Anti-MET antibodies and their applications

Abstract

The present invention relates to agonistic anti-MET antibodies and their use in therapeutic treatment of diseases. The antibodies or antigen-binding fragments thereof bind to human MET protein (hMET) with high affinity and to mouse MET protein (mMET) with high affinity, wherein the antibodies or antigen-binding fragments thereof are hMET agonists and mMET agonists. The antibodies or antigen-binding fragments of the present invention produce molecular and cellular effects similar to HGF binding.

Description

[0001] This application is a divisional application of the Chinese patent application with application date of June 23, 2017, application number 201780040456.2, and invention name “Anti-MET Antibodies and Applications Thereof”. Technical Field

[0002] The present invention relates to antibodies and antigen-binding fragments that bind to human and mouse hepatocyte growth factor (HGF) receptors (also known as MET) with high affinity. The antibodies and antigen-binding fragments are agonists of MET in humans and mice, thereby producing molecular and cellular effects similar to those of HGF binding. The present invention also relates to the therapeutic use of antibodies and antigen-binding fragments as MET agonists. Background Art

[0003] HGF is a pleiotropic cytokine of mesenchymal origin that mediates a characteristic array of biological functions including cell proliferation, motility, differentiation and survival. The HGF receptor, also known as MET, is expressed by a variety of tissues including: all epithelial cells, endothelial cells, muscle cells, neuronal cells, osteoblasts, hematopoietic cells, and various components of the immune system.

[0004] HGF and MET signaling play an important role during embryonic development, where it guides the migration of precursor cells and determines cell survival or death. In adults, HGF / MET signaling is normally quiescent and is restored during wound healing and tissue regeneration. Some cancers and tumors usurp HGF / MET signaling to promote tumor survival and proliferation in the host organism. Therefore, despite limited success, inhibition of the HGF-MET axis has become a popular target for anticancer therapy.

[0005] Due to its role in tissue healing and regeneration, recombinant HGF has also been studied for the treatment of many conditions, including: degenerative diseases, inflammatory diseases, autoimmune diseases, metabolic diseases and transplantation-related conditions. However, recombinant HGF has poor pharmacological properties: it requires proteolytic activation to be biologically active; once activated, it has an extremely short half-life in vivo; and its industrial production is complex and expensive.

[0006] Agonist anti-MET antibodies that activate MET in a manner that mimics HGF have been proposed as an alternative.

[0007] The following antibodies that at least partially mimic the activity of HGF have been described: (i) 3D6 mouse anti-human MET antibody (U.S. Pat. No. 6,099,841); (ii) 5D5 mouse anti-human MET antibody (U.S. Pat. No. 5,686,292); (iii) NO-23 mouse anti-human MET antibody (U.S. Pat. No. 7,556,804 B2); (iv) B7 human naive anti-human MET antibody (U.S. Patent Application No. 2014 / 0193431 A1); (v) DO-24 mouse anti-human MET antibody (Prat et al., Mol Cell Biol. 11, 5954-5962, 1991; Prat et al., J Cell Sci. 111, 237-247, 1998); and (vi) DN-30 mouse anti-human MET antibody (Prat et al., Mol Cell Biol. 11, 5954-5962, 1991; Prat et al., J Cell Sci. 111, 237-247, 1998). Biol. 11, 5954-5962, 1991; Prat et al., J Cell Sci. 111, 237-247, 1998). Summary of the invention

[0008] Agonistic anti-MET antibodies produced to date, such as those described in the Background section, are often obtained as byproducts of processes aimed at identifying antagonistic molecules and are not explicitly designed to be agonist molecules for therapeutic use. In addition, the most significant limitation of prior art anti-MET antibodies is that they have been produced in mouse systems (except for B7, which was identified using a human naive phage library); as a result, these antibodies are unlikely to exhibit cross-reactivity with mouse MET. Even if minor cross-reactivity with self-antigens is in principle possible, these interactions are generally of very low affinity.

[0009] While the absence of cross-reactivity is not a concern for mouse cancer models (because they use human xenografts), cross-reactivity of antibodies between human and mouse MET is an important requirement for regenerative medicine or preclinical mouse models of non-oncology human diseases, which require antibodies to function on mouse tissues and cells.

[0010] Not only is it desirable for an agonistic anti-MET antibody to cross-react with mouse MET in order to evaluate the antibody in preclinical models, it is also desirable for the antibody to bind to mouse MET with the same or similar affinity as it does to human MET, and for the antibody to elicit the same or similar effects in the mouse system as it does in the human system - otherwise experiments conducted in preclinical models will not be predictive of the human situation. As demonstrated in the Examples, none of the prior art anti-MET agonistic antibodies exhibit affinity for mouse MET, and certainly none of the prior art antibodies exhibit the same or similar binding and agonism in both mouse and human systems.

[0011] The present application provides anti-MET agonist antibodies prepared by designing to bind human and mouse MET with high affinity. These antibodies: (i) exhibit agonistic activity in human and mouse MET biological systems - that is, they induce MET signaling - some with similar or higher potency than HGF; (ii) induce a full spectrum of HGF-induced biological activities, thus representing an effective alternative to recombinant HGF; (iii) exhibit excellent binding to mouse MET when directly compared with prior art antibodies; (iv) exhibit biologically significant agonist activity at concentrations as low as 1 pM; (v) exhibit a plasma half-life of several days in mice, and pharmacological saturation concentrations are reached at a dose of 1 μg / kg, which is very low for therapeutic antibodies; (vi) maintain renal function and renal integrity in a mouse model of acute kidney injury; (vii) prevent liver failure in a mouse model of acute liver injury and Antagonize hepatocyte injury; (viii) exhibit anti-fibrotic, anti-inflammatory and pro-regenerative activities in a mouse model of chronic liver injury; (ix) prevent weight loss, reduce intestinal bleeding, maintain colon integrity, inhibit inflammation and promote epithelial regeneration in a mouse model of ulcerative colitis and a mouse model of inflammatory bowel disease; (x) promote insulin-dependent glucose uptake in a mouse model of type I diabetes; (xi) overcome insulin resistance in a mouse model of type II diabetes; (xii) improve fatty liver, inhibit fibrosis and restore liver function in a mouse model of non-alcoholic steatohepatitis (NASH); (xiii) accelerate wound healing in a mouse model of diabetic ulcers; (xiv) interact with MET in Rattus norvegicus and Macaca fascicularis fascicularis) MET, thereby allowing toxicology and pharmacology studies in these two vertebrates that are required before applying for first-in-human trials; and (xv) recognizing epitopes that are conserved in humans, mice, rats, and cynomolgus monkeys, thereby providing greater utility in animal models.

[0012] Therefore, in a first aspect, the present invention provides an antibody or an antigen-binding fragment thereof, which binds to human MET protein (hMET) with high affinity and binds to mouse MET protein (mMET) with high affinity, wherein the antibody or antigen-binding fragment thereof is an hMET agonist and an mMET agonist. In certain embodiments, the antibody or antigen-binding fragment thereof comprises at least one heavy chain variable domain (VH) and at least one light chain variable domain (VL), wherein the VH and VL domains, when tested as Fab fragments, exhibit 1×10 -3 s -1 Up to 1×10 -2 s -1 , optionally 1×10 -3 s-1 Up to 6×10 -3 s -1 The dissociation rate (k measured by Biacore) 解离 ), and showed 1×10 -3 s -1 Up to 1×10 -2 s -1 , optionally 1×10 -3 s -1 Up to 6×10 -3 s -1 The dissociation rate (k measured by Biacore) 解离 ). In certain embodiments, the antibody or antigen-binding fragment thereof has equivalent affinity for hMET and mMET.

[0013] In certain embodiments, the antibody or antigen-binding fragment thereof induces phosphorylation of hMET and induces phosphorylation of mMET. In certain embodiments, the antibody or antigen-binding fragment thereof induces phosphorylation of hMET and induces phosphorylation of mMET at an EC of less than 3.0 nM, optionally less than 2.0 nM. 50 Induces phosphorylation of hMET (as measured by phospho-MET ELISA) and has an EC of less than 3.0 nM, optionally less than 2.0 nM. 50 Induces phosphorylation of mMET (as measured by phospho-MET ELISA). In certain embodiments, the antibody or antigen-binding fragment thereof induces phosphorylation of hMET and mMET equivalently.

[0014] In certain embodiments, the antibody or antigen-binding fragment thereof exhibits high phosphorylation potency for hMET and high phosphorylation potency for mMET. In certain embodiments, the antibody or antigen-binding fragment thereof exhibits high phosphorylation potency for hMET and high phosphorylation potency for mMET. 50 and / or at least 80% of E max Induced phosphorylation of hMET (expressed as a percentage of HGF-induced activation in phospho-MET ELISA) and had an EC of less than 1 nM. 50 and / or at least 80% of E max In some alternative embodiments, the antibody or antigen-binding fragment thereof exhibits low phosphorylation potency for hMET and low phosphorylation potency for mMET. In some such embodiments, the antibody or antigen-binding fragment thereof has an EC of 1 nM to 5 nM. 50 and / or 60-80% E max(expressed as the percentage of HGF-induced activation in phosphorylated MET ELISA) induced phosphorylation of hMET and the EC values ​​were 1 nM to 5 nM. 50 and / or 60-80% E max (Expressed as percentage of HGF-induced activation in phospho-MET ELISA) Induced phosphorylation of mMET.

[0015] In certain embodiments, the antibody or antigen binding fragment thereof induces an HGF-like cell response when in contact with human cells and induces an HGF-like cell response when in contact with mouse cells. In certain embodiments, the antibody or antigen binding fragment thereof completely induces an HGF-like cell response when in contact with human cells and when in contact with mouse cells. In certain embodiments, the complete induction of an HGF-like cell response is measurable when one, any two, or all of the following occur:

[0016] (i) in a cell scattering assay, when the concentration of the antibody or antigen-binding fragment thereof is 0.1 to 1.0 nM, the antibody or antigen-binding fragment thereof induces cell scattering comparable to maximal HGF-induced scattering;

[0017] (ii) In an anti-apoptotic cell assay, the antibody or antigen-binding fragment thereof exhibits EC 50 EC smaller than HGF 50 1.1 times of E max (measured as % of total ATP content of non-apoptotic control cells) was greater than that observed for HGF. max 90%; and / or

[0018] (iii) In a branching morphogenesis assay, cells treated with the antibody exhibited greater than 90% of the number of branches per spheroid induced by the same (non-zero) concentration of HGF

[0019] In certain embodiments, the antibody or antigen-binding fragment thereof partially induces an HGF-like cell response when in contact with human cells and when in contact with mouse cells. In certain embodiments, the partial induction of an HGF-like cell response is measurable when the following (i)-(iii) occur:

[0020] (i) in a cell scattering assay, the antibody or antigen-binding fragment thereof induces at least 25% of the cell scattering induced by 0.1 nM homologous HGF when the concentration of the antibody is 1 nM or less;

[0021] (ii) In an anti-apoptotic cell assay, the antibody or antigen-binding fragment thereof exhibits EC 50 EC no greater than HGF 50 7.0 times, and / or Emax The cell viability is at least 50% of that observed for HGF; and / or

[0022] (ii) in a branching morphogenesis assay, cells treated with the antibody exhibit at least 25% more branches per spheroid than induced by the same (non-zero) concentration of HGF;

[0023] Furthermore, the antibody or antigen-binding fragment cannot completely induce HGF-like cellular response.

[0024] In certain embodiments, the antibody or antigen-binding fragment thereof is an HGF competitor. In certain embodiments, the antibody or antigen-binding fragment thereof has an IC of no greater than 5 nM. 50 and / or at least 50% of I max Competes with hHGF for binding to hMET with an IC of no greater than 5 nM 50 and / or at least 50% of I max In some embodiments, the antibody or antigen-binding fragment thereof competes equally with hHGF and mHGF. In some embodiments, the antibody or antigen-binding fragment thereof is a complete HGF competitor. In some such embodiments, the antibody or antigen-binding fragment thereof has an IC of less than 2 nM. 50 and / or greater than 90% I max Competes with hHGF and has an IC value of less than 2 nM 50 and / or greater than 90% I max In some embodiments, the antibody or antigen-binding fragment thereof is a partial HGF competitor. In some such embodiments, the antibody or antigen-binding fragment thereof has an IC of 2 to 5 nM. 50 and / or 50% to 90% I max Competes with hHGF and has an IC of 2-5 nM 50 and / or 50% to 90% I max Competes with mHGF.

[0025] The antibodies or antigen-binding fragments thereof of the present invention may show cross-reactivity with monkey-derived MET, such as cynomolgus (Macaca fascicularis) MET, and may show cross-reactivity with rat-derived MET (Rattus norvegicus).

[0026] The antibodies or antigen-binding fragments thereof of the present invention can bind to an epitope of human MET between amino acid residues 123 to 223 of human MET (throughout the document, the numbering of human MET refers to GenBank sequence number X54559). Also provided are antibodies or antigen-binding fragments thereof of the present invention, which can bind to an epitope of human MET between amino acids 224-311 of human MET. Also provided are antibodies or antigen-binding fragments thereof of the present invention, which can bind to an epitope of human MET between amino acids 314-372 of human MET. Also provided are antibodies or antigen-binding fragments thereof of the present invention, which can bind to an epitope of human MET between amino acids 546-562 of human MET.

[0027] Also provided are antibodies or antigen-binding fragments thereof of the invention, which can bind to an epitope of human MET comprising amino acid residue Ile367. Also provided are antibodies or antigen-binding fragments thereof of the invention, which can bind to an epitope of human MET comprising amino acid residue Asp372 of human MET. In certain embodiments, the antibody or antigen-binding fragment thereof binds to an epitope of human MET comprising amino acid residues Ile367 and Asp372 of human MET.

[0028] Also provided is an antibody or an antigen-binding fragment thereof of the present invention, which can bind to an epitope of human MET comprising amino acid residue Thr555 of human MET.

[0029] The present invention also provides an antibody or an antigen-binding fragment thereof, wherein the antibody or the antigen-binding fragment thereof comprises a heavy chain variable domain and a light chain variable domain, wherein the heavy chain variable domain comprises H-CDR1, H-CDR2 and H-CDR3, and the light chain variable domain comprises L-CDR1, L-CDR2 and L-CDR3, wherein:

[0030] H-CDR1 comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 2, 9, 16, 23, 30, 37, 44, 51, 58, 65 and 72;

[0031] H-CDR2 comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 4, 11, 18, 25, 32, 39, 46, 53, 60, 67 and 74;

[0032] H-CDR3 comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 6, 13, 20, 27, 34, 41, 48, 55, 62, 69 and 76;

[0033] L-CDR1 comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 79, 86, 93, 100, 107, 114, 121, 128, 135, 142 and 149;

[0034] L-CDR2 comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 81, 88, 95, 102, 109, 116, 123, 130, 137, 144 and 151; and

[0035] L-CDR3 comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 83, 90, 97, 104, 111, 118, 125, 132, 139, 146 and 153.

[0036] [71G2] In one embodiment, the present invention provides an antibody or antigen-binding fragment comprising a heavy chain variable domain and a light chain variable domain, wherein the heavy chain variable domain comprises H-CDR1, H-CDR2 and H-CDR3, and the light chain variable domain comprises L-CDR1, L-CDR2 and L-CDR3, wherein:

[0037] H-CDR1 comprises the amino acid sequence shown in SEQ ID NO:44;

[0038] H-CDR2 comprises the amino acid sequence shown in SEQ ID NO:46;

[0039] H-CDR3 comprises the amino acid sequence shown in SEQ ID NO:48;

[0040] L-CDR1 comprises the amino acid sequence shown in SEQ ID NO: 121;

[0041] L-CDR2 comprises the amino acid sequence shown in SEQ ID NO: 123; and

[0042] L-CDR3 comprises the amino acid sequence shown in SEQ ID NO:125.

[0043] [71G2] In one such embodiment, the heavy chain variable domain of the antibody or fragment comprises the amino acid sequence of SEQ ID NO:167, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto; and the light chain variable domain comprises the amino acid sequence of SEQ ID NO:168, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto.

[0044] [71D6] In another embodiment, the present invention provides an antibody or antigen-binding fragment comprising a heavy chain variable domain and a light chain variable domain, wherein the heavy chain variable domain comprises H-CDR1, H-CDR2 and H-CDR3, and the light chain variable domain comprises L-CDR1, L-CDR2 and L-CDR3, wherein:

[0045] H-CDR1 comprises the amino acid sequence shown in SEQ ID NO:30;

[0046] H-CDR2 comprises the amino acid sequence shown in SEQ ID NO:32;

[0047] H-CDR3 comprises the amino acid sequence shown in SEQ ID NO:34;

[0048] L-CDR1 comprises the amino acid sequence shown in SEQ ID NO: 107;

[0049] L-CDR2 comprises the amino acid sequence shown in SEQ ID NO: 109; and

[0050] L-CDR3 comprises the amino acid sequence shown in SEQ ID NO:111.

[0051] [71D6] In one such embodiment, the heavy chain variable domain of the antibody or antigen-binding fragment comprises the amino acid sequence of SEQ ID NO:163, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto; and the light chain variable domain comprises the amino acid sequence of SEQ ID NO:164, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto.

[0052] [71G3] In another embodiment, the present invention provides an antibody or antigen-binding fragment comprising a heavy chain variable domain and a light chain variable domain, wherein the heavy chain variable domain comprises H-CDR1, H-CDR2 and H-CDR3, and the light chain variable domain comprises L-CDR1, L-CDR2 and L-CDR3, wherein:

[0053] H-CDR1 comprises the amino acid sequence shown in SEQ ID NO:9;

[0054] H-CDR2 comprises the amino acid sequence shown in SEQ ID NO: 11;

[0055] H-CDR3 comprises the amino acid sequence shown in SEQ ID NO: 13;

[0056] L-CDR1 comprises the amino acid sequence shown in SEQ ID NO:86;

[0057] L-CDR2 comprises the amino acid sequence shown in SEQ ID NO: 88; and

[0058] L-CDR3 comprises the amino acid sequence shown in SEQ ID NO:90.

[0059] [71G3] In one embodiment, the heavy chain variable domain of the antibody or antigen-binding fragment comprises the amino acid sequence of SEQ ID NO: 157, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto; and the light chain variable domain comprises the amino acid sequence of SEQ ID NO: 158, or a sequence that is at least 90%, 95%, 97% or 99% identical thereto.

[0060] In another embodiment, the present invention provides an antibody or antigen-binding fragment comprising a heavy chain variable domain comprising H-CDR1, H-CDR2 and H-CDR3 and a light chain variable domain comprising L-CDR1, L-CDR2 and L-CDR3, wherein H-CDR1, H-CDR2 and H-CDR3 are selected from a group of CDRs (CDR1, CDR2 and CDR3) of Fab shown in Table 3, and L-CDR1, L-CDR2 and L-CDR3 are the corresponding CDRs (CDR1, CDR2 and CDR3) of the same Fab shown in Table 4.

[0061] In certain embodiments, the heavy chain variable domain of the antibody or antigen-binding fragment comprises a VH amino acid sequence in Table 5, or a sequence at least 90%, 95%, 97% or 99% identical thereto, and the light chain variable domain comprises a corresponding VL amino acid sequence in Table 5, or a sequence at least 90%, 95%, 97% or 99% identical thereto.

[0062] The amino acid sequence of the VH domain shows an embodiment of less than 100% sequence identity with a defined VH domain amino acid sequence (e.g., SEQ ID NO: x), although it may include the same heavy chain CDRs as HCDR1, HCDR2, and HCDR3 of the VH of SEQ ID NO: x, but shows amino acid sequence variation in the framework region. For example, one or more amino acid residues in the framework region may be replaced by amino acid residues that appear in equivalent positions in the human VH domain encoded by the human germline. Similarly, the amino acid sequence of the VL domain shows an embodiment of less than 100% sequence identity with a defined VL domain amino acid sequence (e.g., SEQ ID NO: y), although it may include the same light chain CDRs as LCDR1, LCDR2, and LCDR1 of the VH of SEQ ID NO: y, but shows amino acid sequence variation in the framework region. For example, one or more amino acid residues in the framework region may be replaced by amino acid residues that appear in equivalent positions in the human VL domain encoded by the human germline.

[0063] The present invention also provides antibodies and antigen-binding fragments, which contain humanized / germline variants of the VH and VL domains of the aforementioned antibodies, and affinity variants and variants of conservative amino acid substitutions as defined herein. Specifically provided are chimeric antibodies, which contain the VH and VL domains of the Fab derived from the above-mentioned llama, or human germline variants thereof fused to the constant domains of human antibodies (particularly human IgG1, IgG2, IgG3 or IgG4). The heavy chain and light chain variable domains of the aforementioned antibodies, or germline variants thereof, affinity variants or conservative variants thereof may be included in conventional four-chain antibodies or other antigen-binding proteins, such as Fab, Fab', F(ab')2, bispecific Fab, and Fv fragments, diabodies, linear antibodies, single-chain antibody molecules, single-chain variable fragments (scFv) and multispecific antibodies. The heavy chain variable domain or germline variants thereof, affinity variants or conservative variants thereof may also be used as single domain antibodies.

[0064] In another aspect, the present invention also provides: an isolated polynucleotide encoding an antibody or antigen-binding fragment according to the present invention; an expression vector comprising the polynucleotide operably linked to a regulatory sequence that allows the antibody or antigen-binding fragment thereof to be expressed in a host cell or a cell-free expression system; and a host cell or a cell-free expression system containing the expression vector. The present invention also provides a method for producing a recombinant antibody or antigen-binding fragment thereof, the method comprising culturing the host cell or cell-free expression system under conditions that allow the expression of the antibody or antigen-binding fragment and recovering the expressed antibody or antigen-binding fragment.

[0065] In another aspect, the present invention provides a pharmaceutical composition comprising the antibody or antigen-binding fragment of the present invention and at least one pharmaceutically acceptable carrier or excipient.

[0066] In another aspect, the present invention provides the use of an antibody or antigen-binding fragment of the present invention or a pharmaceutical composition of the present invention in therapy.

[0067] In another aspect, the present invention provides a method for treating or preventing liver damage in a human patient, wherein the liver damage is optionally acute liver damage or chronic liver damage, the method comprising administering a therapeutically effective amount of a MET agonist antibody to a patient in need thereof. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the present invention.

[0068] In another aspect, the present invention provides a method for treating or preventing renal injury in a human patient, wherein the renal injury is optionally acute renal injury, the method comprising administering a therapeutically effective amount of a MET agonist antibody to a patient in need thereof. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the present invention.

[0069] In another aspect, the present invention provides a method for treating or preventing inflammatory bowel disease in a human patient, the inflammatory bowel disease is optionally ulcerative colitis, the method comprising administering a therapeutically effective amount of a MET agonist antibody to a patient in need thereof. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the present invention.

[0070] In another aspect, the present invention provides a method for treating or preventing diabetes in a human patient, wherein the diabetes is optionally type I or type II diabetes, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET agonist antibody. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the present invention.

[0071] In another aspect, the invention provides a method of treating or preventing nonalcoholic steatohepatitis in a human patient, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET agonist antibody. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the invention.

[0072] In another aspect, the present invention provides a method for treating or promoting wound healing in a human patient, the human patient is optionally a patient with diabetes, the method comprising administering a therapeutically effective amount of a MET agonist antibody to a patient in need thereof. In certain embodiments, the MET agonist antibody is an antibody or antigen-binding fragment according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 : The immune response of llamas immunized with human MET-Fc was determined by ELISA. Human MET ECD (hMET) or mouse MET ECD (mMET) recombinant proteins were immobilized in a solid phase and exposed to serial dilutions of serum from llamas before (PRE) or after (POST) immunization. Binding was revealed using mouse anti-llama IgG1 and HRP-conjugated donkey anti-mouse antibodies. OD: optical density; AU: arbitrary units.

[0074] Figure 2 : Schematic representation of human MET deletion mutants used to identify the MET domain responsible for mAb binding. ECD: extracellular domain; aa: amino acid; L.peptide: leader peptide; SEMA: semaphorin homology domain; PSI or P: plexin-semaphorin-integrin homology domain; IPT: immunoglobulin-transcription factor-plexin homology domain. On the right, the corresponding residues of human MET are reported according to UniProtKB#P08581.

[0075] Figure 3 : Schematic diagram of the llama-human chimeric MET protein used to finely locate the epitope recognized by the anti-MET antibody. The extracellular portion of the llama MET and human MET consists of 931 and 932 amino acids (aa), respectively (llama MET has a short leader peptide of 2 aa but has an insertion after the 163rd aa). Both receptor extracellular domains contain a leader peptide, a semaphorin homology domain (SEMA), a plexin-semaphorin-integrin homology domain (PSI or P), and four immunoglobulin-transcription factor-plexin homology domains (IPT). Chimera CH1-5 has an N-terminal llama portion, followed by a C-terminal human portion. Chimera CH6-7 has an N-terminal human portion, followed by a C-terminal llama portion.

[0076] Figure 4 : Agonist activity of human / mouse equivalent anti-MET antibodies in human and mouse cells as determined by Western blotting. A549 human lung cancer cells and MLP29 mouse liver precursor cells were serum starved and then stimulated with increasing concentrations of mAb or recombinant human HGF (hHGF; A549) or mouse HGF (mHGF; MLP29). MET autophosphorylation was determined by Western blotting using an anti-phospho-MET antibody (tyrosines 1234-1235). The same cell lysates were also analyzed by Western blotting using an anti-whole human MET antibody (A549) or an anti-whole mouse MET antibody (MLP29).

[0077] Figure 5 : The biological activity of human / mouse equivalent anti-MET antibodies was determined by the branching morphogenesis assay using LOC human renal epithelial cells and MLP29 mouse liver precursor cells. Cell spheroids were seeded in a collagen layer and then exposed to increasing concentrations of mAb or recombinant human HGF (LOC) or mouse HGF (MLP29). Branching morphogenesis was observed by microscopy over time and colonies were photographed after 5 days.

[0078] Figure 6 : Comparison with prior art antibodies: Human-mouse cross-reactivity. Human or mouse MET ECD was immobilized in a solid phase and exposed to increasing concentrations of antibody (all in mouse IgG / λ format) in solution. Binding was revealed by ELISA using HRP-conjugated anti-mouse Fc antibody.

[0079] Figure 7: Comparison with prior art antibodies: MET autophosphorylation. A549 human lung cancer cells and MLP29 mouse liver progenitor cells were deprived of serum growth factors for 48 hours and then stimulated with increasing concentrations of antibodies. After 15 minutes of stimulation, cells were lysed and phospho-MET levels were determined by ELISA using anti-MET antibodies for capture and anti-phosphotyrosine antibodies for revealing.

[0080] Figure 8 : Comparison with prior art antibodies: Branching morphogenesis. LOC human renal epithelial cell spheroids were seeded in a collagen layer and then incubated with increasing concentrations of mAbs. Branching morphogenesis was observed microscopically over time and colonies were photographed after 5 days.

[0081] Fig. 9 : Comparison with prior art antibodies: Branching morphogenesis. MLP29 mouse liver precursor cell spheroids were seeded in a collagen layer and then incubated with increasing concentrations of mAb. Branching morphogenesis was observed by microscopy over time and colonies were photographed after 5 days.

[0082] Fig.10 : Plasma stability of human / mouse equivalent anti-MET antibodies. A single bolus of 1 mg / kg or 10 mg / kg antibody was injected intraperitoneally, and blood samples were collected from the tail vein 3, 6, 12 and 24 hours after injection. The blood samples were processed and the antibody concentration in plasma was determined by ELISA. (A) Peak and trough levels of injected antibody. (B) Antibody plasma half-life was calculated by linear fit of the antibody concentration Ln transformation.

[0083] Fig.11 :Acute liver failure model: plasma concentration of liver function markers.Acute liver injury is induced in BALB / c mice by subcutaneous injection of CCl4 solution.Soon after poisoning, mice are randomly divided into 4 large groups (arm), which receive a single push of 71G3, 71D6, 71G2 or only vehicle (PBS). The antibody is administered by intraperitoneal injection at a dose of 5mg / kg. Each large group comprises three groups of mice, which are killed at different times (12, 24 and 48 hours) after poisoning. Blood samples are collected at different times (0, 12, 24 and 48 hours) after injection. At autopsy, blood and liver are collected for analysis. The plasma levels of liver markers aspartate aminotransferase (AST), alanine aminotransferase (ALT) and bilirubin (BIL) are measured by standard clinical biochemical methods.

[0084] Fig.12 : Acute liver failure model: Histological examination of liver sections. Fig.11As described in the legend, acute liver injury was induced in BALB / c mice. At autopsy, livers were extracted and embedded in paraffin for histological analysis. Sections were stained with hematoxylin and eosin and examined by microscopy. Representative images of each treatment group are shown. Magnification: 100 times.

[0085] Fig.13 :Chronic liver injury model:The plasma concentration of liver function markers.Liver damage and fibrosis in BALB / c mice are induced by long-term exposure to CCl4 for several weeks.Soon after the first CCl4 injection, mice are randomly divided into 4 large groups, which accept the processing of 71G3, 71D6, 71G2 or only vehicle (PBS) respectively.Antibody is used by intraperitoneal injection dosage of 1mg / kg three times a week.Other 5th control large group does not accept CCl4 or antibody and is used as healthy control.At chronic CCl4 poisoning after 6 weeks, mice are put to death.When autopsy, blood and liver are collected and analyzed.The plasma level of liver marker aspartate aminotransferase (AST) and alanine aminotransferase (ALT) is measured by standard clinical biochemical method.

[0086] Fig.14 : Chronic liver injury model: Histological examination of liver sections stained with Picro Sirius red. Fig.13 As described in the legend, liver damage and fibrosis in BALB / c mice were induced by long-term exposure to CCl4 for several weeks. At autopsy, the liver was extracted and embedded in paraffin for immunohistochemical analysis. The sections were stained with Picro Sirius red. Representative images of each treatment group are shown. Magnification: 100 times.

[0087] Fig.15 : Chronic liver injury model: Histological examination of liver sections stained with anti-α-smooth muscle actin (α-SMA) antibody. Fig.13 As described in the legend, liver damage and fibrosis in BALB / c mice were induced by chronic exposure to CCl4. At autopsy, livers were extracted and embedded in paraffin for immunohistochemical analysis. Sections were stained with anti-α-smooth muscle actin (α-SMA) antibody. Representative images of each treatment group are shown. Magnification: 100 times.

[0088] Fig.16: Acute kidney injury model: plasma levels of renal function markers.Acute renal failure was induced in BALB / c mice by a single bolus of HgCl2 injected intraperitoneally. Soon after HgCl2 poisoning, mice were randomly divided into 4 large groups, which were subjected to the treatment of 71G3, 71D6, 71G2 or only vehicle (PBS). The antibody was administered at a dose of 5 mg / kg by intraperitoneal injection every 24 hours. Mice were killed after 72 hours of injection of HgCl2. At autopsy, blood and kidneys were collected for analysis. Blood urea nitrogen (BUN) and creatinine (CRE) plasma levels were measured by standard clinical biochemistry methods.

[0089] Fig.17 : Acute kidney injury model: histological analysis of kidney sections. Fig.16 As described in the legend, acute renal failure was induced in BALB / c mice by HgCl2 injection. At autopsy, kidneys were extracted and embedded in paraffin for histological analysis. Kidney sections were stained with hematoxylin and eosin. Representative images of each treatment group are shown. Magnification: 400 times.

[0090] Fig.18 :Ulcerative colitis model: body weight, disease activity index (DAI) and colon length.Ulcerative colitis was induced in BALB / c mice by adding dextran sulfate sodium (DSS) to drinking water for 10 days.On the 10th day, DSS treatment was interrupted and mice were put back into normal water.From the 1st day, mice were randomly divided into 7 large groups, which were treated with 71G3, 71D6, 71G2 (dosage was 1mg / kg or 5mg / kg) or only vehicle (PBS).The other 8th control group did not receive DSS or antibody and was used as a healthy control.On the 12th day, that is, 2 days after interrupting DSS administration, mice were killed.At autopsy, colons were collected, washed, and their length was measured with a ruler.After measurement, colons were embedded in paraffin and processed for histological analysis.During the whole experiment, mice weight was monitored regularly, and the clinical symptoms of ulcerative colitis were evaluated by measuring fecal blood, rectal bleeding and fecal consistency. Each parameter is scored from 0 (no symptoms) to 3 (maximum manifestation of symptoms). The scores for the individual parameters are added together to produce a DAI of 0 to 9. (A) Changes in body weight over time (percentage relative to time 0). (B) Changes in DAI over time. (C) Colon length at autopsy. For clarity, data for the 1 mg / kg and 5 mg / kg groups are shown in separate figures.

[0091] Fig.19 : Ulcerative colitis model: histological analysis of colon sections. Fig.18As described in the legend, ulcerative colitis was induced in BALB / c mice by exposure to dextran sulfate sodium (DSS). At necropsy, the colon was collected, measured, then embedded in paraffin and processed for histological analysis. Colon sections were stained with hematoxylin and eosin, examined by microscopy, and photographed. Experimental groups, antibody doses, and magnifications are indicated adjacent to each image. Please refer to the text for image analysis.

[0092] Fig. 20 : Inflammatory bowel disease model: body weight and colon length. Colon injury and inflammation were induced in C57BLKS / J mice by rectal injection of 2,4,6-trinitrobenzenesulfonic acid (TNBS) dissolved in ethanol. Shortly after TNBS administration, mice were randomly divided into 4 large groups, which were treated with 71G3, 71D6, 71G2 or only vehicle (PBS). The fifth control group did not receive TNBS or antibodies and was used as a healthy control. Mice were killed 5 days after TNBS administration. At autopsy, the colon was collected and measured. After measurement, the colon was embedded in paraffin and processed for histological analysis. Throughout the experiment, the weight of mice was measured every day. (A) Changes in body weight over time (percentage relative to time 0). (B) Colon length at autopsy.

[0093] Fig.21 : Inflammatory bowel disease model: histological analysis of colon sections. Fig. 20 As described in the figure legends, colon injury and inflammation were induced in BALB / c mice by rectal injection of 2,4,6-trinitrobenzenesulfonic acid (TNBS). At necropsy, the colon was collected and measured. After measurement, the colon was embedded in paraffin and processed for histological analysis. Colon sections were stained with hematoxylin and eosin, examined by microscopy, and photographed. See the main text for image analysis.

[0094] Fig. 22:Type I diabetes model: promotion of glucose uptake and combination with insulin in diabetic mice. Pancreatic β-cell degeneration was induced in BALB / c mice by intraperitoneal injection of streptozotocin (STZ). STZ-treated mice showed twice the average basal blood glucose compared with untreated mice. STZ-treated mice were randomly divided into 4 large groups, which were treated with 71G3, 71D6, 71G2 or vehicle (PBS) alone. The fifth control group did not receive STZ or antibodies and was used as a healthy control. Blood glucose concentrations under fasting conditions were monitored at any time for 5 weeks. At the end of the 5th week, glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed. (A) Analysis of basal blood glucose levels over time under fasting conditions. (B) GTT: After oral administration of glucose to fasted animals, blood glucose levels were monitored over time. (C) ITT: After intraperitoneal injection of insulin to partially fasted animals, blood glucose levels were monitored over time.

[0095] Fig.23 : Type I diabetes model: promotion of glucose uptake and association with insulin in cultured cells. C2C12 mouse myoblasts were induced to differentiate into myoblasts and then incubated with human / mouse equivalent agonist anti-MET antibodies (71G3, 71D6, 71G2). After 24 hours, the antibody-treated cells were divided into 3 large groups, which were subjected to acute stimulation with 0nM, 100nM or 1000nM human recombinant insulin for 1 hour in the presence of the fluorescent glucose analog 2-(N-7-nitro-2,1,3-benzoxadiazole-4-amino)-2-deoxy-D-glucose (2-NBDG). 2-NBDG uptake was determined by flow cytometry. (A) Induction of 2-NBDG uptake by human / mouse equivalent agonist anti-MET antibodies or insulin. (B) Induction of 2-NBDG uptake by 71G3 in the absence or presence of insulin. (C) Induction of 2-NBDG uptake by 71D6 in the absence or presence of insulin. (D) Induction of 2-NBDG uptake by 71G2 in the absence or presence of insulin.

[0096] Fig.24: Type II diabetes model: Normalization of blood glucose levels and overcoming insulin resistance in db / db mice. At 8 weeks of age, female db / db mice (C57BLKS / J variant with a point mutation in the leptin receptor gene lepr) were randomly divided into 4 groups, which received treatment with 71G3, 71D6, 71G2 or vehicle only (PBS). The antibodies were administered by intraperitoneal injection twice a week at a dose of 1 mg / kg. Blood glucose concentrations under fasting conditions were monitored every 10 days for 7 weeks. At the end of the treatment, i.e. when the mice were 15 weeks of age, glucose tolerance tests (GTT) and insulin tolerance tests (ITT) were performed using age-matched wild-type C57BLKS / J mice as controls. (A) Changes in blood glucose concentration over time. (B) GTT: Blood glucose levels were monitored over time after oral administration of glucose to fasted animals. (C) ITT: Blood glucose levels were monitored over time after intraperitoneal injection of insulin to partially fasted animals.

[0097] Fig.25 : Mouse model of non-alcoholic steatohepatitis (NASH): Improvement of fatty liver determined by histology. Female db / db mice aged 8 weeks were randomly divided into 4 groups, which were treated with 71G3, 71D6, 71G2 or vehicle (PBS) alone. The antibody was administered by intraperitoneal injection at a dose of 1 mg / kg twice a week. After 8 weeks of treatment, the mice were killed and autopsied. Blood was collected for analysis of liver function markers. The liver was extracted, embedded in paraffin and processed for histological examination. Liver sections were stained with hematoxylin and eosin. The cytoplasm of the adipocytes appears empty and white because the lipids have been washed away during the alcohol treatment of the samples. Representative images of each treatment group are shown. Magnification: 200 times.

[0098] Fig.26 : Mouse model of nonalcoholic steatohepatitis (NASH): inhibition of fibrosis as measured by Picro Sirius red staining. Fig.25 Eight-week-old female db / db mice were randomized and treated as described in the figure legends. At necropsy, livers were processed for histological examination. Liver sections were stained with Sirius red to highlight fibrosis. Representative images of each treatment group are shown. Magnification: 200x.

[0099] Fig. 27 : Mouse model of nonalcoholic steatohepatitis (NASH): normalization of liver function markers. Fig.25As described in the legend, 8-week-old female db / db mice were treated with purified 71G3, 71D6, 71G2 or vehicle alone. After 7 weeks of treatment, blood was collected for analysis of liver function markers. (A) Plasma levels of aspartate aminotransferase (AST). (B) Plasma levels of alanine aminotransferase (ALT).

[0100] Fig.28 : Mouse model of diabetic ulcer: accelerated healing of wounds. 8-week-old db / db diabetic mice were anesthetized and then cut with a 0.8 cm wide circular perforated blade for skin biopsy to produce a circular wound on the right rear side. The entire epidermis was removed. On the second day after surgery, mice were randomly divided into 4 large groups, which received treatment with purified 71G3, 71D6 and 71G2 or only with a carrier (PBS). The antibody was delivered every other day by intraperitoneal injection at a dose of 5 mg / kg. Calipers were used to measure the wound diameter every day. (A) Changes in wound area over time. (B) Average re-epithelialization rate determined by averaging the daily percentage of wound closure.

[0101] Fig.29 : Cross-reactivity with Rattus norvegicus and Macaca fascicularis determined by ELISA. To test pan-species cross-reactivity, a restricted antibody panel representing SEMA conjugates (71D6, 71C3, 71D4, 71A3, 71G2) and PSI conjugates (76H10, 71G3) was selected. 5D5 prior art antibody was used as a control. Human, mouse, rat or monkey MET ECD was fixed in a solid phase and exposed to increasing concentrations of mAb (in its human IgG1 / λ form) in solution. Binding was revealed by using an HRP-conjugated anti-human Fc antibody.

[0102] Fig.30: Amino acid sequence alignment between MET ECD domains from H. sapiens, M. musculus, R. norvegicu, M. fascicularis, and L. glama. (A) Sequence alignment relative to the region recognized by SEMA binding antibodies (71D6, 71C3, 71D4, 71A3, and 71G2) (human MET sequence SEQ ID NO: 239; mouse MET sequence SEQ ID NO: 240; rat MET sequence SEQ ID NO: 241; cynomolgus MET sequence SEQ ID NO: 242; Llama MET sequence SEQ ID NO: 243). Amino acids recognized by the human-llama chimera approach shown in Table 12 are underlined. In this region, there are 5 residues (Ala 327, Ser 336, Phe 343, Ile 367, Asp 372) that are conserved in human and mouse MET but not in alpaca MET. These amino acids are represented by black boxes and progressive numbers 1-5. Among them, 4 residues (Ala 327, Ser 336, Ile 367, Asp 372) are also conserved in rat and cynomolgus monkey MET. The amino acids responsible for binding to the SEMA binding antibody are represented by "S" (for SEMA). The amino acids responsible for binding to 5D5 / Onartuzumab are represented by "O" (for Onartuzumab). (B) Sequence alignment relative to the region recognized by PSI binding antibodies 76H10 and 71G3 (human MET sequence SEQ ID NO: 244; mouse MET sequence SEQ ID NO: 245; rat MET sequence SEQ ID NO: 246; cynomolgus monkey MET sequence SEQ ID NO: 247; llama MET sequence SEQ ID NO: 248). The amino acids recognized by the human-llama chimera method shown in Table 12 are underlined. Within this region, there are three residues (Arg 547, Ser 553, Thr 555) that are conserved in human and mouse MET but not in llama MET. These amino acids are represented by black boxes and progressive numbers 6-8. Among them, two residues (Ser 553 and Thr 555) are also conserved in rat and cynomolgus monkey MET. The amino acids responsible for binding to the PSI binding antibodies are represented by "P" (for PSI).

[0103] Fig.31 : Schematic diagram of MET mutants used for fine epitope mapping. Using human MET ECD as a template, Fig.30The key residues indicated by the progressive numbers 1-8 in were mutagenized in different arrangements to generate mutants AL. Each of these mutants is completely human except for the indicated residues (which are llama). DETAILED DESCRIPTION

[0104] As used herein, the term "immunoglobulin" includes polypeptides having a combination of two heavy chains and two light chains, whether or not they have any relevant specific immunoreactivity. "Antibody" refers to an assembly with significant known specific immunoreactivity to an antigen of interest (e.g., MET). The term "MET antibody" or "anti-MET antibody" herein refers to an antibody that is immunologically specific to the MET protein. Antibodies and immunoglobulins contain light chains and heavy chains, with or without interchain covalent bonds between them. The basic immunoglobulin structure in vertebrate systems is relatively easy to understand.

[0105] The general term "immunoglobulin" includes five different classes of antibodies that can be distinguished biochemically. Although all five classes of antibodies are within the scope of the present invention, the following discussion generally relates to the IgG class of immunoglobulin molecules. With respect to IgG, immunoglobulins contain two identical light polypeptide chains of molecular weight of approximately 23,000 Daltons and two identical heavy chains of molecular weight of 53,000-70,000. The four chains are connected by disulfide bonds in a "Y" configuration, wherein the light chains bracket the heavy chains starting at the mouth of the "Y" and continuing through the variable region.

[0106] The light chain of an antibody is classified as κ or λ (κ, λ). Each heavy chain class can be combined with a κ or λ light chain. Typically, the light chain and the heavy chain are covalently bound to each other, and when a B cell or a genetically engineered host cell produces an immunoglobulin, the "tail" portion of the two heavy chains is interconnected by a covalent disulfide bond or a non-covalent bond. In the heavy chain, the amino acid sequence extends from the N-terminal at the bifurcated end of the Y configuration to the C-terminal at the bottom of each chain. It will be appreciated by those skilled in the art that the heavy chain is classified as γ, μ, α, δ or ε (γ, μ, α, δ, ε), with some subclasses (e.g., γ1-γ4). It is the nature of the chain that determines the "class" of the antibody as IgG, IgM, IgA, IgD or IgE, respectively. Immunoglobulin subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, etc., are well characterized and known to confer functional specialization. Modified forms of each of these classes and isotypes are readily discernible to one skilled in the art in view of the instant disclosure and, therefore, fall within the scope of the instant invention.

[0107] As mentioned above, the variable region of an antibody allows the antibody to selectively recognize and specifically bind to an epitope on an antigen. That is, the VL domain and VH domain of an antibody combine to form a variable region that defines a three-dimensional antigen binding site. This quaternary antibody structure forms an antigen binding site present at the end of each arm of the Y. More specifically, the antigen binding site is defined by three complementary determining regions (CDRs) on each of the VH chain and the VL chain.

[0108] As used herein, the terms "MET protein" or "MET antigen" or "MET" are used interchangeably and refer to a receptor tyrosine kinase that binds hepatocyte growth factor (HGF) in wild-type form. The terms "human MET protein" or "human MET receptor" or "human MET" or "hMET" are used interchangeably to refer to human MET (GenBank Accession No.: X54559), including native human MET proteins naturally expressed in human hosts and / or on the surface of human cultured cell lines, as well as recombinant forms and fragments thereof, and naturally occurring mutant forms. The terms "mouse MET protein" or "mouse MET receptor" or "mouse MET" or "mMET" are used interchangeably to refer to mouse MET (GenBank Accession No.: NM_008591), including native mouse MET proteins naturally expressed in mouse hosts and / or on the surface of mouse cultured cell lines, as well as recombinant forms and fragments thereof, and naturally occurring mutant forms.

[0109] As used herein, the term "binding site" comprises a region of a polypeptide that is responsible for selective binding to a target antigen of interest (e.g., hMET). A binding domain comprises at least one binding site. Exemplary binding domains include antibody variable domains. Antibody molecules of the invention may comprise a single binding site or multiple (e.g., two, three, or four) binding sites.

[0110] As used herein, the term "derived from" a specified protein (e.g., a MET antibody or antigen-binding fragment thereof) refers to the source of the polypeptide. In one embodiment, the polypeptide or amino acid sequence derived from a specific starting polypeptide is a CDR sequence or a sequence related thereto. In one embodiment, the amino acid sequence derived from a specific starting polypeptide is not continuous. For example, in one embodiment, one, two, three, four, five or six CDRs are derived from a starting antibody. In one embodiment, a polypeptide or amino acid sequence derived from a specific starting polypeptide or amino acid sequence has an amino acid sequence that is substantially identical to the amino acid sequence of the starting sequence or a portion thereof (wherein, the portion is composed of at least 3-5 amino acids, at least 5-10 amino acids, at least 10-20 amino acids, at least 20-30 amino acids, or at least 30-50 amino acids), or is recognizable by a person of ordinary skill in the art as originating from the starting sequence. In one embodiment, one or more CDR sequences derived from the starting antibody are changed to produce a variant CDR sequence, e.g., an affinity variant, wherein the variant CDR sequence maintains MET binding activity.

[0111] "Camelid-derived" - In certain preferred embodiments, the MET antibody molecule of the present invention comprises a framework amino acid sequence and / or CDR amino acid sequence derived from a conventional camelid antibody produced by active immunization of a camelid with a MET-derived antigen. However, the MET antibody comprising an amino acid sequence of camelid origin can be engineered to comprise a framework and / or constant region sequence derived from a human amino acid sequence (i.e., a human antibody) or other non-camelid mammalian species. For example, a human or non-human primate framework region, a heavy chain portion, and / or a hinge portion may be included in the subject MET antibody. In one embodiment, one or more non-camelid amino acids may be present in the framework region of a "camelid-derived" MET antibody, for example, a camelid framework amino acid sequence may comprise one or more amino acid mutations in which corresponding human or non-human primate amino acid residues are present. In addition, as described extensively elsewhere herein, the VH and VL domains of camelid origin or humanized variants thereof can be linked to the constant domains of a human antibody to produce a chimeric molecule.

[0112] As used herein, "conservative amino acid substitution" refers to an amino acid substitution in which an amino acid residue is replaced by an amino acid residue with a similar side chain. Families of amino acid residues with similar side chains have been defined in the art, including basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), non-polar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), β-branched side chains (e.g., threonine, valine, isoleucine) and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine). Therefore, a non-essential amino acid residue in an immunoglobulin polypeptide can be replaced by another amino acid residue from the same side chain family. In another embodiment, a string of amino acids can be replaced by a string of amino acids that are structurally similar but different in order and / or composition of the side chain family members.

[0113] As used herein, the term "heavy chain portion" includes an amino acid sequence derived from an immunoglobulin heavy chain constant domain. The polypeptide comprising the heavy chain portion includes at least one of the following: a CH1 domain, a hinge (e.g., an upper, middle and / or lower hinge region) domain, a CH2 domain, a CH3 domain or a variant or fragment thereof. In one embodiment, the antibody or antigen-binding fragment of the present invention may include the Fc portion of an immunoglobulin heavy chain (e.g., a hinge portion, a CH2 domain and a CH3 domain). In another embodiment, the antibody or antigen-binding fragment of the present invention may lack at least a portion of the constant domain (e.g., all or part of the CH2 domain). In certain embodiments, at least one and preferably all constant domains are derived from a human immunoglobulin heavy chain. For example, in a preferred embodiment, the heavy chain portion includes a fully human hinge domain. In other preferred embodiments, the heavy chain portion includes a fully human Fc portion (e.g., hinge, CH2 and CH3 domain sequences from human immunoglobulins). In certain embodiments, the constituent constant domains of the heavy chain portion are from different immunoglobulin molecules. For example, the heavy chain portion of the polypeptide may include a CH2 domain derived from an IgG1 molecule and a hinge region derived from an IgG3 or IgG4 molecule. In other embodiments, the constant domain is a chimeric domain comprising different immunoglobulin molecule portions. For example, the hinge may include a first portion from an IgG1 molecule and a second portion from an IgG3 or IgG4 molecule. As described above, it will be appreciated by those of ordinary skill in the art that the constant domains of the heavy chain portion may be modified to be different from naturally occurring (wild-type) immunoglobulin molecules in terms of amino acid sequence. That is, the polypeptides of the present invention disclosed herein may include changes or modifications to one or more heavy chain constant domains (CH1, hinge, CH2 or CH3) and / or light chain constant regions (CL). Exemplary modifications include adding, deleting or replacing one or more amino acids in one or more domains.

[0114] As used herein, a "chimeric" protein comprises a first amino acid sequence linked to a second amino acid sequence (not naturally linked in nature). The amino acid sequences may normally be present in separate proteins that are brought together in a fusion polypeptide, or they may normally be present in the same protein, but in a new arrangement in the fusion polypeptide. For example, a chimeric protein may be produced by chemical synthesis, or by generating and translating a polynucleotide encoding the peptide regions in a desired relationship. Exemplary chimeric MET antibodies include fusion proteins comprising VH and VL domains of camelid origin fused to the constant domains of a human antibody (e.g., human IgG1, IgG2, IgG3, or IgG4) or fused to the constant domains of a mouse antibody (e.g., mouse IgG1, IgG2a, IgG2b, IgG2c, or IgG3), or humanized variants thereof.

[0115] As used herein, the terms "variable region" and "variable domain" are interchangeable and are intended to have equivalent meanings. The term "variable" refers to: certain parts of the variable domains VH and VL are very different in sequence between antibodies, and are used for the binding and specificity of each specific antibody to its target antigen. However, variability is not evenly distributed throughout the variable domains of the antibody. It is concentrated in three segments called "hypervariable loops" in each VL domain and VH domain, which form a part of the antigen binding site. The first, second and third hypervariable loops of the Vλ light chain domain are referred to herein as L1 (λ), L2 (λ) and L3 (λ), and can be defined as comprising the 24th-33rd residues (L1 (λ), consisting of 9, 10 or 11 amino acid residues), the 49th-53rd residues (L2 (λ), consisting of 3 residues) and the 90th-96th residues (L3 (λ), consisting of 5 residues) (Morea et al., Methods 20, 267-279, 2000). The first, second and third hypervariable loops of the Vκ light chain domain are referred to herein as L1 (κ), L2 (κ) and L3 (κ), and can be defined as comprising residues 25-33 (L1 (κ), consisting of 6, 7, 8, 11, 12 or 13 residues), residues 49-53 (L2 (κ), consisting of 3 residues) and residues 90-97 (L3 (κ), consisting of 6 residues) in the VL domain (Morea et al., Methods 20, 267-279, 2000). The first, second and third hypervariable loops of the VH domain are referred to herein as H1, H2 and H3, and can be defined as comprising residues 25-33 (H1, consisting of 7, 8 or 9 residues), residues 52-56 (H2, consisting of 3 or 4 residues) and residues 91-105 (H3, highly variable in length) in the VH domain (Morea et al., Methods 20, 267-279, 2000).

[0116] Unless otherwise indicated, the terms L1, L2 and L3 refer to the first, second and third hypervariable loops of the VL domain, respectively, and encompass hypervariable loops obtained from Vκ and Vλ isotypes. The terms H1, H2 and H3 refer to the first, second and third hypervariable loops of the VH domain, respectively, and encompass hypervariable loops obtained from any known heavy chain isotype, including γ, ε, δ, α or μ.

[0117] Hypervariable loops L1, L2, L3, H1, H2 and H3 may each comprise a portion of a "complementarity determining region" or "CDR" as defined below. The terms "hypervariable loop" and "complementarity determining region" are not strictly synonymous, because hypervariable loops (HV) are defined based on structure, while complementarity determining regions (CDRs) are defined based on sequence variability (Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991), and the limits of HV and CDR may be different in some VH and VL domains.

[0118] The CDRs of VL and VH domains can generally be defined as comprising the following amino acids: residues 24-34 (CDRL1), residues 50-56 (CDRL2), and residues 89-97 (CDRL3) in the light chain variable domain; and residues 31-35 or 31-35b (CDRH1), residues 50-65 (CDRH2), and residues 95-102 (CDRH3) in the heavy chain variable region (Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991). Thus, HV can be contained within the corresponding CDRs, and references herein to the "hypervariable loops" of VH and VL domains should be interpreted as also including the corresponding CDRs, and vice versa, unless otherwise indicated.

[0119] The more highly conserved parts of the variable domain are referred to as the framework regions (FRs) as defined below. The variable domains of native heavy and light chains each contain four FRs (FR1, FR2, FR3, and FR4, respectively), which are mainly in a β-folded configuration connected by three hypervariable loops. The hypervariable loops in each chain are kept together in close proximity by the FRs, and contribute to the formation of the antigen binding site of the antibody by the hypervariable loops from the other chain. The structural analysis of antibodies reveals the relationship between the sequence and the shape of the binding site formed by the complementary determining region (Chothia et al., Journal of Molecular Biology, 227, pp. 799-817, 1992 (Chothia et al., J.Mol.Biol.227, 799-817, 1992); Tramontano et al., Journal of Molecular Biology, 215, pp. 175-182, 1990 (Tramontano et al., J.Mol.Biol, 215, 175-182, 1990)). Despite their high sequence variability, five of the six loops adopt only a small set of main-chain conformations, referred to as “canonical structures.” These conformations are determined first by the length of the loop and second by the presence of key residues at certain positions in the loop and in the framework regions, which determine the conformation they fill, hydrogen bond, or the ability to assume unusual main-chain conformations.

[0120] As used herein, the term "CDR" or "complementarity determining region" refers to the non-contiguous antigen binding sites found in the variable regions of both heavy and light chain polypeptides. These specific regions have been described by Kabat et al., J. Biol. Chem. 252, 6609-6616, 1977; Kabat et al., Immunol. Protein Sequences, 5th Ed., Public Health Service, National Institutes of Health, Bethesda, MD, 1991; Chothia et al., J. Mol. Biol. 196, 901-917, 1987; and MacCallum et al., J. Mol. Biol. 262, 732-745, 1996. al., J. Mol. Biol. 262, 732-745, 1996), wherein the definitions include overlaps or subsets of amino acid residues when compared to each other. The amino acid residues covering the CDRs defined by each of the above references are listed for comparison. Preferably, the term "CDR" is a CDR defined by Kabat based on sequence comparison.

[0121] Table 1: CDR definitions.

[0122]

[0123] 1 The residue numbering follows the nomenclature of Kabat et al., supra.

[0124] 2 Residue numbering follows the nomenclature of Chothia et al., supra.

[0125] 3 Residue numbering follows the nomenclature of MacCallum et al., supra.

[0126] As used herein, the term "framework region" or "FR region" includes amino acid residues that are part of the variable region but not part of the CDR (e.g., using the Kabat definition of CDR). Thus, the length of the variable region framework is between about 100-120 amino acids, but includes only those amino acids outside the CDR. For the specific example of the heavy chain variable domain and for the CDRs defined by Kabat et al., framework region 1 corresponds to the domain of the variable region containing amino acids 1-30; framework region 2 corresponds to the domain of the variable region containing amino acids 36-49; framework region 3 corresponds to the domain of the variable region containing amino acids 66-94; and framework region 4 corresponds to the domain of the variable region from amino acid 103 to the end of the variable region. The framework region of the light chain is similarly separated by each light chain variable region CDR. Similarly, by using the CDR definitions of Chothia et al. or McCallum et al., the framework region boundaries are separated by the ends of the corresponding CDRs as described above. In a preferred embodiment, the CDRs are as defined by Kabat.

[0127] In naturally occurring antibodies, the six CDRs present in each monomer antibody are short, non-continuous amino acid sequences that are specifically positioned to form an antigen binding site because the antibody presents its three-dimensional configuration in an aqueous environment. The rest of the heavy and light chain variable domains show less intermolecular variability in the amino acid sequence and are referred to as framework regions. The framework region mainly adopts a β-folded conformation and the CDR forms a loop that connects the β-folded structure and forms a part of the β-folded structure in some cases. Therefore, these framework regions play a role in forming a scaffold that is provided for positioning the six CDRs in the correct direction by interchain non-covalent interactions. The antigen binding site formed by the positioned CDRs defines a surface complementary to the epitope on the immunoreactive antigen. The complementary surface promotes the non-covalent binding of the antibody to the immunoreactive antigen epitope. One of ordinary skill in the art can easily identify the position of the CDR.

[0128] As used herein, the term "hinge region" includes the portion of the heavy chain molecule that connects the CH1 domain to the CH2 domain. The hinge region contains about 25 residues and is flexible, thus allowing the two N-terminal antigen binding regions to move independently. The hinge region can be subdivided into three different domains: upper, middle and lower hinge domains (Roux et al., J. Immunol. 161, 4083-4090, 1998). The MET antibody comprising a "full human" hinge region may contain one of the hinge region sequences shown in Table 2 below.

[0129] Table 2: Human hinge sequences

[0130]

[0131] As used herein, the term "CH2 domain" comprises a portion of a heavy chain molecule that, for example, extends from about residue 244 to residue 360 ​​of an antibody using a conventional numbering scheme (Kabat numbering system, residue 244 to residue 360; EU numbering system, residues 231-340; Kabat et al., Immunology Protein Sequences, 5th Edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991). The uniqueness of the CH2 domain is that it is not tightly paired with another domain. Instead, two N-connected branched carbohydrate chains are inserted between the two CH2 domains of a complete native IgG molecule. It has also been well documented that the CH3 domain extends from the CH2 domain to the C-terminus of the IgG molecule and comprises about 108 residues.

[0132] As used herein, the term "fragment" refers to a portion or part of an antibody or antibody chain that contains fewer amino acid residues than a complete or complete antibody or antibody chain. The term "antigen-binding fragment" refers to a polypeptide fragment of an immunoglobulin or antibody that binds to an antigen or competes with a complete antibody for antigen binding (i.e., a complete antibody derived from them) (i.e., specifically binds to hMET and MMET). As used herein, the term "fragment" of an antibody molecule includes an antigen-binding fragment of an antibody, such as an antibody light chain variable domain (VL), an antibody heavy chain variable domain (VH), a single-chain antibody (scFv), a F(ab')2 fragment, a Fab fragment, a Fd fragment, a Fv fragment, and a single-domain antibody fragment (DAb). Fragments can be obtained, for example, by chemical or enzymatic treatment of a complete or complete antibody or antibody chain or by a recombinant method.

[0133] As used herein, the term "valency" refers to the number of potential target binding sites in a polypeptide. Each target binding site specifically binds to one target molecule or specific site on a target molecule. When a polypeptide comprises more than one target binding site, each target binding site may specifically bind to the same or different molecules (e.g., may bind to different ligands or different antigens, or different epitopes on the same antigen). The subject binding molecule has at least one binding site specific for hMET.

[0134] As used herein, the term "specificity" refers to the ability to bind (e.g., immune response) to a given target (e.g., hMET, mMET). A polypeptide can be monospecific and contain one or more binding sites that specifically bind to a target; or a polypeptide can be multispecific and contain two or more binding sites that specifically bind to the same or different targets. In one embodiment, the antibodies of the invention are specific for more than one target. For example, in one embodiment, the multispecific binding molecules of the invention bind hMET and a second target molecule. In this article, the second target molecule is a molecule other than hMET or mMET.

[0135] The term "epitope" refers to the portion of an antigen (eg, human MET) that contacts an antibody. An epitope may be linear, i.e., involving binding to a single amino acid sequence; or may be conformational, i.e., involving binding to two or more amino acid sequences in various regions of the antigen, which may not necessarily be continuous. The antibodies provided herein may bind to different (overlapping or non-overlapping) epitopes within the extracellular domain of the human MET protein.

[0136] As used herein, the term "synthetic" with respect to polypeptides includes polypeptides containing non-naturally occurring amino acid sequences. For example, non-naturally occurring polypeptides are modified forms of naturally occurring polypeptides (e.g., containing mutations such as additions, substitutions or deletions), or contain a first amino acid sequence (which may or may not be naturally occurring) linked in a linear sequence of amino acids to a second amino acid sequence (which may or may not be naturally occurring) (which is not naturally linked in nature).

[0137] As used herein, the term "engineering" includes the manipulation of nucleic acid or polypeptide molecules by synthetic means (e.g., by recombinant techniques, in vitro peptide synthesis, by enzymatic or chemical coupling of peptides, or some combination of these techniques). Preferably, the antibodies of the present invention are engineered, including, for example, humanized antibodies and / or chimeric antibodies, and antibodies that have been engineered to improve one or more properties such as antigen binding, stability / half-life or effector function.

[0138] As used herein, the term "modified antibody" includes: synthetic forms of antibodies that are altered so that they do not exist naturally, for example, antibodies that include at least two heavy chain portions but do not include two complete heavy chains (e.g., domain-deleted antibodies or mini-antibodies); multispecific forms of antibodies (e.g., bispecific, trispecific, etc.) that are altered to bind to two or more different antigens or different epitopes on a single antigen; heavy chain molecules connected to scFv molecules, etc. ScFv molecules are known in the art and are described, for example, in U.S. Pat. No. 5,892,019. In addition, the term "modified antibody" includes multivalent forms of antibodies (e.g., trivalent antibodies, tetravalent antibodies, etc., antibodies that bind to three or more copies of the same antigen). In another embodiment, the modified antibody of the present invention is a fusion protein that includes at least one heavy chain portion lacking a CH2 domain and includes a binding domain of a polypeptide that includes a binding portion of one member of a receptor ligand pair.

[0139] The term "modified antibody" may also be used herein to refer to amino acid sequence variants of the MET antibodies of the present invention. One of ordinary skill in the art will appreciate that the MET antibodies of the present invention may be modified to produce variant MET antibodies that differ in amino acid sequence from the MET antibodies from which they were derived. For example, nucleotide or amino acid substitutions that result in conservative substitutions or changes at "non-essential" amino acid residues may be made (e.g., in CDRs and / or framework residues). Amino acid substitutions may include substitutions of one or more amino acids with naturally occurring or non-natural amino acids.

[0140] As used herein, the term "humanizing substitution" refers to an amino acid substitution in which an amino acid residue present at a specific position in the VH or VL domain of a MET antibody of the invention (e.g., a MET antibody of camelid origin) is replaced by an amino acid residue present at an equivalent position in a reference human VH or VL domain. The reference human VH or VL domain may be a VH or VL domain encoded by a human germline. Humanizing substitutions may be made in the framework regions and / or CDRs of a MET antibody as defined herein.

[0141] As used herein, the term "humanized variant" refers to a variant antibody that contains one or more "humanizing substitutions" compared to a reference MET antibody, wherein a portion of the reference antibody (e.g., the VH domain and / or VL domain or a portion thereof containing at least one CDR) has an amino acid sequence derived from a non-human species, and the "humanizing substitutions" occur within the amino acid sequence derived from the non-human species.

[0142] The term "germline variant" is used herein to specifically refer to a "humanized variant", wherein the "humanized substitution" results in the amino acid residue present at the equivalent position in the reference human VH or VL domain encoded by the human germline replacing one or more amino acid residues present at a specific position in the VH or VL domain of the MET antibody of the present invention (e.g., a MET antibody of camelid origin). Typically, for any given "germline variant", the substituted amino acid residues substituted into the germline variant are completely or mainly from the VH or VL domain encoded by a single human germline. The terms "humanized variant" and "germline variant" are generally used interchangeably herein. One or more "humanized substitutions" are introduced into a VH or VL domain of camelid origin (e.g., llama origin) resulting in a "humanized variant" of a VH or VL domain of camelid (llama) origin. If the substituted amino acid residues are mainly or completely derived from a VH or VL domain sequence encoded by a single human germline, the result can be a "human germline variant" of a VH or VL domain of camelid (llama) origin.

[0143] As used herein, the term "affinity variant" refers to a variant antibody that exhibits one or more changes in the amino acid sequence compared to a reference MET antibody of the invention, wherein the affinity variant exhibits an altered affinity for hMET and / or mMET compared to the reference antibody. Preferably, the affinity variant exhibits an improved affinity for hMET and / or mMET compared to the reference MET antibody. A lower K for hMET and / or for mMET D Or for hMET and / or for mMET slower dissociation rate, this improvement can be obvious. Compared with the reference MET antibody, affinity variants generally show one or more changes in the amino acid sequence in the CDR. Such substitutions can result in different amino acid residues (which can be naturally occurring amino acid residues or non-naturally occurring amino acid residues) replacing the original amino acid present at a given position in the CDR. Amino acid substitutions can be conservative or non-conservative.

[0144] As used herein, an antibody with "high human homology" refers to an antibody comprising a heavy chain variable domain (VH) and a light chain variable domain (VL) that together exhibit at least 90% amino acid sequence identity to the closest matching human germline VH and VL sequences. Antibodies with high human homology may include antibodies containing the VH and VL domains of natural non-human antibodies that exhibit a sufficiently high percentage of sequence identity to human germline sequences, including, for example, antibodies comprising the VH and VL domains of conventional antibodies of camelids, and engineered, particularly humanized or germlined variants of such antibodies as well as "fully human" antibodies.

[0145] In one embodiment, the VH domain of an antibody with high human homology may exhibit 80% or more amino acid sequence identity or sequence homology with one or more human VH domains in the framework regions FR1, FR2, FR3, and FR4. In other embodiments, the amino acid sequence identity or sequence homology between the VH domain of a polypeptide of the invention and the closest matching human germline VH domain sequence may be 85% or more, 90% or more, 95% or more, 97% or more, or up to 99% or even 100%.

[0146] In one embodiment, the VH domain of an antibody with high human homology may contain one or more (eg, 1 to 10) amino acid sequence mismatches in the framework regions FR1, FR2, FR3, and FR4 compared to the closest matching human VH.

[0147] In another embodiment, the VL domain of the antibody with high human homology can show 80% or more sequence identity or sequence homology with one or more human VL domains in the framework regions FR1, FR2, FR3 and FR4. In other embodiments, the amino acid sequence identity or sequence homology between the VL domain of the polypeptide of the invention and the closest matching human germline VL domain sequence can be 85% or more, 90% or more, 95% or more, 97% or more, or up to 99% or even 100%.

[0148] In one embodiment, the VL domain of an antibody with high human homology may contain one or more (e.g., 1 to 10) amino acid sequence mismatches in the framework regions FR1, FR2, FR3, and FR4 compared to the closest matching human VL sequence. Before analyzing the percent sequence identity between an antibody with high human homology and human germline VH and VL, a canonical fold can be determined, which allows identification of a human germline segment family with the same combination of canonical folds of H1 and H2 or L1 and L2 (and L3). Subsequently, a human germline family member with the highest degree of sequence homology to the variable region of the antibody of interest is selected for scoring of sequence homology. Methods for determining the closest matching human germline and determining % sequence identity / homology are well known to those skilled in the art.

[0149] Antibodies with high human homology may comprise hypervariable loops or CDRs with human or human-like canonical folding structures. In one embodiment, at least one hypervariable loop or CDR in a VH domain or VL domain of an antibody with high human homology may be obtained or derived from a VH or VL domain of a non-human antibody, such as a conventional antibody from a species of the family Camelidae, but exhibits a predicted or actual canonical folding structure that is substantially the same as that occurring in human antibodies. In one embodiment, both H1 and H2 in the VH domain of an antibody with high human homology exhibit a predicted or actual canonical folding structure that is substantially the same as that occurring in human antibodies.

[0150] An antibody with high human homology may comprise a VH domain in which the hypervariable loops H1 and H2 form a combination of canonical folding structures that are identical to the combination of canonical structures present in at least one known human germline VH domain. It has been observed that only certain combinations of canonical folding structures of H1 and H2 are actually present in VH domains encoded by human germlines. In one embodiment, H1 and H2 in the VH domain of an antibody with high human homology may be obtained from a non-human species, such as a VH domain of a camelid species, but form a combination of predicted or actual canonical folding structures that are identical to the combination of canonical folding structures present in a human germline or somatically mutated VH domain. In a non-limiting embodiment, H1 and H2 in the VH domain of an antibody with high human homology may be obtained from a non-human species, such as a VH domain of a camelid species, and form one of the following canonical folding combinations: 1-1, 1-2, 1-3, 1-6, 1-4, 2-1, 3-1, and 3-5. An antibody with high human homology may contain a VH domain that exhibits high sequence identity / sequence homology to human VH and contains hypervariable loops that exhibit structural homology to human VH.

[0151] It may be advantageous in terms of overall primary amino acid sequence identity that the canonical folds and combinations thereof present at H1 and H2 in the VH domain of an antibody with high human homology are "correct" for the human VH germline sequence that represents the closest match to the VH domain of an antibody with high human homology. For example, if the closest sequence match is to a human germline VH3 domain, it may be advantageous that H1 and H2 form a combination of canonical folds that also naturally occur in human VH3 domains. This may be particularly important in the case of antibodies with high human homology derived from non-human species, such as antibodies containing VH and VL domains derived from conventional antibodies of Camelids, and especially antibodies containing humanized Camelid VH and VL domains.

[0152] Thus, in one embodiment, the VH domain of a MET antibody with high human homology may exhibit 80% or more, 85% or more, 90% or more, 95% or more, 97% or more, up to 99% or even 100% sequence identity or sequence homology with a human VH domain in the framework regions FR1, FR2, FR3 and FR4; and H1 and H2 in the same antibody are obtained from non-human VH domains (e.g., derived from Camelidae species, preferably Ilama), but form a combination of predicted or actual canonical folding structures that is the same as the known canonical folding combination naturally occurring in the same human VH domain.

[0153] In other embodiments, L1 and L2 in the VL domain of an antibody with high human homology are both obtained from a VL domain of a non-human species (e.g., a VL domain of camelid origin), and both present a predicted or actual canonical folding structure that is substantially identical to the canonical folding structure present in a human antibody. L1 and L2 in the VL domain of an antibody with high human homology may form a combination of predicted or actual canonical folding structures that is identical to the combination of canonical folding structures present in known human germline VL domains. In a non-limiting embodiment, L1 and L2 in the Vλ domain of an antibody with high human homology (e.g., an antibody containing a VL domain of camelid origin or a humanized variant thereof) may form one of the following canonical folding combinations: 11-7, 13-7 (A, B, C), 14-7 (A, B), 12-11, 14-11, and 12-12 (as defined in Williams et al., J. Mol. Biol. 264, 220-232, 1996, and as shown at http: / / www.bioc.uzh.ch / antibody / Sequences / Germlines / VBase_hVL.html). In a non-limiting embodiment, L1 and L2 in the Vκ domain can form one of the following canonical folding combinations: 2-1, 3-1, 4-1 and 6-1 (as defined in Tomlinson et al., EMBO J. 14, 4628-4638, 1995, and as shown in http: / / www.bioc.uzh.ch / antibody / Sequences / Germlines / VBase_hVK.html). In another embodiment, L1, L2 and L3 in the VL domain of an antibody with high human homology can all exhibit a substantially human structure. Preferably, the VL domain of an antibody with high human homology exhibits high sequence identity / sequence homology with a human VL, and the hypervariable loops in the VL domain exhibit structural homology with a human VL.

[0154] In one embodiment, the VL domain of a MET antibody with high human homology can exhibit 80% or more, 85% or more, 90% or more, 95% or more, 97% or more, or up to 99% or even 100% sequence identity with the human VL domain in the framework regions FR1, FR2, FR3 and FR4; and the hypervariable loop L1 and the hypervariable loop L2 can form a combination of predicted or actual canonical folding structures that is the same as the canonical folding combination naturally occurring in the known same human VL domain.

[0155] Of course, it is conceivable that a VH domain showing high sequence identity / sequence homology to a human VH and structural homology to the hypervariable loops of a human VH would be combined with a VL domain showing high sequence identity / sequence homology to a human VL and structural homology to the hypervariable loops of a human VL, to provide an antibody containing a high human homology of a VH / VL pairing (e.g. a VH / VL pairing of camelid origin) and having maximum sequence and structural homology to a human-encoded VH / VL pairing.

[0156] Methods for assessing the presence of a human-like canonical fold structure in a camelid-derived (eg, llama-derived) CDR, VH domain or VL domain are described in WO2010 / 001251 and WO2011 / 080350, the entire contents of which are incorporated herein by reference.

[0157] As used herein, the term "affinity" or "binding affinity" should be understood based on the usual meaning in the art in the context of antibody binding, and reflects the strength and / or stability of the binding between an antigen and a binding site on an antibody or antigen-binding fragment thereof.

[0158] The anti-MET antibodies provided herein are characterized by high affinity binding to human MET (hMET) and high affinity binding to mouse MET (mMET). Binding affinity to hMET and mMET can be assessed using standard techniques known to those skilled in the art.

[0159] In one embodiment, surface plasmon resonance can be used, for example using Biacore TM The binding affinity of Fab clones containing defined VH / VL pairs is evaluated by Biacore. Fab clones containing VH / VL pairs of antibodies and antigen-binding fragments of the invention generally exhibit TM Measured as 1×10 -3 Up to 1×10 -2 s -1 , optionally 1×10 -3 Up to 6×10 -3 s -1 Dissociation rates within this range can be considered as an indication that the Fab and the corresponding bivalent mAb exhibit high affinity binding to hMET. Similarly, as described in the accompanying Examples, Fab clones comprising VH / VL pairings of antibodies and antigen-binding fragments of the invention generally exhibit high affinity binding to hMET as determined by Biacore. TM Measured as 1×10 -3 Up to 1×10 -2 s -1 , optionally 1×10 -3 Up to 6×10 -3 s -1The mMET dissociation rate of 100 μg / mL is 100 μg / mL. The dissociation rate within this range can be regarded as an indication that the Fab and the corresponding bivalent mAb show high affinity binding to mMET. Therefore, the Fab showing the dissociation rate of human and mouse MET falling within the range shows high affinity binding to hMET, and high affinity binding to mMET, that is, Fab is cross-reactive between hMET and mMET. (Individually) a bivalent mAb comprising two Fabs showing the dissociation rate of human and mouse MET falling within the range is also considered to show high affinity binding to human MET and high affinity binding to mouse MET.

[0160] Binding affinity can also be expressed as the dissociation constant, or K, for a specific antibody. D . K D The smaller the value, the stronger the binding interaction between the antibody and its target antigen. For example, the K 结合 and K 解离 The rate determines K D In general, when measured as mAbs, the antibodies and antigen-binding fragments of the invention exhibit a K of 0.05 for mMET and 0.06 for hMET. D Less than 0.1nMol / L.

[0161] Binding affinity to human and murine MET can also be assessed using a cell-based system as described in the accompanying Examples, in which mAbs are tested for binding to mammalian cell lines expressing MET using, for example, ELISA or flow cytometry. High affinity to hMET or mMET can be detected, for example, by EC in the ELISA described in Example 3. 50 Not greater than 0.5 nM is indicated.

[0162] As described above, the present invention relates, at least in part, to antibodies and antigen-binding fragments thereof that bind hMET and mMET with high affinity. The properties and characteristics of the MET antibodies and antibody fragments according to the present invention will now be described in more detail.

[0163] The high affinity hMET and mMET cross-reactive antibodies and antigen-binding fragments described herein are MET agonists. As used herein, MET agonists induce MET signaling (partially or completely) when binding to the MET receptor. MET agonist antibodies and antigen-binding fragments according to the present invention are agonists of hMET and mMET. The agonist activity of the antibodies described herein for hMET or mMET binding can be indicated by (at least partially) mimicking the molecular and / or cellular responses induced when homologous HGF-MET binds (i.e., hMET binding to human HGF, mMET binding to mouse HGF). Antibodies that stimulate such responses are also referred to herein as "anti-MET agonists," "agonist antibodies," and grammatical variants thereof. Similarly, antibodies that partially or completely stimulate such responses are referred to herein as "partial MET agonists" or "partial agonists," or "full MET agonists" or "full agonists," respectively. It should be emphasized that the antibodies and antigen-binding fragments of the present invention induce MET signaling in human and mouse systems, i.e., they are agonists of hMET and mMET. Thus, the following discussion applies both to responses induced by binding of the antibodies and antigen-binding fragments of the invention to hMET and to responses induced by binding of the antibodies and antigen-binding fragments of the invention to mMET.

[0164] The agonism of MET by the antibodies and antigen-binding fragments of the invention can be indicated by molecular responses such as phosphorylation of the MET receptor and / or cellular responses such as those detectable in cell scattering assays, anti-apoptosis assays, and / or branching morphogenesis assays. These molecular and cellular responses are further described as follows:

[0165] (i) Phosphorylation of the MET receptor. In this case, when the binding of the antibody or antigen binding fragment causes autophosphorylation of MET in the absence of receptor-ligand binding (i.e., in the absence of hHGF, the antibody or antigen binding fragment binds to human hMET to cause phosphorylation of hMET and in the absence of mHGF, the antibody or antigen binding fragment binds to mMET to cause phosphorylation of mMET), the MET agonist antibody or antigen binding fragment phosphorylates MET. Phosphorylation of MET can be determined by detection known in the art such as Western blotting or phosphorylation-MET ELISA (as described in Example 6, and as described in Basilico et al., Journal of Clinical Investigation, 124, 3172-3186, 2014 (Basilico et al., J Clin Invest. 124, 3172-3186, 2014)). The antibodies and antigen-binding fragments described herein may exhibit "high phosphorylation efficiency" or "low phosphorylation efficiency" for hMET, and may exhibit "high phosphorylation efficiency" or "low phosphorylation efficiency" for mMET. In this case, when the antibody or fragment exhibits EC 50 Similar to HGF (<1 nM) and / or E MAX was at least 80% (expressed as a percentage of maximal HGF-induced activation) for mMET and exhibited an EC 50 Similar to HGF (<1 nM) and / or E MAX An antibody or antigen-binding fragment exhibits "high phosphorylation potency" when it has a potency for hMET of at least 80% (expressed as a percentage of maximal HGF-induced activation). 50 1nM-5nM and / or E MAX 60-80% (expressed as a percentage of maximal HGF-induced activation) of mMET and exhibited an EC 50 1nM-5nM and / or E MAX An antibody or antigen-binding fragment exhibits "low phosphorylation potency" when the potency towards hMET is 60-80% (expressed as a percentage of maximal HGF-induced activation).

[0166] (ii) Inducing HGF-like cellular responses. MET agonism can be measured using assays such as cell scattering assays, anti-apoptosis assays, and / or branching morphogenesis assays as described in this example. In this context, a MET agonist antibody or antigen-binding fragment according to the present invention induces a response in a cell assay, such as a response similar to (at least in part) that observed after exposure to homologous HGF. For example, a MET agonist can be indicated as: an increase in cell scattering in response to an antibody compared to cells exposed to a control antibody (e.g., IgG1); an increase in EC 10 expression compared to untreated cells; 50Less than 32 nM and / or E max Protection against drug-induced apoptosis with cell viability greater than 20%; and / or an increase in the number of branches per spheroid in a spheroid preparation exposed to the antibody or antigen-binding fragment.

[0167] Depending on the assay used, the antibodies and antigen-binding fragments described herein can "fully induce" or "partially induce" an HGF-like cellular response when contacted with human cells, and can "fully induce" or "partially induce" an HGF-like cellular response when contacted with mouse cells.

[0168] In this case, "complete induction" of an HGF-like cellular response by an antibody or fragment can be measured as:

[0169] In the cell scattering assay, when the antibody concentration is 0.1-1 nM, the antibody or antigen-binding fragment induces an increase in cell scattering at least equivalent to 0.1 nM homologous HGF;

[0170] In anti-apoptosis assays, antibodies or antigen-binding fragments show EC 50 EC no greater than HGF 50 1.1 times and / or E max Cell survival was greater than that observed for HGF. max 90%; and / or

[0171] In the branching morphogenesis assay, cells treated with the antibody or antigen-binding fragment exhibited greater than 90% of the number of branches per spheroid induced by the same (non-zero) concentration of HGF.

[0172] In this case, if the antibody or antigen-binding fragment does not "fully induce" an HGF-like cellular response as defined above, then "partial induction" of an HGF-like cellular response can be measured as:

[0173] In a cell scattering assay, the antibody or antigen-binding fragment induces at least 25% of the level of cell scattering induced by 0.1 nM homologous HGF when the antibody concentration is 1 nM or less;

[0174] In anti-apoptosis assays, antibodies or antigen-binding fragments show EC 50 EC no greater than HGF 50 7.0 times and / or E max Cell viability was at least 50% of that observed for HGF;

[0175] In a branching morphogenesis assay, cells treated with the antibody or antigen-binding fragment exhibit at least 25% of the number of branches per spheroid induced by the same (non-zero) concentration of HGF.

[0176] As already described, the antibodies and antigen-binding fragments according to the present invention are hMET agonists and mMET agonists. Thus, in embodiments in which the antibody (partially or completely) induces an HGF-like cellular response, the HGF-like cellular response is (partially or completely) induced when the antibody or antigen-binding fragment is in contact with human cells and is (partially or completely) induced when the antibody or antigen-binding fragment is in contact with mouse cells.

[0177] Binding region localization (Example 4) demonstrates that the anti-MET antibodies of the present invention recognize an epitope of MET in the PSI domain of MET or the SEMA domain of MET. Therefore, in certain embodiments, the antibodies or antigen-binding fragments of the present invention recognize an epitope in the PSI domain of MET (preferably human MET). In certain alternative embodiments, the antibodies or antigen-binding fragments of the present invention recognize an epitope in the SEMA domain of MET (preferably human MET).

[0178] In certain embodiments, an antibody or antigen-binding fragment that recognizes an epitope in a SEMA domain recognizes an epitope located on a blade of the SEMA β-propeller. In certain embodiments, the epitope is located on blade 4 or 5 of the SEMA β-propeller. In certain such embodiments, the epitope is located between amino acids 314-372 of human MET. In certain embodiments, the epitope is located on blades 1-4 or 1-3 of the SEMA β-propeller. In certain embodiments, the epitope is located between amino acids 27-313 of human MET, or between amino acids 27-225 of human MET.

[0179] In certain embodiments, an antibody or antigen-binding fragment that recognizes an epitope in the PSI domain of MET recognizes an epitope located between amino acids 516-545 of MET (preferably human MET). In certain embodiments, an antibody or antigen-binding fragment that recognizes an epitope in the PSI domain of MET recognizes an epitope located between amino acids 546-562 of MET (preferably human MET).

[0180] In certain aspects, the antibodies described herein recognize an epitope in the extracellular domain of MET, the epitope comprising one or more amino acid residues conserved in human and mouse MET. In a preferred embodiment, the antibodies described herein recognize an epitope in the extracellular domain of MET, the epitope comprising one or more amino acid residues conserved in human MET, mouse MET, rat MET and monkey (e.g., cynomolgus monkey) MET.

[0181] In certain embodiments, the antibodies of the present invention recognize an epitope of human MET located in the region from amino acid residue 123 to residue 223 of human MET. In certain embodiments, the antibodies of the present invention recognize an epitope of human MET located in the region from amino acid residue 224 to residue 311 of human MET. In certain embodiments, the antibodies of the present invention recognize an epitope of human MET located in the region from amino acid residue 314 to residue 372 of human MET. In certain embodiments, the antibodies of the present invention recognize an epitope of human MET located in the region from amino acid residue 546 to residue 562 of human MET.

[0182] In certain embodiments, the antibodies or antigen-binding fragments of the present invention recognize an epitope of human MET comprising amino acid residue Ile367. In certain embodiments, the antibodies or antigen-binding fragments of the present invention recognize an epitope of human MET comprising amino acid residue Asp372 of human MET. In certain embodiments, the antibodies or antigen-binding fragments of the present invention recognize an epitope of human MET comprising amino acid residues Ile367 and Asp372.

[0183] In certain such embodiments, the antibodies or antigen-binding fragments of the invention recognize an epitope of human MET located in the region from amino acid residue 314 to residue 372 of human MET, wherein the epitope comprises amino acid residue Ile367. In certain such embodiments, the antibodies or antigen-binding fragments of the invention recognize an epitope of human MET located in the region from amino acid residue 314 to residue 372 of human MET, wherein the epitope comprises amino acid residue Asp371. In certain such embodiments, the antibodies or antigen-binding fragments of the invention recognize an epitope of human MET located in the region from amino acid residue 314 to residue 372 of human MET, wherein the epitope comprises amino acid residues Ile367 and Asp372.

[0184] In certain embodiments, the antibodies or antigen-binding fragments of the invention bind to an epitope of human MET comprising amino acid residue Thr555 of human MET.

[0185] In certain such embodiments, the antibodies or antigen-binding fragments of the invention recognize an epitope of human MET located in the region from amino acid residue 546 to residue 562 of human MET, wherein the epitope comprises amino acid residue Thr555.

[0186] It should be understood that antibodies or their antigen-binding fragments can recognize epitopes consisting of many amino acid residues. Epitopes can be linear, conformational, or combined. When an epitope is specified as being in a certain region of amino acids, the epitope can be formed by one or more amino acids in the region in contact with the antibody or fragment. Therefore, it should be understood that in certain embodiments of the present invention, antibodies or their fragments can recognize epitopes consisting of multiple amino acid residues (continuous or non-continuous) in a specified region (e.g., amino acids 314-372, or amino acids 546 to 562), provided that the epitope recognized includes specified amino acid residues (e.g., Ile367, Asp372, Thr555). Methods for determining residues identified as part of an antibody epitope are familiar to those skilled in the art, and include, for example, those described in Examples 4 and 26.

[0187] Because the anti-MET antibodies and antigen-binding fragments of the present invention bind to epitopes that overlap or are proximal to the binding domain recognized by HGF, the antibodies and antigen-binding fragments are able to (at least partially) compete with HGF for binding to homologous MET (i.e., compete with human HGF for hMET binding, and compete with mouse HGF for mMET binding). That is, the antibody or antigen-binding fragment directly or indirectly prevents HGF from binding to homologous MET in a binding assay (e.g., the ELISA described in Example 5). Therefore, in certain embodiments, the MET antibodies and antigen-binding fragments of the present invention compete with mouse and human HGF for binding to homologous MET. Antibodies or antigen-binding fragments that compete with HGF in this manner are also referred to herein as "HGF competitors." Assays to determine whether an antibody or antigen-binding fragment competes with HGF for MET binding are well known to those skilled in the art. For example, in a competition ELISA, an HGF competitor may exhibit an IC of no greater than 5 nM. 50 and / or at least 50% of I max (Maximum percent competition at saturation.) The antibodies and antigen-binding fragments of the invention compete with mouse HGF for mMET binding, and with human HGF for hMET binding.

[0188] The antibodies or antigen-binding fragments of the invention can "completely compete" or "partially compete" with HGF for cognate MET binding. In this context, a "complete competitor" can be an agent that exhibits an IC of less than 2 nM in a competition assay such as an ELISA. 50 and / or at least 90% in I max In certain embodiments, a "perfect competitor" exhibits an IC of less than 1 nM. 50 and / or greater than 90% I max A "partial competitor" may be a compound that exhibits an IC in a competition assay such as an ELISA. 50 2-5 nM and / or I maxThe given values ​​apply to competition with mouse HGF and human HGF for binding to homologous MET.

[0189] As already described, the antibodies and antigen-binding fragments of the present invention are advantageous due to their ability to recognize both human and mouse MET. The antibodies or antigen-binding fragments thereof described herein are particularly advantageous when they exhibit the same properties when binding to mMET and hMET. This equivalence allows the analysis of antibodies in preclinical murine disease models, with the expectation that the antibodies will exhibit the same or similar properties in a human setting.

[0190] Thus, in certain embodiments, the antibodies and binding fragments of the invention exhibit equivalent binding affinities for hMET and mMET. In this context, "equivalent binding affinity" is considered to mean that the affinity of the antibody or antigen binding fragment for hMET is 0.5-1.5 times greater than the affinity of the antibody for mMET. In certain embodiments, the antibodies and antigen binding fragments of the invention exhibit an affinity for hMET that is 0.8-1.2 times greater than the affinity of the antibody or antigen binding fragment for mMET.

[0191] By way of illustration and example, an antibody or antigen-binding fragment having equivalent affinity for mMET and hMET may exhibit an off-rate for hMET that is 0.5-1.5 times greater than the off-rate exhibited for mMET when measured as a Fab fragment. For example, an antibody having equivalent affinity for mMET and hMET exhibits an off-rate for mMET of 2.6×10 -3 s -1 , and showed a dissociation rate of 1.3-3.9×10 -3 s -1 By way of further example, an antibody or antigen-binding fragment having equivalent affinity for mMET and hMET may exhibit an EC for hMET. 50 The EC of the antibody or fragment for mMET (e.g., as determined by ELISA or flow cytometry) is 50 For example, an antibody with equivalent affinity for mMET and hMET exhibits an EC for mMET of 50 was 0.1 nMol / L and showed an EC 50 0.05-0.15nMol / L.

[0192] In certain embodiments, the antibodies and antigen-binding fragments of the present invention are equivalent agonists of mMET and hMET. In this context, "equivalent" means that the level of MET agonism induced when binding to hMET is 0.5-1.5 times the level of signaling induced when binding to mMET. In certain embodiments, the antibodies and antigen-binding fragments of the present invention induce MET signaling when binding to hMET, and the level of signaling is 0.8-1.2 times the level of signaling induced when binding to mMET.

[0193] In certain embodiments, the antibodies or antigen-binding fragments of the invention are equivalent mMET and hMET agonists when measured by at least one MET agonism assay described herein. For example, the antibodies or antigen-binding fragments of the invention can induce equivalent MET phosphorylation, exhibit equivalent protective efficacy against drug-induced apoptosis, and / or induce equivalent levels of branching in a branching morphogenesis assay. In certain embodiments, the antibodies or antigen-binding fragments exhibit equivalent MET agonism when measured by all of the assays described.

[0194] As an illustration, equivalent phosphorylation of MET by the antibody of the present invention can be detected as follows: 50 EC for mMET 50 For example, if the EC 50 The EC value for hMET is 2.9 nM. 50 If the concentration of the antibody is 1.45-4.35 nM, the antibody will equivalently induce hMET phosphorylation. Similarly, equivalent MET activation indicated in the anti-apoptotic assay can be detected as E in human cells. max It is E in mouse cells max For example, if the E max The E of human cells is 37.5%. max If the concentration of 18.75-56.25% of the antibody is 18.75-56.25%, then the antibody will be an equivalent hMET agonist. Equivalent MET agonism indicated in a branching morphogenesis assay can be detected as 0.5-1.5 times the number of branches observed after exposure of human cell spheroids to the antibody compared to the number of branches observed after exposure of mouse cell spheroids to the same (non-zero) antibody concentration. For example, if the number of branches exhibited by mouse cells after exposure to 0.5 nM of the antibody is 14, and the number of branches exhibited by human cells after exposure to 0.5 nM of the antibody is 7-21, then the antibody will be an equivalent hMET agonist.

[0195] Similarly, equivalent agonism of hMET and mMET may be indicated by equivalent cell scattering. The nature of the output of this assay means that the use of a 0.5-1.5 factor is inappropriate. In a cell scattering assay, equivalent agonism of hMET and mMET may be indicated by a cell scattering score of + / -1 for human cells exposed to the antibody compared to the cell scattering score of mouse cells exposed to the same (non-zero) concentration of the same antibody. For example, if mouse cells exposed to 0.33 nM of an antibody exhibit a cell scattering score of 2, and human cells exposed to 0.33 nM of the same antibody exhibit a cell scattering score of 1-3, then the antibody would be an equivalent agonist of hHGF.

[0196] In certain embodiments, the antibodies and antigen-binding fragments of the invention exhibit equivalent HGF competition between mMET and hMET. In this case, "equivalent HGF competition" is considered to mean that the level of competition with human HGF exhibited by the antibody or antigen-binding fragment for hMET is 0.5-1.5 times the level of competition with mouse HGF exhibited by the antibody or antigen-binding fragment for mMET. In certain embodiments, the level of competition with human HGF exhibited by the antibodies and antigen-binding fragments of the invention for mMET is 0.8-1.2 times the level of competition with mouse HGF exhibited by the antibody or antigen-binding fragment for mMET.

[0197] As an example, equivalent competition of an antibody with human HGF and mouse HGF can be measured as the IC of an antibody competing with human HGF-hMET binding. 50 IC of the antibody that competes with mouse HGF-mMET binding 50 For example, if the IC of mHGF-mMET binding 50 The IC of hHGF-hMET binding is 0.34 nM. 50 At 0.17-0.51 nM, the antibody competed equally with hHGF and mHGF.

[0198] In certain embodiments, the antibodies and antigen-binding fragments of the invention cross-react with rat MET and / or macaque MET. Cross-reactivity with one or both of rat and macaque MET has the advantage that toxicology studies can be performed in rat and / or macaque model systems. In this regard, whether an antibody exhibits cross-reactivity with cynomolgus monkey or rat MET can be determined by ELISA as described in the accompanying Example 25.

[0199] The antibodies or antigen-binding fragments thereof described herein may comprise at least one hypervariable loop or complementarity determining region obtained from a VH domain or VL domain of a species of Camelidae. In particular, the antibodies or antigen-binding fragments may comprise VH and / or VL domains or CDRs thereof obtained by active immunization of outbred Camelidae (e.g., Ilama) with a human MET antigen.

[0200] "Hypervariable loops or complementary determining regions obtained from the VH or VL domains of a Camelidae species" refers to hypervariable loops (HV) or CDRs having an amino acid sequence identical or substantially identical to an amino acid sequence of a hypervariable loop or CDR encoded by a Camelidae immunoglobulin gene. In this article, "immunoglobulin genes" include germline genes, immunoglobulin genes undergoing rearrangement, and somatic mutation genes. Therefore, the amino acid sequence of the HV or CDR obtained from the VH or VL domains of a Camelidae species may be identical to the amino acid sequence of the HV or CDR present in a mature Camelidae conventional antibody. In this context, the term "obtained from / obtained from" means the following structural relationship: the HV or CDR of the MET antibody comprises an amino acid sequence (or a smaller variant thereof) originally encoded by a Camelidae immunoglobulin gene. However, with respect to the production method for preparing the MET antibody, this does not necessarily mean a specific relationship.

[0201] Camelid-derived MET antibodies may be derived from any Camelid species, including, inter alia, llama, dromedary, alpaca, llama, guanaco or camel.

[0202] MET antibodies comprising VH and VL domains or their CDRs of camelid origin are generally recombinantly expressed polypeptides and may be chimeric polypeptides. The term "chimeric polypeptide" refers to an artificial (non-naturally occurring) polypeptide that is produced by juxtaposing two or more peptide fragments that do not appear consecutively. This definition includes "species" chimeric polypeptides produced by juxtaposing peptide fragments encoded by two or more species (e.g., camelids and humans).

[0203] The camelid-derived CDR may comprise one of the CDR sequences shown in Tables 3 and 4 below.

[0204] In one embodiment, the entire VH domain and / or the entire VL domain may be obtained from a species of the family Camelidae. In a specific embodiment, the VH domain of camelid origin may comprise the amino acid sequence shown in SEQ ID NO: 155, 157, 159, 161, 163, 165, 167, 169, 171, 173 or 175, and the VL domain of camelid origin may comprise the amino acid sequence shown in SEQ ID NO: 156, 158, 160, 162, 164, 166, 168, 170, 172, 174 or 176. The VH domain of camelid origin and / or the VL domain of camelid origin may then be subjected to protein engineering to introduce one or more amino acid substitutions, insertions or deletions into the amino acid sequence of the camelid. These engineered changes preferably comprise amino acid substitutions relative to the camelid sequence. These changes include "humanization" or "germlining", in which one or more amino acid residues in a camelid-encoded VH or VL domain are replaced by the equivalent residue from the homologous human-encoded VH or VL domain.

[0205] The isolated camelid VH and VL domains obtained by active immunization of camelids (e.g., alpacas) with human MET antigens can be used as the basis for engineering MET antibodies according to the present invention. Starting from the complete camelid VH and VL domains, one or more amino acid substitutions, insertions, or deletions that deviate from the starting camelid sequence can be engineered. In certain embodiments, such substitutions, insertions, or deletions may be present in the framework regions of the VH domain and / or VL domain. The purpose of such primary amino acid sequence changes can be to reduce possible unfavorable properties (e.g., immunogenicity in human hosts (so-called humanization), potential product heterogeneity and / or instability (glycosylation, deamidation, isomerization, etc.) sites, or to enhance some other favorable properties of the molecule (e.g., solubility, stability, bioavailability, etc.)). In other embodiments, primary amino acid sequence changes can be engineered in one or more hypervariable loops (or CDRs) of the camelid VH and / or VL domains obtained by active immunization. Such changes may be introduced to enhance antigen binding affinity and / or specificity, or to reduce possible unfavorable properties, for example, immunogenicity in human hosts (so-called humanization), sites of potential product heterogeneity and / or instability, glycosylation, deamidation, isomerization, etc., or to enhance some other favorable property of the molecule, such as solubility, stability, bioavailability, etc.

[0206] Therefore, in one embodiment, the present invention provides a variant MET antibody, which contains at least one amino acid substitution in at least one framework or CDR region of the VH domain or VL domain compared to a VH or VL domain from a camelid, examples of which include but are not limited to: a camelid VH domain comprising the amino acid sequence shown in SEQ ID NO: 155, 157, 159, 161, 163, 165, 167, 169, 171, 173 or 175 and a camelid VL domain comprising the amino acid sequence shown in SEQ ID NO: 156, 158, 160, 162, 164, 166, 168, 170, 172, 174 or 176.

[0207] In certain embodiments, a "chimeric" antibody molecule is provided, which comprises VH and VL domains (or engineered variants thereof) of camelid origin and one or more constant domains from non-camelid antibodies, such as human-encoded constant domains (or engineered variants thereof). In such embodiments, it is preferred that both the VH domain and the VL domain are obtained from the same species of camelids, for example, both VH and VL can be derived from alpacas (before introducing engineered amino acid sequence variations). In such embodiments, both the VH and VL domains can be derived from a single animal, particularly a single animal that has been actively immunized with a human MET antigen.

[0208] In certain embodiments, the present invention may include chimeric camelid / human antibodies, particularly chimeric antibodies in which the VH and VL domains are all-camelid sequences (e.g., llama or alpaca) and the remainder of the antibody is all-human. MET antibodies may include antibodies comprising "humanized" or "germlined" variants of camelid-derived VH and VL domains or their CDRs, as well as camelid / human chimeric antibodies in which the VH and VL domains contain one or more amino acid substitutions in the framework region compared to the camelid VH and VL domains obtained by active immunization of camelids with human MET antigens. This "humanization" increases the percentage of sequence identity with human germline VH or VL domains by replacing mismatched amino acid residues in the starting camelid VH or VL domains with equivalent residues found in human germline-encoded VH or VL domains.

[0209] In certain embodiments, the invention may include chimeric camelid / mouse antibodies, particularly chimeric antibodies in which the VH and VL domains are all camelid sequences (eg, llama or alpaca) while the remainder of the antibody is all mouse sequence.

[0210] The MET antibodies and antigen-binding fragments of the present invention may also be CDR-grafted antibodies, wherein CDRs (or hypervariable loops) derived from camelid antibodies, such as camelid MET antibodies produced by active immunization with human MET protein or encoded by camelid genes, are grafted onto human VH and VL frameworks, and the rest of the antibody is also fully human. Such CDR-grafted MET antibodies may contain CDRs having amino acid sequences shown in Tables 3 and 4 below.

[0211] Camelid-derived MET antibodies include variants in which the hypervariable loops or CDRs of the VH domain and / or VL domain are obtained from conventional camelid antibodies produced against human MET, but in which at least one of the (camelid-derived) hypervariable loops or CDRs has been engineered to include one or more amino acid substitutions, additions or deletions relative to the camelid coding sequence. These changes include "humanization" of the hypervariable loops / CDRs. Camelid-derived HV / CDRs that have been engineered in this way may still exhibit an amino acid sequence that is "substantially identical" to the amino acid sequence of the HV / CDR encoded by the camelid. In this case, "substantial identity" may allow no more than one or no more than two amino acid sequence mismatches with the HV / CDR encoded by the camelid. Specific embodiments of MET antibodies may contain humanized variants of the CDR sequences shown in Tables 3 and 4.

[0212] Conventional camelid (e.g., llama) antibodies provide an advantageous starting point for the preparation of antibodies for use as human therapeutics due to the following factors discussed in US 12 / 497,239, which is incorporated herein by reference:

[0213] 1) High percentage of sequence homology between Camelidae VH and VL domains and their human counterparts;

[0214] 2) High structural homology between the CDRs of camelid VH and VL domains and their human counterparts (i.e., human-like canonical fold structures and human-like combinations of canonical folds).

[0215] The camelid (e.g., llama) platform also offers significant advantages in terms of the functional diversity of available MET antibodies.

[0216] MET antibodies comprising camelid VH and / or camelid VL domains for human therapy can be further improved by "humanization" of native camelid VH and VL domains, for example to render them less immunogenic in human hosts. The overall goal of humanization is to produce molecules in which the VH and VL domains exhibit minimal immunogenicity when introduced into human subjects, while retaining the specificity and affinity of the antigen binding site formed by the parental VH and VL domains.

[0217] One approach to humanization (so called "germlining") involves altering the amino acid sequence of a camelid VH or VL domain to bring it closer to the germline sequence of a human VH or VL domain.

[0218] Determining the homology between the camelid VH (or VL) domain and the human VH (or VL) domain is a critical step in the humanization process, both in terms of selecting the camelid amino acid residues (in a given VH or VL domain) to be altered, and in terms of selecting appropriate replacement amino acid residues.

[0219] A method for germlining conventional antibodies from camelids has been developed based on the alignment of a large number of novel camelid VH (and VL) domain sequences, which are generally somatically mutated VH (or VL) domains known to bind to the target antigen, having human germline VH (or VL) sequences, human VH (and VL) consensus sequences, and germline sequence information available for alpacas (llama pacos).

[0220] The method described in WO2011 / 080350 (the contents of which are incorporated herein by reference) can be applied to: (i) select "camelid" amino acid residues for substitution in camelid-derived VH or VL domains or their CDRs, and (ii) select "human" amino acid residues for substitution when humanizing any given camelid VH (or VL) domain. This method can be used to prepare humanized variants of the camelid-derived CDRs having the amino acid sequences shown in Tables 3 and 4, and also for germlining of camelid-derived VH and VL domains having the sequences shown in Table 5.

[0221] MET antibodies can adopt various different embodiments in which VH domains and VL domains are present. The term "antibody" is used in the broadest sense herein, and includes, but is not limited to: monoclonal antibodies (including full-length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), as long as they exhibit appropriate immunological specificity for human MET protein and mouse MET protein. The term "monoclonal antibody" as used herein refers to an antibody obtained from a substantially homogeneous antibody population, that is, except for a small amount of possible naturally occurring mutations, the individual antibodies constituting the population are the same. Monoclonal antibodies are highly specific for a single antigenic site. In addition, in contrast to conventional (polyclonal) antibody preparations that typically include different antibodies for different determinants (epitopes) on an antigen, each monoclonal antibody is directed against a single determinant or epitope on the antigen.

[0222] "Antibody fragments" include a portion of a full-length antibody, typically its antigen binding domain or variable domain. Examples of antibody fragments include: Fab, Fab', F(ab')2, bispecific Fab, and Fv fragments, diabodies, linear antibodies, single-chain antibody molecules, single-chain variable fragments (scFv), and multispecific antibodies formed by antibody fragments (see Holliger and Hudson, Nature Biotechnol. 23: 1126-1136, 2005, the contents of which are incorporated herein by reference).

[0223] In non-limiting embodiments, the MET antibodies provided herein may comprise a CH1 domain and / or a CL domain, the amino acid sequence of which is completely or substantially human. If the MET antibody is used for human therapeutic purposes, it generally causes the entire constant region of the antibody or at least a portion thereof to have a completely or substantially human amino acid sequence. Thus, one or more or any combination of the CH1 domain, hinge region, CH2 domain, CH3 domain, and CL domain (and CH4 domain, if present) may be completely or substantially human in terms of their amino acid sequence. Such antibodies may be any human isotype, such as IgG1.

[0224] Advantageously, the CH1 domain, hinge region, CH2 domain, CH3 domain and CL domain (and CH4 domain, if present) may all or substantially have human amino acid sequences. In the context of the constant region of a humanized or chimeric antibody or antibody fragment, the term "substantially human" refers to an amino acid sequence identity of at least 90%, or at least 92%, or at least 95%, or at least 97%, or at least 99% with a human constant region. The term "human amino acid sequence" herein refers to an amino acid sequence encoded by a human immunoglobulin gene, including germline, rearranged, and somatically mutated genes. Such antibodies may be any human isotype, with human IgG4 and IgG1 being particularly preferred.

[0225] Also provided are MET antibodies comprising constant domains of "human" sequences that have been altered by one or more amino acid additions, deletions or substitutions to the human sequence, except for those embodiments that specifically require the presence of a "fully human" hinge region.

[0226] The presence of a "fully human" hinge region in the MET antibodies of the invention is beneficial both for minimizing immunogenicity and optimizing stability of the antibodies.

[0227] The MET antibodies provided herein can be of any isotype. Antibodies for human therapeutic use are typically of IgA, IgD, IgE, IgG, IgM type, often of IgG type, in which case they may belong to any of the four subclasses of IgG1, IgG2a and IgG2b, IgG3 or IgG4. Within each of these subclasses, one or more amino acid substitutions, insertions or deletions, or other structural modifications are permitted within the Fc portion, for example to enhance or reduce Fc-dependent functions.

[0228] In a non-limiting embodiment, it is contemplated that one or more amino acid substitutions, insertions or deletions may be made in the constant region of the heavy chain and / or light chain, particularly in the Fc region. Amino acid substitutions may result in replacement of the substituted amino acid with a different naturally occurring amino acid or with a non-natural or modified amino acid. Other structural modifications are also allowed, such as changing the glycosylation pattern (e.g., by adding or deleting N- or O-linked glycosylation sites). Depending on the intended use of the MET antibody, it may be necessary to modify the antibody of the invention with respect to its binding properties to the Fc receptor, for example, to modulate effector function.

[0229] In certain embodiments, the MET antibody may comprise an Fc region of a given antibody isotype, such as human IgG1, modified to reduce or substantially eliminate one or more antibody effector functions naturally associated with the antibody isotype. In non-limiting embodiments, the MET antibody may be substantially free of any antibody effector functions. In this context, "antibody effector functions" include one or more or all of antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and antibody-dependent cellular phagocytosis (ADCP).

[0230] The amino acid sequence of the Fc portion of the MET antibody may contain one or more mutations, such as amino acid substitutions, deletions or insertions, which have the effect of reducing one or more antibody effector functions (compared to a wild-type corresponding antibody without the mutation). Several such mutations are known in the field of antibody engineering. Non-limiting examples suitable for inclusion in the MET antibodies described herein include the following mutations in the Fc domain of human IgG4 or human IgG1: N297A, N297Q, LALA (L234A, L235A), AAA (L234A, L235A, G237A) or D265A (amino acid residues numbered according to the EU numbering system in human IgG1).

[0231] Monoclonal antibodies or antigen-binding fragments thereof that "cross-compete" with the MET antibodies disclosed herein are those monoclonal antibodies or antigen-binding fragments thereof that bind to human MET at sites that are identical or overlapping with sites that bind to the MET antibodies of the present invention and bind to mouse MET at sites that are identical or overlapping with sites that bind to the MET antibodies of the present invention. Competitive monoclonal antibodies or antigen-binding fragments thereof can be identified, for example, by antibody competition assays. For example, a sample of purified or partially purified human MET can be bound to a solid support. Then, an antibody compound or antigen-binding fragment thereof of the present invention and a monoclonal antibody or antigen-binding fragment thereof suspected of being able to compete with such an antibody compound of the present invention are added. One of the two molecules is labeled. If the labeled compound and the unlabeled compound bind to separate and discrete sites on MET, the labeled compound will bind to the same level, regardless of whether the suspected competing compound is present. However, if the sites of interaction are identical or overlapping, the unlabeled compound will compete, and the amount of labeled compound bound to the antigen will be reduced. If the unlabeled compound is present in excess, little (if any) labeled compound will bind. For purposes of the present invention, a competing monoclonal antibody or antigen-binding fragment thereof is one that reduces the binding of an antibody compound of the invention to MET by about 50%, about 60%, about 70%, about 80%, about 85%, about 90%, about 95%, or about 99%. Details of procedures for conducting such competitive assays are well known in the art and can be found in Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1988, pp. 567-569, 1988, ISBN 0-87969-314-2. These assays can be quantified using purified antibodies. A standard curve is established by titrating one antibody by itself, i.e., the same antibody is used for labeling and competitor. Titrate the ability of an unlabeled competing monoclonal antibody or its antigen-binding fragment to inhibit the binding of the labeled molecule to the plate. Plot the results and compare the concentrations required to achieve the desired degree of inhibition of binding.

[0232] The present invention also provides: a polynucleotide molecule encoding the MET antibody of the present invention; an expression vector containing a nucleotide sequence encoding the MET antibody of the present invention operably linked to a regulatory sequence, the regulatory sequence allowing the antibody or its antigen-binding fragment to be expressed in a host cell or a cell-free expression system; and a host cell or a cell-free expression system containing such an expression vector.

[0233] Polynucleotide molecules encoding the MET antibodies of the present invention include, for example, recombinant DNA molecules. As used herein, the terms "nucleic acid," "polynucleotide," or "polynucleotide molecule" are used interchangeably and refer to any single-stranded or double-stranded DNA or RNA molecule, and if single-stranded, also to molecules of its complementary sequence. In discussing nucleic acid molecules, the sequence or structure of a particular nucleic acid molecule may be described herein according to the conventional convention of providing a 5' to 3' sequence. In some embodiments of the present invention, a nucleic acid or polynucleotide is "isolated." When applied to nucleic acid molecules, the term refers to a nucleic acid molecule separated from a sequence immediately continuous with it in the naturally occurring genome of an organism from which it originated. For example, an "isolated nucleic acid" may include a DNA molecule inserted into a vector such as a plasmid or viral vector or integrated into the genomic DNA of a prokaryotic or eukaryotic cell or a non-human host organism. When applied to RNA, the term "isolated polynucleotide" refers primarily to an RNA molecule encoded by an isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been purified / isolated from other nucleic acids associated with it in its natural state (ie, in a cell or tissue). An isolated polynucleotide (DNA or RNA) may further represent a molecule produced directly by a biological or synthetic process, and separated from other components present during its production.

[0234] In order to recombinantly produce the MET antibody according to the present invention, a recombinant polynucleotide encoding it is prepared (using standard molecular biology techniques) and inserted into a replicable vector for expression in a selected host cell or cell-free expression system. Suitable host cells can be prokaryotes, yeast cells or higher eukaryotic cells, particularly mammalian cells. Examples of useful mammalian host cell lines are: monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney cell line (293 cells or 293 cells subcloned for growth in suspension culture, Graham et al., Journal of General Virology, 36, 59-74, 1977); baby hamster kidney cells (BHK, ATCC CCL10); Chinese hamster ovary cells / DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA, 77, 4216, 1980); 77:4216, 1980); mouse Sertoli cells (TM4; Mather, Biol. Reprod., 23, 243-252, 1980 (Mather, Biol. Reprod. 23:243-252, 1980); mouse myeloma cells SP2 / 0-AG14 (ATCC CRL 1581; ATCC CRL 8287) or NS0 (HPA Culture Collection No. 85110503); monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human hepatocytes (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al., Annals NY Acad. Sci. 383:44-68, 1982); MRC 5 cells; FS4 cells; and human hepatoma cell line (Hep G2), and DSM's PERC-6 cell line. Expression vectors suitable for each of these host cells are also well known in the art.

[0235] It should be noted that the term "host cell" generally refers to a cultured cell line. All humans into which an expression vector encoding an antigen-binding polypeptide according to the invention has been introduced are specifically excluded from the definition of "host cell".

[0236] In an important aspect, the present invention also provides a method for producing the MET antibody of the present invention, comprising culturing a host cell (or a cell-free expression system) containing a polynucleotide encoding the MET antibody (e.g., an expression vector) under conditions that allow expression of the MET antibody, and recovering the expressed MET antibody. This recombinant expression method can be used for large-scale production of MET antibodies according to the present invention, including monoclonal antibodies for human therapeutic use. Suitable vectors, cell lines, and production methods for large-scale production of recombinant antibodies suitable for in vivo therapeutic use are generally available in the art and are well known to those skilled in the art.

[0237] The MET antibodies provided herein have therapeutic uses, particularly therapeutic treatment of diseases, particularly conditions that benefit from stimulation of MET function, including but not limited to: degenerative diseases, inflammatory diseases, autoimmune diseases, metabolic diseases, transplantation-related diseases, and wound healing. In this regard, the MET antibodies provided herein are examples of a broader class of MET agonists (e.g., HGF) that have therapeutic utility in treating the conditions.

[0238] Hepatocytes express MET and are the main target of HGF, which promotes their proliferation and protects them from apoptosis. MET antibodies that induce MET signaling are shown herein (Examples 16 and 17) to protect hepatocytes in a mouse model of liver damage (acute liver damage and chronic damage). As already described herein, the antibodies of the present invention exhibit equivalent properties in the human system as in the mouse system, and therefore, similar protective effects can be expected in the case of human liver damage. Therefore, in one aspect, the present invention provides a method for treating or preventing liver damage in a human patient, the method comprising administering a therapeutically effective amount of a MET antibody to a patient in need thereof, the MET antibody inducing MET signaling. In certain embodiments, the method is a method for treating or preventing acute liver damage. In certain embodiments, the method is a method for treating or preventing chronic liver damage. In certain embodiments, the antibody is an antibody as described herein.

[0239] Renal epithelial cells express significant levels of MET and are sensitive to HGF stimulation. MET antibodies that induce MET signaling are shown herein (Example 18) to confer protection in a mouse model of acute kidney injury. As already described herein, the antibodies of the present invention exhibit equivalent properties in the human system as in the mouse system, and therefore, similar protective effects can be expected in the case of human kidney injury. Therefore, in one aspect, the present invention provides a method for treating or preventing kidney injury in a human patient, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET antibody that induces MET signaling. In certain embodiments, the method is a method for treating or preventing acute kidney injury. In certain embodiments, the antibody is an antibody as described herein.

[0240] It is also demonstrated herein (Examples 19 and 20) that administration of a MET antibody that induces MET signaling provides effective treatment in a mouse model of inflammatory bowel disease (IBD), such as in ulcerative colitis. Therefore, in one aspect, the present invention provides a method for treating or preventing IBD in a human patient, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET antibody that induces MET signaling. In certain embodiments, the method is a method for treating or preventing ulcerative colitis. In certain embodiments, the antibody is an antibody as described herein.

[0241] It is further demonstrated herein that the administration of MET antibodies that induce MET signaling can restore metabolic function in diabetes, including type I and type II diabetes (Examples 21 and 22). In particular, in the type I diabetes model (Example 21), MET antibodies were shown to promote glucose uptake. In addition, the administration of MET antibodies and insulin produced a synergistic effect on glucose uptake. In the type II diabetes model (Example 22), MET antibodies were shown to normalize glucose control and reduce insulin resistance. Therefore, in one aspect, the present invention provides a method for treating or preventing diabetes in a human patient, the method comprising administering a therapeutically effective amount of a MET antibody to a patient in need, the MET antibody inducing MET signaling. In certain embodiments, the method is a method for treating or preventing type I diabetes. In certain such embodiments, the method further comprises administering insulin to the patient. In certain embodiments, the method is a method for treating type II diabetes. In certain embodiments, the antibody is an antibody as described herein.

[0242] It is further demonstrated herein that administration of a MET antibody that induces MET signaling can reduce the degree of fatty liver in a mouse model of non-alcoholic steatohepatitis (NASH) (Example 23). In particular, the MET antibody can reduce the number of adipocytes and the level of fibrosis. Therefore, in one aspect, the present invention provides a method for treating or preventing NASH in a human patient, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET antibody that induces MET signaling. In certain embodiments, the antibody is an antibody as described herein.

[0243] It is further demonstrated herein that administration of a MET antibody that induces MET signaling can promote wound healing (Example 24). In addition, the MET antibody can promote wound healing in diabetic mice that exhibit impaired wound healing. Therefore, in one aspect, the present invention provides a method for promoting wound healing in a human patient, the method comprising administering to a patient in need thereof a therapeutically effective amount of a MET antibody that induces MET signaling. In certain embodiments, the human patient suffers from diabetes, optionally type I diabetes. In certain embodiments, the antibody is an antibody as described herein.

[0244] Example

[0245] The present invention will now be further understood with reference to the following non-limiting experimental examples.

[0246] Example 1: Immunization of alpacas

[0247] Immunization of llamas and harvesting of peripheral blood lymphocytes (PBL), as well as subsequent RNA extraction and amplification of antibody fragments, were performed as described above (De Haard et al., J. Bact. 187: 4531-4541, 2005). Two adult llamas (Lama glama) were immunized by intramuscular injection of a chimeric protein consisting of the extracellular domain (ECD) of human MET (MET-Fc; R&D Systems) fused to the Fc portion of human IgG1. Each llama received an injection once a week for six weeks, for a total of six injections. Each injection consisted of 0.2 mg of protein in Freund's incomplete adjuvant, which was divided into more than two bifurcations at the neck.

[0248] Blood samples of 10 ml were collected before and after immunization for studying immune response. About one week after the last immunization, 400 ml of blood was collected and PBL was obtained using the Ficoll-Paque method. Total RNA was extracted by the phenol-guanidine thiocyanate method (Chomczynski et al., Analytical Biochemistry, 162 phases, 156-159 pages, 1987 (Chomczynski et al., Anal. Biochem. 162: 156-159, 1987)) and used as a template for random cDNA synthesis by using the SuperScript™ III first-strand synthesis system kit (Life Technologies). As described above, cDNA encoding the VH-CH1 region and VL-CL domains (κ and λ) of Ilama IgG1 were amplified and subcloned into the phagemid vector pCB3 (de Haard et al., J Biol Chem. 274: 18218-18230, 1999). The recombinant phagemids were used to transform E. coli strain TG1 (Netherland Culture Collection of Bacteria) to generate 4 different Fab expression phage libraries (one λ and one κ library for each immunized Ilama). The diversity was 10 8 -10 9 .

[0249] The immune response to the antigen was studied by ELISA. To this end, we obtained the ECD of human MET (UniProtKB#P08581; aa 1-932) and mouse MET (UniProtKB#P16056.1; aa 1-931) by standard protein engineering techniques. Human or mouse MET ECD recombinant proteins were immobilized in a solid phase (100 ng / well, 96-well plate) and exposed to serial dilutions of serum from alpacas before (day 0) or after (day 45) immunization. Binding was revealed using mouse anti-alpaca IgG1 (Daley et al., Clin. Vaccine Immunol. 12, 2005) and HRP-conjugated donkey anti-mouse antibodies (Jackson Laboratories). As Figure 1 As shown, both alpacas displayed an immune response to human MET ECD. Consistent with the notion that the extracellular portion of human MET displays 87% homology to its mouse homolog, a good degree of cross-reactivity was also observed with mouse MET ECD.

[0250] Example 2: Selection and screening of Fabs binding to human and mouse MET.

[0251] According to the standard phage display scheme, phage expressing Fab from the above library is produced.For selection, phage is first adsorbed onto fixed recombinant human MET ECD, washed, and then eluted with trypsin.After selecting two cycles with human MET ECD, mouse MET ECD is used to perform another two cycles in the same way.In parallel, we also selected phages that alternate human MET ECD cycles with mouse MET ECD cycles, a total of four cycles.By merging the phages selected by two methods, they are then used to infect TG1 Escherichia coli.Single colonies are separated and secretion of Fab is induced using IPTG (Fermentas).Bacteria containing the periplasmic part of Fab are collected, and their ability to bind human and mouse MET ECD is tested by surface plasmon resonance (SPR).By using amine coupling in sodium acetate buffer (GE Healthcare), human or mouse MET ECD is fixed on a CM-5 chip.Periplasmic extracts containing Fab are loaded into a BIACORE 3000 device (GE Healthcare) with a flow rate of 30 μl / min. The Fab dissociation rate (k) was measured over a two-minute period. 解离 ). The binding of the Fab to human and mouse MET was further characterized by ELISA using MET ECD in solid phase and crude periplasmic extracts in solution. Because the Fab was engineered with a MYC tag, an HRP-conjugated anti-MYC antibody (ImTec Diagnostics) was used to reveal binding.

[0252] Fabs that bind to human and mouse MET in SPR and ELISA were selected, and their corresponding phages were sequenced (LGC Genomics). Cross-reactive Fab sequences were divided into families based on VH CDR3 sequence length and content. The internal numbering of the VH family is not based on the IMTG (International Immunogenetics Information System) nomenclature. In summary, we can identify 11 different human / mouse cross-reactive Fabs belonging to 8 VH families. The CDR and FR sequences of the heavy chain variable region are shown in Table 3. The CDR and FR sequences of the light chain variable region are shown in Table 4. The complete amino acid sequences of the heavy and light chain variable regions are shown in Table 5. The complete DNA sequences of the heavy and light chain variable regions are shown in Table 6.

[0253] Table 3: Framework region and CDR sequences of the VH domains of Fabs binding to human and mouse MET.

[0254]

[0255]

[0256] Table 4: Framework region and CDR sequences of the VL domains of Fabs binding to human and mouse MET.

[0257]

[0258]

[0259] Table 5: Variable domain amino acid sequences of Fabs that bind human and mouse MET.

[0260]

[0261]

[0262] Table 6: Variable domain nucleotide sequences of Fabs that bind human and mouse MET.

[0263]

[0264]

[0265]

[0266]

[0267]

[0268]

[0269] The various Fab families and their ability to bind human and mouse MET are shown in Table 7.

[0270] Table 7: Fabs that bind to human MET (hMET) and mouse MET (mMET). Fabs were grouped into families based on their VH CDR3 sequences. Binding of Fabs to human and mouse MET ECD was determined by surface plasmon resonance (SPR) and by ELISA. SPR values ​​represent k dissociation (s -1 ). ELISA values ​​represent the optical density (OD) at 450 nm (AU, arbitrary units). Both SPR and ELISA were performed using crude periplasmic extracts. The Fab concentration in the extracts was not determined. Values ​​are the average of three independent measurements.

[0271]

[0272] Example 3: Chimerization of Fab into mAb

[0273] The cDNAs encoding the VH and VL (κ or λ) domains of the selected Fab fragments were engineered into two separate pUPE mammalian expression vectors (U-protein Express), which contained cDNAs encoding CH1, CH2 and CH3 of human IgG1 or human CL (κ or λ). The complete amino acid sequences of the heavy and light chains of the llama-human chimeric antibody are shown in Table 8.

[0274] Table 8: Complete heavy and light chain amino acid sequences of Ilama-human chimeric mAbs that bind human and mouse MET.

[0275]

[0276]

[0277]

[0278]

[0279] The production (by transient transfection of mammalian cells) and purification (by protein A affinity chromatography) of the resulting chimeric llama-human IgG1 molecules were outsourced to U-protein Express. The binding of the chimeric mAbs to MET was determined by ELISA using hMET or mMET ECD in solid phase and increasing concentrations of antibody in solution (0-20 nM). Binding was revealed by using an HRP-conjugated anti-human Fc antibody (Jackson Immuno Research Laboratories). This analysis revealed that all chimeric llama-human antibodies bound human and mouse MET with picomolar affinities, exhibiting EC values ​​of 0.06 nM to 0.3 nM. 50 Although it may be due to partial exposure of epitopes in the immobilized antigen, the binding capacity (E MAX ) varies from antibody to antibody but is similar in human and mouse settings. 50 and E MAX The values ​​are shown in Table 9.

[0280] Table 9: Binding of chimeric mAbs to human and mouse MET determined by ELISA using MET ECD immobilized in solid phase and increasing concentrations of antibody (0-20 nM) in solution. 50 Values ​​are expressed in nMol / L. MAX Values ​​are expressed as optical density (OD) at 450 nm (AU, arbitrary units).

[0281]

[0282] We also analyzed whether chimeric anti-MET antibodies bind to native human and mouse MET in living cells. To this end, increasing concentrations of antibodies (0-100 nM) were incubated with A549 human lung cancer cells (American Type Culture Collection) or MLP29 mouse liver precursor cells (gift from Prof. Enzo Medico, University of Torino, Strada Provinciale 142 km 3.95, Candiolo, Torino, Italy; Medico et al., Mol Biol Cell 7, 495-504, 1996), both of which express physiological levels of MET. Antibody binding to cells was analyzed by flow cytometry using a phycoerythrin-conjugated anti-human IgG1 antibody (eBioscience) and a CyAn ADP analyzer (Beckman Coulter). As positive controls for human MET binding, we used a commercial mouse anti-human MET antibody (R&D Systems) and a phycoerythrin-conjugated anti-mouse IgG1 antibody (eBioscience). As positive controls for mouse MET binding, we used a commercial goat anti-mouse MET antibody (R&D Systems) and a phycoerythrin-conjugated anti-goat IgG1 antibody (eBioscience). All antibodies showed EC 50 The binding activity of 5-nitropropene to human and mouse cells was 0.2-2.5 nM in a dose-dependent manner. Consistent with the data obtained in ELISA, the maximum binding (E MAX ) varied depending on the antibody but was similar in human and mouse cells. These results indicate that the chimeric llama-human antibody recognizes membrane-bound MET in its native conformation in both human and mouse cell systems. 50 and E MAX The values ​​are shown in Table 10.

[0283] Table 10: Binding of chimeric mAbs to human and mouse cells determined by flow cytometry using increasing concentrations of antibody (0-50 nM). 50 Values ​​are expressed in nMol / L. MAX Values ​​are expressed as percentage relative to the control.

[0284]

[0285] Example 4: Receptor region responsible for antibody binding

[0286] In order to locate the receptor region recognized by antibodies that bind to human and mouse MET (hereinafter referred to as human / mouse equivalent anti-MET antibodies), we measured their ability to bind to a group of engineered proteins derived from the generated human MET (Basilico et al., Journal of Molecular Biology, 283, 21267-21227, 2008). This group includes ( Figure 2 ): entire MET ECD (Decoy MET); MET ECD lacking IPT domains 3 and 4 (SEMA-PSI-IPT 1-2); MET ECD lacking IPT domains 1-4 (SEMA-PSI); isolated SEMA domain (SEMA); fragment containing IPT domains 3 and 4 (IPT 3-4). The engineered MET protein was immobilized in a solid phase and exposed to increasing concentrations of chimeric antibodies (0-50 nM) in solution. Binding was revealed using an HRP-conjugated anti-human Fc antibody (Jackson Immuno Research Laboratories). As shown in Table 11, the analysis showed that 7 mAbs recognized epitopes within the SEMA domain, while the other 4 recognized epitopes within the PSI domain.

[0287] Table 11: Human / mouse equivalent anti-MET antibodies and Figure 2 Binding of the set of MET deletion mutants described in . The MET domain responsible for antibody binding is shown in the last column on the right.

[0288]

[0289] To more finely localize the MET region responsible for antibody binding, we took advantage of the absence of cross-reactivity between our antibodies and llama MET (the organism used to produce these immunoglobulins). To this end, we generated a series of llama-human and human-llama chimeric MET proteins spanning the entire MET ECD as described (Basilico et al., Journal of Clinical Investigation, 124, 3172-3186, 2014). The chimeras ( Figure 3) were immobilized in a solid phase and then exposed to increasing concentrations of mAb (0-20 nM) in solution. Binding was revealed using an HRP-conjugated anti-human Fc antibody (Jackson Immuno Research Laboratories). This analysis revealed that five SEMA-binding mAbs (71D6, 71C3, 71D4, 71A3, 71G2) recognized an epitope located between amino acids 314-372 of human MET, corresponding to the region of blades 4-5 of the 7-blade SEMAβ propeller (Stamos et al., EMBO Journal, 23, 2325-2335, 2004). J. 23, 2325-2335, 2004). The other two SEMA binding mAbs (74C8, 72F8) recognize epitopes located between amino acids 123-223 and 224-311, respectively, corresponding to blades 1-3 and 1-4 of the SEMA β-propeller. PSI binding mAbs (76H10, 71G3, 76G7, 71G12) do not appear to show any significant binding to any of the two PSI chimeras. Considering the results presented in Table 11, these antibodies may recognize an epitope located between amino acids 546-562 of human MET. These results are summarized in Table 12.

[0290] Table 12: Localization of epitopes recognized by human / mouse equivalent anti-MET antibodies as determined by ELISA. Figure 3 Human MET ECD (hMET) or llama MET ECD (1MET) as well as llama-human MET chimeric proteins (CH1-7) described in were immobilized on a solid phase and then exposed to increasing concentrations of mAbs.

[0291]

[0292] Example 5: HGF competition assay

[0293] The above analysis suggests that the epitopes recognized by some human / mouse equivalent anti-MET antibodies may overlap with the epitopes involved in HGF binding to MET (Stamos et al., EMBO J. 23, 2325-2335, 2004; Merchant et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013; Basilico et al., J Clin Invest. 124, 3172-3186, 2014). To study along this line, we tested the competition between mAb and HGF by ELISA. Recombinant human and mouse HGF (R&D Systems) were biotinylated at the N-terminus using NHS-LC-biotin (Thermo Scientific). Human or mouse MET-Fc protein (R&D Systems) was immobilized in a solid phase and then exposed to 0.3 nM biotinylated human or mouse HGF in the presence of increasing concentrations (0-120 nM) of antibody. HGF bound to MET was revealed using HRP-conjugated streptavidin (Sigma-Aldrich). As shown in Table 13, this analysis allowed human / mouse equivalent anti-MET mAbs to be divided into two groups: complete HGF competitors (71D6, 71C3, 71D4, 71A3, 71G2) and partial HGF competitors (76H10, 71G3, 76G7, 71G12, 74C8, 72F8).

[0294] Table 13: The ability of human / mouse equivalent anti-MET antibodies to compete with HGF for binding to MET as determined by ELISA. MET-Fc chimeric proteins (human or mouse) were immobilized in a solid phase and then exposed to a fixed concentration of biotinylated HGF (human or mouse) in the presence of increasing concentrations of antibody. HGF bound to MET was revealed using HRP-conjugated streptavidin. Antibody HGF competition is expressed as IC 50 (concentration to achieve 50% competition) and I MAX (Maximum contention percentage reached at saturation).

[0295]

[0296] As a general rule, SEMA binders replace HGF more effectively than PSI binders. In particular, those antibodies that recognize epitopes within blades 4 and 5 of the SEMAβ propeller are the most effective HGF competitors (71D6, 71C3, 71D4, 71A3, 71G2). This observation is consistent with the view that SEMA blade 5 contains a high-affinity binding site for the α chain of HGF (Merchant et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013). The PSI domain has not been shown to be directly involved in HGF, but has been suggested to act as a "hinge" that regulates the adaptability of HGF between the SEMA domain and the IPT region (Basilico et al., J Clin Invest. 124, 3172-3186, 2014). Therefore, mAbs that bind to PSI (76H10, 71G3, 76G7, 71G12) may hinder the binding of HGF to MET by interfering with this process or by steric hindrance, rather than by direct competition with the ligand. Finally, it has been demonstrated that blades 1-3 of the SEMA β propeller are responsible for low affinity binding of the β chain of HGF, which plays a central role in MET activation, but only partially contributes to the HGF-MET binding strength (Stamos et al., EMBO J. 23, 2325-2335, 2004). This may explain why mAbs (74C8, 72F8) that bind to regions of MET are partial competitors of HGF.

[0297] Example 6: MET activation detection

[0298] Due to their bivalent nature, immunoglobulins against receptor tyrosine kinases can exhibit receptor agonist activity, thereby mimicking the effects of natural ligands. To study along this line, we tested the ability of human / mouse equivalent anti-MET antibodies to promote MET autophosphorylation in receptor activation assays. A549 human lung cancer cells and MLP29 mouse liver precursor cells were deprived of serum growth factors for 48 hours and then stimulated with increasing concentrations (0-5nM) of antibodies or recombinant HGF (A549 cells, recombinant human HGF, R&D Systems; MLP29 cells, recombinant mouse HGF, R&D Systems). After 15 minutes of stimulation, the cells were washed twice with ice-cold phosphate-buffered saline (PBS) and then subjected to the lysis described (Longati et al., Oncogene, 9, 49-57, 1994). Protein lysates were resolved by electrophoresis and then analyzed by Western blotting using antibodies specific for the phosphorylated form of human or mouse MET (tyrosine 1234-1235) (Cell Signaling Technology). The same lysates were also analyzed by Western blotting using anti-whole human MET antibodies (Invitrogen) or anti-whole mouse MET antibodies (R&D Systems). This analysis showed that all human / mouse equivalent antibodies exhibited MET agonist activity. Figure 4 As shown, some antibodies promoted MET autophosphorylation to a degree comparable to HGF (71G3, 71D6, 71C3, 71D4, 71A3, 71G2, 74C8). Some other antibodies (76H10, 76G7, 71G12, 72F8) were less effective, and this was particularly evident at lower antibody concentrations. No clear correlation was observed between MET activation activity and HGF competitive activity.

[0299] In order to obtain more quantitative data, the agonistic activity of the antibody is also characterized by phosphorylation-MAT ELISA. For this reason, A549 and MLP29 cells are serum starved as described above, and then stimulated with mAbs of increasing concentrations (0-25nM). Recombinant human (A549) or mouse (MLP29) HGF is used as a control. As described above (Basilico et al., Journal of Clinical Investigation, 124, 3172-3186, 2014 (Basilico et al., J Clin Invest. 124, 3172-3186, 2014)), cells are lysed and phosphorylation-MET levels are determined by ELISA. In short, 96-well plates are coated with mouse anti-human MET antibodies or rat anti-mouse MET antibodies (both from R&D Systems) and then incubated with cell lysates. After washing, captured proteins were incubated with biotin-conjugated anti-phospho-tyrosine antibody (Thermo Fisher), and binding was revealed using HRP-conjugated streptavidin (Sigma-Aldrich).

[0300] The results of this analysis are consistent with the data obtained by Western blot. As shown in Table 14, 71G3, 71D6, 71C3, 71D4, 71A3, 71G2 and 74C8 strongly activated MET, while 76H10, 76G7, 71G12 and 72F8 caused less obvious effects. In any case, all antibodies showed comparable effects in human and mouse cells.

[0301] Table 14: Agonist activity of human / mouse equivalent anti-MET antibodies in human and mouse cells as determined by ELISA. A549 human lung cancer cells and MLP29 mouse liver precursor cells were serum starved and then stimulated with increasing concentrations of mAbs. Recombinant human HGF (hHGF; A549) or mini-HGF (hHGF; A549) were used as controls. Cell lysates were analyzed by ELISA using anti-whole MET antibody for capture and anti-phosphotyrosine antibody for revealing. Agonist activity is expressed as EC 50 (nM) and E MAX (HGF activity percentage).

[0302]

[0303] Example 7: Dispersion Detection

[0304] To evaluate whether the agonistic activity of human / mouse equivalent anti-MET antibodies can be translated into biological activity, we performed scattering assays on human and mouse epithelial cells. To this end, HPAF-II human pancreatic cancer cells (American Type Culture Collection) and MLP29 mouse liver precursor cells were stimulated with increasing concentrations of recombinant HGF (human or mouse; both from R&D Systems) and cell scattering was determined 24 hours later by microscopic examination as previously described (Basilico et al., J Clin Invest. 124, 3172-3186, 2014). This preliminary analysis revealed that HGF-induced cell scattering was linear until it reached saturation at about 0.1 nM in both cell lines. Based on these HGF standard curves, we elaborated a scoring system from 0 (complete absence of cell scattering in the absence of HGF) to 4 (maximal cell scattering in the presence of 0.1 nM HGF). HPAF-II and MLP29 cells were stimulated with increasing concentrations of human / mouse equivalent anti-MET antibodies and cell scattering was determined 24 hours later using the above scoring system. As shown in Table 15, this analysis revealed that all tested mAbs promoted cell scattering in both human and mouse cell systems, with essentially overlapping results for both species. 71D6 and 71G2 exhibited exactly the same activity as HGF; 71G3 and 71A3 were slightly less potent than HGF; 71C3 and 74C8 required significantly higher concentrations to match the activity of HGF; 71D4, 76G7, 71G12, and 72F8 did not reach saturation in this assay.

[0305] Table 15: Biological activity of human / mouse equivalent anti-MET antibodies determined in a cell-based scattering assay. HPAF-II human pancreatic cancer cells and MLP29 mouse liver precursor cells were stimulated with increasing concentrations of human / mouse equivalent anti-MET antibodies and cell scattering was determined 24 hours later using the scoring system described in the text (0, no cell scattering; 4, maximum cell scattering).

[0306] HPAF-II human pancreatic cancer cells

[0307]

[0308] MLP29 mouse liver progenitor cells

[0309]

[0310]

[0311] Example 8: Protection against drug-induced apoptosis

[0312] Several lines of experimental evidence indicate that HGF exhibits potent anti-apoptotic effects on cells expressing MET (reviewed by Nakamura et al., J Gastroenterol Hepatol. 26 Suppl 1, 188-202, 2011). To test the potential anti-apoptotic activity of human / mouse equivalent anti-MET antibodies, we performed a cell-based drug-induced survival assay. MCF10A human mammary epithelial cells (American Type Culture Collection) and MLP29 mouse hepatic progenitor cells were incubated with increasing concentrations of staurosporine (Sigma Aldrich). After 48 hours, cell viability was determined by measuring total ATP concentration using the CellTiter Glo kit (Promega) and a Victor X4 multilabel plate reader (Perkin Elmer). This preliminary analysis revealed that the drug concentration that induced about 50% cell death was 60nM for MCF10A cells and 100nM for MLP29 cells. Next, we incubated MCF10A cells and MLP29 cells with the above-determined drug concentrations in the presence of increasing concentrations (0-32nM) of anti-MET mAb or recombinant HGF (human or mouse; both from R&D Systems). Cell viability was determined after 48 hours as described above. The results of this analysis provided in Table 16 show that human / mouse equivalent antibodies protect human and mouse cells from staurosporine-induced cell death to a considerable extent. Although some monoclonal antibodies (71G3, 71D6, 71G2) showed similar or higher protective activity than HGF, other molecules (76H10, 71C3, 71D4, 71A3, 76G7, 71G12, 74C8, 72F8) showed only partial protection in human or mouse cell systems.

[0313] Table 16: Biological activity of human / mouse equivalent anti-MET antibodies measured by cell-based drug-induced apoptosis assay. MCF10A human mammary epithelial cells and MLP29 mouse liver precursor cells were incubated with a fixed concentration of staurosporine in the presence of increasing concentrations of anti-MET mAb or recombinant HGF (human or mouse), and total ATP content was determined after 48 hours. Cell viability was calculated as a percentage of total ATP content relative to cells treated with neither staurosporine nor antibody and is expressed as EC 50 and E MAX .

[0314]

[0315] Example 9: Branching morphogenesis assay

[0316] As discussed in the background section, HGF is a pleiotropic cytokine that promotes harmonic regulation of independent biological activities, including cell proliferation, motility, invasion, differentiation, and survival. A cell-based assay that better summarizes all of these activities is the branching morphogenesis assay, which replicates the formation of tubular organs and glands during embryogenesis (Rosário and Birchmeier, Trends Cell Biol. 13, 328-335, 2003). In this assay, spheroids of epithelial cells are seeded in a 3D collagen matrix and stimulated by HGF to grow tubules, which eventually form branching structures. These branching tubules are similar to the hollow structures of epithelial glands, such as mammary glands, which exhibit lumens surrounded by polarized cells. This assay is the most complete HGF assay that can be performed in vitro.

[0317] To test whether human / mouse equivalent anti-MET antibodies exhibit agonistic activity in this assay, we seeded LOC human renal epithelial cells (Michieli et al., Nat Biotechnol. 20, 488-495, 2002) and MLP29 mouse liver progenitor cells onto collagen layers as described (Hultberg et al., Cancer Res. 75, 3373-3383, 2015) and then exposed them to increasing concentrations of mAb or recombinant HGF (human or mouse, both from R&D Systems). Branching morphogenesis was observed over time by microscopy, and colonies were photographed after 5 days. Representative images are shown in Figure 5 Quantification of branching morphogenetic activity was obtained by counting the number of branches per spheroid. As shown in Table 17, all antibodies tested induced a dose-dependent formation of branched tubules. However, consistent with the data obtained in the MET autophosphorylation assay and the cell scattering assay, 71D6, 71A3, and 71G2 exhibited the most potent agonist activity, similar to or better than recombinant HGF.

[0318] Table 17: Branching morphogenesis assay. Cell spheroid preparations of LOC human renal epithelial cells or MLP29 mouse hepatic progenitor cells were seeded on a collagen layer and then incubated with increasing concentrations (0, 0.5, 2.5 and 12.5 nM) of mAb or recombinant HGF (LOC, human HGF; MLP29, mouse HGF). Branching morphogenesis was observed by microscopy over time and colonies were photographed after 5 days. Branching was quantified by counting the number of branches per spheroid (primary branches plus secondary branches).

[0319] LOC cells

[0320] mAbs 0nM 0.5nM 2.5nM 12.5nM 76H10 3.3±1.5 7.3±0.6 11.7±1.5 16.7±1.5 71G3 3.0±1.0 13.7±1.5 19.0±2.6 22.3±2.1 71D6 3.0±1.0 29.0±2.0 29.0±2.6 32.7±1.5 71C3 3.3±0.6 8.7±1.5 12.7±2.1 15.7±2.1 71D4 3.0±1.0 9.0±2.6 15.7±1.2 18.7±1.5 71A3 3.0±1.7 24.0±4.6 30.3±3.2 31.3±1.5 71G2 3.7±1.5 25.3±2.1 29.3±3.5 31.7±3.5 76G7 2.7±0.6 6.7±0.6 13.3±4.2 16.3±5.7 71G12 3.3±0.6 7.0±2.6 15.3±5.5 16.0±4.6 74C8 3.0±1.0 10.3±4.2 17.0±4.6 18.7±4.9 72F8 3.3±1.5 9.0±3.5 12.3±2.1 16.0±3.0 hHGF 3.0±1.0 18.0±2 27.7±2.5 20.3±2.1

[0321] MLP29 cells

[0322] mAbs 0nM 0.5nM 2.5nM 12.5nM 76H10 0.3±0.6 10.7±4.0 14.3±3.2 24.7±6.0 71G3 0.3±0.6 24.7±4.5 34.3±5.5 29.3±8.0 71D6 1.3±1.2 32.7±3.5 39.0±7.5 41.3±8.0 71C3 0.3±0.6 11.7±3.5 15.7±6.5 24.7±6.5 71D4 0.7±1.2 16.0±2.6 14.7±4.5 21.7±5.5 71A3 0.7±0.6 30.3±2.1 42.0±6.2 42.7±8.0 71G2 1.0±1.0 34.0±2.6 46.3±4.7 45.0±7.0 76G7 0.3±0.6 14.7±2.1 18.7±4.5 24.7±6.5 71G12 1.0±1.0 14.0±2.6 14.7±5.5 22.7±6.0 74C8 0.7±0.6 17.3±2.5 15.3±6.0 22.3±9.0 72F8 1.0±1.0 12.7±3.1 11.7±3.5 18.7±2.5 mHGF 0.7±1.2 32.3±4.0 43.7±4.2 36.0±7.2

[0323] Example 10: Human-mouse equivalent agonistic anti-MET antibodies provide ample opportunities to modulate MET activity

[0324] Based on the biochemical and biological assays described so far, we performed a comprehensive analysis aimed at comparing antibody function. The properties of the various mAbs measured in the assays performed are summarized in Table 18. Analysis of this table revealed that human-mouse equivalent agonistic anti-MET antibodies exhibit a wide range of biochemical and biological activities, providing ample opportunity to modulate MET activity in a customized manner. Depending on the chosen translation or clinical application, antibodies can be selected from those that have been identified to compete fully or partially with HGF, which strongly or slightly cause MET activation, strongly or weekly promote cell invasion, or strongly or gently antagonize apoptosis. From this perspective, agonistic antibodies are more versatile and more plastic than HGF because they allow the induction of more gradual responses compared to the on or off properties of HGF.

[0325] From a pharmacological point of view, the possibility of inducing selective biological activity downstream of MET is very useful. For example, certain applications in the field of oncology are beneficial to ligands that dissociate the nutritional properties of HGF from its pro-invasive activity (Michieli et al., Nature Biotechnology, 20, 488-495, 2002). Other applications in the field of hepatology ideally require factors that protect hepatocytes from apoptosis without promoting cell invasion (Takahara et al., Liver Disease, 47, 2010-2025, 2008). In other applications in the field of muscular dystrophy, myoblast differentiation into myoblasts requires, on the one hand, shutting down HGF-induced proliferation and, on the other hand, protection from differentiation-related apoptosis (Cassano et al., PLoS One, 3, e3223, 2008). In all these applications and other similar cases, one can envision the use of partial agonist mAbs that, on the one hand, replace endogenous HGF and, on the other hand, cause mild MET activation, thereby enhancing certain biological activities of HGF and simultaneously reducing others.

[0326] In contrast, various applications in the field of regenerative medicine require strong pro-survival signals and rapid tissue repair to prevent irreversible cell damage or degeneration. This is found, for example, in cases of sudden liver failure, acute kidney injury, or severe pancreatitis (reviewed by Nakamura et al., J Gastroenterol Hepatol. 26 Suppl 1, 188-202, 2011). In all of these applications and other similar situations, one would rather use a fully agonistic mAb that is as effective as possible in promoting tissue healing and regeneration. HGF competition does not really work in this case, because fully agonistic mAbs are as effective as HGF (if not more effective), and can reach pharmacological concentrations logarithmic above the physiological levels at which the endogenous ligand is found.

[0327] In other pathological conditions involving non-canonical, less characterized functions of HGF, such as those involving the immune system (inflammatory diseases, autoimmune diseases, transplant-related complications), the hematopoietic system (stem cell mobilization, hematopoiesis), and the nervous system (nerve growth, neuronal degeneration), the role of the HGF / MET pathway remains poorly studied. Although several lines of experimental evidence suggest that recombinant HGF or HGF gene therapy improves these diseases in preclinical models (reviewed by Nakamura et al., Journal of Gastroenterology and Hepatology, 26 Suppl 1, 188-202, 2011), we do not have enough information to determine whether all or only some of the functions of HGF are beneficial. For these therapeutic applications, the possibility to fine-tune MET activity with a highly diverse panel of MET agonistic antibodies is potentially advantageous compared to HGF (without mentioning the numerous pharmacological issues implicit in using recombinant HGF as a drug as discussed in the Summary of the Invention section).

[0328] In conclusion, we suggest that all identified human-mouse equivalent anti-MET antibodies may have potential for therapeutic applications, regardless of whether they fully or partially compete with HGF and regardless of whether they fully or partially activate the MET receptor.

[0329] Table 18: Key biochemical and biological features of human / mouse equivalent anti-MET antibodies. This table summarizes the following abilities of each mAb: ability to bind to purified MET ECD (ELISA), ability to recognize native MET on MET-expressing cells (FACS), ability to compete with HGF for MET binding (HGF competition), ability to activate MET in a receptor autophosphorylation assay (MET activation), ability to promote cell scattering (scattering assay), ability to protect cells from drug-induced apoptosis (survival assay), and ability to promote branching morphogenesis of epithelial cell spheroids (branching morphogenesis). For each assay, the score is based on the relative activity of any given mAb relative to other antibodies (+, lower limit 50%; ++, upper limit 50%). Since all antibodies exhibited similar activity in human and mouse systems, only one score is shown for each assay.

[0330]

[0331] Example 11: Constant region swapping does not alter the biochemical and biological characteristics of human / mouse equivalent antibodies

[0332] Since the purpose of the present invention is to generate and identify agonistic anti-MET antibodies that work equally well in human and mouse systems, we attempted to determine whether the exchange of human heavy and light chain constant regions with the corresponding mouse constant regions affects the major biochemical and biological activities of a representative panel of antibodies. To this end, we selected three representative molecules from the human / mouse equivalent antibody panel (71G3, a partial competitor of HGF and a partial agonist in a bioassay; 71D6 and 71G2, complete competitors of HGF and complete agonists in a bioassay). The VH and VL regions of 71G3, 71D6 and 71G2 were installed on a mouse IgG1 / λ antibody backbone. The sequences of all mouse immunoglobulin variants are available from public databases such as the ImMunoGeneTics Information System (www.imgt.org). Fusion with the desired variable region can be achieved by standard genetic engineering procedures. The complete amino acid sequences of the heavy and light chains of the generated alpaca-mouse chimeric antibodies are shown in Table 19.

[0333] Table 19: Complete heavy and light chain amino acid sequences of llama-mouse chimeric mAbs that bind human and mouse MET.

[0334]

[0335] Production and purification of recombinant immunoglobulins can be achieved by transient transfection and affinity chromatography in mammalian cells, respectively, according to well-established protocols. We then compared the biochemical and biological activities of the mouse forms of 71G3, 71D6, and 71G2 with those of the human forms of the same antibodies.

[0336] We evaluated the ability of the antibodies to bind to purified human or mouse MET ECD by ELISA, recognize native MET on human or mouse cells by FACS, induce scattering of human and mouse epithelial cells, and promote branching morphogenesis in collagen. The results of this analysis, summarized in Table 20, indicate that swapping the human and mouse constant regions does not substantially affect any of the properties analyzed.

[0337] Table 20: Constant region swapping does not alter the biochemical and biological characteristics of human / mouse equivalent antibodies. Three representative agonistic antibodies (71G3, 71D6 and 71G2) in mouse or human form were subjected to several in vitro assays aimed at characterizing their main biochemical and biological properties.

[0338]

[0339] Example 12: Comparison with antibodies of the prior art: human-mouse cross-reactivity.

[0340] As discussed in detail in the Background section, several other studies have described agonistic anti-MET antibodies that at least partially mimic the activity of HGF. At the time of writing, these include: (i) 3D6 mouse anti-human MET antibody (U.S. Pat. No. 6,099,841); (ii) 5D5 mouse anti-human MET antibody (U.S. Pat. No. 5,686,292); (iii) NO-23 mouse anti-human MET antibody (U.S. Pat. No. 7,556,804 B2); (iv) B7 human naive anti-human MET antibody (U.S. Patent Application No. 2014 / 0193431 A1); (v) DO-24 mouse anti-human MET antibody (Prat et al., Mol. Cell. Biol. 11, 5954-5962, 1991; Prat et al., J. Cell Sci. 111, 237-247, 1998 (Prat et al., Mol Cell Biol. 11, 5954-5962, 1991; Prat et al., J. Cell Sci. 111, 237-247, 1998) Sci. 111, 237-247, 1998)); and (vi) DN-30 mouse anti-human MET antibody (Prat et al., Mol Cell Biol. 11, 5954-5962, 1991; Prat et al., J Cell Sci. 111, 237-247, 1998).

[0341] We obtained all prior art agonistic anti-MET antibodies as follows: 3D6 hybridoma was purchased from the American Type Culture Collection (Catalog No. ATCC-HB-12093). 3D6 antibody was purified from hybridoma conditioned media by standard affinity chromatography protocols.

[0342] The cDNA encoding the variable region of the 5D5 antibody, a bivalent progenitor of the antagonist anti-MET antibody Onartuzumab, was synthesized based on the VH and VL sequences disclosed in U.S. Pat. No. 7,476,724B2 (Merchant et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013). The obtained DNA fragment was fused to the mouse constant IgG1 / λ domain by standard protein engineering protocols and produced as a bivalent monoclonal antibody.

[0343] The NO-23 antibody was obtained from Professor Maria Prat of the University of Novara, Italy (inventor of NO-23; U.S. Patent No. 7,556,804B2). The No-23 antibody can also be obtained by requesting the corresponding hybridoma (Clone No. ICLC 03001) from the Interlab Cell Line Collection (ICLC), an international depository of the Advanced Biotechnology Center (ABC) in Genoa, Italy.

[0344] cDNA encoding the variable regions of the B7 antibody was synthesized based on the VH and VL sequences disclosed in US Patent Application No. 2014 / 0193431 A1. The obtained DNA fragment was fused to the mouse constant IgG1 / λ domain and generated as a bivalent monoclonal antibody as described above.

[0345] DO-24 and DN-30 antibodies were obtained from Professor Maria Prat of the University of Novara, Italy (who first identified and characterized DO-24 and DN-30; Prat et al., Mol Cell Biol. 11, 5954-5962, 1991; Prat et al., J Cell Sci. 111, 237-247, 1998). The now discontinued DO-24 antibody was commercially available from Upstate Biotechnology for many years. The DN-30 antibody can also be obtained by requesting the corresponding hybridoma (Clone No. ICLC PD 05006) from the Interlab Cell Line Collection (ICLC), an international depository of the Advanced Biotechnology Center (ABC) in Genoa, Italy.

[0346] Because the vast majority of animal models of human diseases use mice as hosts, cross-reactivity with mouse antigens is an essential prerequisite for antibodies whose biological activity needs to be validated in preclinical systems. Since all antibodies of the prior art are generated in mice (except B7, which was identified using a human naive phage library), it is unlikely that these molecules will cross-react with mouse MET. Even if minor cross-reactions with self-antigens are in principle possible, these interactions are usually of very low affinity.

[0347] As detailed in U.S. Pat. No. 6,099,841, the 3D6 antibody does not bind mouse MET, and the inventors had to use ferrets and minks to demonstrate that their antibody had in vivo activity. Clearly, these animal models neither represent ideal systems for modeling human disease nor are their uses in preclinical medicine established. Furthermore, the inventors did not provide any quantitative data regarding differences in antibody affinity and activity between the human system and the ferret or mink system.

[0348] It has been clearly shown that the 5D5 antibody and its derivatives do not bind to mouse MET (Merchant et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013). There is no information on its cross-reactivity with other preclinical species.

[0349] Likewise, U.S. Patent Application No. 2014 / 0193431 A1 provides no information regarding the cross-reactivity of the B7 antibody with mouse MET or MET from other species.

[0350] U.S. Patent No. 7,556,804 B2 claims that the NO-23 antibody cross-reacts with mouse, rat, and dog MET, but provides no quantitative experimental evidence to support this claim. The inventors used a single saturating dose of NO-23 to immunoprecipitate MET from mouse, rat, human, or dog cell lysates and then mixed the immunoprecipitated proteins with radioactive 32 After radiolabeling, the incorporated 32 P-ATP visualization. This method is very sensitive and by no means quantitative; it is impossible to know what percentage of cross-reactivity the bands on the gel correspond to.

[0351] Similarly, it is suggested that the DO-24 antibody cross-reacts with mouse MET because Matrigel particles containing DO-24 promote vascular recruitment when implanted into the peritoneal cavity of mice (Prat et al., J Cell Sci. 111, 237-247, 1998). However, this may also be due to increased inflammation and no direct evidence of DO-24 interaction with mouse MET is provided. In another different study, a single saturating dose of DO-24 (20 nM) was shown to cause autophosphorylation of MET in rat cardiomyocyte cell line H9c2 and mouse cardiomyocyte cell line HL-5 (Pietronave et al., Am J Physiol Heart Circulatory Physiol. 298, H1155-65, 2010; Figure 1 )). In the same experiment, a much lower dose of recombinant HGF (0.5 nM) was shown to cause MET phosphorylation to a comparable extent. As the authors themselves acknowledge in the Discussion section, these results indicate that DO-24 is significantly less potent than HGF in these rodent cell lines. Since the same authors claim that DO-24 is a fully agonistic mAb that matches HGF activity in human cell models (Prat et al., J Cell Sci. 111, 237-247, 1998), it should be concluded that DO-24 does not induce the same efficacy or potency in human and mouse cells. In addition, it should be noted that the experiments shown by Pietronave et al. are not quantitative and cannot be used to extract information on the extent of cross-reactivity that occurs between DO-24 and mouse or rat MET, the measurement of which requires head-to-head dose-response studies such as those we have done (see below). In a third work, a mixture of DO-24 and DN-30 antibodies was used to immunoprecipitate MET from mouse mesenchymal stem cell lysates (Forte et al., Stem Cells. 24, 23-33, 2006). Both the presence of DN-30 and the type of assay (immunoprecipitation from cell lysates) prevented obtaining precise information about the ability of DO-24 to interact with native mouse MET. In summary, there is no experimental evidence that the DO-24 antibody elicits comparable biological responses in human and mouse cells.

[0352] Finally, it was clearly shown that the DN-30 antibody does not interact with mouse MET (Prat et al., J Cell Sci. 111, 237-247, 1998; and Petrelli et al., Proc Natl Acad Sci USA 103, 5090-9095, 2006, supplementary material).

[0353] To directly determine whether and to what extent the prior art agonistic anti-MET antibodies cross-react with mouse MET, and to compare them with our human / mouse equivalent anti-MET antibodies, we performed ELISA tests. Since all prior art antibodies were obtained or engineered in mouse IgG / λ format, we used mouse IgG / λ format of 71G3, 71D6 and 71G2. Human or mouse MET ECD was immobilized on a solid phase (100 ng / well, 96-well plate) and exposed to increasing concentrations (0-40 nM) of the antibody in solution. Binding was revealed by using an HRP-conjugated anti-mouse Fc antibody (Jackson Immuno Research Laboratories). As shown in Table 21, the analysis revealed that; although the prior art antibodies had a K of 0.059 nM (B7) to 4.935 nM (3D6), the d 71G3, 71D6, and 71G2 bound to both human and mouse MET, and they did so with indistinguishable affinities and capacities. The complete binding curves for all antibodies are shown in Figure 2. Figure 6 shown.

[0354] Table 21: Binding affinity and capacity of anti-MET antibodies to human and mouse MET as determined by ELISA. Affinity is expressed as EC 50 (nMol / L). Capacity is expressed in E MAX (optical density at 450 nm; nc, not converged). For the entire binding curve, see Figure 6 .

[0355]

[0356] Example 13: Comparison with prior art antibodies: MET autophosphorylation

[0357] In order to compare the agonistic activity of the antibodies of the prior art with the agonistic activity of the human / mouse equivalent anti-MET antibodies, we used human and mouse cells for MET autophosphorylation experiments. A549 human lung cancer cells and MLP29 mouse liver precursor cells were deprived of serum growth factors for 48 hours and then stimulated with increasing concentrations (0-25nM) of antibodies. After 15 minutes of stimulation, the cells were washed twice with ice-cold phosphate-buffered saline (PBS) and then lysed as described (Longati et al., Oncogene, 9, 49-57, 1994). Phospho-MET levels were determined by ELISA using anti-MET antibody for capture (R&D Systems) and anti-phosphotyrosine (R&D Systems) for revealing as described above (Basilico et al., J Clin Invest. 124, 3172-3186, 2014).

[0358] This analysis revealed two major differences between the prior art antibodies and the human / mouse equivalent anti-MET antibodies described in this document. First, consistent with the results obtained in the binding experiments, only 71G3, 71D6 and 71G2 promoted MET autophosphorylation in human and mouse cells. Prior art antibodies (including DO-24 and NO-23) induced MET activation only in human cells; no activity on mouse cells was detected in the system we analyzed. Second, all prior art antibodies always showed lower agonist activity compared to 71G3, 71D6 and 71G2. The most agonist prior art mAbs were 5D5 and B7, which showed slightly lower activity than 71G3, 71D6 and 71G2. The least agonist prior art mAb was 3D6. The other molecules showed intermediate activity. The results of this analysis are shown in Figure 7 middle.

[0359] Example 14: Comparison with prior art antibodies: Branching morphogenesis

[0360] In order to compare the biological activity of the antibodies of the prior art with that of the human / mouse equivalent anti-MET antibodies, we performed a branching morphogenesis assay. This assay summarizes all relevant biological activities of HGF, including cell proliferation, scattering, differentiation and survival. As described above, LOC human renal epithelial cells and MLP29 mouse liver precursor cells were seeded on a collagen layer and then incubated with increasing concentrations of mAb or recombinant HGF (human or mouse, both from R&D Systems). Branching morphogenesis was observed microscopically over time and colonies were photographed after 5 days. Quantification of branching morphogenetic activity was achieved by counting the number of branching tubules sprouting from each spheroid and is shown in Table 22. Representative images of spheroids are shown in Figure 8 (LOC cells) and Fig. 9 (MLP29 cells).

[0361] Table 22: Branching morphogenesis assay. Cell spheroid preparations of LOC human renal epithelial cells or MLP29 mouse liver precursor cells were seeded on a collagen layer and then incubated with increasing concentrations (0, 0.04, 0.2, 1 and 5 nM) of mAb or recombinant human HGF (LOC) or mouse HGF (MLP29). Branching morphogenesis was observed by microscopy over time and colonies were photographed after 5 days. Branching was quantified by counting the number of branches per spheroid (primary branches plus secondary branches).

[0362] LOC cells

[0363] mAbs 0nM 0.04nM 0.2nM 1nM 5nM 71G3 2.7±0.6 9.0±1.0 13.3±1.5 17.7±1.5 20.7±1.2 71D6 2.3±0.6 18.7±3.2 29.3±2.5 30.7±2.1 30.3±1.2 71G2 2.7±1.5 22.3±2.3 26.3±2.1 30.0±2.0 30.3±3.5 3D6 2.3±0.6 4.0±1.0 7.0±1.0 10.7±1.5 19.3±4.2 5D5 4.3±1.5 15.7±1.5 18.3±1.5 21.3±2.1 27.7±1.5 B7 3.3±1.5 8.7±1.5 13.3±1.5 19.7±1.5 24.0±2.0 NO-23 3.3±1.2 6.0±1.0 7.0±1.0 8.7±1.2 8.7±1.5 DO-24 3.3±2.1 8.0±1.0 12.0±1.0 12.3±1.2 17.7±2.1 DN-30 3.3±0.6 6.3±1.5 8.3±1.5 9.7±1.5 10.3±1.5 hHGF 4.7±1.5 10.7±1.5 16.7±1.5 28.3±3.5 24.7±7.6

[0364] MLP29 cells

[0365] mAbs 0nM 0.04nM 0.2nM 1nM 5nM 71G3 0.3±0.6 19.3±1.5 23.7±2.1 32.7±2.5 28.7±1.2 71D6 0.7±0.6 21.0±2.0 32.0±1.0 42.7±5.5 37.0±2.0 71G2 0.0±0.0 15.0±1.7 36.0±4.6 50.7±5.5 48.0±3.6 3D6 0.3±0.6 0.7±0.6 0.7±0.6 0.7±0.6 0.7±0.6 5D5 1.0±1.0 0.7±1.2 0.3±0.6 1.3±1.5 1.0±1.0 B7 0.3±0.6 0.7±0.6 0.3±0.6 1.3±1.5 0.7±1.2 NO-23 0.7±1.2 0.3±0.6 0.7±0.6 1.0±1.0 0.7±0.6 DO-24 1.0±1.0 0.7±1.2 0.7±0.6 0.7±1.2 0.7±1.2 DN-30 0.7±0.6 0.3±0.6 1.0±1.0 0.7±0.6 0.7±0.6 mHGF 0.3±0.6 26.0±4.4 34.0±5.0 46.0±2.6 37.0±2.0

[0366] The data presented lead to the following observations. In human cells, 71D6, 71G2, and 5D5 exhibited activity comparable to that of human HGF; 71G3, 3D6, B7, and DO-24 behaved as partial agonists; NO-23 and DN-30 exhibited very little agonistic activity. In mouse cells, only 71G3, 71D6, and 71G2 effectively induced the formation of branched tubules; all other antibodies, consistent with their inability to bind mouse MET in ELISA, did not induce branched morphogenesis at all.

[0367] We conclude that the prior art antibodies elicit different biological activities in human and mouse systems compared to the human / mouse equivalent anti-MET antibodies.

[0368] Example 15: Plasma Half-Life of Human / Mouse Equivalent Anti-MET Antibodies

[0369] Next, we transferred the selected human / mouse equivalent anti-MET antibodies to in vivo studies. As a preliminary analysis, we determined their peak and trough levels in mice. To this end, we injected affinity-purified 71G3, 71D6 and 71G2 (in its mouse IgG / λ form) into 7-week-old female BALB / c mice (Charles River) by intraperitoneal injection. A single bolus of 1 mg / kg or 10 mg / kg was injected, and blood samples were collected from the tail vein at 3, 6, 12 and 24 hours after injection. The blood samples were processed and the antibody concentration in plasma was determined by ELISA. As described in Example 1, standard 96-well plates were coated with human MET ECD (100 ng / well) and then exposed to incrementally diluted mouse plasma to capture anti-MET antibodies. After repeated washing with PBS, the presence of anti-MET antibodies was revealed using HRP-conjugated donkey anti-mouse antibodies (Jackson Laboratories). To quantify bound antibodies, we established standard curves of purified 71G3, 71D6, and 71G2 under the same conditions.

[0370] The results of this analysis are shown in Fig.10 The antibody concentrations in plasma were similar for all antibodies tested and were proportional to the amount of protein injected. After 24 hours, the antibody concentrations in plasma were approximately 15 nM for a 1 mg / kg bolus and 250 nM for a 10 mg / kg bolus. Considering that the agonist activity of these antibodies in the most demanding assay (branching morphogenesis assay) reaches saturation at concentrations of 5 nM or lower, we can safely conclude that the plasma levels of antibodies obtained by intraperitoneal injection are biologically relevant with boluses as low as 1 mg / kg.

[0371] In addition, we also calculated the plasma half-life of the injected antibody. This is achieved by converting the antibody concentration to natural logarithm (Ln), fitting the data into a line and then calculating the slope of the line. This analysis leads to the estimation that the half-life of 71G3, 71D6 and 71G2 is very similar, and for 1mg / kg push injection corresponds to about 3 days and for 10mg / kg push injection is about 9 days. This is compared with the recombinant HGF (Ido et al., Hepatology Research, 30 phases, 175-181 pages, 2004 (Ido et al., Hepatol Res.30,175-181,2004)) reported to have a half-life of 2.4 minutes in rodents. The plasma stability data of the whole group are summarized in Table 23.

[0372] These data suggest that human / murine equivalent anti-MET antibodies may advantageously replace recombinant HGF in all clinical applications requiring systemic administration of HGF.

[0373] Table 23: plasma stability of human / mouse equivalent antibodies. Single bolus injections (1 mg / kg or 10 mg / kg) of affinity-purified 71G3, 71D6 and 71G2 were administered to 7-week-old female BALB / c mice by intraperitoneal injection. After injection for 3, 6, 12 and 24 hours, blood samples were collected from the tail vein and the antibody concentration in the plasma was determined by ELISA. The plasma half-life was calculated by linear fit of the natural logarithm conversion of the antibody concentration.

[0374]

[0375] Example 16: In vivo activity: protection against acute liver injury

[0376] Hepatocytes express MET and are the main target of HGF, which promotes their proliferation and protects them from apoptosis (reviewed by Nakamura et al., J Gastroenterol Hepatol. 1, 188-202, 2011). Therefore, we tested whether human / mouse equivalent agonistic anti-MET antibodies exhibit protective activity in a mouse model of acute liver failure. To this end, we injected a single dose of CCl4 (0.2 ml of a 10% solution in olive oil; both from Sigma-Aldrich) into the subcutaneous compartment of 7-week-old female BALB / c mice (Charles River). Shortly after the CCl4 injection, the mice were randomly divided into 4 large groups of 6 mice each, and each mouse received a single bolus injection of purified 71G3, 71D6, 71G2 or vehicle only (PBS). Antibody is administered by intraperitoneal injection dosage of 5mg / kg. Blood samples are collected at different times (0,12,24 and 48 hours) after injection. The fifth other control large group comprises 6 mice, and it does not accept CCl4 or antibody and is put to death when the experiment is finished. When autopsy, blood and liver are collected and analyzed. The plasma level of liver marker aspartate aminotransferase (AST), alanine aminotransferase (ALT) and bilirubin (BIL) is measured by standard clinical biochemical method. Liver is embedded in paraffin and standard protocol is used to process it for histological analysis.

[0377] like Fig.11Shown, CCl4 injection in control mice causes the rapid and significant increase of the level of all three blood parameters analyzed, and this increase reaches peak value after poisoning 12-24 hours.In the control large group, CCl4 injection causes AST, ALT and bilirubin levels to rise 286,761 and 13 times respectively.In all antibody large groups, these increases significantly reduce (71G3, 53%, 62% and 46%; 71D6, 37%, 34% and 48%; 71G2, 50%, 39% and 54%).With regard to liver protection, the most effective antibody is 71D6.

[0378] Histological examination of the liver at autopsy revealed that CCl4 caused significant tissue damage around the central vein of each liver module, characterized by eosinophilic staining and large cytoplasm, typically diseased hepatocytes. Cell-cell interactions appear loose, allowing infiltration of red blood cells that leak from damaged blood vessels. In the large group treated with antibodies, these peripheral damaged areas are smaller and show less signs of tolerance, which can be demonstrated by less eosinophilic staining, normal cytoplasmic size and reduced blood cell infiltration. Representative images of liver sections stained with hematoxylin and eosin are shown in Fig.12 .

[0379] These results suggest that human / mouse equivalent agonistic anti-MET antibodies may be used clinically to treat acute liver diseases characterized by rapid progression of liver dysfunction, often leading to abnormal liver biochemical values, jaundice, coagulopathy, cerebral edema, and encephalopathy. These pathological conditions include, but are not limited to, acetaminophen overdose, idiosyncratic reactions to drugs (e.g., tetracycline), drug abuse (ecstasy, cocaine), viral infections (hepatitis A, B, E).

[0380] Example 17: In vivo activity: protection against chronic liver damage

[0381] We also tested whether human / mouse equivalent agonistic anti-MET antibodies exhibit therapeutic effects in a mouse model of chronic liver injury. In fact, HGF is known to have antifibrotic activity in the liver (reviewed by Matsumoto and Nakamura, Ciba Found Symp. 212, 198-211; discussion 211-214, 1997). To this end, 7-week-old female BALB / c mice (Charles River) were chronically exposed to CCl4 for several weeks. In the first week, mice were injected subcutaneously twice with 0.1 ml of a 5% solution of CCl4 in olive oil (both from Sigma-Aldrich). In the following weeks, the dose of CCl4 was increased (0.1 ml of a 10% solution in olive oil), while the injection frequency remained unchanged (twice a week). Soon after the first injection, mice were randomly divided into 4 large groups, 7 mice in every large group, which were accepted by 71G3, 71D6, 71G2 or only vehicle (PBS) respectively. Antibody was administered by intraperitoneal injection three times a week at a dosage of 1mg / kg. The other 5th control large group comprises 7 mice, which do not accept CCl4 or antibody and are used as healthy controls. At chronic CCl4 poisoning mice were put to death after 6 weeks. When autopsy, blood was collected and liver was analyzed. The plasma levels of liver markers aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured by standard clinical biochemical methods. Liver was embedded in paraffin and standard protocol was used to process it for histological analysis.

[0382] like Fig.13 As shown, long-term exposure to CCl4 in control mice resulted in impaired liver function, as determined by higher AST and ALT plasma levels. In contrast to the acute model, which causes a sharp but brief burst in liver marker levels, chronic CCl4 intoxication induced a more moderate increase in AST and ALT levels, about 5-fold compared to untreated mice. Notably, antibody treatment completely blocked the increase in AST concentrations, in fact reducing it to below basal levels. Antibodies also significantly prevented the burst in ALT levels, although not as dramatic as observed for AST.

[0383] Liver sections were stained by various techniques designed to detect fibrotic tissue, including Masson's trichrome staining, Picro Sirius Red, and anti-alpha smooth muscle actin (α-SMA) antibodies. Hematoxylin and eosin staining was also performed to examine the general histological structure. The analysis revealed that chronic CCl4 treatment resulted in the formation of a significant amount of fibrotic tissue in the interlobular spaces, specifically characterized by positive staining for Picro Sirius Red and anti-alpha smooth muscle actin (α-SMA) antibodies. The fibrotic tissue formed a "ribbon" connecting the portal triads, demonstrating the hexagonal shape of the liver unit. Notably, liver sections from animals receiving CCl4 and agonistic anti-MET antibodies exhibited milder fibrosis in terms of staining intensity, and the fibrotic areas appeared to be confined to the periportal space. Representative images of liver sections stained with Picro Sirius Red and anti-α-SMA antibodies are shown in Figure 3, Table 4, and Table 5, respectively. Fig.14 and Fig.15 .

[0384] These data suggest that human / mouse equivalent agonistic anti-MET antibodies may be used clinically to treat pathological conditions associated with chronic liver injury, characterized by progressive destruction and regeneration of the liver parenchyma and leading to cirrhosis and fibrosis. Agonistic anti-MET antibodies may be used to reduce or prevent fibrosis, thereby restoring liver structure and function. They may also be used to suppress inflammatory and immune responses that typically exacerbate chronic liver disease.

[0385] Example 18: In vivo activity: protection against acute kidney injury

[0386] Renal epithelial cells express significant levels of MET and are very sensitive to HGF stimulation (reviewed by Mizuno et al. Front Biosci. 13, 7072-7086, 2008). Therefore, we tested whether human / mouse equivalent agonistic anti-MET antibodies exhibit protective effects in a mouse model of acute renal failure. To this end, we induced tubular injury in 7-week-old female BALB / c mice (Charles River) by intraperitoneal injection of a single bolus of HgCl2 (3 mg / kg). Shortly after HgCl2 intoxication, mice were randomized into 4 large groups, which were treated with 71G3, 71D6, 71G2, or vehicle only (PBS). The antibodies were administered by intraperitoneal injection at a dose of 10 mg / kg every 24 hours. Each large group contained 6 mice, which were sacrificed 72 hours after injection of HgCl2. At autopsy, blood and kidneys were collected for analysis. Blood urea nitrogen (BUN) and creatinine (CRE) plasma levels were determined by standard clinical biochemical methods. Kidneys were processed for histological analysis using standard protocols.

[0387] like Fig.16 As shown, HgCl2 injection into control mice caused a dramatic increase in BUN and CRE levels. In the control group, BUN and CRE increased 6-fold and 12-fold, respectively. These increases were significantly reduced in all antibody groups (respectively: 71G3, 52% and 54%; 71D6, 39% and 30%; 71G2, 45% and 44%). In terms of kidney protection, the most effective antibody was 71D6.

[0388] Histological examination of the kidneys revealed that HgCl2 caused extensive tubular damage characterized by proximal tubular dilatation, atrophy, and necrosis. The glomerular architecture collapsed and detached from the surrounding matrix, substantially increasing the periglomerular space. In the antibody-treated large group, there was less proximal tubular cell necrosis, and the histological architecture of the glomeruli appeared intact. Representative images of kidney sections stained with hematoxylin and eosin are shown in Fig.17 .

[0389] We propose that human / mouse equivalent agonistic anti-MET antibodies can be used clinically to treat pathological conditions associated with acute renal failure, which may be caused by, for example, ischemic or nephrotoxic injury, hypovolemic shock, obstruction of the urine collecting system, atherosclerosis, sepsis, diabetes, autoimmune disease or rhabdomyolysis. Agonistic anti-MET antibodies can be used to prevent or reverse acute renal failure, protect renal tubular epithelial cells from apoptosis, accelerate epithelial cell regeneration and restore renal function.

[0390] Example 19: In vivo activity: protection against acute colonic injury and reduction of inflammation in a mouse model of ulcerative colitis and promote regeneration

[0391] It is well known that intestinal epithelial cells express MET and that HGF plays a key role in the homeostasis and regeneration of the gastrointestinal tract (Nakamura et al., J Gastroenterol Hepatol. 1, 188-202, 2011). Therefore, we tested whether human / mouse equivalent agonistic anti-MET antibodies could promote intestinal protection and regeneration in a mouse model of ulcerative colitis. To this end, we exposed 7-week-old female BALB / c mice (Charles River) to dextran sulfate sodium (DSS) in drinking water for 10 days. On day 10, DSS treatment was discontinued and mice were returned to normal water. Starting from day 1, mice were randomly divided into 7 large groups of 7 mice each and received treatment with 71G3, 71D6, 71G2 (at a dose of 1 mg / kg or 5 mg / kg) or vehicle only (PBS). The antibody was administered by intraperitoneal injection three times a week. An additional eighth control group contained 7 mice that did not receive DSS or antibody and served as healthy controls. On day 12, i.e., 2 days after discontinuation of DSS administration, the mice were sacrificed. At autopsy, the colons were collected, washed, and their lengths were measured with a ruler. After measurement, the colons were embedded in paraffin and processed for histological analysis.

[0392] Throughout the experiment, mice body weight was monitored regularly, and clinical symptoms of ulcerative colitis were assessed by measuring fecal blood, rectal bleeding, and fecal consistency. Quantification was achieved by using a standard scoring system used in preclinical models (Kim et al., Journal of Visualized Experiments, 60, pii: 3678, 2012): each parameter was scored from 0 (no symptoms) to 3 (maximum manifestation of symptoms). Scores relative to individual parameters were added together to produce a disease activity index (DAI) of 0 to 9.

[0393] like Fig.18 As shown, exposure to DSS in the PBS group resulted in a body weight loss of up to 25%; the DAI increased to 4 points or more; and the length of the colon decreased by up to 40%. Remarkably, all the antibodies analyzed reversed these effects in a dose-dependent manner, thus showing significant activity already at the lower doses tested. 71D6 was the most potent antibody: after a transient decrease, it restored body weight to normal values, comparable to those observed in the PBS group; it suppressed the increase in DAI, thus substantially suppressing all clinical symptoms; and it prevented the shortness of the colon, thus limiting it to insignificant changes.

[0394] Colon sections were stained with hematoxylin and eosin and examined microscopically. Fig.19 As shown, DSS administration caused severe damage to the colonic mucosa. The epithelial layer was eroded and infiltrated with lymphocytes. The colonic mucosa was interspersed with occult abscess sites and was severely colonized by foamy macrophages that caused tissue destruction. Peripheral visceral lymph nodes were enlarged. The mucous glands were characterized by atrophy and showed significant mucus consumption, which was replaced by inflammatory infiltrates (including foamy macrophages, lymphocytes, and neutrophils). Several ulcers were significantly invaded by granulocytes or macrophage exudates, resulting in complete disappearance of glandular components. It is worth noting that mice treated with DSS and agonistic anti-MET antibodies showed milder degeneration and inflammatory symptoms. Specifically, there were no components of acute inflammation, including macrophages and granulocytes; the mucosa was only slightly damaged, manifested as sparse gland distortion and sparseness; mucin secretion was restored, and there was no erosion or ulceration at all. Although these protective effects were dose-dependent in all antibody groups, they were already evident at 1 mg / kg, indicating that the antibody concentrations achieved with this dose were very close to saturation (see plasma stability in Example 15). In this model, the most potent antibody appears to be 71D6.

[0395] Example 20: In vivo activity: protection against acute colonic injury, reduction of inflammation and Immunosuppression

[0396] Motivated by the above results, we also tested whether agonistic anti-MET antibodies exhibit therapeutic effects in a more specific mouse model of inflammatory bowel disease. To this end, we induced acute colonic injury in 7-week-old female C57BL / 6 mice (Charles River) by rectal injection of 2,4,6-trinitrobenzenesulfonic acid (TNBS) dissolved in ethanol. The TNBS / ethanol combination is known to induce colorectal inflammation through both immunologic and erosive processes (reviewed by Jones-Hall and Grisham, Pathophysiology 21, 267-288, 2014). TNBS dissolved in 50% ethanol was administered by enema at a dose of 5 mg / mouse. Shortly after TNBS administration, mice were randomly divided into 4 large groups of 6 mice each to receive treatment with purified 71G3, 71D6, 71G2, or vehicle only (PBS). The antibody was administered every other day by intraperitoneal injection at a dose of 1 mg / kg. The fifth control group contained 6 mice that did not receive TNBS or antibody and were used as healthy controls. The weight of mice was monitored daily. Mice were sacrificed 5 days after TNBS administration. As described above, at autopsy, the colon was collected and measured. After measurement, the colon was embedded in paraffin and processed for histological analysis.

[0397] like Fig. 20 As shown, exposure to TNBS resulted in a weight loss of approximately 15% and a reduction in colon length of more than 20%. Although milder than the effects caused by DSS, these effects were significantly different from those observed in all antibody groups. In fact, treatment with 71G3, 71D6, and 71G2 almost completely inhibited TNBS-induced weight loss and colon shortening, making the antibody-treated animals almost indistinguishable from healthy control mice.

[0398] Colon sections were stained with hematoxylin and eosin and examined microscopically. Fig.21 As shown, TNBS administration caused the onset of typical symptoms of lymphocytic colitis, which is characterized by enlargement of peripheral visceral lymph nodes, the presence of lymphocyte aggregates in the submucosal layer and mucosa, and increased lymphocyte infiltration. Several full-depth ulcers were visible, which were associated with stromal hyperproliferation and lymphocyte and neutrophil infiltration. All of these pathological processes were strongly inhibited in the agonist anti-MET antibody group, showing reduced lymphocyte infiltration and reduced mucosal damage. Even if lymphocytes were present, they were not associated with mucus damage or epithelial damage.

[0399] These results and the data reported in the previous examples suggest that human / mouse equivalent agonistic anti-MET antibodies may be used clinically to treat pathological conditions associated with ulcerative colitis or more generally with inflammatory bowel disease. Treatment with agonistic anti-MET antibodies can reduce intestinal damage, promote epithelial cell proliferation, and reduce inflammatory cell infiltration, thereby improving the clinical course of the disease.

[0400] Example 21: In vivo activity: promoting glucose uptake and synergy with insulin in a mouse model of type 1 diabetes

[0401] HGF has been reported to promote insulin-dependent glucose uptake in cultured mouse skeletal muscle cells (Perdomo et al., J Biol Chem. 283, 13700-13706, 2008). Therefore, we tested whether our agonistic anti-MET antibody could reduce hyperglycemia in a mouse model of type I diabetes. To this end, we induced pancreatic β-cell degeneration in 7-week-old female BALB / c mice (Charles River) by intraperitoneal injection of streptozotocin (STZ; Sigma Aldrich). STZ was injected at a dose of 40 mg / kg per day for 5 consecutive days. One week after the last injection, blood glucose levels under fasting conditions were determined using standard glucose strips (GIMA). At this time, STZ-treated mice showed twice the average basal blood glucose compared to untreated mice (240 mg / dL vs. 120 mg / dL). Mice are randomly divided into 4 large groups, 7 mice in each large group, and each mouse receives the treatment of purified 71G3, 71D6, 71G2 or only vehicle (PBS) based on basal blood sugar. Antibody is administered by intraperitoneal injection twice a week at a dosage of 1mg / kg. The fifth control large group in addition comprises 7 mice, which do not receive STZ or antibody and are used as healthy controls. Monitor blood glucose concentration under fasting conditions at any time for 5 weeks. At the end of the 5th week, we carried out glucose tolerance test (GTT) and insulin tolerance test (ITT). GTT includes administration of glucose to fasting animals by oral gavage, and then blood glucose levels are measured at different time points. ITT includes administration of insulin to partially fasted animals by intraperitoneal injection or intravenous injection, and then blood glucose levels are measured at different time points.

[0402] like Fig. 22 As shown in Figure 1A, basal blood glucose levels in STZ-treated mice continued to increase throughout the experiment. This is due to chronic pancreatic inflammation, which gradually worsens organ damage. In contrast, antibody-treated animals showed steadily decreasing blood glucose levels, which eventually reached a plateau after the second week of treatment. Antibody administration did not completely normalize blood glucose, but it reduced it by up to 25%, thus bringing it to approximately halfway between the levels observed in STZ-treated and control mice. Considering that hyperglycemia in this model is due to the lack of β-cell-derived insulin, we wondered whether the lower glucose levels in the antibody group were due to increased insulin levels. However, ELISA testing of blood samples revealed that this was not the case (not shown). In the GTT, although the mice receiving antibody treatment started with lower blood glucose levels, they failed to show a normal glucose uptake curve ( Fig. 22In contrast, antibody-treated mice did show a faster response to insulin in the ITT ( Fig. 22 C). Fifteen minutes after insulin injection, glucose blood levels in mice undergoing chronic antibody treatment dropped to approximately 30-40% relative to time zero, which is significantly lower than levels observed in STZ-treated mice and control animals ( Fig. 22 D in Figure 5). These results suggest that agonistic anti-MET antibodies promote glucose uptake in the absence of insulin. They also suggest that agonistic anti-MET antibodies and insulin synergistically regulate glucose uptake when both are present.

[0403] This hypothesis was tested in a cell-based assay using mouse skeletal muscle cells. According to the provider's recommendation, C2C12 mouse myoblasts (obtained from the American Tissue Type Collection) were induced to differentiate into myoblasts and then incubated with human / mouse equivalent agonistic anti-MET antibodies (71G3, 71D6, 71G2). After 24 hours, the antibody-treated cells were divided into 3 large groups and subjected to acute stimulation of 0nM, 100nM or 1000nM human recombinant insulin (Sigma Aldrich) in the presence of fluorescent glucose analog 2-(N-7-nitro-2,1,3-benzoxadiazole-4-amino)-2-deoxy-D-glucose (2-NBDG; Life Technologies) for 1 hour. 2-NBDG uptake was determined by flow cytometry.

[0404] like Fig.23 As shown, 71G3, 71D6 and 71G2 promote glucose uptake in a dose-dependent manner. Compared with insulin alone and antibodies alone, the combination of insulin and agonist anti-MET antibodies produces a synergistic effect and promotes higher glucose uptake. These data are consistent with the findings in the synergy of HGF and insulin in regulating glucose metabolism in cultured cells (Fafalios et al., Nature Medicine, 17, 1577-1584, 2011), and confirm our hypothesis that agonist anti-MET antibodies can enhance insulin-independent and dependent glucose uptake.

[0405] Example 22: In vivo activity: Normalization of blood glucose levels and overcoming insulin resistance in a mouse model of type II diabetes anti-

[0406] Driven by the observation that human / mouse equivalent agonistic anti-MET antibodies can synergize with insulin to promote glucose uptake, we tested their therapeutic potential in a mouse model of type II diabetes. Type II diabetes is characterized by high blood glucose levels, hyperinsulinemia and insulin resistance. One of the most characterized mouse models of type II diabetes is the db / db mouse, a C57BLKS / J strain with a point mutation in the leptin receptor gene lepr. This mutation leads to a loss of satiety and therefore unlimited food intake, resulting in obesity and the above-mentioned clinical hallmarks of type II diabetes (reviewed by Wang et al., Current Diabetes Review, 10, 131-145, 2014).

[0407] Female db / db mice were obtained from Charles River (JAX) at 7 weeks of age. TM Mice Strain BKS.Cg-Dock7 m + / +Lepr db J) was obtained. One week later, mice were randomly divided into 4 large groups, each with 5 mice, and treated with purified 71G3, 71D6, 71G2 or only carrier (PBS) respectively. The antibody was administered by intraperitoneal injection twice a week at a dose of 1 mg / kg. The blood glucose concentration under fasting conditions was monitored every 14 days for 8 weeks. After 7 weeks of treatment, when mice were 15 weeks old, glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed.

[0408] like Fig.24 Shown, the average basal blood glucose concentration in the PBS large group is about 230mg / dL when randomized, which absolutely corresponds to the level of diabetic patients. These values ​​tend to increase over time, and the average blood glucose concentration in the PBS large group (i.e. after 8 weeks) is about 330mg / dL at the end of the experiment. On the contrary, in the large group receiving antibody treatment, the basal blood glucose under fasting conditions continuously decreases over time. At the end of the experiment, the average blood glucose concentration of 71G3, 71D6 and 71G2 large groups was respectively 173mg / dL, 138mg / dL and 165mg / dL.

[0409] After 7 weeks of treatment, i.e. when the mice were 15 weeks old, we tested their acute response to glucose and insulin attack. It should be remembered that these mice are hyperinsulinemic and show high blood sugar levels compared to mice treated with STZ (see Example 21) because they are insulin resistant. In fact, when challenged with glucose in GTT, the mice of the large group of PBS failed to show a normal glucose uptake curve. All mice showed a sharp increase in blood sugar, which remained elevated during the entire test. Symmetrically, when performing ITT, the same mice showed an abnormal response to insulin, thereby showing a slight and transient increase in glucose levels. At least in preclinical models, this abnormal response is a sign of insulin resistance.

[0410] Mice in the antibody group also did not show a normal glucose uptake curve in the GTT, although they started from a lower basal level, thus indicating that the agonist anti-MET antibody could not neutralize the acute burst in blood glucose levels. However, it is noteworthy that the antibody treatment did significantly improve the response to insulin in the ITT, reversing the paradoxical effect observed in the PBS group and making the ITT curve look more similar to that shown by non-diabetic mice (C57BLKS / J; Charles River). We conclude that long-term treatment with agonist anti-MET antibodies improves type II diabetes in db / db mice and partially overcomes insulin resistance.

[0411] Based on these results and the results provided in the previous examples, we suggest that human / mouse equivalent agonist anti-MET antibodies can be used clinically to treat pathological conditions associated with high blood glucose levels. These may include: type I diabetes, type II diabetes, or other diabetes-like conditions characterized by high glucose and / or insulin resistance (e.g., metabolic syndrome).

[0412] Example 23: In vivo activity: Improvement of fatty liver in a mouse model of non-alcoholic steatohepatitis

[0413] It has been shown that targeted genetic deletion of MET in the liver leads to the development of severe non-alcoholic steatohepatitis (NASH) in mice (Kroy et al. J Hepatol. 61, 883-890, 2014). In an independent study (Kosone et al. Am J Physiol Gastrointest Liver Physiol. 293, G204-210, 2007), HGF improved high-fat diet-induced fatty liver in mice by activating microsomal triglyceride transfer protein (MTP) and apolipoprotein B (ApoB), thereby minimizing the storage of fatty acids.

[0414] High insulin db / db mice are also widely used as a model of NASH, and more generally as a model of fatty liver disease. On a normal diet, these mice accumulate large amounts of lipids in their hepatocytes, leading to hepatic steatosis, fibrosis, and chronic liver failure. This condition can be further exacerbated by placing mice on a high-fat diet (reviewed by Anstee and Goldin, Int J Exp Pathol. 87, 1-16, 2006).

[0415] Motivated by the above observations and considerations, we tested whether human / mouse equivalent agonistic anti-MET antibodies could improve moderate hepatic steatosis in db / db mice on a normal diet. To this end, we obtained female db / db mice as described above. When the animals were 8 weeks old, they were randomly divided into 4 large groups of 6 mice each and treated with purified 71G3, 71D6, 71G2 or vehicle only (PBS). The antibodies were administered by intraperitoneal injection twice a week at a dose of 1 mg / kg. After 8 weeks of treatment, the mice were killed and autopsied. The liver was extracted, embedded in paraffin and processed for histological examination. Blood was collected for analysis of liver function markers.

[0416] Liver sections were stained with hematoxylin and eosin or Picro Sirius red to highlight fibrosis. Fig.25 As shown, livers from the PBS group showed marked steatosis, often concentrated around the central veins. The hepatocytes were significantly enlarged and rich in lipids. Fatty hepatocytes were intermixed with normal hepatocytes, and steatosis occupied up to 60% of the pericentral space. In contrast, livers from antibody-treated animals contained significantly fewer fat cells and looked very normal overall. Fig.26 As shown, Picro Sirius red staining confirmed moderate periportal fibrosis in the PBS group, characterized by thickening of the stromal layer surrounding the hepatic triad (portal vein, hepatic artery, and bile duct), sometimes extending into the interlobular spaces. Notably, liver sections from all antibody groups showed much lower fibrosis, if any. Analysis of plasma liver function markers AST and ALT confirmed these observations (see Fig. 27 ). In fact, animals treated with agonistic anti-MET antibodies showed abnormally low AST and ALT plasma concentrations, which were approximately 2.5-fold lower than those in the PBS group and 2-fold lower than the average AST and ALT levels in normal mice.

[0417] These data suggest that human / mouse equivalent agonistic anti-MET antibodies may be used clinically to treat NASH or other pathological conditions associated with fatty liver. Agonistic anti-MET antibodies may be used to inhibit lipid accumulation in hepatocytes, prevent or reverse hepatic hyperplasia, and inhibit the vicious cycle between fatty acid accumulation and macrophage infiltration. Chronic inflammation always leads to the deposition of extracellular matrix. Therefore, agonistic anti-MET antibodies may also be used to reduce fibrosis associated with steatosis.

[0418] Example 24: In vivo activity: wound healing in diabetic mice

[0419] Clinically relevant diabetic complications are manifested as increased ulcers and impaired wound healing. Since HGF is associated with wound healing (Nakamura et al., Journal of Gastroenterology and Hepatology, 1, 188-202 pages, 2011 (Nakamura et al., J Gastroenterol Hepatol. 1, 188-202, 2011)), we tried to determine whether human / mouse equivalent anti-MET antibodies can promote wound healing in a diabetic background. To this end, we obtained db / db diabetic mice as described above. At 8 weeks of age, we anesthetized the animals and then used a circular perforated blade to cut a 0.8 cm wide circular wound on the right rear side for skin biopsy (GIMA). Remove the entire epidermis. On the second day after surgery, the mice were randomly divided into 4 large groups and treated with purified 71G3, 71D6 and 71G2 or only with a carrier (PBS). The antibody was delivered every other day by intraperitoneal injection at a dose of 5 mg / kg. The wound diameter was measured using a caliper every day.

[0420] like Fig.28 As shown, antibody treatment significantly accelerated wound closure and re-epithelialization. While the control group repaired the experimental wounds at an average rate of 5% per day, this value increased to 8% in the 71G3 group, 12% in the 71D6 group, and 11% in the 71G2 group.

[0421] We propose that human / mouse equivalent agonistic anti-MET antibodies can be used clinically to treat diabetes-related ulcers and wounds that generally exhibit impaired healing. Diabetes-related ulcers represent an unmet medical need. Diabetes is the leading cause of non-traumatic lower limb amputations in the United States. Agonistic anti-MET antibodies can be used to accelerate healing, improve re-epithelialization, and promote vascularization of hyperglycemia-induced ulcers.

[0422] Example 25: Rattus norvegicus (Rattus norvegicus) and cynomolgus monkey (Macaca fascicularis) MET Cross-reactivity

[0423] Since the vast majority of animal models of human diseases use mice as hosts, cross-reactivity with mouse antigens is a prerequisite for antibodies, which need to be validated in preclinical systems. This is the rationale that motivates us to identify human-mouse equivalent anti-MET antibodies. However, some preclinical procedures (such as organ transplantation and other experimental practices requiring complex surgical procedures) are preferably performed in larger rodents or primates. In addition, pharmacodynamic and pharmacokinetic studies are preferably performed in higher vertebrates (usually rats and monkeys). Last but not least, toxicological evaluation of therapeutic antibodies is ideally performed in monkeys; or, if this is not possible, in two different rodent species. Therefore, cross-reactivity with rats and monkeys is also ideally desired.

[0424] To this end, we investigated whether our human / mouse equivalent anti-MET antibodies cross-react with MET from other species, including Rattus norvegicus and Macaca fascicularis. Rat MET ECD (NCBI#NP_113705.1; aa 1-931) and monkey MET ECD (NCBI#XP_005550635.2; aa 1-948) were obtained by standard protein engineering techniques. Human and mouse MET ECDs were used as controls. A restricted panel of antibodies representing SEMA binders (71D6, 71C3, 71D4, 71A3, 71G2) and PSI binders (76H10, 71G3) was selected. The 5D5 prior art antibody was used as a control. MET ECD protein was immobilized on a solid phase (100 ng / well, 96-well plate) and exposed to increasing concentrations (0-40 nM) of antibodies (with human constant regions) in solution. Binding was revealed by using an HRP-conjugated anti-human Fc antibody (Jackson Immuno Research Laboratories). Fig.29As shown, all human / mouse equivalent antibodies tested bind human, mouse, rat and monkey MET with similar affinity and capacity, while 5D5 binds only human and simian MET. We conclude that the 71D6, 71C3, 71D4, 71A3, 71G2, 76H10 and 71G3 antibodies bind human MET, mouse MET, rat MET and simian MET with similar affinity and capacity, at least as determined by ELISA.

[0425] Example 26: Fine epitope mapping

[0426] To fine-tune the MET epitope recognized by human / mouse equivalent anti-MET antibodies, we adopted the following strategy. We reasoned that if an antibody raised in Ilama against human MET cross-reacted with mouse MET, then this antibody might recognize residues that are conserved in H. sapiens and M. musculus but not in H. sapiens, M. musculus, and L. glama. The same reasoning can be extended to R. norvegicus and M. fascicularis.

[0427] To investigate along this line, we aligned and compared the amino acid sequences of human (UniProtKB#P08581; aa 1-932), mouse (UniProtKB#P16056.1; aa 1-931), rat (NCBI#NP_113705.1; aa 1-931), cynomolgus monkey (NCBI#XP_005550635.2; aa 1-948), and alpaca MET (GenBank#KF042853.1; aa 1-931) with each other ( Fig.30 ). With reference to Table 12, we focus on the region of MET responsible for binding to 71D6, 71C3, 71D4, 71A3 and 71G2 antibodies (aa 314-372 of human MET) and binding to 76H10 and 71G3 antibodies (aa 546-562 of human MET). In the front region of human MET (aa 314-372), there are 5 residues (Ala 327, Ser 336, Phe 343, Ile 367, Asp 372) that are conserved in human and mouse MET but not in alpaca MET. Fig.30In the figure, these amino acids are indicated by black boxes and progressive numbers 1-5. Among them, four residues (Ala 327, Ser 336, Ile367, Asp 372) are also conserved in rat and cynomolgus monkey MET. In the back region of human MET (aa 546-562), there are three residues (Arg 547, Ser 553, Thr555) that are conserved in human and mouse MET but not in alpaca MET. Fig.30 In the figure, these amino acids are indicated by black boxes and progressive numbers 6 to 8. Among them, two residues (Ser 553 and Thr 555) are also conserved in rat and cynomolgus monkey MET.

[0428] Using human MET as a template, we mutagenized each of these residues in different arrangements, thereby generating a series of fully human MET mutants except for a specific residue (Ilama). A schematic diagram of the mutants is shown in Fig.31 . Next, we tested the affinity of the selected SEMA binding mAbs (71D6, 71C3, 71D4, 71A3, 71G2) and PSI binding mAbs (76H10 and 71G3) to these MET mutants by ELISA. To this end, various MET proteins were fixed to a solid phase (100 ng / well, 96-well plate) and then exposed to increasing concentrations of antibody (0-50 nM) solutions. Since the antibodies used were in their human constant region form, HRP-conjugated anti-human Fc secondary antibodies (Jackson Immuno Research Laboratories) were used to reveal the binding. Wild-type human MET was used as a positive control. The results of this analysis are listed in Table 24.

[0429] Table 24: Epitopes of MET responsible for agonistic antibody binding Represents residues conserved in Homo sapiens, Mus musculus, Rattus norvegicus, Macaca fascicularis but not conserved between the same species and Ilama. Residues conserved in human, mouse, rat, Macaca fascicularis but not conserved in Ilama MET were tested for relevance to binding to agonistic mAbs by ELISA. Wild-type (WT) or mutant (MT) human MET ECD was immobilized in a solid phase and exposed to increasing concentrations of mAbs in solution. Binding was revealed using an anti-human Fc secondary antibody. All binding values ​​were normalized to WT protein and expressed as percent binding compared to WT MET (E MAX ). Each mutant (AL) contains at least 2 mutations (1-8), such as Fig.31 shown.

[0430]

[0431] The results presented above provide a clear and unambiguous picture of the relevant residues for binding of our agonistic antibodies.

[0432] All tested SEMA binders (71D6, 71C3, 71D4, 71A3, 71G2) appear to bind to the same epitope, which contains two key amino acids located within blade 5 of the SEMA β-propeller that are conserved in human, mouse, cynomolgus monkey, and rat MET but not in alpaca MET: Ile 367 and Asp 372. In fact, mutations of Ala 327, Ser 336, or Phe 343 did not affect binding at all; mutations of Ile 367 partially impaired binding; and mutations of Ile 367 and Asp 372 completely abolished binding. We conclude that Ile 367 and Asp 372 of human MET are critical for binding of the SEMA-directed antibodies tested. These three residues are Fig.30 Indicated by "S" (for SEMA).

[0433] In addition, the tested PSI binders (76H10, 71G3) appear to bind to the same epitope. However, in contrast to the SEMA epitope, the PSI epitope contains only one key amino acid that is also conserved in human, mouse, cynomolgus monkey and rat MET but not in alpaca MET: Thr 555. In fact, mutations of Arg 547 or Ser 553 did not affect binding at all, whereas mutations of Thr 555 completely abolished binding. We conclude that Thr 555 represents a crucial determinant for the binding of the tested PSI-directed antibodies. This residue is Fig.30 It is denoted by "P" (for PSI).

[0434] Example 27: Uniqueness of human / mouse equivalent agonistic antibodies

[0435] The fine epitope mapping results presented in Example 26 provide molecularly detailed demonstration that the agonistic antibodies provided by the present invention possess unique features not possessed by any prior art molecules. This uniqueness can best be understood by performing the following analysis.

[0436] For most of the prior art antibodies discussed in Examples 12-14, there is no available information about the precise epitopes they recognize on MET. However, we know that these epitopes must be different from those recognized by our antibodies, as none of the prior art molecules cross-react with mouse MET. An instructive example of this diversity is provided by 5D5 / Onartuzumab, which is the only anti-MET antibody in the prior art annotated with detailed molecular information about its interaction with MET. 5D5 / Onartuzumab recognizes 4 different residues located within blade 5 of the SEMA β-propeller, very close to the amino acids responsible for interaction with our SEMA binding antibody (Merchant et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013). Fig.30 These residues indicated by "O" (for Onartuzumab) in FIG. 3 correspond to Gln 328, Arg 331, Leu 337, and Asn 338.

[0437] It is noteworthy that these residues are not conserved between Homo sapiens and Mus musculus. This is fully consistent with the concept of using mice as a host for the generation of 5D5 / Onartuzumab and explains why it does not cross-react with mouse MET (Merchant et al. (supra) and our Figure 6 and Fig.29 ). Furthermore, none of these residues are conserved between Homo sapiens and Rattus norvegicus, but they are conserved between Homo sapiens and Macaca fascicularis. This explains why 5D5 / Onartuzumab does not bind rat MET, but it does bind Macaca fascicularis MET ( Fig.29 ).

[0438] In contrast to 5D5 / Onartuzumab and all other prior art molecules discussed herein, our human / mouse equivalent antibodies were generated using llamas as hosts and were explicitly screened for their ability to cross-react with human and mouse MET. Since most of the few amino acids that are conserved between Homo sapiens and Mus musculus but not in Homo sapiens, Mus musculus, and Ilamas ( Fig.30 ) are also conserved in Rattus norvegicus and Mus musculus, and chance determined that the selected antibodies also cross-reacted with rats and monkeys.

[0439] In summary, both the immunization strategy and the screening design make the antibodies of the present invention unique. On the one hand, the species used for immunization (L. glama) is sufficiently distant from Homo sapiens, Mus musculus, Rattus norvegicus and Macaca fascicularis to ensure that there are sufficient mismatches in the amino acid sequence of L. glama MET compared to MET from other species (see Fig.30 ). These mismatches are crucial because the immunized host cannot generate antibodies against epitopes that it recognizes as "self". On the other hand, the human / mouse double screening protocol forces the selection of antibodies that recognize epitopes that are conserved between these two species. This step is also necessary because in the absence of two species panning, one would simply select the most representative antibodies or those that show higher affinity, but not necessarily cross-reactivity. The introduction of these two criteria (the fifth species and the 'double dipping' protocol) allowed us to identify antibodies with new and unique characteristics.

[0440] References

[0441] Anstee QM and Goldin RD, Int J Exp Pathol. 87, 1-16, 2006

[0442] Basilico C et al., J Biol. Chem. 283: 21267-21227, 2008

[0443] Basilico C et al., J Clin Invest. 124, 3172-3186, 2014

[0444] Cassano M et al., PLoS One 3, e3223, 2008

[0445] Chomczynski P et al., Anal. Biochem. 162: 156-159, 1987

[0446] Daley LP et al., Clin. Vaccine Immunol. 12, 2005

[0447] de Haard H et al., J Biol Chem. 274:18218-18230, 1999

[0448] De Haard H et al., J. Bact. 187:4531-4541, 2005

[0449] Fafalios A et al. Nat Med. 17, 1577-1584, 2011

[0450] Forte G et al., Stem Cells. 24, 23-33, 2006

[0451] Hultberg A et al., Cancer Res. 75, 3373-3383, 2015

[0452] Ido A et al., Hepatol Res. 30, 175-181, 2004

[0453] Jones-Hall YL and Grisham MB, Pathophysiology 21, 267-288, 2014

[0454] Kim JJ et al., J Vis Exp. 60, pii:3678, 2012

[0455] Kosone T et al., Am J Physiol Gastrointest Liver Physiol. 293, G204-210, 2007

[0456] Kroy DC et al. J Hepatol. 61, 883-890, 2014

[0457] Longati P et al., Oncogene 9, 49-57, 1994

[0458] Matsumoto K and Nakamura T,Ciba Found Symp.212,198-214,1997

[0459] Medico E et al., Mol Biol Cell 7, 495-504, 1996

[0460] Merchant M et al., Proc Natl Acad Sci USA 110, E2987-2996, 2013

[0461] Michieli P et al. Nat Biotechnol. 20, 488-495, 2002

[0462] Mizuno S et al. Front Biosci. 13, 7072-7086, 2008

[0463] Nakamura T et al., J Gastroenterol Hepatol. 26 Suppl 1, 188-202, 2011

[0464] Perdomo G et al., J Biol Chem. 283, 13700-13706, 2008

[0465] Petrelli A et al., Proc Natl Acad Sci USA 103, 5090-9095, 2006

[0466] Pietronave S et al., Am J Physiol Heart Circ Physiol. 298, H1155-65, 2010

[0467] Prat M et al., Mol Cell Biol. 11, 5954-5962, 1991

[0468] Prat M et al., J Cell Sci. 111, 237-247, 1998

[0469] Rosário M and Birchmeier W, Trends Cell Biol. 13, 328-335, 2003

[0470] Stamos J et al., EMBO J. 23, 2325-2335, 2004

[0471] Takahara et al., Hepatology, 47, 2010-2025, 2008

[0472] Wang B et al. Curr Diabetes Rev. 10, 131-145, 2014

Claims

1. An anti-MET antibody or an antigen-binding fragment thereof, comprising a heavy chain variable domain and a light chain variable domain, wherein the heavy chain variable domain comprises H-CDR1, H-CDR2 and H-CDR3, and the light chain variable domain comprises L-CDR1, L-CDR2 and L-CDR3, wherein: H-CDR1 consists of the amino acid sequence shown in SEQ ID NO:2; H-CDR2 consists of the amino acid sequence shown in SEQ ID NO:4; H-CDR3 consists of the amino acid sequence shown in SEQ ID NO:6; L-CDR1 consists of the amino acid sequence shown in SEQ ID NO:79; L-CDR2 consists of the amino acid sequence shown in SEQ ID NO:81; and L-CDR3 consists of the amino acid sequence shown in SEQ ID NO:

83.

2. The anti-MET antibody or antigen-binding fragment thereof according to claim 1, wherein The heavy chain variable domain consists of a sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 155; and the light chain variable domain consists of a sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:

156.

3. The anti-MET antibody or antigen-binding fragment thereof according to claim 1 or 2, which meets one or more of the following conditions: - the anti-MET antibody or antigen-binding fragment thereof contains the hinge region, CH2 domain and / or CH3 domain of human IgG; - the anti-MET antibody or antigen-binding fragment thereof has a sequence that is at least 90%, 95%, 97% or 99% identical to human IgG; - the heavy chain variable domain and / or light chain variable domain or one or more CDRs of the anti-MET antibody or antigen-binding fragment thereof are derived from camelids; - the heavy chain variable domain and / or light chain variable domain or one or more CDRs of the anti-MET antibody or antigen-binding fragment thereof are derived from camelids, wherein The animal is a llama.

4. The anti-MET antibody or antigen-binding fragment thereof according to claim 3, wherein The IgG is IgG1.

5. The anti-MET antibody or antigen-binding fragment thereof according to claim 3, wherein The CH2 and / or CH3 domains have been modified to reduce or substantially eliminate one or more antibody effector functions.

6. An affinity variant or human germline variant of the anti-MET antibody or antigen-binding fragment thereof according to claim 1 or 2.

7. An isolated polynucleotide encoding the anti-MET antibody or antigen-binding fragment thereof according to claim 1 or 2.

8. An expression vector comprising the polynucleotide of claim 7 operably linked to a regulatory sequence that allows the anti-MET antibody or antigen-binding fragment thereof to be expressed in a host cell or a cell-free expression system.

9. A host cell or a cell-free expression system, comprising the expression vector according to claim 8.

10. A method for producing a recombinant anti-MET antibody or an antigen-binding fragment thereof, the method comprising culturing the host cell or cell-free expression system according to claim 9 under conditions allowing expression of the anti-MET antibody or the antigen-binding fragment thereof and recovering the expressed anti-MET antibody or the antigen-binding fragment thereof.

11. A pharmaceutical composition comprising the anti-MET antibody or antigen-binding fragment thereof according to claim 1 or 2 and at least one pharmaceutically acceptable carrier or excipient.

12. Use of the anti-MET antibody or antigen-binding fragment thereof according to claim 1 or 2 or the pharmaceutical composition according to claim 11 in the preparation of a medicament for the following diseases: - Treating or preventing liver damage in human patients; - Treating or preventing renal damage in human patients; - treating or preventing inflammatory bowel disease in human patients; - Treating or preventing diabetes in human patients; - treating or preventing non-alcoholic steatohepatitis in a human patient; or - Treating or promoting wound healing in human patients.

13. The use according to claim 12, wherein: - The liver injury is acute liver injury or chronic liver injury; - The renal injury is acute renal injury; - The inflammatory bowel disease is ulcerative colitis; - the diabetes is type I diabetes or type II diabetes; or - The human patient is a patient suffering from diabetes.

Images