Downstream processing of bispecific antibody constructs

Abstract

The present invention provides an improved manufacturing process for producing a bispecific antibody product comprising at least two binding domains. The process comprises at least a process comprising a step of refolding a construct, wherein one or more domains comprise disulfide bonds, and wherein the construct is contacted with a refolding buffer to refold the construct into its native form, the buffer comprising (i) an oxidizing and / or reducing agent at a concentration of at least 0.1 mM, and (ii) a chaotropic agent at a concentration > 1 M, and wherein the pH of the refolding buffer corresponds to + / - 4.5 of the pi value of the first domain or + / - 4.0 of the pi value of the entire construct.

Description

Technical Field

[0001] This invention relates to methods of biotechnology, and more particularly to downstream manufacturing methods for producing bispecific antibody constructs. Background Technology

[0002] For years, protein-based drugs have been among the fastest-growing and most promising therapeutic agents, playing a significant role in virtually every area of ​​medicine and being among the fastest-growing therapeutic agents in clinical (pre-)development and as commercial products (Leader, Nature Reviews Drug Discovery, January 7, 2008, 21-39). Compared to small chemical drugs, protein drugs exhibit high specificity and activity at relatively low concentrations and typically provide therapies for high-impact diseases such as various cancers, autoimmune diseases, and metabolic disorders (Roberts, Trends Biotechnol, July 2014; 32(7):372-80; Wang, Int J Pharm, August 20, 1999; 185(2):129-88).

[0003] Advances in commercial-scale purification methods have enabled the production of protein-based drugs, such as recombinant proteins, in high purity on the first manufacturing run. However, proteins are only marginally stable and are highly susceptible to both chemical and physical degradation, even during upstream manufacturing processes. Chemical degradation refers to modifications involving covalent bonds, such as deamidation, oxidation, cleavage or formation of new disulfide bridges, hydrolysis, isomerization, or deglycosylation. Physical degradation includes protein unfolding, unwanted adsorption to surfaces, and aggregation. Addressing these physical and chemical instabilities is one of the most challenging tasks in protein drug development (Chi et al., Pharm Research, Vol. 20, No. 9, Sep 2003, pp. 1325-1336; Roberts, Trends Biotechnol, July 2014; 32(7):372-80).

[0004] Therefore, despite advancements in manufacturing, new protein-based drugs still require innovative new manufacturing methods to avoid product quality issues such as protein aggregation. This impacts upstream manufacturing, downstream manufacturing, storage, and application.

[0005] These new protein-based drugs include, for example, bispecific (monoclonal) antibodies. Bispecific antibodies are artificial proteins that can bind to two different types of antigens simultaneously. They are known in several structural forms and their applications in cancer immunotherapy and drug delivery have been explored (Fan, Gaowei; Wang, Zujian; Hao, Mingju; Li, Jinming (2015). "Bispecific antibodies and their applications". Journal of Hematology & Oncology. 8: 130).

[0006] Generally speaking, bispecific antibodies can be IgG-like, i.e., full-length bispecific antibodies, or non-IgG-like bispecific antibodies that are non-full-length antibody constructs. Full-length bispecific antibodies typically retain the structure of a traditional monoclonal antibody (mAb) with two Fab arms and an Fc region, except that the two Fab sites bind different antigens. Non-full-length bispecific antibodies lack the entire Fc region. These include chemically linked Fabs, Fab regions alone, and various types of bivalent and trivalent single-chain variable fragments (scFvs). Fusion proteins that mimic the variable domains of two antibodies also exist. Among these newer forms, the most likely to be further developed is the bispecific T-cell conjugate (BiTE®) (Yang, Fa; Wen, Weihong; Qin, Weijun (2016). "Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies". International Journal of Molecular Sciences. 18 (1): 48).

[0007] Bispecific molecules such as BiTE ® The antibody construct is a recombinant protein construct prepared from two flexibly linked antibody-derived binding domains. (BiTE) ® One binding domain of the antibody construct is specific to selected tumor-associated surface antigens on target cells; the second binding domain is specific to CD3 (a subunit of the T cell receptor complex on T cells). Through its specific design, BiTE... ®The antibody construct is uniquely suited for transiently binding T cells to target cells while simultaneously and potently activating the inherent cytolytic potential of T cells against target cells. This is the first-generation BiTE antibody developed for clinical use as AMG 103 and AMG 110. ® An important further development of the antibody constructs (see WO 99 / 54440 and WO 2005 / 040220) is the provision of a bispecific antibody construct (WO 2008 / 119567) that binds to a background-independent epitope at the N-terminus of the CD3ε chain. BiTE binding to this selected epitope... ® The antibody construct not only showed efficacy against humans and marmosets (… Callithrix jacchus, velvety-crowned tamarin ( Saguinus oedipus ) or squirrel monkey ( Saimiri The CD3ε chain of the sciureus exhibits cross-species specificity, and because it recognizes this specific epitope (rather than the epitope of the CD3-binding in previously described bispecific T-cell conjugating molecules), it does not nonspecifically activate T cells to the same extent as observed with previous generations of T-cell conjugating antibodies. This reduction in T-cell activation is associated with fewer or reduced T-cell redistribution in patients, which has been identified as a risk of side effects.

[0008] Currently, approximately 25% (m / m) or more of bispecific antibody constructs, such as bispecific T-cell conjugates (BiTE®) molecules (ScFc-BiTE®) conjugated with single-chain Fc and produced via industrial mammalian CHO cell cultures (expression, secretion, collection from a bioreactor, and purification, e.g., via a protein A column), can be obtained in aggregate form, which reduces yield and increases production costs. The next downstream purification step (typically cation exchange chromatography (CEX)) removes the aggregates. Therefore, downstream aggregate reduction is necessary.

[0009] Traditionally, redox refolding occurs after proteins aggregate in inclusion bodies in microbial systems (E. coli). In these cases, high concentrations of denaturing agents (6M guanidine, urea, etc.) are typically required. Oxidizing agents (copper, dissolved oxygen, etc.) are used to form disulfide bonds, or redox agents (cysteine / cystine, cysteine / cystamine, etc.) are used to rearrange scrambled disulfide bonds. Although it is generally assumed that industrial mammalian cells express proteins with the desired, correct higher-order structure (folding) and disulfide bond linkages, this is not always the case.

[0010] Several cases have been described in the literature involving further processing of proteins and antibodies produced in mammalian cells with the aim of refolding the proteins and generating or rearranging disulfide bonds. For example, the TNF-Fc fusion protein (etanercept) was treated with cysteine / cystine redox to form and rearrange scrambled disulfide bonds, thereby minimizing the subsequent elution of HIC fractions (Sassenfeld, HM, Remmele, RL, & McCoy, RE 2007, Increased recovery of active proteins, US 7157557; Sassenfeld, HM, Remmele, RL, & McCoy, RE 2009, Increased recovery of active proteins (antibodies), US 7544874 B2). In another case, a high percentage (16%) of open disulfide bonds were observed in the Fab region of the IgG1 antibody (Chaderjian, WB, Chin, ET, Harris, RJ, & Etcheverry, TM 2005. Effect of copper sulfateon performance of a serum-free CHO cell culture process and the level of freethiol in the recombinant antibody expressed. Biotechnol. Prog., 21, (2) 550-553). The formation of disulfide bonds was accomplished by oxidation through the addition of 50-100 uM copper sulfate to the bioreactor (Chaderjian, Chin, Harris, & Etcheverry 2005).Redox refolding of IgG2 antibodies to enrich desired IgG2A or IgG2B disulfide isoforms (Dillon, TM, Speed-Ricci, M., Vezina, C., Flynn, GC, Liu, YD, Rehder, DS, Plant, M., Henkle, B., Li, Y., Varnum, B., Wypych, J., Balland, A., & Bondarenko, PV 2008. Structural and functional characterization of disulfide isoforms of the human IgG2 subclass. J. Biol. Chem., 283, 16206-16215, Dillon, TM, Rehder, DS, Bondarenko, PV, Ricci, M., Gadgil, HS, Banks, D., Zhou, J., Lu, Y., Goetze, A., & Zhang, Y. 2011. Methods for refolding of recombinant antibodies. Patent US7928205. IgG2 molecules immobilized on protein A columns were also subjected to redox refolding and redox treatment to enrich the desired IgG2 disulfide isoform (Zhou, J. & Hong, T. 2006, Method for refolding polypeptides, WO 2006 / 060083).Weaker denaturing conditions were used for single-chain Fv antibodies containing 1M urea (Chen, LH, Huang, Q., Wan, L., Zeng, LY, Li, SF, Li, YP, Lu, XF, & Cheng, JQ 2006. Expression, purification, and in vitro refolding of a humanized single-chain Fv antibody against humanCTLA4 (CD152). Protein Expr. Purif., 46, (2) 495-502). However, no applicable refolding conditions have been previously described for refolding aggregated bispecific antibody constructs, such as BiTE® antibody constructs, which allow for rapid, versatile, and resource-efficient steps that can be incorporated into routine downstream processing to increase overall yield. Even larger bispecific antibody constructs, containing domains such as extended half-life and typically at least 100 kDa, present even greater processing challenges in downstream processing. For example, single-chain Fc (ScFc)BiTE® antibody constructs are typically large (approximately 105 kDa) and contain long linkers designed for protein engineering work. In contrast, almost no single-chain natural mammalian proteins (less than 5%) are larger than 100 kDa. In this regard, those skilled in the art consider standard IgG antibodies to consist of two heavy chains of approximately 50 kDa and two light chains of approximately 25 kDa, which fold separately and then bond together. In contrast, single-chain antibody constructs such as scFc BiTE® antibody constructs fold into only one heavily engineered chain. Therefore, prior learning on downstream processing is needed for heavily engineered single-chain antibody constructs, and further refinement is required for non-natural high-molecular-weight constructs. Summary of the Invention

[0011] Surprisingly, a downstream method can be provided that ensures improved quality of bispecific antibody constructs, thereby increasing product yield, and can be integrated into existing downstream processing steps.

[0012] Therefore, in one aspect, within the context of this invention, a downstream method for purifying a bispecific antibody construct comprising:

[0013] • The first domain, which binds to the target, preferably the tumor target, with a pI of 5.0 to 9.5;

[0014] • A second domain that binds to an extracellular domain of the human and macaque CD3ε chain; and

[0015] • Optionally, a third structural domain comprising two polypeptide monomers, each polypeptide monomer comprising a hinge, a CH2 domain, and a CH3 domain, wherein the two polypeptide monomers are fused together via a peptide linker, wherein the construct comprises at least one open disulfide bond, the method comprising the step of refolding the construct, wherein one or more domains comprise disulfide bonds, and wherein

[0016] The construct is contacted with a refolding buffer to refold the construct into its native form, the buffer comprising (i) a redox and / or oxidizing agent at a concentration of at least 0.1 mM, and (ii) a liquefying agent at a concentration > 1 M, and wherein the pH of the refolding buffer corresponds to + / - 4.5 of the pI value of the first domain or + / - 4.0 of the pI value of the entire construct.

[0017] According to the aforementioned aspects, it is also envisioned that the oxidant be selected from the group consisting of copper(II), dissolved oxygen and (L)-dehydroascorbic acid (DHA), preferably copper(II).

[0018] Based on the aforementioned aspects, it is also conceivable that the oxidizing agent is copper(II) sulfate.

[0019] According to the aforementioned aspects, it is also envisioned that the oxidant is present at a concentration of 0.1 to 10 mM, preferably 0.1 to 1 mM.

[0020] According to the aforementioned aspects, it is also envisioned that the redox agent be selected from the group consisting of: reduced glutathione / oxidized glutathione, cysteine / cystine, cysteine / cystamine, cysteine / cystamine, dithiothreitol (DTT), β-mercaptoethanol and glutathione.

[0021] According to the aforementioned aspects, it is also envisioned that the redox agent is present at a concentration of 0.1 to 10 mM, preferably 0.1 to 1 mM.

[0022] Based on the aforementioned aspects, it is also envisioned that the refolding buffer contains an oxidizing agent but not a redox agent.

[0023] According to the aforementioned aspects, it is also envisioned that the liquid ionizer be selected from the group consisting of: arginine, urea, dimethylurea, methylurea, ethylurea, organic solvents, detergents, high temperature (preferably above 40, 50, 60, 70 or 80ºC) and guanidine (or its protonated form, guanidinium), preferably guanidinium.

[0024] Based on the aforementioned aspects, it is also envisioned that the liquid agent exists at a concentration in the range of 1.2 to 2 M.

[0025] According to the aforementioned aspects, a pH value of at least 4.9 is also envisioned, preferably in the range of pH 5.0 to 9.5, more preferably about 5.0 to 6.0 or 6.5 to 9.0, and more preferably 5.0 or 7.0 to 8.0.

[0026] Based on the aforementioned aspects, it is also envisioned that the first structural domain combines CD33, CDH19, MSLN, FLT3, BCMA, CD19, MUC17, EpCAM, EGFRviii, DLL3, CLDN 18, CDH3, and / or PSMA.

[0027] According to the aforementioned aspects, it is also envisioned that, compared with the monomer content before redox refolding, the monomer content of the construct, as determined by size exclusion chromatography (monomer peak percentage) after the application of a refolding buffer, increases by at least 25%, and preferably by at least 40%, more preferably at least 60%, 65%, at least 70%, or even at least 85%, of the monomer content.

[0028] Based on the aforementioned aspects, it is also envisioned that the method be carried out at room temperature (approximately 25ºC).

[0029] Based on the aforementioned aspects, it is also envisioned that the incubation time of the method is 1 to 24 hours, preferably 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 hours, and more preferably 4 hours.

[0030] According to the aforementioned aspects, it is also envisioned that the contact between the construct and the refolding buffer be carried out (i) in the processing fluid (pool) during a downstream purification step or (ii) on a chromatographic column during a downstream purification step.

[0031] According to the aforementioned aspects, it is also envisioned that the concentration of the construct (i) in the liquid cell is from 0.5 mg / ml to 15 mg / ml, preferably from 0.7 mg / ml to 12 mg / ml or (ii) on the chromatographic column is from 5 mg / ml to 15 mg / ml, preferably from 10 mg / ml to 15 mg / ml.

[0032] Based on the aforementioned aspects, it is also envisioned that the liquid pool is harvested cell culture medium (HCCF), chromatographic eluent pool, protein A, protein L or protein G affinity chromatography eluent pool, filtered virus inactivation pool, cation exchange chromatography (CEX) monomer pool and CEX high molecular weight post-peak pool.

[0033] Based on the aforementioned aspects, it is also envisioned that the method of the present invention preferably includes the following steps:

[0034] (a) Provide harvested cell culture medium (HCCF) containing the bispecific antibody construct secreted by mammalian cells;

[0035] (b) Under conditions suitable for association between the construct and the separation matrix, the HCCF is contacted with a first separation matrix for chromatographic purification, wherein the separation matrix is ​​an affinity resin selected from the group consisting of: protein A, protein G, protein L, and synthetic analog affinity resins.

[0036] (c) Wash the first separation matrix;

[0037] (d) Elute the construct from the first separation matrix;

[0038] (e) Inactivate the virus in the eluent containing the construct;

[0039] (f) Under conditions suitable for association between the construct and the separation matrix, an eluent containing the construct is contacted with a second separation matrix for chromatographic purification, wherein the separation matrix is ​​a cation exchange chromatographic matrix;

[0040] (g) Wash the second separation matrix;

[0041] (h) The construct is eluted from the second separation matrix, wherein the construct is separated into two fractions: (i) a monomer pool containing construct monomers and (ii) a high molecular weight pool containing construct aggregates.

[0042] (i) The virus is separated from the eluent containing fraction (i) by filtration; and

[0043] (j) Optionally, the eluent may be subjected to further ultrafiltration and / or percolation steps.

[0044] The construct is contacted with a refolding buffer in the cell culture medium before harvest or in the HCCF in step (a) to refold the construct back to its native form.

[0045] According to the aforementioned aspect, alternatively, in at least one of steps (b) or (f) as described above, the construct is brought into contact with a refolding buffer on the separation matrix (on the column) to refold the construct back into its natural form.

[0046] According to the aforementioned aspect, alternatively, in at least one of steps (d), (e), or (h) as described above, the construct is brought into contact with a refolding buffer in the processing fluid (pool) to refold the construct back into its natural form.

[0047] According to the aforementioned aspects, it is also envisioned that the first binding structural domain of the construct includes a VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from the following groups, and a VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from the following groups:

[0048] (a) CDR-H1 as depicted in SEQ ID NO: 4, CDR-H2 as depicted in SEQ ID NO: 5, CDR-H3 as depicted in SEQ ID NO: 6, CDR-L1 as depicted in SEQ ID NO: 1, CDR-L2 as depicted in SEQ ID NO: 2, and CDR-L3 as depicted in SEQ ID NO: 3,

[0049] (b) CDR-H1 as depicted in SEQ ID NO: 29, CDR-H2 as depicted in SEQ ID NO: 30, CDR-H3 as depicted in SEQ ID NO: 31, CDR-L1 as depicted in SEQ ID NO: 34, CDR-L2 as depicted in SEQ ID NO: 35, and CDR-L3 as depicted in SEQ ID NO: 36,

[0050] (c) CDR-H1 as depicted in SEQ ID NO: 42, CDR-H2 as depicted in SEQ ID NO: 43, CDR-H3 as depicted in SEQ ID NO: 44, CDR-L1 as depicted in SEQ ID NO: 45, CDR-L2 as depicted in SEQ ID NO: 46, and CDR-L3 as depicted in SEQ ID NO: 47,

[0051] (d) CDR-H1 as depicted in SEQ ID NO: 53, CDR-H2 as depicted in SEQ ID NO: 54, CDR-H3 as depicted in SEQ ID NO: 55, CDR-L1 as depicted in SEQ ID NO: 56, CDR-L2 as depicted in SEQ ID NO: 57, and CDR-L3 as depicted in SEQ ID NO: 58,

[0052] (e) CDR-H1 as depicted in SEQ ID NO: 65, CDR-H2 as depicted in SEQ ID NO: 66, CDR-H3 as depicted in SEQ ID NO: 67, CDR-L1 as depicted in SEQ ID NO: 68, CDR-L2 as depicted in SEQ ID NO: 69, and CDR-L3 as depicted in SEQ ID NO: 70,

[0053] (f) CDR-H1 as depicted in SEQ ID NO: 83, CDR-H2 as depicted in SEQ ID NO: 84, CDR-H3 as depicted in SEQ ID NO: 85, CDR-L1 as depicted in SEQ ID NO: 86, CDR-L2 as depicted in SEQ ID NO: 87, and CDR-L3 as depicted in SEQ ID NO: 88,

[0054] (g) CDR-H1 as depicted in SEQ ID NO: 94, CDR-H2 as depicted in SEQ ID NO: 95, CDR-H3 as depicted in SEQ ID NO: 96, CDR-L1 as depicted in SEQ ID NO: 97, CDR-L2 as depicted in SEQ ID NO: 98, and CDR-L3 as depicted in SEQ ID NO: 99,

[0055] (h) CDR-H1 as depicted in SEQ ID NO: 105, CDR-H2 as depicted in SEQ ID NO: 106, CDR-H3 as depicted in SEQ ID NO: 107, CDR-L1 as depicted in SEQ ID NO: 109, CDR-L2 as depicted in SEQ ID NO: 110, and CDR-L3 as depicted in SEQ ID NO: 111,

[0056] (i) CDR-H1 as depicted in SEQ ID NO: 115, CDR-H2 as depicted in SEQ ID NO: 116, CDR-H3 as depicted in SEQ ID NO: 117, CDR-L1 as depicted in SEQ ID NO: 118, CDR-L2 as depicted in SEQ ID NO: 119, and CDR-L3 as depicted in SEQ ID NO: 120,

[0057] (j) CDR-H1 as depicted in SEQ ID NO: 126, CDR-H2 as depicted in SEQ ID NO: 127, CDR-H3 as depicted in SEQ ID NO: 128, CDR-L1 as depicted in SEQ ID NO: 129, CDR-L2 as depicted in SEQ ID NO: 130, and CDR-L3 as depicted in SEQ ID NO: 131,

[0058] (k) CDR-H1 as depicted in SEQ ID NO: 137, CDR-H2 as depicted in SEQ ID NO: 138, CDR-H3 as depicted in SEQ ID NO: 139, CDR-L1 as depicted in SEQ ID NO: 140, CDR-L2 as depicted in SEQ ID NO: 141, and CDR-L3 as depicted in SEQ ID NO: 142,

[0059] (l) CDR-H1 as depicted in SEQ ID NO: 152, CDR-H2 as depicted in SEQ ID NO: 153, CDR-H3 as depicted in SEQ ID NO: 154, CDR-L1 as depicted in SEQ ID NO: 155, CDR-L2 as depicted in SEQ ID NO: 156, and CDR-L3 as depicted in SEQ ID NO: 157,

[0060] (m) CDR-H1 as depicted in SEQ ID NO: 167, CDR-H2 as depicted in SEQ ID NO: 168, CDR-H3 as depicted in SEQ ID NO: 169, CDR-L1 as depicted in SEQ ID NO: 170, CDR-L2 as depicted in SEQ ID NO: 171, and CDR-L3 as depicted in SEQ ID NO: 172,

[0061] (n) CDR-H1 as depicted in SEQ ID NO: 203, CDR-H2 as depicted in SEQ ID NO: 204, CDR-H3 as depicted in SEQ ID NO: 205, CDR-L1 as depicted in SEQ ID NO: 206, CDR-L2 as depicted in SEQ ID NO: 207, and CDR-L3 as depicted in SEQ ID NO: 208;

[0062] (o) CDR-H1 as depicted in SEQ ID NO: 214, CDR-H2 as depicted in SEQ ID NO: 215, CDR-H3 as depicted in SEQ ID NO: 216, CDR-L1 as depicted in SEQ ID NO: 217, CDR-L2 as depicted in SEQ ID NO: 218, and CDR-L3 as depicted in SEQ ID NO: 219;

[0063] (p) CDR-H1 as depicted in SEQ ID NO: 226, CDR-H2 as depicted in SEQ ID NO: 227, CDR-H3 as depicted in SEQ ID NO: 228, CDR-L1 as depicted in SEQ ID NO: 229, CDR-L2 as depicted in SEQ ID NO: 230, and CDR-L3 as depicted in SEQ ID NO: 231; and

[0064] (q) CDR-H1 as depicted in SEQ ID NO: 238, CDR-H2 as depicted in SEQ ID NO: 239, CDR-H3 as depicted in SEQ ID NO: 240, CDR-L1 as depicted in SEQ ID NO: 241, CDR-L2 as depicted in SEQ ID NO: 242, and CDR-L3 as depicted in SEQ ID NO: 243.

[0065] Based on the aforementioned aspects, it is also envisioned that the method be applied to constructs produced through upstream continuous manufacturing methods.

[0066] According to another aspect, a device is also envisioned to perform the method of the present invention.

[0067] According to another aspect, it is also envisioned to provide bispecific antibody constructs purified by the method of the present invention. Attached Figure Description

[0068] Figure 1 : Figure 1 An example setup of the downstream purification method steps for purifying the bispecific antibody construct according to the invention is shown. The names of the produced materials correspond to the processing liquids (cells) after the respective downstream purification steps, and samples of these liquids are collected for size exclusion chromatography (SEC) analysis. Apart from the percentage of monomers determined by SEC, the remaining percentage of the material is aggregates. Low molecular weight (LMW) substances constitute only a small portion. Samples from the virus inactivation cell (FVIP) filtered for protein A and the cation exchange chromatography (CEX) HMW cell were used for redox refolding.

[0069] Figure 2 : Figure 2 Size exclusion chromatogram of the CEX HMW (post-peak) cell of an exemplary CD33xCD3 bispecific antibody construct aggregated after treatment with guanidine salt only (GuHCl) at pH 8 is shown.

[0070] Figure 3 : Figure 3Size exclusion chromatograms of the HMW (post-peak) cells of an exemplary CD33xCD3 bispecific antibody construct after treatment with GuHCl and copper(II) or dehydroascorbic acid are shown.

[0071] Figure 4 : Figure 4 The size exclusion chromatogram of the aggregated exemplary CD33xCD3 bispecific antibody construct protein A FVIP cell after treatment with GuHCl is shown.

[0072] Figure 5 : Figure 5 Size exclusion chromatogram of the CEX HMW (post-peak) cell of an exemplary CD19xCD3 bispecific antibody construct after treatment with GuHCl and copper(II) is shown.

[0073] Figure 6 : Figure 6 Size exclusion chromatogram of the CEX HMW (post-peak) cell of an exemplary CLDNxCD3 bispecific antibody construct after treatment with GuHCl and copper(II) is shown. Detailed Implementation

[0074] Preferably, this document describes a downstream method for providing optimized redox refolding conditions for bispecific antibody constructs, preferably single-chain antibody constructs, such as bispecific T-cell conjugate (BiTE®) antibody constructs, especially scFc BiTE® antibody constructs, which advantageously improve the quality of the antibody construct product. This includes reducing aggregation, increasing the yield of monomeric antibody constructs, and / or reducing the occurrence of isotypes, thereby improving overall process economics. This method can be generically added to several standard downstream processing steps, such as chromatographic purification, virus inactivation, cation exchange chromatography, virus filtration, and / or ultrafiltration / percolation. Surprisingly, specific redox refolding conditions have been found to effectively reduce the aggregation of bispecific antibody constructs and increase the yield of their monomeric products. These conditions include, for example, a concentration of guanidine, also referred to herein as a protonated form of guanidine salt, at a concentration higher than that expected in the literature, and low concentrations of redox agents, preferably oxidants such as copper(II), and / or low concentrations of redox reagents (such as cysteine / cysteine). In some embodiments, oxidants are preferred.

[0075] Typically, high percentages, such as exceeding 25% or even 50% or 70%, of bispecific antibody constructs, such as the single-chain Fc-conjugated BiTE® antibody construct (ScFc-BiTE®) produced by industrial mammalian CHO cells (expressed, secreted, collected from a bioreactor, and purified via a protein A column), are adversely obtained downstream as aggregates, which significantly reduces yield and increases production costs. These aggregates are typically removed by a subsequent downstream purification step (typically cation exchange chromatography (CEX)). However, in methods for producing bispecific antibody constructs (e.g., ScFc-BiTE®) from mammalian cells (or microbial systems), redox refolding is used as an additional downstream processing step, which includes a redox refolding step as described herein, increasing the yield of the desired monomeric protein product (bispecific antibody construct) and reducing aggregates. Numerous refolding conditions, with or without reducing / oxidizing (redox) agents, have been evaluated for bispecific antibody constructs.

[0076] Preferably, the text describes a downstream method for processing an antibody construct generated upstream, preferably a bispecific antibody construct (e.g., the ScFc-BiTE® antibody construct), comprising contacting a protein already produced by a mammalian cell or microbial system with a clinker and a redox agent at a preferred pH, said pH depending on the pI of the bispecific antibody construct. In one embodiment, the preferred pH depends on the pI of the target domain of the bispecific antibody construct and is typically in the pH range of 5 to 8. Such pH values ​​correspond to typical pH values ​​in the pool or on-column conditions during the downstream processing of the bispecific antibody construct. Several clinker (denaturing, unfolding) agents, including guanidine, arginine, and elevated temperatures, were evaluated. Guanidine and arginine are preferred as clinker agents, with guanidine being the most preferred. For example, exemplary guanidine concentrations evaluated were in the range of 0.5 to 2.5, where a significant increase in monomer content was observed at concentrations of clinker above 1 M, preferably in the range of 1.2 to 2 M. Higher clinker concentrations typically carry an increased risk of product aggregation. Conversely, amounts below 1 M typically do not show significant refolding effects.

[0077] According to the invention, ionizing agents, such as guanidine salts, especially when combined with redox agents, such as copper II, surprisingly reduce the occurrence of downstream bispecific construct aggregation, and preferably further reduce the occurrence of isotype and / or heterotype antibody constructs. This is typically demonstrated by smaller shoulders in size exclusion chromatography, where shoulders should be understood as deviations from peaks with a typically Gaussian curve-like shape, for example by partial broadening or tailing of the peak. Accordingly, according to the invention, the occurrence of non-aggregated, i.e., monomeric bispecific antibodies is significantly increased in downstream processing leading to a more efficient overall manufacturing method of the antibody constructs according to the invention. Typically, the monomer content representing the desired product fraction is increased by at least 25%, preferably at least 40, 50, 60, 70, 80, 90, 100%, or even greater than 100%, relative to the monomer percentage before the application of the invention. For example, where the monomer content (monomer peaks in each size exclusion chromatography) increases from about 25 to about 50%, an increase of about 100% in monomer content has been achieved relative to the monomer content before the application of the invention. The actual final monomer content achievable as a percentage can vary depending on the specific molecule. However, for antibody constructs typically described herein, a percentage increase of at least 25%, and typically at least about 50%, of monomer content is typically observed. In this regard, to increase the percentage of monomer, it is preferable to add a redox agent along with a dissociation agent (e.g., guanidine) to a refolding buffer. When cysteine ​​and cysteine ​​form a redox agent in a refolding buffer having a total redox agent concentration in the range of at least 1 M, for example 1 to 2.5 M, preferably 1.2 to 2 M, for example at molar ratios of 6:1, 14:1, and 33:1 (Cys:Cyss), the percentage of monomer can be increased from 39% (without redox agent) to 52%, 61%, and 69%, respectively, i.e., an increase of up to about 76%. When used with a redox agent, such as an oxidizing agent, like copper(II) sulfate or dehydroascorbate, such as 0.5 mM copper(II) sulfate, for example, 1.2 M to 2.0 M guanidine, a surprising increase in monomer content with a high degree of refolding was observed, up to about 90%. The percentage of monomer was also increased with dehydroascorbate, for example, at concentrations ranging from 0.1 M to 1 M, preferably 0.5% to up to about 73%, accompanied by a clinker ranging from about 1.2 M to 2 M, such as about 1.2 M to 2.0 M guanidine. However, the refolding procedure described herein is also suitable for earlier downstream purification, for example, for the scFc-BiTE® CEX post-peak pool. Such earlier downstream product purification for efficient purification could include a protein A pool, for example, where a protein A FVIP sample contains approximately 73% monomer and 27% aggregated antibody constructs.If the refolding according to the invention is performed, for example with 2.0 M guanidine, the percentage of monomer can be increased to up to about 94% at a pH range of about pH 5.0 to pH 8.0.

[0078] In the context of this invention, a defined pH range has been surprisingly achieved, in which integrating the refolding step into downstream processing methods is preferably effective. The pH range depends on the pI of the first binding domain of the (tumor) target and / or the pI of the entire antibody construct. The first domain of the antibody construct as described herein typically has a pI of at least 4.9. For example, regarding the first domain, i.e., the target-binding domain, the CD33xCD3 antibody construct as described herein has a pI of about 4.9. Other pI values ​​are typically about 6.0 for the CD19xCD3 antibody construct as described herein, about 6.4 for the BCMAxCD3 antibody construct as described herein, about 8.7 for the CD19xCD3 antibody construct as described herein, about 9.4 for the MSLNxCD3 antibody construct as described herein, and about 9.2 for the FLT3xCD3 antibody construct as described herein. Therefore, the pH of the refolding buffer should at least correspond to the typically lowest pI of the first binding domain of the antibody construct of this invention, which is typically, but not necessarily, 4.9. Therefore, for example, in the context of this invention, a refolding buffer of about pH 5 is generally suitable.

[0079] In the context of this invention, the pI of the second binding domain, namely the CD3ε binding domain, of the antibody construct according to the invention typically has a higher pI than that of the target binding domain, for example, at least 9 pI, typically 9.3. In the presence of the third domain according to the invention, the pI is typically at least 7, typically 7.2. Therefore, the pI of the entire antibody construct will be higher than that of the first binding domain. Typically, the pI of the entire antibody construct is about 7.0 to 9.0, i.e., in the neutral range. Therefore, in the context of this invention, a neutral pH, such as pH 7.0 to 9.0, or for example 8.0, is preferred, where the refolding buffer or environment in which refolding occurs. For example, the CD33xCD3 antibody construct as described herein has a pI of about 7.2. For the CD19xCD3 antibody construct as described herein, other pI values ​​are typically around 8.0, around 8.0 for the BCMAxCD3 antibody construct as described herein, around 8.6 for the CD19xCD3 antibody construct as described herein, around 8.8 for the MSLNxCD3 antibody construct as described herein, and around 8.8 for the FLT3xCD3 antibody construct as described herein.

[0080] In this invention, a "pool" is understood as a downstream processing fluid containing the processed product (i.e., a bispecific antibody construct) and not associated with a matrix (e.g., a column). According to the method of the invention, a refolding buffer can be applied to several different pools or columns. Pools may include, for example, a filtered virus inactivation pool (FVIP) after protein A purification, or a high molecular weight (HMW) pool after CEX via buffer exchange, or consistent with the filtered virus inactivation step.

[0081] The concentration range of the bispecific antibody construct can be from 0.5 to 10 mg / ml, preferably close to 1 mg / ml. The duration can be several hours, for example from 2 to 48 hours, preferably up to 24 hours, more preferably about 10, 9, 8, 7, 6, 5, 4, 3 or 2 hours, for example 4 hours.

[0082] According to the method of the invention, the temperature range can be from 4ºC to 40ºC, for example from 10ºC to 30ºC, preferably from about 20ºC to about 25ºC, for example, room temperature of about 25ºC. Such a temperature selection avoids uncontrolled and undesirable aggregation while ensuring a suitable reaction time.

[0083] In the context of this invention, several oxidizing and redox agents can be used. The refolding buffer preferably uses an "oxidizing agent" comprising copper (i.e., copper(II)), dehydroascorbic acid, and dissolved oxygen. A preferred concentration range is 0.1 to 10 mM, preferably 0.5 to 1 mM, which is higher than the concentration range taught in the art regarding the closure of open disulfide bonds. Lower oxidizing agent concentrations (e.g., copper(II)) typically increase the time required for oxidation, i.e., the rebuilding of disulfide bonds. Higher oxidizing agent concentrations (e.g., copper(II)) may adversely lead to the oxidation and truncation of methionine residues. If the oxidation and truncation of methionine is in a functionally critical protein region, it should be minimized. Generally, the oxidizing agent (e.g., copper(II)) concentration according to the invention is chosen as a trade-off between residual open disulfide bonds and truncation. The copper concentration should also be determined by protein concentration. Previously, lower copper concentrations have been described in the art, for example, 0.005 to 0.1 mM copper sulfate in Harris et al. 2005, and about 0.01 mM in Treuheit et al., US6808902, and US7723490. However, the present invention relates to a method comprising 0.1 mM to 10 mM of an oxidant, preferably 0.5 mM to 1 mM, such as copper(II) (preferably copper(II) sulfate), which results in a faster reaction for rebuilding disulfide bonds by using the ionizing agent according to the invention. Unwilling to be bound by theory, without using an ionizing agent, only the surface-exposed open area, i.e., the reduced disulfide, can be accessed by an oxidant or redox agent to close (i.e. oxidize) the disulfide to restore bonds and tertiary structure, thereby advantageously reducing the amount of aggregates.

[0084] Several "redox agents" can be used, either as an alternative or in lieu of other agents, including the following pairs: cysteine / cystine, cysteine / cystamine, and reduced glutathione / oxidized glutathione. Cysteine ​​or reduced glutathione can also be used alone, as they will produce their oxidized forms, for example, at near-neutral pH. Surprisingly, the use of cysteine ​​and cystine has been found to significantly reduce aggregates. However, when using cysteine ​​and cystine, cystosineization of both cysteine ​​residues (one open disulfide bond) has been observed. Preferably, the cystine concentration is about 1 mM, and the cysteine ​​concentration is n times higher than the cysteine ​​concentration. For example, the cystine concentration can be in the range of 0.1 mM to 10 mM, preferably 1 mM. The cysteine:cystine ratio can be any value from 1:1 to 33:1, preferably about 20:1. Cystamine can be used instead of cystine at comparable concentrations and ratios. Reduced and oxidized glutathione can also be used at similar concentrations and ratios.

[0085] Adding an oxidizing agent according to the invention, such as copper or dehydroascorbic acid, along with a dissociation agent advantageously reduces the amount of undesirable product aggregates. Typically, significant reductions in aggregation can be achieved, for example, by refolding, from 26% to 6% of the CD33xCD3 bispecific antibody construct in the Protein A FVIP pool. Preferred conditions may include, for example, contacting the aggregates with a refolding buffer (acetate buffer) at pH 5.1 and 2.0 M guanidine at room temperature for 4 hours. Significant reductions in aggregation can also be achieved in the CEX HMW pool, for example by applying a refolding buffer at pH 8.0 containing 2.0 M guanidine and 0.5 mM copper(II), or, for example, a refolding buffer at pH 8.0 containing 2.0 M guanidine and 0.5 mM dehydroascorbic acid.

[0086] The present invention also relates to refolding buffers that can be advantageously used in the methods according to the invention.

[0087] The refolding buffer according to the invention is also suitable for refolding on downstream chromatographic columns (e.g., protein A columns). Generally, the redox refolding method described herein can be carried out in a container, on a column, in cell culture medium after harvest, or in cell culture medium in a bioreactor before harvest by adding the aforementioned redox and ionizing agents.

[0088] Bispecific antibody constructs according to the present invention, such as ScFc-BiTE® molecules, typically contain an Fc region whose size, pI, and hydrophobicity are similar to the corresponding region of IgG antibodies. General downstream processing of IgG antibodies from host cell proteins (HCPs), other bioreactor impurities and reagents, high molecular weight (HMW) substances or aggregates, low molecular weight substances, or shear products is known in the art (Shukla et al. 2006). Such downstream processing has been adapted to purify bispecific antibody constructs, such as scFc-BITE®, which typically includes the following steps: cell culture harvesting; protein A chromatography; filtration for virus inactivation; a second column chromatography purification step or both, such as CEX; virus filtration; UF / DF. Bispecific antibody construct materials obtained upstream (e.g., in continuous manufacturing processes) are typically obtained through application... Figure 1Purification is performed as shown in the steps described. The method for achieving redox refolding according to the invention can be carried out in any of the harvesting and purification steps listed above, typically containing a large percentage of undesirable aggregates. For example, redox refolding can be performed in a vessel after protein A chromatography, which purifies the host cell protein and reagents but still contains aggregates. Denaturing reagents can be added to the protein A pool, or the protein A pool can be buffer-exchanged into a denaturing-refolding buffer. Refolding on a protein A column can also be performed by washing and incubating proteins immobilized on the column with a ionizing agent (e.g., guanidine, arginine, urea). Oxidizing agents (e.g., copper or dehydroascorbic acid) or redox reagents can also be added to perform on-column refolding.

[0089] In the context of this invention, "chaotrope" or "chaotrope agent" should be understood as a denaturing or unfolding agent, such as guanidine, arginine, urea, extreme pH, organic solvents such as alcohols, and / or detergents (e.g., SDS), which disrupt the hydrogen bond network and van der Waals interactions of macromolecules (e.g., bispecific antibody constructs). This increases the entropy of the macromolecule, which is associated with an increase in the kinetic energy or motion of the macromolecule. At the low concentrations preferably applied according to this invention, the chaotrope agent can partially or transiently disrupt the tertiary external structure of proteins. For full-length monoclonal antibodies (mAbs), the effects of chaotrope agents at concentrations of 0 M to 2 M guanidine hydrochloride (GdnHCl) were investigated prior to 2014 using various spectroscopic techniques, size exclusion chromatography (SEC), and analytical ultracentrifugation. Moderate perturbation of the mAb tertiary structure was detected on day 0. On day 1 (after 24 hours at 37ºC), major perturbations of the tertiary structure were observed at clinker concentrations above 1.2 M, particularly above 1.8 M. This led to an increase in high molecular weight (HMW) material after 6 and 11 days of incubation. Above 1.2 M guanidine, mAb aggregation was detected only in solutions where partial unfolding was induced. Mehta et al. (2014) noted, “Differential scanning calorimetry studies showed that the CH2 domain of the antibody unfolded in antibody molecules incubated at 1.2 M and higher concentrations of GuHCl. These results indicate that the unfolding of the CH2 domain leads to aggregation.”

[0090] Based on the teachings of the art, those skilled in the art would be reluctant to use a dissociative agent with a concentration higher than 1 M, such as 1.2 M guanidine, to treat antibodies or bispecific antibody constructs when aggregation should be prevented. Therefore, surprisingly, in the method of the present invention, the bispecific antibody of the present invention is treated with a dissociative agent concentration higher than 1 M for refolding purposes, and a non-aggregated product is still obtained. However, the bispecific antibody construct of the present invention contains at least one additional disulfide bond in each of the three binding domains, i.e., at least one in the target binding domain, CD3 binding domain, or scFc binding domain. Preferably, an additional disulfide bond is present in the CH2 of the third domain, which makes the bispecific antibody construct more stable, i.e., more stable than the CH2 of glycosylated mAb. As a supporting feature, the method of the present invention preferably applies milder reaction conditions, i.e., reaction times shorter than 24 hours, e.g., 4 hours, and temperatures below 37ºC, e.g., 25ºC. Not wishing to be bound by theory, but based on the specificity claimed herein, applying the method according to the present invention does not lead to major perturbations of the tertiary structure, which would ultimately result in aggregation. Instead, the bispecific antibody constructs aggregated by the method of the present invention allow oxidants and / or redox agents to reach the “open” disulfide bonds and “close” them again through partial and transitional unfolding of the ionizing agent.

[0091] In the context of this invention, "closing" refers to the reconstruction of the covalent bond of the -SS- disulfide bond. Therefore, the tertiary structure of the bispecific antibody construct is reconstructed and the previous aggregation is eliminated or at least reduced. As a result, the product yield of downstream processing using the method of this invention is improved.

[0092] In the context of this invention, "cell culture" or "culture" refers to the growth and reproduction of cells outside of multicellular organisms or tissues. Suitable culture conditions for mammalian cells are known in the art. See, for example, *Animal Cell Culture: A Practical Approach*, edited by D. Rickwood, Oxford University Press, New York (1992). Mammalian cells can be cultured in suspension or attached to a solid substrate.

[0093] The term "mammalian cell" refers to any cell that is derived from or originates from any mammal (e.g., human, hamster, mouse, green monkey, rat, pig, cow, or rabbit). For example, a mammalian cell can be an immortalized cell. In some embodiments, a mammalian cell is a differentiated cell. In some embodiments, a mammalian cell is an undifferentiated cell. Non-limiting examples of mammalian cells are described herein. Further examples of mammalian cells are known in the art.

[0094] As used herein, the term “cell culturing medium” (also known as “culture medium,” “cell culture media,” or “tissue culture medium”) refers to any nutrient solution used to grow cells (e.g., animal or mammalian cells) and typically provides at least one or more of the following components: energy (usually in the form of carbohydrates, such as glucose); one or more of all essential amino acids, typically twenty basic amino acids plus cysteine; vitamins and / or other organic compounds typically required in low concentrations; lipids or free fatty acids; and trace elements, such as inorganic compounds or naturally occurring elements, typically required in very low concentrations (usually in the micromolar range).

[0095] Cell culture media include those that are typically used and / or known to be used in any cell culture method, such as, but not limited to, batch, extended batch, fed-batch, and / or perfusion or continuous cell culture.

[0096] “Growth” cell culture medium or fed medium refers to a cell culture medium typically used during the exponential growth phase (“growth phase”) and sufficiently intact to support cell culture during this phase. Growth cell culture media may also contain selectants that confer selective marker resistance or viability to host cell lines. Such selectants include, but are not limited to, genimycin (G4118), neomycin, hygromycin B, puromycin, bleomycin, methionine sulfinimide, methotrexate, glutamine-free cell culture media, glycine-deficient cell culture media, hypoxanthine and thymidine, or thymidine alone.

[0097] "Production" cell culture medium, or fed culture medium, refers to cell culture typically used during the transition period at the end of exponential growth and during subsequent transition and / or production phases when protein production takes over. Such cell culture media are sufficiently intact to maintain the required cell density, viability, and / or product titers during this phase.

[0098] "Perfusion" cell culture medium, or fed culture medium, is a cell culture medium typically used to maintain cell culture via perfusion or continuous culture methods, and is sufficiently intact to support cell culture during this process. Perfusion cell culture medium formulations can be richer or more concentrated than basal cell culture medium formulations to suit methods for removing used medium. Perfusion cell culture medium can be used during both the growth and production phases.

[0099] The term "culture" or "cell culture" refers to the maintenance or proliferation of mammalian cells under a set of controlled physical conditions.

[0100] The term "mammalian cell culture" refers to a liquid culture medium containing multiple mammalian cells that is maintained or proliferated under a set of controlled physical conditions.

[0101] The term "liquid culture medium" refers to a fluid containing sufficient nutrients to allow cells (e.g., mammalian cells) to grow or proliferate in vitro. For example, a liquid culture medium may contain one or more of the following: amino acids (e.g., 20 amino acids), purines (e.g., hypoxanthine), pyrimidines (e.g., thymidine), choline, inositol, thiamine, folic acid, biotin, calcium, nicotinamide, pyridoxine, riboflavin, thymine, cyanocobalamin, pyruvate, lipoic acid, magnesium, glucose, sodium, potassium, iron, copper, zinc, and sodium bicarbonate. In some embodiments, the liquid culture medium may contain serum from mammals. In some embodiments, the liquid culture medium does not contain serum from mammals or another extract (defined liquid culture medium). In some embodiments, the liquid culture medium may contain trace metals, mammalian growth hormones, and / or mammalian growth factors. Another example of a liquid culture medium is a basic culture medium (e.g., a medium containing only inorganic salts, a carbon source, and water). Non-limiting examples of liquid culture media are described herein. Further examples of liquid culture media are known in the art and are commercially available. Liquid culture media may contain mammalian cells at any density. For example, as used in this paper, a certain volume of liquid culture medium removed from a bioreactor may be substantially free of mammalian cells.

[0102] The second liquid culture medium can be the same as the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is a concentrated form of the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is added as a dry powder.

[0103] The term "antibody product" refers to a "secreted protein" or "secreted recombinant protein," and specifically to a protein (e.g., a recombinant protein) that initially contains at least one secretion signal sequence during translation in mammalian cells and is at least partially secreted into the extracellular space (e.g., liquid culture medium) by enzymatic cleavage of the secretion signal sequence in mammalian cells. Skilled practitioners will understand that a "secreted" protein does not need to be completely dissociated from the cell to be considered a secreted protein.

[0104] The term bispecific antibody product encompasses bispecific antibodies, such as full-length IgG-based antibodies and fragments thereof, which are typically referred to herein as bispecific antibody constructs.

[0105] The term "antibody construct" refers to a molecule whose structure and / or function are based on the structure and / or function of an antibody (e.g., a full-length or intact immunoglobulin molecule) and / or extracted from the variable heavy chain (VH) and / or variable light chain (VL) domains of an antibody or fragment thereof. Therefore, the antibody construct is capable of binding to its specific target or antigen. Furthermore, the binding domains of the antibody construct according to the invention contain minimum structural requirements for antibody binding to the target. These minimum requirements can be defined, for example, by the presence of at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region), preferably all six CDRs. Alternative methods for defining the minimum structural requirements of an antibody include defining the antibody epitope within the specific target structure, the protein domain of the target protein constituting the epitope region (epitope cluster), or by referencing a specific antibody competing with the defined antibody epitope. The antibodies upon which the constructs according to the invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies.

[0106] The binding domain of the antibody construct according to the invention may, for example, include the CDR group mentioned above. Preferably, those CDRs are contained within the framework of the antibody light chain variable region (VL) and the antibody heavy chain variable region (VH); however, it does not necessarily include both. For example, the Fd fragment has two VH regions and typically retains some antigen-binding functionality of the intact antigen-binding domain. Other examples of antibody fragments, antibody variants, or binding domains include (1) Fab fragments, a monovalent fragment having VL, VH, CL, and CH1 domains; (2) F(ab')2 fragments, a bivalent fragment having two Fab fragments connected in the hinge region by disulfide bridges; (3) Fd fragments having two VH and CH1 domains; (4) Fv fragments having VL and VH domains in an antibody single arm; (5) dAb fragments having a VH domain (Ward et al., (1989) Nature [Nature] 341:544-546); (6) separated complementarity-determining regions (CDRs); and (7) single-chain Fv (scFv), the latter being preferred (e.g., derived from scFV libraries). Examples of embodiments of the antibody constructs according to the present invention are described, for example, in the following: WO 00 / 006605, WO 2005 / 040220, WO 2008 / 119567, WO 2010 / 037838, WO 2013 / 026837, WO 2013 / 026833, US2014 / 0308285, US 2014 / 0302037, WO 2014 / 144722, WO 2014 / 151910 and WO 2015 / 048272.

[0107] Additionally, the term "binding domain" or "bound domain" refers to fragments of full-length antibodies, such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab')2, or "r IgG" ("half antibody"). Antibody constructs according to the invention may also contain modified fragments of the antibody (also called antibody variants), such as scFv, di-scFv or di(s)-scFv, scFv-Fc, scFv-zipper, scFab, Fab2, Fab3, biantibodies, single-chain biantibodies, tandem biantibodies (Tandabs), tandem di-scFv, tandem tri-scFv, "multi-antibodies" (such as tri- or tetra-antibodies), and single-domain antibodies such as nanobodies or monovariable domain antibodies containing only one variable domain, which may be VHH, VH, or VL that specifically binds to antigens or epitopes independently of other V regions or domains.

[0108] As used herein, the terms “single-chain Fv,” “single-chain antibody,” or “scFv” refer to single-peptide chain antibody fragments that contain variable regions from both the heavy and light chains but lack constant regions. Generally, single-chain antibodies further include a polypeptide linker between the VH and VL domains, which allows them to form the desired structure that will permit antigen binding. Single-chain antibodies are discussed in detail in: Pluckthun, *The Pharmacology of Monoclonal Antibodies*, Vol. 113, edited by Rosenburg and Moore, Springer-Verlag, New York, pp. 269–315 (1994). Various methods for generating single-chain antibodies are known, including those described in the following: U.S. Patent Nos. 4,694,778 and 5,260,203; International Patent Application Publication No. WO 88 / 01649; Bird (1988) Science 242:423-442; Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883; Ward et al. (1989) Nature 334:54454; Skerra et al. (1988) Science 242:1038-1041. In specific embodiments, the single-chain antibody may also be bispecific, multispecific, human and / or humanized, and / or synthetic.

[0109] Furthermore, the definition of the term "antibody construct" includes monovalent, divalent, and polyvalent / multivalent constructs, and therefore includes bispecific constructs that specifically bind to only two antigen structures, and multispecific constructs that specifically bind to more than two (e.g., three, four, or more) antigen structures through different binding domains. Additionally, the definition of the term "antibody construct" includes molecules consisting of only one polypeptide chain and molecules consisting of more than one polypeptide chain, which can be identical (homodimer, homotrimer, or homooligomer) or different (heterodimer, heterotrimer, or heterooligomer). Examples of the antibodies and their variants or derivatives identified above are particularly described in the following: Harlow and Lane, Antibodies alaboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999); Kontermann and Dübel, Antibody Engineering, Springer, 2nd edition 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

[0110] As used herein, the term "bispecific" refers to an antibody construct that is "at least bispecific," meaning it contains at least a first binding domain and a second binding domain, wherein the first binding domain binds to one antigen or target (e.g., a target cell surface antigen), and the second binding domain binds to another antigen or target (e.g., CD3). Therefore, antibody constructs according to the invention contain specificity against at least two different antigens or targets. For example, the first domain preferably does not bind to the extracellular epitopes of one or more of the species described herein. The term "target cell surface antigen" refers to an antigenic structure expressed by a cell and present on the cell surface such that it is accessible to antibody constructs as described herein. It can be a protein, preferably the extracellular portion of a protein, or a carbohydrate structure, preferably the carbohydrate structure of a protein, such as a glycoprotein. It is preferably a tumor antigen. The term "bispecific antibody construct" of the invention also encompasses multispecific antibody constructs, such as trispecific antibody constructs, which include three binding domains, or constructs having more than three (e.g., four, five…) specificities.

[0111] Given that the antibody constructs according to the invention are (at least) bispecific, they are not naturally occurring and are distinctly different from naturally occurring products. Therefore, a “bispecific” antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin having at least two distinct binding ends with different specificities. Bispecific antibody constructs can be produced by a variety of methods, including hybridoma fusion or Fab' fragment linkage. See, for example, Songsivilai and Lachmann, Clin. Exp. Immunol. [Clinical Experimental Immunology] 79:315-321 (1990).

[0112] The antibody constructs of the present invention may or may not contain peptide linkers (spacer peptides) at least two binding domains and variable domains (VH / VL). According to the present invention, the term "peptide linker" comprises an amino acid sequence through which the amino acid sequences of one (variable and / or binding) domain and another (variable and / or binding) domain of the antibody construct of the present invention are linked to each other. Peptide linkers can also be used to fuse a third domain with other domains of the antibody construct of the present invention. A fundamental technical feature of such peptide linkers is that they do not contain any polymerization activity. Suitable peptide linkers are those described in U.S. Patents 4,751,180 and 4,935,233 or WO 88 / 09344. Peptide linkers can also be used to attach other domains or modules or regions (such as half-life extension domains) to the antibody constructs of the present invention.

[0113] The antibody construct of the present invention is preferably an "in vitro generated antibody construct." This term refers to an antibody construct as defined above, wherein all or a portion of the variable region (e.g., at least one CDR) is generated in a non-immune cell selection process, such as in vitro phage display, protein microarray, or any other method in which the ability of a candidate sequence to bind to an antigen can be tested. Therefore, this term preferably excludes sequences generated solely by genomic rearrangements in animal immune cells. A "recombinant antibody" is an antibody produced using recombinant DNA technology or genetic engineering.

[0114] As used herein, the term "monoclonal antibody" (mAb) or monoclonal antibody construct refers to an antibody derived from a substantially homogeneous population of antibodies, meaning that individual antibodies comprising that population are identical except for possibly small amounts of naturally occurring mutations and / or post-translational modifications (e.g., isomerization, amidation). Monoclonal antibodies exhibit high specificity against a single antigenic side or determinant on an antigen, compared to conventional (polyclonal) antibody formulations that typically comprise different antibodies targeting different determinants (or epitopes). In addition to their specificity, monoclonal antibodies also have the advantage of being synthesized via hybridoma culture and thus free from contamination by other immunoglobulins. The modifier "monoclonal" indicates the characteristic of antibodies derived from a substantially homogeneous population of antibodies and should not be construed as requiring the antibody to be produced by any particular method.

[0115] For the preparation of monoclonal antibodies, any technique that provides antibodies produced from continuous cell line cultures can be used. For example, the monoclonal antibody to be used can be prepared by the hybridoma method first described in Koehler et al., Nature, 256: 495 (1975), or by a recombinant DNA method (see, for example, U.S. Patent No. 4,816,567). Examples of other techniques for producing human monoclonal antibodies include the three-source hybridoma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss (1985), 77-96).

[0116] Hybridomas can then be screened using standard methods such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (BIACORE™) analysis to identify one or more hybridomas that produce antibodies that specifically bind to a specified antigen. Any form of the relevant antigen can be used as an immunogen, such as recombinant antigens, naturally occurring forms, any variants or fragments thereof, and their antigenic peptides. Surface plasmon resonance, as used in the BIAcore system, can be used to increase the efficiency of phage antibodies binding to epitopes of target cell surface antigens (Schier, Human Antibodies Hybridomas 7 (1996), 97-105; Malmborg, J. Immunol. Methods 183 (1995), 7-13).

[0117] Another exemplary method for preparing monoclonal antibodies includes screening protein expression libraries, such as phage display or ribosome display libraries. Phage display is described, for example, in: Ladner et al., U.S. Patent No. 5,223,409; Smith (1985) Science 228:1315-1317; Clackson et al., Nature 352:624-628 (1991); and Marks et al., J. Mol. Biol. 222: 581-597 (1991).

[0118] In addition to using display libraries, relevant antigens can be used to immunize non-human animals, such as rodents (e.g., mice, hamsters, rabbits, or rats). In one embodiment, the non-human animal includes at least a portion of the human immunoglobulin gene. For example, large fragments of human Ig (immunoglobulin) gene loci may be used to engineer mouse strains with antibody production defects. Using hybridoma technology, antigen-specific monoclonal antibodies derived from genes having the desired specificity can be generated and selected. See, for example, XENOMOUSE™, Green et al. (1994) Nature Genetics 7:13-21, US 2003-0070185, WO 96 / 34096, and WO 96 / 33735.

[0119] Monoclonal antibodies can also be derived from non-human animals and then modified using recombinant DNA techniques known in the art, such as humanization, deimmunization, and chimerism. Examples of modified antibody constructs include humanized variants of non-human antibodies, “affinity-matured” antibodies (see, for example, Hawkins et al. J. Mol. Biol. [Journal of Molecular Biology] 254, 889-896 (1992) and Lowman et al., Biochemistry [Biochemistry] 30, 10832-10837 (1991)) and antibody mutants with altered functions of one or more effectors (see, for example, U.S. Patent 5,648,260, Kontermann and Dübel (2010), cited above, and Little (2009), cited above).

[0120] In immunology, affinity maturation is a process through which B cells produce antibodies with increased affinity for antigens during an immune response. With repeated exposure to the same antigen, the host produces antibodies with progressively greater affinity. Like the natural prototype, in vitro affinity maturation is based on the principles of mutation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random mutations are introduced into the CDR using radiation, chemical mutagens, or error-prone PCR. Furthermore, genetic diversity can be increased through strand shuffling. Two or three rounds of mutation and selection using display methods (such as phage display) typically yield antibody fragments with affinity in the low nanomolar range.

[0121] Preferred types of amino acid substitution changes in antibody constructs involve substituting one or more hypervariable residues of the parent antibody (e.g., humanized or human antibody). Generally, one or more resulting variants selected for further development will have improved biological properties relative to the parent antibody that produced them. A convenient method for generating such substituted variants involves affinity maturation using phage display. Briefly, several hypervariable side ends (e.g., 6-7 side ends) are mutated to produce all possible amino acid substitutions at each side end. The resulting antibody variants are displayed monovalently from filamentous phage particles as fusions with the gene III product of M13 packaged within each particle. The biological activity (e.g., binding affinity) of the phage-displayed variants is then screened as disclosed herein. To identify candidate hypervariable side ends for modification, alanine scanning mutagenesis can be performed to identify hypervariable residues that significantly contribute to antigen binding. Alternatively or additionally, analyzing the crystal structure of the antigen-antibody complex to identify contact points between the binding domain and, for example, human target cell surface antigens may be advantageous. Such contact residues and adjacent residues are candidates for substitution according to the techniques described herein. Once such variants are generated, they are screened as described herein, and antibodies that exhibit superior properties in one or more relevant assays can be selected for further development.

[0122] The monoclonal antibodies and antibody constructs of the present invention specifically include “chimeric” antibodies (immunoglobulins) wherein a portion of the heavy chain and / or light chain is identical or homologous to a corresponding sequence in an antibody derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of one or more chains is identical or homologous to a corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass, and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences], 81: 6851-6855 (1984)). Chimeric antibodies of interest herein include “primate-derived” antibodies, which contain a variable domain antigen-binding sequence derived from non-human primates (e.g., Old World monkeys, apes, etc.) and a human constant region sequence. Various methods for preparing chimeric antibodies have been described. See, for example, Morrison et al., Proc. Natl. Acad. Sci USA, 81:6851, 1985; Takeda et al., Nature, 314:452, 1985; Cabilly et al., U.S. Patent No. 4,816,567; Boss et al., U.S. Patent No. 4,816,397; Tanaguchi et al., EP 0171496; EP 0173494; and GB 2177096.

[0123] Antibodies, antibody constructs, antibody fragments, or antibody variants can also be modified by specifically deleting human T-cell epitopes using methods disclosed, for example, in WO 98 / 52976 or WO 00 / 34317 (referred to as "deimmunization" methods). In short, the heavy and light chain variable domains of antibodies can be analyzed for peptides that bind to MHC class II; these peptides represent potential T-cell epitopes (as defined in WO 98 / 52976 and WO 00 / 34317). To detect potential T-cell epitopes, a computer modeling method called "peptide threading" can be applied, and a database of human MHC class II binding peptides can be searched for motifs present in the VH and VL sequences, as described in WO 98 / 52976 and WO 00 / 34317. These motifs bind to any of the 18 major MHC class II DR allotypes and thus constitute potential T-cell epitopes. Detected potential T-cell epitopes can be eliminated by substituting a few amino acid residues in the variable domain, or preferably by a single amino acid substitution. Typically, conserved substitutions are performed. Generally, but not exclusively, amino acids shared with positions in human germline antibody sequences may be used. Human germline sequences are disclosed, for example, in: Tomlinson et al. (1992) J. MoI. Biol. [Journal of Molecular Biology] 227:776-798; Cook, GP et al. (1995) Immunol. Today [Contemporary Immunology] Vol. 16 (5):237-242; and Tomlinson et al. (1995) EMBO J. [Journal of the European Society for Molecular Biology] 14:14:4628-4638. The V BASE catalog provides a comprehensive catalog of human immunoglobulin variable region sequences (edited by Tomlinson, LA. et al. MRC Centre for Protein Engineering [Medical Research Council Centre for Protein Engineering], Cambridge, UK). These sequences can be used as sources of human sequences, for example, for the frame region and CDR. A shared human frame area may also be used, such as that described in U.S. Patent No. 6,300,064.

[0124] "Humanized" antibodies, antibody constructs, variants, or fragments thereof (such as Fv, Fab, Fab', F(ab')2, or other antigen-binding sequences of antibodies) are antibodies or immunoglobulins with a predominantly human sequence containing one or more minimal sequences derived from non-human immunoglobulins. For the most part, humanized antibodies are human immunoglobulins (receptor antibodies) in which residues from the hypervariable region (also known as the CDR) of the receptor are replaced by residues from the hypervariable region (donor antibody) of a non-human (e.g., rodent) species (such as mouse, rat, hamster, or rabbit) with the desired specificity, affinity, and ability. In some cases, Fv framework region (FR) residues of human immunoglobulins are replaced by corresponding non-human residues. Furthermore, as used herein, "humanized antibodies" may also include residues not found in either the receptor antibody or the donor antibody. These modifications are made to further improve and optimize antibody performance. Humanized antibodies may also contain at least a portion of the immunoglobulin constant region (Fc) of a human immunoglobulin. For further details, see Jones et al., Nature, 321: 522-525 (1986); Reichmann et al., Nature, 332: 323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2: 593-596 (1992).

[0125] Humanized antibodies or fragments thereof can be generated by replacing the sequence of an Fv variable domain that does not directly participate in antigen binding with an equivalent sequence of a human Fv variable domain. Exemplary methods for generating humanized antibodies or fragments thereof are provided by: Morrison (1985) Science 229:1202-1207; Oi et al. (1986) BioTechniques 4:214; and US 5,585,089; US 5,693,761; US ​​5,693,762; US 5,859,205; and US 6,407,213. Those methods involve isolating, manipulating, and expressing nucleic acid sequences encoding all or part of the immunoglobulin Fv variable domain from at least one of the heavy or light chains. Such nucleic acids can be obtained from hybridomas that produce antibodies against a predetermined target as described above, as well as other sources. The recombinant DNA encoding the humanized antibody molecule can then be cloned into a suitable expression vector.

[0126] Humanized antibodies can also be produced using transgenic animals, such as mice expressing human heavy and light chain genes but not endogenous mouse immunoglobulin heavy and light chain genes. Winter describes an exemplary CDR transplantation method (US Patent No. 5,225,539) that can be used to prepare the humanized antibodies described herein. All CDRs of a particular human antibody can be replaced with at least a portion of non-human CDRs, or only some CDRs can be replaced with non-human CDRs. Only the number of CDRs required to bind the humanized antibody to the predetermined antigen needs to be replaced.

[0127] Humanized antibodies can be optimized by introducing conserved substitutions, shared sequence substitutions, germline substitutions, and / or reversion mutations. Such modified immunoglobulin molecules can be prepared using any of several techniques known in the art (e.g., Teng et al., Proc. Natl. Acad. Sci. USA, 80: 7308-7312, 1983; Kozbor et al., Immunology Today, 4: 7279, 1983; Olsson et al., Meth. Enzymol., 92: 3-16, 1982, and EP 239 400).

[0128] The terms “human antibody,” “human antibody construct,” and “human binding domain” include antibodies, antibody constructs, and binding domains having antibody regions that substantially correspond to variable and constant regions or domains of human immunoglobulin sequences known in the art, including, for example, those described by Kabat et al. (1991) (cited above). The human antibodies, antibody constructs, or binding domains of the present invention may contain, for example, amino acid residues not encoded by human immunoglobulin sequences in CDRs, particularly CDR3 (e.g., mutations introduced by random or flanking specific mutagenesis in vitro or by somatic mutations in vivo). Human antibodies, antibody constructs, or binding domains may have at least one, two, three, four, five, or more positions substituted with amino acid residues not encoded by human immunoglobulin sequences. However, the definitions of human antibodies, antibody constructs, and binding domains as used herein also encompass “fully human antibodies,” which contain only human antibody sequences that are not artificially and / or genetically altered, such as those derived using techniques or systems like Xenomouse. Preferably, the “fully human antibody” does not contain amino acid residues that are not encoded by human immunoglobulin sequences.

[0129] In some embodiments, the antibody constructs of the present invention are “isolated” or “substantially pure” antibody constructs. When used to describe the antibody constructs disclosed herein, “isolated” or “substantially pure” means that the antibody construct has been identified, isolated, and / or recovered from components of its production environment. Preferably, the antibody construct does not associate with, or substantially does not associate with, any other components from its production environment. Contaminating components of its production environment, such as those produced by recombinant transfected cells, are typically substances that interfere with the diagnostic or therapeutic use of the peptide and may include enzymes, hormones, and other protein or non-protein solutes. The antibody construct may, for example, comprise at least about 5% by weight or at least about 50% by weight of the total protein in a given sample. It should be understood that, depending on the circumstances, isolated protein may comprise from 5% by weight to 99.9% by weight of the total protein content. By using inducible promoters or overexpression promoters, peptides can be prepared at significantly higher concentrations, such that they are prepared at increased concentration levels. This definition includes the production of antibody constructs in a variety of organisms and / or host cells known in the art. In a preferred embodiment, the antibody construct is (1) purified to a degree sufficient to obtain at least 15 N-terminal or internal amino acid residues using a spin-cup sequencer, or (2) purified to homogenization by SDS-PAGE under non-reducing or reducing conditions using Coomassie blue or, preferably, silver staining. However, isolated antibody constructs are typically prepared by at least one purification step.

[0130] The term "binding domain" in this invention characterizes a domain that (specifically) binds to / interacts with / recognizes a given target epitope or a given target side terminal on a target molecule (antigen) (e.g., CD33 and CD3, respectively). The structure and function of the first binding domain (recognizing, for example, CD33) and preferably also the structure and / or function of a second binding domain (e.g., recognizing CD3) are based on the structure and / or function of an antibody, such as a full-length or intact immunoglobulin molecule, and / or extracted from the variable heavy chain (VH) and / or variable light chain (VL) domains of the antibody or a fragment thereof. Preferably, the first binding domain is characterized by the presence of three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). The second binding domain preferably also includes the minimum structural requirements of the antibody that allow target binding. More preferably, the second binding domain comprises at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). It is envisioned that the first binding domain and / or the second binding domain are generated or available through phage display or library screening methods, rather than by transplanting CDR sequences from a pre-existing (monoclonal) antibody into a scaffold.

[0131] According to the present invention, the binding domain is in the form of one or more polypeptides. Such polypeptides may include protein portions and non-protein portions (e.g., chemical linkers or chemical cross-linking agents, such as glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments, and peptides generally having fewer than 30 amino acids) comprise two or more amino acids coupled to each other via covalent peptide bonds (forming an amino acid chain).

[0132] As used herein, the term "peptide" describes a group of molecules that typically consist of more than 30 amino acids. Peptides can further form polymers, such as dimers, trimers, and higher oligomers, i.e., molecules composed of more than one polypeptide molecule. The polypeptide molecules forming such dimers, trimers, etc., can be identical or different. Thus, the corresponding higher-order structures of such polymers are called homodimers or heterodimers, homotrimers or heterotrimers, etc. An example of a heteropolymer is an antibody molecule in its native form consisting of two identical light-chain polypeptide chains and two identical heavy-chain polypeptide chains. The terms "peptide," "polypeptide," and "protein" also refer to naturally modified peptides / polypeptides / proteins, where such modification is achieved through, for example, post-translational modifications (such as glycosylation, acetylation, phosphorylation, etc.). When referred to herein, "peptide," "polypeptide," or "protein" can also be chemically modified, such as PEGylation. Such modifications are well known in the art and are described below.

[0133] Preferably, the binding domains for target cell surface antigens and / or the binding domains for CD3ε are human binding domains. Antibodies and antibody constructs containing at least one human binding domain avoid some of the problems associated with antibodies or antibody constructs having variable and / or constant regions from non-human animals (e.g., rodents, rats, hamsters, or rabbits). The presence of such rodent-derived proteins can lead to rapid clearance of the antibody or antibody construct or can induce an immune response against the antibody or antibody construct in the patient. To avoid using rodent-derived antibodies or antibody constructs, human or fully human antibodies / antibody constructs can be generated by introducing human antibody function into rodents to induce the production of fully human antibodies in rodents.

[0134] The ability to clone and recombinant megabase-sized human loci in YAC and introduce them into mouse lines provides a powerful method for elucidating the functional components of very large or coarsely located loci and for generating useful human disease models. Furthermore, replacing mouse loci with their human equivalents using this technique can provide unique insights into the expression and regulation of human gene products during development, their communication with other systems, and their involvement in disease induction and progression.

[0135] A key practical application of this strategy is the "humanization" of the mouse humoral immune system. Introducing human immunoglobulin (Ig) loci into mice where endogenous Ig genes are inactivated provides an opportunity to study the fundamental mechanisms of programmed antibody expression and assembly, as well as their role in B cell development. Furthermore, this strategy can provide an ideal source for the production of fully human monoclonal antibodies (mAbs)—a significant milestone contributing to the prospects of antibody therapy in human diseases. Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic responses inherent in mouse or mouse-derived mAbs, thereby increasing the efficacy and safety of administered antibody / antibody constructs. The use of fully human antibodies or antibody constructs can be expected to offer significant advantages in treating chronic and relapsing human diseases requiring repeated compound administration, such as inflammation, autoimmune diseases, and cancer.

[0136] One approach to achieving this goal is to engineer mouse strains deficient in antibody production using large fragments of human Ig loci. These mice are expected to generate a large library of human antibodies in the absence of mouse antibodies. Large human Ig fragments will maintain significant variable genetic diversity and appropriate regulation of antibody production and expression. By leveraging mouse mechanisms for antibody diversification and selection, and by exploiting the lack of immune tolerance to human proteins, the regenerated human antibody libraries in these mouse strains should produce high-affinity antibodies against any antigen of interest, including human antigens. Using hybridoma technology, antigen-specific human mAbs with desired specificity can be readily generated and selected. This general strategy is demonstrated by the generation of the first XenoMouse mouse strain (see Green et al., Nature Genetics, 7:13-21 (1994)). The XenoMouse strain was engineered with yeast artificial chromosomes (YACs) containing 245 kb and 190 kb germline fragments of human heavy chain and κ light chain loci, respectively, with core variable and constant region sequences. YACs containing human Ig have been shown to be compatible with mouse systems for rearranging and expressing antibodies, and to replace inactivated mouse Ig genes. This is demonstrated by their ability to induce B cell development, produce an adult-like human repertoire of fully human antibodies, and generate antigen-specific human mAbs. These results also show that introducing human Ig loci containing a greater number of V genes, additional regulatory elements, and a larger portion of the human Ig constant region can substantially reproduce a complete repertoire that characterizes the humoral response to infection and immunity. The work of Green et al. has recently been extended to introducing germline YAC fragments containing megabase-sized human heavy chain loci and κ light chain loci, respectively, to introduce human antibody repertoires greater than approximately 80%. See Mendez et al., Nature Genetics, 15:146-156 (1997) and U.S. Patent Application Serial No. 08 / 759,620.

[0137] The origin of the XenoMouse mouse is further discussed and described in the following: US Patent Application Serial Nos. 07 / 466,008, 07 / 610,515, 07 / 919,297, 07 / 922,649, 08 / 031,801, 08 / 112,848, 08 / 234,145, 08 / 376,279, 08 / 430,938, 08 / 464,584, 08 / 46 Serial numbers 4,582, 08 / 463,191, 08 / 462,837, 08 / 486,853, 08 / 486,857, 08 / 486,859, 08 / 462,513, 08 / 724,752, and 08 / 759,620; and U.S. Patent Nos. 6,162,963, 6,150,584, 6,114,598, 6,075,181, and 5,939,598; and Japanese Patent Nos. 3,068,180 B2, 3,068,506 B2, and 3,068,507 B2. See also Mendez et al., Nature Genetics, 15:146-156 (1997) and Green and Jakobovits, J. Exp. Med., 188:483-495 (1998), EP 0 463 151 B1, WO 94 / 02602, WO 96 / 34096, WO 98 / 24893, WO 00 / 76310 and WO 03 / 47336.

[0138] In an alternative approach, other companies, including GenPharm International, Inc., utilize the “microlobe” approach. In the microlobe approach, exogenous Ig loci are mimicked by incorporating fragments (individual genes) from Ig loci. Thus, one or more VH genes, one or more DH genes, one or more JH genes, a mu constant region, and a second constant region (preferably a γ constant region) are formed into a construct for insertion into animals. The method is described in the following: U.S. Patent Nos. 5,545,807 and 5,545,806; 5,625,825; 5,625,126; 5,633,425; 5,661,016; 5,770,429; 5,789,650; 5,814,318; 5,877,397; 5,874,299; and 6,255,458 (each by Lonberg and Kay); U.S. Patent Nos. 5,591,669 and 6,023,010 by Krimpenfort and Berns; and U.S. Patent Nos. 5,612,205 by Berns et al.; 5,7 21,367; and 5,789,215, and U.S. Patent Nos. 5,643,763 to Choi and Dunn, and U.S. Patent Application Serial Nos. 07 / 574,748, 07 / 575,962, 07 / 810,279, 07 / 853,408, 07 / 904,068, 07 / 990,860, 08 / 053,131, 08 / 096,762, 08 / 155,301, 08 / 161,739, 08 / 165,699, and 08 / 209,741 to GenPharm. See also EP 0 546 073 B1, WO 92 / 03918, WO 92 / 22645, WO 92 / 22647, WO 92 / 22670, WO 93 / 12227, WO 94 / 00569, WO 94 / 25585, WO 96 / 14436, WO 97 / 13852, and WO 98 / 24884, and U.S. Patent No. 5,981,175. Further see Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994), Tuaillon et al. (1995), and Fishwild et al. (1996).

[0139] Kirin has also demonstrated the generation of human antibodies from mice in which large segments or entire chromosomes have been introduced via microcell fusion. See European patent applications 773 288 and 843 961. Xenerex Biosciences is developing a technology for the potential generation of human antibodies. In this technology, SCID mice are recombined with human lymphocytes (e.g., B and / or T cells). The mice are then immunized with an antigen and an immune response against the antigen is generated. See U.S. Patents 5,476,996; 5,698,767; and 5,958,765.

[0140] Human anti-mouse antibody (HAMA) responses have led the industry to produce chimeric or other humanized antibodies. However, certain human anti-chimeric antibody (HACA) responses are expected, particularly with long-term or multiple-dose use of antibodies. Therefore, it is desirable to provide antibody constructs containing human binding domains against target cell surface antigens and human binding domains against CD3ε to eliminate the problems and / or effects of HAMA or HACA responses.

[0141] The terms “bind to (specifically)”, “recognize (specifically)”, “target (specifically)”, and “react with (specifically)” mean, according to the present invention, that the binding domain interacts or specifically interacts with a given epitope or a given target side end on a target molecule (antigen) (here: target cell surface antigen and CD3ε, respectively).

[0142] The term "epitope" refers to a side of an antigen on which a binding domain (such as an antibody or immunoglobulin, or a derivative, fragment, or variant of an antibody or immunoglobulin) specifically binds. An epitope is antigenic, and therefore the term epitope is sometimes also referred to herein as an "antigen structure" or "antigen determinant." Thus, the binding domain is the "antigen interaction side." The binding / interaction is also understood to define "specific recognition."

[0143] An epitope can be formed by consecutive amino acids or by discontinuous amino acids juxtaposed through the ternary folding of a protein. A linear epitope is an epitope in which the primary sequence of amino acids contains the identified epitope. Linear epitopes typically include at least 3 or at least 4, and more commonly at least 5 or at least 6 or at least 7, such as about 8 or about 10 amino acids, in a unique sequence.

[0144] In contrast to linear epitopes, "conformal epitopes" are epitopes in which the primary sequence of the amino acids constituting the epitope is not the sole defining component of the recognized epitope (e.g., the primary sequence of amino acids is not necessarily the epitope recognized by the binding domain). Typically, conformational epitopes contain an increased number of amino acids compared to linear epitopes. Regarding the recognition of conformational epitopes, the binding domain recognizes the three-dimensional structure of an antigen, preferably a peptide or protein, or a fragment thereof (in the context of this invention, an antigenic structure with a binding domain is included within an antigenic protein on the target cell surface). For example, when a protein molecule folds to form a three-dimensional structure, certain amino acids and / or polypeptide backbones forming the conformational epitope are juxtaposed, enabling the antibody to recognize the epitope. Methods for determining epitope conformation include, but are not limited to, X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, and site-directed spin labeling and electron paramagnetic resonance (EPR) spectroscopy.

[0145] The following describes a method for epitope localization: When a region (a continuous amino acid stretcher) in a human target cell surface antigen protein is exchanged / replaced with a corresponding region of a non-human, non-primate target cell surface antigen (e.g., mouse target cell surface antigen, but others such as chicken, rat, hamster, rabbit, etc. are also possible), a decrease in binding of the binding domain is expected, unless the binding domain is cross-reactive with the non-human, non-primate target cell surface antigen used. This decrease is preferably at least 10%, 20%, 30%, 40%, or 50% compared to binding to the corresponding region in the human target cell surface antigen protein; more preferably at least 60%, 70%, or 80%; and most preferably 90%, 95%, or even 100%, thereby setting the binding to the corresponding region in the human target cell surface antigen protein to 100%. It is envisioned that the above-described human target cell surface antigen / non-human target cell surface antigen chimera is expressed in CHO cells. It is also envisioned that human target cell surface antigen / non-human target cell surface antigen chimeras fuse with the transmembrane and / or cytoplasmic domains of different membrane-bound proteins (such as EpCAM).

[0146] In alternative or additional methods for epitope localization, several truncated forms of extracellular domains of human target cell surface antigens can be generated to identify specific regions recognized by binding domains. In these truncated forms, different extracellular target cell surface antigen domains / subdomains or regions are progressively deleted starting from the N-terminus. It is envisioned that the truncated target cell surface antigen forms can be expressed in CHO cells. It is also envisioned that the truncated target cell surface antigen forms can fuse with transmembrane domains and / or cytoplasmic domains of different membrane-binding proteins (such as EpCAM). It is also envisioned that the truncated target cell surface antigen forms may encompass a signal peptide domain at their N-terminus, such as a signal peptide derived from the mouse IgG heavy chain signal peptide. Further envisioning that the truncated target cell surface antigen forms may encompass a v5 domain at their N-terminus (after the signal peptide), which would allow verification of their correct expression on the cell surface. Those truncated target cell surface antigen forms that no longer encompass the target cell surface antigen region recognized by the binding domain are expected to experience reduced or lost binding. The binding reduction is preferably at least 10%, 20%, 30%, 40%, or 50%; more preferably at least 60%, 70%, or 80%, and most preferably 90%, 95%, or even 100%, thereby setting the binding to the entire human target cell surface antigen protein (or its extracellular region or domain) to 100%.

[0147] Another method for determining the contribution of specific residues of target cell surface antigens to the recognition of antibody constructs or binding domains is alanine scanning (see, for example, Morrison KL and Weiss GA. Cur Opin Chem Biol. [New Insights in Chemical Biology] 2001 June; 5(3):302-7), in which each residue to be analyzed is replaced with alanine, for example, via site-directed mutagenesis. Alanine is used because it has a non-giant, chemically inert methyl functional group, but still mimics the secondary structure references of many other amino acids. In cases where the size of conserved mutant residues is required, sometimes giant amino acids (such as valine or leucine) can be used. Alanine scanning is a well-established technique that has been used for a long time.

[0148] The interaction between a binding domain and an epitope or region containing an epitope means that the binding domain exhibits considerable affinity for the epitope / epitope-containing region on a specific protein or antigen (here: target cell surface antigen and CD3, respectively), and typically does not show significant reactivity with proteins or antigens other than target cell surface antigen or CD3. "Considerable affinity" includes approximately 10... -6 M(KD) or stronger affinity binding. Preferably, when the binding affinity is about 10... -12 Up to 10 -8 M, 10 -12 Up to 10 -9M, 10 -12 Up to 10 -10 M, 10 -11 Up to 10 -8 M, preferably about 10 -11 Up to 10 -9 When M is present, binding is considered specific. In particular, whether the binding domain specifically reacts with or binds to the target can be readily tested by comparing the reaction of the binding domain with the target protein or antigen with the reaction of the binding domain with proteins or antigens other than the target cell surface antigen or CD3. Preferably, the binding domains of the present invention substantially or substantially do not bind to proteins or antigens other than the target cell surface antigen or CD3 (i.e., the first binding domain preferably does not bind to proteins other than the target cell surface antigen, and the second binding domain does not bind to proteins other than CD3). It is envisioned that the antibody construct according to the present invention is characterized by superior affinity characteristics compared to other HLE forms. Therefore, this superior affinity indicates a prolonged half-life in vivo. The longer half-life of the antibody construct according to the present invention can reduce the duration and frequency of administration that typically contributes to improving patient compliance. This is particularly important because the antibody construct of the present invention is especially beneficial for highly debilitated or even multiple cancer patients.

[0149] The terms “substantially / truly non-binding” or “cannot bind” mean that the binding domain of the present invention does not bind to proteins or antigens other than target cell surface antigens or CD3, that is, it does not show more than 30%, preferably no more than 20%, more preferably no more than 10%, particularly preferably no more than 9%, 8%, 7%, 6% or 5% reactivity with proteins or antigens other than target cell surface antigens or CD3, thereby setting the binding to target cell surface antigens or CD3 to 100% respectively.

[0150] It is believed that specific binding is achieved through binding domains and specific motifs in the amino acid sequence of the antigen. Therefore, binding is achieved due to its primary, secondary, and / or tertiary structures, as well as secondary modifications of said structures. The specific interaction between the antigen-interacting side and its specific antigen can lead to simple binding of said side to the antigen. Furthermore, the specific interaction between the antigen-interacting side and its specific antigen can alternatively or additionally lead to signal initiation, for example, due to induced changes in antigen conformation, antigen oligomerization, etc.

[0151] The term "variable" refers to the portion of an antibody or immunoglobulin domain that exhibits sequence variability and participates in determining the specificity and binding affinity of a particular antibody (i.e., "one or more variable domains"). The pairing of variable heavy chains (VH) and variable light chains (VL) together forms a single antigen-binding side end.

[0152] Variability is not uniformly distributed throughout the variable domains of an antibody; it is concentrated in subdomains of each of the heavy and light chain variable regions. These subdomains are called “hypervariable regions” or “complementarity-determining regions” (CDRs). The more conserved (i.e., non-hypervariable) portions of the variable domains are called “framework” regions (FRMs or FRs) and provide a scaffold for the six CDRs in three-dimensional space to form an antigen-binding surface. The naturally occurring variable domains of the heavy and light chains each contain four FRM regions (FR1, FR2, FR3, and FR4), which primarily utilize the β-sheet configuration and are connected by three hypervariable regions that form loops connecting the β-sheet structure and, in some cases, form part of the β-sheet structure. The hypervariable regions in each chain are closely clustered together by the FRMs and, together with hypervariable regions from the other chain, contribute to the formation of antigen-binding side ends (see Kabat et al., cited above).

[0153] The term "CDR" and its plural "CDR" refer to the complementarity-determining regions of the three binding features constituting the light chain variable region (CDR-L1, CDR-L2, and CDR-L3) and the three binding features constituting the heavy chain variable region (CDR-H1, CDR-H2, and CDR-H3). CDRs contain most of the residues responsible for the antibody-antigen specific interaction and thus contribute to the functional activity of the antibody molecule: they are the major determinants of antigen specificity.

[0154] The precise definition of CDR boundaries and lengths is subject to different classification and numbering systems. Therefore, CDRs can be referenced by Kabat, Chothia, contact, or any other boundary definition (including the numbering system described herein). Despite the different boundaries, each of these systems has a degree of overlap in terms of constituting the so-called “hypervariate region” within a variable sequence. Therefore, CDR definitions according to these systems can differ in length and boundary region relative to adjacent frame regions. See, for example, Kabat (a method based on cross-species sequence variability), Chothia (a method based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al., cited above; Chothia et al., J. MoI. Biol [Journal of Molecular Biology], 1987, 196: 901-917; and MacCallum et al., J. MoI. Biol [Journal of Molecular Biology], 1996, 262: 732). Another standard for characterizing antigen-binding flanking ends is the AbM definition used by the AbM antibody modeling software from Oxford Molecular. See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains in: Antibody Engineering Lab Manual (edited by Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). Regarding the use of two residue identification techniques to define overlapping rather than identical regions, they can be combined to define heterozygous CDRs. However, numbering according to the so-called Kabat system is preferred.

[0155] Typically, CDR formation can be classified as canonical ring structures. The term "canonical structure" refers to the backbone conformation used by the antigen-binding (CDR) ring. From comparative structural studies, five of the six antigen-binding rings have been found to have only a limited library of available conformations. Each canonical structure can be characterized by the torsion angle of the polypeptide backbone. Thus, corresponding rings between antibodies can have very similar three-dimensional structures, but most of the rings exhibit high amino acid sequence variability (Chothia and Lesk, J. MoI. Biol. [Journal of Molecular Biology], 1987, 196: 901; Chothia et al., Nature [Nature], 1989, 342: 877; Martin and Thornton, J. MoI. Biol. [Journal of Molecular Biology], 1996, 263: 800). Furthermore, there is a relationship between the ring structure used and the surrounding amino acid sequence. The conformation of a particular canonical class is determined by the ring length and the amino acid residues located in key positions within the ring and within the conserved framework (i.e., outside the ring). Therefore, specific canonical categories can be assigned based on the presence of these key amino acid residues.

[0156] The term "canonical structure" can also include considerations regarding the linear sequence of the antibody, such as those cataloged by Kabat (Kabat et al., cited above). The Kabat numbering scheme (system) is a widely used standard for numbering the amino acid residues of variable domains of antibodies in a consistent manner and is the preferred scheme for the application of this invention, as also mentioned elsewhere herein. Additional structural considerations can also be used to determine the canonical structure of an antibody. For example, those differences not fully reflected by Kabat numbering can be described by the numbering system of Chothia et al., and / or revealed by other techniques, such as crystallography and two- or three-dimensional computational modeling. Thus, a given antibody sequence can be placed in a canonical category, which in particular allows for the identification of appropriate chassis sequences (e.g., based on the expectation of including multiple canonical structures in a library). Kabat numbering of antibody amino acid sequences and structural considerations, as described by Chothia et al., cited above, and their significance for interpreting canonical aspects of antibody structure are described in the literature. The subunit structures and three-dimensional conformations of different classes of immunoglobulins are well known in the art. For a review of antibody structures, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, edited by Harlow et al., 1988.

[0157] The CDR3 of the light chain, and especially the CDR3 of the heavy chain, can constitute the most important determinants of antigen binding within the variable regions of both the light and heavy chains. In some antibody constructs, the heavy chain CDR3 appears to constitute the primary contact region between the antigen and antibody. In vitro selection schemes that modify CDR3 individually can be used to alter antibody binding properties or determine which residues contribute to antigen binding. Therefore, CDR3 is typically the largest source of molecular diversity within the antibody binding side. For example, H3 can be as short as two amino acid residues or more than 26 amino acids.

[0158] In classic full-length antibodies or immunoglobulins, each light (L) chain is linked to a heavy (H) chain by a covalent disulfide bond, while two H chains are linked to each other by one or more disulfide bonds, depending on the H chain isotype. The CH domain closest to the VH is usually named CH1. The constant (“C”) domain does not directly participate in antigen binding but exhibits various effector functions, such as antibody-dependent, cell-mediated cytotoxicity, and complement activation. The Fc region of an antibody is included within the heavy chain constant domain and can, for example, interact with Fc receptors located on the cell surface.

[0159] The sequences of antibody genes are highly altered after assembly and somatic mutation, and it is estimated that these altered genes encode 10. 10 Different antibody molecules (Immunoglobulin Genes, 2nd Edition, edited by Jonio et al., Academic Press, San Diego, CA, 1995). Therefore, the immune system provides an immunoglobulin repertoire. The term "repertoire" refers to at least one nucleotide sequence, wholly or partially derived from at least one sequence encoding at least one immunoglobulin. One or more sequences can be generated in vivo by rearrangement of the V, D, and J segments of the heavy chain and the V and J segments of the light chain. Alternatively, one or more sequences can be generated from cells in response to rearrangement, such as in vitro stimulation. Alternatively, some or all of one or more sequences can be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods, see, for example, U.S. Patent 5,565,332. A repertoire may include only one sequence or may include multiple sequences, including sequences from a genetic diversity set.

[0160] The term "Fc moiety" or "Fc monomer" in this invention refers to a polypeptide comprising at least one domain functional with a CH2 domain and at least one domain functional with a CH3 domain of an immunoglobulin molecule. It is evident from the term "Fc monomer" that a polypeptide comprising those CH domains is a "polypeptide monomer." An Fc monomer can be a polypeptide comprising at least a fragment of an immunoglobulin constant region excluding the first constant region immunoglobulin domain (CH1) of the heavy chain, but retaining at least a functional portion of a CH2 domain and a functional portion of a CH3 domain, wherein the CH2 domain is at the amino terminus of the CH3 domain. In a preferred aspect of this definition, the Fc monomer can be a polypeptide constant region comprising a portion of an Ig-Fc hinge region, a CH2 region, and a CH3 region, wherein the hinge region is at the amino terminus of the CH2 domain. It is envisioned that the hinge region of the invention promotes dimerization. For example, but not limited to, such Fc polypeptide molecules can be obtained by digesting the immunoglobulin region with papain (generating, of course, a dimer of two Fc polypeptides). In another aspect of this definition, the Fc monomer can be a polypeptide region comprising portions of both the CH2 and CH3 regions. For example, but not limited to, such Fc polypeptide molecules can be obtained by digesting immunoglobulin molecules with pepsin. In one embodiment, the polypeptide sequence of the Fc monomer is substantially similar to the following Fc polypeptide sequences: IgG1 Fc region, IgG2 Fc region, IgG3 Fc region, IgG4 Fc region, IgM Fc region, IgA Fc region, IgD Fc region, and IgE Fc region. (See, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993)). Because there are some variations among immunoglobulins, and for clarity only, an Fc monomer refers to the last two heavy chain constant regions of the immunoglobulin domains of IgA, IgD, and IgG, and the last three heavy chain constant regions of the immunoglobulin domains of IgE and IgM. As mentioned above, the Fc monomer may also include flexible hinges at the N-termini of these domains. For IgA and IgM, the Fc monomer may include a J chain. For IgG, the Fc moiety comprises immunoglobulin domains CH2 and CH3, and a hinge between the first two domains and CH2. Although the boundaries of the Fc moiety can vary, an example of a human IgG heavy chain Fc moiety comprising the functional hinge, CH2, and CH3 domains can be defined, for example, as P476 comprising residue D231 (a residue of the hinge domain—corresponding to D234 in Table 1 below) to the carboxyl terminus of the CH3 domain, respectively, L476 (for IgG4), where according to Kabat numbering. A third domain of the antibody construct of the present invention is defined by two Fc moietyes or Fc monomers fused together via a peptide linker; this third domain can also be defined as an scFc domain.

[0161] In one embodiment of the invention, it is envisioned that the scFc domain, as disclosed herein, is included only in the third domain of the antibody construct, with each Fc monomer fused to the other.

[0162] According to the present invention, the IgG hinge region can be identified by analogy using the Kabat numbers listed in Table 1. Based on the above, it is contemplated that the hinge domain / region of the present invention comprises amino acid residues of the IgG1 sequence stretcher corresponding to D234 to P243 according to the Kabat numbers. Similarly, it is contemplated that the hinge domain / region of the present invention comprises or is composed of the IgG1 hinge sequence DKTHTCPPCP (SEQ ID NO: 182) (corresponding to stretchers D234 to P243 as shown in Table 1 below – variations of this sequence are also contemplated, as long as the hinge region still promotes dimerization). In a preferred embodiment of the present invention, the glycosylation site at Kabat position 314 of the CH2 domain in the third domain of the antibody construct is removed by N314X substitution, where X is any amino acid other than Q. The substitution is preferably N314G substitution. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (according to the Kabat positions): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0163] It is also envisioned that the third domain of the antibody construct of the present invention comprises or consists of the following in the amino-to-carboxyl sequence: DKTHTCPPCP (SEQ ID NO: 182) (i.e., hinge)-CH2-CH3-linker-DKTHTCPPCP (SEQ ID NO: 182) (i.e., hinge)-CH2-CH3. In a preferred embodiment, the peptide linker of the above-described antibody construct is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, wherein x is an integer of 5 or greater (e.g., 5, 6, 7, 8, etc. or greater), preferably 6 ((Gly4Ser)6). The construct may further comprise the above-described substituted N314X, preferably N314G, and / or additional substituted V321C and R309C. In a preferred embodiment of the antibody construct of the present invention as defined above, it is envisioned that the second domain binds to the extracellular surface site of the CD3ε chain in humans and / or macaques.

[0164] Table 1: Kabat numbers of amino acid residues in the hinge region

[0165]

[0166] In another embodiment of the invention, the hinge domain / region comprises or is composed of: the IgG2 subtype hinge sequence ERKCCVECPPCP (SEQ ID NO: 183), the IgG3 subtype hinge sequence ELKTPLDTTHTCPRCP (SEQ ID NO: 184) or ELKTPLGDTTHTCPRCP (SEQ ID NO: 185), and / or the IgG4 subtype hinge sequence ESKYGPPCPSCP (SEQ ID NO: 186). The IgG1 subtype hinge sequence may be one of the following EPKSCDKTHTCPPCP (as shown in Table 1 and SEQ ID NO: 183). Therefore, these core hinge regions are also contemplated in the context of the invention.

[0167] The location and sequence of the IgG CH2 and IgG CD3 domains can be identified by analogy using the Kabat numbers listed in Table 2:

[0168] Table 2: Kabat numbers of amino acid residues in the CH2 and CH3 regions of IgG

[0169]

[0170] In one embodiment of the invention, the amino acid residues highlighted in bold are removed from the CH3 domain of the first or both Fc monomers.

[0171] The peptide linker where the peptide monomers of the third domain (“Fc moiety” or “Fc monomer”) are fused together preferably contains at least 25 amino acid residues (25, 26, 27, 28, 29, 30, etc.). More preferably, the peptide linker contains at least 30 amino acid residues (30, 31, 32, 33, 34, 35, etc.). Also preferably, the linker contains up to 40 amino acid residues, more preferably up to 35 amino acid residues, and most preferably exactly 30 amino acid residues. A preferred embodiment of the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, where x is an integer of 5 or greater (e.g., 6, 7, or 8). Preferably, the integer is 6 or 7, more preferably 6.

[0172] When using a linker to fuse a first domain with a second domain or to fuse a first or second domain with a third domain, the linker preferably has a length and sequence sufficient to ensure that each of the first and second domains can independently retain its differential binding specificity. For peptide linkers connecting at least two binding domains (or two variable domains) in the antibody construct of the present invention, those containing only a small number of amino acid residues (e.g., 12 amino acid residues or less) are preferred. Therefore, peptide linkers with 12, 11, 10, 9, 8, 7, 6, or 5 amino acid residues are preferred. Contemplated peptide linkers having fewer than 5 amino acids contain 4, 3, 2, or 1 amino acid, wherein Gly-rich linkers are preferred. Preferred embodiments of peptide linkers for fusing the first and second domains are depicted in SEQ ID NO:1. A preferred linker embodiment for a peptide linker for fusing the second and third domains is the (Gly)4-linker, specifically the G4-linker.

[0173] In the context of the aforementioned “peptide linker,” a particularly preferred “single” amino acid is Gly. Therefore, the peptide linker may consist of a single amino acid, Gly. In a preferred embodiment of the invention, the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or greater (e.g., 2 or 3). Preferred linkers are depicted in SEQ ID Nos: 1 to 12. Features of peptide linkers that do not promote secondary structures are known in the art and described, for example, by Dall'Acqua et al. (Biochem. [Biochemistry] (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol [Molecular Immunology] (1992) 29, 21-30), and Raag and Whitlow (FASEB [Journal of the Federation for Experimental Biology of America] (1995) 9(1), 73-80). Furthermore, peptide linkers that do not promote any secondary structures are preferred. The connection between the domains can be provided, for example, by genetic engineering, as described in the examples. Methods for preparing fused and operatively linked bispecific single-stranded constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., WO 99 / 54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York, 2001).

[0174] In a preferred embodiment of the antibody construct of the present invention, the first and second domains are formed in the form of (scFv)2, scFv-single-domain mAb, biantibody, and oligomers of any of these forms.

[0175] According to particularly preferred embodiments and as described in the appended examples, the first and second domains of the antibody construct of the present invention are “bispecific single-chain antibody constructs,” more preferably bispecific “single-chain Fv” (scFv). Although the two domains, VL and VH, of the Fv fragment are encoded by independent genes, they can be joined together by a synthetic linker using recombinant methods, as described above, which allows them to be prepared as a single protein chain in which the VL and VH regions pair to form a monovalent molecule; see, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA [Proceedings of the National Academy of Sciences] 85:5879-5883. These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the function of the fragments is evaluated in the same manner as that of complete or full-length antibodies. Thus, a single-chain variable fragment (scFv) is a fusion protein of the variable regions of the heavy chain (VH) and light chain (VL) of an immunoglobulin, typically using a short linker peptide of about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linker is typically enriched with glycine for flexibility and with serine or threonine for solubility, and can connect the N-terminus of VH to the C-terminus of VL, or vice versa. Despite the removal of the constant region and the introduction of the linker, the protein retains the specificity of the original immunoglobulin.

[0176] Bispecific single-chain antibody constructs are known in the art and described in the following: WO 99 / 54440, Mack, J. Immunol. [Journal of Immunology] (1997), 158, 3965-3970, Mack, PNAS [Proceedings of the National Academy of Sciences of the United States of America], (1995), 92, 7021-7025, Kufer, Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy], (1997), 45, 193-197, Löffler, Blood, (2000), 95, 6, 2098-2103, Brühl, Immunol. [Immunology], (2001), 166, 2420-2426, Kipriyanov, J. Mol.Biol. [Journal of Molecular Biology], (1999), 293, 41-56. The techniques described for generating single-chain antibodies (see in particular U.S. Patent 4,946,778; Kontermann and Dübel (2010), cited above, and Little (2009), cited above) can be applied to generate single-chain antibody constructs that specifically recognize one or more selected targets.

[0177] Bivalent (also known as divalent) or bispecific single-chain variable fragments (di-scFvs or bi-scFvs having the form (scFv)2) can be engineered by linking two scFv molecules (e.g., using a linker as described above). If the two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule will preferably be called bivalent (i.e., having two valences for the same target epitope). If the two scFv molecules have different binding specificities, the resulting (scFv)2 molecule will preferably be called bispecific. Linking can be performed by generating tandem scFvs by producing a single peptide chain with two VH regions and two VL regions (see, for example, Kufer P. et al., (2004) Trends in Biotechnology 22(5):238-244). Another possibility is to generate scFv molecules with linker peptides that are too short for the two variable regions to fold together (e.g., about five amino acids), thus forcing the scFvs to dimerize. This genotype is called a biantibody (see, for example, Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90 (14): 6444-8).

[0178] According to the present invention, the first domain, the second domain, or both the first and second domains may comprise a single-domain antibody, specifically a variable domain or at least a CDR of the single-domain antibody. A single-domain antibody comprises only one (monomer) antibody variable domain capable of selectively binding to a specific antigen independently of other V regions or domains. The first single-domain antibody is engineered from heavy-chain antibodies found in camels, and these are referred to as V regions. H H fragment. Cartilaginous fish also possess heavy chain antibodies (IgNARs), from which a substance called V can be obtained. NAR Single-domain antibodies are fragments. An alternative approach is to split the dimeric variable domain from common immunoglobulins, such as those from humans or rodents, into monomers, thus obtaining VH or VL as single-domain antibodies (Abs). Although most research on single-domain antibodies is currently based on heavy-chain variable domains, nanobodies derived from light chains have also been shown to specifically bind to target epitopes. Examples of single-domain antibodies are so-called sdAbs, nanobodies, or single-variable-domain antibodies.

[0179] Therefore, (single-domain mAb)2 is a monoclonal antibody construct consisting of (at least) two single-domain monoclonal antibodies, each selected individually from the group consisting of: V H V L V H H and V NAR The linker is preferably in the form of a peptide linker. Similarly, the “scFv-single-domain mAb” is a monoclonal antibody construct consisting of at least one single-domain antibody as described above and an scFv molecule as described above. Again, the linker is preferably in the form of a peptide linker.

[0180] Whether an antibody construct competitively binds to another given antibody construct can be measured in competitive assays such as competitive ELISAs or cell-based competitive assays. Avidin-conjugated microparticles (beads) can also be used. Similar to avidin-coated ELISA plates, each of these beads can serve as a substrate on which an assay can be performed when reacted with a biotinylated protein. The antigen is coated onto the beads, and then pre-coated with a first antibody. A second antibody is added, and any additional binding is determined. Possible means for readout include flow cytometry.

[0181] T cells, or T lymphocytes, are a type of lymphocyte (which is itself a type of white blood cell) that plays a central role in cell-mediated immunity. Several subgroups of T cells exist, each with a distinct function. T cells can be distinguished from other lymphocytes (such as B cells and NK cells) by the presence of the T cell receptor (TCR) on their cell surface. The TCR is responsible for recognizing antigens that bind to the major histocompatibility complex (MHC) molecule and is composed of two distinct protein chains. In 95% of T cells, the TCR consists of alpha (α) and beta (β) chains. When the TCR binds to the antigenic peptide and MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by associated enzymes, co-receptors, specialized adaptor molecules, and activated or released transcription factors.

[0182] The CD3 receptor complex is a protein complex composed of four chains. In mammals, the complex contains a CD3γ (gamma) chain, a CD3δ (delta) chain, and two CD3ε (epsilon) chains. These chains associate with the T cell receptor (TCR) and the so-called ζ (z-chain) to form the T cell receptor CD3 complex and generate an activation signal in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (epsilon) chains are highly associated cell surface proteins of the immunoglobulin superfamily containing a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif essential for TCR signaling, called the immunoreceptor tyrosine-based activation motif, or ITAM for short. The CD3ε molecule is a polypeptide that, in humans, is formed by the activating motif located on chromosome 11. CD3E The gene encodes CD3ε. The most preferred epitope of CD3ε is contained within amino acid residues 1-27 of the extracellular domain of human CD3ε. It is envisioned that the antibody constructs according to the invention typically and advantageously exhibit less nonspecific T cell activation, which is undesirable in specific immunotherapies. This implies a reduced risk of side effects.

[0183] Recruiting T cells via multispecific (or at least bispecific) antibody constructs for redirected lysis of target cells involves cytolytic synapse formation and delivery of perforin and granzymes. The conjugated T cells are capable of continuous target cell lysis and are unaffected by immune escape mechanisms that interfere with peptide antigen processing and presentation or clonal T cell differentiation; see, for example, WO 2007 / 042261.

[0184] Cytotoxicity mediated by the antibody constructs of this invention can be measured in various ways. Effector cells can be, for example, stimulated and enriched (human) CD8-positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMCs). If the target cells are of macaque origin or expressed or transfected with a macaque target cell surface antigen bound to a first domain, the effector cells should also be of macaque origin, such as a macaque T cell line, e.g., 4119LnPx. Target cells should express a target cell surface antigen (at least an extracellular domain), e.g., a human or macaque target cell surface antigen. Target cells can be cell lines (e.g., CHO) stably or transiently transfected with a target cell surface antigen (e.g., a human or macaque target cell surface antigen). Alternatively, target cells can be naturally expressed target cell surface antigen-positive cell lines. For target cell lines expressing high levels of target cell surface antigen on their cell surfaces, EC is expected to be... 50 Values ​​are typically low. The effector cell to target cell (E:T) ratio is usually around 10:1, but this can vary. The cytotoxic activity of the target cell surface antigen xCD3 bispecific antibody construct can be measured... 51Cr-release assay (approximately 18 hours of incubation) or FACS-based cytotoxicity assay (approximately 48 hours of incubation) may be performed. Modifications to the assay incubation time (cytotoxic response) are also possible. Other methods for measuring cytotoxicity are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays (including bioluminescent assays), sulforhodamine B (SRB) assays, WST assays, clonogenic assays, and ECIS techniques.

[0185] Preferably, the cytotoxic activity mediated by the xCD3 bispecific antibody construct of the target cell surface antigen of the present invention is measured in a cell-based cytotoxicity assay. It can also be used in… 51 Cr- release was measured in the assay. It was determined by EC. 50 The value indicates that it corresponds to the half-maximal effective concentration (MCC) of the antibody construct (the concentration of the antibody construct that induces a cytotoxic response midway between baseline and maximum). Preferably, the ECC of the target cell surface antigen xCD3 bispecific antibody construct is... 50 The value is ≤ 5000 pM or ≤ 4000 pM, more preferably ≤ 3000 pM or ≤ 2000 pM, even more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 400 pM or ≤ 300 pM, even more preferably ≤ 200 pM, even more preferably ≤ 100 pM, even more preferably ≤ 50 pM, even more preferably ≤ 20 pM or ≤ 10 pM, and most preferably ≤ 5 pM.

[0186] The EC given above 50 The value can be measured in different assays. Those skilled in the art will know that when using stimulation / enrichment of CD8... + When T cells act as effector cells, compared to unstimulated PBMCs, EC can be expected. 50 The values ​​are lower. Furthermore, it can be expected that, compared to rats with low target expression, EC values ​​will be lower when target cells express high levels of target cell surface antigens. 50 The value is low. For example, when using stimulation / enrichment of human CD8. + When T cells are used as effector cells (and target cells are cells transfected with target cell surface antigens, such as CHO cells or target cell surface antigen-positive human cell lines), the EC50 of the target cell surface antigen xCD3 bispecific antibody construct... 50 The preferred value is ≤ 1000 pM, more preferably ≤ 500 pM, even more preferably ≤ 250 pM, even more preferably ≤ 100 pM, even more preferably ≤ 50 pM, even more preferably ≤ 10 pM, and most preferably ≤ 5 pM. When human PBMCs are used as effector cells, the EC of the target cell surface antigen xCD3 bispecific antibody construct... 50The value is preferably ≤ 5000 pM or ≤ 4000 pM (especially when the target cells are human cell lines positive for target cell surface antigen), more preferably ≤ 2000 pM (especially when the target cells are cells transfected with target cell surface antigen, such as CHO cells), more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 200 pM, even more preferably ≤ 150 pM, even more preferably ≤ 100 pM, and most preferably ≤ 50 pM or lower. When using a macaque T cell line such as LnPx4119 as the effector cell and a macaque target cell surface antigen transfected cell line such as CHO cells as the target cell line, the EC of the target cell surface antigen xCD3 bispecific antibody construct is... 50 The value is preferably ≤ 2000 pM or ≤ 1500 pM, more preferably ≤ 1000 pM or ≤ 500 pM, even more preferably ≤ 300 pM or ≤ 250 pM, even more preferably ≤ 100 pM, and most preferably ≤ 50 pM.

[0187] Preferably, the target cell surface antigen xCD3 bispecific antibody construct of the present invention does not induce / mediate lysis or substantially does not induce / mediate lysis of target cell surface antigen-negative cells such as CHO cells. The terms "does not induce lysis," "substantially does not induce lysis," "does not mediate lysis," or "substantially does not mediate lysis" mean that the antibody construct of the present invention does not induce or mediate more than 30%, preferably no more than 20%, more preferably no more than 10%, particularly preferably no more than 9%, 8%, 7%, 6%, or 5% of the lysis of target cell surface antigen-negative cells, thereby setting the lysis of target cell surface antigen-positive human cell lines as 100%. This is generally applicable to antibody constructs at concentrations up to 500 nM. Those skilled in the art know how to measure cell lysis effortlessly. Furthermore, this specification teaches specific instructions on how to measure cell lysis.

[0188] The difference in cytotoxic activity between the monomeric and dimeric isoforms of a single target cell surface antigen xCD3 bispecific antibody construct is called the "potency gap." This potency gap can be calculated, for example, as the EC50 between the monomeric and dimeric forms of the molecule. 50 The ratio between values. The potency gap of the target cell surface antigen xCD3 bispecific antibody construct of the present invention is preferably ≤5, more preferably ≤4, even more preferably ≤3, even more preferably ≤2, and most preferably ≤1.

[0189] The first binding domain and / or the second (or any other) binding domain of the antibody construct of the present invention preferably have cross-species specificity for members of the primate mammalian order. Cross-species specific CD3 binding domains are described, for example, in WO 2008 / 119567. According to one embodiment, in addition to binding to human target cell surface antigens and human CD3, the first binding domain and / or the second binding domain will also bind to target cell surface antigens / CD3 of primates, including (but not limited to) New World primates (such as marmosets, woolly tamarins, or squirrel monkeys), Old World primates (such as baboons and rhesus macaques), gibbons, and non-human hominids.

[0190] In one embodiment of the antibody construct of the present invention, the first binding domain binds to a human target cell surface antigen and further binds to a macaque target cell surface antigen (such as the target cell surface antigen of a cynomolgus monkey), and more preferably, binds to a macaque target cell surface antigen expressed on surface macaque cells. The affinity of the first binding domain for the macaque target cell surface antigen is preferably ≤ 15 nM, more preferably ≤ 10 nM, even more preferably ≤ 5 nM, even more preferably ≤ 1 nM, even more preferably ≤ 0.5 nM, even more preferably ≤ 0.1 nM, and most preferably ≤ 0.05 nM or even ≤ 0.01 nM.

[0191] Preferably, the affinity gap (as determined, for example, by BiaCore or by Scatchard analysis) of the antibody construct according to the invention for binding macaque target cell surface antigen to human target cell surface antigen [ma target cell surface antigen: hu target cell surface antigen] is < 100, preferably < 20, more preferably < 15, further preferably < 10, even more preferably < 8, more preferably < 6, and most preferably < 2. The preferred range for the affinity gap of the antibody construct according to the invention for binding macaque target cell surface antigen to human target cell surface antigen is between 0.1 and 20, more preferably between 0.2 and 10, even more preferably between 0.3 and 6, even more preferably between 0.5 and 3 or between 0.5 and 2.5, and most preferably between 0.5 and 2 or between 0.6 and 2.

[0192] The second (binding) domain of the antibody construct of the present invention binds to human CD3ε and / or macaque CD3ε. In a preferred embodiment, the second domain further binds to the CD3ε of marmosets, woolly-crowned tamarins, or squirrel monkeys. Both marmosets and woolly-crowned tamarins belong to the subfamily Marmosetinae (Marmosetinae). Callitrichidae New World primates, while squirrel monkeys belong to the family Cypriniidae (…). Cebidae New World primates.

[0193] Preferably, in the antibody construct of the present invention, the second domain that binds to the extracellular domain of human and / or macaque CD3 comprises a VL region containing a subset of CDR-L1, CDR-L2 and CDR-L3:

[0194] (a) CDR-L1 as described in SEQ ID NO: 27 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 28 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 29 of WO 2008 / 119567;

[0195] (b) CDR-L1 as described in SEQ ID NO: 117 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 118 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 119 of WO 2008 / 119567; and

[0196] I. CDR-L1 as described in SEQ ID NO: 153 of WO 2008 / 119567, CDR-L2 as described in SEQ ID NO: 154 of WO 2008 / 119567, and CDR-L3 as described in SEQ ID NO: 155 of WO 2008 / 119567.

[0197] In a similarly preferred embodiment of the antibody construct of the present invention, the second domain that binds to the extracellular domain of the human and / or macaque CD3ε chain comprises a VH region containing a subset of CDR-H1, CDR-H2, and CDR-H3:

[0198] (a) CDR-H1 as described in SEQ ID NO: 12 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 13 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 14 of WO 2008 / 119567;

[0199] (b) CDR-H1 as described in SEQ ID NO: 30 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 31 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 32 of WO 2008 / 119567;

[0200] I. CDR-H1 as depicted in SEQ ID NO: 48 of WO 2008 / 119567, CDR-H2 as depicted in SEQ ID NO: 49 of WO 2008 / 119567, and CDR-H3 as depicted in SEQ ID NO: 50 of WO 2008 / 119567;

[0201] (d) CDR-H1 as described in SEQ ID NO: 66 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 67 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 68 of WO 2008 / 119567;

[0202] I. CDR-H1 as depicted in SEQ ID NO: 84 of WO 2008 / 119567, CDR-H2 as depicted in SEQ ID NO: 85 of WO 2008 / 119567, and CDR-H3 as depicted in SEQ ID NO: 86 of WO 2008 / 119567;

[0203] (f) CDR-H1 as described in SEQ ID NO: 102 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 103 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 104 of WO 2008 / 119567;

[0204] (g) CDR-H1 as described in SEQ ID NO: 120 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 121 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 122 of WO 2008 / 119567;

[0205] (h) CDR-H1 as described in SEQ ID NO: 138 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 139 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 140 of WO 2008 / 119567;

[0206] (i) CDR-H1 as described in SEQ ID NO: 156 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 157 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 158 of WO 2008 / 119567; and

[0207] (j) CDR-H1 as described in SEQ ID NO: 174 of WO 2008 / 119567, CDR-H2 as described in SEQ ID NO: 175 of WO 2008 / 119567, and CDR-H3 as described in SEQ ID NO: 176 of WO 2008 / 119567.

[0208] In a preferred embodiment of the antibody construct of the present invention, the above three groups of VL CDRs and the above ten groups of VH CDRs are combined in the second binding domain to form (30) groups, each group containing CDR-L 1-3 and CDR-H 1-3.

[0209] Preferably, for the antibody construct of the present invention, the second domain binding to CD3 includes a VL region selected from the group consisting of: such as the VL region depicted in SEQ ID NO: 17, 21, 35, 39, 53, 57, 71, 75, 89, 93, 107, 111, 125, 129, 143, 147, 161, 165, 179 or 183 of WO 2008 / 119567 or the VL region depicted in SEQ ID NO: 200.

[0210] Also preferably, the second domain associated with CD3 includes a VH region selected from the group consisting of: such as the VH region depicted in SEQ ID NO: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177 or 181 of WO2008 / 119567 or the VH region depicted in SEQ ID NO: 201.

[0211] Most preferably, the antibody construct of the present invention is characterized in that the second domain binding to CD3 includes a VL region and a VH region selected from the group consisting of:

[0212] (a) The VL region as depicted in SEQ ID NO: 17 or 21 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 15 or 19 of WO 2008 / 119567;

[0213] (b) The VL region as depicted in SEQ ID NO: 35 or 39 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 33 or 37 of WO 2008 / 119567;

[0214] I. The VL region as depicted in SEQ ID NO: 53 or 57 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 51 or 55 of WO 2008 / 119567;

[0215] (d) The VL region as depicted in SEQ ID NO: 71 or 75 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 69 or 73 of WO 2008 / 119567;

[0216] The VL region as depicted in SEQ ID NO: 89 or 93 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 87 or 91 of WO 2008 / 119567;

[0217] (f) The VL region as depicted in SEQ ID NO: 107 or 111 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 105 or 109 of WO 2008 / 119567;

[0218] (g) The VL region as depicted in SEQ ID NO: 125 or 129 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 123 or 127 of WO 2008 / 119567;

[0219] (h) The VL region as described in SEQ ID NO: 143 or 147 of WO 2008 / 119567 and the VH region as described in SEQ ID NO: 141 or 145 of WO 2008 / 119567;

[0220] (i) the VL region as depicted in SEQ ID NO: 161 or 165 of WO 2008 / 119567 and the VH region as depicted in SEQ ID NO: 159 or 163 of WO 2008 / 119567; and

[0221] (j) The VL region as described in SEQ ID NO: 179 or 183 of WO 2008 / 119567 and the VH region as described in SEQ ID NO: 177 or 181 of WO 2008 / 119567.

[0222] Preferably, the second domain binding to CD3 in the antibody construct of the present invention comprises the VL region as depicted in SEQ ID NO: 200 and the VH region as depicted in SEQ ID NO: 201.

[0223] According to a preferred embodiment of the antibody construct of the present invention, the first and / or second domains have the following form: the pair of VH and VL regions are in the form of single-chain antibodies (scFv). The VH and VL regions are arranged in the order of VH-VL or VL-VH. Preferably, the VH region is located at the N-terminus of the adapter sequence, and the VL region is located at the C-terminus of the adapter sequence.

[0224] The preferred embodiment of the antibody construct of the present invention is characterized in that the second domain binding to CD3 comprises an amino acid sequence selected from the group consisting of: SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 of WO 2008 / 119567 or as depicted in SEQ ID NO: 202.

[0225] Covalent modifications of antibody constructs are also included within the scope of this invention and are typically, but not always, performed post-translational. For example, covalent modifications of several species of antibody constructs can be introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizing agent capable of reacting with selected side chain or N- or C-terminal residues.

[0226] Cysteine ​​residues most commonly react with α-haloacetic esters (and corresponding amines), such as chloroacetic acid or chloroacetamide, to yield carboxymethyl or carboxamide-methyl derivatives. Cysteine ​​residues can also be derived by reacting with bromotrifluoroacetone, α-bromo-β-(5-imidazolyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl-2-pyridyl disulfide, p-chloromercuric benzoate, 2-chloromercuric-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0227] Histidine residues are derived by reaction with diethyl pyrocarbonate at pH 5.5–7.0, as this formulation has relative specificity for the histidine side chain. p-Bromobenzoylmethyl bromide is also useful; this reaction is preferably carried out at pH 6.0 in 0.1 M sodium dimethylarsinate. Lysidine residues and amino-terminal residues react with succinic anhydride or other carboxylic anhydrides. Derivatization with these agents has the effect of reversing the charge of the lysidine residue. Other suitable reagents for derivatizing α-amino residues include imine esters such as methylpyridinium imide; pyridoxal phosphate; pyridoxal; chlorine borohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reactions with glyoxylates.

[0228] Arginyl residues are modified by reaction with one or more conventional reagents, including benzoylcarboxaldehyde, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Due to the high pKa of the guanidine functional group, the derivatization of arginine residues requires the reaction to be carried out under basic conditions. Furthermore, these reagents can react with lysine groups as well as the ε-amino group of arginine.

[0229] Specific modifications can be made to tyrosine acyl residues, with particular interest in introducing spectral labeling into tyrosine acyl residues through reactions with aromatic diazo compounds or tetranitromethane. Most commonly, N-acetylimidazolium and tetranitromethane are used to form O-acetyltyrosine acyl compounds and 3-nitro derivatives, respectively. 125 I or 131 The chloramine-T method described above is suitable for preparing labeled proteins for radioimmunoassays by iodizing tyrosine residues.

[0230] The carboxyl side group (aspartic or glutamic) is selectively modified by reacting with a carbodiimide (R'—N=C=N–R'), wherein R and R' are optionally different alkyl groups, such as 1-cyclohexyl-3-(2-morpholino-4-ethyl)carbodiimide or 1-ethyl-3-(4-azainium-4,4-dimethylpentyl)carbodiimide. Furthermore, the aspartic and glutamic residues are converted to asparaginic and glutamic residues by reacting with ammonium ions.

[0231] Derivatization with bifunctional agents can be used to crosslink the antibody constructs of the present invention to a water-insoluble carrier matrix or surface for use in a variety of methods. Commonly used crosslinking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters (e.g., esters with 4-azidosalicylic acid), homodifunctional imine esters, including disuccinimide esters such as 3,3'-dithiobis(succinimide propionate), and bifunctional maleimides such as bis-N-maleimide-1,8-octane. Derivatizing agents such as methyl 3-[(p-azidophenyl)dithio]propionimide produce photoactivated intermediates capable of forming crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates and reactive substrates, as described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440, are used for protein fixation.

[0232] Glutamine acyl residues and asparagine acyl residues are typically deamidated to their respective glutamine acyl residues and asparagine acyl residues. Alternatively, these residues are deamidated under weakly acidic conditions. Both forms of these residues are within the scope of this invention.

[0233] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl groups of serine or threonyl residues, methylation of the α-amino groups of the lysine, arginine, and histidine side chains (TE Creighton, Proteins: Structure and Molecular Properties, WH Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of N-terminal amines, and amidation of any C-terminal carboxyl group.

[0234] Covalent modification of another species of antibody construct included within the scope of this invention comprises altering the glycosylation pattern of the protein. As is known in the art, the glycosylation pattern can be dependent on the protein sequence (e.g., the presence or absence of specific glycosylated amino acid residues discussed below) or the host cell or organism from which the protein is produced. Specific expression systems are discussed below.

[0235] Glycosylation of peptides is typically N-linked or O-linked. N-linking refers to the attachment of the carbohydrate moiety to a side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of either of these tripeptide sequences in a peptide creates a potential glycosylation site. O-linked glycosylation involves attaching one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyl amino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

[0236] Adding glycosylation sites to the antibody construct can be conveniently accomplished by altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences (for N-linked glycosylation sites). Modifications can also be made by adding or substituting one or more serine or threonine residues into the starting sequence (for O-linked glycosylation sites). For convenience, the amino acid sequence of the antibody construct is preferably altered at the DNA level, particularly by mutating the DNA encoding the polypeptide at preselected bases to generate a codon that will be translated into the desired amino acid.

[0237] Another approach to increasing the amount of carbohydrate moiety on antibody constructs is by chemically or enzymatically coupling glycosides to proteins. The advantage of these procedures is that they do not require the generation of proteins in host cells with glycosylation capabilities for N- and O-linking. Depending on the coupling method used, one or more sugars may be linked to (a) arginine and histidine, (b) a free carboxyl group, (c) a free thiol group, such as those of cysteine, (d) a free hydroxyl group, such as those of serine, threonine, or hydroxyproline, (f) an aromatic residue, such as those of phenylalanine, tyrosine, or tryptophan, or (f) an amide group of glutamine. These methods are described in WO 87 / 05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem. [CRC Critical Review of Biochemistry], pp. 259–306.

[0238] The removal of carbohydrate moieties present on the starting antibody construct can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposing the protein to the compound trifluoromethanesulfonic acid, or an equivalent compound. This treatment results in the cleavage of most or all sugars except the linking sugar (N-acetylglucosamine or N-acetylglucosamine), while keeping the polypeptide intact. Chemical deglycosylation was developed by Hakimuddin et al., 1987. Arch.Biochem. Biophys. [Journal of Biochemistry and Biophysics] 259:52 and Edge et al., 1981, Anal. Biochem. [Analytical Biochemistry] 118:131 describes the enzymatic cleavage of carbohydrate moieties on polypeptides, which can be achieved using various endoglucosidases and exoglucosidases, as described by Thotakura et al., 1987, and Meth. Enzymol. [Enzymatic Methods] 138:350. Glycosylation at potential glycosylation sites can be prevented by using the compound tunicamycin, as described by Duskin et al., 1982. J. Biol. Chem. As described in [Journal of Biochemistry] 257:3105, tunicamycin blocks the formation of protein-N-glycosidic bonds.

[0239] This article also considers other modifications to the antibody construct. For example, another type of covalent modification of the antibody construct includes linking the antibody construct to various non-protein polymers, including but not limited to various polyols such as polyethylene glycol, polypropylene glycol, polyoxyethylene, or copolymers of polyethylene glycol and polypropylene glycol, in a manner illustrated in U.S. Patent Nos. 4,640,835, 4,496,689, 4,301,144, 4,670,417, 4,791,192, or 4,179,337. Furthermore, as is known in the art, amino acid substitutions can be made at different positions within the antibody construct, for example, to facilitate the addition of polymers such as PEG.

[0240] In some embodiments, covalent modification of the antibody construct of the present invention includes the addition of one or more markers. The marker group can be coupled to the antibody construct via spacer arms of various lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and can be used in carrying out the present invention. The terms “marker” or “marker group” refer to any detectable marker. Generally, markers fall into several categories depending on the assay in which they will be detected – examples include, but are not limited to:

[0241] a) Isotope labeling, which can be radioactive isotopes or heavy isotopes, such as radioactive isotopes or radionuclides (e.g. 3 H, 14 C 15 N、 35 S, 89 Zr、 90 Y、 99 Tc, 111 In、 125 I, 131 I)

[0242] b) Magnetic markings (e.g., magnetic particles)

[0243] c) Redox-active components

[0244] d) Optical dyes (including but not limited to chromophores, phosphors, and fluorophores), such as fluorophores (e.g., FITC, rhodamine, lanthanide phosphors), chemiluminescent groups, and fluorophores, which may be "small molecule" fluorescent agents or protein fluorescent agents.

[0245] e) Enzymatic groups (e.g., horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase)

[0246] f) Biotinylated groups

[0247] g) A predetermined polypeptide epitope recognized by the second reporter (e.g., leucine zipper pair sequence, binding side of the second antibody, metal-binding domain, epitope tag, etc.).

[0248] "Fluorescent label" refers to any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include, but are not limited to, fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methyl-coumarin, pyrene, malachite green, stilbene, fluorescein yellow, waterfall blue J, Texas red, IAEDANS, EDANS, BODIPY FL, LC red 640, Cy5, Cy5.5, LC red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), waterfall blue, waterfall yellow, and R-phycoerythrin (PE) (Molecular Probes, Eugene, Oregon). OR), FITC, Rhodamine and Texas Red (Pierce, Rockford, IL), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, PA). Suitable optical dyes (including fluorophores) are described in Richard P. Haugland's Molecular Probes Handbook.

[0249] Suitable protein fluorescent labeling includes, but is not limited to, green fluorescent protein (GFP), including GFP from species such as *Renilla*, *Ptilosarcus*, or *Aequorea* (Chalfie et al., 1994, *Science* 263:802-805); EGFP (Clontech Laboratories, Inc., Genebank Registry No. U55762); blue fluorescent protein (BFP, Quantum Biotechnologies, Inc., 1801 de Maisonneuve Blvd. West, 8th Floor, H3H 1J9, Montreal, Quebec, Canada); Stauber, 1998, *Biotechniques* 24:462-471; Heim et al., 1996, *Curr. Biol.* 6:178-182), Enhanced Yellow Fluorescent Protein (EYFP, Crotec Laboratories, Inc.), Luciferase (Ichiki et al., 1993, J. Immunol. [Journal of Immunology] 150:5408-5417), β-galactosidase (Nolan et al., 1988, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America]). 85:2603-2607) and sea kidney (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, US Patent Nos. 5,292,658, 5,418,155, 5,683,888, 5,741,668, 5,777,079, 5,804,387, 5,874,304, 5,876,995, 5,925,558).

[0250] The antibody constructs of the present invention may also include additional domains, such as those that facilitate the separation of the molecule or relate to adaptive pharmacokinetic distribution of the molecule. Domains facilitating the separation of the antibody construct may be selected from peptide motifs or auxiliaryly introduced portions that can be captured in a separation method (e.g., a separation column). Non-limiting examples of such additional domains include peptide motifs referred to as Myc-tags, HAT-tags, HA-tags, TAP-tags, GST-tags, chitin-binding domains (CBD-tags), maltose-binding protein (MBP-tags), Flag-tags, Strep-tags and their variants (e.g., StrepII-tags), and His-tags. All antibody constructs disclosed herein characterized by an identified CDR may include a His-tag domain, typically referred to as a continuous His residue repeating sequence, preferably five, and more preferably six His residues (hexahistidines), in the amino acid sequence of the molecule. The His-tag may be located, for example, at the N or C terminus of the antibody construct, preferably at the C terminus. Most preferably, the hexahistine tag (HHHHHH) (SEQ ID NO:199) is linked to the C-terminus of the antibody construct according to the invention via a peptide bond. Additionally, the PLGA-PEG-PLGA conjugate system can be combined with the polyhistine tag for sustained-release application and improved pharmacokinetic distribution.

[0251] Amino acid sequence modifications of the antibody constructs described herein have also been considered. For example, it may be necessary to improve the binding affinity and / or other biological properties of the antibody construct. Amino acid sequence variants of the antibody construct are prepared by introducing appropriate nucleotide changes into the nucleic acid of the antibody construct or by peptide synthesis. All amino acid sequence modifications described below should produce antibody constructs that retain the desired biological activity (binding to target cell surface antigens and CD3) of the unmodified parent molecule.

[0252] The term "amino acid" or "amino acid residue" typically refers to an amino acid having its recognized definition in the art, such as an amino acid selected from the group consisting of: alanine (Ala or A); arginine (Arg or R); asparagine (Asn or N); aspartic acid (Asp or D); cysteine ​​(Cys or C); glutamine (Gin or Q); glutamic acid (Giu or E); glycine (Giy or G); histidine (His or H); isoleucine (He or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (Val or V), although modified, synthetic, or rare amino acids may be used as needed. Generally speaking, amino acids can be grouped into those with nonpolar side chains (e.g., Ala, Cys, He, Leu, Met, Phe, Pro, Val); those with negatively charged side chains (e.g., Asp, Giu); those with positively charged side chains (e.g., Arg, His, Lys); or those with uncharged polar side chains (e.g., Asn, Cys, Gin, Giy, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0253] Amino acid modifications include, for example, the deletion and / or insertion and / or substitution of residues within the amino acid sequence of the antibody construct. Any combination of deletions, insertions, and substitutions may be performed to achieve the final construct, provided that the final construct possesses the desired characteristics. Amino acid changes can also alter the post-translational processes of the antibody construct, such as changing the number or location of glycosylation sites.

[0254] For example, 1, 2, 3, 4, 5, or 6 amino acids (depending on their length) can be inserted, substituted, or deleted in each CDR, while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids can be inserted, substituted, or deleted in each FR. Preferably, amino acid sequence insertion into the antibody construct includes fusions with amino and / or carboxyl termini of polypeptides containing 100 or more residues in length ranging from 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 residues, as well as intra-sequence insertions of single or multiple amino acid residues. Corresponding modifications can also be made within the third domain of the antibody construct of the present invention. Insertion variants of the antibody construct of the present invention include fusions with the N-terminus or C-terminus of an enzyme antibody construct or with a polypeptide.

[0255] Sites of most interest in substitution mutagenesis include (but are not limited to) the CDRs of the heavy and / or light chains, particularly hypervariable regions, but FR alterations of the heavy and / or light chains are also considered. Substitutions are preferably conservative substitutions as described herein. Preferably, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids can be substituted in the CDR, while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids can be substituted in the frame region (FR), depending on the length of the CDR or FR. For example, if the CDR sequence covers 6 amino acids, it is conceivable that 1, 2, or 3 of these amino acids will be substituted. Similarly, if the CDR sequence covers 15 amino acids, it is conceivable that 1, 2, 3, 4, 5, or 6 of these amino acids will be substituted.

[0256] A useful method for identifying certain residues or regions in an antibody construct that are preferred mutagenic sites is called "alanine scanning mutagenesis," as described by Cunningham and Wells in Science, 244: 1081-1085 (1989). Here, residues or target residue groups (e.g., charged residues such as arg, asp, his, lys, and glu) in the antibody construct are identified and replaced with neutral or negatively charged amino acids (preferably alanine or polyalanine) to influence the interaction between the amino acid and the epitope.

[0257] Then, those amino acid positions exhibiting functional sensitivity to substitution are refined by introducing further or other variants at or to the substitution site. Therefore, while the sites or regions used to introduce amino acid sequence changes are predetermined, the nature of the mutation itself does not need to be predetermined. For example, to analyze or optimize the performance of mutations at a given site, alanine scans or random mutagenesis can be performed at the target codon or region, and variants of the expressed antibody construct can be screened for the optimal combination of desired activities. Techniques for substitution mutations at predetermined sites in DNA with known sequences are well known, such as M13 primer mutagenesis and PCR mutagenesis. Mutants are screened using assays of antigen-binding activity (such as target cell surface antigens or CD3 binding).

[0258] Generally, if an amino acid is substituted in one or all of the CDRs in the heavy chain and / or light chain, it is preferred that the subsequently obtained "substituted" sequence has at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, and particularly preferably 90% or 95% identity with the "initial" CDR sequence. This means that the substitution depends on the degree of identity between the CDR length and the "substituted" sequence. For example, a CDR having 5 amino acids is preferably 80% identical with its substituted sequence in order to substitute at least one amino acid. Therefore, the CDR of the antibody construct can have different degrees of identity with its substituted sequence; for example, CDRL1 can have 80% identity, while CDRL3 can have 90%.

[0259] Preferred substitutions (or alternatives) are conservative substitutions. However, any substitution (including non-conservative substitutions or one or more of the “exemplary substitutions” listed in Table 3 below) is contemplated, provided that the antibody construct retains its ability to bind to the target cell surface antigen via the first domain and to CD3, CD3ε, via the second domain, and / or that its CDR has identity with the subsequently substituted sequence (having at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, and particularly preferably 90% or 95% identity with the “original” CDR sequence).

[0260] Conservative substitutions are shown under the heading “Preferred Substitutions” in Table 3. If such substitutions result in a change in biological activity, they may be designated as “Exemplary Substitutions” in Table 3 or more substantial changes may be introduced as further described below with reference to the amino acid categories, and products may be screened to obtain the desired characteristics.

[0261] Table 3: Amino Acid Substitutions

[0262]

[0263] The substantial modification of the biological properties of the antibody constructs of the present invention is accomplished by selecting substitutions that are significantly different in maintaining the following effects: (a) the structure of the polypeptide backbone in the substitution region, such as folded or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the majority of the side chain. Naturally occurring residues are grouped based on common side chain characteristics: (1) hydrophobic: leucine, met, ala, val, leu, ile; (2) neutral hydrophilic: cys, ser, thr, asn, gln; (3) acidic: asp, gln; (4) basic: his, lys, arg; (5) residues affecting chain orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

[0264] Non-conservative substitutions would require replacing members of one class with members of another. Any cysteine ​​residue that does not participate in maintaining the proper conformation of the antibody construct can generally be substituted with a serine residue to improve the oxidative stability of the molecule and prevent aberrant cross-linking. Conversely, one or more cysteine ​​bonds can be added to the antibody to improve its stability (especially when the antibody is an antibody fragment, such as an Fv fragment).

[0265] For amino acid sequences, sequence identity and / or similarity are determined using standard techniques known in the art, including but not limited to Smith and Waterman, 1981. Adv. Applied Math. [Advanced Applied Mathematics] 2:482, Local Sequence Identity Algorithm, Needleman and Wunsch, 1970. J. Mol. Biol. [Journal of Molecular Biology] 48:443, Sequence Identity Alignment Algorithm, Pearson and Lipman, 1988. Proc. Nat. Acad. Sci. USA [Proceedings of the National Academy of Sciences] 85:2444. Retrieval of similarity methods, computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Drive, Madison, Wisconsin), Devereux et al., 1984. Nucl. Acid Res. [Nucleic Acid Research] The optimal matching sequence procedure described in 12:387-395 preferably uses the default settings or is checked. Preferably, the identity percentage is calculated by FastDB based on the following parameters: mismatch penalty of 1; gap penalty of 1; gap size penalty of 0.33; and connection penalty of 30, “Current Methods in Sequence Comparison and Analysis”, Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp. 127-149 (1988), Alan R. Liss, Inc.

[0266] A useful example of an algorithm is PILEUP. PILEUP uses progressive pairwise alignment to create multiple sequence alignments from a set of related sequences. It can also draw a tree diagram showing the clustering relationships used to create the alignments. PILEUP uses Feng and Doolittle, 1987.J. A simplification of the progressive alignment method described in Mol. Evol. [Journal of Molecular Evolution] 35:351-360; this method is similar to that of Higgins and Sharp, 1989. CABIOS The method described in 5:151-153. Useful PILEUP parameters include a default empty space weight of 3.00, a default empty space length weight of 0.10, and a weighted terminal empty space.

[0267] Another example of a useful algorithm is the BLAST algorithm, described in: Altschul et al., 1990. J. Mol. Biol. [Journal of Molecular Biology] 215:403-410; Altschul et al., 1997, Nucleic AcidsRes. [Nucleic Acid Research] 25:3389-3402; and Karin et al., 1993, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America] 90:5873-5787. A particularly useful BLAST procedure is from Altschul et al., 1996. Methods The WU-BLAST-2 procedure was obtained in Enzymology [Enzymological Methods] 266:460-480. WU-BLAST-2 uses several search parameters, most of which are set to default values. Adjustable parameters are set to the following values: overlap interval = 1, overlap fraction = 0.125, word threshold (T) = II. The HSP S and HSP S2 parameters are dynamic values ​​and are determined by the procedure itself based on the composition of a specific sequence and a specific database used to search for sequences of interest; however, these values ​​can be adjusted to increase sensitivity.

[0268] Another useful algorithm is that of Altschul et al., 1993. Nucl. A vacancy BLAST was reported in Acids Res. [Nucleic Acid Research] 25:3389-3402. The vacancy BLAST used BLOSUM-62 instead of a score; the threshold parameter T was set to 9; a double-click method was used to trigger non-vacancy expansion, incurring a cost of 10+k for a vacancy length of k; Xu was set to 16, and Xg was set to 40 (for the database search phase) and 67 (for the algorithm's output phase). Vacancy matching was triggered by a score corresponding to approximately 22 bits.

[0269] Generally, the amino acid homology, similarity, or identity between the various variant CDR or VH / VL sequences is at least 60% with respect to the sequences described herein, and more typically at least 65% or 70%, more preferably at least 75% or 80%, and even more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, and almost 100% with preferred increased homology or identity. Similarly, the "percentage (%) nucleic acid sequence identity" relative to the nucleic acid sequence of the binding protein identified herein is defined as the percentage of nucleotide residues in the candidate sequence that are identical to nucleotide residues in the coding sequence of the antibody construct. Specifically, the BLASTN module of WU-BLAST-2 was used with default parameters, and the overlap interval and overlap fraction were set to 1 and 0.125, respectively.

[0270] Generally speaking, the nucleotide sequence encoding the various variant CDR or VH / VL sequence has at least 60% homology, similarity, or identity with the nucleotide sequence described herein, and more typically has at least 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, and preferably an increased homology or identity of almost 100%. Therefore, the “variant CDR” or “variant VH / VL region” is one that has specified homology, similarity or identity with the parental CDR / VH / VL of the present invention and shares biological functions, including but not limited to at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the specificity and / or activity of the parental CDR or VH / VL.

[0271] In one embodiment, the percentage of identity between the antibody construct according to the invention and the human lineage is ≥ 70% or ≥ 75%, more preferably ≥ 80% or ≥ 85%, even more preferably ≥ 90%, and most preferably ≥ 91%, ≥ 92%, ≥ 93%, ≥ 94%, ≥ 95%, or even ≥ 96%. Identity with the human antibody lineage gene product is considered an important characteristic for reducing the risk of therapeutic proteins inducing an immune response against the drug in patients during treatment. Hwang and Foote (“Immunogenicity of engineered antibodies”; Methods 36 (2005) 3-10) demonstrated that reducing the non-human portion of the drug antibody construct leads to a reduced risk of inducing anti-drug antibodies in patients during treatment. A comparison of numerous clinically evaluated antibody drugs and corresponding immunogenicity data shows the following trend: humanization of the V region of the antibody results in lower protein immunogenicity (average 5.1% of patients) than antibodies carrying the unchanged non-human V region (average 23.59% of patients). Therefore, V-region-based protein therapeutics in antibody construct form need to have a high degree of identity with human sequences. For the purpose of determining germline identity, the V region of VL can be aligned with the amino acid sequences of human germline V and J regions (http: / / vbase.mrc-cpe.cam.ac.uk / ) using VectorNTI software, and the amino acid sequence (in percentage) is calculated by dividing identical amino acid residues by the total number of amino acid residues in VL. The same applies to the VH region (http: / / vbase.mrc-cpe.cam.ac.uk / ), except that VH CDR3 can be excluded due to the high diversity of VHCDR3 and the lack of existing human germline VH CDR3 alignment pairs. Recombinant techniques can then be used to increase sequence identity with human antibody germline genes.

[0272] In another embodiment, the bispecific antibody construct of the present invention exhibits high monomer yield under standard research-scale conditions, such as in a standard two-step purification process. Preferably, the monomer yield of the antibody construct according to the present invention is ≥ 0.25 mg / L supernatant, more preferably ≥ 0.5 mg / L, even more preferably ≥ 1 mg / L, and most preferably ≥ 3 mg / L supernatant.

[0273] Similarly, the yield of the isotype of the dimeric antibody construct can be determined, and the monomer percentage (i.e., monomer:(monomer + dimer)) can be determined accordingly. The yield of the monomeric and dimeric antibody constructs and the calculated monomer percentage can be obtained, for example, in an SEC purification step from culture supernatant produced on a standardized research scale in a rolling flask. In one embodiment, the monomer percentage of the antibody construct is ≥ 80%, more preferably ≥ 85%, even more preferably ≥ 90%, and most preferably ≥ 95%.

[0274] In one embodiment, the preferred plasma stability of the antibody construct (the ratio of EC50 with plasma to EC50 without plasma) is ≤ 5 or ≤ 4, more preferably ≤ 3.5 or ≤ 3, even more preferably ≤ 2.5 or ≤ 2, and most preferably ≤ 1.5 or ≤ 1. The plasma stability of the antibody construct can be determined by incubating the construct in human plasma at 37°C for 24 hours, followed by... 51 The EC50 was determined in the chromium-releasing cytotoxicity assay for testing. Effector cells in the cytotoxicity assay can be stimulated, enriched human CD8-positive T cells. Target cells can be, for example, CHO cells transfected with human target cell surface antigen. The effector cell to target cell (E:T) ratio can be selected as 10:1. The human plasma bank used for this purpose was derived from blood collected from healthy donors using an EDTA-coated syringe. Cellular components were removed by centrifugation, and the supernatant plasma phase was collected and subsequently pooled. As a control, the antibody construct was diluted immediately prior to the cytotoxicity assay in RPMI-1640 medium. Plasma stability was calculated as the ratio of EC50 (after plasma incubation) to EC50 (control).

[0275] Furthermore, it is preferable that the monomer-to-dimer conversion of the antibody construct of the present invention is low. The conversion can be measured under different conditions and analyzed by high-performance size exclusion chromatography. For example, incubation of the monomeric isoform of the antibody construct can be carried out in an incubator for 7 days at 37°C and concentrations such as 100 µg / ml or 250 µg / ml. Under these conditions, it is preferable that the antibody construct of the present invention exhibits a dimer percentage of ≤ 5%, more preferably ≤ 4%, even more preferably ≤ 3%, even more preferably ≤ 2.5%, even more preferably ≤ 2%, even more preferably ≤ 1.5%, and most preferably ≤ 1% or ≤ 0.5% or even 0%.

[0276] Preferably, the bispecific antibody construct of the present invention exists with very low dimerization after multiple freeze / thaw cycles. For example, the antibody construct monomer is adjusted to a concentration of 250 µg / ml in, for example, a universal formulation buffer, and subjected to three freeze / thaw cycles (freezing at -80°C for 30 min, followed by thawing at room temperature for 30 min), followed by high-performance SEC to determine the percentage of the original monomeric antibody construct that has been converted into a dimer antibody construct. Preferably, for example, after three freeze / thaw cycles, the dimerization percentage of the bispecific antibody construct is ≤ 5%, more preferably ≤ 4%, even more preferably ≤ 3%, even more preferably ≤ 2.5%, even more preferably ≤ 2%, even more preferably ≤ 1.5%, and most preferably ≤ 1% or even ≤ 0.5%.

[0277] The bispecific antibody constructs of the present invention preferably exhibit favorable thermal stability with aggregation temperatures ≥ 45°C or ≥ 50°C, more preferably ≥ 52°C or ≥ 54°C, even more preferably ≥ 56°C or ≥ 57°C, and most preferably ≥ 58°C or ≥ 59°C. The thermal stability parameter can be determined based on the antibody aggregation temperature as follows: An antibody solution with a concentration of 250 µg / ml is transferred to a disposable cuvette and placed in a dynamic light scattering (DLS) device. The sample is heated from 40°C to 70°C at a constant heating rate of 0.5°C / min, and the measured radius is obtained. The antibody aggregation temperature is calculated using the increase in radius indicating the melting of the protein and aggregate.

[0278] Alternatively, the intrinsic biophysical protein stability of the antibody construct can be determined by temperature melting profiles using differential scanning calorimetry (DSC). These experiments were performed using a VP-DSC device from MicroCal LLC (Northampton, MA, USA). Energy uptake of samples containing the antibody construct was recorded from 20°C to 90°C compared to samples containing only the formulation buffer. The antibody construct was adjusted to a final concentration of 250 μg / ml, for example, in SEC running buffer. To record the corresponding melting profiles, the sample temperature was gradually increased. At each temperature T, the energy uptake of the sample and the formulation buffer reference was recorded. The difference between the energy uptake Cp (kcal / mol / °C) of the sample and the reference was plotted against the corresponding temperature. The melting temperature was defined as the temperature at which the first maximum energy uptake occurred.

[0279] It is also envisioned that the target cell surface antigen xCD3 bispecific antibody construct of the present invention has a turbidity of ≤ 0.2, preferably ≤ 0.15, more preferably ≤ 0.12, even more preferably ≤ 0.1 and most preferably ≤ 0.08 (as measured by OD340 after concentrating the purified monomeric antibody construct to 2.5 mg / ml and incubating overnight).

[0280] Furthermore, it is envisioned that the bispecific antibody constructs of the present invention exhibit therapeutic efficacy or antitumor activity. This can be evaluated, for example, in studies disclosed in examples of advanced human tumor xenograft models:

[0281] Those skilled in the art know how to modify or adjust certain parameters of the study, such as the number of injected tumor cells, the injection site, the number of transplanted human T cells, the amount of bispecific antibody construct to be applied, and the timeline, while still obtaining meaningful and reproducible results. Preferably, the tumor growth inhibition T / C [%] is ≤ 70 or ≤ 60, more preferably ≤ 50 or ≤ 40, even more preferably ≤ 30 or ≤ 20, and most preferably ≤ 10 or ≤ 5 or even ≤ 2.5.

[0282] In a preferred embodiment of the antibody construct of the present invention, the antibody construct is a single-chain antibody construct.

[0283] Furthermore, in a preferred embodiment of the antibody construct of the present invention, the third structural domain comprises, in the order of amino to carboxyl groups:

[0284] Hinge-CH2-CH3-Joint-Hinge-CH2-CH3.

[0285] Additionally, in one embodiment of the invention, one or preferably each (two) polypeptide monomers' CH2 domains of the third domain contain an intradomain cysteine ​​disulfide bridge. As is known in the art, the term "cysteine ​​disulfide bridge" refers to a structure having the general formula... R–S–S–R The functional group. This bond is also called an SS bond or a disulfide bridge, and is derived by coupling two thiol groups of a cysteine ​​residue. For the antibody construct of the present invention, it is particularly preferred that the cysteine ​​that forms the cysteine ​​disulfide bridge in the mature antibody construct be introduced into the amino acid sequence corresponding to the CH2 domains of 309 and 321 (Kabat number).

[0286] In one embodiment of the invention, the glycosylation site at Kabat position 314 of the CH2 domain is removed. Preferably, the glycosylation site is removed by N314X substitution, where X is any amino acid other than Q. The substitution is preferably N314G substitution. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (according to the Kabat position): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0287] Preferred features of the antibody construct of the present invention, compared to, for example, bispecific heterologous Fc antibody constructs known in the art, may particularly relate to the introduction of the aforementioned modifications into the CH2 domain. Therefore, for the construct of the present invention, it is preferred that the CH2 domain in the third domain of the antibody construct of the present invention contains intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321 and / or the glycosylation site at Kabat position 314 is removed by the aforementioned N314X substitution, preferably by N314G substitution.

[0288] In another preferred embodiment of the invention, the CH2 domain in the third domain of the antibody construct of the invention contains intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321, and the glycosylation site at Kabat position 314 is removed by N314G substitution.

[0289] In one embodiment, the present invention provides an antibody construct, wherein:

[0290] (182) The first domain contains two antibody variable domains, and the second domain contains two antibody variable domains;

[0291] (ii) The first domain contains an antibody variable domain, and the second domain contains two antibody variable domains;

[0292] (iii) The first domain contains two antibody variable domains, and the second domain contains one antibody variable domain; or

[0293] (iv) The first domain contains an antibody variable domain, and the second domain contains an antibody variable domain.

[0294] Therefore, the first and second domains can each be binding domains containing two antibody variable domains (such as VH and VL domains). Examples of such binding domains containing two antibody variable domains have been described above and include, for example, the Fv fragment, scFv fragment, or Fab fragment described above. Alternatively, one or both of these binding domains may contain only a single variable domain. Examples of such single-domain binding domains have been described above and include, for example, nanobodies or single-variable-domain antibodies containing only one variable domain, which can be VHH, VH, or VL that binds to antigens or epitopes independently of other V regions or domains.

[0295] In a preferred embodiment of the antibody construct of the present invention, the first and second domains are fused to the third domain via a peptide linker. Preferred peptide linkers have been described above and are characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or greater (e.g., 2 or 3). Particularly preferred linkers for the fusion of the first and second domains with the third domain are depicted in SEQ ID No: 1.

[0296] In a preferred embodiment, the antibody construct of the present invention is characterized by comprising, in the order from amino to carboxyl groups:

[0297] (a) First structural domain;

[0298] (b) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187-189;

[0299] I. Second structural domain;

[0300] (d) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID NO: 187, 188, 189, 195, 196, 197 and 198;

[0301] The first polypeptide monomer of the third structural domain;

[0302] (f) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID Nos: 191, 192, 193, and 194; and

[0303] (g) The second polypeptide monomer of the third structural domain.

[0304] In one aspect of the invention, the target cell surface antigen bound by the first domain is a tumor antigen, an antigen specific to an immune disorder, or a viral antigen. The term "tumor antigen," as used herein, can be understood as those antigens presented on tumor cells. These antigens can be presented on the cell surface having an extracellular portion, which is typically combined with transmembrane and cytoplasmic portions of the molecule. These antigens can sometimes be presented only by tumor cells and never by normal cells. Tumor antigens may be expressed only on tumor cells or may represent tumor-specific mutations compared to normal cells. In this case, they are referred to as tumor-specific antigens. More commonly, antigens are presented by both tumor cells and normal cells, and these are referred to as tumor-associated antigens. These tumor-associated antigens may be overexpressed compared to normal cells, or may be accessible to antibody binding in tumor cells due to the less compact structure of tumor tissue compared to normal tissue. Non-limiting examples of tumor antigens used in this article are CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, MUC17, CLDN18.2, CDH3, CD70, BCMA, and PSMA.

[0305] In the context of this invention, other target cell surface antigens specific to immune disorders include, for example, TL1A and TNF-α. These targets are preferably addressed by the bispecific antibody constructs of this invention, which are preferably full-length antibodies. In a highly preferred embodiment, the antibody of this invention is a heterologous IgG antibody.

[0306] In a preferred embodiment of the antibody construct of the present invention, the tumor antigen, preferably the tumor antigen, is selected from the group consisting of: CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, MUC17, CLDN18.2, CDH3, CD70, BCMA and PSMA.

[0307] In one aspect of the invention, the antibody construct comprises, in the order of amino to carboxyl groups:

[0308] (a) A first domain having an amino acid sequence selected from the group consisting of: SEQ ID No: 7, 8, 17, 27, 28, 37, 38, 39, 40, 41, 48, 49, 50, 51, 52, 59, 60, 61, 62, 63, 64, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 89, 90, 91, 92, 93, 100, 101, 102, 103, 104, 113, 114, 121, 122, 123, 1 24, 125, 131, 132, 133, 134, 135, 136, 143, 144, 145, 146, 147, 148, 149, 150, 151, 158, 159, 160, 161, 162, 163, 164, 165, 166, 173, 174, 175, 176, 177, 178, 179, 180, 181, 223, 235, and 246.

[0309] (b) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187-189;

[0310] I. A second structural domain having an amino acid sequence selected from the group consisting of: SEQ ID No: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 or SEQ ID NO: 202 of WO 2008 / 119567;

[0311] (d) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID No: 187, 188, 189, 195, 196, 197 and 198;

[0312] I. A first polypeptide monomer with a third structural domain, the first polypeptide monomer having a polypeptide sequence selected from the group consisting of: WO2017 / 134140 SEQ ID No: 17-24;

[0313] (f) A peptide linker having an amino acid sequence selected from the group consisting of: SEQ ID Nos: 191, 192, 193, and 194; and

[0314] (g) A second polypeptide monomer with a third domain having a polypeptide sequence selected from the group consisting of SEQ ID No: 17-24 of WO2017 / 134140.

[0315] In one aspect, the bispecific antibody construct of the present invention is characterized by having an amino acid sequence selected from the group consisting of the following and targeting the corresponding target cell surface antigen:

[0316] (a)SEQ ID No: 27, 28, 37 to 41; CD33

[0317] (b) Each of SEQ ID Nos. 48 to 52; EGFRvIII

[0318] (c) Each of SEQ ID No: 59 to 64; MSLN

[0319] (d) Each of the CDH19 in SEQ ID No: 71 to 82

[0320] (e) Each of DLL3 in SEQ ID No: 100 to 104

[0321] (f)SEQ ID NO: 7, 8, 17, 113 and 114CD19

[0322] (g) Each of the FLT3s in SEQ ID No: 89 to 93

[0323] (h) Each of the CDH3 in SEQ ID No: 121 to 125

[0324] (i) Each of the BCMAs in SEQ ID No: 132 to 136

[0325] (j) Each of the PSMAs in SEQ ID No: 143 to 151, 158 to 166 and 173 to 181

[0326] (k)SEQ ID NO:213MUC17

[0327] (l) Each of CLDN18.2 in SEQ ID NO: 225 and 237 and

[0328] (m)SEQ ID No:248CD70

[0329] The present invention further provides a polynucleotide / nucleic acid molecule encoding the antibody construct of the present invention. A polynucleotide is a biopolymer composed of 13 or more nucleotide monomers covalently bonded in a chain. DNA (such as cDNA) and RNA (such as mRNA) are examples of polynucleotides with different biological functions. A nucleotide is an organic molecule that acts as a monomer or subunit of a nucleic acid molecule such as DNA or RNA. The nucleic acid molecule or polynucleotide can be double-stranded or single-stranded, linear or circular. It is preferably contained in a vector, which is preferably contained in a host cell. The host cell, for example, can express the antibody construct after transformation or transfection with the vector or polynucleotide of the present invention. For this purpose, the polynucleotide or nucleic acid molecule is operatively linked to a control sequence.

[0330] The genetic code is a set of rules that translates the information encoded within genetic material (nucleic acids) into proteins. Biological decoding in living cells is accomplished by ribosomes linking amino acids in a sequence specified by mRNA, using tRNA molecules to carry the amino acids and read them out as three nucleotides at a time. This code defines how the sequence of these nucleotide triplets (called codons) specifies which amino acid will be added next during protein synthesis. With some exceptions, trinucleotide codons in a nucleic acid sequence specify a single amino acid. Because the vast majority of genes are encoded using the exact same codon, this particular codon is often called the canonical or standard genetic code. While the genetic code determines the protein sequence of a given coding region, other genomic regions can influence when and where these proteins are produced.

[0331] Furthermore, this invention provides vectors containing the polynucleotide / nucleic acid molecules of this invention. A vector is a nucleic acid molecule used as a medium for transferring (foreign) genetic material into cells. The term "vector" encompasses, but is not limited to, plasmids, viruses, granules, and artificial chromosomes. Generally, engineered vectors contain an origin of replication, a multiple cloning site, and a selectability marker. The vector itself is typically a nucleotide sequence, usually a DNA sequence containing an insert (transgenic gene) and a larger sequence that acts as the "backbone" of the vector. In addition to the transgenic insert and backbone, modern vectors may encompass other features such as promoters, genetic markers, antibiotic resistance, reporter genes, target sequences, and protein purification tags. Vectors called expression vectors (expression constructs) are particularly used for expressing transgenes in target cells and typically contain control sequences.

[0332] The term "control sequence" refers to the DNA sequence necessary for the expression of an operable coding sequence in a particular host organism. For example, control sequences applicable to prokaryotes include promoters, optional operon sequences, and ribosome-binding sides. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

[0333] When a nucleic acid is functionally related to another nucleic acid sequence, that nucleic acid is "operably linked." For example, if the DNA of a pre-sequence or secretory leader sequence is expressed as a pre-protein involved in polypeptide secretion, then the DNA of the pre-sequence or secretory leader sequence is operably linked to the DNA of the polypeptide; if a promoter or enhancer affects the transcription of a coding sequence, then the promoter or enhancer is operably linked to that sequence; or if the ribosome-binding side is positioned to facilitate translation, then the ribosome-binding side is operably linked to the coding sequence. Generally speaking, "operably linked" means that the linked DNA sequences are contiguous, and in the case of a secretory leader sequence, they are contiguous and in the reading phase. However, enhancers do not need to be contiguous. Ligation is accomplished by joining at a convenient restriction site. If such a site is not available, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

[0334] "Transfection" is the intentional introduction of nucleic acid molecules or polynucleotides (including vectors) into target cells. This term is primarily used for non-viral methods in eukaryotic cells. Transduction is generally used to describe virus-mediated transfer of nucleic acid molecules or polynucleotides. Transfection in animal cells typically involves opening transient pores or "holes" in the cell membrane to allow for the uptake of substances. Transfection can be performed using calcium phosphate, via electroporation, by cell extrusion, or by mixing cationic lipids with substances to create liposomes (which fuse with the cell membrane and store their cargo inside).

[0335] The term "transformation" is used to describe the non-viral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and into non-animal eukaryotic cells (including plant cells). Therefore, transformation is a genetic alteration in bacteria or non-animal eukaryotic cells that results from the direct uptake and subsequent incorporation of exogenous genetic material (nucleic acid molecules) from their surroundings via one or more cell membranes. Transformation can be achieved artificially. For transformation to occur, the cell or bacteria must be in a competent state, which may occur as a time-limited response to environmental conditions such as starvation and cell density.

[0336] Furthermore, this invention provides a host cell that has been transformed or transfected with the polynucleotide / nucleic acid molecule or vector of this invention. As used herein, the terms "host cell" or "recipient cell" are intended to include any single cell or cell culture that may be, or is already, a receptor for a vector, exogenous nucleic acid molecule, or a polynucleotide encoding an antibody construct of this invention, and / or a receptor for the antibody construct itself. The appropriate substance is introduced into the cell by transformation, transfection, or the like. The term "host cell" is also intended to include the progeny or potential progeny of a single cell. Because certain modifications may occur in the progeny due to natural, accidental, or intentional mutations or due to environmental influences, such progeny may not be completely identical to the parent cell (in morphology, genome, or complete DNA complement), but are still included within the scope of the terminology used herein. Suitable host cells include prokaryotic or eukaryotic cells, and also include, but are not limited to, bacterial, yeast, fungal, plant, and animal cells such as insect cells and mammalian cells such as mouse, rat, macaque, or human cells.

[0337] The antibody constructs of the present invention can be produced in bacteria. After expression, the antibody constructs of the present invention are separated from E. coli cell paste as soluble fractions and can be purified by, for example, affinity chromatography and / or size exclusion. The final purification can be performed in a manner similar to that used for purifying antibodies expressed, for example, in CHO cells.

[0338] Besides prokaryotes, eukaryotic microorganisms (such as filamentous fungi or yeast) are suitable cloning or expression hosts for the antibody constructs of this invention. Saccharomyces cerevisiae (Saccharomyces cerevisiae) Saccharomyces cerevisiae *Saccharomyces cerevisiae* or common baker's yeast are the most commonly used among microorganisms of lower eukaryotic hosts. However, many other genera, species, and strains are generally available and can be used in this article, such as *Schizosaccharomyces cerevisiae*. Schizosaccharomyces pombe), Kluyveromyce hosts, such as Lactobacillus lactis ( K. Kluyveromyces lactis, K. fragilis (ATCC 12424), Kluyveromyces bulgaricus ( K. bulgaricus (ATCC 16045), Wickkluwer yeast ( K. wickeramii (ATCC 24178), Valtiruvi yeast ( K. Waltii (ATCC 56500), Kluyveromyces waltii ( K. drosophilarum (ATCC36906), heat-resistant Kluyveromyces ( K. thermotolerans and Max Kluyveromyces ( K.marxianus); Yersinia (EP 402 226); Pichia pastoris (EP 183 070); Candida; Trichoderma reesei (EP 244 234); Neurospora crassa; Schwanniomyces, such as Schwanniomyces serrata ( Schwanniomyces occidentalis); and filamentous fungi, such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts, such as Aspergillus nidus ( A. nidulans) and Aspergillus niger ( A. niger).

[0339] Suitable host cells for expressing the glycosylated antibody constructs of the present invention are derived from multicellular organisms. Examples of invertebrate cells include plant cells and insect cells. Cells derived from organisms such as the fall armyworm (Pterocarya stenoptera) have been identified. Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito) Aedes Aedes albopictus (mosquito), Aedes albopictus (mosquito), and Drosophila melanogaster (mosquito). Drosophila melanogaster (fruit fly) and silkworm ( Bombyx Many baculovirus strains and variants of *Mori*, as well as corresponding permitted insect host cells, are available for transfection. Several viral strains used for transfection are publicly available, such as those from *Alfalfa Silver-striped Noctuidae* (*Mori*). Autographa The L-1 variant of californica NPV and the Bm-5 strain of silkworm NPV, and according to the present invention, such viruses can be used as the viruses described herein, particularly for transfecting fall armyworm cells.

[0340] Plant cell cultures of cotton, maize, potato, soybean, petunia, tomato, Arabidopsis, and tobacco can also be used as hosts. Cloning and expression vectors that can be used to produce proteins in plant cell cultures are known to those skilled in the art. See, for example, Hiatt et al., Nature (1989) 342: 76-78; Owen et al. (1992) Bio / Technology 10: 790-794; Artsaenko et al. (1995) The Plant Journal 8: 745-750; and Fecker et al. (1996) Plant Mol Biol 32: 979-986.

[0341] However, the greatest interest is in vertebrate cells, and the propagation of vertebrate cells in cultures (tissue cultures) has become a routine procedure. Examples of useful mammalian host cell lines include the monkey kidney CV1 line transformed from SV40 (COS-7, ATCC CRL1651); the human embryonic kidney line (293 cells or subclones for growth in suspension culture, Graham et al., J. Gen Virol. [Journal of General Virology] 36: 59 (1977)); juvenile hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences] 77: 4216 (1980)); mouse saturated hamster cells (TM4, Mather, Biol. Reprod. [Reproductive Biology] 23: 243-251 (1980)); monkey kidney cells (CVI ATCC CCL 70); and African green monkey kidney cells (VERO-76, ATCC). CRL1587); human cervical cancer cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL34); Buffalo rat hepatocytes (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human hepatocytes (Hep G2, 1413 8065); mouse mammary tumor (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals N. Y Acad. Sci. [Annals of the New York Academy of Sciences] (1982) 383: 44-68); MRC 5 cells; FS4 cells; and human hepatocellular carcinoma cell line (Hep G2).

[0342] In another embodiment, the present invention provides a method for generating antibody constructs of the present invention, the method comprising culturing host cells of the present invention under conditions that allow expression of antibody constructs of the present invention and recovering the resulting antibody constructs from the culture.

[0343] As used herein, the term "culture" refers to the in vitro maintenance, differentiation, growth, proliferation, and / or reproduction of cells in a culture medium under suitable conditions. The term "expression" includes any step involved in generating the antibody constructs of the present invention, including but not limited to transcription, post-transcriptional modification, translation, post-translational modification, and secretion.

[0344] When using recombinant technology, antibody constructs can be generated intracellularly in the periplasmic space or secreted directly into the culture medium. If the antibody constructs are generated intracellularly, particulate debris from the host cell or lysed fragments is removed as a first step, for example, by centrifugation or ultrafiltration. Carter et al., Bio / Technology 10: 163-167 (1992) describes a procedure for isolating antibodies secreted into the periplasmic space of *E. coli*. Briefly, the cell paste is thawed in the presence of sodium acetate (pH 3.5), EDTA, and benzyl sulfonyl fluoride (PMSF) over approximately 30 min. Cell debris can be removed by centrifugation. In cases where antibodies are secreted into the culture medium, the supernatant from such expression systems is typically concentrated first using a commercially available protein concentrator, such as an Amicon or Millipore Pellicon ultrafiltration unit. Any of the foregoing steps may include a protease inhibitor (such as PMSF) to inhibit proteolysis, and may include antibiotics to prevent the growth of foreign contaminants.

[0345] The antibody constructs of the present invention prepared from host cells can be recovered or purified using, for example, hydroxyapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography. Depending on the antibody to be recovered, other techniques for protein purification can also be used, such as fractionation on an ion-exchange column, ethanol precipitation, reversed-phase HPLC, chromatography on silica, and chromatography on heparin (SEPHAROSE). TM Chromatography performed on an anion or cation exchange resin (such as a polyaspartic acid column), chromatographic focusing, SDS-PAGE, and ammonium sulfate precipitation are all possible methods. When the antibody construct of the present invention contains a CH3 domain, Bakerbond ABX resin (JT Baker, Phillipsburg, NJ) can be used for purification.

[0346] Affinity chromatography is the preferred purification technique. The most common matrix for affinity ligand attachment is agarose, but other matrices are also available. Mechanically stable matrices such as controllably porous glass or poly(divinyl styrene)benzene allow for faster flow rates and shorter processing times than achievable with agarose.

[0347] Furthermore, the present invention provides pharmaceutical compositions comprising the antibody construct of the present invention or an antibody construct produced by the method according to the present invention. Preferably, the homogeneity of the antibody construct in the pharmaceutical compositions of the present invention is ≥ 80%, more preferably ≥ 81%, ≥ 82%, ≥ 83%, ≥ 84% or ≥ 85%, further preferably ≥ 86%, ≥ 87%, ≥ 88%, ≥ 89% or ≥ 90%, even more preferably ≥ 91%, ≥ 92%, ≥ 93%, ≥ 94% or ≥ 95%, and most preferably ≥ 96%, ≥ 97%, ≥ 98% or ≥ 99%.

[0348] As used herein, the term "pharmaceutical composition" refers to a composition suitable for administration to a patient, preferably a human patient. Particularly preferred pharmaceutical compositions of the present invention comprise a preferred therapeutically effective amount of one or more antibody constructs of the present invention. Preferably, the pharmaceutical composition further comprises a suitable formulation of one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives, and / or adjuvants. The acceptable components of the composition are preferably non-toxic to the recipient at the dose and concentration used. Pharmaceutical compositions of the present invention include, but are not limited to, liquid, freeze-dried, and lyophilized compositions.

[0349] The compositions of this invention may comprise pharmaceutically acceptable carriers. Generally, as used herein, "pharmaceutically acceptable carrier" means any and all aqueous and non-aqueous solutions, sterile solutions, solvents, buffers (e.g., phosphate-buffered saline (PBS) solutions), water, suspensions, emulsions (e.g., oil / water emulsions), wetting agents of various species, liposomes, dispersion media, and coatings compatible with drug administration, particularly parenteral administration. The use of such media and agents in pharmaceutical compositions is well known in the art, and compositions comprising such carriers can be formulated by well-known conventional methods.

[0350] Some embodiments provide pharmaceutical compositions comprising the antibody construct of the present invention and one or more additional excipients, such as those illustratively described in this section and elsewhere herein. In this respect, excipients can be used in the present invention for a variety of purposes, such as adjusting the physical, chemical, or biological properties of formulations, such as adjusting viscosity and / or the methods of the present invention to improve efficacy and / or stabilize such formulations and methods against degradation and spoilage caused, for example, by stresses occurring during manufacturing, transport, storage, pre-use preparation, application, and post-treatment.

[0351] In some embodiments, the pharmaceutical composition may contain formulation substances intended to alter, maintain, or preserve aspects of the composition such as pH, osmotic pressure, viscosity, transparency, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or permeability (see REMINGTON'S PHARMACEUTICAL SCIENCES, 18th edition, edited by AR Genrmo, 1990, Mack Publishing Company). In such embodiments, suitable formulation substances may include, but are not limited to:

[0352] • Amino acids, such as glycine, alanine, glutamine, asparagine, threonine, proline, 2-phenylalanine, including charged amino acids, preferably lysine, acetylsine, arginine, glutamate and / or histidine.

[0353] • Antimicrobial agents, such as antibacterial and antifungal agents

[0354] • Antioxidants, such as ascorbic acid, methionine, sodium sulfite, or sodium bisulfite;

[0355] • Buffers, buffering systems, and buffers used to maintain a composition at or slightly below a physiological pH; examples of buffers are borates, bicarbonates, Tris-HCl, citrates, phosphates or other organic acids, succinates, phosphates, and histidines; for example, Tris buffers with a pH of approximately 7.0–8.5;

[0356] • Non-aqueous solvents, such as propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate;

[0357] • Aqueous carriers include water, alcohol / aqueous solutions, emulsions, or suspensions, including brine and buffered media;

[0358] • Biodegradable polymers, such as polyester;

[0359] • Expanding agents, such as mannitol or glycine;

[0360] • Chelating agents, such as ethylenediaminetetraacetic acid (EDTA);

[0361] •Isotonics and absorption delay agents;

[0362] • Complexing agents (such as caffeine, polyvinylpyrrolidone, β-cyclodextrin or hydroxypropyl-β-cyclodextrin);

[0363] • Filler;

[0364] • Monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose, or dextrin); the carbohydrates may be non-reducing sugars, preferably trehalose, sucrose, octetrate, sorbitol, or xylitol;

[0365] • (Low molecular weight) proteins, peptides or protein carriers, such as human or bovine serum albumin, gelatin or immunoglobulins, preferably of human origin;

[0366] • Coloring agents and flavorings;

[0367] • Sulfur-containing reducing agents, such as glutathione, lipoic acid, sodium thioglycolate, thioglycerol, [α]-monothioglycerol, and sodium thiosulfate.

[0368] • Diluent;

[0369] • Emulsifiers;

[0370] • Hydrophilic polymers, such as polyvinylpyrrolidone;

[0371] • Salt formation against counterions, such as sodium;

[0372] • Preservatives, such as antimicrobial agents, antioxidants, chelating agents, inert gases, etc.; examples are: benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide.

[0373] • Metal complexes, such as Zn-protein complexes;

[0374] • Solvents and co-solvents (such as glycerol, propylene glycol, or polyethylene glycol);

[0375] • Sugars and sugar alcohols, such as trehalose, sucrose, octasulfate, mannitol, sorbitol or xylitol, stachyose, mannose, sorbitol, xylose, ribose, myoinisitose, galactose, lactitol, ribitol, myoinisitol, galactitol, glycerol, cyclic polyols (e.g., inositol), polyethylene glycol; and polyols;

[0376] • Suspension agent;

[0377] • Surfactants or wetting agents (such as pluronics, PEG, sorbitan, polysorbates (such as polysorbate 20, polysorbate), tritium nuclei, tromethamine, lecithin, cholesterol, tyloxapal)); the surfactant may be a detergent, preferably with a molecular weight > 1.2 KD, and / or a polyether, preferably with a molecular weight > 3 KD; preferred non-limiting examples of detergents are Tween 20, Tween 40, Tween 60, Tween 80 and Tween 85; preferred non-limiting examples of polyethers are PEG 3000, PEG 3350, PEG 4000 and PEG 5000;

[0378] • Stabilizers, such as sucrose or sorbitol;

[0379] • Tension enhancers (such as alkali metal halides, preferably sodium chloride or potassium chloride, mannitol or sorbitol);

[0380] • Parenteral delivery media, including sodium chloride solution, Ringer's dextran, dextran and sodium chloride, lactated Ringer's solution or non-volatile oils;

[0381] • Intravenous delivery mediators, including fluid and nutritional supplements, and electrolyte supplements (such as those based on Ringer's dextran).

[0382] It will be apparent to those skilled in the art that, for example, different components of a pharmaceutical composition (e.g., those listed above) can have different effects, and amino acids can act as buffers, stabilizers, and / or antioxidants; mannitol can act as swelling agents and / or tension enhancers; sodium chloride can act as a delivery medium and / or tension enhancer; and so on.

[0383] In addition to the polypeptides defined herein, the compositions of the present invention are envisioned to contain other bioactive agents, depending on the intended use of the composition. Such agents may be drugs acting on the gastrointestinal system, drugs acting as cell inhibitors, drugs preventing hyperuricemia, drugs suppressing immune responses (e.g., corticosteroids), drugs modulating inflammatory responses, drugs acting on the circulatory system, and / or agents known in the art such as cytokines. The use of the antibody constructs of the present invention in co-therapies, i.e., in combination with another anticancer drug, is also envisioned.

[0384] In some embodiments, the optimal pharmaceutical composition will be determined by those skilled in the art based on, for example, the intended route of administration, delivery method, and desired dosage. See, for example, REMINGTON'S PHARMACEUTICAL SCIENCES, ibid. In some embodiments, such compositions can affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the antibody constructs of the present invention. In some embodiments, the primary mediator or carrier in the pharmaceutical composition may be aqueous or non-aqueous in nature. For example, suitable mediators or carriers may be water for injection, physiological saline solution, or artificial cerebrospinal fluid, possibly supplemented with other substances common in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are other exemplary mediators. In some embodiments, the antibody construct compositions of the present invention can be prepared for storage by mixing selected components having the desired purity with optional formulations (REMINGTON'S PHARMACEUTICAL SCIENCES, ibid.) in the form of lyophilized cakes or aqueous solutions. Furthermore, in some embodiments, the antibody constructs of the present invention can be formulated into lyophilized products using appropriate excipients (such as sucrose).

[0385] When considering parenteral administration, the therapeutic compositions of the present invention can be provided as pyrogen-free, parenterally acceptable aqueous solutions containing the desired antibody construct of the present invention in a pharmaceutically acceptable medium. Particularly suitable media for parenteral injection are sterile distilled water, in which the antibody construct of the present invention is formulated into a suitably preserved sterile isotonic solution. In some embodiments, the preparation may involve formulating the desired molecule with an agent that can provide controlled or sustained release of a product (such as injectable microspheres, bio-erosive particles, polymeric compounds such as polylactic acid or polyglycolic acid), beads, or liposomes that can be delivered via reservoir injection. In some embodiments, hyaluronic acid, which has the effect of promoting circulation duration, may also be used. In some embodiments, an implantable drug delivery device may be used to introduce the desired antibody construct.

[0386] Other pharmaceutical compositions will be apparent to those skilled in the art, including formulations involving the antibody constructs of the present invention in sustained or controlled delivery / release formulations. Techniques for formulating various other sustained or controlled delivery methods (such as liposome carriers, bioerectable microparticles or porous beads and reservoir injections) are also known to those skilled in the art. See, for example, International Patent Application No. PCT / US93 / 00829, which describes the controlled release of porous polymer microparticles for delivering pharmaceutical compositions. Sustained-release formulations may include a semi-permeable polymer matrix in the form of a molded article (e.g., a membrane or microcapsule). Sustained-release matrices may include polyesters, hydrogels, polylactide (as disclosed in U.S. Patent No. 3,773,919 and European Patent Application Publication No. EP 058481), copolymers of L-glutamic acid and γ-L-glutamic acid ethyl ester (Sidman et al., 1983, Biopolymers 2:547-556), poly(2-hydroxyethyl-methacrylate) (Langer et al., 1981, J. Biomed. Mater. Res. 15:167-277 and Langer, 1982, Chem. Tech. 12:98-105), ethylene vinyl acetate (Langer et al., 1981, ibid.), or poly-D(-)-3-hydroxybutyric acid (European Patent Application Publication No. EP 133,988). Sustained-release compositions may also include liposomes prepared by any of several methods known in the art. See, for example, Eppstein et al., 1985, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences of the United States of America] 82:3688-3692; European Patent Application Publication Nos. EP 036,676; EP 088,046 and EP 143,949.

[0387] Antibody constructs can also be embedded in microcapsules prepared by, for example, coagulation techniques or interfacial polymerization (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly(methyl methacrylate) microcapsules, respectively), in colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules), or in coarse-drop emulsions. Such techniques are disclosed in Remington's Pharmaceutical Sciences, 16th edition, edited by Oslo, A. (1980).

[0388] Pharmaceutical compositions intended for internal administration are typically provided as sterile formulations. Sterilization can be achieved through filtration using a sterile filter membrane. When the composition is lyophilized, sterilization can be performed using this method before or after lyophilization and reconstitution. Compositions intended for parenteral administration can be stored in lyophilized form or in solution. Parenteral compositions are typically placed in containers with sterile access ports (e.g., intravenous solution bags or vials with stoppers that can be punctured by a hypodermic needle).

[0389] Another aspect of the invention includes self-buffered antibody construct formulations of the invention, which can be used as pharmaceutical compositions, as described in International Patent Application WO 06138181A2 (PCT / US2006 / 022599). Useful protein stabilization and formulation materials and methods in this regard are available in a wide variety of descriptions, such as Arakawa et al., “Solvent interactions in pharmaceutical formulations,” Pharm Res. 8(3): 285-91 (1991); Kendrick et al., “Physical stabilization of proteins in aqueous solution,” in RATIONAL DESIGN OF STABLE PROTEIN FORMULATIONS: THEORY AND PRACTICE, edited by Carpenter and Manning in Pharmaceutical Biotechnology 13: 61-84 (2002); and Randolph et al., “Surfactant-protein interactions,” Pharm Biotechnol. 13: 159-75. (2002), with particular reference to the relevant sections concerning excipients and methods for self-buffered protein formulations according to the invention, especially concerning protein pharmaceutical products and methods for veterinary and / or human medical use.

[0390] According to certain embodiments of the invention, salts can be used to, for example, adjust the ionic strength and / or isotonicity of the formulation and / or improve the solubility and / or physical stability of proteins or other components of the compositions according to the invention. It is well known that ions can stabilize the native state of proteins by binding to charged residues on the surface of proteins and by shielding charged and polar groups in proteins and reducing the strength of their electrostatic interactions, attractive and repulsive interactions. Ions can also stabilize the denatured state of proteins, particularly by binding to denatured peptide bonds (--CONHs). Furthermore, ionic interactions with charged and polar groups in proteins can reduce intermolecular electrostatic interactions and thus prevent or reduce protein aggregation and insolubility.

[0391] Ions have significantly different effects on proteins. Several classification and rating systems have been developed for ions that can be used to formulate pharmaceutical compositions according to the present invention and their effects on proteins. One example is the Hofmeister series, which rates ions and polar nonionic solutes based on their effect on the conformational stability of proteins in solution. Stable solutes are referred to as “lyophilic.” Unstable solutes are referred to as “agnostic.” High concentrations of lyophilic agents (e.g., >1 mol ammonium sulfate) are typically used to precipitate proteins from solution (“salting out”). Agnostic agents are typically used to denature and / or dissolve proteins (“salting in”). The relative effectiveness of ions in “salting in” and “salting out” defines their position in the Hofmeister series.

[0392] According to various embodiments of the present invention, free amino acids can be used in the antibody construct formulations of the present invention as swelling agents, stabilizers, antioxidants, and other standard uses. Lysine, proline, serine, and alanine can be used to stabilize proteins in the formulation. Glycine can be used for lyophilization to ensure proper cake structure and properties. In both liquid and lyophilized formulations, arginine can be used to inhibit protein aggregation. Methionine can be used as an antioxidant.

[0393] Polyols include sugars such as mannitol, sucrose, and sorbitol, as well as polyols such as glycerol and propylene glycol, and for the purposes of this discussion, polyethylene glycol (PEG) and related substances. Polyols are lyophilic. They are useful stabilizers in both liquid and lyophilized formulations to protect proteins from physical and chemical degradation processes. Polyols can also be used to adjust the stress of formulations. In selected embodiments of the invention, mannitol is a useful polyol, which is commonly used to ensure the structural stability of the cake in lyophilized formulations. It ensures the structural stability of the cake. It is often used in conjunction with lyophilization protectants such as sucrose. Sorbitol and sucrose are preferred agents for adjusting stress and acting as stabilizers to prevent freeze-thaw stress during transport or to prevent the formation of clumps during manufacturing. Reducing sugars (containing free aldehyde or ketone groups), such as glucose and lactose, can glycosylate surface lysine and arginine residues. Therefore, they are generally not preferred polyols for use according to the invention. Furthermore, sugars that form such reactive substances are not preferred polyols in this invention, such as sucrose, which hydrolyzes under acidic conditions to fructose and glucose, thus resulting in glycosylation. PEG can be used to stabilize proteins and as a cryoprotectant, and can be used in this invention.

[0394] Examples of antibody construct formulations of the present invention further include surfactants. Protein molecules can readily adsorb onto surfaces and denature, subsequently accumulating at air-liquid, solid-liquid, and liquid-liquid interfaces. These effects are generally inversely proportional to protein concentration. These detrimental interactions are generally inversely proportional to protein concentration and are typically exacerbated by physical oscillations, such as those generated during product transport and handling. Surfactants are conventionally used to prevent, minimize, or reduce surface adsorption. Surfactants useful in this regard include polysorbate 20, polysorbate 80, other fatty acid esters of dehydrated sorbitol polyethoxylates, and poloxamer 188. Surfactants are also commonly used to control protein conformational stability. The use of surfactants in this respect is protein-specific, as any given surfactant typically stabilizes some proteins and destabilizes others.

[0395] Polysorbates are readily oxidatively degraded and are typically supplied in quantities sufficient to induce oxidation of protein residue side chains, particularly methionine. Therefore, polysorbates should be used with caution and at their lowest effective concentration. In this respect, polysorbates exemplify the general rule that excipients should be used at their lowest effective concentration.

[0396] Examples of antibody construct formulations of the present invention further comprise one or more antioxidants. Harmful oxidation of proteins in pharmaceutical formulations can be prevented to some extent by maintaining appropriate levels of ambient oxygen and temperature and avoiding exposure to light. Antioxidant excipients can also be used to prevent oxidative degradation of proteins. In this regard, useful antioxidants are reducing agents, oxygen / free radical scavengers, and chelating agents. Antioxidants used in therapeutic protein formulations according to the present invention are preferably water-soluble and maintain their activity throughout the shelf life of the product. In this regard, EDTA is a preferred antioxidant according to the present invention. Antioxidants can damage proteins. For example, reducing agents, such as glutathione in particular, can break intramolecular disulfide bonds. Therefore, the antioxidants selected for use in the present invention are particularly intended to eliminate or sufficiently reduce the possibility of themselves damaging proteins in the formulation.

[0397] Formulations according to the present invention may contain metal ions that are protein cofactors and are essential for the formation of protein coordination complexes, such as zinc, which is essential for the formation of certain insulin suspensions. Metal ions can also inhibit some processes that degrade proteins. However, metal ions also catalyze physical and chemical processes that degrade proteins. Magnesium ions (10-120 mM) can be used to inhibit the isomerization of aspartic acid to isoaspartic acid. +2 Ions (up to 100 mM) can increase the stability of human deoxyribonuclease. However, Mg +2 Mn +2 and Zn +2 This can destabilize rhDNase. Similarly, Ca... +2 and Sr +2 It can stabilize factor VIII, which can be due to Mg +2 Mn +2 and Zn +2 Cu +2 and Fe +2 To stabilize, and its aggregation can be achieved through Al +3 The number of ions increases.

[0398] Examples of antibody construct formulations of the present invention further comprise one or more preservatives. Preservatives are essential when the development involves multiple-dose parenteral formulations extracted from the same container more than once. Their primary function is to inhibit microbial growth and ensure the sterility of the product throughout its shelf life or intended use. Commonly used preservatives include benzyl alcohol, phenol, and m-cresol. While preservatives have a long history of use in the parenteral administration of small molecules, the development of protein formulations containing preservatives can be challenging. Preservatives almost always have an unstable effect (aggregation) on proteins, and this has become a major factor limiting their use in multi-dose protein formulations. To date, most protein drugs have been formulated for single-use only. However, when multi-dose formulations are possible, they offer the added advantages of patient convenience and increased marketability. A good example is human growth hormone (hGH), where the development of preservative formulations has led to the commercialization of more convenient, reusable injection pens. At least four such pen devices containing hGH preservative formulations are currently commercially available. Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech), and Genotropin (lyophilized, two-chamber cartridge, Pharmacia & Upjohn) contain phenol, while Somatrope (Eli Lilly) is formulated with m-cresol. Several aspects need to be considered during the formulation and development of preservative dosage forms. The effective preservative concentration in the drug product must be optimized. This requires testing the given preservative in the dosage form at concentration ranges that confer antimicrobial efficacy without compromising protein stability.

[0399] As can be expected, developing liquid formulations containing preservatives is more challenging than developing lyophilized formulations. Lyophilized products can be lyophilized without preservatives and reconstituted with a preservative-containing diluent upon use. This shortens the time the preservative is in contact with the protein, thus significantly minimizing the associated stability risks. In the case of liquid formulations, the effectiveness and stability of the preservative should be maintained throughout the product's shelf life (approximately 18 to 24 months). It is important to note that the effectiveness of the preservative should be demonstrated in the final formulation containing the active pharmaceutical ingredient and all excipient components.

[0400] The antibody constructs disclosed herein can also be formulated as liposomes. A “liposome” is a small vesicle composed of various types of lipids, phospholipids, and / or surfactants, which can be used to deliver drugs to mammals. The components of a liposome are typically arranged in a bilayer, similar to the lipid arrangement of a biological membrane. Liposomes containing antibody constructs are prepared by methods known in the art, such as Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688 (1985); Hwang et al., Proc. Natl. Acad. Sci. USA, 77: 4030 (1980); U.S. Patent Nos. 4,485,045 and 4,544,545; and WO 97 / 38731. Liposomes with extended circulation times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter with defined pore sizes to produce liposomes with the desired diameter. The Fab' fragment of the antibody construct of the present invention can be conjugated to liposomes via a disulfide exchange reaction, as described in Martin et al. J. Biol. Chem. 257: 286-288 (1982). Chemotherapy agents are optionally contained within the liposomes. See Gabizon et al. J. National Cancer Institute 81 (19) 1484 (1989).

[0401] Once a pharmaceutical composition is formulated, it can be stored in sterile vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations can be stored in ready-to-use form or in a form that can be reconstituted before application (e.g., lyophilized form).

[0402] The bioactivity of the pharmaceutical compositions defined herein can be determined, for example, by cytotoxicity assays, as described in the following examples, WO 99 / 54440, or by Schlereth et al. (Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy] 20 (2005), 1-12). As used herein, “efficacy” or “in vivo efficacy” means a response to treatment with the pharmaceutical compositions of the present invention using, for example, standardized NCI response criteria. The success of a therapy using the pharmaceutical compositions of the present invention or its in vivo efficacy refers to the effectiveness of the composition for its intended use, i.e., the composition’s ability to produce its desired effect, i.e., the depletion of pathological cells (e.g., tumor cells). In vivo efficacy can be monitored by established standard methods for the corresponding disease entity, including but not limited to white blood cell counts, differential cytometry, fluorescence-activated cell sorting, and bone marrow aspiration. In addition, various disease-specific clinical chemistry parameters and other established standard methods can be used. In addition, computer-assisted computed tomography, X-ray, and magnetic resonance imaging (MRI) can be used (e.g., for response assessment based on National Cancer Institute standards [Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM, Lister TA, Vose J, Grillo-Lopez A, Hagenbeek A, Cabanillas F, Klippensten D, Hiddemann W, Castellino R, Harris NL, Armitage JO, Carter W, Hoppe R, Canellos GP. Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol. [Journal of Clinical Oncology]). April 1999; 17(4):1244]), positron emission tomography (PET), white blood cell count, differential, fluorescence-activated cell sorting, bone marrow aspiration, lymph node biopsy / histology and various lymphoma-specific clinical chemistry parameters (e.g., lactate dehydrogenase) and other established standard methods.

[0403] Another major challenge in developing drugs (such as the pharmaceutical compositions of the present invention) is the predictable modulation of pharmacokinetic properties. To this end, pharmacokinetic profiles of candidate drugs can be established, i.e., profiles of pharmacokinetic parameters that affect the ability of a particular drug to treat a given condition. Pharmacokinetic parameters affecting the ability of a drug to treat a disease entity include, but are not limited to, half-life, distribution capacity, hepatic first-pass metabolism, and serum binding. The efficacy of a given agent can be affected by each of the parameters mentioned above. The contemplated features of the antibody constructs of the present invention provide for specific FC patterns they contain, such as differences in pharmacokinetic behavior. Targeting antibody constructs with extended half-life according to the present invention preferably exhibit a surprisingly increased in vivo retention time compared to the “canonical” non-HLE form of said antibody constructs.

[0404] "Half-life" refers to the time it takes for 50% of an administered drug to be eliminated through biological processes (such as metabolism and excretion). "First-pass metabolism" refers to the tendency of a drug to be metabolized during its first contact with the liver, i.e., its first passage through the liver. "Volume of distribution" refers to the degree to which a drug remains in various compartments of the body (such as intracellular and extracellular spaces, tissues, and organs) and its distribution within these compartments. "Serium binding" refers to the tendency of a drug to interact with and bind to serum proteins (such as albumin), leading to a reduction or loss of its biological activity.

[0405] Pharmacokinetic parameters also include bioavailability, lag time (T lag), Tmax, absorption rate, onset time, and / or Cmax for a given amount of drug administered. "Bioavailability" refers to the amount of drug in the blood compartment. "Lack time" refers to the time delay between drug administration and its detectability and measurability in blood or plasma. "Tmax" is the time after which the drug reaches its maximum blood concentration, and "Cmax" is the maximum blood concentration obtained with a given drug. The time required to reach the blood or tissue concentration of the drug required to achieve its biological effect is influenced by all parameters. Pharmacokinetic parameters of bispecific antibody constructs exhibiting cross-species specificity (which can be determined in preclinical animal testing in non-chimpanzee primates as outlined above) are also presented, for example, in the publications of Schlereth et al. (Cancer Immunol. Immunother. [Cancer Immunology Immunotherapy] 20 (2005), 1-12).

[0406] In a preferred aspect of the invention, the pharmaceutical composition is stable at about -20°C for at least four weeks. It is apparent from the accompanying examples that the quality of the antibody constructs of the present invention can be tested using different systems compared to the quality of corresponding prior art antibody constructs. These tests are understood to conform to the "ICH Harmonised Tripartite Guideline: Stability Testing of Biotechnological / Biological Products Q5C and Specifications: Test procedures and Acceptance Criteria for Biotech / Biological Products Q6B [ICH Tripartite Harmonized Guidelines: Stability Testing of Biotechnological / Biological Products Q5C and Specification: Test Procedures and Acceptance Criteria for Biotechnological / Biological Products Q6B], and are therefore selected to provide stability indicator curves to identify changes in the properties, purity, and potency of the product. The term purity is generally accepted as relative. Due to the effects of glycosylation, deamidation, or other heterogeneity, the absolute purity of biotechnological / biological products should typically be assessed using more than one method, and the derived purity value depends on the method. For stability testing purposes, purity testing should focus on the method used to identify degradation products.

[0407] To assess the quality of a pharmaceutical composition containing the antibody construct of the present invention, it can be analyzed, for example, by analyzing the content of soluble aggregates in solution (HMWS excluded according to size). Preferably, stability at about -20°C for at least four weeks is characterized by a content of less than 1.5% HMWS / preferably less than 1% HMWS.

[0408] Preferred formulations of antibody constructs for pharmaceutical compositions may, for example, comprise components of the formulation described below:

[0409] •Ingredients:

[0410] Potassium phosphate, L-arginine hydrochloride, trehalose, and polysorbate 80, at pH 6.0

[0411] Further examples for evaluating the stability of the antibody constructs of the present invention in pharmaceutical composition form are provided in the appended Examples 4-12. In those examples, embodiments of the antibody constructs of the present invention are tested under different stress conditions in different pharmaceutical formulations, and the results are compared with other extended half-life (HLE) forms of bispecific T-cell conjugating antibody constructs known in the art. Generally, it is envisioned that antibody constructs according to the present invention, provided with a specific FC mode, are typically more stable under a wide range of stress conditions (such as temperature and light stress) compared to antibody constructs with different HLE forms and those without any HLE forms (i.e., “canonical” antibody constructs). This temperature stability can involve both decreasing temperatures (below room temperature, including freezing) and increasing temperatures (above room temperature, including temperatures up to or above body temperature). As those skilled in the art will recognize, this improved stability with respect to stress, which is difficult to avoid in clinical practice, makes the antibody constructs safer because fewer degradation products occur in clinical practice. As a result, the increased stability implies increased safety.

[0412] One embodiment provides an antibody construct of the present invention or an antibody construct produced according to the method of the present invention for the prevention, treatment or relief of proliferative diseases, neoplastic diseases, viral diseases or immune disorders.

[0413] The formulations described herein are pharmaceutical compositions intended for use in patients in need of treating, alleviating, and / or preventing the pathological medical conditions described herein. The term "treatment" refers to both therapeutic and prophylactic (or preventative) measures. Treatment includes applying or administering the formulation to the body, isolated tissues, or cells of a patient suffering from a disease / disorder, symptoms of a disease / disorder, or a predisposition to a disease / disorder, with the aim of curing, restoring, alleviating, altering, remedying, relieving, improving, or influencing the disease, its symptoms, or the predisposition to a disease.

[0414] As used herein, the term "remission" means any improvement in the disease state of a patient suffering from a tumor or cancer or metastatic cancer as described below, achieved by administering an antibody construct according to the invention to a subject in need. Such improvement may also be considered as slowing or halting the progression of the patient's tumor or cancer or metastatic cancer. As used herein, the term "prevention" means preventing the occurrence or recurrence of a patient suffering from a tumor or cancer or metastatic cancer as specified below, achieved by administering an antibody construct according to the invention to a subject in need.

[0415] The term "disease" refers to any condition that would benefit from treatment with the antibody constructs or pharmaceutical compositions described herein. This includes both chronic and acute disorders or diseases, including pathological features that make mammals susceptible to the diseases under consideration.

[0416] A “vesicle” is an abnormal growth of tissue that usually, but not always, forms a lump. When it does form a lump, it is usually called a “tumor.” Vesicles or tumors can be benign, potentially malignant (precancerous), or malignant. Malignant vesicles are usually called cancer. They often invade and destroy surrounding tissues and can metastasize, meaning they spread to other parts of the body, tissues, or organs. Therefore, the term “metastatic cancer” covers metastases to tissues or organs other than the original tumor. Lymphoma and leukemia are lymphatic vesicles. For the purposes of this invention, they are also covered in the terms “tumor” or “cancer.”

[0417] The term "viral disease" describes a disease that results from a viral infection in a subject.

[0418] As used in this article, the term "immunological disorder" describes immunological disorders according to the common definitions of the term, such as autoimmune diseases, hypersensitivity reactions, and immunodeficiency.

[0419] In one embodiment, the present invention provides a method for treating or alleviating proliferative diseases, neoplastic diseases, viral diseases, or immune disorders, the method comprising the step of administering an antibody construct of the present invention or an antibody construct generated according to the method of the present invention to a subject in need.

[0420] The terms "subjects in need" or "those in need of treatment" include those who already have the disorder as well as those who need to prevent the disorder. Subjects in need, or "patients," include people and other mammalian subjects who receive preventative or therapeutic treatment.

[0421] The antibody constructs of the present invention are typically designed for specific routes and methods of administration, specific dosages and frequencies of administration, specific treatments for specific diseases, bioavailability, and duration of action. The components of the composition are preferably formulated at concentrations acceptable for the administration site.

[0422] Therefore, formulations and compositions can be designed according to the present invention for delivery via any suitable route of administration. In the context of the present invention, routes of administration include, but are not limited to, those mentioned above.

[0423] • Local routes (such as epidermal, inhalation, nose, eye, ear (auricular / aural), vagina, mucous membrane);

[0424] • Enteric routes (e.g., oral, gastrointestinal, sublingual, sublipal, buccal, rectal); and

[0425] • Extra-gastrointestinal routes (such as intravenous, intra-arterial, intra-bone, intramuscular, intracerebral, intraventricular, epidural, intrathecal, subcutaneous, intraperitoneal, extra-amniotic, intra-articular, intracardiac, intradermal, intralesional, intrauterine, intrabladder, intravitreal, percutaneous, intranasal, transmucosal, intrasynovial, intraluminal).

[0426] The pharmaceutical compositions and antibody constructs of the present invention are particularly suitable for parenteral administration, such as subcutaneous or intravenous delivery, such as by injection (e.g., pellet injection) or by infusion (e.g., continuous infusion). The pharmaceutical compositions can be administered using medical devices. Examples of medical devices for administering pharmaceutical compositions are described in U.S. Patent Nos. 4,475,196; 4,439,196; 4,447,224; 4,447,233; 4,486,194; 4,487,603; 4,596,556; 4,790,824; 4,941,880; 5,064,413; 5,312,335; 5,312,335; 5,383,851; and 5,399,163.

[0427] In particular, the present invention provides uninterrupted administration of suitable compositions. As a non-limiting example, uninterrupted or substantially uninterrupted (i.e., continuous) administration can be achieved through a small pump system worn by the patient for metering the inflow of a therapeutic agent into the patient's body. Pharmaceutical compositions comprising antibody constructs of the present invention can be administered using said pump system. Such pump systems are generally known in the art and typically rely on the periodic replacement of cartridges containing the therapeutic agent to be infused. When the cartridge in such a pump system is replaced, a temporary interruption may occur in the otherwise uninterrupted inflow of the therapeutic agent into the patient's body. In this case, the administration phase before and after cartridge replacement will still be considered within the meaning of pharmaceutical means, and the method of the present invention together constitutes one instance of "uninterrupted administration" of such a therapeutic agent.

[0428] Continuous or uninterrupted administration of the antibody construct of the present invention can be performed intravenously or subcutaneously via a fluid delivery device or a small pump system, which includes a fluid drive mechanism for displacing fluid from a reservoir and an actuation mechanism for actuating the drive mechanism. A pump system for subcutaneous administration may include a needle or cannula for penetrating the patient's skin and delivering a suitable composition into the patient's body. The pump system may be directly attached to or connected to the patient's skin independently of a vein, artery, or blood vessel, thereby allowing direct contact between the pump system and the patient's skin. The pump system may be connected to the patient's skin for 24 hours to several days. The pump system may be small in size and have a small-volume reservoir. As a non-limiting example, the reservoir volume of the suitable pharmaceutical composition to be administered may be from 0.1 to 50 ml.

[0429] Continuous application can also be performed transdermally via patches worn on the skin, and replaced at regular intervals. Those skilled in the art are familiar with patch systems suitable for drug delivery for this purpose. Notably, transdermal application is particularly suitable for uninterrupted application because the replacement of the first exhausted patch can advantageously be accomplished simultaneously with placing a new second patch, for example, immediately adjacent to the skin surface of the first exhausted patch, and just before the first exhausted patch is removed. There are no issues of flow interruption or battery failure.

[0430] If the pharmaceutical composition has been lyophilized, the lyophilized material should be reconstituted in a suitable liquid before administration. This reconstitution can be achieved in, for example, bacteriostatic water for injection (BWFI), physiological saline, phosphate-buffered saline (PBS), or the same formulation in which the protein was previously in contact with the lyophilized material.

[0431] The compositions of the present invention can be administered to subjects at a suitable dose, which can be determined, for example, through dose-escalation studies by administering increasing doses of the antibody constructs of the present invention exhibiting the cross-species specificity described herein to non-chimpanzee primates (e.g., rhesus monkeys). As described above, the antibody constructs of the present invention exhibiting the cross-species specificity described herein can advantageously be used in the same form for preclinical testing in non-chimpanzee primates and for use as a drug in humans. Dosing regimens will be determined by the attending physician and clinical factors. As is well known in the medical field, the dose for any given patient depends on many factors, including patient size, body surface area, age, the specific compound to be administered, sex, time and route of administration, general health condition, and other concurrently administered medications.

[0432] The term "effective dose" or "effective dosage" is defined as an amount sufficient to achieve or at least partially achieve the desired effect. The term "therapeutic effective dose" is defined as an amount sufficient to cure or at least partially stop the disease and its complications in a patient with an existing condition. The effective amount or dosage for this purpose will depend on the condition to be treated (indication), the antibody construct delivered, the treatment context and goals, the severity of the disease, prior therapy, the patient's clinical history and response to the therapeutic agent, the route of administration, the patient's body size (weight, body surface or organ size) and / or condition (age and general health status), and the general state of the patient's autoimmune system. Appropriate dosages may be adjusted according to the attending physician's judgment so that it can be administered to the patient once or in a series of administrations to achieve the best therapeutic effect.

[0433] Based on the factors mentioned above, typical dosage ranges can be from about 0.1 µg / kg to up to about 30 mg / kg or more. In specific embodiments, dosage ranges can be from 1.0 µg / kg to about 20 mg / kg, optionally from 10 µg / kg to about 10 mg / kg, or from 100 µg / kg to about 5 mg / kg.

[0434] The therapeutically effective amount of the antibody construct of the present invention preferably results in a reduction in the severity of disease symptoms, an increase in the frequency or duration of symptom-free periods, or prevention of damage or disability caused by disease suffering. For treating tumors expressing target cell antigens, the therapeutically effective amount of the antibody construct of the present invention, such as an anti-target cell antigen / anti-CD3 antibody construct, preferably inhibits cell growth or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% relative to untreated patients. The ability of the compound to inhibit tumor growth can be evaluated in animal models where efficacy is predicted.

[0435] The pharmaceutical compositions can be administered as a standalone therapeutic agent or in combination with other therapies, such as anticancer therapies as desired, such as other protein and non-protein drugs. These pharmaceuticals can be administered simultaneously with compositions comprising antibody constructs of the present invention as defined herein, or separately at time-defined intervals and dosages before or after the administration of the antibody construct.

[0436] As used herein, the term "effective and non-toxic dose" refers to a tolerable dose of the antibody construct of the present invention that is sufficiently high to induce pathological cell depletion, tumor elimination, tumor shrinkage, or disease stabilization without or substantially without major toxic effects. Such an effective and non-toxic dose can be determined, for example, by dose escalation studies described in the art, and should be below the dose that induces serious adverse side effects (dose-limiting toxicity, DLT).

[0437] As used herein, the term "toxicity" refers to the toxic effects of a drug as manifested in adverse events or serious adverse events. These adverse events may refer to a lack of systemic drug tolerance and / or a lack of local tolerance after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

[0438] As used herein, the terms “safety,” “in vivo safety,” or “tolerability” define the administration of a drug without immediately inducing serious adverse events (local tolerance) or without inducing serious adverse events over a longer period of drug administration. For example, “safety,” “in vivo safety,” or “tolerability” can be evaluated at regular intervals during treatment and follow-up periods. Measurements include clinical evaluations, such as organ manifestations, and screening for laboratory abnormalities. Clinical evaluations can be performed, and deviations from normal findings can be recorded / coded according to NCI-CTC and / or MedDRA standards. Organ manifestations can include criteria such as allergy / immunology, blood / bone marrow, arrhythmias, coagulation, etc., as described, for example, in the Common Terminology Standard for Adverse Events v3.0 (CTCAE). Laboratory parameters that can be tested include, for example, hematology, clinical chemistry, coagulation profiles, and urinalysis, as well as examinations of other body fluids (such as serum, plasma, lymph or cerebrospinal fluid, fluids, etc.). Therefore, safety can be assessed, for example, through physical examination, imaging techniques (i.e., ultrasound, X-ray, CT scan, magnetic resonance imaging (MRI), and other measures with technical devices (i.e., electrocardiography)), vital signs, by measuring laboratory parameters, and by recording adverse events. For example, adverse events in non-chimpanzee primates according to the uses and methods of the present invention can be examined by histopathological and / or histochemical methods.

[0439] The above terms are also mentioned in the following: for example, Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6; ICH Harmonised Tripartite Guideline; ICH Steering Committee meeting on July 16, 1997.

[0440] Finally, the present invention provides a kit comprising the antibody construct of the present invention or an antibody construct produced according to the method of the present invention, the pharmaceutical composition of the present invention, the polynucleotide of the present invention, the vector of the present invention, and / or the host cell of the present invention.

[0441] In the context of this invention, the term "kit" means two or more components (one of which corresponds to the antibody construct, pharmaceutical composition, vector, or host cell of this invention) packaged together in a container, receptor, or other container. Therefore, a kit can be described as a group of products and / or devices sufficient to achieve a particular objective, which can be sold as a single unit.

[0442] The kit may include one or more receivers (such as vials, ampoules, containers, syringes, pouches) having any suitable shape, size, and material (preferably waterproof, such as plastic or glass), which contain an appropriate dose of the antibody construct or pharmaceutical composition of the present invention (see above). The kit may additionally include instructions for use (e.g., in the form of a booklet or instruction manual), means for administering the antibody construct of the present invention (such as syringes, pumps, infusion sets, etc.), means for reconstructing the antibody construct of the present invention, and / or means for diluting the antibody construct of the present invention.

[0443]

[0444] The present invention also provides a kit for a single-dose administration unit. The kit of the present invention may further contain a first acceptor comprising a dried / lyophilized antibody construct and a second acceptor comprising an aqueous formulation. In some embodiments of the present invention, kits comprising single-compartment and multi-compartment pre-filled syringes (e.g., liquid syringes and lyophilized syringes) are provided.

[0445] It should be noted that, unless the context clearly indicates otherwise, as used herein, the singular forms “a,” “an,” and “the” include a plural of indicators. Thus, for example, a reference to “a reagent” includes one or more of such different reagents, and a reference to “the method” includes reference to equivalent steps and methods known to those skilled in the art that can modify or replace the methods described herein.

[0446] Unless otherwise indicated, the term "at least" preceding a series of elements should be understood to refer to each element in the series. Those skilled in the art will recognize or be able to determine many equivalents of the specific embodiments of the invention described herein using only conventional experimentation. Such equivalents are intended to be covered within the scope of this invention.

[0447] The term “and / or” as used herein includes the meaning of “and,” “or,” and “all or any other combination of elements connected by the term.”

[0448] As used herein, the term “about” or “approximately” means within 20%, preferably within 10%, and more preferably within 5% of a given value or range. However, it also includes specific numbers, such as about 20 including 20.

[0449] The terms “less than” or “greater than” include specific numbers. For example, “less than 20” means less than or equal to. Similarly, “more than” or “greater than” means more than or equal to / or greater than or equal to, respectively.

[0450] Throughout this specification and the claims, unless the context otherwise requires, the word "comprise" and its variations such as "comprises" or "comprising" shall be understood to implicitly include the said integer or step or group of integers or steps, but not exclude any other integer or step or group of integers or steps. When used herein, the term "comprise" may be replaced by the terms "containing" or "comprising," or sometimes by the term "having."

[0451] When used herein, "consisting of" excludes any element, step, or component not specified in the elements of the claim. When used herein, "consisting substantially of" does not exclude materials or steps that do not substantially affect the essential and novel features of the claim.

[0452] In each example in this article, any one of the terms “comprising / including,” “substantially consisting of,” and “consisting of” can be replaced by any of the other two terms.

[0453] It should be understood that the present invention is not limited to the specific methods, schemes, materials, reagents, and substances described herein, and therefore can be varied. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the invention as defined solely by the claims.

[0454] All publications and patents (including all patents, patent applications, scientific publications, manufacturers' specifications, instructions, etc.) referenced throughout this specification are incorporated herein by reference in their entirety, whether preceding or following. Nothing herein shall be construed as an admission that the invention is not entitled to any prior art as a result of such disclosures. Where material incorporated by reference conflicts or is inconsistent with this specification to any extent, this specification shall supersede any such material.

[0455] A better understanding of the invention and its advantages will be gained from the following examples, which are for illustrative purposes only. These examples are not intended to limit the scope of the invention in any way.

[0456] Example 1: Determining the optimal refolding conditions for the CD33 x CD3 BiTE® antibody construct

[0457] Materials and Methods: The CD33 x CD3 x scFc bispecific antibody construct (ScFc-BiTE®; SEQ ID NO: 41) was used in this study. Figure 1 The ScFc-BiTE® material was purified using the following steps. Protein A-filtered virus-inactivation pool (FVIP) and CEX HMW pool samples were then used for redox refolding.

[0458] Size exclusion chromatography (SEC) was used to determine the percentages of monomeric and aggregate substances in ScFc-BiTE®, under the following conditions: Column: G3000SWxl TOSOH Bioscience, 7.8 mm ID x 30 cm column (catalog number 08541), at room temperature or 25ºC. Mobile phase: 100 mM sodium phosphate, 250 mM NaCl, pH 6.8. Flow rate: 0.5 ml / min. Run time: 35 min. Unless otherwise specified, the refolding procedure was performed by stepwise dialysis using a Thermo Scientific Slide-A-Lyzer™ dialysis cartridge (20 kWh CO) at room temperature. After treatment with the refolding buffer, the sample was dialyzed back to 100 mM NaOAc, pH 5.1, overnight at 4ºC.

[0459] Results: A series of pH and guanidine concentrations were applied to increase the monomer concentration in the ScFc-BiTE® CEX post-peak cell, which contained only 25%–27% monomer. When the ScFc-BiTE® CEX post-peak sample cell was buffered with a control buffer (100 mM NaOAc, pH 5.1), the percentage of monomer remained at approximately 25%–27%. However, when the guanidine concentration in the refolding buffer was higher than 1 M and reached up to 2 M, the percentage of ScFc-BiTE® monomer increased in a concentration-dependent manner, up to 60% monomer. A detailed list of refolding conditions and the resulting percentage of monomer is provided in Table 4, and the effect of guanidine concentration on refolding is shown in [Table 4]. Figure 2 The SEC chromatograms are shown. To assess the effect of pH on refolding, refolding buffers containing 2.0M and 2.5M guanidine were adjusted to pH 4.2, 5.1, or 8.0 using acetic acid or Tris base. No significant difference was observed between pH 5.1 and pH 8.0. However, when incubated at pH 4.2, ScFc-BiTE® samples aggregated after the CEX peak (Table 4).

[0460] To further increase the percentage of monomers, redox reagents were added to the refolding buffer along with guanidine. When cysteine ​​and cysteine ​​were added to the 1.2 M guanidine refolding buffer at molar ratios of 6:1, 14:1, and 33:1 (Cys:Cyss), the percentage of monomers increased from 39% (without redox reagent) to 52%, 61%, and 69%, respectively (Table 4). Therefore, refolding treatment with Cys / Cyss warrants further investigation.

[0461] Oxidizing agents, such as copper(II) sulfate and dehydroascorbate, were evaluated. 0.5 mM copper(II) sulfate was found to have the greatest refolding effect when combined with 1.2 M or 2.0 M guanidine, increasing the monomer percentage to 90% and 88%, respectively (see Table 4). Figure 3 The percentage of monomers treated with 0.5 mM dehydroascorbate and 1.2 M and 2.0 M guanidine also increased to 66% and 73%, respectively.

[0462] To demonstrate that the refolding procedure reported above for the scFc-BiTE® CEX post-peak pool is also applicable to earlier downstream purification, such as the protein A pool, a protein A FVIP sample containing approximately 73% monomers and 27% aggregates was refolded using 2.0 M guanidine. At pH 5.1 and pH 8.0, the percentage of monomers increased to 94% and 89%, respectively (Table 5 and...). Figure 4 ).

[0463] Table 4 shows the refolding conditions and results of the downstream CEX HMW (post-peak) pool, which contained only 25% monomer and 75% high molecular weight material before the refolding treatment, and up to 89.6% after refolding treatment using the exemplary CD33xCD3 bispecific antibody according to the present invention.

[0464]

[0465] Table 5: Conditions and results of ScFc-BiTE® protein A FVIP pool refolding.

[0466]

[0467] Example 2: Determining the optimal refolding conditions for the DLL3 x CD3 antibody construct on a column.

[0468] On-column redox refolding of the DLL3 x CD3 x scFc (SEQ ID NO: 104) bispecific antibody construct was effectively examined. For this purpose, a ProA column with isocratic elution was used. The loading was 10 g / Lr (according to the DLL3xCD3 bispecific antibody construct method) HCCF. Wash buffer conditions were: 25 mM Tris + 1.2 M guanidine, 2 M guanidine, and copper(II) sulfate in 1.2 M guanidine*; control at 25 mM Tris + Arg*, all at pH 8.0. Wash with 2 column volumes of buffer; hold for 2 h; collect wash tanks separately. Elute protein at pH 3.7 (according to the DLL3xCD3 scFc-BiTE® method); sample. Virus inactivation: titrate to pH 3.6, hold for 60 min, neutralize to pH 5.1 with 2 M Tris base; sample. Tanks: wash tank; elution tank at pH 3.9; neutralized virus inactivation tank. The analytical tools used were: SEC, CHOP, and Leach ProA.

[0469] Example 3: Determining the conditions for successful refolding of CD19 and CLDN bispecific antibody constructs

[0470] At pH 8.0 and room temperature, the post-CEX pool of CD19 x CD3 x scFc (SEQ ID NO: 114) and CLND x CD3 x scFc antibody constructs was refolded for 4 hours with 2.0 M guanidine and 0.5 mM copper(II), after which the buffer was replaced with acetate buffer at pH 5.1. For the CD19 x CD3 x scFc antibody construct, the percentage of monomers increased from 24.0% to 48.7% (i.e., an increase of approximately 100%) (see [link to relevant documentation]). Figure 5 (and Table 6), and for the CLND x CD3 x scFc antibody construct, it increased from 23.4% to 38.5% (i.e., an increase of approximately 64%) (see Table 6). Figure 6 (See Table 7). In some cases, the application method of the present invention significantly reduced aggregate formation and increased the monomer percentage. Improved monomer structural homogeneity (reduced monomer isotypes) was also observed.

[0471] Table 6: Experimental conditions for CD19 x CD3 x scFc molecules

[0472]

[0473] Table 7: Experimental conditions for CLDN x CD3 x scFc molecules

[0474]

[0475] In addition, refolding experiments were conducted at pH 7.0. At pH 7.0, when incubated with 2.0 M guanidine and 0.5 mM copper(II), the monomeric isoform decreased from 10.2% to 0% (see Table 8).

[0476] Table 8: Experimental conditions for CD19 x CD3 x scFc molecules

[0477]

[0478] Table 9: Sequence List

[0479]

[0480]

[0481]

[0482]

[0483]

[0484]

[0485]

[0486]

[0487]

[0488]

[0489]

[0490]

[0491]

[0492]

[0493]

[0494]

[0495]

[0496]

[0497]

[0498]

[0499]

[0500]

[0501]

[0502]

[0503]

[0504]

[0505]

[0506]

[0507]

[0508]

[0509]

[0510]

[0511]

[0512]

[0513]

[0514]

[0515] sequence list <110> Amgen Inc. <120> Downstream processing of bispecific antibody constructs <130> 32243 / 53572A / PC <150> US 62 / 744,595 <151> 2018-10-11 <160> 248 <170> PatentIn version 3.5 <210> 1 <211> 15 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VL CDR1 <400> 1 Lys Ala Ser Gln Ser Val Asp Tyr Asp Gly Asp Ser Tyr Leu Asn 1 5 10 15 <210> 2 <211> 7 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VL CDR2 <400> 2 Asp Ala Ser Asn Leu Val Ser 1 5 <210> 3 <211> 9 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VL CDR3 <400> 3 Gln Gln Ser Thr Glu Asp Pro Trp Thr 1 5 <210> 4 <211> 5 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VH CDR1 <400> 4 Ser Tyr Trp Met Asn 1 5 <210> 5 <211> 17 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VH CDR2 <400> 5 Gln Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe Lys 1 5 10 15 Gly <210> 6 <211> 15 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VH CDR3 <400> 6 Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp Tyr 1 5 10 15 <210> 7 <211> 111 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VL <400> 7 Asp Ile Gln Leu Thr Gln Ser Pro Ala Ser Leu Ala Val Ser Leu Gly 1 5 10 15 Gln Arg Ala Thr Ile Ser Cys Lys Ala Ser Gln Ser Val Asp Tyr Asp 20 25 30 Gly Asp Ser Tyr Leu Asn Trp Tyr Gln Gln Ile Pro Gly Gln Pro Pro 35 40 45 Lys Leu Leu Ile Tyr Asp Ala Ser Asn Leu Val Ser Gly Ile Pro Pro 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Asn Ile His 65 70 75 80 Pro Val Glu Lys Val Asp Ala Ala Thr Tyr His Cys Gln Gln Ser Thr 85 90 95 Glu Asp Pro Trp Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys 100 105 110 <210> 8 <211> 124 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19 VH <400> 8 Gln Val Gln Leu Gln Gln Ser Gly Ala Glu Leu Val Arg Pro Gly Ser 1 5 10 15 Ser Val Lys Ile Ser Cys Lys Ala Ser Gly Tyr Ala Phe Ser Ser Tyr 20 25 30 Trp Met Asn Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Gln Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe 50 55 60 Lys Gly Lys Ala Thr Leu Thr Ala Asp Glu Ser Ser Ser Thr Ala Tyr 65 70 75 80 Met Gln Leu Ser Ser Leu Ala Ser Glu Asp Ser Ala Val Tyr Phe Cys 85 90 95 Ala Arg Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp 100 105 110 Tyr Trp Gly Gln Gly Thr Thr Val Thr Val Ser Ser 115 120 <210> 9 <211> 5 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VH CDR1 <400> 9 Arg Tyr Thr Met His 1 5 <210> 10 <211> 17 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VH CDR2 <400> 10 Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe Lys 1 5 10 15 Asp <210> 11 <211> 10 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VH CDR3 <400> 11 Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr 1 5 10 <210> 12 <211> 10 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VL CDR1 <400> 12 Arg Ala Ser Ser Ser Val Ser Tyr Met Asn 1 5 10 <210> 13 <211> 7 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VL CDR2 <400> 13 Asp Thr Ser Lys Val Ala Ser 1 5 <210> 14 <211> 9 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VL CDR3 <400> 14 Gln Gln Trp Ser Ser Asn Pro Leu Thr 1 5 <210> 15 <211> 119 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VH <400> 15 Asp Ile Lys Leu Gln Gln Ser Gly Ala Glu Leu Ala Arg Pro Gly Ala 1 5 10 15 Ser Val Lys Met Ser Cys Lys Thr Ser Gly Tyr Thr Phe Thr Arg Tyr 20 25 30 Thr Met His Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe 50 55 60 Lys Asp Lys Ala Thr Leu Thr Thr Asp Lys Ser Ser Ser Thr Ala Tyr 65 70 75 80 Met Gln Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr Trp Gly Gln Gly 100 105 110 Thr Thr Leu Thr Val Ser Ser 115 <210> 16 <211> 108 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD3 VL <400> 16 Val Asp Asp Ile Gln Leu Thr Gln Ser Pro Ala Ile Met Ser Ala Ser 1 5 10 15 Pro Gly Glu Lys Val Thr Met Thr Cys Arg Ala Ser Ser Ser Val Ser 20 25 30 Tyr Met Asn Trp Tyr Gln Gln Lys Ser Gly Thr Ser Pro Lys Arg Trp 35 40 45 Ile Tyr Asp Thr Ser Lys Val Ala Ser Gly Val Pro Tyr Arg Phe Ser 50 55 60 Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu Thr Ile Ser Ser Met Glu 65 70 75 80 Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gln Gln Trp Ser Ser Asn Pro 85 90 95 Leu Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lys 100 105 <210> 17 <211> 504 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD19xCD3 scFv BLINCYTO incl connector and his-label <400> 17 Asp Ile Gln Leu Thr Gln Ser Pro Ala Ser Leu Ala Val Ser Leu Gly 1 5 10 15 Gln Arg Ala Thr Ile Ser Cys Lys Ala Ser Gln Ser Val Asp Tyr Asp 20 25 30 Gly Asp Ser Tyr Leu Asn Trp Tyr Gln Gln Ile Pro Gly Gln Pro Pro 35 40 45 Lys Leu Leu Ile Tyr Asp Ala Ser Asn Leu Val Ser Gly Ile Pro Pro 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Asn Ile His 65 70 75 80 Pro Val Glu Lys Val Asp Ala Ala Thr Tyr His Cys Gln Gln Ser Thr 85 90 95 Glu Asp Pro Trp Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys Gly 100 105 110 Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Val 115 120 125 Gln Leu Gln Gln Ser Gly Ala Glu Leu Val Arg Pro Gly Ser Ser Val 130 135 140 Lys Ile Ser Cys Lys Ala Ser Gly Tyr Ala Phe Ser Ser Tyr Trp Met 145 150 155 160 Asn Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile Gly Gln 165 170 175 Ile Trp Pro Gly Asp Gly Asp Thr Asn Tyr Asn Gly Lys Phe Lys Gly 180 185 190 Lys Ala Thr Leu Thr Ala Asp Glu Ser Ser Ser Thr Ala Tyr Met Gln 195 200 205 Leu Ser Ser Leu Ala Ser Glu Asp Ser Ala Val Tyr Phe Cys Ala Arg 210 215 220 Arg Glu Thr Thr Thr Val Gly Arg Tyr Tyr Tyr Ala Met Asp Tyr Trp 225 230 235 240 Gly Gln Gly Thr Thr Val Thr Val Ser Ser Gly Gly Gly Gly Ser Asp 245 250 255 Ile Lys Leu Gln Gln Ser Gly Ala Glu Leu Ala Arg Pro Gly Ala Ser 260 265 270 Val Lys Met Ser Cys Lys Thr Ser Gly Tyr Thr Phe Thr Arg Tyr Thr 275 280 285 Met His Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile Gly 290 295 300 Tyr Ile Asn Pro Ser Arg Gly Tyr Thr Asn Tyr Asn Gln Lys Phe Lys 305 310 315 320 Asp Lys Ala Thr Leu Thr Thr Asp Lys Ser Ser Ser Thr Ala Tyr Met 325 330 335 Gln Leu Ser Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys Ala 340 345 350 Arg Tyr Tyr Asp Asp His Tyr Cys Leu Asp Tyr Trp Gly Gln Gly Thr 355 360 365 Thr Leu Thr Val Ser Ser Val Glu Gly Gly Ser Gly Gly Ser Gly Gly 370 375 380 Ser Gly Gly Ser Gly Gly Val Asp Asp Ile Gln Leu Thr Gln Ser Pro 385 390 395 400 Ala Ile Met Ser Ala Ser Pro Gly Glu Lys Val Thr Met Thr Cys Arg 405 410 415 Ala Ser Ser Ser Val Ser Tyr Met Asn Trp Tyr Gln Gln Lys Ser Gly 420 425 430 Thr Ser Pro Lys Arg Trp Ile Tyr Asp Thr Ser Lys Val Ala Ser Gly 435 440 445 Val Pro Tyr Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu 450 455 460 Thr Ile Ser Ser Met Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gln 465 470 475 480 Gln Trp Ser Ser Asn Pro Leu Thr Phe Gly Ala Gly Thr Lys Leu Glu 485 490 495 Leu Lys His His His His His 500 <210> 18 <211> 14 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-L1 <400> 18 Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 1 5 10 <210> 19 <211> 7 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-L2 <400> 19 Gly Thr Lys Phe Leu Ala Pro 1 5 <210> 20 <211> 9 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-L3 <400> 20 Val Leu Trp Tyr Ser Asn Arg Trp Val 1 5 <210> twenty one <211> 5 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-H1 <400> twenty one Lys Tyr Ala Met Asn 1 5 <210> twenty two <211> 19 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-H2 <400> twenty two Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp Ser 1 5 10 15 Val Lys Asp <210> twenty three <211> 14 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C CDR-H3 <400> twenty three His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp Ala Tyr 1 5 10 <210> twenty four <211> 125 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C VH <400> twenty four Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 1 5 10 15 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 20 25 30 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 35 40 45 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 50 55 60 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 65 70 75 80 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 85 90 95 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 100 105 110 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser 115 120 125 <210> 25 <211> 109 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C VL <400> 25 Gln Thr Val Val Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly 1 5 10 15 Thr Val Thr Leu Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly 20 25 30 Asn Tyr Pro Asn Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly 35 40 45 Leu Ile Gly Gly Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe 50 55 60 Ser Gly Ser Leu Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val 65 70 75 80 Gln Pro Glu Asp Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn 85 90 95 Arg Trp Val Phe Gly Gly Gly Thr Lys Leu Thr Val Leu 100 105 <210> 26 <211> 249 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> I2C VH-VL <400> 26 Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 1 5 10 15 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 20 25 30 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 35 40 45 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 50 55 60 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 65 70 75 80 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 85 90 95 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 100 105 110 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser Gly Gly Gly 115 120 125 Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Thr Val Val 130 135 140 Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly Thr Val Thr Leu 145 150 155 160 Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 165 170 175 Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly Leu Ile Gly Gly 180 185 190 Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe Ser Gly Ser Leu 195 200 205 Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val Gln Pro Glu Asp 210 215 220 Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn Arg Trp Val Phe 225 230 235 240 Gly Gly Gly Thr Lys Leu Thr Val Leu 245 <210> 27 <211> 122 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11's CD33 ccVH <400> 27 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Cys Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser 115 120 <210> 28 <211> 122 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11's CD33 VH <400> 28 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Gly Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser 115 120 <210> 29 <211> 5 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11's CD33 HCDR1 <400> 29 Asn Tyr Gly Met Asn 1 5 <210> 30 <211> 17 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11's CD33 HCDR2 <400> 30 Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe Gln 1 5 10 15 Gly <210> 31 <211> 13 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11's CD33 HCDR3 <400> 31 Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr 1 5 10 <210> 32 <211> 113 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 CC VL <400> 32 Asp Ile Val Met Thr Gln Ser Pro Asp Ser Leu Thr Val Ser Leu Gly 1 5 10 15 Glu Arg Thr Thr Ile Asn Cys Lys Ser Ser Gln Ser Val Leu Asp Ser 20 25 30 Ser Thr Asn Lys Asn Ser Leu Ala Trp Tyr Gln Gln Lys Pro Gly Gln 35 40 45 Pro Pro Lys Leu Leu Leu Ser Trp Ala Ser Thr Arg Glu Ser Gly Ile 50 55 60 Pro Asp Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr 65 70 75 80 Ile Asp Ser Pro Gln Pro Glu Asp Ser Ala Thr Tyr Tyr Cys Gln Gln 85 90 95 Ser Ala His Phe Pro Ile Thr Phe Gly Cys Gly Thr Arg Leu Glu Ile 100 105 110 Lys <210> 33 <211> 113 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 VL <400> 33 Asp Ile Val Met Thr Gln Ser Pro Asp Ser Leu Thr Val Ser Leu Gly 1 5 10 15 Glu Arg Thr Thr Ile Asn Cys Lys Ser Ser Gln Ser Val Leu Asp Ser 20 25 30 Ser Thr Asn Lys Asn Ser Leu Ala Trp Tyr Gln Gln Lys Pro Gly Gln 35 40 45 Pro Pro Lys Leu Leu Leu Ser Trp Ala Ser Thr Arg Glu Ser Gly Ile 50 55 60 Pro Asp Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr 65 70 75 80 Ile Asp Ser Pro Gln Pro Glu Asp Ser Ala Thr Tyr Tyr Cys Gln Gln 85 90 95 Ser Ala His Phe Pro Ile Thr Phe Gly Gln Gly Thr Arg Leu Glu Ile 100 105 110 Lys <210> 34 <211> 17 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 LCDR1 <400> 34 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 1 5 10 15 Ala <210> 35 <211> 7 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 LCDR2 <400> 35 Trp Ala Ser Thr Arg Glu Ser 1 5 <210> 36 <211> 9 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 LCDR3 <400> 36 Gln Gln Ser Ala His Phe Pro Ile Thr 1 5 <210> 37 <211> 250 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Features not yet classified <223> CD33 HL CC of E11 <400> 37 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Cys Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys<000189�>85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser Gly Gly Gly Gly Ser Gly 115 120 125 Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Val Met Thr Gln Ser 130 135 140 Pro Asp Ser Leu Thr Val Ser Leu Gly Glu Arg Thr Thr Ile Asn Cys 145 150 155 160 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 165 170 175 Ala Trp Tyr Gln Gln Lys Pro Gly Gln Pro Pro Lys Leu Leu Leu Ser 180 185 190 Trp Ala Ser Thr Arg Glu Ser Gly Ile Pro Asp Arg Phe Ser Gly Ser 195 200 205 Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Asp Ser Pro Gln Pro Glu 210 215 220 Asp Ser Ala Thr Tyr Tyr Cys Gln Gln Ser Ala His Phe Pro Ile Thr 225 230 235 240 Phe Gly Cys Gly Thr Arg Leu Glu Ile Lys 245 250 <210> 38 <211> 250 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> E11 CD33 HL <400> 38 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Gly Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser Gly Gly Gly Gly Ser Gly 115 120 125 Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Val Met Thr Gln Ser 130 135 140 Pro Asp Ser Leu Thr Val Ser Leu Gly Glu Arg Thr Thr Ile Asn Cys 145 150 155 160 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 165 170 175 Ala Trp Tyr Gln Gln Lys Pro Gly Gln Pro Pro Lys Leu Leu Leu Ser 180 185 190 Trp Ala Ser Thr Arg Glu Ser Gly Ile Pro Asp Arg Phe Ser Gly Ser 195 200 205 Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Asp Ser Pro Gln Pro Glu 210 215 220 Asp Ser Ala Thr Tyr Tyr Cys Gln Gln Ser Ala His Phe Pro Ile Thr 225 230 235 240 Phe Gly Gln Gly Thr Arg Leu Glu Ile Lys 245 250 <210> 39 <211> 505 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD33 CC E11 HL x I2C HL Bispecific Molecule <400> 39 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Cys Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser Gly Gly Gly Gly Ser Gly 115 120 125 Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Val Met Thr Gln Ser 130 135 140 Pro Asp Ser Leu Thr Val Ser Leu Gly Glu Arg Thr Thr Ile Asn Cys 145 150 155 160 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 165 170 175 Ala Trp Tyr Gln Gln Lys Pro Gly Gln Pro Pro Lys Leu Leu Leu Ser 180 185 190 Trp Ala Ser Thr Arg Glu Ser Gly Ile Pro Asp Arg Phe Ser Gly Ser 195 200 205 Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Asp Ser Pro Gln Pro Glu 210 215 220 Asp Ser Ala Thr Tyr Tyr Cys Gln Gln Ser Ala His Phe Pro Ile Thr 225 230 235 240 Phe Gly Cys Gly Thr Arg Leu Glu Ile Lys Ser Gly Gly Gly Gly Ser 245 250 255 Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 260 265 270 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 275 280 285 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 290 295 300 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 305 310 315 320 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 325 330 335 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 340 345 350 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 355 360 365 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser Gly Gly Gly 370 375 380 Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Thr Val Val 385 390 395 400 Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly Thr Val Thr Leu 405 410 415 Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 420 425 430 Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly Leu Ile Gly Gly 435 440 445 Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe Ser Gly Ser Leu 450 455 460 Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val Gln Pro Glu Asp 465 470 475 480 Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn Arg Trp Val Phe 485 490 495 Gly Gly Gly Thr Lys Leu Thr Val Leu 500 505 <210> 40 <211> 511 <212> PRT <213> Artificial Sequence <220> <223> Synthetic polypeptide <220> <221> Features not yet classified <223> CD33 E11 HL x I2C HL <400> 40 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Gly Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser Gly Gly Gly Gly Ser Gly 115 120 125 Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Val Met Thr Gln Ser 130 135 140 Pro Asp Ser Leu Thr Val Ser Leu Gly Glu Arg Thr Thr Ile Asn Cys 145 150 155 160 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 165 170 175 Ala Trp Tyr Gln Gln Lys Pro Gly Gln Pro Pro Lys Leu Leu Leu Ser 180 185 190 Trp Ala Ser Thr Arg Glu Ser Gly Ile Pro Asp Arg Phe Ser Gly Ser 195 200 205 Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Asp Ser Pro Gln Pro Glu 210 215 220 Asp Ser Ala Thr Tyr Tyr Cys Gln Gln Ser Ala His Phe Pro Ile Thr 225 230 235 240 Phe Gly Gln Gly Thr Arg Leu Glu Ile Lys Ser Gly Gly Gly Gly Ser 245 250 255 Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 260 265 270 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 275 280 285 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 290 295 300 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 305 310 315 320 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 325 330 335 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 340 345 350 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 355 360 365 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser Gly Gly Gly 370 375 380 Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Thr Val Val 385 390 395 400 Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly Thr Val Thr Leu 405 410 415 Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 420 425 430 Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly Leu Ile Gly Gly 435 440 445 Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe Ser Gly Ser Leu 450 455 460 Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val Gln Pro Glu Asp 465 470 475 480 Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn Arg Trp Val Phe 485 490 495 Gly Gly Gly Thr Lys Leu Thr Val Leu His His His His His 500 505 510 <210> 41 <211> 993 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> CD33 CC x I2C-scFc bispecific HLE molecule <400> 41 Gln Val Gln Leu Val Gln Ser Gly Ala Glu Val Lys Lys Pro Gly Glu 1 5 10 15 Ser Val Lys Val Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Asn Tyr 20 25 30 Gly Met Asn Trp Val Lys Gln Ala Pro Gly Gln Cys Leu Glu Trp Met 35 40 45 Gly Trp Ile Asn Thr Tyr Thr Gly Glu Pro Thr Tyr Ala Asp Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Thr Thr Asp Thr Ser Thr Ser Thr Ala Tyr 65 70 75 80 Met Glu Ile Arg Asn Leu Gly Gly Asp Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Trp Ser Trp Ser Asp Gly Tyr Tyr Val Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Ser Val Thr Val Ser Ser Gly Gly Gly Gly Ser Gly 115 120 125 Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Ile Val Met Thr Gln Ser 130 135 140 Pro Asp Ser Leu Thr Val Ser Leu Gly Glu Arg Thr Thr Ile Asn Cys 145 150 155 160 Lys Ser Ser Gln Ser Val Leu Asp Ser Ser Thr Asn Lys Asn Ser Leu 165 170 175 Ala Trp Tyr Gln Gln Lys Pro Gly Gln Pro Pro Lys Leu Leu Leu Ser 180 185 190 Trp Ala Ser Thr Arg Glu Ser Gly Ile Pro Asp Arg Phe Ser Gly Ser 195 200 205 Gly Ser Gly Thr Asp Phe Thr Leu Thr Ile Asp Ser Pro Gln Pro Glu 210 215 220 Asp Ser Ala Thr Tyr Tyr Cys Gln Gln Ser Ala His Phe Pro Ile Thr 225 230 235 240 Phe Gly Cys Gly Thr Arg Leu Glu Ile Lys Ser Gly Gly Gly Gly Ser 245 250 255 Glu Val Gln Leu Val Glu Ser Gly Gly Gly Leu Val Gln Pro Gly Gly 260 265 270 Ser Leu Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Asn Lys Tyr 275 280 285 Ala Met Asn Trp Val Arg Gln Ala Pro Gly Lys Gly Leu Glu Trp Val 290 295 300 Ala Arg Ile Arg Ser Lys Tyr Asn Asn Tyr Ala Thr Tyr Tyr Ala Asp 305 310 315 320 Ser Val Lys Asp Arg Phe Thr Ile Ser Arg Asp Asp Ser Lys Asn Thr 325 330 335 Ala Tyr Leu Gln Met Asn Asn Leu Lys Thr Glu Asp Thr Ala Val Tyr 340 345 350 Tyr Cys Val Arg His Gly Asn Phe Gly Asn Ser Tyr Ile Ser Tyr Trp 355 360 365 Ala Tyr Trp Gly Gln Gly Thr Leu Val Thr Val Ser Ser Gly Gly Gly 370 375 380 Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gln Thr Val Val 385 390 395 400 Thr Gln Glu Pro Ser Leu Thr Val Ser Pro Gly Gly Thr Val Thr Leu 405 410 415 Thr Cys Gly Ser Ser Thr Gly Ala Val Thr Ser Gly Asn Tyr Pro Asn 420 425 430 Trp Val Gln Gln Lys Pro Gly Gln Ala Pro Arg Gly Leu Ile Gly Gly 435 440 445 Thr Lys Phe Leu Ala Pro Gly Thr Pro Ala Arg Phe Ser Gly Ser Leu 450 455 460 Leu Gly Gly Lys Ala Ala Leu Thr Leu Ser Gly Val Gln Pro Glu Asp 465 470 475 480 Glu Ala Glu Tyr Tyr Cys Val Leu Trp Tyr Ser Asn Arg Trp Val Phe 485 490 495 Gly Gly Gly Thr Lys Leu Thr Val Leu Gly Gly Gly Gly Asp Lys Thr 500 505 510 His Thr Cys Pro Pro Cys Pro Ala Pro Glu Leu Leu Gly Gly Pro Ser 515 520 525 Val Phe Leu Phe Pro Pro Lys Pro Lys Asp Thr Leu Met Ile Ser Arg 530 535 540 Thr Pro Glu Val Thr Cys Val Val Val Asp Val Ser His Glu Asp Pro 545 550 555 560 Glu Val Lys Phe Asn Trp Tyr Val Asp Gly Val Glu Val His Asn Ala 565 570 575 Lys Thr Lys Pro Cys Glu Glu Gln Tyr Gly Ser Thr Tyr Arg Cys Val 580 585 590 Ser Val Leu Thr Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu Tyr 595 600 605 Lys Cys Lys Val Ser Asn Lys Ala Leu Pro Ala Pro Ile Glu Lys Thr 610 615 620 Ile Ser Lys Ala Lys Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr Leu 625 630 635 640 Pro Pro Ser Arg Glu Glu Met Thr Lys Asn Gln Val Ser Leu Thr Cys 645 650 655 Leu Val Lys Gly Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser 660 665 670 Asn Gly Gln Pro Glu Asn Asn Tyr Lys Thr Thr Pro Pro Val Leu Asp 675 680 685 Ser Asp Gly Ser Phe Phe Leu Tyr Ser Lys Leu Thr Val Asp Lys Ser 690 695 700 Arg Trp Gln Gln Gly Asn Val Phe Ser Cys Ser Val Met His Glu Ala 705 710 715 720 Leu His Asn His Tyr Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly Lys 725 730 735 Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly 740 745 750 Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Asp Lys 755 760 765 Thr His Thr Cys Pro Pro Cys Pro Ala Pro Glu Leu Leu Gly Gly Pro 770 775 780 Ser Val Phe Leu Phe Pro Pro Lys Pro Lys Asp Thr Leu Met Ile Ser 785 790 795 800 Arg Thr Pro Glu Val Thr Cys Val Val Val Asp Val Ser His Glu Asp 805 810 815 Pro Glu Val Lys Phe Asn Trp Tyr Val Asp Gly Val Glu Val His Asn 820 825 830 Ala Lys Thr Lys Pro Cys Glu Glu Gln Tyr Gly Ser Thr Tyr Arg Cys 835 840 845 Val Ser Val Leu Thr Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu 850 855 860 Tyr Lys Cys Lys Val Ser Asn Lys Ala Leu Pro Ala Pro Ile Glu Lys 865 870 875 880 Thr Ile Ser Lys Ala Lys Gly Gln Pro Arg Glu Pro Gln Val Tyr Thr 885 890 895 Leu Pro Pro Ser Arg Glu Glu Met Thr Lys Asn Gln Val Ser Leu Thr 900 905 910 Cys Leu Val Lys Gly Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu 915 920 925 Ser Asn Gly Gln Pro Glu Asn Asn Tyr Lys Thr Thr Pro Pro Val Leu 930 935 940 Asp Ser Asp Gly Ser Phe Phe Leu Tyr Ser Lys Leu Thr Val Asp Lys 945 950 955 960 Ser Arg Trp Gln Gln Gly Asn Val Phe Ser Cys Ser Val Met His Glu 965 970 975 Ala Leu His Asn His Tyr Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly 980 985 990 Lys <210> 42 <211> 5 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> EGFRvIIIxCD3-scFc VH CDR1 <400> 42 Asn Tyr Gly Met His 1 5 <210> 43 <211> 17 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> EGFRvIIIxCD3-scFc VH CDR2 <400> 43 Val Ile Trp Tyr Asp Gly Ser Asp Lys Tyr Tyr Ala Asp Ser Val Arg 1 5 10 15 Gly <210> 44 <211> 15 <212> PRT <213> Artificial sequence <220> <223> Synthetic peptides <220> <221> Unclassified features <223> EGFRvIIIxCD3-scFc VH CDR3 <400> 44 Asp Gly Tyr Asp Ile Leu Thr Gly Asn Pro Arg Asp Phe Asp Tyr 1 5 10 15 ...

Claims

1. A method for purifying a single-chain bispecific antibody construct produced from mammalian cells, said construct comprising at least two domains: • A first structural domain with a pI of 5.0 to 9.5, which is combined with CD33, CD19, CLDN18.2, or DLL3. • A second domain that binds to an extracellular domain of the human and macaque CD3ε chain; and • A third structural domain comprising two polypeptide monomers, each polypeptide monomer comprising a hinge, a CH2 domain, and a CH3 domain, wherein the two polypeptide monomers are fused together via a peptide linker, and wherein the construct comprises at least one open disulfide bond, wherein one or more of the structural domains contain disulfide bonds. The single-chain bispecific antibody construct comprises a sequence selected from SEQ ID NO: 41, 104, 114 and 225; The method includes the step of refolding the construct, wherein the construct is contacted with a refolding buffer to refold the construct back to its natural form. The buffer comprises (i) a redox agent and / or an oxidizing agent at a concentration of at least 0.1 mM, wherein the redox agent is cysteine / cysteine, wherein the redox agent is present at a concentration of 0.1 to 10 mM, and wherein the oxidizing agent is selected from the group consisting of copper(II) and (L)-dehydroascorbic acid, wherein the oxidizing agent is present at a concentration of 0.1 to 10 mM, and (ii) a dissociation agent with a concentration of 1.2 M to 2.0 M, wherein the dissociation agent is a guanidine salt. Furthermore, the pH of the refolding buffer corresponds to + / - 4.5 of the pI value of the first domain or + / - 4.0 of the pI value of the entire construct.

2. The method of claim 1, wherein the oxidant is copper(II).

3. The method of claim 1, wherein the oxidizing agent is copper(II) sulfate.

4. The method of claim 1, wherein the refolding buffer comprises an oxidizing agent but not a redox agent.

5. The method of claim 1, wherein the pH value is 5.0 to 6.0 or 7.0 to 9.

0.

6. The method of claim 1, wherein the monomer content of the construct, as determined after application of the refolding buffer, is at least 60%, at least 65%, at least 70%, or at least 85%.

7. The method of claim 6, wherein the monomer content of the construct is determined by size exclusion chromatography.

8. The method of claim 7, wherein the monomer content of the construct is determined by monomer peak percentage.

9. The method of claim 1, wherein the method is performed at a temperature between 4ºC and 30ºC.

10. The method of claim 1, wherein the incubation time is 1 to 24 h.

11. The method of claim 1, wherein contacting the construct with the refolding buffer is performed (i) in a processing fluid during a downstream purification step or (ii) on a chromatographic column during a downstream purification step.

12. The method of claim 1, wherein the concentration of the construct (i) in the liquid cell is from 0.5 mg / ml to 15 mg / ml or (ii) on the chromatographic column is from 5 mg / ml to 15 mg / ml.

13. The method of claim 12, wherein the liquid pool is selected from the group consisting of: harvested cell culture medium (HCCF), chromatographic eluent pool, protein A, protein L or protein G affinity chromatography eluent pool, filtered virus inactivation pool, cation exchange chromatography (CEX) monomer pool and CEX high molecular weight post-peak pool.

14. The method of claim 1, further comprising the following steps: (a) Provide harvested cell culture medium (HCCF) containing the bispecific antibody construct secreted by mammalian cells; (b) Under conditions suitable for association between the construct and the separation matrix, the HCCF is contacted with a first separation matrix for chromatographic purification, wherein the first separation matrix is ​​an affinity resin selected from the group consisting of: protein A, protein G, protein L and synthetic analog affinity resins. (c) Wash the first separation matrix; (d) Elute the construct from the first separation matrix; (e) Inactivate the virus in the eluent containing the construct; (f) Under conditions suitable for association between the construct and the second separation matrix, an eluent containing the construct is contacted with the second separation matrix for chromatographic purification, wherein the second separation matrix is ​​a cation exchange chromatographic matrix; (g) Wash the second separation matrix; (h) The construct is eluted from the second separation matrix, wherein the construct is separated into two fractions: (i) a monomer pool containing construct monomers and (ii) a high molecular weight pool containing construct aggregates. (i) The virus is separated from the eluent containing fraction (i) by filtration; as well as (j) Optionally, the eluent may be subjected to further ultrafiltration and / or percolation steps. Among them (1) In cell culture medium prior to harvest or in HCCF in step (a), and / or (2) On at least one of the separation substrates in step (b) or (f), and / or (3) In the processing fluid in at least one step (d), (e), or (h) The construct is brought into contact with the refolding buffer to refold the construct back into its natural form.

15. The method of claim 14, wherein the separation matrix is ​​a column.

16. The method of claim 1, wherein the method is applied to a construct produced by an upstream continuous manufacturing method.

Images