.5	0.9885
				M55	181.6	153.0 to 215.6	0.9869
M42	191.3	93.67 to 390.7	0.9607
				M62	233.5	206.4 to 264.1	0.9947
M53	236.6	196.3 to 285.2	0.9859
				M6	650.3	241.7 to 1750	0.9540
M9	1218.0	352.8 to 4208	0.9614
				M85	-		0.4454
IL13	-		0.3438

Example 5: biological activity

Macrophage-chondrocyte co-culture assay

M71 was evaluated in a macrophage-chondrocyte co-culture system. Differentiated macrophages are known to produce oncostatin M (Hasegawa et al, Rheumatology 38: 612-. OSM reduces the synthesis of the highly sulfated proteoglycan polysaccharide white glycans (GAGs) that make up an important part of the cartilage matrix. .

Discovery of anti-human oncostatin M antibodies using recombinant HEK-produced His-Avi-tagged human and cynomolgus monkey oncostatin M, and use of these molecules and antibodies derived from R&The activity of these antibodies was evaluated using recombinant human oncostatin M of bacterial origin from the D system (295-OM). These are all different from native endogenous human oncostatin M due to his-avi tag (Osm produced by HEK) or lack of glycosylation (bacterial recombinant Osm). Macrophages are known to secrete oncostatin M (Grove et al, J Lipid Res (J.Lipid. Res.) 32: 1889-97, 1991) and oncostatin M reduces proteoglycan synthesis in human chondrocytes (Sanchez et al, OA and Cart. ("osteoarthritis and cartilage") 12: 810-10, 2004). Thus, macrophage-chondrocyte co-culture systems were used to determine the ability of anti-human oncostatin M antibodies to neutralize endogenous or native human oncostatin M. Briefly, a single alginate bead containing 40,000 normal human articular chondrocytes was cultured in the presence of differentiated human macrophages for 72 hours. These experiments were performed in the presence of anti-oncostatin M antibodies at a range of doses. At the end of the experiment, by incorporation of radioactivity³⁵5O₄To measure proteoglycan synthesis.

CD14+ peripheral blood mononuclear cells were obtained from all cells. Monocytes were cultured at 2.5 × 105 cells/well in 0.5ml macrophage medium (RPMI + glutamine with 10% heat-inactivated FBS, 1% NEAA, and 1% penicillin-streptomycin) on 48-well plates. Cells were treated with macrophage-colony stimulating factor (M-CSF) at 100 ng/ml. After 48 hours, the medium was replaced to remove non-adherent cells. On day 6, the macrophage medium containing M-CSF was replaced with macrophage medium without M-CSF. On day 8, macrophage medium was replaced with chondrocyte medium (50% Ham's F-12/50% DMEM with 10% fetal bovine serum) and a single alginate bead chondrocyte culture (particulate Engineering # CDD-H-2200) was added to each well. Aliquots of macrophage conditioned medium were retained and stored at-80 ℃ for analysis of oncostatin M levels using the R & D system human oncostatin M DuoSet (DY 295).

The alginate bead-macrophage co-culture was maintained in the presence of 20. mu.g/ml of anti-human oncostatin M antibody (M64, M71, M55, M69) or in the presence of a dose range (5. mu.g/ml to 0.00076. mu.g/ml; 1: 3 dilution) of antibody (M71and M55). In addition, a human IgG1 isotype control (CNTO6234) was included on the plates at the highest concentration of anti-oncostatin M antibody tested. Other controls were chondrocytes only (no co-culture) and in the presence of 2ng/ml human oncostatin M. In addition, wells containing only macrophages were maintained in order to measure oncostatin M production. After 72 hours of co-incubation, 10. mu. Ci/ml radioactive 35SO4(Perkin-Elmer NEX041H002MC) was added to each well and held for an additional 20 hours.

Measurement Using CPC precipitation technique (MP BIomedicals (#190177)³⁵SO₄And (3) doping. In and with³⁵SO₄After 20 hours of incubation, the labeled medium was removed and each bead was washed twice with DPBS containing Ca and Mg. After the second wash, 200. mu.l of citrate buffer (150mM NaCl, 55mM sodium citrate, ph6.8) was added to each bead. Plates were incubated at 37 ℃ for 10 to 15 minutes until the column was dissolved. A 100 μ l aliquot from each well was transferred to a Millipore multi-layer sieve 96-well filter plate pre-wetted with 1% CPC, and then 10 μ l of 10% CPC was added to each well and held for 5 minutes. The plate was then evacuated until the filter was dry. Each well was then washed 2 times with 200 μ l of 1% CPC, and vacuum was applied to each well until the filter was dry. The plastic bottom was then removed from the plate and replaced with a seal (Perkin Elmer # 6005185). Scintillation fluid (50 μ l, Perkin Elmer #6013621) was added to each well and a plate seal was applied to the top of the plate. The plates are then counted on a Top Count reader (Top Count reader).

At 20 μ g/ml, M64, M71, M55 and M69 increased proteoglycan synthesis above the levels observed in the absence of antibody, while isotype control had no effect (fig. 3). In a separate experiment, M71 increased proteoglycan synthesis in a dose-dependent manner to the level exhibited by chondrocytes alone (defined as 100% neutralization), with an EC50 of 30ng/ml, whereas the isotype control had no effect. These data show that macrophage-derived Osm reduces proteoglycan synthesis in co-cultured chondrocytes, and that anti-human oncostatin M antibodies neutralize primary oncostatin M.

Human lung fibroblast phospho-STAT 3 assay

OSM induces proliferation and collagen production in normal human lung fibroblasts (Scaffidi et al, Br. J. Pharmacol (J. England. Pharmacol.) 136: 793-801, 2002). Excessive collagen production by fibroblasts is a key feature of many pathological conditions (Lim et al, Oncogene 23 (39): 5416-25, 2006; Huang et al, J Cell Biochem 81 (1): 102-13, 2001). Oncostatin M Receptor Signaling activates the JAK-STAT pathway, and phosphorylation of STAT3 is an early event in the Signaling pathway (August et al (1997) Signaling of Type II Oncostatin M Receptor (Signaling of Type II Oncostatin M Receptor) J Biol Chem (J. Biochem.) 272: 15760-15764). The ability of oncostatin M to produce pSTAT3 was determined in Normal Human Lung Fibroblasts (NHLF) using the R & D system human/mouse pSTAT3Duoset (DYC 4607-5). This assay was then used to determine the ability of M55 and M71 to neutralize oncostatin M signaling.

These experiments were performed using NHLF from Lonza (CC-2512) cultured on Lonza proprietary FGM-2(CC-3132) medium. Briefly, cells were seeded at 25,000 cells/well in FGM-2 on a plate and cultured for 24 hours. The cells were then treated with oncostatin M or antibody plus oncostatin M for 10 minutes. To avoid temperature dependent effects during this 10 minute incubation, all solutions used to prepare the treatments were pre-heated and maintained at 37 ℃. After 10 min treatment, the medium was aspirated off and replaced with complete lysis buffer. Lysis buffer (pH 7.2) consisted of 1% NP-40, 1% sodium deoxycholate, 0.1% SDS, 0.15M NaCl, and 0.01M sodium phosphate and was stored at 4 ℃. In use, the complete lysis buffer was prepared as follows: 1 tablet of protease inhibitor cocktail (Roche, 11836153001) and 110. mu.L of HALT phosphatase inhibitor (Thermo Scientific78420) were added to 11ml of lysis buffer. After a 10 minute lysis step, the resulting lysate was ready for detection in pSTAT 3.

To determine the oncostatin M dose-response, NHLF cells were treated with a range of doses of OSM (100ng/ml to 0.024ng/ml w/1: 4 diluted in triplicate). Dilution plates were prepared with pre-warmed PBS + 1% BSA, where the Osm concentration in each well was 10X higher than the final concentration required for treatment, and medium-only wells were included as untreated controls. The medium was completely removed from the plate and replaced with 180. mu.l of pre-warmed FGM-2. The timer was started for 10 minutes and then 20 μ Ι was transferred from each well in the dilution plate to the corresponding well in the culture plate. After 10 minutes, the treatment solution in the cell plate was completely removed by aspiration and replaced with 100 μ l of complete lysis buffer. To minimize the difference in incubation time, lysis buffer was added to the wells in the same order as the treatments. The assay plate was then placed on a shaker for 10 minutes. After shaking, the lysates were frozen at-80 ℃ for later testing or transferred directly to ELISA plates coated with anti-pSTAT 3. ELISA (R & D system human/mouse pSTAT3Duoset) was performed according to the manufacturer's instructions except 1) only 90 μ l of lysate or standard was added to each well and 2) SuperSignal Pico (ThermoScientific37069) was used as HR substrate. Luminescence was read on a Victor3 plate reader for ELISA plates.

To evaluate the ability of M55 and M71 to neutralize OSM signaling in NHLF, cells were treated with dose ranges of anti-OSM antibodies (triplicate wells, 500ng/ml to 0.005ng/ml, diluted 1-10 or 50ng/ml to 1.563ng/ml, diluted 1: 2) in the presence of 2ng/ml human OSM. Dilution plates were prepared with PBS for antibody dose-response treatment at 20X final concentration, including wells for isotype control (highest concentration of anti-OSM antibody), and untreated control and cells treated with OSM alone contained no antibody. Human oncostatin M was prepared with FGM-2 alone at 40ng/ml (20X final concentration). The 20X antibody and OSM solution were mixed in equal volumes in dilution plates to form 10X treatment solution, and the plates were incubated at 37 ℃ for 1 hour to allow OSM to bind to the antibody. After 1 hour, the medium was removed from the plate and replaced with 180 μ l of pre-warmed FGM-2. The timer was started for 10 minutes and 20 μ Ι was transferred from each well in the dilution plate to the corresponding well in the culture plate. After 10 minutes, the solution in the cell plate was completely removed by aspiration and replaced with 100 μ l of complete lysis buffer. To minimize the difference in incubation time, lysis buffer was added to the wells in the same order as the treatments. The assay plate was then placed on a shaker for 10 minutes. After shaking, the lysates were frozen at-80 ℃ for later testing or transferred directly to ELISA plates coated with anti-pSTAT 3. ELISA (R & D system human/mouse pSTAT3Duoset) was performed according to the manufacturer's instructions except 1) only 90 μ l of lysate or standard was added to each well and 2) SuperSignal Pico (ThermoScientific37069) was used as HR substrate. Luminescence was read on a Victor3 plate reader for ELISA plates.

Human oncostatin M increases pSTAT3 in NHLF cells with EC50 of about 1 ng/ml. An example of an oncostatin M dose response is provided in fig. 4A. For this example, EC50 was 0.90ng/ml with a 95% confidence interval of 0.70 to 1.17 ng/ml. The neutralizing capacity of anti-oncostatin M antibodies was determined in the presence of 2ng/ml of oncostatin M, and all data normalized to luminescence in the presence of 2ng/ml of OSM and in the absence of antibody. Figure 4B shows dose-dependent neutralization of M71 on OSM-induced STAT3 phosphorylation. As the concentration of M71 increased, the extent of phosphorylation of STAT3 decreased. Dose-response curves were calculated from data from six separate experiments and the IC50 calculated for M71 was 8.9ng/ml with 95% confidence intervals of 6.9 to 11.6 ng/ml.

Example 6: in vivo Activity

After systemic administration of human oncostatin M, M71 was evaluated for its ability to block cytokine production in vivo. Intraperitoneal injection of human oncostatin M increases the levels of certain serum cytokines, presumably by interacting with murine leukemia inhibitory factor receptors.

Systemic (i.p.) administration of human oncostatin M to mice developed a model for evaluating the neutralizing capacity of anti-oncostatin M monoclonal antibodies in an in vivo environment. Mice were i.p. injected with 10 μ g of human oncostatin M or PBS vehicle control in 200 μ l of PBS. After 1 hour, mice were anesthetized with CO2 and blood was collected by end heart puncture. Each blood sample was allowed to clot on ice for 20 minutes and then spun at 3500rpm for 10-15 minutes. Serum samples were cryopreserved until analyzed using Milliplex mouse MAP cytokine/chemokine multiplex (32) panels according to the manufacturer's instructions. Sample analysis showed that human oncostatin M significantly increased serum levels of murine KC, IP-10, MCP-1, IL-6 and eotaxin compared to vehicle control, with no effect on other cytokines in the panel. These data show that injection of human oncostatin M induces cytokine release, presumably by interacting with the murine leukemia inhibitory factor receptor (Richards et al, J Immunol. (. J. Immunol.) 159: 2431-37, 1997; Lindberg et al, Mol Cell Biol. (molecular Cell biology) 18: 3357-3367, 1988), which can be used to test the in vivo neutralizing capacity of anti-oncostatin M antibodies.

M71and M55 were evaluated in a mouse systemic administration model. Briefly, mice were dosed subcutaneously with M71 or M55(20, 2.0 or 0.2mg/kg human IgG1 anti-human Osm), CNTO6234(huIgG1 isotype control, 20mg/kg) or PBS in a volume of 10. mu.l/g. After 24 hours, each mouse was then injected i.p. with 10 μ g of self-produced CHO cell-derived recombinant human oncostatin M w/0.1% mouse serum albumin (Sigma a3559) in PBS (Sigma D8357) or PBS-MSA vehicle control alone (200 μ L total volume). After 1 hour, mice were anesthetized with CO2 and blood was collected by end heart puncture. Each blood sample was allowed to clot on ice for 20 minutes and then spun at 3500rpm for 10-15 minutes. Serum samples were cryopreserved until analyzed using Milliplex mouse MAP cytokine/chemokine multiplex (32) panels. Serum samples were analyzed according to the manufacturer's instructions.

Human oncostatin M induced a significant increase in serum levels (unpaired student's t-test) of five of the cytokines detected by the kit: eotaxin, IL-6, IP-10, KC and MCP-1. Pre-administration of the isotype control CNTO6234 at 20mg/kg had no effect on the release of cytokines induced by oncostatin M. However, pre-administration of M71 significantly reduced serum levels of IP-10, MCP-1, IL-6 and eotaxin (administered at 2.0 and 20mg/kg) and KC (administered at 20 mg/kg). M71 at 0.2mg/kg had no effect on any cytokine. The effects of M71and isotype control on IP-10 and MCP-1 are shown in FIGS. 5A and B, respectively. It was observed that M55 has a less robust neutralization, since 20 and 2.0mg/kg of neutralization were seen only in IP-10. Serum levels of IL-6, eotaxin and MCP-1 were reduced with 20mg/kg M55, whereas no reduction in KC levels was seen with any dose of M55. These data show the ability of anti-oncostatin M antibodies to neutralize the biological effects of exogenous human oncostatin M in a murine model of systemic administration.

IP-10, interferon gamma inducible protein 10kDa or inducible small cytokine B10, is a protein encoded by the CXCL10 gene (C-X-C motif chemokine 10(CXCL10)) in humans. Various effects have been attributed to CXCL10, such as chemotropism of monocytes/macrophages, T cells, NK cells and dendritic cells, as well as promotion of T cell adhesion to endothelial cells. KC, now called chemokine (C-X-C motif) ligand 1(CXCL1), is a small cytokine belonging to the CXC chemokine family, previously known as GRO1 oncogene, GRO α, neutrophil activating protein 3(NAP-3) and melanoma growth stimulating activity α (MSGA- α). In humans, this protein is encoded by the CXCL1 gene. CXCL1 is expressed by macrophages, neutrophils and epithelial cells and has neutrophil chemotactic activity.

Example 7: co-crystallography

Fab fragments comprising V regions H17(SEQ ID NO: 51) and L180(SEQ ID NO: 55) were crystallized from residues 26-212 of human OSM (SEQ ID NO: 10).

OSM shares its four-helix bundle three-dimensional structure with other members of the gp130 cytokine family. The four-helix bundle configuration is characterized by four alpha helical segments connected by relatively loose loops, these helical segments are designated A (residues 10-37), B (residues 67-90), C (residues 105-131) and D (residues 159-185). OSM interacts with gp130 via an epitope called site II, which includes the amino acid residues located in helices A and C (Q16, Q20, G120, N123, N124) (Deller et al, Structure 8(8) (863-. It is believed that site III, the epitope responsible for the interaction of OSM with OSMR β and LIFR α, is defined primarily by residues located in helix D (Deller et al, Structure 8 (8): 863-; 874, 2000).

Crystallization of

Crystallization of the complexes was performed by the sink vapor diffusion method at 20 ℃ using an Oryx4 protein crystallization manipulator (Douglas Instruments) to dispense equal volumes of 0.2. mu.L of protein complex (10.95mg/mL) and 0.2. mu.L of stock solution. And (5) performing multiple crystallization screening. Most of the droplets remained clear, reflecting the high solubility of the complex. Crystals were obtained from 0.1M MES pH6.5, 2.4M ammonium sulfate and 0.1M Tris pH8.5, 3.5M sodium formate.

Results of the crystal structure solution

The OSM residues contacted with H14/L180Fab constitute the binding epitope. Antibody residues that contact OSM constitute the binding paratope. All six CDRs of both variable domains are involved in OSM binding. The contacted residues are given in table 13 and shown in figure 5. The long CDR-L1 forms a valley-like antigen binding site along with the CDRs of heavy chain variable region H1, with a small ridge at the bottom (fig. 4C, left panel). When joined, the two sides of the valley enclose the OSM four-helix bundle along helices a and C with the bottom ridge joined between the two helices (fig. 4C, right panel). Antibody and antigen binding interface maskingWith a solvent accessible surface (for Ab isFor Ag is). Although there are multiple charged residues at the interface, there is no charge-charge pairing, suggesting that vdw and H bonding play the most important role in antibody and antigen interaction.

Watch 13：

Distance cut-off for hydrogen bonds (highlighted in bold)And for vdw contact is

Thus the H17/L180Fab has previously shown residues to be contacted by gp 130; q20 and G120, and contacts the OSM along the a and C helices.

Example 8: pharmacokinetics

OSM is a soluble target associated with inflammatory processes, unlike cell surface displayed targets on cells. The properties of intact IgG comprising a binding region as found herein and additionally having an Fc domain can thus be tailored to the purpose and therapeutic specifications associated with its use using methods of engineering Fc with mutations conferring altered FcR binding.

In the compositions of the invention, the sustained release activity and persistence in circulation is a beneficial specification for the treatment of monoclonal IgG. Thus, the Fc domain has enhanced affinity for the neonatal receptor (FcRn).

Mutant and wild-type antibodies were constructed using standard recombinant techniques using the previously described mutation M428L (MedImmune U.S. patent No.7670600) in combination with N434S (US7371826, WO 2006/053301). The two mabs, M71and M71L/S, were compared in a standard activity assay and their persistence in the circulation of non-human primates was compared.

Measurement of

M71and M71L/S were compared in parallel in the A375-S2 proliferation assay. Dose-response was evaluated from the starting concentration 1. mu.g/ml to 0.00032. mu.g/ml (diluted 1: 5). At a concentration of 1. mu.g/ml, the isotype controls for these antibodies, namely CNTO3930 and CNTO8852, respectively, had no effect on the ability of 2ng/ml of human oncostatin M to inhibit proliferation. Both M71and M71L/S completely neutralized the effect of oncostatin M at 1. mu.g/ml, and there was no measurable difference in IC 50.

M71and M71L/S were compared in a mouse systemic administration model. Briefly, mice were dosed subcutaneously with M71and M71L/S (20, 10 or 5.0mg/kg), CNTO3930(huIgG1 isotype control, 20mg/kg), CNTO8852 (isotype control for Fc mutant variants, 20mg/kg) or PBS in a volume of 10. mu.l/g. After 24 hours, each mouse was injected i.p. with 10 μ g of self-produced CHO cell-derived recombinant human oncostatin M w/0.1% mouse serum albumin (Sigma a3559) in PBS (Sigma D8357) or PBS-MSA vehicle control alone (200 μ L total volume). After 1 hour, mice were anesthetized with CO2 and blood was collected by end heart puncture. Each blood sample was allowed to clot on ice for 20 minutes and then spun at 3500rpm for 10-15 minutes. Serum samples were cryopreserved until analyzed using a custom Millipore mouse multiplex consisting of beads specific for IL-6, MCP-1, eotaxin, KC, and IP-10. The isotype control had no effect on cytokine release induced by human oncostatin M. However, both M71and M71L/S neutralized oncostatin M-induced cytokine release and there was no significant difference in potency or efficacy.

Pharmacokinetic analysis

Serum half-lives of M71and M71L/S were compared in a non-terminal cynomolgus pharmacokinetic study. The study included a total of 12 cynomolgus monkeys and evaluated for each antibody for subcutaneous (s.c., n-3) and intravenous (i.v., n-3) administration. The antibody was administered at a dose of 3mg/kg and the study was conducted for more than 60 days. Blood samples were taken at 1 hour (group IV only) and 6 hours and at days 1, 2, 4, 6, 8, 12, 16, 30, 37, 45 and 60. Serum from the samples was frozen at-80 ℃ until tested. Antibody levels in serum were determined using ELISA optimized for the MesoScale Discovery platform in cynomolgus monkey serum. The biotinylated capture antibody was an anti-idiotypic antibody raised against M71 (mouse anti-M71). The detection antibody was ruthenium-labeled anti-human IgG and the readout was mesoscaledicovery chemiluminescence.

The results of this study are shown in fig. 6A and B. FIG. 6A shows a plot from i.v. dosing where the serum half-life of M71 was 15.21+/-3.0 days, while the half-life of M71L/S was 29.4+/-2.3 days. Similar results were obtained from s.c. dosing (FIG. 6B), where the serum half-life of M71 was 15.4+/-4 days, while the serum half-life of M71L/S was 32.0+/-5.9 days.

The results show an approximately two-fold increase in M71L/S t1/2 compared to M71.

Claims (42)

1. An isolated antibody that specifically binds human OSM protein and modulates the interaction between human OSM protein and gp130 protein, comprising heavy chain complementarity determining region 3 (H-CDR3) of an amino acid sequence selected from the group consisting of SEQ ID NOS: 20 and 21.

2. The antibody of claim 1, further comprising a heavy chain framework 1sequence, a H-CDR 1sequence, a framework 2 sequence, a H-CDR2 sequence, and a framework 3 sequence having the amino acid sequence of SEQ ID NO 48.

3. The antibody of claim 2, wherein heavy chain residue S31 contacts OSM protein having the amino acid sequence of SEQ ID No.11 at position Q20, and heavy chain residues T30 and Y52 contact OSM protein having the amino acid sequence of SEQ ID No.11 at position G120.

4. The antibody of claim 2, wherein residues of the human variable light chain framework are derived from a human vk germline gene.

5. The antibody of claim 4, wherein the vk germline gene is the IGKV4 germline gene sequence.

6. The antibody of claim 2, further comprising a light chain variable region comprising SEQ ID NO 8, wherein X₁Y, S or A; x₂K, E or N; x₃Y, W or F; and X₄Is W.

7. The antibody of claim 6, wherein the L-CDR3 sequence comprises the amino acid sequence of SEQ ID NO. 29.

8. The antibody of claim 7, wherein the L-CDR3 sequence comprises an amino acid sequence selected from the group consisting of SEQ ID NOs 27, 28, 45, and 46.

9. The antibody of claim 6, wherein the L-CDR 1sequence comprises an amino acid sequence selected from the group consisting of SEQ ID NOs 23-25 and 30-41.

10. The antibody of claim 9, wherein the L-CDR2 sequence comprises an amino acid sequence selected from SEQ ID NOs 26 and 42-44.

11. An isolated antibody that specifically binds human OSM and modulates the interaction between human OSM and human gp130 protein, comprising a light chain variable region amino acid sequence selected from SEQ ID NOS: 49-53 and/or a heavy chain variable region amino acid sequence selected from SEQ ID NOS: 54 and 55.

12. The isolated antibody of claim 11, wherein the antibody comprises the light chain variable region amino acid sequence of SEQ ID NO: 51 and the heavy chain variable region amino acid sequence of SEQ ID NO: 55.

13. The isolated antibody of claim 11, wherein the antibody comprises the light chain variable region amino acid sequence of SEQ ID NO: 53 and the heavy chain variable region amino acid sequence of SEQ ID NO: 54.

14. The isolated antibody of claim 11, wherein the paratope contacts epitope residues of human OSM having the amino acid sequence of SEQ ID No.11 at positions Q16, Q20, and G120.

15. The isolated antibody of any one of claims 1-14, further comprising a human heavy chain constant region selected from the group consisting of IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, and IgM.

16. The isolated antibody of claim 15, wherein the constant region comprises a human IgG isotype.

17. The isolated antibody of claim 16, wherein the isotype is IgG 1.

18. The isolated antibody of claim 17, wherein the constant region is mutated such that the antibody is non-lytic.

19. The isolated antibody of claim 16, wherein the constant region is mutated to enhance the affinity of the antibody for a neonatal receptor (FcRn) as compared to an antibody having a wild-type IgG1 constant domain sequence.

20. The isolated antibody of claim 19, wherein the constant region is mutated at positions 428 and 434, wherein the numbering is according to the Kabat EU numbering.

21. The isolated antibody of claim 20, wherein the mutations are M428L and N434S, wherein the numbering is according to the Kabat EU numbering.

22. An antigen-binding fragment of any one of the antibodies of claims 1-21.

23. The antigen-binding fragment of claim 22, wherein the fragment is selected from the group consisting of Fab, Fab', Fd, F (ab)2, and ScFv.

24. A pharmaceutical composition comprising the isolated antibody or antigen-binding fragment of said antibody of any one of claims 1-23, and a pharmaceutically acceptable excipient or carrier.

25. A method of treating a human patient suffering from a disease or condition responsive to modulation of the interaction between human OSM protein and human gp130 protein, said method comprising the step of administering to said patient a therapeutically effective amount of the composition of claim 24.

26. The method of claim 25, wherein the disease or disorder is an arthropathy selected from the group consisting of: osteoarthritis, rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, neuropathic joint disease, reactive arthritis, and rotator cuff tear arthropathy.

27. A method of treating a human patient suffering from a disease or condition characterized by the release of pro-inflammatory cytokines and chemokines by macrophages and monocytes comprising the step of administering to said patient a therapeutically effective amount of the composition of claim 24.

28. The method of claim 25, wherein the disease or condition is selected from the group consisting of: rheumatoid arthritis, juvenile arthritis, psoriatic arthritis, ankylosing spondylitis, psoriasis, chronic plaque-type disease, lupus erythematosus, inflammatory lung diseases, idiopathic pulmonary fibrosis, sepsis, preeclampsia, COPD, asthma, and multiple sclerosis.

29. The method of treating a human patient according to claim 25, wherein the patient has a fibrotic disease selected from: atherosclerosis, diabetic nephropathy, pulmonary fibrosis, idiopathic pulmonary fibrosis, systemic sclerosis, and cirrhosis of the liver.

30. An isolated polynucleotide encoding the heavy and/or light chain of the therapeutic antibody of any one of claims 1-21, or an antigen-binding fragment of said therapeutic antibody.

31. A stably transformed or transfected recombinant host cell comprising the isolated polynucleotide of claim 30.

32. The stably transformed or transfected recombinant host cell of claim 31 comprising a vector comprising a polynucleotide encoding the light chain variable amino acid sequence of SEQ ID No. 53 and a second polynucleotide encoding the heavy chain variable amino acid sequence of SEQ ID No. 54.

33. The stably transformed or transfected recombinant host cell of claim 31 comprising a vector comprising a polynucleotide encoding the light chain variable amino acid sequence of SEQ ID No. 51 and a second polynucleotide encoding the heavy chain variable amino acid sequence of SEQ ID No. 55.

34. The host cell of any one of claims 32 or 33, wherein the cell is mammalian.

35. The host cell of claim 34, wherein the cell is CHO.

36. A method of making an antibody comprising the steps of culturing the host cell of claim 34 and recovering said antibody from said cell.

37. A kit comprising a sterile formulation of the isolated antibody or fragment of any one of claims 1-21 and instructions for administering the antibody to a subject in need thereof.

38. An isolated antibody that specifically binds human OSM protein, comprising:

a. 23-25 and 30-41 selected from the group consisting of light chain complementarity determining region 1 (L-CDR1) amino acid sequence;

b. 26, 42-44, and a light chain complementarity determining region 2 (L-CDR2) amino acid sequence selected from SEQ ID NOs; and

c. 10, 27-29, 45, 47, light chain complementarity determining region 3 (L-CDR3) amino acid sequence.

39. An isolated antibody that specifically binds OSM, comprising:

a. a heavy chain complementarity determining region 1 (H-CDR1) amino acid sequence selected from SEQ ID NOs 14 and 15;

b. a heavy chain complementarity determining region 2(H-CDR 2) amino acid sequence selected from SEQ ID NOS 16-18; and

c. a heavy chain complementarity determining region 3 (H-CDR3) amino acid sequence selected from SEQ ID NOS 19-22.

40. An isolated antibody comprising the light chain L-CDRs 1, 2, and 3 of claim 38 and the heavy chain H-CDRs 1, 2, and 3 of claim 39.

41. An isolated antibody that specifically binds OSM, comprising:

H-CDR1 having the amino acid sequence of SEQ ID NO. 14;

H-CDR2 having the amino acid sequence of SEQ ID NO. 17;

H-CDR3 having the amino acid sequence of SEQ ID NO. 21;

L-CDR1 having the amino acid sequence of SEQ ID NO 38;

L-CDR2 having the amino acid sequence of SEQ ID NO. 43; and

46 has the amino acid sequence of SEQ ID NO. 3.

42. Any invention described herein.