Joining Members

Abstract

The present invention relates to Lewis Y (Le y The present invention also relates to the use of said binding members in medicine and to nucleic acids encoding said binding members.

Description

[Technical field]

[0001] The present invention relates to Lewis Y. y The present invention also relates to the use of said binding members in medicine and to nucleic acids encoding said binding members. [Background technology]

[0002] Glycan structures are present in both protein and glycolipid backbones and can be significantly overexpressed in cancer due to altered expression of glycosyltransferases. Glycolipids consist of a lipid tail with a carbohydrate head and constitute 5% of lipid molecules in the outer monolayer. There are three major series of glycolipids: globo, neolacto, and ganglio lipids. Glycolipids are assumed to be very good targets due to their high surface density, mobility, and association with membrane microdomains, all of which contribute to strong cellular interactions. However, the generation of anti-glycolipid antibodies is a challenging task, since there is no T-cell help and monoclonal antibodies (mAbs) are usually of low affinity and IgM subclass (Durrant, Noble et al. 2012, Rabu, McIntosh et al. 2012). mAbs against glycans expressed on proteins overcome this issue, but they present new challenges because mAbs rarely see only small glycans and usually recognize glycans on specific proteins, providing very restricted expression.

[0003] Only a limited number of antibodies that recognize glycans have been described. Several anti-Lewis (Le) carbohydrate antigen mAbs have been generated to date, but these often have cross-reactivity with a range of glycans expressed in normal tissues. Lewis carbohydrate antigens are formed by the sequential addition of fucose onto chains of oligosaccharide precursors on glycoproteins or glycolipids through the action of a set of glycosyltransferases (Yuriev, Farrugia et al. 2005). They are divided into two groups: type I (Le a and Le b ) and type II (Le x and Le y ) can be divided into Le a and Le b The antigens are considered to be blood group antigens, and the Le x and Le y Lewis Y has been reported to be expressed only in certain epithelial cells (Scott, Geleick et al. 2000), and is considered a tumor-associated marker (Soejima and Koda 2005), representing an attractive target for cancer treatment, including antibody-based immunotherapy (Scott, Geleick et al. 2000). Most normal tissues do not express Lewis Y, except epithelial cells of the gastrointestinal tract (which, however, express it at very low density) (Schuster, Umana et al. 2005). However, Lewis Y (Fuca1-2Galb1-4(Fuca1-3)GlcNAc) has been reported to be expressed at high density in almost 70% of human epithelial cancers, and is therefore an attractive candidate target for cancer immunotherapy. The recognition of Lewis Y as a good therapeutic target has led to the development of many anti-Lewis Y antibodies. y These anti-Le y Some of the mAbs target other Lewis antigens, e.g. b / y It cross-reacts with mAb 692 / 29, which recognizes IL-1, and has demonstrated an antitumor response against colon tumors in xenograft models (Noble, Spendlove et al. 2013).

[0004] Le y Several other mAbs that bind to Lewis antigens are known in the art, but these cross-react with other related Lewis antigens. y and B-7-2. B-7-2 has also been shown to be associated with tumor cells (European Patent Publication No. 0285059). U.S. Patent No. 5,869,045 to Lewis et al. y and Lewis x Lewis et al. disclose an antibody, BR-96, which binds to both haptens. y and Lewis b Antibodies that bind to both of these antigens are known. The C14 monoclonal antibody is Lewis y and Lewis b Studies have shown that the antibody recognizes and binds to both extended and non-extended forms of antigens (Durrant at al., Hybridoma, 12, 647-660 (1996)). International Patent Publication No. WO 02 / 092126 describes the antibody binding to Lewis y and Lewis b The present inventors disclose mAbs that bind to both haptens of the above.

[0005] Mouse or chimeric anti-Le y There are many clinical trials using antibodies (Scott, Geleick et al. 2000). Unfortunately, in most cases, the antibodies are y The mAb cross-reacts with Lewis X and HII structures that are more commonly expressed in normal tissues (Furukawa, Welt et al. 1990, Yin, Finstad et al. 1996). y Antibodies have also been identified, but a consistent problem with these antibodies is the Le x, and the degree of cross-reactivity with H2 structures, which causes red blood cell agglutination and gastrointestinal toxicity (Kitamura, Stockert et al. 1994, Pai, Wittes et al. 1996, Tolcher, Sugarman et al. 1999). y The fine specificity of the antibody is important for successful clinical treatment. Summary of the Invention

[0006] In a first aspect, the present invention provides Fuca1-2Galb1-4(Fuca1-3)GlcNAc(Le y The present invention provides an isolated specific binding member capable of specifically binding to the .

[0007] The inventors have provided specific binding members that display potent anti-tumour activity in vivo. The specific binding members of the invention display potent immune-mediated cytotoxic activity against human colon cancer cells in vitro via antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC). They also have the property of directly inducing cell death without the need for immune effector cells. By immunising with glycolipids rather than tumour glycoproteins or whole cells, the inventors have generated ultraspecific anti-Leu antibodies that do not cross-react with any other glycans. y Examples of mAbs have been generated, including an IgG3k murine mAb known herein as "FG27.10" and an IgG1k murine mAb known herein as "FG27.18", a chimeric hIgG1 (known as "CH27"), and a humanized hIgG1 (known as "Hu27").

[0008] Anti-Leu cross-reacts with other glycans yIn contrast to previous studies with mAbs, the specific binding members of the present invention have a very restricted normal distribution and bind only to subpopulations of cells in normal stomach, lung, tonsil, pancreas, and duodenum. Furthermore, they failed to stain colon, jejunum, breast, kidney, or ileum. They bind strongly to a broad range of tumors, including 80% of gastric tumors, 100% of colorectal tumors, 95% of ovarian tumors, and 85% of breast tumors. The specific binding members of the present invention, e.g., FG27.10 / 18, bind to Le y The inventors have shown that the specific binding member according to the invention was unable to bind to the cell line Colo21. In contrast, y / x mAb binds strongly and y / b The mAb bound moderately, indicating that Colo201 binds to Le x and Le b However, Le y The results show that FG27 does not express any of the cross-reactive Les in the human IgG1-specific binding members, confirming the specificity of the specific binding members of the invention. This specificity is reflected in the similar but distinct sequences of the FG27 variable heavy and light chain regions (Figure 13a, b). FG27 does not express any of the cross-reactive Les in the human IgG1-specific binding members of ... y The heavy chain CDRs show 16-68% amino acid sequence differences, the light chain CDR sequences show 4.3-16% amino acid sequence differences, the heavy chain variable region amino acid sequence shows 8.6-49.5% amino acid sequence differences, and the light chain variable region amino acid sequence shows 2-24% amino acid sequence differences from the different mAbs (see Table 4 herein). Most of the variation is in CDRH2, with 3-5 of the 8 amino acids varying between the different mAbs.

[0009] Specific binding members of the invention may bind to lipid or protein backbones. The protein backbone may have a molecular weight of approximately 50-150 kDa, as determined by SDS-PAGE.

[0010] Specific binding members of the invention preferably comprise one or more binding domains selected from binding domains having an amino acid sequence substantially as set out as residues 27-38 (CDRH1), 54-65 (CDRH2) or 105-116 (CDRH3) of Figure 1a or 1b. A specific binding member may comprise a binding domain comprising an amino acid sequence substantially as set out as residues 105-116 (CDRH3) of the amino acid sequence of Figure 1a or 1b. Such specific binding members may further comprise one or both, preferably both, of binding domains having an amino acid sequence substantially as set out as residues 27-38 (CDRH1) and residues 56-65 (CDRH2) of the amino acid sequence shown in Figures 1a and 1b.

[0011] A binding member may comprise an amino acid sequence substantially as set out as 1-127(VH) of Figure 1a or 1b.

[0012] A specific binding member may comprise one or more binding domains selected from binding domains having the amino acid sequence of residues 27-38 (CDRL1), 56-65 (CDRL2), or 105-113 (CDRL3) of Figure 1c. A binding member may comprise a binding domain having an amino acid sequence substantially as set out as residues 105-113 (CDRL3) of the amino acid sequence of Figure 1c. Such a specific binding member may further comprise one or both, preferably both, of binding domains having an amino acid sequence substantially as set out as residues 27-38 (CDRL1) and residues 56-65 (CDRL2) of the amino acid sequence shown in Figure 1c.

[0013] Specific binding members comprising multiple binding domains of the same or different sequence, or combinations thereof, are included within the scope of the present invention. At least one binding domain may be carried by a human antibody framework. For example, one or more binding domains may be replaced with the complementarity determining regions (CDRs) of a whole human antibody or its variable regions.

[0014] One isolated specific binding member of the invention comprises an amino acid sequence substantially as set out as residues 1 to 124 (VL) of the amino acid sequence shown in Figure 1c.

[0015] Isolated specific binding members of the invention may comprise one or more, preferably all, of a binding domain having the amino acid sequence substantially as set out as residues 27-38 (CDRH1), 54-65 (CDRH2) or 105-116 (CDRH3) of Figure 1a or 1b combined with one or more, preferably all, of a binding domain having the amino acid sequence substantially as set out as residues 27-38 (CDRL1), 56-65 (CDRH2) or 105-113 (CDRL3) of Figure 1c.

[0016] A binding member may comprise an amino acid sequence substantially as set out as residues 1 to 127 (VH) of the amino acid sequence of Figure 1a or 1b, and an amino acid sequence substantially as set out as residues 1 to 124 (VL) of the amino acid sequence of Figure 1c.

[0017] Since a single prototype mAb, e.g., FG27 mAb, with the desired properties described herein has been isolated, it is easy to generate other mAbs with similar properties by using methods known in the art. For example, the method of Jespers et al., 1994 (Jespers, Roberts et al. 1994) can be used to guide the selection of mAbs with the same epitope and therefore similar properties as the prototype mAb. Using phage display, the first heavy chain of the prototype antibody is paired with a repertoire of (preferably human) light chains to select mAbs that bind to glycans, and then the new light chains are paired with a repertoire of (preferably human) heavy chains to select mAbs that bind to (preferably human) glycans with the same epitope as the prototype mAb.

[0018] The specific binding member may be an antibody or antibody fragment, Fab, (Fab')2, scFv, Fv, dAb, Fd, or diabody. The antibody may be a polyclonal antibody. The antibody may be a monoclonal antibody (mAb). The antibody of the invention may be a humanized antibody, a chimeric antibody, or a veneered antibody, or may be a non-human antibody of any species.

[0019] Murine or chimeric antibodies pose a high risk of harmful anti-mouse antibody (HAMA) reactions in patients (Schroff et al. 1985; Azinovic et al. 2006; Miotti et al. 1999; D'Arcy and Mannik 2001). Therefore, most approved therapeutic mAbs are humanized or fully human IgG antibodies.

[0020] A specific binding member of the present invention may comprise a heavy chain having an amino acid sequence substantially as shown in Figure 2a and a light chain having an amino acid sequence substantially as shown in Figure 2b.

[0021] A specific binding member of the present invention may comprise a heavy chain having an amino acid sequence substantially as shown in Figure 3a and a light chain having an amino acid sequence substantially as shown in Figure 3b.

[0022] A specific binding member of the present invention may comprise a heavy chain having an amino acid sequence substantially as set out in FIG. 33 and / or a light chain having an amino acid sequence substantially as set out in FIG.

[0023] The present invention further relates to an antibody comprising a VH chain having the amino acid sequence of residues 1 to 127 of FIG. 1a or 1b and a VL chain having the amino acid sequence of residues 1 to 124 of FIG. 1c, and y Binding members that compete for binding to the contained glycan are provided.

[0024] Le yIncluded in the invention are specific binding members which are capable of specifically binding to Le and which are at least 90%, at least 95%, or at least 99% identical in the VH and / or VL domain to the VH or VL domain of Figures 1, 2 or 3. y Included in the invention are specific binding members capable of specifically binding to and which are at least 90%, at least 95%, or at least 99% identical to the heavy and / or light chains of Figure 33. Preferably, such antibodies differ from the sequences of Figures 1, 2, or 3 by small numbers of functionally insignificant amino acid substitutions (e.g. conservative substitutions), deletions, or insertions.

[0025] Specific binding members of the present invention may carry a detectable or functional label.

[0026] In a further aspect, the present invention provides an isolated nucleic acid encoding a specific binding member of the present invention and a method of preparing a specific binding member of the present invention comprising expressing said nucleic acid under conditions which result in expression of said binding member and recovering the binding member. y Isolated nucleic acids encoding specific binding members capable of specifically binding to and which are at least 90%, at least 95%, or at least 99% identical to the sequences provided herein are encompassed by the present invention.

[0027] The specific binding members of the invention may be used in methods of treatment or diagnosis of the human or animal body, for example a method of treating a tumour in a patient, preferably a human, comprising the step of administering to said patient an effective amount of a specific binding member of the invention. The invention also provides specific binding members of the invention for use in medicine, preferably for use in the treatment of tumours, and the use of a specific binding member of the invention in the manufacture of a medicament for the diagnosis or treatment of a tumour. The tumour may be a gastric, colorectal, pancreatic, lung, ovarian or breast tumour.

[0028] Disclosed herein are antigens to which the specific binding members of the invention bind. y may be provided. y may be provided in isolated form and used in screening to develop additional specific binding members thereto. For example, a library of compounds may be prepared by screening Le y The method can be screened for members of the library that specifically bind to Le. y The nucleosome may be on a lipid or protein backbone. If on a protein backbone, it may have a molecular weight of about 50-150 kDa, as determined by SDS-PAGE.

[0029] In a further aspect, the present invention relates to a method for the diagnosis or prognosis of gastric, colorectal, pancreatic, lung, ovarian and breast tumours, preferably comprising administering to said patient a therapeutically effective amount of the Le y An isolated specific binding member capable of binding to the contained glycan is provided.

[0030] Furthermore, the present invention relates to a method for the detection of Le y A method of diagnosing cancer is provided which involves using a specific binding member of the invention to detect contained glycans, where the pattern of glycans detected by the binding member can be used to stratify treatment options for an individual. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0031] These and other aspects of the invention are described in further detail below.

[0032] As used herein, a "specific binding member" is a member of a pair of molecules that have binding specificity for each other. The members of a specific binding pair may be naturally occurring or wholly or partially synthetically produced. One member of the pair of molecules has an area on its surface, which may be a protrusion or a cavity, that specifically binds to, and is therefore complementary to, a particular spatial and polar organization of the other member of the pair of molecules. Thus, the members of the pair have the property of specifically binding to each other. Examples of types of specific binding pairs are antigen-antibody, biotin-avidin, hormone-hormone receptor, 5 receptor-ligand, enzyme-substrate. The present invention generally relates to antigen-antibody type reactions, but it also concerns small molecules that bind to antigens as defined herein.

[0033] As used herein, "treatment" includes any regime that can benefit a human or non-human animal, preferably a mammal. Treatment may be in respect of an existing condition or may be preventative (prophylactic treatment).

[0034] As used herein, a "tumor" is an abnormal growth of tissue. It may be localized (benign) or may invade nearby tissues (malignant) or distant tissues (metastatic). Tumors include neoplastic growths that cause cancer, as well as esophageal, colorectal, gastric, breast, ovarian, and endometrial tumors, and cancerous tissues or cell lines, including, but not limited to, white blood cells. As used herein, "tumor" also includes within its scope endometriosis.

[0035] The term "antibody" as used herein refers to immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, i.e., molecules that contain an antigen binding site that specifically binds an antigen, whether natural or partially or wholly synthetically produced. The term also encompasses all polypeptides or proteins having a binding domain that is or is homologous to an antibody binding domain. They may be derived from natural sources or they may be partially or wholly synthetically produced. Exemplary antibodies of the invention include immunoglobulin isotypes (e.g., IgG, IgE, IgM, IgD, and IgA) and their isotypic subclasses; fragments that contain the antigen binding domain, e.g., Fab, scFv, Fv, dAb, Fd; and diabodies. Antibodies can be polyclonal or monoclonal. Monoclonal antibodies can be designated as "mAbs."

[0036] It is possible to take monoclonal and other antibodies and use recombinant DNA technology to produce other antibodies or chimeric molecules that retain the specificity of the original antibody. Such technology may involve introducing DNA encoding the variable region of an immunoglobulin or the CDRs of an antibody into the constant region or constant region plus framework regions of a different immunoglobulin. See, for example, EP 184187, GB 2188638, or EP 239400. Hybridomas or other cells that produce the antibodies may be subject to genetic mutations or other changes that may or may not alter the binding specificity of the antibodies produced.

[0037] Since antibodies can be modified in many ways, the term "antibody" should be interpreted to encompass all specific binding members or substances that have a binding domain with the required specificity. The term thus includes antibodies, antibody fragments, derivatives, functional equivalents, and homologs of humanized antibodies, including all polypeptides that contain an immunoglobulin binding domain, whether natural or wholly or partially synthetic. Thus, chimeric molecules or equivalents that contain an immunoglobulin binding domain fused to another polypeptide are included. Cloning and expression of chimeric antibodies are described in European Patent Publications 0120694 and 0125023. Humanized antibodies can be modified antibodies that have the variable regions of a non-human, e.g., mouse, antibody and the constant regions of a human antibody. Methods for making humanized antibodies are described, for example, in U.S. Pat. No. 5,225,539.

[0038] It has been shown that fragments of a whole antibody can perform the function of binding to an antigen. Examples of binding fragments are: (i) a Fab fragment consisting of the VL, VH, CL, and CH1 domains; (ii) an Fd fragment consisting of the VH and CH1 domains; (iii) an Fv fragment consisting of the VL and VH domains of a single antibody; (iv) a dAb fragment consisting of the VH domain (Ward, Gussow et al. 1989); (v) an isolated CDR region; (vi) a F(ab)'2 fragment, which is a bivalent fragment containing two linked Fab fragments; (vii) a single chain Fv molecule (scFv), in which the VH and VL domains form an antigen binding site by linking them together with a peptide linker that connects the two domains (Bird, Hardman et al. 1988, Huston, Levinson et al. 1988); (viii) bispecific single chain Fv dimers (WO 92 / 09965); and (ix) "diabodies", multivalent or multispecific fragments constructed by gene fusion (WO 94 / 13804; (Holliger, Prospero et al. 1993)).

[0039] Diabodies are multimers of polypeptides in which each polypeptide comprises a first domain comprising the binding region of an immunoglobulin light chain and a second domain comprising the binding region of an immunoglobulin heavy chain, and the two domains are linked (e.g., by a peptide linker) but cannot bind to each other to form an antigen-binding site: the antigen-binding site is formed by the binding of the first domain of one polypeptide in the multimer to the second domain of another polypeptide in the multimer (WO 94 / 13804).

[0040] When bispecific antibodies are used, they may be conventional bispecific antibodies that can be produced in a variety of ways (Holliger and Winter 1993), for example prepared chemically or from hybrid hybridomas, or may be any of the bispecific antibody fragments mentioned above. It may be preferable to use scFv dimers or diabodies rather than whole antibodies. Diabodies and scFvs can be constructed without an Fc region using only the variable domains, presumably reducing the effects of anti-idiotypic reactions. Other forms of bispecific antibodies include the single chain "Janusins" described in (Traunecker, Lanzavecchia et al. 1991).

[0041] Bispecific diabodies may also be useful because, in contrast to bispecific whole antibodies, they can be readily constructed and expressed in E. coli. Diabodies (and many other polypeptides, e.g., antibody fragments) of appropriate binding specificity can be readily selected from libraries using phage display (WO 94 / 13804). If one arm of the diabody remains constant, e.g., with specificity for antigen X, this allows the creation of libraries in which the other arm is varied and antibodies of appropriate specificity can be selected.

[0042] A "binding domain" is a portion of a specific binding member that contains an area that specifically binds and is complementary to part or all of an antigen. If the binding member is an antibody or an antigen-binding fragment thereof, the binding domain may be a CDR. If the antigen is large, the antibody may only bind to a specific part of the antigen, this part being called an epitope. An antigen-binding domain may be provided by one or more antibody variable domains. An antigen-binding domain may include an antibody light chain variable region (VL) and an antibody heavy chain variable region (VH).

[0043] "Specific" is generally used to describe the situation where one member of a specific binding pair does not exhibit any significant binding to molecules other than its specific binding partner, e.g., has less than about 30%, preferably less than 20%, less than 10%, or less than 1% cross-reactivity with any other molecule. The term is also applicable, for example, when an antigen-binding domain is specific for a particular epitope carried by many antigens, in which case the specific binding member carrying the antigen-binding domain can bind to a variety of antigens carrying the epitope. A specific binding member of the present invention exhibits a binding activity against the glycan of Leu in the absence of detectable binding to any other antigen (e.g., any other glycan) when binding is tested by the protocol presented in the "Glycome Analysis" section of the Examples herein. y In some cases, the antibody may be capable of specifically binding to the

[0044] "Isolated" refers to the situation where a specific binding member of the invention or a nucleic acid encoding said binding member is preferably according to the invention. Generally, the members and nucleic acids are free or substantially free from naturally associated substances such as other polypeptides or nucleic acids found in their natural environment or the environment (e.g. cell culture) in which they are prepared (where said preparation is by recombinant DNA techniques carried out in vitro or in vivo). The specific binding members and nucleic acids may be formulated with a diluent or adjuvant, or may even be isolated for practical purposes - for example, the members will typically be mixed with gelatin or other carrier when used to coat microtiter plates for use in immunoassays, or mixed with a pharma- ceutically acceptable carrier or diluent when used diagnostically or therapeutically. The specific binding members may be glycosylated, either naturally or by heterologous eukaryotic systems, or they may be unglycosylated (for example when produced by expression in prokaryotic cells).

[0045] "Substantially as set out" means that the amino acid sequence of the present invention is identical or significantly homologous to the amino acid sequence described. By "significantly homologous" it is contemplated that 1-5, 1-4, 1-3, 2 or 1 substitutions can be made in the sequence.

[0046] The present invention also includes within its scope polypeptides having the amino acid sequences presented in Figures 1, 2, or 3, polynucleotides having the nucleic acid sequences presented in Figures 1 or 2, and sequences having substantial identity thereto, for example at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 99% identity thereto. The percent identity of two amino acid sequences or two nucleic acid sequences is generally determined by aligning the sequences for optimal comparison (e.g., gaps may be introduced in the first sequence for best alignment with the second sequence) and comparing the amino acid residues or nucleotides at corresponding positions. The "best alignment" is the alignment of the two sequences that results in the highest percent identity. The percent identity is determined by comparing the number of identical amino acid residues or nucleotides in the sequences (i.e., % identity = number of identical positions / total number of positions x 100).

[0047] The determination of percent identity between two sequences can be achieved using a mathematical algorithm known to those skilled in the art. One example of a mathematical algorithm for comparing two sequences is the algorithm of Karlin and Altschul, 1990 (Karlin and Altschul 1990), modified as in Karlin and Altschul, 1993 (Karlin and Altschul 1993). The NBLAST and XBLAST programs of Altschul et al., 1990 (Altschul, Gish et al. 1990) incorporate such an algorithm. BLAST nucleotide searches can be performed with the NBLAST program (score=100, wordlength=12) to obtain nucleotide sequences homologous to the nucleic acid molecules of the present invention. BLAST protein searches can be performed with the XBLAST program (score=50, wordlength=3) to obtain amino acid sequences homologous to the protein molecules of the present invention. To obtain gapped alignments for comparison, Gapped BLAST can be used as described in Altschul et al., 1997 (Altschul, Madden et al. 1997). Alternatively, PSI-Blast can be used to perform an iterated search that detects distant relationships between molecules (Id.). When using BLAST, Gapped BLAST, and PSI-Blast programs, the default parameters of each program (e.g., XBLAST and NBLAST) can be used. See http: / / www.ncbi.nlm.nih.gov. Another example of a mathematical algorithm used for sequence comparison is the algorithm of Myers and Miller, 1989 (Myers and Miller 1989). The ALIGN program (version 2.0), which is part of the GCG sequence alignment software package, incorporates such an algorithm.Other algorithms for sequence analysis known in the art include ADVANCE and ADAM, described in Torellis and Robotti, 1994 (Torelli and Robotti 1994); and FASTA, described in Pearson and Lipman, 1988 (Pearson and Lipman 1988). Within FASTA, ktup is a control option that sets the sensitivity and speed of the search.

[0048] The isolated specific binding members of the present invention are y It can bind to carbohydrates, which is y The binding domain may be a ceramide or may be a protein portion. The binding domain may comprise the amino acid sequence substantially as set forth in residues 105-116 (CDRH3) of Figures 1a and 1b and 105-113 of Figure 1c. y It may be carried in a structure that allows for binding to carbohydrates.

[0049] The structures for carrying the binding domains of the invention are generally the sequences or substantial parts of the heavy or light chains of antibodies in which the binding domains are located at positions corresponding to the CDR3 regions of naturally occurring VH and VL antibody variable domains encoded by rearranged immunoglobulin genes. The structures and positions of immunoglobulin variable domains can be determined by reference to http: / / www.imgt.org / . The amino acid sequence substantially as set out in residues 105-116 of Figures 1a and 1b may be carried as the CDR3 or substantial part of a human heavy chain variable domain, and the amino acid sequence substantially as set out in residues 105-113 of Figure 1c may be carried as the CDR3 or substantial part of a human light chain variable domain.

[0050] The variable domain may be derived from any germline or rearranged human variable domain, or may be a synthetic variable domain based on consensus sequences of known human variable domains. The CDR3-derived sequences of the present invention may be introduced into a repertoire of variable domains lacking CDR3 regions using recombinant DNA technology. For example, Marks et al., 1992 (Marks, Griffiths et al. 1992) describes a method for generating a repertoire of antibody variable domains in which a consensus primer directed to or adjacent to the 5' end of the variable domain region is used in conjunction with a consensus primer against the third framework region of the human VH gene to provide a repertoire of VH variable domains lacking CDR3. Furthermore, Marks et al., 1992 (Marks, Griffiths et al. 1992) describes how this repertoire can be combined with the CDR3 of a particular antibody. Using similar techniques, CDR3 derived sequences of the invention may be shuffled with a repertoire of VH or VL domains lacking CDR3, and the shuffled complete VH or VL domains can be combined with the cognate VL or VH domains to provide a specific binding member of the invention. This repertoire can then be displayed in a suitable host system, such as the phage display system of WO 92 / 01047, from which suitable specific binding members can be selected. The repertoire can be displayed in a suitable host system, such as the phage display system of WO 92 / 01047, from which suitable specific binding members can be selected. 4 or more individual members, e.g. 10 6 ~10 8 or 10 10 The committee may consist of members of

[0051] Similar shuffling or combinatorial techniques have also been disclosed by Stemmer, 1994 (Stemmer 1994), who describes the technique in relation to the β-lactamase gene, but observes that this approach can be used to generate antibodies. A further alternative is to use random mutagenesis of, for example, the FG27 VH or VL gene to generate mutations in the entire variable domain, to generate novel VH or VL regions carrying sequences derived from the CDR3 of the invention. Such a technique has been described by Gram et al., 1992 (Gram, Marconi et al. 1992) using error-prone PCR.

[0052] Another method that can be used is to carry out mutagenesis on the CDR regions of the VH or VL genes. Such techniques are disclosed by Barbas et al., 1994 (Barbas, Hu et al. 1994) and Schier et al., 1996 (Schier, McCall et al. 1996). A substantial portion of an immunoglobulin variable domain generally comprises at least three CDR regions together with their intervening framework regions. This portion may also comprise at least about 50% of either or both of the first and fourth framework regions, the 50% being the C-terminal 50% of the first framework region and the N-terminal 50% of the fourth framework region. Additional residues at the N- or C-terminus of the substantial portion of the variable domain may not be normally associated with naturally occurring variable domain regions. For example, construction of specific binding members of the invention by recombinant DNA techniques may result in the introduction of N- or C-terminal residues encoded by linkers introduced to facilitate cloning or other manipulation steps, including the introduction of linkers to join variable domains of the invention to further protein sequences including immunoglobulin heavy chains, other variable domains (e.g. in the production of diabodies) or protein tags, as discussed in more detail below.

[0053] The present invention provides specific binding members comprising a pair of binding domains based on the amino acid sequences of the VL and VH regions substantially as depicted in Figure 1, namely amino acids 1-127 (VH) of Figure 1a or 1b and amino acids 1-124 (VL) of Figure 1c. Single binding domains based on any of these sequences form further aspects of the invention. In the case of binding domains based on the amino acid sequence of the VH region substantially as depicted in Figures 1a and 1b, such binding domains may be used as targeting agents, since it is known that immunoglobulin VH domains are capable of binding to target antigens in a specific manner. In the case of any of the single chain specific binding domains, these domains may be used to screen for complementary domains capable of forming two-domain specific binding members with in vivo properties as good as or comparable to those of the FG27 antibody disclosed herein.

[0054] This can be achieved by phage display screening methods using the so-called hierarchical dual combinatorial approach disclosed in WO 92 / 01047, in which individual colonies containing either H or L chain clones are used to infect the entire library of clones encoding the other chain (L or H), and specific binding members of the resulting double chain are selected by phage display techniques such as those disclosed in this reference. This technique is also disclosed in Marks et al., 1992 (Marks, Griffiths et al. 1992).

[0055] Specific binding members of the invention may further comprise an antibody constant region or part thereof. For example, specific binding members based on the VL region shown in Figure 1c may bind at their C-terminus to an antibody light chain constant domain. Similarly, specific binding members based on the VH region shown in Figures 1a and 1b may bind at their C-terminus to all or part of an immunoglobulin heavy chain from any antibody isotype, e.g. IgG, IgA, IgE and IgM, and isotype subclasses, in particular IgG1, IgG2 and IgG4.

[0056] Specific binding members of the present invention may be used in methods of diagnosis and treatment of tumours in human or animal subjects.

[0057] When used diagnostically, the specific binding members of the invention may be coupled to a detectable label, e.g. 131 I or 99 The antibodies may be labeled with a radioactive label such as Tc, which may be attached to the specific binding members of the invention using conventional chemical techniques known in the art of antibody imaging. Labels also include enzyme labels such as horseradish peroxidase. Labels further include chemical moieties such as biotin that can be detected via binding to a specific cognate detectable moiety, for example labeled avidin.

[0058] Although the specific binding members of the invention have been shown by themselves to be effective in killing cancer cells, they may further be labelled with a functional label. Functional labels include substances that are designed to target the site of a cancer to effect the destruction of the cancer cell. Such functional labels include toxins such as ricin and enzymes such as bacterial carboxypeptidase or nitroreductase, which can convert a prodrug into an effective drug. Additionally, specific binding members may be labelled with chemotherapeutic or cytotoxic agents such as maytansine (DM1 and DM4), onides, auristatins, calicheamicin, duocamycin, doxorubicin, or 90 Y or 131It may be bound to or otherwise associated with a radioactive label, such as I.

[0059] Furthermore, the specific binding members of the invention may be administered alone or in combination with other treatments, either simultaneously or sequentially depending on the condition being treated. Thus, the invention further provides a product comprising a specific binding member of the invention and an active agent as a combined preparation for simultaneous, separate or sequential use in the treatment of tumours. The active agent may include chemotherapeutic or cytotoxic agents including 5-fluorouracil, cisplatin, mitomycin C, oxaliplatin and tamoxifen, which may act synergistically with the binding members of the invention. Other active agents may include analgesics in suitable doses, such as non-steroidal anti-inflammatory drugs (e.g. aspirin, paracetamol, ibuprofen or ketoprofen) or opiates, such as morphine, or anti-emetics.

[0060] Without wishing to be bound by theory, the property of the binding members of the invention to act synergistically with effective agents to enhance tumor killing may not be due to immune effector mechanisms, but rather to the binding of Leu bound to the cell surface. y This may be a direct result of binding of the binding member to the glycan. Cancer immunotherapy involving antibodies against immune checkpoint molecules has shown efficacy in combination with different immuno-oncology treatment modalities against a variety of malignancies.

[0061] The specific binding members of the present invention are usually administered in the form of a pharmaceutical composition, which may contain at least one component in addition to the specific binding member. In addition to the active ingredient, the pharmaceutical composition may contain pharma- ceutically acceptable excipients, diluents, carriers, buffers, stabilizers, or other substances well known to those skilled in the art. Such substances should be non-toxic and should not interfere with the efficacy of the active ingredient. The exact nature of the carrier or other substances will vary depending on the route of administration, which may be oral or by injection, e.g., intravenous injection. Injection is the primary route for therapeutic administration of the composition, although it is envisaged that delivery via a catheter or other surgical tube may also be used. Some suitable routes of administration include intravenous, subcutaneous, intraperitoneal, and intramuscular administration. Liquid formulations may be utilized after reconstitution from powder formulations.

[0062] For intravenous injection or injection at the affected site, the active ingredient is in the form of a parenterally acceptable aqueous solution that is pyrogen-free and has suitable pH, isotonicity, and stability.Those skilled in the art can easily prepare suitable solutions using isotonic vehicles such as sodium chloride injection, Ringer's solution, lactated Ringer's solution, etc.Preservatives, stabilizers, buffers, antioxidants, and / or other additives may be included as necessary.

[0063] Pharmaceutical compositions for oral administration may be in the form of tablets, capsules, powders, or liquids. Tablets may contain a solid carrier or adjuvant, such as gelatin. Liquid pharmaceutical compositions generally contain a liquid carrier, such as water, petroleum, animal or vegetable oils, mineral oil, or synthetic oil. Physiological saline, dextrose, or other sugar solutions or glycols, such as ethylene glycol, propylene glycol, or polyethylene glycol, may be included. When the formulation is liquid, it may be, for example, a physiological salt solution, including a non-phosphate buffer of pH 6.8-7.6, or a lyophilized powder.

[0064] The composition can also be administered via microspheres, liposomes, other microparticulate delivery systems, or sustained release formulations that are placed in certain tissues, including blood.Suitable examples of sustained release carriers include semipermeable polymer matrices in the form of common articles, such as suppositories or microcapsules.Implantable or microencapsulated sustained release matrices include polylactides (U.S. Pat. No. 3,773,919; European Patent Publication No. 0058481), copolymers of L-glutamic acid and gamma-ethyl-L-glutamate (Sidman, Steber et al. 1983), poly(2-hydroxyethyl-methacrylate). Liposomes containing polypeptides are prepared by well-known methods: DE 3,218, 121A; (Eppstein, Marsh et al. 1985); (Hwang, Luk et al. 1980); EP 0052522; EP 0036676; EP 0088046; EP 0143949; EP 0142541; JP 83-11808; U.S. Patents 4,485,045 and 4,544,545. Essentially, liposomes are small (about 200-800 angstroms) unilamellar type, in which the lipid content is greater than about 30 mol% cholesterol, and the selected ratio is adjusted for optimal rate of polypeptide leakage. The compositions may be administered in a localized manner to the site of a tumor or other desired location, or may be delivered in a manner that targets tumor or other cells.

[0065] The composition is preferably administered to an individual in a "therapeutically effective amount", which is an amount sufficient to show benefit to the individual. The actual amount administered, as well as the rate and time course of administration, will vary depending on the nature and severity of what is being treated. Prescription of treatment, such as determining dosage, is within the responsibility of general practitioners and other physicians, and will usually take into account the disorder being treated, the individual patient's condition, the site of delivery, the method of administration, and other factors known to practitioners. The compositions of the present invention are particularly relevant for the treatment of existing tumors, particularly cancer, and for preventing the recurrence of such conditions after initial treatment or surgery. Examples of the techniques and protocols mentioned above can be found in Remington's Pharmaceutical Sciences, 16 th edition, Oslo, A. (ed), 1980 (Remington 1980).

[0066] The optimal dosage can be determined by the physician based on a number of parameters, including, for example, age, sex, weight, severity of the condition being treated, the active ingredient being administered, and the route of administration. Generally, a serum concentration of the polypeptide and antibody that allows for receptor saturation is desired. A concentration of greater than about 0.1 nM is usually sufficient. For example, 100 mg / m 2 This dose of antibody provides a serum concentration of about 20 nM for about 8 days.

[0067] As a rough guideline, antibody doses range from 10 to 300 mg / m 2 An equivalent dose of the antibody fragment may be administered weekly in an amount of Le yMore frequent intervals are used to maintain serum levels above concentrations that allow carbohydrate saturation. The dose of the composition will vary depending on the properties of the binding member, such as its binding activity and in vivo plasma half-life, the concentration of the polypeptide in the formulation, the route, site and rate of administration, the clinical tolerance of the patient involved, the pathology afflicting the patient, etc., and is within the skill of the physician. For example, a dose of 300 μg of antibody per patient per administration is preferred, but doses can range from about 10 μg to 6 mg per administration. Different doses are utilized during a series of successive inoculations. The practitioner can administer an initial inoculation and then boost with a relatively small amount of antibody.

[0068] The present invention is also directed to optimizing immunization schedules to enhance protective immune responses against cancer.

[0069] Binding members of the invention may be made in whole or in part by chemical synthesis. They may be readily prepared by well-established standard solution or preferably solid phase peptide synthesis techniques, general descriptions of which are widely available (see, for example, JM Stewart and JD Young, 1984 (Stewart and Young 1984), M. Bodanzsky and A. Bodanzsky, 1984 (Bodanzsky and Bodanzsky 1984)), or they may be prepared in solution, by solution phase methods or any combination of solid-phase, solution phase and solution chemical techniques, for example by first completing each peptide moiety and then, if desired and appropriate, after removing any protecting groups present, introducing residue X by reaction of the respective carbonate or sulfonic acid or their reactive derivatives.

[0070] Another convenient way to produce a binding member according to the invention is to express nucleic acid encoding it by using nucleic acid in an expression system.

[0071] The present invention further provides isolated nucleic acids encoding the specific binding members of the present invention. Nucleic acids include DNA and RNA. In a preferred embodiment, the present invention provides nucleic acids encoding the specific binding members of the present invention as defined above. Examples of such nucleic acids are shown in Figures 1 and 2. The skilled artisan can determine substitutions, deletions and / or additions to the nucleic acid that still provide the specific binding member of the present invention.

[0072] The invention also provides constructs in the form of plasmids, vectors, transcription or expression cassettes comprising at least one nucleic acid as described above. The invention also provides recombinant host cells comprising one or more constructs as described above. As mentioned, nucleic acids encoding specific binding members of the invention form an aspect of the invention, as do methods of producing specific binding members, comprising expression from the encoding nucleic acid. Expression may conveniently be achieved by culturing recombinant host cells comprising the nucleic acid under appropriate conditions. Following production by expression, the specific binding member may be isolated and / or purified using any suitable technique and then used as appropriate.

[0073] Systems for cloning and expressing polypeptides in a variety of different host cells are well known. Suitable host cells include bacteria, mammalian cells, yeast, and baculovirus systems. Mammalian cell lines available in the art for the expression of heterologous polypeptides include Chinese hamster ovary cells, HeLa cells, baby hamster kidney cells, NSO mouse melanoma cells, and many others. A common preferred bacterial host is E. coli. The expression of antibodies and antibody fragments in prokaryotic cells such as E. coli is well established in the art. For a review, see, for example, Pluckthun, 1991 (Pluckthun 1991). Expression of eukaryotic cells in culture is also available to those skilled in the art as an option for the production of specific binding members. For recent reviews, see, for example, Reff, 1993 (Reff 1993); Trill et al., 1995 (Trill, Shatzman et al. 1995).

[0074] Suitable vectors can be selected or constructed, including suitable control sequences, including promoter sequences, transcription termination sequences, polyadenylation sequences, enhancer sequences, marker genes and other sequences.Vector can be plasmid, virus, such as phage or phagemid, as appropriate.For further details, see, for example, Sambrook et al., 1989 (Sambrook 1989).Many known techniques and protocols for the manipulation of nucleic acid, for example in the preparation of nucleic acid constructs, mutagenesis, sequencing, introduction of DNA into cells, and gene expression and protein analysis are detailed in Ausubel et al., 1992 (Ausubel 1992).

[0075] Thus, a further aspect of the invention provides a host cell comprising a nucleic acid as disclosed herein. A further aspect provides a method comprising introducing the nucleic acid into a host cell. The introduction may use any available technique. For eukaryotic cells, suitable techniques may include calcium phosphate transfection, DEAE-dextran, electroporation, liposome-mediated transfection, and transduction using retroviruses or other viruses, such as vaccinia virus, or, for insect cells, baculovirus. For bacterial cells, suitable techniques may include calcium chloride transformation, electroporation, and transfection using bacteriophages. After the introduction, expression from the nucleic acid may be caused or allowed, for example, by culturing the host cell under conditions for gene expression.

[0076] The nucleic acids of the invention can be integrated into the genome (e.g., chromosome) of the host cell. Integration can be facilitated by the inclusion of sequences that facilitate recombination with the genome by standard techniques.

[0077] The present invention also provides a method comprising using a construct as described above in an expression system for expressing a specific binding member or polypeptide as described above.

[0078] Preferred features of each embodiment of the invention are as for each other embodiment mutatis mutandis. Prior art documents referred to herein are incorporated to the fullest extent permitted by law.

[0079] Unexpectedly, the inventors found that Le y We found that the FG27 family of mAbs that recognize glycans containing leukocytes induced non-apoptotic direct cell death without involving immune effector cells. y Drugs that bind to the glycans and induce non-apoptotic cell death are provided.

[0080] FG27 mAb induced membrane damage in cells, resulting in cell aggregation, loss of microvilli, uptake of small molecular weight dyes, and pore formation. Over time, cells developed large pores and lysed by a mechanism similar to necrosis, as with many mAbs that recognize other glycans (Hellstrom, Garrigues et al. 1990, Chou, Takematsu et al. 1998, Zhong, Manzi et al. 2001, Alvarez-Rueda, Leprieur et al. 2007, Loo, Pryer et al. 2007, Zhang, Zhang et al. 2010). FG27 mAb also exhibited potent cytotoxic activity in vitro via ADCC and CDC. Administration of FG27 (0.1 mg intravenously (iv) twice weekly for 9 weeks) to mice with established metastases in the liver and peritoneal cavity resulted in complete tumor eradication and cure of approximately 30% of the mice. The potential of FG27 mAb in eradicating well-established tumors without concurrent chemotherapy was demonstrated by Le y These results demonstrate their promise as monotherapy agents for the treatment of human solid tumors expressing IL-1, IL-2, and IL-3.

[0081] The invention will now be further described by the following non-limiting examples and the accompanying drawings. [Brief description of the drawings]

[0082] [Figure 1]Figure 1a: Complete amino acid and nucleotide sequences of mouse FG27.10 IgG3 heavy chain. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). Figure 1b: Complete amino acid and nucleotide sequences of mouse FG27.18 IgG1 heavy chain. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). Figure 1c: Complete amino acid and nucleotide sequences of mouse FG27.10 and FG27.18 kappa chains. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). [Figure 1-2] Same as above. [Figure 1-3] Same as above. [Figure 2-1] Figure 2a: Complete amino acid and nucleotide sequence of the murine FG27 heavy chain variable region chimerized with a human IgG1 heavy chain constant region. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). Figure 2b: Complete amino acid and nucleotide sequence of the murine FG27 kappa chain variable region chimerized with a human kappa chain constant region. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). [Figure 2-2] Same as above. [Figure 3-1] Figure 3a: Complete amino acid and nucleotide sequence of the humanized FG27 heavy chain variable region. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). Figure 3b: Complete amino acid and nucleotide sequence of the humanized FG27 kappa chain variable region. The numbers represent the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). [Figure 3-2] Same as above. [Figure 4] Figure 4: Binding of FG27.10 and FG27.18 to ST16, HUVEC, and PBMC. Negative controls for this assay were NSO supernatant for ST16 and HUVEC, or cells alone for PBMC, and positive controls were anti-sialyl-diLea mAb, 505 / 4, or W6 / 32, which recognize HLA. [Diagram 5] Figure 5: Binding of FG27.10 and FG27.18 to glycolipid extracts from colorectal (C170), ovarian (OVCAR-3, OVCAR-4, OVCA433, OAW28, OAW42) and gastric cell lines (ST16, MKN45, AGS). Glycolipids extracted from a range of cancer cell lines were plated on ELISA plates. Cells and glycolipids were then incubated with supernatants of FG27.10 or FG27.18. Binding to glycolipids was probed with anti-IgG-HRP / TMB. [Figure 6] Figure 6: Binding of FG27.10 to Lewis antigens (a, b, x, and y) coupled to HSA assessed by ELISA. 505 / 4 (anti-sialyl-di-Lea), FG88.2 (Lea-x), FG88.7 (Lea-x), and 225-Le (Leb) were included as positive controls, and mouse HRPO was included as a negative control. Antibody activity was measured by absorbance at 405 nm. [Figure 7-1] Figure 7: Binding of FG27.10 and FG27.18 was screened against the glycan array of The Consortium for Functional Glycomics, which consists of 610 mammalian glycan targets. Fine specificity was compared between a) FG27.10, b) FG27.18, c) 692 (Ley / b), and d) BR96 (Ley / x): where a=Lea, b=Leb, y=Ley, x-Lex, D-=di, T-=tri, S-=sialyl, Ex-=extended, φ represents mannose containing glycan, and yx=Ley-Lex. Note that the glycan numbers may not be the same for each mAb, since the mAbs were screened on different versions of the glycan array. [Figure 7-2] Same as above. [Figure 8] Figure 8: Binding of FG27 to Ley bound to total proteins or lipids extracted from a panel of tumor cell lines. Western blot detection of FG27 binding to Ley associated with total glycoproteins or glycolipids extracted from lysates (equivalent of 1x105 cells) of tumor cell lines, AGS cells, HCT15 cells, OVCAR3 cells, MCF7 cells, and H322 cells. [Figure 9] Figure 9: Comparison of anti-Ley and Leb mAbs across a range of cell lines. SC101 (692 / 29) and BR96 were titrated from 0.1 to 30 μg / ml across a range of cell lines (ST16, HT29, C170, LoVo, Colo201, OVCAR-3, AGS) where FG27 supernatant was diluted 1:30 from neat. Binding was probed with anti-IgG-FITC and analyzed by flow cytometry. [Figure 10] Figure 10: Representative IHC images of normal esophagus, stomach, duodenum, ileum, colon, pancreas, lung, and kidney stained with 692 / 29 (Ley / b), BR96 (Ley / x), and FG27 (Ley). [Figure 11] Figure 11a: Antitumor activity of FG27 in vivo. The percentage of tumor growth over the study period is shown in the C170HM2 bioluminescent mouse tumor model used to evaluate the antitumor activity of murine FG27 mAb compared to vehicle-only control (PBS). In this model, bioluminescence represents viability. Group n>9; the study was terminated on day 32. Treatment with FG27 resulted in a significant reduction in bioluminescent tumor burden by the last day of the study compared to vehicle control (p=0.014). Figure 11b: Tumor burden at the injection site assessed by BLI shows a significant reduction in tumor growth at the injection site with FG27 treatment (p=0.05). [Figure 12]Figure 12: Antitumor activity of FG27 in vivo. The percentage of tumor growth is shown in the C170HM2 bioluminescence mouse tumor model used to evaluate the antitumor activity of FG27 compared to positive control mAb and vehicle only control (PBS). Bioluminescence represents the viability of tumor cells in this model. Group n>8; treatment with FG27 resulted in a significant reduction in the percentage of tumor growth by day 51 compared to the treated control (p=0.009). Figure 12b: Analysis by log-rank Mantel-Cox test represents significant survival of the FG27 treated group compared to the vehicle only control (p=0.035). Treatment was terminated at day 120. [Figure 13-1] Figure 13a: Amino acid and nucleotide sequences of mouse FG27 heavy chain variable region compared to BR96, H18A, Hu3S193, and SC101 using the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). Figure 13b: Amino acid and nucleotide sequences of mouse FG27 kappa chain variable region compared to BR96, SC101, H18A, Hu3S193 using the standardized IMGT system for numbering antibody sequences (Lefranc, Giudicelli et al. 2009). [Figure 13-2] Same as above. [Figure 13-3] Same as above. [Figure 13-4] Same as above. [Figure 14] Figure 14: Binding of chimeric IgG1 (5 μg / ml) and IgG2 (5 μg / ml) mAbs to the Ley-expressing tumor cell line AGS, assessed by indirect immunofluorescence and flow cytometry analysis, was compared to the murine versions of the antibodies FG27.10 (5 μg / ml) and FG27.18 (5 μg / ml). The anti-HLA mAb W6 / 32 (1 / 100) was included as a positive control. [Figure 15-1]Figure 15a: PI uptake in ST16 cells mediated by mouse IgG3 mAb FG27.10, but not the IgG1 variant FG27.18, induced robust PI uptake even at 4°C. Figure 15b: PI uptake mediated by FG27.10 and FG27.18 in a panel of tumor cell lines (AGS, Colo201, C170, C170HM2, MCF7, OVCA4, LoVo, MKN45, 791T, and Skov3). Figure 15c: PI uptake mediated by FG27.10, FG27.18, CH27 IgG2, and CH27 IgG1 in C170 cells. [Figure 15-2] Same as above. [Figure 16] Figure 16: Internalization data of FG27.10, FG27.18, and CH27 over a range of concentrations (0.01 nM-30 nM) in a panel of cancer cell lines, AGS, MCF7, H322, and C170. Cellular internalization of mAbs was monitored using saporin-conjugated goat antibody against mouse IgG (ATSbio, IT#48) or saporin-conjugated goat antibody against human IgG (ZAP) (F'ab fragment) 75 ng / well and assessed by inhibition of thymidine incorporation (added during the final 24 h incubation). Control wells consisted of cells incubated with culture medium alone, irrelevant positive control IgG3 mAb FG88, and secondary conjugate alone. [Figure 17-1]Figure 17a: Example of FG27.10 mIgG3 mediating ADCC / CDC. Target cells were incubated with PBMC isolated from human blood with FG27.10 supernatant or irrelevant IgG (ADCC) or complement (CDC). Error bars are present representing standard deviation of triplicate wells, but may be unclear depending on the data point. Figure 17b: ADCC of MCF7 tumor cell line upon incubation with FG27.10, FG27.18, 692 / 29, CH27 IgG1, and CH27 IgG2. These are results from one experiment, but represent two replicates. Figure 17c: Mean percentage killing by CDC of ST16 tumor cell line after incubation with FG27.10, FG27.18, 692 / 29, CH27 IgG1, and CH27 IgG2 with human serum after 4 hours. Herceptin was included as a control. The assay was performed with serial dilutions of 3, 1, 0.3, and 0.1 μg / ml, and each dilution was incubated in triplicate. Figure 17d: Mean percentage killing by CDC of AGS cells after incubation with FG27.10, 692 / 29, CH27 IgG1, and CH27 IgG2 with human serum after 4 hours. The assay was performed with serial dilutions of 3, 1, 0.3, and 0.1 μg / ml, and each dilution was incubated in triplicate. [Figure 17-2] Same as above. [Figure 17-3] Same as above. [Figure 18] Figure 18: Gibbs Analysis I of the FG27 Heavy Chain: Sequence Identity. Structurally important residues are highlighted: proline residues are outlined in white, cysteine ​​residues are outlined in black, and asparagine residues are outlined in grey. [Figure 19] Figure 19: Gibbs Analysis II of the FG27 heavy chain: Identity and similarity scores, 4 Å proximal residues and loop lengths of the CDRs. Cysteine ​​residues are outlined in white. [Figure 20] FIG. 20: Human heavy chain donor sequence germline analysis. [Figure 21]Figure 21: Structurally important residues of the FG27 heavy chain humanization strategy are highlighted: proline residues are outlined in white, cysteine ​​residues are outlined in black, and asparagine residues are outlined in grey. Grey outlined and italicized residues represent back-translation to the murine residue. [Figure 22] Figure 22:27 Gibbs Analysis of the Kappa Light Chain I: Sequence Identity. Structurally important residues are highlighted: proline residues are outlined in white, cysteine ​​residues are outlined in black, and asparagine residues are outlined in grey. [Figure 23] Figure 23: Gibbs Analysis II of FG27 kappa light chain: identity and similarity scores, 4 Å proximal residues and CDR loop lengths. [Figure 24] FIG. 24: Human kappa light chain donor sequence germline analysis. [Diagram 25] Figure 25: FG27 kappa light chain humanization strategy. Structurally important residues are highlighted: proline residues are outlined in white and cysteine ​​residues are outlined in black. Bold residues represent back-translation to the murine residue. [Figure 26] Figure 26: Comparison of antibody binding to LewisY-HSA antigen in antibody combinations designated as LewRHA with LewRKA, LewRKB, LewRKC, or LewRKD, and LewRHB with LewRKA, LewRKB, LewRKC, or LewRKD, including the cFG27 positive control antibody and the c1210 negative control antibody. [Figure 27] FIG. 27. Thermal stability of LewRHA with LewRKA, LewRKB, LewRKC, or LewRKD, and LewRHB with LewRKA, LewRKB, LewRKC, or LewRKD, with the cFG27 positive control antibody. [Figure 28] FIG. 28. Thermal shift analysis of Lewisy FG27 humanized variants hLewAB, hLewAC, hLewAD, hLewBC and control mouse FG27. [Figure 29]Figure 29. BiaCore analysis of antibody affinity of FG27 humanized variants hLewAB, hLewAC, hLewAD, hLewBC and control murine FG27. [Diagram 30] Figure 30. Non-specific protein-protein interactions in FG27 humanized variants hLewAB, hLewAC, hLewAD, hLewBC and control mouse FG27 as measured by CIC (Cross-Interaction Chromatography). [Diagram 31] Figure 31. Binding of FG27.10, FG27.18, ch27IgG1, and humanized variant hLewAC to a panel of cell lines. [Diagram 32] Figure 32: Significantly increased growth inhibition by i27G1 compared to 27hIgG1 in MCF7 (panel i) and AGS (panel ii). Significantly increased functional affinity (SPR) by i27G1 compared to 27hIgG1 (panel iii). i27G1 maintains comparable ADCC (panel iv) and CDC (panel v) activity in MCF7. Significance compared to parental constructs was estimated from 2-way ANOVA (direct cytotoxicity) or 1-way ANOVA (functional affinity) with Dunnett's correction for multiple comparisons. [Diagram 33] Figure 33: Amino acid and nucleic acid sequences of each light and heavy chain of the "i27G1" antibody. EXAMPLES

[0083] method Binding to tumor cell lines: 1×10 5Cancer cells were incubated with 50 μl of primary antibody for 1 h at 4°C. Cells were washed with 200 μl of RPMI 10% NBCS (new born calf serum) (NBCS: Sigma, Poole, UK) and spun at 1,000 rpm for 5 min. The supernatant was discarded and 50 μl of FITC-conjugated anti-mouse IgG Fc-specific mAb (Sigma; 1 / 100 in RPMI 10% NBCS) was used as the secondary antibody. Cells were incubated in the dark at 4°C for 1 h, then washed with 200 μl of RPMI 10% NBCS and spun at 1,000 rpm for 5 min. After discarding the supernatant, cells were fixed using 0.4% formaldehyde. Samples were analyzed on a Beckman coulter FC-500 flow cytometer (Beckman Coulter, High Wycombe, UK). WinMDI 2.9 software was used to analyze and plot the raw data.

[0084] Binding to blood: 50 μl of healthy donor blood was incubated with 50 μl of primary antibody at 4° C. for 1 hour. The blood was washed with 150 μl of RPMI 10% NBCS and spun at 1,000 rpm for 5 minutes. The supernatant was discarded and 50 μl of FITC-conjugated anti-mouse IgG Fc-specific mAb (1 / 100 in RPMI 10% NBCS) was used as the secondary antibody. The cells were incubated in the dark at 4° C. for 1 hour, then washed with 150 μl of RPMI 10% NBCS and spun at 1,000 rpm for 5 minutes. After discarding the supernatant, red blood cells were lysed using 50 μl / well Cal-Lyse (Invitrogen, Paisley, UK) followed by 500 μl / well distilled water. The blood was then spun at 1,000 rpm for 5 minutes. The supernatant was discarded and the cells were fixed using 0.4% formaldehyde. Samples were analyzed on a FC-500 flow cytometer (Beckman Coulter). WinMDI 2.9 software was used to analyze and plot the raw data.

[0085] Extraction of plasma membrane glycolipids: Cell pellets (5 × 10 7 100 cells) were resuspended in 500 μl of mannitol / HEPES buffer (50 mM mannitol, 5 mM HEPES, pH 7.2, both Sigma) and passed through three needles (23G, 25G, 27G) with 30 pulses each. 25 μl of 1 M CaCl was added to the cells and passed through three needles with 30 pulses each as above. The sheared cells were incubated on ice for 20 min and then spun at 3,000 g for 15 min at room temperature (RT). The supernatant was collected and spun at 48,000 g for 30 min at 4 °C and the supernatant was discarded. The pellet was resuspended in 1 ml of methanol, then resuspended in 1 ml of chloroform and incubated at RT for 30 min with rotation. The sample was then spun at 1,200 g for 10 min to remove precipitated proteins. The supernatant containing the cell membrane glycolipids was collected and stored at -20 °C.

[0086] Glycomic analysis: To clarify the fine specificity of FG27 mAb, the antibodies were FITC-labeled and sent to the Consortium for Functional Glycomics (http: / / www.functionalglycomics.org / static / consortium / resources / resourcecoreh8.shtml), where they were screened against over 600 natural and synthetic glycans. Briefly, synthetic and mammalian glycans with amino linkers were printed on glass microscope slides that were activated with N-hydroxysuccinimide (NHS) to form amide bonds. This work was performed by Core H of CFG at Emory University. FG27 was tested at room temperature at 1 μg / ml in PBS. Briefly, antibody samples were applied to the printed surface of the microarray and incubated for 1 h in a humidified chamber. The slides were then rinsed 4 times with PBS, after which fluorescently labeled (Alexa Fluor 488) anti-rabbit IgG was added and incubated for 1 h. The slides were then rinsed four times with PBS and fluorescence was measured on a Perkin-Elmer microarray XL4000 scanner and analyzed using Imagene software (BioDiscovery).

[0087] Further details of the protocol used are provided below.

[0088] Glycan-binding assay using unlabeled monoclonal antibodies 1. Introduction: The primary objective of 1.1.Core H is to determine the glycan binding proteins (GBPs) and binding specificities of various organisms to which glycans submitted by researchers are bound using printed glycan microarrays.

[0089] 2. Reference address: www.functionalglycomics.org

[0090] 3. Materials needed: 3.1. Slide where glycans are printed on the slide surface with white etched barcodes and black marks (Core D) - Do not touch this area. 3.2.Cover glass (Fisher Scientific, 12-545F) 3.3. A moistened slide processing chamber (Fisher scientific, NC9091416), or a homemade system using a petri dish; with a moist paper towel at the bottom of the chamber. 3.4. 100 ml Coplin jar for washing slides 3.5.Tris-HCl (Fisher scientific, BP152-1) 3.6.NaCl (Fisher scientific, S271-3) 3.7.CaCl2(Fisher scientific, C79-500) 3.8.MgCl2(Fisher scientific, BP214-500) 3.9. Potassium dihydrogen phosphate (Fisher scientific, P285-3) 3.10.dH20 3.11. Cyanine 5-streptavidin (ZYMED 43-4316) 3.12. Appropriate secondary antibodies, fluorescently labeled where possible. 3.13.BSA(Fisher scientific, Bp1600-100) 3.14.Tween-20 (EMD Biosciences, 655205) 3.15. Sodium azide (Fisher Scientific, S227-500) 3.16.ProScanArray Scanner(Perkin Elmer)

[0091] 4. Buffer: 4.1.TSM = 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM CaCl2, 2 mM MgCl2 4.2.TSM Wash Buffer (TSMW) = TSM Buffer + 0.05% Tween-20 4.3.TSM Binding Buffer (TSMBB) = TSM buffer + 0.05% Tween 20 + 1% BSA

[0092] 5. Protocol: 5.1. Make working stocks of wash buffers (TSM, TSM Wash Buffer, and HO) or collect reagents and bring to room temperature if they have been in the refrigerator. 5.1.1.Buffer (A) TSM - 20 mM Tris-HCl, pH 7.4 150 mM NaCl, 2 mM CaCl2, 2 mM MgCl2 5.1.2.Buffer(B)TSM Wash Buffer(TSMW)-TSM Buffer+0.05% Tween-20 5.1.3.Buffer (C) TSM Binding Buffer (TSMBB) - TSM buffer + 0.05% Tween 20 + 1% BSA 5.1.4.dH2O 5.2. Prepare 100 μl sample by diluting antibody in TSMBB or appropriate binding buffer to a final concentration of 5-50 μg / ml or the appropriate concentration required for the assay based on antibody properties. 5.3. Remove the slide from the desiccator and label the sample with the name on the slide outside the black mark, near the barcode. 5.4. Hydrate the slides by placing them in a glass Coplin staining jar containing 100 ml TSMW for 5 minutes. 5.5. Remove excess liquid from the slide by standing the slide upright and allowing the liquid to drain. 5.6. Carefully apply 70 μl of sample (see 5.2) near the left-most slide between the black marks. 5.7. Gently place the cover slip on the slide, making sure not to form bubbles in the sample under the cover slip. If necessary, remove bubbles by gently tapping the cover slip with the pipette tip or by slowly lifting one side of the cover slip. Make sure the cover slip is between the black marks.

[0093] SDS-PAGE and Western blot analysis: Briefly, 5 or 10 6 Cell equivalents (cell lysates, membrane lipid extracts) were analyzed for FG27 binding. Tumor cell membrane lipid extracts and cell lysates were reduced with dithiothreitol (DTT; Pierce Biotechnology, ThermoFisher, Loughborough, UK), subjected to SDS-PAGE using NOVEX 4%-12% Bis-Tris gels (Invitrogen), and transferred to Hybond-P PVDF membranes (GE Healthcare, Amersham, UK) using 1× transfer buffer (20x, Invitrogen) and 20 (v / v)% methanol for 1 h at 30 V. Membranes were blocked with 5 (w / v)% non-fat dry milk in 0.05 (v / v)% Tween-PBS for 1 h and then probed with primary antibodies diluted in Tween-PBS, 2% BSA for 1 h. Binding of the primary antibody was detected using a biotin-conjugated anti-mouse IgG Fc-specific secondary antibody (Sigma; 1 / 2000 dilution in Tween-PBS, 2% BSA) for 1 h and visualized using IRDye 800CW streptavidin (LICOR Biosciences, UK; 1 / 1000 in Tween-PBS 2% BSA).

[0094] Immunohistochemical evaluation of FG27: To determine the therapeutic value of FG27, it was screened in microarrays of gastric, colorectal, ovarian, and breast tumor tissues by immunohistochemistry (IHC).

[0095] Methods: Immunohistochemistry was performed using a standard avidin-biotin peroxidise method. Paraffin-embedded tissue sections were placed on a 60°C hot block to melt the paraffin. Tissue sections were deparaffinised with xylene and rehydrated through graded alcohols. The sections were then immersed in 500 ml of citrate buffer (pH 6) and heated in a microwave (Whirlpool) for 20 min to retrieve antigens. Endogenous peroxidase activity was blocked by incubating tissue sections with endogenous peroxidase solution (Dako Ltd, Ely, UK) for 5 min. Normal swine serum (NSS; Vector Labs, CA, USA; 1 / 50 PBS) was added to each section for 20 min to block non-specific binding of the primary antibody. All sections were incubated with Avidin D / Biotin blocking kit (Vector Labs) for 15 min to block non-specific binding of avidin and biotin, respectively. Sections were blocked again with NSS (1 / 50 PBS) for 5 min. Tissue sections were then incubated with primary antibodies for 1 h at RT, and anti-β-2-microglobulin (Dako Ltd; 1 / 100 in PBS) mAb and PBS alone were used as positive and negative controls, respectively. Tissue sections were washed with PBS and incubated with biotinylated goat anti-mouse / rabbit immunoglobulin (Vector Labs; 1 / 50 in NSS) for 30 min at RT. Tissue sections were washed with PBS and incubated with preformed streptavidin-biotin / horseradish peroxidase complex (Dako Ltd) at 1 / 50 (PBS) for 30 min at RT. 3,3'-Diaminobenzidine tetrahydrochloride (DAB) was used as a substrate. Each section was incubated twice with 100 μl of DAB solution for 5 min.Finally, sections were lightly counterstained with hematoxylin (Sigma-Aldrich, Poole Dorset, UK) before being dehydrated in graded alcohols, cleared by immersion in xylene, and mounted on slides using DPX (Distyrene, plasticiser, xylene) mounting medium (Sigma).

[0096] Identification of the heavy and light chain variable regions of FG27.10 and FG27.18 Cell source and preparation of total RNA: Approximately 5 × 10 cells from hybridomas FG27.10 and FG27.18. 6 Cells were harvested from tissue culture, washed once with PBS, and cell pellets were treated with 500 μl of Trizol (Invitrogen). After dispersing the cells in the reagent, they were stored at −80°C until RNA was prepared according to the manufacturer's protocol. RNA concentration and purity were determined by Nanodrop. Prior to cDNA synthesis, RNA was treated with DNase I to remove genomic DNA contamination according to the manufacturer's recommendations (DNase I recombinant, RNase-free, Roche Diagnostics, Burgess Hill, UK).

[0097] cDNA synthesis: First strand cDNA was prepared from 3 μg of total RNA using a first strand cDNA synthesis kit and AMV reverse transcriptase according to the manufacturer's protocol (Roche Diagnostics). After cDNA synthesis, reverse transcriptase activity was destroyed by incubation at 90° C. for 10 min, and cDNA was stored at −20° C.

[0098] GAPDH PCR to assess cDNA quality: PCR was used to assess the quality of the cDNA; primers specific for the mouse GAPDH housekeeping gene (5'-TTAGCACCCCTGGCCAAGG-3' and 5'-CTTACTCCCTTGGAGGCCATG-3') were used with hot start Taq polymerase (NEB, Hitchen, UK) for 35 cycles (95°C for 3 min, followed by 35 cycles of 94°C / 30 sec, 55°C / 30 sec, 72°C / 1 min: a final polishing step of 72°C for 10 min). Amplified products were assessed by agarose gel electrophoresis.

[0099] PCR primer design for cloning of FG27 variable regions: Primers were designed to amplify the heavy and light chain variable regions based on sequence data of the PCR products. Primers were designed to allow cloning of the relevant chains into unique restriction enzyme sites in the hIgG1 / κ dual expression vector pDCOrig-hIgG1. Each 5' primer was targeted to a defined variable region start codon and leader peptide, with Kozak consensus immediately 5' of the start codon. Each 3' primer was designed to be complementary to the binding region of the antibody sequence, to maintain the reading frame after cloning of the chains, and to preserve the amino acid sequence usually found at the binding region / constant region junction. All primers were purchased from Eurofins MWG.

[0100] Heavy chain variable region PCR: Immunoglobulin heavy chain variable region usage was determined using PCR with a previously published set of primers (Jones and Bendig 1991). Previous results using a mouse mAb isotyping test kit (Serotec, Oxford, UK) indicate that FG27.10 is a mouse IgG3 antibody and FG27.18 is a mouse IgG1 antibody. Therefore, the appropriate constant region reverse primer was used to amplify from the constant region. PCR amplification was performed with 5' and 3' primers specific for the 12 mouse VH regions, specific for the antibody subclasses previously determined, using hot start Taq polymerase for 35 cycles (94°C for 5 min, then 35 cycles of 94°C / 1 min, 60°C / 1 min, 72°C / 2 min; final polishing step at 72°C for 20 min). Amplified products were assessed by agarose gel electrophoresis. Positive amplification was obtained with the VH4 primer.

[0101] PCR of light (κ) chain variable regions: Immunoglobulin light chain variable region usage was determined using PCR with a previously published set of primers (Jones and Bendig 1991). Previous results using a mouse mAb isotyping test kit indicate that both FG27.10 and FG27.18 used κ light chains. PCR amplification was performed with primers specific for mouse Vκ region 5' and 3' to mouse Cκ using hot start Taq polymerase for 35 cycles (94°C for 5 min, then 35 cycles of 94°C / 1 min, 60°C / 1 min, 72°C / 2 min; final polishing step at 72°C for 20 min). Amplification products were assessed by agarose gel electrophoresis. Positive amplification was provided by Vκ1 and Vκ2 primers for both FG27.10 and FG27.18.

[0102] Purification and sequencing of PCR products: PCR products were purified using a Qiaquick PCR purification kit (Qiagen, Crawley, UK). The concentration of the resulting DNA was determined by Nanodrop and purity was assessed by agarose gel electrophoresis. PCR products were sequenced using the original 5' and 3' PCR primers at the University of Nottingham DNA sequencing facility (http: / / www.nottingham.ac.uk / life-sciences / facilities / dna-sequencing / index.aspx). Sequences were analyzed (V-region identification, junction analysis) using the IMGT database search facility (http: / / www.imgt.org / IMGT_vquest / vquest?livret=0&Option=mouseIg). Sequencing showed that FG27.10 and FG27.18 share identical heavy and light chain variable regions. Sufficient constant residues were present in the heavy chain sequences to confirm that FG27.10 is of mIgG3 subclass and FG27.18 is of mIgG1 subclass, both resulting from a single splenocyte NSO fusion event.

[0103] Cloning strategy: Direct cloning of PCR products into pDCOrig-hIgG1 using restriction sites built into the PCR primers was known to be relatively inefficient from previous Scancell experience. Therefore, a dual cloning strategy was adopted: PCR products generated using a proofreading polymerase were cloned simultaneously into both pDCOrig-hIgG1 and a TA vector (pCR2.1; Invitrogen), where the product cloned into the TA vector served as a backup source of easily amplified material in case the initial pDCOrig-hIgG1 cloning failed.

[0104] FG27.18 heavy / light chain PCR for cloning: PCR amplification was performed using the above mentioned proofreading polymerase (Phusion; NEB) and cloning primers using the FG27.18 cDNA template for 35 cycles (98°C for 3 min, then 35 cycles of 98°C / 30 sec, 58°C / 30 sec, 72°C / 45 sec; final polishing step at 72°C for 3 min). Successful amplification was confirmed by agarose gel electrophoresis.

[0105] Method 1 - Direct light chain cloning: The amplified light chain of FG27.18 was sequentially digested with the restriction enzymes BsiWI and BamHI according to the manufacturer's instructions (NEB). The vector (pDCOrig-hIgG1 containing the V-regions from the previously cloned antibody) was digested at the same time. The vector DNA was purified by agarose gel using the QIAquick gel extraction kit (Qiagen) and the insert DNA was purified using a PCR purification kit. After DNA quantification by Nanodrop, the vector DNA was treated with phosphatase according to the manufacturer's recommendations (Antarctic Phosphatase, NEB) and the light chain insert was ligated into the vector (T4 DNA ligase, NEB). The ligated DNA was transformed into chemically competent TOP10F' cells (Invitrogen) and plated onto LB agar plates supplemented with 35 μg / ml Zeocin (Invitrogen, Toulouse, France), then incubated overnight at 37°C.

[0106] Method 2 - TOPO light chain cloning: The amplified light chain of FG27.18 was treated with Taq polymerase (NEB) for 15 min at 72°C to add "A" overhangs compatible with TA cloning. The treated PCR product was incubated with TOPO TA vector pCR2.1 (Invitrogen) and transformed into chemically competent TOP10F' cells according to the manufacturer's instructions. The transformed bacteria were plated on LB agar plates supplemented with ampicillin (80 μg / ml) and then incubated overnight at 37°C. Colonies were grown in liquid culture (LB supplemented with 80 μg / ml ampicillin) and plasmid DNA was prepared (spin miniprep kit, Qiagen). The presence of the insert was confirmed by sequential digestion with BsiWI and BamHI and agarose gel electrophoresis. Sequencing was performed on miniprep DNA from colonies using T7 and M13rev primers. The DNA insert from one such colony had the predicted FG27.18 light chain sequence; a 300 ml LB / ampicillin culture of the bacteria was grown overnight and plasmid DNA was prepared by maxiprep (plasmid maxi kit, Qiagen). The insert of the maxiprep DNA was confirmed by sequencing.

[0107] Cloning of TOPO heavy chain: The amplified FG27.18 heavy chain was treated with Taq polymerase (NEB) for 15 min at 72°C to add "A" overhangs. The treated PCR product was incubated with TOPO TA vector pCR2.1 and transformed into chemically competent TOP10F' cells as described above. The transformed bacteria were plated on LB agar plates supplemented with ampicillin and then incubated overnight at 37°C. Colonies were grown in liquid culture (LB / ampicillin) and plasmid DNA was prepared (spin miniprep kit). The presence of the insert was confirmed by digestion with HindIII and AfeI and agarose gel electrophoresis. Sequencing was performed on miniprep DNA from many colonies using T7 and M13rev primers. The DNA insert from one such colony had the expected FG27.18 heavy strand sequence; a 300 ml LB / ampicillin culture of the bacteria was grown overnight and plasmid DNA was prepared by maxiprep (plasmid maxi kit, Qiagen). The maxiprep DNA insert was confirmed by sequencing.

[0108] Cloning of the light chain of the pDCOrig-hIgG1 dual expression vector: The FG27.18 light chain was digested from the TOPO vector pCR2.1 by sequential digestion with BsiWI and BamHI, and the 400 bp insert DNA agarose gel purified using the QIAquick Gel Extraction Kit (Qiagen) according to the manufacturer's recommendations. This insert was ligated into the previously prepared pDCOrig-hIgG1 vector (see above) and transformed into chemically competent TOP10F' cells. The transformation was plated on LB agar plates supplemented with 35 μg / ml Zeocin, then incubated overnight at 37°C.

[0109] Colonies were grown in liquid culture (LB supplemented with 35 μg / ml Zeocin) and plasmid DNA was prepared (spin miniprep kit, Qiagen). Sequencing was performed on miniprep DNA from all colonies using the P6 sequencing primer located in the human kappa constant region. DNA inserts from colonies had the predicted FG27.18 light chain sequence correctly inserted into pDCOrig-hIgG1; 300 ml bacterial LB / zeocin cultures were grown overnight and plasmid DNA was prepared by maxiprep (plasmid maxi kit, Qiagen).

[0110] Cloning of pDCOrig-hIgG1 dual expression vector heavy chain: The FG27.18 heavy chain insert was digested from the TOPO vector pCR2.1 by digestion with HindIII and AfeI. The vector (pDCOrig-hIgG1-27.18k) containing the FG27.18 kappa light chain (prepared above) was also digested with HindIII and AfeI. The vector DNA was then treated with phosphatase according to the manufacturer's recommendations (Antarctic Phosphatase, NEB). After agarose gel electrophoresis, the 6.5 kb pDCOrig-hIgG1 vector band and the 400 bp FG27.18H insert band were isolated using a QIAquick gel extraction kit (Qiagen) according to the manufacturer's recommendations. The insert was ligated into the pDCOrig-hIgG1 vector and transformed into chemically competent TOP10F' cells. Transformants were plated on LB agar plates supplemented with 35 μg / ml Zeocin and then incubated overnight at 37°C. Colonies were grown in liquid culture (LB supplemented with 35 μg / ml Zeocin) and plasmid DNA was prepared (spin miniprep kit, Qiagen). The presence of the insert was confirmed by digestion with HindIII and AfeI and agarose gel electrophoresis. Sequencing was performed on miniprep DNA from a number of colonies using the P3rev sequencing primer located in the human IgG1 constant region. The DNA insert from one of the colonies had the predicted FG27.18 heavy chain sequence correctly inserted into pDCOrig-hIgG1; 300 ml bacterial LB / zeocin cultures were grown overnight and plasmid DNA was prepared by maxiprep (plasmid maxi kit, Qiagen). Sequencing was used to confirm the location of both the heavy and light chains.

[0111] Expression, purification, and characterization of chimeric antibody constructs: Methods for expression and purification of chimeric antibodies according to the present invention can be achieved using methods well known in the art. Briefly, antibodies can be purified from supernatants harvested from transiently or subsequently stably transfected cells by standard protocols, such as Protein A or Protein G affinity chromatography according to Sambrook et al., 2001 (Sambrook and Russell 2001).

[0112] QuikChange Lightning Site-Directed Mutagenesis Kit (Stratagene) Prepare the reaction as described below: 5 μl 10x reaction buffer 0.12 μl (25 ng) of RHA or RKA template 1.3 μl (125 ng) of oligonucleotide mutagenesis primer For 1.3 μl (125 ng) of oligonucleotide mutagenesis primer Rev 1 μl of dNTP mix 1.5 μl QuikSolution reagent ddH2O (to a final volume of 50 μl) 1 μl QuikChange Lightning Enzyme

[0113] 1. Cycle each reaction using the cycling parameters outlined in the table below. [Table 1]

[0114] 2. Add 2 μl of Dpn I restriction enzyme. 3. Mix each reaction gently and thoroughly, microfuge briefly, and immediately incubate at 37°C for 5 minutes to digest parental dsDNA. 4. Transform 2 μl of DpnI-treated DNA from each reaction into separate 45 μl (+ 2 μl β-ME) aliquots of XL10-Gold ultracompetent cells (see Transformation of TOP10™ E. coli). 5. Screen colonies using Phusion method, miniprep, and sequence to confirm the exact mutation.

[0115] Q5® Site-Directed Mutagenesis Kit (NEB Protocol): 1. Assemble the following reagents in a thin-walled PCR tube: [Table 2]

[0116] cycle [Table 3]

[0117] Kinase, ligase, and DpnI (KLD) treatment 2. Add the following reagents and incubate for 5 min at RT: 1 μl PCR product 5 μl 2X KLD Reaction Buffer 1 μl 10X KLD Enzyme Mix 3 μl nuclease-free water 3. Transform 5 μl of the KDL mix from each reaction into separate 50 μl NEB 5-α Competent E. coli. Carefully tap the tube 4-5 times to mix. Do not vortex. 4. Keep the mixture on ice for 30 minutes. 5. Heat shock at 42°C for 30 seconds. Place on ice for 5 minutes. 6. Pipette 950 μl of RT SOC into the mixture. 7. Incubate with shaking (250 rpm) at 37°C for 60 minutes. 8. Mix the cells thoroughly by tapping and inverting the tube, then plate 50 μl onto a kanamycin plate and incubate at 37° C. overnight. 9. Screen colonies using the GoTaq Green method, miniprep, and sequence to confirm the exact mutation.

[0118] IgG quantification: Coat each well of a 96-well immunoplate with a 100 μl aliquot of 0.4 μg / ml goat anti-human IgG antibody, diluted in PBS and incubated overnight at 4° C. (The plate can be stored for one month at this stage). Similarly, coat another blank plate with BSA / PBS blocking solution. Remove excess coating solution and wash the plate three times with 200 μl / well of washing buffer. In the blank plate, distribute 120 μl of SEC buffer to all wells except wells in column 2 rows B-G. Prepare a solution of human IgG1 / κ antibody at 1 μg / ml in SEC buffer to serve as a standard. Pipette 240 μl / well into wells in column 2 rows B and C. Centrifuge the medium from the transfected cells (250 g, 5 min) and save the supernatant. Pipette 240 μl of supernatant from the "no DNA" control (where cos cells were transfected in the absence of DNA) into wells in column 2 row D. Pipette 240 μl / well of experimental supernatant into wells in column 2 rows E, F, and G. Mix 240 μl aliquots of wells in column 2, rows B-G and transfer 120 μl to adjacent wells in column 3. Continue to column 11 with a series of 2-fold dilutions of standards, controls, and experimental samples. Transfer 100 μl from each well of the anti-IgG-coated plate to the corresponding well. Incubate for 1 h at 37 °C. Wash all wells 3 times with wash buffer (200 μl). Dilute goat anti-human kappa light chain peroxidase conjugate 5000-fold in SEC buffer and add 100 μl to each well. Repeat incubation and wash steps (step 9). Add 150 μl of K-BLUE substrate to each well and incubate for 10 min at RT in the dark. Stop the reaction by adding 50 μl of RED STOP solution to each well. Read the optical density at 655 nm.

[0119] Lewis y Binding ELISA Coat each well of a 94-well MaxiSorp plate (Nunc) with 100 ng / well Lewis Y HSA peptide in PBS and incubate overnight at 4° C. Wash 3 times with PBS-T (0.1% Tween 20). Block untreated plates with 250 μl per well of PBS / 0.2% BSA / 0.05% Tween 20 and incubate for 1 h at RT. Wash 3 times with PBS-T. Add 240 μl of antibody (diluted in PBS / 0.2% BSA / 0.05% Tween 20 if necessary) to wells in column 1; add 120 μl of buffer (PBS / 0.2% BSA / 0.05% Tween 20) to other wells. Transfer 120 μl from column 1 to adjacent wells in column 2. Continue to column 12 with a series of two-fold dilutions of the experimental samples. Transfer 100 μl per well from the dilution plate to the experimental plate. Incubate 1 h at RT. Wash wells 3 times with PBS-T. Dilute goat anti-human Fc peroxidase conjugate 10000-fold in PBS / 0.2% BSA / 0.05% Tween 20 (or dilute anti-mouse 10000-fold) and add 100 μl to each well. Incubate 1 h at RT and repeat washing step. Add 150 μl substrate (K-Blue) per well and incubate 150 min at RT. Stop the reaction by adding 50 μl RED STOP solution to each well. Read optical density at 650 nm.

[0120] thermal stability Dilute fully humanized antibodies and chimeric controls to 1 μg / ml in TBS / 0.2% Tween and aliquot into PCR tubes in volumes appropriate for EC80 concentration. Bring volume up to 100 μl with same buffer. Heat each tube separately between 30° C. and 85° C. for 10 min with 5° C. intervals and cool to 4° C. Freeze 1 μg / ml stocks for 1 h and then dilute to EC80 concentration. Perform binding assays against Lewis Y HSA peptide using 100 μl of each antibody per well in a 96-well plate (assay in duplicate at each temperature).

[0121] Thermal Shift Comparison Prepare samples directly in white PCR plates of 96-well plates in a final volume of 25 μl (final concentrations of purified antibodies of 1 and 2 μM). Make a 1:100 stock in Sypro Orange-PBS buffer, then add 1:10 to the final sample (e.g. 2.5 μl to 25 μl). Load into qPCR instrument using MxPro software, SYBR Green method, (filter=FRROX, no reference dye). Set thermal profile -71 cycles with 1°C increment. Plot results and determine Tm.

[0122] Purification of humanized mab Equipment: GE Healthcare AKTAxpress(TM) Purification System Software:UNICORN Column: HiTrap MabSelect SuRe, 1ml; HiLoad 16 / 600 Superdex 200pg Mobile phase: IgG elution buffer, Dulbecco's 1x PBS Sample prep: filtered through 0.22 μm Injection volume: 200 ml of Expi293 conditioned medium in DPBS (1:1) Flow rate: sample loading at 0.5 ml / min; gel filtration at 1.5 ml / min; Elute at 1ml / min

[0123] Comparison of aggregation Equipment: Agilent 1260 infinity HPLC system with thermostatted column compartment, Wyatt Technology Dawn Heleos, Wyatt Technology Optilab TRex Analysis software: Wyatt Technology Astra version 6.1.1.17 Column: Acquity UPLC BEH200 SEC, 4.6 x 150mm, 1.7μm Mobile phase: Dulbecco's 1x PBS containing 0.05% sodium azide Sample prep: filtered through PES 0.22 μm Sample concentration: Various Injection volume: 20μl Analysis temperature: 30℃ Flow rate: 0.4ml / min

[0124] CIC(Cross-interaction chromatography) Cross-interaction chromatography (CIC) analysis was performed to assess propensity for non-specific protein-protein interactions and to provide identification of any solubility issues that may result in downstream manufacturing problems.

[0125] Samples were analyzed by two separate 20 μl injections (0.5 mg / ml); first on 1 ml of NHS-activated resin (GE Healthcare) coupled with 30 mg of human polyclonal IgG (Sigma 14506), and second on a blank of 1 ml of NHS-activated resin coupled as a control column. The mobile phase consisted of Dulbecco's PBS (Sigma D8537) containing 0.01% sodium azide (0.1 ml / min) and all experiments were performed at 25° C. Eluted samples were detected by UV absorbance (Agilent 1260 Infinity HPLC system with thermostatted column compartment) and data were analyzed using Wyatt Technology ASTRA software (version 6.1.2.83) to determine the retention times of sample peaks. These were then used to calculate the retention factor k'.

number

[0126] Surface Plasmon Resonance (SPR): Lewis Y HSA conjugates were coupled to a CM5 chip and FG27.10 and FG27.18 mAbs (at concentrations ranging from 0.01 μM to 1 μM) were injected at 30 μl / min. Binding data were fitted to a heterogeneous ligand model using BIA evaluation software.

[0127] Antibody internalization: Cancer cells were cultured at 2×10 3 Cells were plated at a density of 1000 cells / well and allowed to adhere overnight. Cells were cultured in 96-well plates at 37°C for 72 h and treated with the indicated concentrations of mAb in combination with secondary antibodies: goat antibody (F'ab fragment) against mouse IgG conjugated with saporin (ATSbio, IT#48) or goat antibody against human IgG conjugated with saporin. Control wells consisted of cells incubated with culture medium alone and secondary conjugate alone. The amount of secondary antibody per well was 75 ng and the concentration of primary antibody varied from 0.01 nM to 30 nM. Saporin internalization was assessed via inhibition of thymidine incorporation (thymidine was added during the last 24 h of incubation).

[0128] ADCC and CDC: cells (5 × 10 3 ) were incubated with 100 μl of PBMC, 10% autologous serum or medium alone, or simultaneously with a range of mAbs. Maximal spontaneous release was assessed after incubating labeled cells with medium alone or 10% Triton X-100, respectively. After 4 h of incubation, 50 μl of supernatant from each well was transferred to a lumaplate in a 96-well plate. Plates were dried overnight and counted in a Topcount NXT counter (Perkin Elmer, Cambridge, UK). The mean percentage of lysis of target cells was calculated by the following formula:

number

[0129] Assay of PI uptake: FG27.10 and FG27.18 were incubated with AGS cells and tested for uptake of the small molecular weight dye propidium iodide (PI, Sigma) at various concentrations. The number of PI-positive cells represents the number of dying / dead cells. Tumor cells (5×10 4 ) were incubated with 50 μl of primary antibody for 2 h at RT on a 96-well round-bottom microtiter plate. 1 μg of PI was added and cells were incubated for 30 min at RT. Samples were analyzed on an FC-500 flow cytometer (Beckman Coulter). WinMDI 2.9 software was used to analyze and plot the raw data. For comparison, mAbs SC104 and 791T / 36, known to induce membrane damage, were also included as internal controls in the experiment.

[0130] Growth inhibition assay: Growth inhibition by the constructs was assessed by using the water-soluble tetrazolium salt WST-8 (CCK8 kit, Sigma-Aldrich) to measure the activity of cellular hydrogenase, which is directly proportional to the number of viable cells. Briefly, after plating cancer cells (2000 cells / 90 μl / well) overnight, constructs were added at different concentrations in a final volume of 10 μl / well and plates were incubated at 37°C (5% CO2) for 72-96 h. WST-8 reagent was then added (10 μl / well) and after an additional 3 h incubation, plates were read at 450 nm (Tecan Infinite F50) and the percentage of inhibition was calculated. EC 50 Values ​​were determined using nonlinear regression (curve fitting) with GraphPad Prism v 8.0 (GraphPad Inc, La Jolla, Calif.).

[0131] In vivo model: This study was performed under a UK Home Office Licence and followed the NCRI guidelines for the welfare and use of animals in cancer research, the LASA good practice guidelines, and the FELAS working group on pain and distress guidelines. Endotoxin-free (<10EU / ml) FG27 mAb was supplied in pre-formulated aliquots for administration and stored at -20°C until use. Age-matched male MF-1 nude mice were obtained from Harlan Laboratories (Bichester, UK) and each group consisted of n>8 animals, FG27, control mAb or vehicle control.

[0132] Mice were implanted with C170HM2 DLuX cells and monitored by optical imaging to determine tumor establishment and suitability for entry into the study. Mice were dosed with either FG27 or positive control mAb (505 / 4, FG88.2, or FG88.7 (1 mg / ml) as appropriate) at 0.1 mg, 100 μl intravenously (iv) twice weekly until termination, or with mAb vehicle control PBS at 100 μl, iv, twice weekly until termination. Bioluminescence imaging was performed on all mice weekly to obtain pre- and post-treatment tumor measurements. In this way, each mouse provided a pre-treatment control reading against which tumor growth could be compared.

[0133] All measurements and readings were transferred from the original dictation / notation to an Excel (tab-delimited) format for data processing in SPSS v16.0. Data integrity was checked using the explore and describe functions. Where erroneous points were identified they were cross-referenced to the original data and corrected accordingly. The data were screened for outliers and distribution profiles; data points outside the 95% confidence limits (outliers) were excluded from the analysis, but the data sheets were retained for reference.

[0134] Mice were imaged weekly (BLI) for bioluminescent tumor burden over the study period as follows: 60 mg / kg D-luciferin substrate was administered subcutaneously (sc), mice were anesthetized, and BLI readings were taken 15 min after substrate administration with open filter block (2D) and sequential emission filters (3D reconstruction in DLIT). Ventral and dorsal imaging was performed: the optimal location for imaging was the uppermost abdomen. BLI was measured over the entire abdomen, one region of interest (ROI) for each mouse that included all lesions present. Each mouse had pre-dose or baseline images taken to allow calculation of percentage tumor growth over time; these data were averaged per group. BLI readings were also taken post-termination to identify tumors in PM tissues.

[0135] Example 1 - Generation and initial characterization of FG27 mAb FG27 was generated by immunization of gastric tumor cells with glycolipids. Analysis of antibody response to immunization: Antibody titers were initially monitored by enzyme-linked immunosorbent assay (ELISA), followed by thin-layer chromatography (TLC) analysis using ST16 total lipid extracts and cell membrane lipid extracts, flow cytometry (FACS) using ST16 tumor cells, and Western blot using ST16 whole cell extracts, total lipid extracts, and cell membrane lipid extracts. Mice were boosted intravenously (iv) with ST16 cell membrane lipid extracts before fusion to those deemed to have the best response compared to pre-bleed serum controls.

[0136] Binding of FG27 hybridoma supernatants to a panel of tumor cell lines was analyzed by direct immunofluorescence and FACS analysis. Both FG27.10 and FG27.18 bound to ST16, compared to the positive control anti-HLA mAb W6 / 32 (eBioscience, CA, USA), and the negative control, but not to human umbilical vein endothelial cells (HUVEC) or peripheral blood mononuclear cells (PBMC) (Figure 4). Both FG27.10 and FG27.18 were cloned. FG27.10 was IgG3κ and FG27.18 was subclass IgG1κ. To ensure that both mAbs bound to glycolipids, glycolipids were extracted from a range of cell lines, dried onto ELISA plates, and then incubated with FG27.10 and FG27.18. Binding was seen with glycolipids derived from C170, ST16, and AGS with both mAbs, but not with cell lines that showed a lack of binding to whole cells (MKN-45, OAW28, OVCAR-3, OVCAR-4, and OAW42; Figure 5 ).

[0137] Example 2 - Definition of the epitope recognized by FG27 mAb To determine the degree of specificity of FG27.10, the antibody was first coupled to Leu2+ coupled to HSA. a , Le b , Le x , and Le y The binding of FG27.10 to the antigen was evaluated by ELISA. a , Le a-x , or Le b I didn't recognize Le y To further define the fine specificity of the FG27 mAb, a screen of these was performed by the Consortium for Functional Glycomics against over 600 natural and synthetic glycans. Binding of the FG27.10 and FG27.18 mAbs to the glycan array demonstrated that both mAbs strongly recognized Le y For comparison, the binding of 692 / 29 was shown to be specific to Leb , Le b It binds most avidly to glycans containing tri-Leu y and its variants. In contrast, BR96 binds to Le y and a range of Le y Variant (Le y-x , Le y-x-x ) and binds to Le x showed weaker binding to .

[0138] FG27.10 and FG27.18 were also screened on a panel of tumor cell lines by SDS / PAGE / Western blot for their properties of recognizing glycoproteins and glycolipids (Figure 8). In addition to binding glycolipids, both mAbs also recognized a range of glycoproteins, as evidenced by binding at the dye front.

[0139] FG27 is Le y Since the antibody is specific for Le, its binding to the cell line (Figure 9) was y and Le x mAb BR96 and Le y and Le b All three mAbs bound to C170 and AGS cells. In contrast, BR96 bound strongly to Colo201, 692 / 29 bound moderately, and FG27 did not bind (Figure 9). This suggests that this cell line is a good candidate for Leukocyte markers. y Although it does not express Le x and Le b These results indicate that FG27.10 expresses OVCAR3 and ST16 cells, which could not be concluded using non-specific mAbs. HT29 cells were unable to bind to any of the mAbs, whereas FG27.10 bound best to OVCAR3 and ST16 cells, and 692 / 29 bound to LoVo cells.

[0140] Example 3 - Immunohistochemical evaluation of FG27 To assess the binding of FG27 to normal human tissues, it was screened in a panel of tissue types. y / b (692 / 29) and Le y / x In contrast to the mAb that cross-reacted with (BR96), only FG27 bound to subpopulations of cells in normal stomach, lung, tonsil, pancreas, and duodenum (Table 1). y In contrast to the cross-reactive mAbs, FG27 failed to stain the colon, jejunum, breast, kidney, or ileum. Representative IHC images of normal staining of the esophagus, stomach, duodenum, ileum, colon, pancreas, lung, and kidney are shown in Figure 10.

[0141] Table 1: Binding of FG27 to normal human tissues assessed by immunohistochemistry (IHC). Staining of microarrays (TMA) of these tissues was analyzed via novel viewer software 2010, providing a semiquantitative score according to the intensity of staining of the tumor tissue. Strong staining was provided with a score of 3, moderate staining with a score of 2, weak staining with a score of 1, and negative score of 0. The results also show differential staining of specific cell types within these tissues. [Table 4] TIFF0007604452000007.tif26162

[0142] To determine the therapeutic value of FG27, it was screened on human gastric, colorectal, ovarian, and breast TMAs by IHC, which stained with different intensities 80% (69 / 86) of gastric tumor tissues, 100% (55 / 55) of colorectal tumor tissues, 95% (36 / 38) of ovarian tumor tissues, and 85% (39 / 46) of breast tumor tissues (Table 2).

[0143] Table 2. Binding of FG27 mAb to human gastric, colorectal, ovarian, and breast tumor tissues assessed by immunohistochemistry. Staining of these tissue microarrays was analyzed via the novel viewer software 2010, and a semiquantitative score was provided according to the intensity of staining of the tumor tissue. Strong staining was provided with a score of 3, moderate staining with a score of 2, weak staining with a score of 1, and negative score with a score of 0. [Table 5]

[0144] Scoring indicates that FG27 binds broadly to gastric tumors, where 59% of tumors were positively stained and 18% of tumors were strongly stained. 19% of tumors were moderately stained and 22% were weakly stained. The results indicate that FG27 has the potential to target a broad range of gastric tumors and strongly bind to these 18%. Kaplan-Meier analysis did not reveal an association between FG27 antigen expression and survival (data not shown). However, there was a correlation between FG27 antigen expression and M30 (apoptotic marker) expression (Pearson's chi-square, p=0.049). There was also a correlation between FG27 antigen expression and tumor site (Pearson's chi-square, p=0.008), where 65% of gastric specific tumors were stained with FG27 and 26% were strongly stained. In contrast, only 53% of tumors in the gastroesophageal junction stained, but of these, 37% stained strongly.46% of tumors in the lower third of the esophagus stained with FG27, and of these, 53% stained strongly.

[0145] Cross-reactivity of FG27 with ovarian cancer cell lines was screened against ovarian cancer TMAs with 360 tumors by immunohistochemistry. Scoring showed that FG27.18 bound poorly to ovarian tumors, with 85% of tumors staining negatively. Only 15% of tumors stained positively and only 3% stained strongly. 3% of tumors stained moderately and 9% stained weakly.

[0146] Kaplan-Meier survival analysis showed that patients whose tumors expressed the FG27 antigen had significantly better (log-rank, p=0.049) survival than those without expression. There was a correlation between FG27 antigen expression and FIGO stage (Pearson's chi-square, p=0.024). There was also a correlation between FG27 antigen expression and ovarian cancer type (Pearson's chi-square, p=0.024), with more than one-third of mucinous and endometrioid tumors expressing FG27, versus less than 14% of serous clear cell and undifferentiated tumors staining with this mAb. Cox multivariate analysis showed that FG27 provided no additional prognostic value across FIGO stage, residual disease, and response to chemotherapy.

[0147] Table 3: Comparison of tumor staining with Lewis y-specific mab FG27 and other Lewisy cross-reactive mabs BR96 (Lewisy / x) and 692 / 29 (lewisy / b) [Table 6] TIFF0007604452000010.tif63162

[0148] The monospecific Lewis y mab, FG27.10, stained 3-39% more tumors than the Lewis y / b 692 / 29 mab and 6-48% more tumors than the Lewis y / x Br96 mab, reflecting the inherent specificity of FG27.10. The greatest changes were seen in ovarian and breast tumors.

[0149] Unlike FG27, which showed only moderate binding to normal stomach, pancreas, and duodenum in one of two samples, and weak binding to lung and tonsils in one of two samples, BR96 showed strong binding to many tissues, including esophagus, ileum, jejunum, stomach, breast, lung, tonsils, pancreas, and duodenum. BR96 also showed moderate binding to placenta and gallbladder and weak binding to rectum, kidney, and colon. In contrast, 692 / 29 bound strongly to the jejunum, moderately to the ileum, stomach, and tonsils, and weakly to the gallbladder, thymus, lung, colon, pancreas, and duodenum (Table 1). This limited normal binding profile of FG27, especially with respect to gastrointestinal tissues, did not mitigate the binding of the antibody to tumor tissues, with 80% of gastric tumor tissues and 100% of colorectal tumor tissues showing antigen expression. In contrast, 692 / 29 and BR96 showed good distribution in binding of colorectal tumors in 82% and 90%, respectively (Noble, Spendlove et al. 2013), a similar figure to the latter, 15 / 18 samples, i.e. 83%, also reported in US Pat. No. 5,491,088. 39% and 49% of these tumors were moderately or strongly stained compared to 85% with FG27. 2-25 LE showed poor distribution, binding to only 52% of colorectal tumors, with 19% staining moderately and not strongly (Noble, Spendlove et al. 2013).

[0150] Thus, FG27 has a significantly better normal binding profile, especially in gastrointestinal tissues, and each Le y The differential binding between the mAbs is shown to represent distinct properties of the present invention.

[0151] Example 4 - Antitumor activity of FG27 in vivo in an established colorectal metastatic xenograft model Comparison of therapeutic efficacy of mAb FG27 in C170HM2 DLuX human liver metastasis model: The murine C170HM2 DLuX human liver metastasis tumor model was used to investigate the antitumor activity of murine FG27 mAb. The C170HM2 DLuX cell line is a bioluminescent variant of a liver metastatic colon tumor cell line that is implanted into the peritoneal cavity before starting mAb treatment and passaged to invade the liver parenchyma when left in situ for 10 days to metastasize to the liver. The proliferation and distribution / location of such labeled cells and tissues can be assessed non-invasively in real time in appropriate optical imaging systems and in resected tissues post-mortem (PM).

[0152] Antitumor Data: FG27 administered twice weekly (100 μg, iv) reduced peritoneal cavity and associated tumor growth compared to vehicle controls when compared by bioluminescence intensity (FIG. 11a). The study was terminated on day 32. Treatment with FG27 resulted in a significant reduction in bioluminescent tumor burden by the last day of the study compared to vehicle controls (p=0.014). There was a significant reduction in tumor growth at the injection site with FG27 treatment (p=0.05; FIG. 11b).

[0153] In a second in vivo study with extended treatment duration, FG27 demonstrated a significant reduction in the percentage of tumor growth by day 51 compared to the treated control group (p=0.019) (Figure 12a). Analysis by log-rank Mantel-Cox test showed significantly higher survival in the FG27-treated group (p=0.0251) compared to the vehicle-only control (Figure 12b).

[0154] Example 5 - Chimeric mAb The term "chimeric antibody" is intended to refer to an antibody whose variable region sequence is derived from one species and whose constant region sequence is derived from another species, such as an antibody whose variable region sequence is derived from a mouse antibody and whose constant region sequence is derived from a human antibody. The chimeric (or humanized) antibody of the present invention can be prepared based on the sequence of the mouse mAb prepared as described above. When the genetic and amino acid sequences of the antibodies are analyzed, it was shown that FG27.10 (Figure 1) and FG27.18 (Figure 2) have identical variable regions but different constant regions IgG3 and IgG1, respectively. This suggested that a single B cell had switched subclasses and then expanded during the immunization protocol, and both of these clones successfully fused.

[0155] Compared to the closest germline sequence, FG27 contains seven mutations (all non-silent) in its heavy chain and three mutations (all non-silent) in its light chain. Four amino acid differences relative to germline are in the heavy chain CDRs and three in FR3. In CDR1, there is an amino acid change from D to H at position 36. In CDR2, there are amino acid changes from G to D at positions 59 and 63, and an amino acid change from S to N at position 64 (Figure 2a). In the light chain, there are two amino acid differences with the CDR1 germline, S to I at position 28 and S to T at position 32, with one unique difference, K51N, in FR2.

[0156] There are several other mabs that bind Lewis y but cross-react with other related Lewis antigens. FG27.10 / 18 is the only antibody with monospecificity for Lewis y. This is reflected in the similar but distinct sequences of the FG27 variable heavy and light chain regions (Fig. 13a, b). FG27 shows 16-68% differences in amino acid sequence CDRs in the heavy chain, 4.3-16% differences in amino acid sequence in the light chain CDR sequences, 8.6-49.5% differences in amino acid sequence in the heavy chain variable region, and 2-24% differences in amino acid sequence in the light chain variable region from other cross-reactive lewis y mabs (Table 4). Most of the changes are in CDRH2, with 3-5 of 8 amino acids varying between the different mabs.

[0157] Table 4. Amino acid sequence differences in the variable heavy and light chain regions of FG27 derived from anti-Lewis y cross-reactive mab [Table 7]

[0158] The heavy and light chain variable regions of FG27.18 were cloned into human IgG1 and IgG2 expression vectors, which were transfected into CHO-S cells, and the human antibody was purified with protein G. The chimeric mAb was expressed in the gastric cell line AGS-containing Leukocyte. y The antibody bound to tumor cell lines expressing IgG1 (Figure 14).

[0159] Example 6 - Direct Cell Killing Many anti-glycan mAbs have been shown to induce direct cell death in antigen-positive cell lines without the need for effector cells or complement. This may increase their efficacy in vivo, possibly because tumors may develop mechanisms to evade immune-mediated cell death. This property is primarily associated with glycan-binding mAbs of the mIgG3 isotype, and is therefore lost after chimerization or humanization.

[0160] To determine whether FG27.10 and FG27.18 have direct cell death-inducing properties, ST16 cells were incubated with FG27.10, FG27.18, and anti-sialyl Le. a After 2 h incubation at RT with mAb SC104, cell death was measured by the uptake of propidium iodide, a DNA intercalating agent that is only taken up by dead cells (Figure 15a). Interestingly, despite having the same variable region, only FG27.10 was able to induce direct cell death, and PI uptake was not observed with FG27.18. Figure 15b shows the direct killing of a range of antigen-positive (AGS, Colo201, C170, C170HM, MCF-7, OVCAR4) cell lines, where the two mAbs show variable amounts of killing. None of the mAbs showed killing of antigen-negative cell lines (LoVo, MKN45, 791T, SKOV3).

[0161] Figure 15c shows PI uptake in the colorectal cell line C170 by a range of mAbs. FG27.10 and the positive control antibody SC101 showed strong titratable killing. FG27.18 showed weak, non-titratable killing. Chimeric human IgG1 showed moderate titratable killing, whereas the IgG2 variant did not.

[0162] Example 6 - Internalization FG27.10, FG27.18, and CH27 were all evaluated for internalization using various tumor cell lines (Figure 16). FG27.18 showed dose-dependent internalization in most cell lines tested, with reductions in cell viability ranging from 60% (C170) to 75% (MCF7) and 85% (AGS) at the highest concentrations tested. FG27.10 showed the best internalization in AGS (up to 50% reduction in viability) and MCF7 (up to 70% reduction in viability). Internalization in the H322 cell line was more variable with both mouse mAbs. CH27 mAb was efficiently internalized by MCF7, where a reduction in cell viability of up to 75% was observed, and by H322 where a reduction in cell viability of approximately 40% was observed.

[0163] Example 7 - Antitumor activity of FG27 in vitro ADCC and CDC: The properties of murine and chimeric FG27 mAbs to induce tumor cell death via ADCC were screened. Human PBMCs were used as a source of effector cells, and a panel of tumor cells served as target cells. First, ADCC analysis was performed on the supernatants of FG27.10 IgG3 (Figure 17a). The degree of ADCC observed corresponded to the expression of Lewis Y antigen, and was thus seen in ST16 and OVCAR3, where only low levels were observed in C170, and simply background in LoVo, HT29, or Colo201 (Figure 17a). The extent of ADCC MCF7 and OVCAR3 target cell killing by murine FG27.10 (mIgG3), FG27.18 (mIgG1), CH27 IgG1, and IgG2 was measured and compared after 18 hours of incubation at 37° C.; CH27 IgG1 provided the highest level of killing of the cell lines MCF7 and OVCAR3 (FIG. 17b).

[0164] CDC is known to be an important mechanism involved in the elimination of tumor cells in vivo. We assayed the characteristics of ST16 cell killing by CDC induced by mAbs FG27.10, FG27.18, CH27 IgG1, and CH27 IgG2 in the presence or absence of human serum as a source of complement for 18 hours at 37°C. As shown in Figure 17c, among the FG27 mAbs, FG27.10 (mIgG3) provided the highest level of killing at 3 μg / ml, whereas no significant level was seen with CH27 IgG2, which was expected given the nature of the IgG2 antibody. Similarly, when using AGS cells, FG27.10 provided the highest level of killing, followed by CH27 IgG1. Again, no killing was seen with CH27 IgG2 (Figure 17d).

[0165] In summary, FG27 mAb potently induced ADCC and significant CDC using human PBMCs as effector cells with human serum as a complement source.

[0166] Example 9 - Creation of a humanized version of the FG27 antibody Human VH and VK cDNA database A database of human immunoglobulin sequences in the Kabat alignment was compiled using human and mouse immunoglobulin protein sequences from the International Immunogenetics Database 2009 (Lefranc 307-10) and the Kabat Database Release 5 of Sequences of Proteins of Immunological interest (last update 17-Nov-1999) (Kabat et al. 1-3242). Our database contains 10,906 VH sequences and 2,912 VK sequences.

[0167] FG27 human framework selection Humanization requires the identification of suitable human V regions. The sequence analysis program Gibbs was used to search a database of human VH and VK with the FG27 VH and VK protein sequences using various selection criteria. The program Discovery Studio (Accelrys) was used to identify FW residues (Kabat definition) within 4 Å of the CDR residues of the structure of the mouse FG27 antibody and designated "4 Å Proximity Residues". The human VH sequences with the highest identity to FG27 VH at the 4 Å Proximity Residues are shown in FIG. 19, along with a list of these envelope residues and VCIs, and the number of residues in the FW, VCI, or 4 Å Proximity Residues that are identical to the mouse equivalent positions. FIG. 18 shows the alignment and residue identity of all VH sequences in FIG. 19. The number of obvious somatic mutations in each FW identified by comparison with the most identical human germline V gene is shown in FIG. 20.

[0168] Humanized, murine and incomplete sequences were excluded from this analysis. Sequence AJ579110 was selected as a human donor candidate due to the high sequence identity and similarity of this sequence, which does not have somatic mutations from its germline ABO19439 or HM855436.

[0169] Design of FG27 RHA and FG27 RHB With the appropriate human framework identified, synthetic protein and DNA sequences can be designed. The initial design of the humanized version of FG27 is the grafting of CDR1, 2, and 3 from FG27 VH to the FW of acceptor AJ579110. One VCI+4Å proximal residue at Kabat position 24 is not conserved in FG27 RHA and is backmutated to the mouse equivalent residue to produce the humanized version of FG27 RHB. Figure 21 compares the mouse and humanized versions of the FG27 VH protein sequence.

[0170] FG27 RKA Framework Selection To humanize the light chain, a human kappa chain was identified in a similar process to the heavy chain. Figures 22, 23 show the 4 Å proximal residues and the number of residues in the FW or 4 Å proximal residues that are identical to the mouse equivalent positions, alignment of the VK sequences, and comparison with human germline. Initial analysis found several possible donor candidates, but all of these proved to be human VK4, which showed poor expression. Expanding the analysis to include CDR1 with one less residue resulted in a single candidate, X72449, which showed a higher degree of sequence homology to the mouse antibody than the VK4 candidate. This candidate had a single somatic mutation from the human VK2 germline, but no somatic mutations from the human VK2D germline (Figure 24).

[0171] Design of FG27 RKA and FG27 RKB The X72449-derived framework was used to design the DNA and protein of the humanized construct. CDRs 1, 2, and 3 from FG27 VK have been shown to be grafted into the acceptor FW of X72449 to produce an initial version of humanized FG27. In FG27 RKA, there is a single mismatched 4 Å proximal residue 5, but in variant FG27 RKB, this residue was backmutated to the equivalent murine residue (Figure 25).

[0172] Generation and properties of a humanized version of FG27 Generation of FG27 humanized antibody The genes for FG27 RHA and FG27 RKC were synthesized by Genewiz and codon optimized for the human sequences (Figure 3ab). The native human framework sequences M65092 and X72449, respectively, for the heavy and light chains, and the native mouse CDR sequences were assembled in silico and named FG27 RHA~FG27 RHB and FG27 RKA~FG27 RKD. Using Genewiz's proprietary software algorithms, the sequences were optimized by silent mutagenesis to use codons preferentially utilized and synthesized by human cells. The RHA / B and RKA / B constructs were amplified by PCR using primers specific for the expression vector + insert (as described previously for the chimeric version) and inserted into pHuG1 and pHuK in a ligase-independent cloning reaction and used to transform TOP10 bacteria. RKA and RHA were then modified by PCR mutagenesis to obtain all human variants depicted in FIGS.

[0173] Clones were sequenced and expression plasmid DNA was prepared using QIAGEN Plasmid Miniprep Kit or Qiagen Plasmid Maxiprep kit. Expression construct sequences (RHA, RHB, RKA, and RKB) are shown in Figures 21 and 25. Expression plasmid preparations encoding (humanized or chimeric) VH and VK were used to transfect Expi293 cells and cultured in serum-free medium for 10 days, after which conditioned medium containing secreted antibody was harvested.

[0174] Antibody expression The concentration of IgG1κ antibodies in the conditioned medium of Expi293 cells was measured by ELISA and is shown in Table 5. Most antibodies were produced at sufficient levels, except for those highlighted in red.

[0175] Table 5. IgG levels in conditioned medium of transfected Expi293 cells [Table 8]

[0176] Binding of LewisY-HSA by FG27 antibody The binding activity to LewisY-HSA antigen was measured by binding ELISA. Initial experiments showed that there was essentially no difference between the binding potency of the murine or chimeric antibodies. This result gave us some confidence in the success of the humanization process.

[0177] The data shown in Figure 26 shows the binding potency of the first versions of humanized FG27. Although all versions bound to LewisY-HSA, versions of the antibody consisting of RHA / RKB, RHA / RKC, RHA / RKD, and RHB / RKC were selected as possible lead candidates.

[0178] Thermal stability of humanized candidate antibodies up to high temperatures The purpose of this experiment is to compare the thermal stability of humanized antibodies when subjected to high temperatures varying from 30-85°C for 10 minutes, cooled to 4°C, and used in an ELISA assay at the EC80 concentration of each candidate. All antibodies tested appeared to be equally stable (Figure 27), and all were inactive above 72°C.

[0179] Determination of humanized candidate antibody Tm To determine the melting temperature of the chimeric and humanized lead antibodies, all antibodies of RHA / RKB, RHA / RKC, RHA / RKD, and RHB / RKC were purified by a two-step affinity chromatography and gel filtration system and tested in a thermal shift assay. Samples were incubated with fluorescent dye (Sypro Orange) for 71 cycles with an increment of 1°C per cycle in a qPCR thermal cycler. The Tm of the chimeric and four humanized antibodies is shown in Figure 28. All of the candidate and chimeric antibodies have the same Tm (72°C).

[0180] Affinity of humanized candidate antibodies Determination of antibody affinity using SPR analysis was performed using a Biacore T3000. Lewis Y-HSA was covalently immobilized on a CM5 chip (Cat:BR-1005-30) according to the manufacturer's instructions. Chimeric FG27, RHA / RKB, RHA / RKC, RHA / RKD, and RHB / RKC antibodies were diluted in buffer at concentrations ranging from 200 nM to 0.274 nM (3-fold serial dilutions) and then injected at a flow rate of 50 μl / min. The association time was set to 120 s and the dissociation time was set to 600 s.

[0181] The binding / dissociation kinetics were analyzed by best fitting to a heterologous ligand model using the BIAcore 3000 evaluation software package. An overlay of the 200 nM binding curves is shown in the sensogram (Figure 29). RHA / RKB, RHA / RKC, and RHB / RKC show similar affinities to the chimeric FG27 (RHA / RKB>RHA / RKC>RHB / RKC>>RHA / RKD). These results suggest that humanization successfully retains binding activity within the desired parameters, with RHA / RKB or RHA / RKC having the highest affinity.

[0182] Aggregation of humanized candidate antibodies Samples were injected at 0.4 ml / min onto a size exclusion column in an HPLC system and analyzed by multi-angle light scattering to determine absolute molar mass and confirm aggregation. All variants show no signs of aggregation with average molar molecular weights in the range of 140.19 kDa, which is the expected range for IgG monomers in the context of this analysis. All samples are monodispersed (Mw / Mn<1.05). Mass recovery (calculated mass / injected mass) is 100% which represents good protein recovery and the samples do not appear to contain insoluble aggregates that stick to the column or are retained by the guard column. The overall data suggests that there is no concern of aggregation in any of the anti-FG27 antibody samples analyzed.

[0183] Nonspecific Protein-Protein Interactions (CICs) Cross-Interaction Chromatography (CIC) using bulk purified human polyclonal IgG is a technique for monitoring non-specific protein-protein interactions and can be used to distinguish between soluble and insoluble antibodies. A high retention index (k') indicates a tendency for self-interaction and low solubility. RHA / RKB, RHA / RKC, and RHB / RKC antibodies showed retention indices below 0.044, indicating a low tendency for non-specific interactions and good solubility (Figure 30).

[0184] The antibody has been engineered and expressed as a fully humanized antibody without loss of Lewis Y HSA binding potency (Figures 26 and 29). Experiments with a chimeric antibody consisting of a mouse variable region on a human constant region showed similar Lewis Y HSA binding ELISA potency as the mouse antibody. This result provides some confidence in the positive outcome of the humanization process.

[0185] Initial experiments show that fully humanized heavy and light chains, i.e., heavy and light chains that do not contain framework mutations to introduce the murine 4 Å proximal residues, do not bind the Lewis Y HSA peptide, and that a version of the chimeric positive control antibody but containing a single back mutation binds comparably to the chimeric positive control. Assays measuring antibody solubility, thermal stability, and aggregation suggest that all humanized candidates have acceptable biophysical characteristics.

[0186] In our view, the combination of excellent binding, expression, thermal stability, affinity, and cell binding makes the RHA heavy chain version combined with the RKC light chain version the best candidate for further development.

[0187] Example 10 - FG27 Binding Assay The humanized mAb (humanized 27) and the chimeric mAb (CH27) showed similar cell binding patterns at dose setting in cancer cell lines (Figure 31). The equilibrium affinity constants are generally in the same order of magnitude, but slightly higher compared to the parental mouse antibody, and the maximum binding capacities of humanized 27 and CH27 are greater compared to the values ​​of the mouse mAb. This result suggests a successful outcome of the humanization process.

[0188] Example 11 - Direct killing of humanized mabs To enhance the direct cell killing of the human IgG1 27 chimera (27hIgG1), we transferred selected mIgG3 constant region residues into the hIgG1 Fc domain to create an improved "i27G1" Lewis chimera with improved direct cell killing capacity in vitro. Y A glycan-binding mAb was generated. Figure 33 shows the amino acid and nucleic acid sequences of each light and heavy chain of the "i27G1" antibody. This antibody forms another aspect of the present invention.

[0189] Figure 32 shows the improved direct cell killing, functional affinity, and effector function of i27G1. The mIgG3 27 antibody was synthesized by Lewis y i27G1 has direct cell killing ability in AGS (i) and MCF7 (ii) cell lines expressing mIgG3, whereas chimeric 27hIgG1 does not. Our i27G1, which contains selected mIgG3 constant region residues, exhibits similar direct cell killing as 27mG3 mAb and significantly improved direct cell killing compared to 27hIgG1 in both AGS (i) and MCF7 (ii). Furthermore, i27G1 has significantly improved functional Lewis mAb activity compared to 27hIgG1 (iii). y Importantly, the effector function, ADCC, and CDC of i27G1 were all comparable to those of 27hIgG1 in MCF7 (iv and v). Collectively, these results demonstrate that transferring selected regions from the mIgG3 constant region into the 27hIgG1 backbone resulted in a hybrid mAb with direct cell killing capacity, increased functional affinity, and potent immune effector function.

[0190] References Altschul, S. F., W. Gish, W. Miller, E. W. Myers and D. J. Lipman (1990). ”Basic local alignment search tool.” J Mol Biol 215(3): 403-410. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller and D. J. Lipman (1997). ”Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.” Nucleic Acids Res 25(17): 3389-3402. Alvarez-Rueda, N., S. Leprieur, B. Clemenceau, S. Supiot, V. Sebille-Rivain, A. Faivre-Chauvet, F. Davodeau, F. Paris, J. Barbet, J. Aubry and S. Birkle (2007). ”Binding activities and antitumor properties of a new mouse / human chimeric antibody specific for GD2 ganglioside antigen.” Clin Cancer Res 13(18 Pt 2): 5613s-5620s. Ausubel, F. (1992). Short protocols in molecular biology, John Wiley & Sons. Barbas, C. F., 3rd, D. Hu, N. Dunlop, L. Sawyer, D. Cababa, R. M. Hendry, P. L. Nara and D. R. Azinovic, I., G. L. DeNardo, K. R. Lamborn, G. Mirick, D. Goldstein, B. M. Bradt, and S. J. DeNardo. 2006. ’Survival benefit associated with human anti-mouse antibody (HAMA) in patients with B-cell malignancies’, Cancer Immunol Immunother, 55: 1451-8. Burton (1994). ”In vitro evolution of a neutralizing human antibody to human immunodeficiency virus type 1 to enhance affinity and broaden strain cross-reactivity.” Proc Natl Acad Sci U S A 91(9): 3809-3813. Bird, R. E., K. D. Hardman, J. W. Jacobson, S. Johnson, B. M. Kaufman, S. M. Lee, T. Lee, S. H. Pope, G. S. Riordan and M. Whitlow (1988). ”Single-chain antigen-binding proteins.” Science 242(4877): 423-426. Bodanzsky, M. and A. Bodanzsky (1984). The practice of peptide synthesis. New York, Springer Verlag. Chou, H. H., H. Takematsu, S. Diaz, J. Iber, E. Nickerson, K. L. Wright, E. A. Muchmore, D. L. Nelson, S. T. Warren and A. Varki (1998). ”A mutation in human CMP-sialic acid hydroxylase occurred after the Homo-Pan divergence.” Proc Natl Acad Sci U S A 95(20): 11751-11756. D’Arcy, C. A., and M. Mannik. 2001. ’Serum sickness secondary to treatment with the murine-human chimeric antibody IDEC-C2B8 (rituximab)’, Arthritis Rheum, 44: 1717-8. Durrant, L. G., P. Noble and I. Spendlove (2012). ”Immunology in the clinic review series; focus on cancer: glycolipids as targets for tumour immunotherapy.” Clin Exp Immunol 167(2): 206-215. Eppstein, D. A., Y. V. Marsh, M. van der Pas, P. L. Felgner and A. B. Schreiber (1985). ”Biological activity of liposome-encapsulated murine interferon gamma is mediated by a cell membrane receptor.” Proc Natl Acad Sci U S A 82(11): 3688-3692. Furukawa, K., S. Welt, B. W. Yin, H. J. Feickert, T. Takahashi, R. Ueda and K. O. Lloyd (1990). ”Analysis of the fine specificities of 11 mouse monoclonal antibodies reactive with type 2 blood group determinants.” Mol Immunol 27(8): 723-732. Gram, H., L. A. Marconi, C. F. Barbas, 3rd, T. A. Collet, R. A. Lerner and A. S. Kang (1992). ”In vitro selection and affinity maturation of antibodies from a naive combinatorial immunoglobulin library.” Proc Natl Acad Sci U S A 89(8): 3576-3580. Hellstrom, I., H. J. Garrigues, U. Garrigues and K. E. Hellstrom (1990). ”Highly tumor-reactive, internalizing, mouse monoclonal antibodies to Le(y)-related cell surface antigens.” Cancer Res 50(7): 2183-2190. Holliger, P., T. Prospero and G. Winter (1993). ”“Diabodies”: small bivalent and bispecific antibody fragments.” Proc Natl Acad Sci U S A 90(14): 6444-6448. Holliger, P. and G. Winter (1993). ”Engineering bispecific antibodies.” Curr Opin Biotechnol 4(4): 446-449. Huston, J. S., D. Levinson, M. Mudgett-Hunter, M. S. Tai, J. Novotny, M. N. Margolies, R. J. Ridge, R. E. Bruccoleri, E. Haber, R. Crea and et al. (1988). ”Protein engineering of antibody binding sites: recovery of specific activity in an anti-digoxin single-chain Fv analogue produced in Escherichia coli.” Proc Natl Acad Sci U S A 85(16): 5879-5883. Hwang, K. J., K. F. Luk and P. L. Beaumier (1980). ”Hepatic uptake and degradation of unilamellar sphingomyelin / cholesterol liposomes: a kinetic study.” Proc Natl Acad Sci U S A 77(7): 4030-4034. Jespers, L. S., A. Roberts, S. M. Mahler, G. Winter and H. R. Hoogenboom (1994). ”Guiding the selection of human antibodies from phage display repertoires to a single epitope of an antigen.” Biotechnology (N Y) 12(9): 899-903. Jones, S. T. and M. M. Bendig (1991). ”Rapid PCR-cloning of full-length mouse immunoglobulin variable regions.” Biotechnology (N Y) 9(1): 88-89. Karlin, S. and S. F. Altschul (1990). ”Methods for assessing the statistical significance of molecular sequence features by using general scoring schemes.” Proc Natl Acad Sci U S A 87(6): 2264-2268. Karlin, S. and S. F. Altschul (1993). ”Applications and statistics for multiple high-scoring segments in molecular sequences.” Proc Natl Acad Sci U S A 90(12): 5873-5877. Kitamura, K., E. Stockert, P. Garin-Chesa, S. Welt, K. O. Lloyd, K. L. Armour, T. P. Wallace, W. J. Harris, F. J. Carr and L. J. Old (1994). ”Specificity analysis of blood group Lewis-y (Le(y)) antibodies generatedagainst synthetic and natural Le(y) determinants.” Proc Natl Acad Sci U S A 91(26): 12957-12961. Lefranc, M. P., V. Giudicelli, C. Ginestoux, J. Jabado-Michaloud, G. Folch, F. Bellahcene, Y. Wu, E. Gemrot, X. Brochet, J. Lane, L. Regnier, F. Ehrenmann, G. Lefranc and P. Duroux (2009). ”IMGT, the international ImMunoGeneTics information system.” Nucleic Acids Res 37(Database issue): D1006-1012. Loo, D., N. Pryer, P. Young, T. Liang, S. Coberly, K. L. King, K. Kang, P. Roberts, M. Tsao, X. Xu, B. Potts and J. P. Mather (2007). ”The glycotope-specific RAV12 monoclonal antibody induces oncosis in vitro and has antitumor activity against gastrointestinal adenocarcinoma tumor xenografts in vivo.” Mol Cancer Ther 6(3): 856-865. Marks, J. D., A. D. Griffiths, M. Malmqvist, T. P. Clackson, J. M. Bye and G. Winter (1992). ”By-passing immunization: building high affinity human antibodies by chain shuffling.” Biotechnology (N Y) 10(7): 779-783. Miotti, S., D. R. Negri, O. Valota, M. Calabrese, R. L. Bolhuis, J. W. Gratama, M. I. Colnaghi, and S. Canevari. 1999. ’Level of anti-mouse-antibody response induced by bi-specific monoclonal antibody OC / TR in ovarian-carcinoma patients is associated with longer survival’, Int J Cancer, 84: 62-8. Myers, E. W. and W. Miller (1989). ”Approximate matching of regular expressions.” Bull Math Biol 51(1): 5-37. Noble, P., I. Spendlove, S. Harding, T. Parsons and L. G. Durrant (2013). ”Therapeutic targeting of Lewis(y) and Lewis(b) with a novel monoclonal antibody 692 / 29.” PLoS One 8(2): e54892. Pai, L. H., R. Wittes, A. Setser, M. C. Willingham and I. Pastan (1996). ”Treatment of advanced solid tumors with immunotoxin LMB-1: an antibody linked to Pseudomonas exotoxin.” Nat Med 2(3): 350-353. Pearson, W. R. and D. J. Lipman (1988). ”Improved tools for biological sequence comparison.” Proc Natl Acad Sci U S A 85(8): 2444-2448. Pluckthun, A. (1991). ”Antibody engineering: advances from the use of Escherichia coli expression systems.” Biotechnology (N Y) 9(6): 545-551. Rabu, C., R. McIntosh, Z. Jurasova and L. Durrant (2012). ”Glycans as targets for therapeutic antitumor antibodies.” Future Oncol 8(8): 943-960. Reff, M. E. (1993). ”High-level production of recombinant immunoglobulins in mammalian cells.” Curr Opin Biotechnol 4(5): 573-576. Remington, R. (1980). Remington’s pharmaceutical sciences, Mack Pub. Co. Sambrook, J. (1989). Molecular cloning: A laboratory manual, Cold Spring Harbor Laboratory Press. Sambrook, J. and D. Russell (2001). Molecular cloning: A laboratory manual. Cold Spring Harbor, N.Y., Cold Spring Harbor Laboratory Press. Schier, R., A. McCall, G. P. Adams, K. W. Marshall, H. Merritt, M. Yim, R. S. Crawford, L. M. Weiner, C. Marks and J. D. Marks (1996). ”Isolation of picomolar affinity anti-c-erbB-2 single-chain Fv by molecular evolution of the complementarity determining regions in the center of the antibody binding site.” J Mol Biol 263(4): 551-567. Schuster, M., P. Umana, C. Ferrara, P. Brunker, C. Gerdes, G. Waxenecker, S. Wiederkum, C. Schwager, H. Loibner, G. Himmler and G. C. Mudde (2005). ”Improved effector functions of a therapeutic monoclonal Lewis Y-specific antibody by glycoform engineering.” Cancer Res 65(17): 7934-7941. Scott, A. M., D. Geleick, M. Rubira, K. Clarke, E. C. Nice, F. E. Smyth, E. Stockert, E. C. Richards, F. J. Carr, W. J. Harris, K. L. Armour, J. Rood, A. Kypridis, V. Kronina, R. Murphy, F. T. Lee, Z. Liu, K. Kitamura, G. Ritter, K. Laughton, E. Hoffman, A. W. Burgess and L. J. Old (2000). ”Construction, production, and characterization of humanized anti-Lewis Y monoclonal antibody 3S193 for targeted immunotherapy of solid tumors.” Cancer Res 60(12): 3254-3261. Sidman, K. R., W. D. Steber, A. D. Schwope and G. R. Schnaper (1983). ”Controlled release of macromolecules and pharmaceuticals from synthetic polypeptides based on glutamic acid.” Biopolymers 22(1): 547-556. Schroff, R. W., K. A. Foon, S. M. Beatty, R. K. Oldham, and A. C. Morgan, Jr. 1985. ’Human anti murine immunoglobulin responses in patients receiving monoclonal antibody therapy’, Cancer Res, 45: 879-85. Soejima, M. and Y. Koda (2005). ”Molecular mechanisms of Lewis antigen expression.” Leg Med (Tokyo) 7(4): 266-269. Stemmer, W. P. (1994). ”Rapid evolution of a protein in vitro by DNA shuffling.” Nature 370(6488): 389-391. Stewart, J. and J. Young (1984). Solid phase peptide synthesis. Rockford, Illinois, Pierce Chemical Company. Tolcher, A. W., S. Sugarman, K. A. Gelmon, R. Cohen, M. Saleh, C. Isaacs, L. Young, D. Healey, N. Onetto and W. Slichenmyer (1999). ”Randomized phase II study of BR96-doxorubicin conjugate in patients with metastatic breast cancer.” J Clin Oncol 17(2): 478-484. Torelli, A. and C. A. Robotti (1994). ”ADVANCE and ADAM: two algorithms for the analysis of global similarity between homologous informational sequences.” Comput Appl Biosci 10(1): 3-5. Traunecker, A., A. Lanzavecchia and K. Karjalainen (1991). ”Bispecific single chain molecules (Janusins) target cytotoxic lymphocytes on HIV infected cells.” EMBO J 10(12): 3655-3659. Trill, J. J., A. R. Shatzman and S. Ganguly (1995). ”Production of monoclonal antibodies in COS and CHO cells.” Curr Opin Biotechnol 6(5): 553-560. Ward, E. S., D. Gussow, A. D. Griffiths, P. T. Jones and G. Winter (1989). ”Binding activities of a repertoire of single immunoglobulin variable domains secreted from Escherichia coli.” Nature 341(6242): 544-546. Yin, B. W., C. L. Finstad, K. Kitamura, M. G. Federici, M. Welshinger, V. Kudryashov, W. J. Hoskins, S. Welt and K. O. Lloyd (1996). ”Serological and immunochemical analysis of Lewis y (Ley) blood group antigen expression in epithelial ovarian cancer.” Int J Cancer 65(4): 406-412. Yuriev, E., W. Farrugia, A. M. Scott and P. A. Ramsland (2005). ”Three-dimensional structures of carbohydrate determinants of Lewis system antigens: implications for effective antibody targeting of cancer.” Immunol Cell Biol 83(6): 709-717. Zhang, G., H. Zhang, Q. Wang, P. Lal, A. M. Carroll, M. de la Llera-Moya, X. Xu and M. I. Greene (2010). ”Suppression of human prostate tumor growth by a unique prostate-specific monoclonal antibody F77 targeting a glycolipid marker.” Proc Natl Acad Sci U S A 107(2): 732-737. Zhong, L. T., A. Manzi, E. Skowronski, L. Notterpek, A. L. Fluharty, K. F. Faull, I. Masada, S. Rabizadeh, M. Varsanyi-Nagy, Y. Ruan, J. D. Oh, L. L. Butcher and D. E. Bredesen (2001). ”A monoclonal antibody that induces neuronal apoptosis binds a metastasis marker.” Cancer Res 61(15): 5741-5748.

Claims

1. Fuca1-2Galb1-4(Fuca1-3)GlcNAc(Le ｙ 2. An isolated antibody or antigen-binding fragment thereof that specifically binds to The antibody or antigen-binding fragment thereof a CDRH1 having the amino acid sequence GFTFSHYY, a CDRH2 having the amino acid sequence ISNDGDNT, and a CDRH3 having the amino acid sequence ARGKYDGAWFAY, and CDRL1 having the amino acid sequence QIIVHTNGNTY, CDRL2 having the amino acid sequence KVS, and CDRL3 having the amino acid sequence FQGSHVPFT; Including, An antibody or an antigen-binding fragment thereof.

2. An antibody or antigen-binding fragment thereof described in claim 1, wherein the CDRs are supported by a human antibody framework.

3. Amino acid sequence: EVKLVESGGGLVQPGGSLKLSCATSGFTFSHYYMYWVRQTPEKRLEWVAYISNDGDNTYYPDTIRGRFTISRDNARNTLYLQMSRLKSEDTAMYYCARGKYDGAWFAYWGQGTLVTVSA and a VH having the amino acid sequence: DVLMTQTPLSLPVSLGDQASISCRSSQIIVHTNGNTYLEWYLQKPGQSPNLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEDLGVYYCFQGSHVPFTFGSGTKLEIKR and a VL having the following structure:

4. The antibody or antigen-binding fragment thereof according to any one of claims 1 to 3, further comprising a human constant region.

5. Amino acid sequence: EVKLVESGGGLVQPGGSLKLSCATSGFTFHYYMYWVRQTPEKRLEWVAYISNDGDNTYYPDTIRGRFTISRDNARNTLYLQMSRLKSEDTAMYYCARGKYDGAWFAYWGQGT LVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKT HTCPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKT ISKAKGQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK and a heavy chain having the amino acid sequence: DVLMTQTPLSLPVSLGDQASISCRSSQIIVHTNGNTYLEWYLQKPGQSPNLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEDLGVYYCFQGSHVPFTFGSGTKL EIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC and a light chain having the following structure:

6. Amino acid sequence: QVQLVESGGGVVQPGRSLRLSCAASGFTFSHYYMYWVRQAPGKGLEWVAYISNDGDNTYYPDTIRGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARGKYDGAWFAYWGQG TLVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDK THTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEK TISKAKGQPREPQVYTLPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK and a heavy chain having the amino acid sequence: MEAPAQLLFLLLLWLPDTTGDVVMTQSPLSLPVTPGEPASISCRSSQIIVHTNGNTYLEWYLQKPGQSPQLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEEDVGVYYCFQGSHV PFTFGQGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC and a light chain having the following structure:

7. The antibody or antigen-binding fragment thereof of any one of claims 1 to 6, which is linked or conjugated to a chemotherapeutic or cytotoxic agent.

8. A pharmaceutical composition comprising an antibody or antigen-binding fragment thereof described in any one of claims 1 to 7, and a pharma- ceutically acceptable excipient, diluent, carrier, buffer, or stabilizer.

9. An antibody or antigen-binding fragment thereof according to any one of claims 1 to 7 for use in medicine.

10. The antibody or antigen-binding fragment thereof according to any one of claims 1 to 7 for use in a method for treating or preventing a tumor.

11. A product comprising an antibody or antigen-binding fragment thereof according to any one of claims 1 to 7 and an active agent as a combined preparation for simultaneous, separate or sequential use in the treatment of tumors.

12. The antibody or antigen-binding fragment thereof of claim 10 or the product of claim 11, wherein the tumor is a colorectal, gastric, pancreatic, lung, ovarian, or breast tumor.

13. A nucleic acid encoding the antibody or antigen-binding fragment thereof according to any one of claims 1 to 6.

14. Amino acid sequence: MNFWLSLIFLVLVLKGVQCEVKLVESGGGLVQPGGSLKLSCATSGFTFSHYYMYWVRQTPEKRLEWVAYISNDGDNTYYPDTIRGRFTISRDNARNTLYLQMSRLKSEDTAMYYCAR GKYDGAWFAYWGQGTLVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDK KVEPKSCDKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHTAWTQPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKPKGRAQTPQVYTIPPPREQMSKKKVSLTCLVTNFFSEAISVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK and the heavy chain having the amino acid sequence: MKLPVRLLLVLMFWIPASISDVLMTQTPLSLPVSLGDQASISCRSSQIIVHTNGNTYLEWYLQKPGQSPNLLIYKVSNRFSGVPDRFSGSGSGTDFTLKISRVEAEDLGVYYCFQGSHVP FTFGSGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC An antibody or antigen-binding fragment thereof comprising a light chain having the following structure:

15. The nucleic acid sequence: atgaacttctggctcagcttgattttccttgtccttgttttaaaaggtgtccagtgtgaagtgaagctggtggagtctgggggaggcttagtgcagcctggagggtccctgaaactctcctgtgcaacctctggattcactttcagtcactattacatgtattgggttcgccagactccagagaagaggctggagtgggtcgcatacattagtaatgatggtgataacacctattatccagacactataaggggccgattcaccatctccagagacaatgccaggaacaccctgtacctgcaaatgagccgtctgaagtctgaggacacagccatgtattactgtgcaagagggaagtacgacggggcctggtttgcttactggggccaagggactctggtcactgtctctagcgcttccaccaagggcccatcggtcttccccctggcaccctcctccaagagcacctctgggggcacagcggccctgggctgcctggtcaaggactacttccccgaaccggtgacggtgtcgtggaactcaggcgccctgaccagcggcgtgcacaccttcccggctgtcctacagtcctcaggactctactccctcagcagcgtggtgaccgtgccctccagcagcttgggcacccagacctacatctgcaacgtgaatcacaagcccagcaacaccaaggtggacaagaaagttgagcccaaatcttgtgacaaaactcacacatgcccaccgtgcccagcacctgaactcctggggggaccgtcagtcttcctcttccccccaaaacccaaggacaccctcatgatctcccggacccctgaggtcacatgcgtggtggtggacgtgagccacgaagaccctgaggtcaagttcaactggtacgtggacggcgtggaggtgcatacagcctggacacagccccgtgaagagcagtacaacagtacctaccgagtggtcagtgtcctcaccgtcctgcaccaggactggctgaatggcaaggagtacaagtgcaaggtctccaacaaagccctcccagcccccatcgagaaaaccatctccaaacccaaaggaagagcccagacacctcaagtatacaccatacccccacctcgtgaacaaatgtccaagaagaaggttagtctgacctgcctggtcaccaacttcttctctgaagccatcagtgtggagtgggagagcaatgggcagccggagaacaactacaagaccacgcctcccgtgctggactccgacggctccttcttcctctacagcaagctcaccgtggacaagagcaggtggcagcaggggaacgtcttctcatgctccgtgatgcatgaggctctgcacaaccactacacgcagaagagcctctccctgtctccgggtaaa The heavy chain and nucleic acid sequence encoded by: atgaagttgcctgttaggctgttggtgctgatgttctggattcctgcttccatcagtgatgttttgatgacccaaactccactctccctgcctgtcagtcttggagatcaagcctccatctcttgcagatctagtcagatcattgtacatactaatggaaacacctatttagaatggtacctgcagaaaccaggccagtctccaaacctcctgatctacaaagtttccaaccgattttctggggtcccagacaggttcagtggcagtggatcagggacagatttcacactcaagatcagcagagtggaggctgaggatctgggagtttattactgctttcaaggttcacatgttccattcacgttcggctcggggacaaagttggaaataaaacgtacggtagcggccccatctgtcttcatcttcccgccatctgatgagcagttgaaatctggaactgcctctgttgtgtgcctgctgaataacttctatcccagagaggccaaagtacagtggaaggtggataacgccctccaatcgggtaactcccaggagagtgtcacagagcaggacagcaaggacagcacctacagcctcagcagcaccctgacgctgagcaaagcagactacgagaaacacaaagtctacgcctgcgaagtcacccatcagggcctgagctcgcccgtcacaaagagcttcaacaggggagagtgt The antibody or antigen-binding fragment thereof according to claim 14, comprising a light chain encoded thereby.

Images